Start / Teknik / Environment Variables

Environment Variables

Each episode we discuss the latest news regarding how to reduce the emissions of software and how the industry is dealing with its own environmental impact.

102 avsnitt • Längd: 40 min • Månadsvis

Om podden

Each episode we discuss the latest news regarding how to reduce the emissions of software and how the industry is dealing with its own environmental impact. Brought to you by The Green Software Foundation.

The podcast Environment Variables is created by Green Software Foundation. The podcast and the artwork on this page are embedded on this page using the public podcast feed (RSS).

Avsnitt

Backstage: Green Software Patterns

27 mars 2025 | 12 min

In this episode, Chris Skipper takes us backstage into the Green Software Patterns Project, an open-source initiative designed to help software practitioners reduce emissions by applying vendor-neutral best practices. Guests Franziska Warncke and Liya Mathew, project leads for the initiative, discuss how organizations like AVEVA and MasterCard have successfully integrated these patterns to enhance software sustainability. They also explore the rigorous review process for new patterns, upcoming advancements such as persona-based approaches, and how developers and researchers can contribute.

Learn more about our people:

Chris Skipper: LinkedIn | Website
Franziska Warncke: LinkedIn
Liya Mathew: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Green Software Patterns | GSF [00:23]
GitHub - Green Software Patterns | GSF [ 05:42]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Chris Skipper: Welcome to Environment Variables, where we bring you the latest news from the world of sustainable software development. I am the producer of the show, Chris Skipper, and today we're excited to bring you another episode of Backstage, where we uncover the stories, challenges, and innovations driving the future of green software.

In this episode, we're diving into the Green Software Patterns Project, an open source initiative designed to curate and share best practices for reducing software emissions.

The project provides a structured approach for software practitioners to discover, contribute, and apply vendor-neutral green software patterns that can make a tangible impact on sustainability. Joining us today are Franziska Warncke and Liya Mathew, the project leads for the Green Software Patterns Initiative.

They'll walk us through how the project works, its role in advancing sustainable software development, and what the future holds for the Green Software Patterns. Before we get started, a quick reminder that everything we discuss in this episode will be linked in the show notes below. So without further ado, let's dive into our first question about the Green Software Patterns project. My first question is for Liya. The project is designed to help software practitioners reduce emissions in their applications.

What are some real world examples of how these patterns have been successfully applied to lower carbon footprints?

Liya Mathew: Thanks for the question, and yes, I am pretty sure that there are a lot of organizations as well as individuals who have greatly benefited from this project. A key factor behind the success of this project is the impact that these small actions can have on longer runs. For example, AVEVA has been an excellent case of an organization that embraced these patterns.

They created their own scoring system based on Patterns which help them measure and improve their software sustainability. Similarly, MasterCard has also adopted and used these patterns effectively. What's truly inspiring is that both AVEVA and MasterCard were willing to share their learnings with the GSF and the open source community as well.

Their contributions will help others learn and benefit from their experiences, fostering a collaborative environment where everyone can work towards a more sustainable software.

Chris Skipper: Green software patterns must balance general applicability with technical specificity. How do you ensure that these patterns remain actionable and practical across different industries, technologies and software architectures?

Liya Mathew: One of the core and most useful features of patterns is the ability to correlate the software carbon intensity specification. Think of it as a bridge that connects learning and measurement. When we look through existing catalog of patterns, one essential thing that stands out is their adaptability.

Many of these patterns not only align with sustainability, but also coincide with security and reliability best practices. The beauty of this approach is that we don't need to completely rewrite our software architecture to make it more sustainable. Small actions like catching static data or providing a dark mode can make significant difference.

These are simple, yet effective steps that can lead us a long way towards sustainability. Also, we are nearing the graduation of Patterns V1. This milestone marks a significant achievement and we are already looking ahead to the next exciting phase: Patterns V2. In Patterns V2, we are focusing on persona-based and behavioral patterns, which will bring even more tailored and impactful solutions to our community.

These new patterns will help address specific needs and behaviors, making our tools even more adaptable and effective.

Chris Skipper: The review and approval process for new patterns involves multiple stages, including subject matter expert validation and team consensus. Could you walk us through the workflow for submitting and reviewing patterns?

Liya Mathew: Sure. The review and approval process for new patterns involve multiple stages, ensuring that each pattern meets a standard before integration. Initially, when a new pattern is submitted, it undergoes an initial review by our initial reviewers. During this stage, reviewers check if the pattern aligns with the GSF's mission of reducing software emissions, follows the GSF Pattern template, and adheres to proper formatting rules. They also ensure that there is enough detail for the subject matter expert to evaluate the pattern. If any issue arises, the reviewer provides clear and constructive feedback directly in the pull request, and the submitter updates a pattern accordingly.

Once the pattern passes the initial review, it is assigned to an appropriate SME for deeper technical review, which should take no more than a week, barring any lengthy feedback cycles. The SME checks for duplicate patterns validates the content as assesses efficiency and accuracy of the pattern in reducing software remission.

It also ensures that the pattern's level of depth is appropriate. If any areas are missing or incomplete, the SME provides feedback in the pull request. If the patterns meet all the criteria, SME will then remove the SME review label and adds a team consensus label and assigns this pull request back to the initial reviewer.

Then the Principles and Patterns Working Group has two weeks to comment or object to the pattern, requiring a team consensus before the PR can be approved and merged in the development branch. Thus the raw process ensures that each pattern is well vetted and aligned with our goals.

Chris Skipper: For listeners who want to start using green software patterns in their projects, what's the best way to get involved, access the catalog, or submit a new pattern?

Liya Mathew: All the contributions are made via GitHub pull requests. You can start by submitting a pull request on our repository. Additionally, we would love to connect with everyone interested in contributing. Feel free to reach out to us on LinkedIn or any social media handles and express your interest in joining our project's weekly calls.

Also, check if your organization is a member of the Green Software Foundation. We warmly welcome contributions in any capacity. As mentioned earlier, we are setting our sights on a very ambitious goal for this project, and your involvement would be invaluable.

Chris Skipper: Thanks to Liya for those great answers. Next, we had some questions for Franziska. The Green Software Patterns project provides a structured open source database of curated software patterns that help reduce software emissions. Could you give us an overview of how the project started and its core mission?

Franziska Warncke: Great question. The Green Software Patterns project emerged from a growing recommendation of the environmental impact of software and the urgent need for sustainable software engineering practices. As we've seen the tech industry expand, it became clear that while hardware efficiency has been a focal point for sustainability, software optimization was often overlooked.

A group of dedicated professionals began investigating existing documentation, including resources like the AWS Well-Architected Framework, and this exploration laid to groundwork for the project. This allows us to create a structured approach to the curating of the patterns that can help reduce software emissions.

We developed a template that outlines how each pattern should be presented, ensuring clarity and consistency. Additionally, we categorize these patterns into the three main areas, cloud, web, and AI.

Chris Skipper: Building an open source knowledge base and ensuring it remains useful, requires careful curation and validation. What are some of the biggest challenges your team has faced in developing and maintaining the green software patterns database?

Franziska Warncke: Building and maintaining an open source knowledge base like the Green Software Patterns database, comes with its own set of challenges. One of the biggest hurdles we've encountered is resource constraints. As an open source project, we often operate with limited time personnel, which makes it really, really difficult to prioritize certain tasks over others.

Despite this challenge, we are committed to continuous improvement, collaboration, and community engagement to ensure that the Green Software Patterns database remains a valuable resource for developers looking to adopt more sustainable practices.

Chris Skipper: Looking ahead, what are some upcoming initiatives for the project? Are there any plans to expand the pattern library or introduce new methodologies for evaluating and implementing patterns?

Franziska Warncke: Yes, we have some exciting initiatives on the horizon. So one of our main focuses is to restructure the patterns catalog to adopt the persona-based approach. This means we want to create tailored patterns for various worlds within the software industry, like developers, project managers, UX designers, and system architects.

By doing this, we aim to make the patents more relevant and accessible to a broader audience. We are also working on improving the visualization of the patterns. We recognize that user-friendly visuals are crucial for helping people understand and adopt these patterns in their own projects, which was really missing before.

In addition to that, we plan to categorize the patterns based on different aspects. Such as persona type, adoptability and effectiveness. This structured approach will help users quickly find the patterns that are most relevant to their roads and their needs, making the entire experience much more streamlined. Moreover, we are actively seeking new contributors to join us.

And we believe that the widest set of voices and perspective will enrich our knowledge base and ensure that our patterns reflect a wide range of experience. So, if anyone is interested, we'd love to hear from you.

Chris Skipper: The Green Software Patterns Project is open source and community-driven. How can developers, organizations, and researchers contribute to expanding the catalog and improving the quality of the patterns?

Franziska Warncke: Yeah,

the Green Software Patterns Project is indeed open source and community driven, and we welcome contributions from developers, organizations, and researchers to help expand our catalog and improve the quality of the patterns. We need people to review the existing patterns critically and provide feedback.

This includes helping us categorize them for a specific persona, ensuring that each pattern is tailored to each of various roles in the software industry. Additionally, contributors can assist by adding more information and context to the patterns, making them more comprehensive and useful. Visuals are another key area where we need help.

Creating clear and engaging visuals that illustrate how to implement these patterns can significantly enhance their usability. Therefore, we are looking for experts who can contribute their skills in design and visualization to make the patterns more accessible. So if you're interested, then we would love to have you on board.

Thank you.

Chris Skipper: Thanks to Franziska for those wonderful answers. So we've reached the end of the special backstage episode on the Green Software Patterns Project at the GSF. I hope you enjoyed the podcast. To listen to more podcasts about green software, please visit podcast.greensoftware.foundation. And we'll see you on the next episode.

Bye for now.

The Week in Green Software: Sustainable AI Progress

20 mars 2025 | 50 min

For this 100th episode of Environment Variables, guest host Anne Currie is joined by Holly Cummins, senior principal engineer at Red Hat, to discuss the intersection of AI, efficiency, and sustainable software practices. They explore the concept of "Lightswitch Ops"—designing systems that can easily be turned off and on to reduce waste—and the importance of eliminating zombie servers. They cover AI’s growing energy demands, the role of optimization in software sustainability, and Microsoft's new shift in cloud investments. They also touch on AI regulation and the evolving strategies for balancing performance, cost, and environmental impact in tech.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Holly Cummins: LinkedIn | GitHub | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Data-driven grid decarbonization - Webinar | March 19, 2025
The First Eco-Label for Sustainable Software - Frankfurt am Main | March 27, 2025

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Holly Cummins: Demand for AI is growing, demand for AI will grow indefinitely. But of course, that's not sustainable. Again, you know, it's not sustainable in terms of financially and so at some point there will be that correction.

I'm your host, Chris Adams.

Anne Currie: So hello and welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software. Now, today you're not hearing the dulcet tones of your usual host, Chris Adams. I am a guest host on this, a common guest, a frequent guest host, Anne Currie. And my guest today is somebody I've known for quite a few years and I'm really looking forward to chatting to, Holly.

So do you want to introduce yourself, Holly?

Holly Cummins: So I'm Holly Cummins. I work for Red Hat. My day job is that, I'm a senior principal engineer and I'm helping to develop Quarkus, which is Java middleware. And I'm looking at the ecosystem of Quarkus, which sounds really sustainability oriented, but actually the day job aspect is I'm more looking at

the contributors and, you know, the extensions and that kind of thing. But one of the other things that I do end up looking a lot at is the ecosystem aspect of Quarkus in terms of sustainability. Because Quarkus is a extremely efficient Java runtime. And so when I joined the team, one of the things we asked well, one of the things I asked was, can we, know this is really efficient. Does that translate into an environmental, you know, benefit? Is it actually benefiting the ecosystem? You know, can we quantify it? And so we did that work and we were able to sort of validate our intuition that it did have a much lower carbon footprint, which was nice.

But some things of what we did actually surprised us as well, which was also good because it's always good to be challenged in your assumptions. And so now part of what I'm doing as well is sort of broadening that focus from, instead of measuring what we've done in the past, thinking about, well, what does a sustainable middleware architecture look like?

What kind of things do we need to be providing?

Anne Currie: Thank you very much indeed. That's a really good overview of what I really primarily want to be talking about today. We will be talking about a couple of articles as usual on AI, but really I want to be focused on what you're doing in your day job because I think it's really interesting and incredibly relevant.

So, as I said, my name is Anne Currie. I am the CEO of a learning and development company called Strategically Green. We do workshops and training around building green software and changing your systems to align with renewables. But I'm also one of the authors of O'Reilly's new book, Building Green Software, and Holly was probably the most, the biggest single reviewer/contributor to that book, and it was in her best interest to do so because, we make, I make tons and tons of reference to a concept that you came up with.

I'm very interested in the backstory to this concept, but perhaps you can tell me a little bit more about it because it is, this is something I've not said to you before, but it is, this comes up in review feedback, for me, for the book, more than any other concept in the book. Lightswitch Ops. People saying, "Oh, we've put in, we've started to do Lightswitch Ops."

If anybody says "I've started to do" anything, it's always Lightswitch Ops. So tell us, what is Lightswitch Ops?

Holly Cummins: So Lightswitch Ops, it's really, it's about architecting your systems so that they can tolerate being turned off and on, which sounds, you know, it sounds sort of obvious, but historically that's not how our systems have worked. And so the first step is architect your system so that they can tolerate being turned off and on.

And then the next part is once you have that, actually turn them off and on. And, it sort of, it came about because I'm working on product development now, and I started my career as a performance engineer, but in between those two, I was a client facing consultant, which was incredibly interesting.

And it was, I mean, there was, so many things that were interesting, but one of the things that I sort of kept seeing was, you know, you sort of work with clients and some of them you're like, "Oh wow, you're, you know, you're really at the top of your game" and some you think, "why are you doing this way when this is clearly, you know, counterproductive" or that kind of thing.

And one of the things that I was really shocked by was how much waste there was just everywhere. And I would see things like organizations where they would be running a batch job and the batch job would only run at the weekends, but the systems that supported it would be up 24/7. Or sometimes we see the opposite as well, where it's a test system for manual testing and people are only in the office, you know, nine to five only in one geo and the systems are up 24 hours.

And the reason for this, again, it's sort of, you know, comes back to that initial thing, it's partly that we just don't think about it and, you know, that we're all a little bit lazy, but it's also that many of us have had quite negative experiences of if you turn your computer off, it will never be the same when it comes back up.

I mean, I still have this with my laptop, actually, you know, I'm really reluctant to turn it off. But now we have, with laptops, we do have the model where you can close the lid and it will go to sleep and you know that it's using very little energy, but then when you bring it back up in the morning, it's the same as it was without having to have the energy penalty of keeping it on overnight. And I think, when you sort of look at the model of how we treat our lights in our house, nobody has ever sort of left a room and said, "I could turn the light off, but if I turn the light off, will the light ever come back on in the same form again?"

Right? Like we just don't do that. We

have a great deal of confidence that it's reliable to turn a light off and on and that it's low friction to do it. And so we need to get to that point with our computer systems. And you can sort roll with the analogy a bit more as well, which is in our houses, it tends to be quite a manual thing of turning the lights off and on.

You know, I turn the light on when I need it. In institutional buildings, it's usually not a manual process to turn the lights off and on. Instead, what we end up is, we end up with some kind of automation. So, like, often there's a motion sensor. So, you know, I used to have it that if I would stay in our office late at night, at some point if you sat too still because you were coding and deep in thought, the lights around you would go off and then you'd have to, like, wave your arms to make the lights go back on.

And it's that, you know, it's this sort of idea of like we can detect the traffic, we can detect the activity, and not waste the energy. And again, we can do exactly this our computer systems. So we can have it so that it's really easy to turn them off and on. And then we can go one step further and we can automate it and we can say, let's script to turn things off at 5pm because we're only in one geo.

And you know, if we turn them off at 5pm, then we're enforcing quite a strict work life balance. So...

Anne Currie: Nice, nice work.

Holly Cummins: Yeah. Sustainable. Sustainable pace. Yeah. Or we can do sort of, you know, more sophisticated things as well. Or we can say, okay, well, let's just look at the traffic and if there's no traffic to this, let's turn it off.

off

Anne Currie: Yeah, it is an interestingly simple concept because it's,

when people come up with something which is like, in some ways, similar analogies, a light bulb moment of, you know, why don't people turn things off? Becasue, so Holly, everybody is an unbelievably good public speaker.

One of the best public speakers out there at the moment. And we first met because you came and gave talks at, in some tracks I was hosting on a variety. Some on high performance code, code efficiency, some on, being green. One of the stories you told was about your Lightswitch moment, the realization that actually this was a thing that needed to happen.

And I thought it was fascinating. It was about how, I know everybody, I've been in the tech industry for a long time, so I've worked with Java a lot over the years and many years ago. And one of the issues with Java in the old days was always, it was very hard to turn things off and turn them back on again.

And that was fine in the old world, but you talked about how that was no longer fine. And that was an issue with the cloud because the cloud, using the cloud well, turning things on and off and things, doing things like auto scaling is utterly key to the idea of the cloud. And therefore it had to become part of Quarkus, part of the future of Java. Am I right in that understanding?

Holly Cummins: Yeah, absolutely. And the cloud sort of plays into both parts of the story, actually. So definitely we, the things that we need to be cloud native, like being able to support turning off and on again, are very well aligned to what you need to support Lightswitch Ops. And so the, you know, there with those two, we're pulling in the same direction.

The needs of the cloud and the needs of sustainability are both driving us to make systems that, I just saw yesterday, sorry this is a minor digression, but I was looking something up, and we used to talk a lot about the Twelve-Factor App, and you know, at the time we started talking about Twelve-Factor Apps, those characteristics were not at all universal. And then someone came up with the term, the One-Factor App, which was the application that could just tolerate being turned off and on.

And sometimes even that was like too much of a stretch. And so there's the state aspect to it, but then there's also the performance aspect of it and the timeliness aspect of it. And that's really what Quarkus has been looking at that if you want to have any kind of auto scaling or any kind of serverless architecture or anything like that, the way Java has historically worked, which is that it eats a lot of memory and it takes a long time to start up, just isn't going to work.

And the sort of the thing that's interesting about that is quite often when we talk about optimizing things or becoming more efficient or becoming greener, it's all about the trade offs of like, you know, "oh, I could have the thing I really want, or I could save the world. I guess I should save the world." But sometimes what we can do is we can just find things that we were paying for, that we didn't even want anymore. And that's, I think, what Quarkus was able to do. Because a lot of the reason that Java has a big memory footprint and a lot of the reason that Java is slow to start up is it was designed for a different kind of ops.

The cloud didn't exist. CI/CD didn't exist. DevOps didn't exist. And so the way you built your application was you knew you would get a release maybe once a year and deployment was like a really big deal. And you know, you'd all go out and you'd have a party after you successfully deployed because it was so challenging.

And so you wanted to make sure that everything you did was to avoid having to do a deployment and to avoid having to talk to the ops team because they were scary. But of course, even though we had this model where releases happen very rarely, or the big releases happen very rarely, of course, the world still moves on, you know, people still had defects, people, so what you ended up with was something that was really much more optimized towards patching.

So can we take the system and without actually taking, turning it off and on, because that's almost impossible, can we patch it? So everything was about trying to change the engine of the plane while the plane was flying, which is really clever engineering. If you can support that, you know, well done you.

It's so dynamic. And so everything was optimized so that, you know, you could change your dependencies and things would keep working. And, you know, you could even change some fairly important characteristics of your dependencies and everything would sort of adjust and it would ripple back through the system.

But because that dynamism was baked into every aspect of the architecture, it meant that everything just had a little bit of drag, and everything had a little bit of slowdown that came from that indirection. And then now you look at it in the cloud and you think, well, wait a minute. I don't need that. I don't need that indirection.

I don't need to be able to patch because I have a CI/CD pipeline, and if I'm going into my production systems and SSHing in to change my binaries, something has gone horribly wrong with my process. And you know, I need to, I have all sorts of problems. So really what Quarkus was able to do was get rid of a whole bunch of reflection, get rid of a whole bunch of indirection,

do more upfront at build time. And then that gives you much leaner behavior at runtime, which is what you want in a cloud environment.

Anne Currie: Yeah. And what I love about this and love about the story of Quarkus is, it's aligned with something, non functional requirements. It's like, it's an unbelievably boring name, and for something which is a real pain point for companies. But it's also, in many ways, the most important thing and the most difficult thing that we do.

It's like, being secure, being cost effective, being resilient. A lot of people say to me, well, you know, actually all you're doing with green is adding another non functional requirement. We know those are terrible. But I can say, no, we need to not make it another non functional requirements. It's just a good, another motivator for doing the first three well, you know. Also scaling is about resilience. It's about cost saving, and it's about being green. And it's about, and being able to pave rather than patch, I think is, was the term. It's more secure, you know. Actually patching is much less secure than repaving, taking everything down and bringing it back up.

All the modern thinking about being more secure, being faster, being cheaper, being more resilient is aligned or needs to be aligned with being green and it can be, and it should be, and it shouldn't just be about doing less.

Holly Cummins: Absolutely. And, you know, especially for the security aspect, when you look at something like tree shaking, that gives you more performance by getting rid of the code that you weren't using. Of course, it makes you more secure as well because you get rid of all these code paths and all of these entry points and vulnerabilities that had no benefit to you, but were still a vulnerability.

Anne Currie: Yeah, I mean, one of the things that you've talked about Lightswitch Ops being related to is, well, actually not Lightswitch Ops, but the thing that you developed before Lightswitch Ops, the concept of zombie servers. Tell us a little bit about that because that not only is cost saving, it's a really big security improvement.

So tell us about zombie, the precursor to Lightswitch Ops.

Holly Cummins: Yeah, zombie servers are again, one of those things that I sort of, I noticed it when I was working with clients, but I also noticed it a lot in our own development practices that what we would do was we would have a project and we would fire up a server in great excitement and you know, we'd register something on the cloud or whatever.

And then we'd get distracted and then, or then we, you know, sometimes we would develop it but fail to go to production. Sometimes we'd get distracted and not even develop it. And I looked and I think some of these costs became more visible and more obvious when we move to the cloud, because it used to be that when you would provision a server, once it was provisioned, you'd gone through all of the pain of provisioning it and it would just sit there and you would keep it in case you needed it.

But with the cloud, all of a sudden, keeping it until you needed it had a really measurable cost. And I looked and I realized, you know, I was spending, well, I wasn't personally spending, I was costing my company thousands of pounds a month on these cloud servers that I'd provisioned and forgotten about.

And then I looked at how Kubernetes, the sort of the Kubernetes servers were being used and some of the profiles of the Kubernetes servers. And I realized that, again, there's, each company would have many clusters. And I was thinking, are they really using all of those clusters all of the time?

And so I started to look into it and then I realized that there had been a lot of research done on it and it was shocking. So again, you know, the sort of the, I have to say I didn't coin the term zombie servers. I talk about it a lot, but, there was a company called the Antithesis Institute.

And what they did, although actually, see, now I'm struggling with the name of it because I always thought they were called the Antithesis Institute. And I think it's actually a one letter variant of that, which is much less obvious as a word, but much more distinctive. But I've, every time I talked about them, I mistyped it.

And now I can't remember which one is the correct one, but in any case, it's something like the Antithesis Institute. And they did these surveys and they found that, it was something like a third of the servers that they looked at were doing no work at all. Or rather no, no useful work. So they're still consuming energy, but there's no work being done.

And when they say no useful work as well, that sounds like a kind of low bar. Because when I think about my day job, quite a lot of it is doing work that isn't useful. But they had, you know, it wasn't like these servers were serving cat pictures or that kind of thing. You know, these servers were doing nothing at all.

There was no traffic in, there was no traffic out. So you can really, you know, that's just right for automation to say, "well, wait a minute, if nothing's going in and nothing's coming out, we can shut this thing down." And then there was about a further third that had a utilization that was less than 5%.

So again, you know, this thing, it's talking to the outside world every now and then, but barely. So again, you know, it's just right for a sort of a consolidation. But the, I mean, the interesting thing about zombies is as soon as you talk about it, usually, you know, someone in the audience, they'll turn a little bit green and they'll go, "Oh, I've just remembered that server that I provisioned."

And sometimes, you know, I'm the one giving the talk and I'm like, Oh, while preparing this talk, I just realized I forgot a server, because it's so easy to do. And the way we're measured as well, and the way we measure our own productivity is we give a lot more value to creating than to cleaning up.

Anne Currie: Yeah. And in some ways that makes sense because, you know, creating is about growth and cleaning up you know, it's about degrowth. It's about like, you know, it's like you want to tell the story of growth, but I've heard a couple of really interesting, sales on zombie servers since you started, well, yeah, since you started talking about it, you may not have invented it, but you popularized it. One was from, VMware, a cost saving thing. They were, and it's a story I tell all the time about when they were moving data centers in Singapore, setting up a new data center in Singapore.

They decided to do a review of all their machines to see what had to go across. And they realized that 66 percent of their machines did not need to be reproduced in the new data center. You know, they had a, and that was VMware. People who are really good at running data centers. So imagine what that's like.

But moving data centers is a time when it often gets spotted. But I will say, a more, a differently disturbing story from a company that wished to remain nameless. Although I don't think they need to because I think it's just an absolutely bog standard thing. They were doing a kind of thriftathon style thing of reviewing their data center to see if there was stuff that they could save money on, and they found a machine that was running at 95, 100 percent CPU, and they thought, they thought, Oh my God, it's been hacked.

It's been hacked. Somebody's mining Bitcoin on this. It's, you know, or maybe it's attacking us. Who knows? And so they went and they did some searching around internally, and they found out that it was somebody who turned on a load test, and then forgot to turn it off three years previously. And And the, I would say that obviously that came up from the cost, but it also came up from the fact that machine could have been hacked.

You know, it could be, could have been mining Bitcoin. It could have been attacking them. It could have been doing anything. They hadn't noticed because it was a machine that no one was looking at. And I thought it was an excellent example. I thought those two, excellent examples of the cost and the massive security hole that comes from machines that nobody is looking at anymore.

So, you know, non functional requirements, they're really important. And

Holly Cummins: Yeah.

Anne Currie: doing better on them is also green. And also, they're very, non functional requirements are really closely tied together.

Holly Cummins: Yeah. I mean, oh, I love both of those stories. And I've heard the VMware one before, but I hadn't heard the one about the hundred percent, the load test. That is fantastic. One of the reasons I like talking about zombies and I think one of the reasons people like hearing about it I mean, it's partly the saving the world.

But also I think when we look at greenness and sustainability, some of it is not a very cheerful topic, but the zombie servers almost always when you discover the cases of them, they are hilarious. I mean, they're awful, but they're hilarious And you know, it's just this sort of stuff of, "how did this happen?

How did we allow this to happen?" Sometimes it's so easy to do better. And the examples of doing bad are just something that we can all relate to. And, but on the same time, you know, you sort of think, oh, that shouldn't have happened. How did that happen?

Anne Currie: But there's another thing I really like about zombie servers, and I think you've pointed out this yourself, and I plagiarized from your ideas like crazy in Building Green Software, which is one of the reasons why I got you to be a reviewer, so you could complain about it if you wanted to early on. The,

Holly Cummins: It also means I would agree with you a lot. Yes. Oh This is very, sensible. Very sensible. Yes.

Anne Currie: One of the things that we, that constantly comes up when I'm talking to people about this and when we're writing the book and when we're going out to conferences, is people need a way in. And it's often that, you know, that people think the way into building green software is to rewrite everything in C and then they go, "well, I can't do that.

So that's the end. That's the only way in. And I'm not going to be able to do it. So I can't do anything at all." Operations and zombie servers is a really good way in, because you can just do it, you can, instead of having a hackathon, you can just do a thrift a thon, get everybody to have a little bot running that doesn't need to be running, instantly halve your, you know, it's not uncommon for people to find ways to halve their life.

Yeah. carbon emissions and halve their hosting costs simultaneously in quite a short period of time and it'd be the first thing they do. So I quite like it because it's the first thing they do. What do you think about that? It's, is it the low hanging fruit?

Holly Cummins: Yeah, absolutely, I think, yeah, it's the low hanging fruit, it's easy, it's, kind of entertaining because when you find the problems you can laugh at yourself, and there's, again, there's no downside and several upsides, you know, so it's, you know, it's this double win of I got rid of something I wasn't even using, I have more space in my closet, and I don't have to pay for it.

Anne Currie: Yeah, I just read a book that I really should have read years and years ago, and I don't know why I didn't, because people have been telling me to read it for years, which was the goal. Which is, it's not about tech, but it is about tech. It's kind of the book that was the precursor to the Phoenix Projects, which I think a lot read.

And it was, it's all about TPS, the Toyota Production System. In a kind of like an Americanized version of it, how are the tires production system should be brought to America. And it was written in the 80s and it's all about work in progress and cleaning your environment and getting rid of stuff that gets in your way and just obscures everything.

, you can't see what's going on. Effectively, it was a precursor to lean, which I think is really very well aligned. Green and lean, really well aligned. And, it's something that we don't think about, that cleaning up waste just makes your life much better in ways that are hard to imagine until you've done it.

And zombie, cleaning zombie servers up just makes your systems more secure, cheaper, more resilient, more everything. It's a really good thing to do.

Holly Cummins: Yeah. And there's sort of another way that those align as well, which I think is interesting because I think it's not necessarily intuitive. Which is, sometimes when we talk about zombie servers and server waste, people's first response is, this is terrible. The way I'm going to solve it is I'm going to put in barriers in place so that getting a server is harder.

And that seems really intuitive, right? Because it's like, Oh yes, we need to solve it. But of course, but it has the exact opposite effect. And again it seems so counterintuitive because it seems like if you have a choice between shutting the barn door before the horses left and shutting the barn door after the horses left, you should shut the barn door before the horses left.

But what happens is that if those barriers are in place, once people have a server, if they had to sweat blood to get that server, they are never giving it up. It doesn't matter how many thriftathons you do, they are going to cling to that server because it was so painful to get. So what you need to do is you need to just create these really sort of low friction systems where it's easy come, easy go.

So it's really easy to get the hardware you need. And so you're really willing to give it up and that kind of self service model, that kind of low friction, high automation model is really well aligned again with lean. It's really well aligned with DevOps. It's really well aligned with cloud native.

And so it has a whole bunch of benefits for us as users as well. If it's easier for me to get a server, that means I'm more likely to surrender it, but it also means I didn't have to suffer to get it, which is just a win for me personally.

Anne Currie: It is. And there's something at the end of the goal in the little bit at the end, which I thought was my goodness, the most amazing, a bit of a lightswitch moment for me, when it was talking to this still about 10 years ago, but it was, it's talking about, ideas about stuff that, basically underpin the cloud, underpin modern computing, underpin factories and also warehouses and because I worked for a long time in companies that had warehouses, so you kind of see that there are enormous analogies and it was talking about how a lot of the good modern practice in this has been known since the 50s.

And, it, even in places like japan, where it's really well known, I mean, Toyota is so, the Toyota production system is so well managed, almost everybody knows it, and everybody wants to, every company in Japan wants to be operating in that way. Still, the penetration of companies that actually achieve it is very low, it's only like 20%.

I thought, it's interesting, why is that? And then I realised that you'd been kind of hinting why it was throughout. And if you look on the Toyota website, they're quite clear about it. They say the Toyota production system is all about trial and error. Doesn't matter, you can't read a book that tells you what we did, and then say, "oh well if I do that, then I will achieve the result."

They say it's all about a culture of trial and error. And then you achieve, then you build something which will be influenced by what we do, and influenced by what other people do, and influenced by a lot of these ideas. But fundamentally, it has to be unique to you because anything complicated is context-specific.

Therefore, you are going to have to learn from it. But one of the, one of the key things for trial and error is not making it so hard to try something and so painful if you make an error that you never do any trial and error. And I think that's very aligned with what you were saying about if you make it too hard, then nobody does any trial and error.

Holly Cummins: Yeah. Absolutely.

Anne Currie: I wrote a new version of it, called the cloud native attitude, which was all about, you know, what are people doing? You know, what's the UK enterprise version of the TPS system, and what are the fundamentals and what are people actually doing?

And what I realized was that everybody was doing things that were quite different, that was specific to them, that used some of the same building blocks and were quite often in the cloud because that reduced their bottlenecks over getting hardware. Because that's always, that's a common bottleneck for everybody.

So they wanted to reduce the bottleneck there of getting the access to hardware. But what they were actually doing was built trial and error wise, depending on their own specific context. And every company is different and has a different context. And, yeah, so you have to be able to, that is why failure is so, can't be a four letter word.

Holly Cummins: Yeah. Technically, it's a seven letter word if you say failure, but...

Anne Currie: And it should be treated that way.

Yeah.

I'm very aware that actually our brief for this was to talk about three articles on AI.

Holly Cummins: I have to say, I did have a bit of a panic when I was reviewing the articles because they were very deep into the sort of the intricacies of, you know, AI policy and AI governance, which is not my specialty area.

Anne Currie: No, neither is it mine. All that and when I was reading it, I thought quite a lot about what we've just talked about. It is a new area. It's something that, as far as AI is concerned, I love AI. I have no problem with AI. I think it's fantastic. It's amazing what it can produce.

And if you are not playing around on the free version of ChatGPT, then you are not keeping on top of things because it changes all the time. And it's, very like managing somebody. You get out of it what you put in. If you put in, if you make a very cursory, ask it a couple of cursory questions, you'll get a couple of cursory answers.

If you, you know, leaning back on Toyota again, you almost need to five wise it. You need to No, go, no, but why? Go a little bit deeper. Now go a little bit deeper. Now go a little bit deeper. And then you'll notice that the answers get better and better, like a person, better and better.

So if you, really do, it is worth playing around with it.

Holly Cummins: Just on that, I was just reading an article from Simon Willison this morning and he, was talking about sort of, you know, a similar idea that, you know, you have to put a lot into it and that to get good, he was talking about it for coding assistance that, you know, to get good outputs, it's not trivial.

And a lot of people will sort of try it and then be disappointed by their first result and go, "Oh, well, it's terrible" and dismiss it. But he was saying that one of the mistakes that people make is to anthropomorphize it. And so when they see it making mistakes that a human would never make, they go, "well, this is terrible" and they don't think about it in terms of, well, this has some weaknesses and this has some strengths and they're not the same weaknesses and strengths as a person would have.

And so I can't just see this one thing that a human would never do and then dismiss it. I, you know, you need to sort of adapt how you use it for its strengths and weaknesses, which I thought was really interesting. The sort of the, you know, it's so tempting to anthropomorphize it because it is so human ish in its outputs because it's trained on human inputs, but it is not, it does not have the same strengths and weaknesses as a person.

Anne Currie: Well, I would say the thing is, it can be used in lots of different ways. There are ways you can use it which, actually, it can react like a person, and therefore does need to be called. I mean, if you ask it to do creative things, it's quite human like. And it will come up with, and it will blag, and it will, you know, it's, you just have to treat it to certainly, certain creative things.

You have to go, "is that true?" Can you double check that? Is that, I appreciate your enthusiasm there, but it might not be right. Can you just double check that? In the same way that you would do for, with a very enthusiastic graduate. And you wouldn't have fired them because they said something that seemed plausible

and, well, unless you'd said, do not tell me anything that seems plausible, then you don't double check. Because to a certain extent, they're always enthused. And that's where ideas come from. Stretching what's saying, well, you know, I don't know if this is happening, but this could happen. You have to be a little bit out there to generate new ideas and have new thoughts. I heard a very interesting podcast yesterday where one of the Reeds, I can never remember if it was Reed Hastings or Reed Hoffman, you know, it's like it was talking about AI, it was AI energy use.

And he was saying, we're not stupid, you know, if there's, basically, there are two things that we know are coming. One is AI and one is climate change. We're not going to build, to try and create an AI industry that's requires the fossil fuel industry because that would be crazy talk, you know, we do all need to remember that climate change is coming and it is a different model for how, and, you know, if you are building an AI system that relies on fossil fuels, then you are an idiot because, the big players are not. You know, it's, I love looking at our world in data and looking at what is growing in the world?

And if you look to a chart that's really interesting to look at, if you ever feel depressed about climate change is to look at the global growth in solar power in solar generated power. It's going up like it's not even exponential. It's, you know, it's, it looks vertically asymptotic.

You know, it's super exponential. It's going faster than exponential, nothing else is developing that way. Except maybe AI, but AI from a from a lower point and, actually I think the AI will, and then you've got things with AI, you've got stuff like DeepSeek that's coming out of field and saying, "do you know?

You just didn't need to write this so inefficiently. You could, you know, you could do this on a lot less, and it'd be a lot cheaper, and you could do things on the edge that you didn't know that you could do." So, yeah, I'm not too worried about AI. I think that DeepSeek surprised me.

Holly Cummins: Yeah, I agree.

I think we have been seeing this, you know, sort of enormous rise in energy consumption, but that's not sustainable, and it's not sustainable in terms of climate, but it's also not sustainable financially. And so financial corrections tend to come before the climate corrections.

And so what we're seeing now is architectures that are designed to reduce the energy costs because they need to reduce the actual financial costs. So we get things like DeepSeek where there's the sort of fundamental efficiency in the model of the architecture or the architecture of the model rather.

But then we're also seeing things as well, like you know, up until maybe a year ago, the way it worked was that the bigger the model, the better the results. Just, you know, absolutely. And now we're starting to see things where the model gets bigger. And the results get worse and you see this with RAG systems as well, where when you do your RAG experiment and you feed in just two pages of data, it works fantastically well and then you go, "okay, I'm going to proceed."

And then you feed in like 2000 pages of data and your RAG suddenly isn't really working and it's not really giving you correct responses anymore. And so I think we're seeing an architectural shift away from the really big monolithic models to more orchestrated models. Which is kind of bad in a way, right?

Because it means we as engineers have to do more work. We can't just like have one big monolith and say, "solve everything." But on the other hand, what do engineers love? We love engineering. So it means that there's opportunities for us. So, you know, a pattern that we're seeing a lot now is that you have your sort of orchestrator model that takes the query in and triages it.

And it says, "is this something that should go out to the web? Because, actually, like, that's the best place for this news topic. Or is this something that should go to my RAG model? Is this something..."

You know, and so it'll choose the right model. Those models are smaller, and so they have a much more limited scope.

But, within that scope, they can give you much higher quality answers than the huge supermodel, and they cost much less to run. So you end up with a system, again, it's about the double win, where you have a system which maybe took a little bit more work to architect, but gives you better answers for a lower cost.

Anne Currie: That is really interesting and more aligned as well with how power is being developed potentially, you know, that there is, that you really want to be doing more stuff at the edge, which that you want, and you want people to be doing stuff at home on their own devices, you know, rather than just always having to go to, as you say, Supermodels are bad.

We all disapprove of supermodels.

Holly Cummins: Yeah. and in terms of, you know, that aligns with some of the sort of the, you know, the privacy concerns as well, which is, you know, people want to be doing it at home and certainly organizations want to be keeping their data in house. And so then that means that they need the more organization local model to be keeping their, dirty secrets in house.

Anne Currie: Well, it is true. I mean, the thing is you, it is very hard to keep things secure and sometimes just do want to keep things in house, some of your data in house, you don't necessarily even want to stick it on Amazon if you can avoid it. But yes, so that's been a really interesting discussion and we have completely gone off topic and we've hardly talked at all about, the AI regulation.

I think we both agree that AI regulation, it's quite soon to be doing it. It's interesting. I can see why, the Americans have a tendency to take a completely different approach to the EU. If you look at their laws and I have to, I did do some lecturing in AI ethics and legalities and American laws do tend to be like, well, something goes wrong, you know, get your pantsuit off and fix it. EU laws tend to be about, don't even, don't do it. You know, as you said before, close the door before the horse has, you know, has bolted. And the American law is about bringing it back.

But in some ways, that is, that exemplifies why America grows much faster than Europe does. ,

Holly Cummins: Yeah.

I was, when I was looking at some of the announcements that did come out of the AI summit, I think, yeah, I have really mixed feelings about it because I think I generally feel that regulation is good, but I also agree with you that it can have a stifling effect on growth, but one thing that I think is fairly clearly positive that did seem to be emphasized in the announcements as well is the open source aspect.

So, like, we're, I mean, we have, you know, sort of open source models now, but they're not as open source as, you know, open source software in terms of how reproducible they are, how accessible they are for people to see the innards of, but I think I was thinking a little bit again when I was sort of the way the AI summit is

is making these sort of bodies that have like the public private partnerships, which isn't anything new, but you know, we're sort of seeing quite a few governments coming together. So like the current AI announcement, I think had nine governments and dozens of companies, but it reminded me a little bit of the sort of the birth of radio. When we had this resource which was the airwaves, the frequencies that, you know, had, nobody had cared about. And then now all of a sudden it was quite valuable and there was potentially, you know, the sort of wild west of like, okay, who can take this and exploit it commercially? And then government stepped in and said, "actually, no, this is a resource that belongs to all of us.

And so it needs to be managed." Who has access to it and who can just grab it. And I feel a bit like, even though in a technical sense, the data all around us isn't all of ours. It's, you know, a lot of it is copyrighted and that kind of thing. But if you look at the sort of aggregate of like

all of the data that humanity has produced, that is a collective asset.

And so it should be that how it gets used is for a collective benefit and that regulation, and making sure that it's not just one or two organizations that have the technical potential to leverage that data is a collectively good

thing.

Anne Currie: Especially at the moment, we don't want everything to be happening in the US, because, maybe the US is not the friendly partner that we would always thought it would be, it's, diversity

Holly Cummins: diversity is good. Diversity of geographic interests.

Anne Currie: Indeed. Yeah, it is. So yeah, it's, but it is early days. I'm not an anti AI person by any stretch. In fact, I love AI. I think it's really is an amazing thing. And we just need to align it with the interests of the rest of the humanity in terms

Holly Cummins: Yes.

Anne Currie: but it is interesting. They're saying that in terms of being green, the big players are not idiots. They know that things need to be aligned. But in terms of data, they certainly will be acting in their best interests. So, yeah, I can see they, yeah, indeed. Very interesting. So, we are now coming to time, we've done quite a lot, we've done quite a lot. There won't be much to edit out from what we've talked about today.

I think it's great, it's very good. But,

Holly Cummins: Shall we talk about the Microsoft article though? Cause that, I thought that was really interesting.

Anne Currie: oh yeah, go for it, Yes,

Holly Cummins: Yeah, so one of the other articles that we have is, It said that Microsoft had, was reducing its investment in data centers, which was, I was quite shocked to read that because it's the exact opposite of all of the news articles that we normally see, including one I saw this morning that said that, you know, the big three are looking at increasing their investment in nuclear.

But I thought it was sort of interesting because we've, I think we always tend to sort of extrapolate from the current state and extrapolate it indefinitely forward. So we say demand for AI is growing, demand for AI will grow indefinitely, but of course, that's not sustainable. Again you know, it's not sustainable in terms of financially and so at some point there will be that correction and it seems like, Microsoft has perhaps looked at how much they've invested in data centers and said "oh, perhaps this was a little bit much, perhaps let's rollback that investment just a little bit, because now we have an over capacity on data centers."

Anne Currie: Well, I mean, I wonder how much of DeepSeek had an effect on which is that everybody was looking at it and going, the thing is, I mean, Azure is, it's, not, well, I say this is a public story. So I could, because I have it in the book, the story of during the pandemic, the team, the Microsoft Teams folks looking at what they were doing and saying, "could this be more efficient?" And the answer was yes, because had really no effort in whatsoever to make what they were doing efficient. Really basic efficiency stuff they hadn't done. And so there was tons of waste in that system. And the thing is, when you gallop ahead to do things, you do end up with a lot of waste.

DeepSeek was a great example of, you know this AI thing, we can do it on like much cheaper chips and much fewer machines. And you don't have to do it that way. So I'm hoping that this means that Microsoft have decided to start investing in efficiency. It's a shame because they used to have an amazing team who were fantastic at this kind of stuff, who used it, so we, I was saying, Holly spoke at a conference I did last year about code efficiency. And Quarkus being a really good example of a more efficient platform for running Java on. The first person I had on that used to work for Azure. And he used to, was probably the world's expert in actual practical code efficiency. He got made redundant. Yeah. Because, Microsoft at the time were not interested in efficiency. So "who cares? Pfft, go on, out." But he's now working at NVIDIA doing all the efficiency stuff there. Because some people are not, who paying attention to, I, well I think the lesson there is that maybe Microsoft were not paying that much attention to efficiency, the idea that actually you don't need 10 data centers. A little bit of easy, well, very difficult change to make it really efficient. But quite often there's a lot of low hanging fruit in efficiency.

Holly Cummins: Absolutely. And you need to remember to do it as well, because I think that, I think probably it is a reasonable and correct flow to say, innovate first, optimize second. So, you know, you, don't have be looking at that efficiency as you're innovating because that stifles the efficiency and you know, you might be optimizing something that never becomes anything, but you have to then remember once you've got it out there to go back and say, "Oh, look at all of these low hanging fruit. Look how much waste there is here. Let's, sort it out now that we've proven it's a success."

Anne Currie: Yeah. Yeah, it is. Yes. It's like "don't prematurely optimize does" not mean "never optimize."

Holly Cummins: Yes. Yes.

Anne Currie: So, I, my strong suspicion is that Microsoft are kind of waking up to that a little bit. The thing is, if you have limitless money, and you just throw a whole load of money at things, then, it is hard to go and optimize. As you say, it's a bit like that whole thing of going in and turning off those zombie machines.

You know, you have to go and do it know, it's, you have to choose to do it. If you have limitless money, you never do it, because it's a bit boring, it's not as exciting as a new thing. Yeah, but yeah, limitless money has its downsides as well as up.

Holly Cummins: Yes. Who knew?

Anne Currie: Yeah, but so I think we are at the end of our time. Is there anything else you want to say before you, it was an excellent hour.

Holly Cummins: Nope. Nope. This has been absolutely fantastic chatting to you Anne.

Anne Currie: Excellent. It's been very good talking to you as always. And so my final thing is, if anybody who's listening to this podcast has not read building green software from O'Reilly, you absolutely should, because a lot of what we just talked about was covered in the book. Reviewed by Holly.

Holly Cummins: I can recommend the book.

Anne Currie: I think your name is somewhere as a, some nice thing you said about it somewhere on the book cover, but, so thank you very much indeed. And just a reminder to everybody, everything we've talked about all the links in the show notes at the bottom of the episode. And, we will see, I will see you again soon on the Environment Variables podcast.

Goodbye.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

AI Energy Measurement for Beginners

6 mars 2025 | 57 min

Host Chris Adams is joined by Charles Tripp and Dawn Nafus to explore the complexities of measuring AI's environmental impact from a novice’s starting point. They discuss their research paper, A Beginner's Guide to Power and Energy Measurement and Estimation for Computing and Machine Learning, breaking down key insights on how energy efficiency in AI systems is often misunderstood. They discuss practical strategies for optimizing energy use, the challenges of accurate measurement, and the broader implications of AI’s energy demands. They also highlight initiatives like Hugging Face’s Energy Score Alliance, discuss how transparency and better metrics can drive more sustainable AI development and how they both have a commonality with eagle(s)!

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Dawn Nafus: LinkedIn
Charles Tripp: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

NREL Partner Forum Agenda | 12-13 May 2025

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Charles Tripp: But now it's starting to be like, well, we can't build that data center because we can't get the energy to it that we need to do the things we want to do with it. we haven't taken that incremental cost into account over time, we just kind of ignored it. And now we hit like the barrier, right?

I'm your host, Chris Adams.

Welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host Chris Adams. If you follow a strict media diet, you switch off the Wi-Fi on your house and you throw your phone into the ocean, you might be able to avoid the constant stream of stories about AI in the tech industry. For the rest of us, though, it's basically unavoidable. So having an understanding of the environmental impact of AI is increasingly important if you want to be a responsible practitioner navigating the world of AI, generative AI, machine learning models, DeepSeek, and the rest. Earlier this year, I had a paper shared with me with the intriguing title A Beginner's Guide to Power and Energy Measurement, an Estimation for Computing and Machine Learning. And it turned out to be one of the most useful resources I've since come across for making sense of the environmental footprint of AI. So I was over the moon when I found out that two of the authors were both willing and able to come on to discuss this subject today. So joining me today are Dawn Nafus and Charles Tripp, who worked on the paper and did all this research. And well, instead of me introducing them, well, they're right here. I might as well let them do the honors themselves, actually. So, I'm just going to work in alphabetical order. Charles, I think you're slightly ahead of Dawn. So, if I, can I just give you the room to, like, introduce yourself?

Charles Tripp: Sure. I'm a machine learning researcher and

Stanford

algorithms researcher, and I've been programming pretty much my whole life since I was a little kid, and I love computers. I researched machine learning and reinforcement learning in particular at Stanford, started my own company, but kind of got burnt out on it.

And then I went to the National Renewable Energy Lab where I applied machine learning techniques to energy efficiency and renewable energy problems there. And while I was there, I started to realize that computing energy efficiency was a risingly, like, an increasingly important area of study on its own.

So I had the opportunity to sort of lead an effort there to create a program of research around that topic. And it was through that work that I started working on this paper, made these connections with Dawn. And I worked there for six years and just recently changed jobs to be a machine learning engineer at Zazzle.

I'm continuing to do this research. And, yeah.

Chris Adams: Brilliant. Thank you, Charles. Okay, so national, that's NREL that some people refer

Charles Tripp: That's right. It's one of the national labs.

Chris Adams: Okay. Brillinat. And Dawn, I guess I should give you the space to introduce yourself, and welcome back again, actually.

Dawn Nafus: Thank you. Great to be here. My name is Dawn Nafus. I'm a principal engineer now in Intel Labs. I also run the Socio Technical Systems Lab. And I also sit on Intel's Responsible AI Advisory Council, where we look after what kinds of machine learning tools and products do we want to put out the door.

Chris Adams: Brilliant, thank you, Dawn. And if you're new to this podcast, I mentioned my name was Chris Adams at the beginning of the podcast. I work at the Green Web Foundation. I'm the director of technology and policy there. I'm one of the authors of a report all about the environmental impact of AI last year, so I have like some background on this. I also work as the policy chair in the Green Software Foundation Policy Working Group as well. So that's another thing that I do. And if you, if there, we'll do our best to make sure that we link to every single paper and project on this, so if there are any particular things you find interesting, please do follow, look for the show notes. Okay, Dawn, I'm, let's, shall we start? I think you're both sitting comfortably, right? Shall I begin?

Okay, good. So, Dawn, I'm really glad you actually had a chance to both work on this paper and share and let me know about it in the first place. And I can tell when I read through it, there was quite an effort to, like, do all the research for this.

So, can I ask, like, what was the motivation for doing this in the first place? And, like, was there any particular people you feel really should read it?

Dawn Nafus: Yeah, absolutely. We primarily wrote this for ourselves. In a way. And I'll explain what I mean by that. So, oddly, it actually started life in my role in Responsible AI, where I had recently advocated that Intel should adopt a Protect the Environment principle alongside our suite of other Responsible AI principles, right?

Bias and inclusion, transparency, human oversight, all the rest of it. And so, the first thing that comes up when you advocate for a principle, and they did actually implement it, is "what are you going to do about it?" And so, we had a lot of conversation about exactly that, and really started to hone in on energy transparency, in part because, you know, from a governance perspective, that's an easy thing to at least conceptualize, right? You can get a number.

Chris Adams: Mmm.

Dawn Nafus: You know, it's the place where people's heads first go to. And of course it's the biggest part of, or a very large part of the problem in the first place. Something that you can actually control at a development level. And so, but once we started poking at it, it was, "what do we actually mean by measuring? And for what? And for whom?" So as an example, if we measured, say, the last training run, that'll give you a nice guesstimate for your next training run, but that's not a carbon footprint, right? A footprint is everything that you've done before that, which folks might not have kept track of, right?

So, you know, we're really starting to wrestle with this. And then in parallel, in labs, we were doing some socio technical work on, carbon awareness. And there too, we had to start with measuring. Right? You had to start somewhere. And so that's exactly what the team did. And they found interestingly, or painfully depending on your point of view, look, this stuff ain't so simple, right?

If what you're doing is running a giant training run, you stick CodeCarbon in or whatever it is, sure, you can get absolutely a reasonable number. If you're trying to do something a little bit more granular, a little bit trickier, it turns out you actually have to know what you're looking at inside a data center, and frankly, we didn't, as machine learning people primarily. And so, we hit a lot of barriers and what we wanted to do was to say, okay, there are plenty of other people who are going to find the same stuff we did, so, and they shouldn't have to find out the hard way. So that was the motivation.

Chris Adams: Well, I'm glad that you did because this was actually the thing that we found as well, when we were looking into this, it looks simple on the outside, and then it turned, it feels a bit like a kind of fractal of complexity, and there's various layers that you need to be thinking about. And this is one thing I really appreciated in the paper that we actually, that, that was kind of broken out like that.

So you can at least have a model to think about it. And Charles, maybe this is actually one thing I can, like, hand over to you because I spoke about this kind of hierarchy of things you might do, like there's

stuff you might do at a data facility level or right all the way down to a, like, a node level, for example.

Can you take me through some of the ideas there? Because I know for people who haven't read the paper yet, that seemed to be one of the key ideas behind this, that there are different places where you might make an intervention. And this is actually a key thing to take away if you're trying to kind of interrogate this for the first time.

Charles Tripp: Yeah, I think it's, both interventions and measurement, or I should, it's really more estimation at any level. And it also depends on your goals and perspective. So it, like, if you are operating a data center, right? You're probably concerned with the entire data center, right? Like the cooling systems, the idle power draw, the, converting power to different levels, right?

Like transformer efficiency, things like that. Maybe even the transmission line losses and all of these things. And you may not really care too much about, like, the code level, right? So the types of measurements you might take there or estimates you might make are going to be different. They're gonna be at, like, the system level.

Like, how much is my cooling system using in different conditions, different operating conditions, environmental conditions? From a user's perspective, you might care a lot more about, like, how much energy, how much carbon is this job using? And that's gonna depend on those data center variables. But there's also a degree of like, well, the data center is going to be running whether or not I run my job.

Right? So I really care about my jobs impact more. And then I might be caring about much shorter term, more local estimates, like ones that, might be from measuring the nodes that I'm running on's power or which was what we did it at NREL or, much higher frequency, but less accurate measurements that come from the hardware itself.

Most modern computing hardware has a way to get these hardware estimates of the current power consumption. And you could log those. And there's also difficulties. Once you start doing that is the measurement itself can cause energy consumption. Right? And also potentially interfere with your software and cause it to run more slowly and potentially use more energy.

And so, like, there's difficulties there at that level. Yeah, but there's a whole suite of tools that are appropriate for different uses and purposes, right? Like measuring the power at the wall, going into the data center may be useful at the data center or multiple data center level. Still doesn't tell you all the story, right?

Like the losses in the transmission lines and where did that power come from are still not accounted for, right? But it also doesn't give you a sense for, like, what happens that I take interventions at the user level? It's very hard to see that from that high level, right? Because there's many things running on the system, different conditions there. From the user's point of view, they might only care about, like, you know, this one key piece of my software that's running, you know, like the kernel of this deep learning network.

How much energy is that taking? How much additional energy is that taking? And that's like a very different thing that very different measurements are appropriate for and interventions, right?

Like changing that little, you know, optimizing a little piece of code versus like, maybe we need to change the way our cooling system works on the whole data center or the way that we schedule jobs. Yeah, and the paper goes through many of these levels of granularity.

Chris Adams: Yeah, so this is one thing that really kind of struck out at me because when you, it started at the kind of facility level, which is looking at an entire building where you mentioned things like say, you know, power coming into the entire facility. And then I believe you went down to looking at say the, within that facility, there might be one or more data centers, then you're going down to things like a rack level and then you're going down to

kind of at a node level and then you're all even going all the way down to like a particularly tight loop or the equivalent for that. And when you're looking at things like this, there are questions about like what you what...

if you would make something particularly efficient at, say, the bottom level, the node level, that doesn't necessarily impact, it might not have an impact higher up, for example, because that capacity might be just reallocated to someone else.

For example, it might just be that there's a certain kind of minimum amount of power draw that you aren't able to have much of an impact on. I mean, like, this is, these are some of the things

I was surprised by, or not surprised by, but I really appreciated breaking some of that, these out, because one thing that seemed to, one thing that was, I guess, counterintuitive when I was looking at this was that things you might do at one level can actually be counter, can hinder steps further down, for example, and vice versa.

Charles Tripp: Yeah, that's right. I mean, I think, two important sort of findings are, yeah, like battle scars that we got from doing these measurements. And one data set we produced is called BUTTER-E, which is like a really large scale measurement of energy consumption of training and testing neural networks and how the architecture impacts it.

And we were trying to get reasonable measurements while doing this. And, of the difficulties is in comparing measurements between runs on different systems, even if they're identically configured, can be tricky because different systems based on, you know, manufacturing variances, the heat, you know, like how warm is that system at that time?

Anything that might be happening in the background or over the network, anything that might be just a little different about its environment can have, real measurable impacts on the energy consumed. So, like comparing energy consumption between runs on different nodes, even with identical configurations, we had to account for biases and they're like, oh, this node draws a little bit more power than this one at idle.

And we have to like, adjust for that in order to make a clear comparison of what the difference was. And this problem gets bigger when you have different system configurations or even same configuration, but running in like a totally different data center. So that was like one tricky finding. And I think two other little ones I can mention, maybe we could go into more detail later. But,

another one, like you mentioned, is the overall system utilization and how that's impacted by a particular piece of software running a particular job running is going to vary a lot on what those other users of the system are doing and how that system is scheduled.

So, you can definitely get in the situations where, yeah, I reduced my energy consumption, but that total system is just going to, that energy is going to be used some other time, especially if the energy consumption savings I get are from shortening the amount of time I'm using a resource and then someone else.

But it does mean that the computing is being done more efficiently, right? Like, if everyone does that, then more computing can be done within the same amount of energy. But it's hard to quantify that. Like, what is my impact? It's hard to say, right?

Chris Adams: I see, yeah, and Dawn, go on, I can, see you nodding, so I want you to come in now.

Dawn Nafus: If I can jump in a bit, I mean, I think that speaks to one of the things we're trying to bring out, maybe not literally, but make possible, is this. Those things could actually be better aligned in a certain way, right? Like, the energy that is, you know, for example, when there is idle time, right?

I mean, there are things that data center operators can do to reduce that, right? you know, you can bring things into lower power states, all the rest of it, right? So, in a way, kind of, but at the same time, the developer can't control it, but if they don't actually know that's going on, and it's just like, well, it's there anyway, there's nothing for me to do, right, that's also a problem, right?

So in a way, you've got two different kinds of actors looking at it in very different perspectives. And the clearer we can get about roles and responsibilities, right, you can start to do things like reduce your power when things are idling. Yes, you do have that problem of somebody else is going to jump in. But Charles, I think as your work shows, you know, there's still some idling going on, even though you wouldn't think, so maybe you could talk a little bit about that.

Charles Tripp: Yeah, so one really interesting thing that I didn't expect going into doing these measurements in this type of analysis was, well, first, I thought, "oh great, we can just measure the power on each node, run things and compare them." And we ran into problems immediately. Like, you couldn't compare the energy consumption from two identically configured systems directly, especially if you're collecting a lot of data, because one is just going to use like slightly more than the other because of the different variables I mentioned.

And then when you compare them, you're like, well, that run used way more energy, but it's not because of anything about how the job was configured. It's just, that system used a little bit more. So if I switch them, I'd get the opposite result. So that was one thing. But then, as we got into it and we were trying to figure out, okay, well, now that we figured out a way to account for these variations, let's see what the impact is of running different software with different configurations, especially like neural networks, different configurations on energy consumption and our initial hypothesis was that it was based on mainly the size of the neural network and, you know, like how many parameters basically, like how many calculations, these sorts of things.

And if you look in the research, A lot of the research out there about making neural networks and largely algorithms in general more efficient focuses on how many operations, how many flops does this take, you know? And look, we reduced it by a huge amount. So that means that we get the same energy consumption reductions.

We kind of thought that was probably true for the most part. But as we took measurements, we found that had almost no connection to how much energy was consumed. And the reason was that the amount of energy consumed had way more to do with how much data was moved around on the computer. So how much data was loaded from the network?

How much data was loaded from disc? How much data was loaded from disc into memory, into GPU RAM for using the GPU, into the different caching levels and red, even the registers? So if we computed like how much data got moved in and out of like level two cache on the CPU, we could see that had a huge correlation, like almost direct correlation with energy consumption. Not the number of calculations.

Now, you could get in a situation where, like, basically no data is leaving cache, and I'm doing a ton of computing on that data. In that case, probably a number of calculations does matter, but in most cases, especially in deep learning, has almost no connections, the amount of data moved. So then we thought, okay, well, it's amount of data moved.

It's the data moving. The data has a certain cost. But then we look deeper, and we saw that actually. The amount of data moved is not really what's causing the energy to be consumed. It's the stalls while the system is waiting to load the data. It's waiting for the data to come from, you know, system memory into level three cache.

It needs to do some calculations on that data. So it's pulling it out while it's sitting there waiting. It's that idle power draw. Just it could be for like a millisecond or even a nanosecond or something, right? But it adds up if you have, you know, billions of accesses. Each of those little stalls is drawing some power, and it adds up to be quite a significant amount of power.

So we found that actually the driver of the energy consumption, the primary driver by far in what we were studying in deep learning was

the idle power draw while waiting for data to move around the system. And this was like really surprising because we started with number of calculations, it turns out almost irrelevant.

Right. And then we're like, well, is it the amount of data moved around? It's actually not quite the amount of data moved around, but that does like cause the stalls whenever I need to access the data, but it's really that idle power draw. And and I think that's probably true for a lot of software.

Chris Adams: Yes. I think that does sound about right.

I'm just gonna try if I follow that, because there was, I think there was a few quite key important ideas there. But there's also, if you aren't familiar with how computers are designed, you it might, there. I'll try to paraphrase it. So we've had this idea that the main thing is like, the number of calculations being done. That's like what we thought was the key idea.

But,

Charles Tripp: How much work, you know.

Chris Adams: Yeah, exactly. And, what we actually, what we know about is inside a computer you have like multiple layers of, let's call them say, caches or multiple layers at where you might store data so it's easy and fast to access, but that starts quite small and then gets larger and larger, which a little bit slower over time.

So you might have, like you said, L2 cache, for example, and that's going to be smaller, much, much faster, but smaller than, say, the RAM on your system, and then if you go a bit further down, you've got like a disk, which is going to be way, what larger, and then that's going to be somewhat slower still, so moving between these stages so that you can process, that was actually one of the things that you were looking at, and then it turned out that actually, the thing that, well, there is some correlation there, one of the key drivers actually is the chips kind of in a ready state, ready to actually waiting for that stuff to come in.

They can't really be asleep because they know the data is going to have to come in, have to process it. They have to be almost like anticipating at all these levels. And that's one of the things we, that's one of the big drivers of actually the resource use and

the energy use.

Charles Tripp: I mean, so, like, what we saw was, we actually estimated how much energy it took, like, per byte to move data from, like, system RAM to level three cache to level two to level one to a register at each level. And at some cases, it was so small, we couldn't even really estimate it. But in most cases, we were able to get an estimate for the For that, but a much larger cost was initiating the transfer, and even bigger than that was just the idle power draw during the time that the program executed and how long it executed for. And by combining those, we were able to estimate that most of that power consumption, like 99 percent in most cases was from that idle time, even those little micro stalls waiting for the data to move around. And that's because moving the data while it does take some energy doesn't take that much in comparison to the amount of energy of like keeping the ram on and the data is just like alive in the ram or keeping the CPU active, right?

Like CPUs can go into lower power states, but generally, at least part of that system has to shut down. So like doing it like at a very, fine grain scale is not really feasible.

Many systems can change power state at a like a faster rate than you might imagine, but it's still a lot slower than like out of, you know, per instruction per byte level of, like, I need to load this data.

Like, okay, shut down the system and wait a second, right? Like, that's, it just, not a second, like a few nanoseconds. It's just not practical to do that. And it's so it's just keeping everything on during that time. That's sucking up most of the power. the So one strategy, simple strategy, but it's difficult to implement in some cases is to initiate that load that transfer earlier.

So if you can prefetch the data into the higher levels of memory before you hit the stall where you're waiting to actually use it,

you can probably significantly reduce this power consumption, due to that idle wait. But it's difficult to figure out how to properly do that prefetching.

Chris Adams: Ah, I see. Thanks, charles. So it sounds like, okay, they, we might kind of approach this and there might be some things which feel kind of intuitive but it turns out there's quite a few counterintuitive things.

And like, Dawn, I can see you nodding away sagely here and I suspect there's a few things that you might have to add on this. Because this is, I mean, can I give you a bit of space, Dawn, to kind of talk about some of this too, because I know that this is something that you've shared with me before, is that yeah, there are maybe some rules of thumb you might use, but it's never that simple, basically, or you realise actually that there's quite a bit more to it than that, for example.

Dawn Nafus: Exactly. Well, I think what I really learned out of this effort is that measurement can actually recalibrate your rules of thumbs, right? So you don't actually have to be measuring all the time for all reasons, but even just that the simple, I mean, not so simple story that Charles told like, okay, you know, so I spent a lot of time talking with developers and trying to understand how they work and at a developer perception level, right?

What do they feel like? What's palpable to them, right? Send the stuff off, go have a cup of coffee, whatever it is, right? So they're not seeing all that, you know, and, you know, when I talk to them, most of them aren't thinking about the kinds of things that were just raised, right? Like how much data are you looking at a time?

You can actually set and tweak that.

And that's the kind of, you know, Folks develop an idea about that, and they don't think too hard about it usually, right. So, with measuring, you can start to actually recalibrate the things you do see, right? I think this also gets back to, you know, why is it counterintuitive that, you know, some of these mechanisms and how you actually are training, as opposed to how many flops you're doing, how many parameters, why is that counterintuitive?

Well, at a certain level, you know, the number of flops do actually matter, right? If we do actually have a gigantic, you know, I'm gonna call myself a foundation model type size stuff, I'm gonna build out an entire data center for it, it does matter. But as you get, you know, down and down and more specific, it's a, different ball game.

And there are these tricks of scale that are sort of throughout this stuff, right? Like the fact that, yes, you can make a credible claim, that foundation model will always be more energy intensive than, you know, something so small you can run on a laptop, right? That's always going to be true, right? No measurement necessary, right? You keep going down and down, and you're like, okay, let's get more specific. You can get to actually where this, where our frustration really started was, you, if you try to go to the extreme, right, try to chase every single electron through a data center, you're not going to do it. It feels like physics, it feels objective, it feels true, but at minimum you start to hit the observer effect, right, that, you know, which is what we did.

We were, my colleague Nicole Beckage was trying to measure at an epoch level, right, sort of essentially round, you know, mini round of training. And what she found was that, you know, she was trying to sample so often that she's pulling energy out of the processing and it just, it messed up the numbers, right? So you can try to get down, you know, into that, you know, what feels like more accuracy and then all of a sudden you're in a different ballpark. So these, tricks of like aggregation and scale and what can you say credibly at what level, I think are fascinating, but you kind of got to get a feel for it in the same way that you can get a feel for, "yep, if I'm sending my job off, I know I have at least, you know, however many hours or however many days," right?

Charles Tripp: There's also so much variation that's out of your control, right? Like one run to another one system to another, even different times where you ran on the same system can cause measureable and in some cases significant variations in the energy consumption.

So it's more about, I think about understanding what's causing the energy consumption.

I think that's the more valuable thing to do. But it's easy to like, be like, "I already understand it." And

I think there's a, there's like a historical bias towards number of operations because in old computers without much caching or anything like this, right? Like I restore old computers and, like an old 386 or IBM XT, right?

Like it's running, it has registers in the CPU and then it has main memory. And it, and almost everything is basically how many operations I'm doing is going to closely correlate with how fast the thing runs and

probably how much energy it uses, because most of the energy consumption on those systems Is just basically constant, no matter what I'm doing, right?

It's just it doesn't like idle down the processor while it's not working, right? And there's a historical bias. It's built up over time that, like, was focused on the, you know, and it's also at the programmer level. Like, I'm thinking about what is the computer doing?

Chris Adams: What do I have controll over?

Charles Tripp: But it's only through it's only through actually measuring it that you gain a clearer picture of, like, what is actually using energy.

And I think if you get that picture, then you'll gain an understanding more of

how can I make this software or the data center or anything in between like job allocation more energy efficient, but it's only through actually measuring that we can get that clear picture. Because if we guess, especially using kind of our biases from how we learn to use computers, how we learn about how computers work, we're actually very likely to get an incorrect understanding, incorrect picture of what's driving the energy consumption.

It's much less intuitive than people think.

Chris Adams: Ah, okay, there's a couple of things I'd like to comment on, and then Dawn, i might give you a bit of space on this, because, you said, so there's one, so we're just talking about like flops as a thing that people, okay, are used to looking at, and are like, it's literally written into the AI Act, like, things above a certain number of flops are considered, you know, foundational models, for example, so, you know, that's a really good example of what this actually might be.

And I guess the other thing that I wanted to kind of like touch on is that, I work in the kind of web land, and like, I mean, the Green Web Foundation is a clue in our organization's name. We've had exactly the same thing, where we've been struggling to understand the impact of, say, moving data around, and whether, how much credence you should give to that versus things happening inside a browser, for example.

It looks like you've got some similar kinds of issues and things to be wrestling, with here. But Dawn, I wanted to give you a bit of space because both of you alluded to this, about this idea of having an understanding of what you can and what you can't control and, how you might have a bias for doing one thing without, and then miss something really much larger elsewhere, for example.

Can I maybe give you a bit of space to talk about this idea of, okay, well, which things do you, should you be focusing on, and also understanding of what's within your sphere of influence? What can you control? What can't you control, for example?

Dawn Nafus: Exactly. I think it's in a sense you've captured the main point, which is, you know, that measurements are most helpful when they are relevant to the thing you can control, right? So as a very simple example, you know, there are plenty of AI developers who have a choice in what data centers they can use.

There are plenty who don't, right? You know, when Charles works or worked at NREL, right. The supercomputer was there. That was it. You're not moving, right?

So, if you can move, you know, that overall data center efficiency number that really matters because you can say, alright, "I'm putting my stuff here and not there." If you can't move, like, there's no need to mess with. It it is what it is, right? At the same, and this gets us into this interesting problem, again, a tension between what you might look at it from a policy perspective versus what a developer might look at. We had a lot of kind of, you know, can I say, come to Jesus?

We had a little moment

where we, is that on a podcast? I think I can. Where there was this question of,

are we giving people a bum steer by focusing at, you know, granular developer level stuff, right? Where it's so much actually is on how you run the data center, right? So you, again, you talk about tricks of scale. On the one hand, you know, the amount of energy that you might be directly saving just by, you know, not using or not using, by the time all of those things move through the grid and you're talking about coming, you know, energy coming off of the transmissions cables, right, in aggregate might not actually be directly that big. It might be, but it might not be. And then you flip that around and you think about what aggregate demand looks like and the fact that so much of AI demand is, you know, that's what's putting pressure on our electricity grid.

Right? Then that's the most effective thing you could do, is actually get these, you know, very specific individual jobs down and down, right? So, again, it's all about what you can control, but there are these, whatever perspective you take is just going to flip your, you know, your understanding of the issue around.

Chris Adams: So this was actually one thing I quite appreciated from the paper. There were a few things saying, and it does touch on this idea, that yeah, you, might be focusing on the thing that you feel that you're able to control, but just because you're able to, like, Make very efficient part of this spot here that doesn't necessarily translate into a saving higher up in the system. Simply because if it's, if you don't, if higher up in the system isn't set to actually take advantage of that, then you might never achieve some of these savings It's a little bit like when you're working in cloud, for example, people tell you do all these things to kind of optimize your cloud savings. But if people are not turning data centers off, at best, you might be slowing the growth of infrastructure rollout in future, and like these are, and these are much, much harder things to kind of claim responsibility for, or say that, "yeah, it was definitely, if it weren't for me doing those things, we wouldn't have had that happen."

This is one of the things that I appreciated the paper just making some allusions to and saying, look, yeah, this is, you know, this is why I mean, to be honest, when I was reading this, I was like, wow, there is, there was obviously some stuff for, beginners, but there's actually quite a lot here, which is quite meaty for people who are thinking of it as a much larger systemic level.

So there's definitely things like experts could take away from this as well. So, I just want to check, are there any particular takeaways the two of you would like to kind of draw people's attention to beyond what we've been discussing so far? Because I quite enjoyed the paper and there's a few kind of nice ideas from this. Charles, if I just give you a bit of space to, kind of, come in.

Charles Tripp: Yeah. I've got, kind of two topics that I think build on what we talked about before, but could be really useful for people to be aware of. So one is, sort of one of the outcomes of our studying of the impact of different architectures, data sets, hyper parameter settings on deep neural network energy consumption was that the most efficient networks, most energy efficient networks, and largely that correlates with most time efficient as well, but not always, the most efficient ones were not the smallest ones, and they were not the biggest ones, right?

The biggest ones were just required so much data movement. They were slow. The smallest ones, they took a lot more iterations, right? It took a lot more for them to learn the same thing. And the most efficient ones were the ones where the working sets, where the amount of data that was moved around, matched the different cache sizes.

So as you made the network bigger, it got more efficient because it learned faster. Then when it got so big that the data in like between layers, the communication between layers, for example, started to spill out of a cache level. Then it became much less energy efficient, because of that data movement stall happening.

So we found that like there is like an optimum point there. And for most algorithms, this is probably true where if the working set is sized appropriately for the memory hierarchy, you gain the most efficiency, right? Because generally, like, as I can use more data at a time, I can get my software to work better, right, more efficiently. But there's a point where it falls out of the cache and that becomes less efficient. Exactly what point is going to depend on the software. But I think focusing on that working set size and how it matches to the hardware is a really key piece for almost anyone looking to optimize software for energy efficiency is to think about that. How much data am I moving around and how does that map to the cache? So that's like a practical thing.

Chris Adams: Can I stop you Because I find that quite interesting, in that a lot of the time as developers we're kind of taught to kind of abstract away from

the underlying hardware, and that seems to be going the other way. That's saying, "no, you do need to be thinking about this.

You can't.

There, you know, there's no magic trick."

Charles Tripp: Right? And so, like, for neural networks, that could mean sizing my layers so that those working sets match the cache hierarchy, which is something that no one even considers. It's not even close in most architectures. Like, no one has even thought about this. The other thing is on your point about data center operations and kind of the different perspectives,

one thing that we started to think about as we were doing some of this work was it might make sense to allocate time or in the case of like commercial data center, commercial cloud operator, even like charge field based on at least partly the energy rather than the time, as to incentivize them to use less energy, right?

Like make things more energy efficient. Those can be correlated, but not always right. And another piece of it that I want to touch on of that same puzzle is, from a lot of data center operators perspective, they want to show their systems fully utilized, right? Like there's demand for the system, so we should build an even bigger system and a better system. When it comes to energy consumption.

That's probably not the best way to go, because that means that those systems are sitting there probably doing inefficient things. Maybe even idling a lot of time, right? Like a user allocated the node, but it's just sitting there doing nothing, right? It may be more useful instead of thinking about, like, how much is the system always being utilized?

But think about how much, how much computation or how many jobs or whatever your, like, utilization metric is, do I get, like, per unit energy, right? And you may think about how much, or per unit carbon, right? And you may also think about, like, how much energy savings can I get by doing things like shutting down nodes when they're unlikely to be utilized and more about like having a dynamic capacity, right?

Like full tilt. I can use I can do how many flops or whatever, right? But I can also scale that down to reduce my idle power draw by, you know, 50 percent in low demand conditions. And if you have that dynamic capacity, you may actually be able to get even more throughput. But it's with less energy because when there's no demand, I'm like shutting,

I'm like scaling down my data center, right? And then when there's demand, I'm scaling it up. But these are things that are requiring cultural changes in data center operations to happen.

Chris Adams: I'm glad you mentioned this thing here because, Dawn, I know that you had some notes about, it sounds like in order for you to do that, you need, you probably need different metrics exposed or different kinds of transparency to what we have right now.

Probably more actually. Dawn, can I give you a bit of space to talk about this? Because this is one thing that you told me about before and it's something that is actually touched on in the paper quite a few times actually.

Dawn Nafus: Yeah, I mean, I think we can notice a real gap in a way between the kinds of things that Charles brings his attention to, and the kinds of things that show up in policy environments, in responsible AI circles, right, where I'm a bit closer, we can be a bit vague, and I think we are at the stage where, at least my read on the situation, is that, you know, there's, regardless of where you sit in the debates, and there are rip roaring debates about what to do about the AI energy situation, but I think transparency is probably the one thing we can get the most consensus on, but then, like, just back to that, what the heck does that mean? And I think we need a little, like a, more beats than are currently given to actually where, what work are those measurements doing?

You know, some of the feedback we've gotten is, you know, "well, can't you just come up with a standard?" Like, what's the right standard? It's like, well, no, actually, if data centers aren't standard, and there are many different ways to build a model, then, yes, you can have a standard as a way of having a conversation across a number of different parties to do a very specific thing, like for example, Charles's example, you know, suggested that if we're charging on a per energy basis, that changes a whole lot. Right? But what you can't do is to say, this is the standard that is the right way to do it, and then that meets the requirement, because that's, you know, what we found is that clearly the world is far more, you know, complicated and specific than that.

So, I, you know, I would really encourage the responsible AI community to start to get very specific very quickly, which I don't yet see happening, but I think it's just on the horizon.

Chris Adams: Okay. Well I'm glad you mentioned about maybe taking this a little bit wider 'cause we've dived quite spent a lot of time talking about this paper, but there's other things happening in the world of AI actually, and I wanna give you folks a bit of space to kind of talk about anything that like, or things that you are, that you would like to kind of direct some attention to or you've seen that really you found particularly interesting.

Charles, can I give you some space first and then give Dawn the same, to like say it to like I know, either shout out or point to some particular things that, if they've found this conversation interesting so far, what they might want to be looking at. More data.

Charles Tripp: Yeah. I mean, I think, both in like computer program, computer science at large and especially in machine learning, we've kind of had an attitude, especially within deep learning within machine learning, an attitude of throwing more compute at the problem, right? And more data. The more data that we put through a model and the bigger, the more complicated the model is, the more capable it can be.

But this brute force approach is one of the main things that's driving this increasing computing energy consumption. Right? And I think that it is high time that we start taking a look at making the algorithms we use more energy efficient instead of just throwing more compute. It's easy to throw more compute at it, which is why it's been done.

And also because there hasn't been a significant like material incremental cost of like, Oh, you know, now we need. Twice made GPUs. I don't big deal. But now we're starting to hit constraints because we haven't thought about that incremental energy costs. We haven't had to, as an industry at large, right?

Like, but

now it's starting to be like, well, we can't build that data center because we can't get the energy to it that we need to do the things we want to do with it because we haven't taken that incremental cost into account over time, we just kind of ignored it. And now we hit like the barrier, right?

And so I think thinking about, the energy costs and probably this means investing in more finding more efficient algorithms, more efficient approaches as well as more efficient ways to run data centers and run jobs. That's gonna become increasingly important, even as our compute capacity continues to increase.

The energy costs are likely to increase along with that as we use more and more, and we need create more generation capacity, right? Like, it's expensive at some point where we're really driving that energy production, and that's going to be increasingly an important cost as well as it is now, like, starting to be a constraint to what kind of computing we can do.

So I think investing in more efficient approaches is going to be really key in the future.

Chris Adams: There's one thing that I, that I think Dawn might come in on this actually, is that, you're talking about, it seems that you're talking about having more of a focus on surfacing some of the kind of efficiency or the fact that resource efficiency is actually going to be something that we probably need to value or sharpen, I mean, because as I understand it so far, it's not particularly visible in benchmarks or anything like that right now, like, and if you have benchmarks deciding, what counts as a good model or a good use of this until that's included. You're not going to have anything like this. Is that the kind of stuff you're kind of suggesting we should probably have? Like, some more recognition of, like, or even like, you're taking at the energy efficiency of something and being that thing that you draw attention to or you include in counting something as good or not, essentially.

Dawn Nafus: You know, I have a particular view of efficiency. I suspect many of your listeners might, as well. You know, I think it's notable that at the moment when we're seeing the, you know, the the model of the month, apparently, or the set of models of DeepSeek has come onto the scene and immediately we're starting to see, for the first time, you know, a Jevons paradox showing up in the public discourse.

So this is the paradox that when you make things more efficient, you can also end up stimulating so much demand...

Chris Adams: Absolute use grows even though it gets individually more efficient.

Dawn Nafus: Yeah, exactly. Again, this is like this topsy turvy world that we're in. And so, you know, now the Jevons paradoxes is front page news, you know, my view is that yes, you know, again, we need to be particular about what sorts of efficiencies are we looking for where and not, you know, sort of willy nilly, you know, create an environment where, which I'm not saying you're doing Charles, but you know, what we don't want to do is create an environment where if you can just say it's more efficient, then, somehow, you know, we're all good, right. Which is, you know, what some of the social science of Energy Star has actually suggested that, that stuff is going on.

With that said, right,

I am a big fan of the Hugging Face Energy Star initiative. That looks incredibly promising. And I think one of the things that's really promising about it, so this is, you know, you know, leaderboards when, you know, people put their models up on Hugging Face. There's some energy measurement that happens, some carbon measurement, and then, you know, leaderboards are created and all the rest of it. And I think one of the things that's really good at, right, I can imagine issues as well, but you're A, you know, creating a way to give some people credit for actually looking. B, you're creating a way of distinguishing between two models very clearly, right? So in that context, do you have to be perfect about how many kilowatts or watts or whatever it is? No, actually, right? Right? You know, you're looking at more or less in comparable models. But C, it also interjects this kind of path dependence. Like, who is the next person who uses it? Right?

That really matters. If you're setting up something early on, yes, they'll do something a little bit different. They might not just run inference on it. But you're, changing how models evolve over time and kind of steering it towards even, you know, having energy presence at all. So that's pretty cool to my mind.

So I'm looking forward to...

Chris Adams: Cool. We'll share a link to the Hugging Face. I think they, I think, do you know what they were called? I think it's the, you might be, I think it's, it was initially called the Energy Star Alliance, and then I think they've been told that they need to change the name to the Energy Score Alliance from this, because I

think it, Energy Star turned out to be a trademark, but we can definitely add a link to that in the show notes, because, these, this actually, I think it's something that is officially visible now. It's something that people have been working on late last year, and now there is, we'll share a link to the actual GitHub repo, to the code on GitHub to kind of run this, because this works for both closed source models and open source models. So it does give some of that visibility. Also in France, there is the Frugal LLM challenge, which also sounds similar to what you're talking about, this idea of essentially trying to emphasize more than just the, you know, like to pay a bit more attention to the energy efficiency aspect of this and I'm glad you mentioned the DeepSeek thing as well because suddenly everyone in the world is an armchair expert on William Stanley Jevons paradox stuff.

Everybody knows! Yeah.

Dawn Nafus: Actually, if I could just add one small thing, since you mentioned the Frugal effort in France, there's a whole computer science community, sort of almost at a step's length from the AI development community that's really into just saying, "look, what, you know, what is the purpose of the thing that I'm building, period."

And even, and that, you know, frugal computing, computing within limits, all of that world really about how do we get, you know, just something that somebody is going to actually value, as opposed to, you getting to the next, you know, score on a benchmark leaderboard somewhere. so I think that's kind of also lurking in the background here.

Chris Adams: I'm glad you mentioned this, what we'll do, we'll add a we'll add links to both of those and, you immediately make me think of, there is this actual, so we're technologists mostly, the three of us, we're talking about this and I work in a civil society organization and, just this week, there was a big announcement, like a kind of set of demands from civil society about AI that's being shared at the AI Action Summit, this big summit where all the great and good are meeting in Paris, as you alluded to, next week to talk about what should we do about this? And, they, it's literally called Within Bounds, and we'll share a link to that. And it does talk about this, like, well, you know, if we're going to be using things like AI, what do, we need to have a discussion about what they're for. And that's the first thing I've seen which actually has discussions about saying, well, we should be actually having some concrete limits on the amount of energy for this, because we've seen that if this is a constraint, it doesn't stop engineers.

It doesn't stop innovation. People are able to build new things. What we should also do is we should share a link to, I believe, Vlad Coraoma. he did an interview with him all about Jevons paradox a few, I think, late last year, and that's a really nice deep dive for people who want to basically sound knowledgeable in these conversations on LinkedIn or social media right now, it's a really useful one there as well.

Okay, so we spoke a little bit about these ones here. Charles, are there any particular projects you'd like to kind of like name check before we start to wrap up? Because I think we're coming up to the hour now, actually.

Charles Tripp: I don't know, not particular, but I did mention earlier, you know, we published this BUTTER-E data set and a paper along with it, as well as a larger one without energy measurements called BUTTER. Those are available online. You can just search for it and you'll find it right away. I think, if that's of interest to anyone hearing this, you know, there's a lot of measurements and analysis in there, including, you know, all the details of analysis that I mentioned where we, had this journey from number of compute cycles to, like, amount of stall, in terms of what drives energy consumption.

Chris Adams: Ah, it's visible so people can see it. Oh, that's really cool. I didn't realize about that. Also, while you're still here, Charles, while I have access to you, before we did this interview, you mentioned, there's a whole discussion about wind turbines killing birds, and you were telling me this awesome story about how you were able to model the path of golden eagles to essentially avoid these kind of bird strike stuff happening.

Is that in the public domain? Is something, can we link to that? That sounded super cool.

Charles Tripp: There's several, papers. I'll have to dig up the links, but there's several papers we published and some software also to create these models.

But yeah, I worked on a project where we looked at, we took, eagle biologists and computational fluid dynamics experts and machine learning experts.

And we got together and we created some models based off of real data, real telemetry of tracking, golden eagle flight paths through, well, in many locations, including at wind sites, and match that up with the atmospheric conditions, the flow field, like, or graphic updrafts, which is where the wind hits, you know, like a mountain or a hill and it, some of it blows up.

Right. And golden eagles take advantage of this as well as thermal updrafts caused by heating at the ground. Right. Causing the air to rise to fly. Golden eagles don't really like flapping. They like gliding. And because of that, golden eagles and other soaring birds, their flight paths are fairly easy to predict, right?

Like, you may not know, like, oh, are they going to take a left turn here or right turn there, but generally they're going to fly in the places where there's strong updrafts and using actual data and knowledge from the eagle biologists and simulations of the flow patterns, we were able to create a model that allows wind turbines to be cited and also operate, right?

Like, what, under what conditions, like, what wind conditions in particular and what time of year, which also affects the eagles' behavior, should I perhaps reduce my usage of certain turbines to reduce bird strikes? And in fact, we showed that it could be done without significantly, or even at all, impacting the energy production of a wind site.

You could significantly reduce the chances of colliding with a bird.

Chris Adams: And it's probably good for the birds too, as well, isn't it? Yeah.

Alright, we definitely need to find some links for that. That's, going to be absolute catnip for the nerdy listeners who put, who are into this. Dawn, can I just give you the last word? Are there any particular things that you'd like to, I mean actually I should ask like, we'll add links to like you and Charles online, but if there's anything that you would draw people's attention to before we wrap up, what would you pay, what would you plug here?

Dawn Nafus: I actually did want to just give a shout out to National Renewable Energy Lab, period. One of the things that are amazing about them, speaking of eagles, a different eagle is, they have a supercomputer called Eagle. I believe they've got another one now. It is lovingly instrumented with all sorts of energy measurements, basically anything you can think to measure.

I think you can do it in there. There's another data set from another one of our co authors, Hilary Egan, that has some sort of jobs data. You can dig in and explore like what a real world data center job, you know, situation looks like.

So I just want to give all the credit in the world to National Renewable Energy Lab and the stuff they do on the computing side.

It's just phenomenal.

Chris Adams: Yes, I think that's a really, I would echo that very much. I'm a big fan of NREL and the output for them. It's a really like a national treasure Folks, I'm really, thank you so much for taking me through all of this work and diving in as deeply as we did and referring to things that soar as well, actually, Charles. I hope we could do this again sometime soon, but otherwise, have a lovely day, and thank you once again for joining us. Lovely seeing you two again.

Charles Tripp: Good seeing you.

Chris Adams: Okay, ciao!

Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

The Week in Green Software: Transparency in Emissions Reporting

27 februari 2025 | 53 min

For this episode of TWiGS, Chris and Asim discuss the latest developments in emissions reporting, AI energy efficiency, and green software initiatives. They explore the AI Energy Score project by Hugging Face, which aims to provide an efficiency benchmark for AI models, and compare it with other emissions measurement approaches, including the Software Carbon Intensity (SCI) for AI. The conversation also touches on key policy shifts, such as the U.S. executive order on AI data center energy sourcing, and the growing debate on regulating the data center industry. Plus, they dive into the Beginner's Guide to Power and Energy Measurement for Computing and Machine Learning, a must-read for anyone looking to understand energy efficiency in AI.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Asim Hussain: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Resources:

CodeCarbon [06:00]
Optimum Benchmark | Hugging Face [06:12]
SCI for AI | GSF [06:40]
ITU [07:07]
Responsible AI Institute [10:24]
EcoLogits [15:07]
NREL Data Catalog [25:50]
Kepler | CNCF [30:14]
Environment Variables Ep97: How to Tell When Energy is Green with Killian Daly [33:52]
The Problem of Jevons' Paradox in AI's Polarized Environmental Debate | Sasha Luccioni [49:32]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Asim Hussain: There's this assumption out there that we're trying to hunt for the right, true essentialist value of measurement, and it really isn't like that

I'm your host, Chris Adams.

Hello, and welcome to another edition of This Week in Green Software. I'm your host, Chris Adams. Today, we're tackling an ongoing conversation in software today, predicting, measuring, and accurately reporting emissions data, particularly in AI. And as AI adoption skyrockets, so does its energy footprint.

Putting pressure on data infrastructure and sustainability goals. So today we'll be looking at a few new reports, what's going on, and generally doing a kind of roundup of the news and recent events along this. Because it's not all doom and gloom, although there is some. I'm also joined today by my friend and frequent collaborator, Asim Hussain.

Asim, can I give you some space to introduce yourself before we do our weekly, well, semi weekly, news roundup?

Asim Hussain: Not so weakly, anymore. Yeah. Hi. I'm Asim Hussain. I'm the Executive Director of the Green Software Foundation. So we are a standards organization and our mission is a future where software has zero harmful environmental impacts. And you might not be surprised to hear that we believe one of the best paths forwards is developing standards through consensus of multiple organizations.

Because through setting those standards, you can direct billions of dollars into the right places. And if you do it wrong, you can direct billions of dollars into the wrong places. So let's do it right.

Chris Adams: Okay. Thank you for that Asim. If you're new to this podcast, my name is Chris. I'm the director of technology and policy now at the Green Web Foundation, which is not the same as the Green Software Foundation. It's a small Dutch nonprofit, although we are members, founding members of the Green Software Foundation, along with a number of other much, much larger technology giants.

And I'm the host of this podcast and I'll also be doing my best to compile all the links and stories that we have so that if there's anything that has caught your interest as you listen to this, possibly whilst you're washing your dishes, you've got something to follow up with later. Alright!

Asim Hussain: Is it time for my yearly apology for naming it the Green Software Foundation and causing this constant confusion?

Chris Adams: I think it might be, but sometimes it works in our favor as well, because when people speak to us, like a scrappy startup, a scrappy kind of wacky little non profit, then they say, "oh, we've heard a bunch about you folks. Oh, we thought you were bigger," you know, so it's, we do have, it opens interesting doors. We sometimes do, I have had the odd conversation where people thought I was the Green Web Foundation.

Yeah. So this is,

Asim Hussain: Yeah, let's wear the hats that benefit us at any given moment.

Chris Adams: Pretty much, yeah. So this is what we're going to have and I think that we are doomed to have this mix up and the fact that we are speaking to each other on a regular basis probably doesn't help us, actually. Maybe we should, I don't know, have some big dramatic fallout or something.

Asim Hussain: Oh yeah, let's do like a fake fallout on the internet, yeah.

Chris Adams: We're not that keen for engagement, are we, mate? Let's not do that, alright? Okay. So, I was going to ask if you're sitting, comfortably, Asim, but I can see that you're on a standing desk, so I think you're now standing comfortably, presumably, right?

Asim Hussain: At attention.

Chris Adams: All right, well in that case, shall we start and look at the first story and then see where we go from there?

All right, so the first story that's kind of shown up on the radar is the AI Energy Score from Hugging Space. Sorry, Hugging Face, not hugging space, god. Yeah, so this is, this is actually essentially a project that is being spearheaded by folks at Hugging Face, but with also involvement from companies you've heard of like Salesforce and so on, to essentially work out something that might be a little bit like an Energy Star for AI.

Now, you probably, it's probably not called Energy Star because Energy Star is a trademark, but the general idea is, essentially, if we're going to have various AI models and things, then we should be thinking about them being efficient, and there are tools available to make this possible, actually. Asim, I know you had a chance to look at some of this, and you've had quite a few conversations with Boris Gamazaychikov the at Salesforce.

They're the kind of one of the AI leads. I'm mentioning Boris because he's quite involved in the GSF. There are lots of other people involved with the Hugging Face project, but Boris is the person who we know, so that's why we've got that named.

Asim Hussain: He's not, he's, so just to be clear like he's not a member Salesforce is not a member of the green software foundation. But yeah, I've just been chatting to boris obviously because we want to, one of the things we try and do is chat to everybody who's doing something in the AI measurement space so that we can at least try and coordinate and have like a common voice.

That's kind of one of the one of the things that we've been doing. Yeah

Chris Adams: Cool, and if I understand it correctly, we'll share a link to both the GitHub.io, the kind of public facing site with all this information about how the Energy Score project is working, plus the leaderboard, which has various closed and open source models. It's actually showing how efficient they are at performing particular tasks.

We'll also share a link to the GitHub repo, which actually shows how it's made because it's using tools that you may have heard of if you've ever messed around with AI models yourself. So it's using Code Carbon, which is pretty much the default tool that people use to work out the environmental footprint of a training run or anything like that.

And I believe the set, the other tool is Optimum or optimal Benchmark. I can never remember, but these two

Asim Hussain: that the actual benchmark tool? That's the thing that actually runs the benchmark, yeah.

Chris Adams: Exactly. So this is not like wacky stuff. This is stuff that you probably should have heard of or you are likely to come across, to see. And there is actually a Docker container for people who aren't able to publish their entire open models, with the idea being that you can run some of this

behind the file, as it were, so you can then share some of the numbers back And, Asim, I can't, while I've got you, I wanted to ask you about this because I know that the, I've been kind of tracking the AI Energy Score project for a few months, but I know there was some work inside the GSF to create a Software Carbon Intensity for AI

Asim Hussain: Oh yeah.

Chris Adams: these aren't competing, but they do overlap and maybe you could actually share a little bit more to explain what these two things are or even what is this SEI for AI in this context.

Asim Hussain: And there's also others as well. So we're talking Sir Joseph, the head of R&D is also sitting in with meetings at the ITU, International Telecoms Union, and so they're working on work themselves. There's EcoLogits from Samuel Rice. There's, there's other ones as well. And I probably just want to preface this by saying something, and I'm going to try and put some words to these thoughts.

I've internalized a lot of how I think about measurement and through conversations with others, I just want to make sure,

I want to try and get my point across, which is there isn't one true way of measuring everything. It's not like there's one winner and one loser.

What it is, is that different measurement systems have different trade offs. They incentivize certain things, they disincentivize other things, they have broader scopes and narrower scopes. And one of the things I've realized is you, it's almost impossible to create a measurement system which ticks every single box. Like it's almost impossible to have a measurement system which has the ability to measure like a broad spectrum of stuff and yet still also be consistent and repeatable and all these other areas, all these trade offs.

So yeah, I love AI Energy Score but there's also other ones as well. I just want to preface it by saying every single measure is designed for a particular audience and a particular problem. And I think that's kind of like one of the one of the ways I like to talk to people about it because I do get concerned that people they're always,

there's this assumption out there that we're trying to hunt for the right true essentialist value of measurement and it really isn't like that

so take all of my feedback on everything just with that, you know that context in mind yeah, so I think and I think that's kind of like one of the one of the ways that we look at it.

So what's really good about the AI, do you want me to talk about it? You know a

Chris Adams: please do. Yeah,

Asim Hussain: I mean

Chris Adams: I'm listening to more because I'm, I've got some things to share, but I'm, I haven't heard that much about this. And I haven't been, and I know that the GSF had these workshops going on where people have been exploring this stuff. And I haven't been in those, but I suspect I know you've been in beside them.

And I suspect there've been some good, interesting conversations as a

can't

result.

Asim Hussain: I can't

dive too much deep into it because we're still in progress and we had the agreement not to, you know, give too much information about in-progress stuff.

So if someone has a crazy idea, we're not going to publish it and We'll allow people to have these private conversations But I think there's some stuff I can share that one of the things that's come out from our conversations is there's a really, almost one of the most strongest feelings from the group is for a measure that really has a broad scope for a lot of different AI systems, but also for the breadth of the AI life cycle as well.

So, you know, not just inference and also not just training, but also like, the model as it's deployed in an infrastructure. So it's an end to end computation that includes everything across the chain from edge devices all the way over to data preparation. And so there's various scores, so for instance, there's something called the green, the Green AI Index from the Responsible AI Institute, which is also another measure, and that kind of focuses on a pretty broad spectrum.

There's AI Energy Score, which is excellent because it is focusing on just the model itself. And so when you think of the life cycle, it's not like, it's not gonna, it's just focusing on the model. And they've made, they've done a great job of making it a type of measure, which is consistent and repeatable. And they've done that by, you know, you've got your model. Here's our, here is the benchmark you run. You have no, you've got to run this benchmark. Yeah. you also have to run it on this particular hardware because you can't just get a better score by just running on a better hardware. You want to try and measure the model.

Like you've got to, you've got to, you've got to turn variables into constants to kind of get some sort of measure from that perspective and it's really interesting related to the next thing I'm going to talk about the beginner's guide to, it's a report that's coming out because they, I think they did a really good job they're trying to summarize different types of measurements and I think they put it as a system measurement was kind of very big picture It's kind of what one of the things I think maybe where the SCI for AI is going to be talking about. Then they're kind of job/application specific measurements where you kind of make more of those variables constants.

And then there's kind of what we call a code measurements, which are I want to measure, you know, the emissions of this piece of code. In order to do that, you really need to turn a lot of other variables into constants, so you can know that if you turned a for loop into a while loop, what the actual, like, impact would be.

And where I'd say AI Engine Scores is in terms of that taxonomy, it lands more on the code one. But I'm not saying that's a, I'm saying that is the only way you can get something that is consistent where you can actually have a model that, that, you can really give a score to. And it does incentivize a lot of things.

It incentivizes a lot of the almost code based

patterns to improve model efficiency. But it, because of the way it's worked, it won't incentivize other things. Like, it won't incentivize running compute in cleaner regions. Yeah, cause,

Chris Adams: different kinds of energy, or different cooling, for example, you're only looking at the, just the code part specifically.

Asim Hussain: And that's fine.

Yeah, that's fine. Because if you included that, then you wouldn't be able to have a measure that is going to tell you, okay, is Llama better than DeepSeek? They kind of just want to know that from a, you need to turn these things into variables. So, it's very good from that perspective. And I think it's one of the most advanced ones. It's the best one that does it's job. It does do it's job by being a, by, and they admit this, by having like a narrow bandwidth.

Chris Adams: There's one card it uses, I think it's an NVIDIA H100. I'm, I believe it's that, but I'm not sure I would know an NVIDIA H100 if it was dropped on my feet, so I need to be very clear that I'm at the limits of my expertise when it comes to hardware there. Okay, and the other thing we should probably mention, though, that this was one of the projects that was announced at the AI Action Summit in Paris that happened earlier on, I believe this month, actually, which has all kinds of announcements, so, in Europe, there is a, I think two, I think it's a 200 billion, yeah, a 200 billion euro fund specifically for rolling out AI across Europe.

There was a something that was kind of like a European take on this whole ridiculous Stargate thing. A ginormous French data center thing.

Asim Hussain: Yeah.

Chris Adams: That was Macron giving him some me too. And there was even actually for civil society, there was 400 million euro fund to kind of try and get an idea of the unintended consequences or talk about how you might reign in some of the worst excesses of this new technology that's being kind of deployed in all these places, sometimes where you're asking for it, sometimes where you might not be asking for it.

Asim Hussain: So 0.2 percent of the 200 billion is for

Chris Adams: Yeah. It's

Asim Hussain: the question of whether this is a

Chris Adams: It does speak volumes about our priorities, about who are we serving here, basically, I suppose, or whose needs are being prioritized when you have something like that. But yes, this is, this is some of the kind of ongoing conversations we, I guess, we actually have. there's just two things I just want to check because you used to, you mentioned a couple of projects that people might not be aware of that may be relevant for this conversation.

So you spoke about Ecologits, as I understand it, this is if you're using AI right now and you don't have a model, for example, I mean, you don't have like a whole training setup, you can use something like Ecologics to get an idea of inference. So that's, is that the case?

Asim Hussain: Yeah, I think, it does have a methodology as well. So you can actually just take their methodology and, I think he actually asked us to use the word estimate, but like, cause it's all not, none of this is direct measurements, right? So estimate the emissions of a model, but they also have like an API.

So if you have a named model you can call the API and it will kind of give you information about the, I do believe it's only carbon, it might be carbon and water, I can't quite remember, but it kind of gives you

Chris Adams: French, they have, there's like five specific impact, kind of impact factors. There's like water, ADP, like abiotic depletion, something like that. There's basically five things, and one of them is carbon, and one of them is energy, I believe. And this, you don't need to be, like, if you're already using Claude, or you're already using AI, OpenAI, this is just like a one Python package that essentially wraps the function calls you make to, to that API to get some of the numbers back.

So,

Asim Hussain: I don't think, I don't, I think Ecologist is just for models itself, I don't think it's for,

Chris Adams: Oh no, it is for inference. because we, we put a funding bid to the European AI Act Implementation Fund, where they were basically looking for this stuff. And the thing we realized was that if you are, if you're not doing any training, but you're just doing inference, this is one of the Python packages that will give you an idea about the numbers.

But it is very much,

Asim Hussain: inference only,

Chris Adams: yeah, exactly, inference

Asim Hussain: That's one of the conversations, yeah. Like the biggest conversation we're having in this side for AI right now is to include training or not to include training. And like one of the things the AI Energy Score and Ecologits is that it doesn't include training. The Green AI Index does include training. And, you know, that's it's a very, It's a very, oh god, it's such a hard question, it's like so much nuance to it.

Chris Adams: Well, yeah, because if you're including training, then whose training are you including, right? So if I'm using, say, Llama, should I be saying, should some of Llama's footprint, which was training, and we know, should that be allocated to me, or should it not be? And like, we can point to existing protocols that like say maybe you should, but in this case maybe that isn't.

So yeah, this is an open question right now.

Asim Hussain: Well if you, this is where my brain is so stuck in this area. Because if like, if you include open sources. I want open sources models in yours. It doesn't incentivize the reuse of models. If you don't include an open sources, if you're saying it's open source, I'm not going to include it. You can be a company that just goes, "I open sourced this model so I don't have any emissions." So there's like so many different ways it can be. This is a very, hard question that we need to solve. I also think it's very interesting because it's a, I think it's, I think it's, the training question is.

I, I suspect us figuring out or getting consensus on the training question, a very nuanced discussion and conclusion to the training question will actually help in many, other areas of like, how do you actually measure software? Because I think it's, it's, such a difficult question to answer.

I think the solution will inform so many other areas as well, which are kind of slightly simpler.

Chris Adams: It's almost as if using generally accepted accounting practices first developed hundreds of years ago might not be all that useful for thinking about how you use open source models and open weight models in

Asim Hussain: yeah, advanced technology systems.

Chris Adams: Okay,

Asim Hussain: it's something to do with cloning. Like, if you can clone something, a click of a button, you can't clone a chip. I don't know. I haven't got fully refined thoughts on this yet. So, let's move on.

Chris Adams: We'll wait with bated breath for these, the outputs from the workshops as you do them. All right. So. that gave us a lot of time to chat about that stuff. The other thing I'll just quickly name check for the AI Action Summit was there was a statement called within, the Within Bounds Statement.

I'll share a link to that. This was something that, actually my organization worked with or the organization I'm part of. So, Michelle Thorne, who's my colleague and normally sits next to me, she was working with 120 different civil society groups to basically lay out a set of demands to say, look, if we're talking about AI and we're allocating literally hundreds of billions of euros or dollars to this stuff, can we talk about what it's for and who's benefiting from this stuff?

We'll share a link to that because it's actually, in my view, quite well written and it does a very good job of actually talking about some of the issues that we might not be talking about all the time as people in industry to see how the rest of the world is actually like having to respond to some of this, I suppose.

So we'll share a link to that. But the juicy one now, Asim, is the one that you wanted to talk about, and that we both were nerding out a lot, was A Beginner's Guide to Power and Energy Measurement, an Estimation for Computing and Machine Learning. This is the next story that we have inside it, and I believe you've shared a link to the archive, the archive link for this pre print, because it's a really cool looking paper, and it's publicly available for everyone right now, but it might, I think it's going to be going to some journal, but I'm not quite sure, and figured But

Asim Hussain: I thought it got published in the, in an NREL journal. I don't know. Maybe it's not maybe it's not in a real journal or maybe now that I understand how journals what journals are maybe doesn't really matter

Chris Adams: So NREL here being the National Renewable Energy Labs of the United States of America. That's what NREL was in this case here. We've shared a link to it and, you did talk a little bit about why you like this, but can I give you a bit more space to talk about why you've enjoyed this? Because you don't need to be a beginner to actually appreciate this as far as I understand it, right?

Asim Hussain: No, it goes into a lot of detail. I mean, it says beginner, I'd say it goes a beginner's guide. Probably a little bit of imposter syndrome there, because I'd actually call it, like, it's very well written, so a beginner could start it, but I think it goes into very advanced topics that not many people know at all.

So, I think it goes from beginner to advanced. Yeah, I'm quite proud, Akshaya is the lead author of it, and Dawn Nafus is there, these are two people I worked very closely with at Intel. Very proud of this piece of work from them and the people, people over there. I share this with my team, so we're all working on kind of like thinking about how to measure energy.

And it's just exciting to see, just see how her and everybody else kind of rationalize this all into a very easy to understand, you know, set of concepts. As I said before, like they, they, you know, the first thing they go through to try and come up with this taxonomy, you know, are you measuring for a system?

Are you measuring for a job or are you measuring for code? And I think they've done a really good job of trying to like explain the difference they talk about are you measuring directly versus are you measuring versus proxies? I love the fact that she even goes down and said, there's this idea that we have is there's I always say like everything's a model like you can't, there's actually no such thing as direct measurements.

There's just a very advanced model. and she even goes down into, you know, even if you're using a watt meter and not against a wall, you've actually really got to consider like many of the rare areas because you've got to calibrate it. If you don't calibrate it, it's not going to really go, you calibrate a model, right?

It's not going to like, you know, actually turn out the right numbers and gives you a lot of cautionary tales, you know, where, what to think through. And it really just goes into just a lot of these. I don't know if it's worthwhile going into all of it, but there's just a lot of detail about the things to consider, you know, idle power draw, you know, not only that, but like when you run things, when you run, we always knew that like, it was challenging to measure when you're on shared infrastructure, but then they go into like other details, which is like, it gets even more challenging because the, like, the information you're getting from the socket might actually contain information from the energy draw from the memory and it's hard to, like, disambiguate all of this stuff.

There's ways in which, if you're accessing memory, it increases the idle power of a CPU. There is so much great information here, and a lot of little tips as well.

Chris Adams: Yeah, I think I would agree. It's if you are a beginner, there is some stuff that you can take away, but there is a lot of depth inside this. It's, I actually really enjoyed it too. I enjoyed reading it so much that, actually Dawn sent me, she emailed, I think, emailed me at the beginning of this year, actually, saying, "hey Chris, Check out this cool paper" and I really enjoyed reading it and we were going to do an interview.

We've actually got an interview lined up with Dawn Nafus and one of the other authors, Charles Tripp, who was writing for this. And I believe was at NREL and then has left NREL because,

Asim Hussain: because?

Chris Adams: of yeah, basically, this was the way that we could actually get some people speaking about it.

Because since we've had a change in administration, if you're a federal employee it's much, much more for you, difficult for you to talk about anything relating to, well, sustainability and technology, which is a real shame, especially when, like, it's useful to be able to draw upon expertise for people who do this kind of stuff, right?

So, maybe that's a question we should ask ourselves, like, are we okay with the people we're asking of these questions to not be able to talk to the public about this kind of stuff? But, what we do have, but to go back to the actual paper. I agree with you. I found it really, useful and this hierarchy of interventions was really useful because one of the key things that it kind of highlighted was basically where you have some control and where you don't have some control and give you a real chance to actually say, well, if I'm not able to do this, what, and what are my options?

If I'm still trying to make a meaningful and measurable, yeah, change. Because in many cases, you do have to think about some of the trade offs. The things you might do at a data center level to make some parts maybe slightly more energy efficient or maybe more carbon efficient can have knock on effects elsewhere, for example, further down the kind of, the list,

like further down the chain, basically. And this is what they do talk about. It's a really fun read if you're interested in AI. There's so much depth and the nice thing is the thing that one thing that's really quite nice about NREL specifically is that they've shared all the data to back up a bunch of this stuff.

So in the podcast interview that we have where we dive into this a bit more, we'll be showing there are some links to all the data sets that NREL was using when they were doing all these constant training runs to figure out what their, what the footprint of x might be and everything like that. So it's probably one of the most useful when open data sets we've seen for people who are trying to get

an idea about what the environmental footprint of using, I mean, AI directly, what the direct footprint of this might actually be.

Asim Hussain: I'd argue this is like a seminal piece, and you know, if there's like, I imagine this is going to be like essential reading for Green Software courses around the world. If you really want to like major software, you should this paper.

Chris Adams: Awesome work. I don't work with Akshaya, but I guess, awesome work Akshaya and friends, for that, but probably not just for beginners. So please do not be turned off by the beginners part. It's definitely not just for beginners. There's loads there.

Asim Hussain: They probably put beginners in to make sure the beginners read it but advanced people might think "I already know" so I already know tdp so I don't need to know this.

Chris Adams: Yes, by TDP, you're referring to the Thermal density. Oh, what does it stand for? But that's

Asim Hussain: I thought it's thermal design power

Chris Adams: I think it might be actually you're right. This is the amount of power that gets used at certain amounts of utilization, right? So if I'm using the chip at maximum output, it's going to use this much power. But if it's only using half it's going to be something like that.

Asim Hussain: Yeah, but it's also like Akshaya that kind of opened my eyes to understanding kind of how these power curves, she goes into detail here like how those, you know, we hear about these power curves which tell you 10% utilization is this, that's 30% is this. If you, i'm not going to go into details if you read the paper and you realize how those power curves are made they are very rough estimates of what it like looks like, you know, like you don't really know you don't really, you just, there's no register which is telling you I'm 50% percent like, you're just seeing how much throughput, you're just seeing how much you, basically...

Should I go into it? You basically chuck like a benchmark at it and you keep on hitting you keep on going like okay, dude, it was a website benchmark. Okay, do one hit per second.

Okay, it's fine. You keep on doing it until the benchmark can't go any higher and it's now like 500,000 page views a second. "Okay, I can't seem to do more than 500,000. I must be at 100 percent utilization." That's how that calculation works. And then you think to yourself, "Okay, what does 90% utilization mean?"

If I did 500,000, I'm just going to do 450,000 requests. And that's like the approximated idea of what 90% utilization means. But, what it really kind of ends up meaning is that it, the, it depends on the benchmark because an AI benchmark will have a different energy consumption, your pseudo 90% than a database benchmark, than this benchmark.

When you actually look at the big benchmark providers like, Esper, CERT and all these other ones, they're collections of different types of applications. And the power curve is the average of those. Which is why, like, if you know you're running, and that's why if you're using like a power curve based over, that's what I think it's saying, if you're using a power curve based off of a CERT benchmark, and you're saying that's what your AI consumption is, it might not be.

You really want a power curve which has been generated just by running a, an AI workload. Because the AI workload might just trigger different parts of the chip in different ways. It's very complicated. Yeah, and it, so, it's one of the things we were like, talking about, It's actually one of the reasons I kind of really like the way Kepler works.

Because Kepler,

Chris Adams: Sorry, I'm going to you there. before you go on this, the reason it's, I'm actually, the reason I'm quite happy to give some space for this, is that people who have listened to this might not know that you were literally working at Intel trying to figure this stuff out when you were doing a bunch of the green software stuff, so it's okay, listen, you know, I, like, you do have some prior art in this stuff, right?

Asim Hussain: Yeah. Yeah, we're basically diving into all this stuff. And I kind of learned so much while I was over there. How Kepler works is quite interesting. Is, So Kepler is this kubernetes based system which does a whole bunch of things but one really intelligent thing it does is it tries to figure out what your energy consumption is from the actual stuff that's running on the chips that you're running on. So it has like a machine learning model that, I think it's got, I think it's got some, if you start off Kepler with nothing and it doesn't know anything it will tell you energy numbers but it kind of learns and improves and fine tunes itself based upon A, your actual chips, B, how your chips were configured, C, what you're actually running on your chips.

So you kind of get a more accurate power reading from Kepler. One of the things I think would be great for them to do is to kind of just take that out of Kubernetes. And, because that doesn't necessarily need to be a Kubernetes piece, but it's baked into that infrastructure. Because that would be generally useful everywhere. Yeah.

Chris Adams: We will share links to both of those, and Asim, you're able to find a link for some of this power curve nerdery, that would be very, helpful, because I do know...

Asim Hussain: This paper's got it, yeah.

Chris Adams: Well, okay, in that case, we'll use that, because I do know that, well, some of the work I'm doing outside of being on podcasts with you, for example, I'm aware of, like, there are people putting together procurement guidelines where they speak specifically about this kind of stuff like please tell us what the figures are going to be for this power curve based on these ideas here and being able to refer to some of the actual literature is actually very helpful for people to understand why a government buyer might be asking for this stuff and why that's being used as one way to figure out some of the environmental footprints of the use of digital services.

All right, we'll add some links to that one and then we'll see what we're doing for time. Can I share one? I want to share a story from me. So this one, this is actually, it's not so much about, it kind of is about technology. This is actually an executive order from the USA called Advancing United States Leadership in Artificial Intelligence Infrastructure.

We've shared a link to this and the reason I shared this is because I think it's actually because I work in the policy working group inside the GSF and because we speak a lot about the carbon intensity of power and stuff like that. It's often quite rare to find really good, quite well written and detailed examples of kind of policy.

And this is one that, for a short, beautiful short period of days, was actually publicly available. So this was, I think,

Asim Hussain: I see the link is, oh, no, it's a real link. No, it is way back machine.

Chris Adams: It's webarchive.org, whitehouse.gov, briefing room, presidential actions, on the 14th of January. Just before the new guy came in, there was an executive order all about essentially, deploying AI, and this was specifically about if you're going to deploy AI on public land, what, and in the US

there's lots and lots of federally owned public land, what kind of criteria do you actually want to require as condition of people being able to put things on your land like this? So just the same way that people who have private land, they can say, you can run a datacenter here, as long as you do X, Y, and Z.

This pretty much lays out, okay, here's what you should be looking for. And this stuff includes a bunch of really, in my view, interesting and like very insightful and incisive policy, pieces of policy inside this. So when we talk about the carbon intensity of power, we've spoken before on this podcast multiple times about how in the hydrogen sector, we already have a very rigorous way of talking about how energy can really be green.

And done a recent podcast interview with Killian Daly from EnergyTag talking about this idea, like three pillars, the idea that energy has to be timely. So you can't have power at night being greened with like solar because they're two separate times a day. Deliverable, like you need to be able to have the generation on the same grid as you're consuming from because otherwise it's not very convincing that it's really powering it. And additional, you need to have new power coming in. This literally is name checking every single one of these inside this. Like the actual wording they use

Asim Hussain: in terms of power, in terms of more generally applying that

Chris Adams: this is specifically for data centers. So if all data centers are like, I'll read some of the kind of quotes from this. Basically, like, as part of ongoing work, the Secretary of Defense and Secretary of Energy shall, Blah, blah, blah, blah, blah, will require concurrent like any AI data centers on a federal site will have procured sufficient new clean power generation with capacity value to meet the data centers needs.

And they've, literally explicitly said "has to be deliverable and has to be matched on an hourly basis." So those are the three things right there. They've actually been more explicit about additional elsewhere. So this is like the three things that already in place in other industries, for the first time, really laid out for how the, how you should be doing this for AI data centers.

So if you're a policymaker outside the USA, just copy this link. This is probably some of the best stuff of particularly relating to policy, to energy policy. When

Asim Hussain: it, but does it say, by the way, shall, you know, you know, the shall means, so just everyone who is listing, shall is a very important term. Shall in the standard space. I presume the policy

Chris Adams: You don't get not do to Basically what they're saying.

Asim Hussain: You gotta

Chris Adams: is mandatory if you want to things on federal land. Elsewhere, yeah,

Asim Hussain: should is different. The, the, so just to, the reason you're talking about, as I presume it's the what's mandated is clean energy. Or is what's mandated,

Chris Adams: yeah, sufficient new clean energy power generation is they use, they, and later on, they actually talk about what counts as clean energy in this because there's a bunch of stuff, it's quite a long executive order, and we've had this new guy come in power, who's basically, who's rescinded every other executive order, apart from this one, even though it's not visible, so there's some stuff inside this,

Asim Hussain: into this one. There's something which benefits, benefits something else.

Chris Adams: So there is the whole thing here about, for example, this does say, well, if we're going to have clean energy, we're going to call it carbon free, and we're going to talk about not just renewable, like wind and solar, they talk about, say, the deployment of nuclear, which America, in America, people tend to be more receptive to, or in some places at least. So there's a part, there's a part there. But they even talk about, say, if you're going to have fossil generation, it needs to be 90 percent carbon capture, right? Now, this is a very high bar to hit, because there, right now, there's basically nowhere in any kind of at scale operation which is hitting 90 percent capture of this.

So if you were to have gas and you were to have this is probably about as rigorous as you can reasonably ask. And if anyone is actually, in the year 2025, when we know all the science available to us, you're not saying something like this, got to ask, okay, who's captured, who is captured here?

Because that is a really, like, there, there is just, it's, you need to have this if you're going to be talking about the use of fossil fuels inside this. And really, you probably shouldn't be using fossil fuels at all anyway. But like, this is examples of, yeah, this is what policy does look like.

If you're going to do this, do this properly.

Asim Hussain: Yeah, but at the same time I think what we're seeing is, I mean, it's interesting that the up, I don't know if I've got time to go into it, but the uptime report talks about the,

the increasing demands is forcing organizations to, you know, like you utility, there's so much demand from data centers.

It's not really a question of, you know, you've got to use clean energy. It's like, you don't have the energy or you now have to be a good place. You go to demand response.

But there's also then driving up pressure for those organizations. They're kind of walking back a lot of the stuff previously and there's a lot of fossil fuel generation being thrown out.

Asim Hussain: I have not verified this at all, but today I saw something on my feed. Which said that, I don't like, anyway, which, which said that, Elon's,

Chris Adams: You might be about the x.ai datacenter, the one in Memphis, running

Asim Hussain: in Memphis, there's gonna be, there's, like 15

Chris Adams: yes!

Asim Hussain: to power it. Which, you know, probably is because the utility said to him, "You're not putting an unbelievable load on our grid. We do not have the capacity for you." And he probably went, "ah, I'll build my own gas generators without asking anybody."

Chris Adams: There is a bit of a story behind this. So essentially, the, there was a datacenter, the x.ai datacenter was built very quickly by datacenter standards. And

usually, if you want to have power for a data center, you're going to have to wait some time if it isn't already available.

And, the, basically the approach that was taken was to essentially deploy a bunch of mobile gas turbines to provide the extra megawatts of power such that you could power that. Now the problem is these are really bad for local air quality. So you're shortening the lives of all the people who live around there, for a start, for the sake of this.

And, the other thing that, one of the reasons you're able to do this is because, they count as a mobile generators, they're not covered by the same clean air laws. So you wouldn't able to, yeah, exactly. So essentially this is stuff which has a real human cost, right? This is an already marginalized and kind of racialized community that it already has very bad air and has like elevated cases of asthma and all the stuff like that.

So there is a real human cost being paid here. And the decision has been made. "We're going to use this because we've decided that's more important than the lives of people around here." So, like, that's essentially what coming down to.

Asim Hussain: But also, I mean that, I'm guessing from the fact that this was an active executive order as a, you know, a few months ago that, that wasn't on federal land and therefore, or something like that must be, or

Chris Adams: This is somewhat separate. I mean, for a start, this, the, for the things, for the, xAI case in particular,

you don't, any of the local air guidelines or the local air kind of, laws about air, about air quality, don't apply to mobile providers.

Asim Hussain: Oh,

Chris Adams: providers.

Yeah. Yeah.

Asim Hussain: Even with this executive order, you can always get around it by just playing on mobile?

Chris Adams: So this was, this executive order came later. So we've had this things in xAI. That's been something that we saw last summer. All right. This was only published in January and they, and then it was literally on the White House website for seven days before the new guy came in and it down while pointing to the previous one.

And it's also worth bearing in mind that executive orders are not law. So even though someone can say they need to do this, that doesn't mean that it overrules existing law, example. So absent any other law, this is what you can ask for. And this is why they're able to say for federal law, this is the things we'd be doing.

There's actually a bunch of other really good stuff inside this, in particular, the air quality stuff. So the, as a contrast to saying, "It's okay to use this stuff. Who cares whose lives are shortened?" On the environmental justice, there's a whole piece in this about saying you, if you're going to deploy data centers in public land, then you need to have constant monitoring, all this visit, and have this visible everyone else to see as well.

So like these are the things that I think we don't see that you could totally take as examples away from this. And, they've also literally said. If you're going to deploy, you can't deploy in places which have had traditionally poor air quality below this, this air toxic, AirTox Screening. So basically, places which have already been harmed already, you don't get to deploy them in these places anymore.

And like, this is why I think this is actually quite well written stuff, because it does take into account all these things which we've had, which have been coming up again and again. So if you were trying to come up with some policy for deciding how you deploy, there is so much you can lift from this yourself, for your own corporate policies or anything like that.

Asim Hussain: There's very few benefits to a local community for having a data center built near you, there's very few jobs. There's like very, like, there's a couple of people walking around this giant warehouse and there's all, they've sucked all your electricity, and they, and there's, and you know. I don't know. The data center industry needs to, I was, it was fascinating to me when I was chatting,

I was at an infrastructure conference last year and I was chatting to a gentleman, won't name his name, from the utility sector, and he was saying to me something very interesting. He was saying to me, he believes the data center industry, this is before, he who shall not be named apparently, entered office.

So, this is before that happened, but he was saying he thinks the data center industry is headed right towards full regulation the same way utilities are regulated. So if you want to do a power plant, you can't just go "Oh, it's gonna make me a lot of money. I'm gonna build a power plant here." You have to go through so many checks and balances.

Your profit is limited. Everything is limited. And he was saying based upon the conversations that are happening, you know, you're claiming that this technology is so fundamental to life and existence that it therefore is a commodity, therefore it's something that's you know similar to energy. Energy utilities can't just say "ah we're going to rack up our prices 40, 43 percent because everybody wants it." You've got to, they'd be regulated for that.

So he was really putting a very convincing argument to me that if the data center industry is not careful It's going to get regulated that way and then they don't want to get regulated that way. It's not fun, apparently. And so I think things like this really matter.

Chris Adams: Yeah.

Asim Hussain: Really do matter.

Yeah, you to think about it. If you're with a data center, you can't be, you can't not think about the impacts of the region that you're in. You've got to really put effort into where you need to be a positive net benefit to the place you're being installed, you know, locally as well.

Chris Adams: So this is actually one thing that, so what I think you're, the argument you're making is that if you can, if you're going to present yourself as a utility, something which is what foundational to everything running on, then you probably, maybe there, then you should expect utility style profits rather than SaaS style profits, Because the margins that you might see, when you're from certain, tech giant companies is like 30 percent for example. That's not the same as utilities might be looking at like around 10 to 15 percent for example. And you have different kinds of oversight being introduced.

So yes, this is a conversation that we might have. I suspect it might be longer than we have given the time we have available, but yes, this is something we might point to. Just following on from this, there's a, you did mention this, uptime report, Uptime Institute Report. We'll share a link to that as well.

And I think there is, we might be in a situation where we have a bit of a fight on our hands, or we might be seeing a fight taking place because we do see like in Europe, for example, where, which is probably, arguably, the place where you see fights around data center deployment the strongest. We've just seen new laws be published about what criteria you need to actually have if you're going to connect to the data centers.

This was published, I think, last week, and we'll share a link to that. Where, in contrast to what we've just talked about here, where the US policy was very clear and was very good, we now see, essentially, a guideline saying you can connect data centers to the grid, but you need to have your own generation and you need to integrate nicely with the grid, but there's no mention of climate change, no mention of local environment or anything like that.

This is literally going to likely incentivize even more on site fossil based generation for this, absent no other criteria being in place. So we might see this being challenged, but I think I agree with you. We currently do have this case where, yes, you got all this new technology being deployed,but there is the kind of, we have a fight where there's almost zero regulation and it doesn't feel like it's going to last. It, I can't see how

Asim Hussain: You don't think, you don't think the absence of regulation is gonna last?

Chris Adams: I think what's going to happen is that if you continue to go through this stuff, you're, what's probably going to happen is that you will end up with so much pushback that you will end up with much, much more heavy handed regular legislative responses to this. Because, right now, there's been this push to kind of, essentially, neuter any kind of meaningful science based or data informed discussion around this.

All that does is play into the hands of a much, much more, a much, much more dramatic response later on. So I think it's, if you want to deploy stuff, then this does feel kind of long term, not very helpful for them. But then again, there's a question about, do we need this, how much do we actually need to be deployed?

There's probably a democratic discussion to actually have about that.

Asim Hussain: Well, we haven't even spoken about DeepSeek and its impact on this whole question and kinda how it has, if I'm going back to that conversation, how that person utilities that their big question was. Because

the data center providers, everybody's telling them we need a lot more energy in the future.

And they're going, "well, my God, do "we actually put the effort in to try and roll out this new capacity? And then only to find out on the day in the two years later like "ah, we got it wrong I'm sorry, we won't need that." They're asking the and I'm just a, they're asking the "is it BS?" question because they need really to figure it out and I was thinking okay, they might have just been convinced. Then DeepSeek comes along and Now you know everybody's asking the question "huh, will we need this capacity upgrade?" And now, as soon as DeepSeek came along, everybody said, "yeah, that's great. Now we are gonna do even more AI. We do, we definitely need the capacity, but now we can do more with it." And you're like, well, hang on. Because there is oftentimes a thing that goes on in, you have to create the hype to get the funding. You have to create the hype to get the funding. If you want to convince like investors to invest in your organization, if you want to convince them, you have to create the hype. And what DeepSeek's done is it's just popped it and I don't know how much it's popped it because only the investors know how much it's been popped.

But it's popped it and I wonder if it's really popped it quite significantly and whether we are going to see like a significant pullback. Is Stargate really going to happen or does it really not, really matter? They just want to hand money out to, it's just a reason to hand out 500 billion because you know, why not?

Chris Adams: We can share a link to, there's a good paper from Sasha Luccioni talking a little bit, and friends, about Jevons Paradox. I've actually written a blog post about this as well, particularly for DeepSeek, to kind of make this accessible for people who are trying to understand. Does this, is this going to reduce the footprint or is it going to increase the footprint?

Because there's a few different criteria you want to take into account. Just saying,

Asim Hussain: pops the bubble, it will decrease the footprint, I think.

Chris Adams: That's, this is the thing we can look into and decide. Because the flip side is that if this makes it more likely that they'll, you'll, if this lowers the barrier so that more people are able to use it in more places, that can lead to an absolute increase.

So there are different, there are two different, there are different ways and different takes on this and it's very much case of, okay, this is, yeah, this is one thing we'll show a link to. Asim, I think we've gone down a bit of rabbit hole so we should probably look at the events there's anything particularly we have here.

Asim Hussain: So, there's a couple events coming up.

There is the Practical Advice for Responsible AI on February the 27th at 6pm in London. , it's a UK event. And it's gonna be held in person in the Adaptavist offices, and it's gonna talk about green AI with Charles Humble and AI governance Team with Jovita Tam. There's the GSF Oslo meetup happening on again, February 27th at 5:00 PM. It is in person in the Accenture offices from 5 to 8 PM.

And they're going to talk about how to leverage data and technology to drive sustainability initiatives and enhance security measures, dive into green AI, obviously. There's going to be talks from Abhishek Dewangan and Jonny Mauland. I do apologize.

Chris Adams: Sorry,

Asim Hussain: Read them. I'm sorry, Johnny.

Sorry, Abhishek. Details in the podcast notes. And think that's it. I think I'll pass over to you Chris.

Chris Adams: Yeah. Okay, then. I think that takes us to the end of what we have for this. I assume if there's a particular free resource you would point people to right now on green software as a final thing, what would you point people to as a parting?

Asim Hussain: Oh, honestly, it's that beginner's guide. I don't know if it's I don't know if it's, it is very good, I read the Beginner's Guide to Power and Energy Measurement and Estimation for Computing and the last word.

Chris Adams: Wow, Akshaya better be getting a promotion after this, man. This is just like, this is, so yes, this, I agree. It was a really fun read. If you want to basically sound knowledgeable about AI, this is probably the most useful thing to read. And that's as someone who's written a report all about the environmental impact of AI ourselves, where we work.

All right, Asim, it's really lovely to see you again, mate. Thank you so much coming on. I hope the people who did listen to this were able to stay with us and we didn't go get too self indulgent. And if we did, please do tell us and we'll make sure that we don't do it too much next time. And otherwise I'll see you in one of the future episodes of This Week in Green Software.

Thanks, mate.

Asim Hussain: See you later, mate.

Chris Adams: Toodle oo!

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

How to Tell When Energy is Green with Killian Daly

20 februari 2025 | 61 min

In this episode, host Chris Adams is joined by Killian Daly, Executive Director of EnergyTag, to explore the complexities of green energy tracking and carbon accounting. They discuss the challenges of accurately measuring and claiming green energy use, including the flaws in current carbon accounting methods and how EnergyTag is working to improve transparency through time-based and location-based energy tracking. Killian shares insights from his experience managing large-scale energy procurement and highlights the growing adoption of 24/7 clean energy practices by major tech companies and policymakers. They also discuss the impact of green energy policies on industries like hydrogen production and data centers, emphasizing the need for accurate, accountable energy sourcing and we find out just how tubular Ireland can actually be!

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Killian Daly: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

GHG Protocol [09:15]
Environment Variables Podcast | Ep 82 Electricity Maps w/ Oliver Corradi [32:22]
Masdar Sustainable City [58:28]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Killian Daly: We need to think about this kind of properly and do the accounting correctly.

And unfortunately, we don't do the accounting very well today.

I'm your host, Chris Adams.

Hello, and welcome to another edition of Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. When we write software, there are some things we can control directly. For example, we might be able to code in a tight loop ourselves, or design a system that scales to zero when it's not in use.

And if we're buying from a cloud vendor, like many of us do now, we're often buying digital resources, like gigabytes of RAM and disk, or maybe virtual CPUs, rather than physical servers. It's a little bit less direct, but we still know we have a lot of scope for the decisions, to control the impact of their decisions and what kind of environmental consequences come about from that.

However, if we look one level further down the stack, like how the energy powering our kit is sourced, our control is even more indirect. We rarely, if ever, directly choose the kind of generation that powers data centers that our code runs in. But we know it still has an impact. So if we want to source energy responsibly, how do we do it?

If you want to know this, it's a really good idea to talk to someone whose literal job for years has been buying lots and lots of clean energy and is intimately familiar with the standards involved in doing so and who has spent a lot of time thinking about how to make sure you can tell when the energy you're buying really is green.

Fortunately, today I'm joined by just that person, Killian Daly, the Executive Director of the standards organization, EnergyTag. Killian, it's really, nice to have you on the pod. Thanks for coming on.

Killian Daly: Yeah, thanks. Thanks very much for having me, Chris. great to be on the pod and, an avid listener, also. So it's always nice to contribute.

Chris Adams: Thank you very much. Killian, I'm going to give you a bit of space to introduce yourself, and I've just mentioned that you're involved in EnergyTag, and we'll talk a little bit about what EnergyTag does. Because I know you and because, well, I met you maybe three years ago, I figured it might just be, it might be worth just talking a little bit about our lives outside of green software and sustainability.

So, we were in this accelerator with the Green Web Foundation talking about a fossil free internet, and you were talking about EnergyTag and why it's important to track the provenance of energy. I remember you telling, we were asked about our passions. And, you told me about surfing and I never ever thought about Ireland as a place where you would surf because I didn't think it was all that warm. So can you maybe tell me a little bit like enlighten me here because it's not the first country I think of when I think of surfing and when you said that I was like he's" having a joke, right?"

Killian Daly: Yeah. Well, I do like to joke, but this is not actually one of the jokes, Well, it doesn't need to be warm to surf. You just need to have waves, I suppose. So, yeah, it's something since I was really very young. I've always gone to the west coast of Ireland. Beautiful County Clare near the Cliffs of Moher.

Maybe people know of them. And so we go every year. And my cousins, since a very young age, started surfing. We just, you know, solve these big waves and there's other people out there, surfing, bodyboarding and we're like, "Hey, let's try that out. That looks really cool." So, yeah, since I don't know, 6 or 7 years old, I've been going there every year, in summer, also in winter, me and my cousins also go, yeah.

We go at New Year's get into the frigid cold Atlantic. And, yeah, it's magic, really. If you have the right, if you have the right wetsuit, you can kind of, you can get through anything,

Chris Adams: So there's no such thing as cold was it bad weather, just bad clothing that also applies to wetsuits.

Killian Daly: Yeah. Yeah. Yeah. It couldn't apply. Couldn't apply anymore. And obviously, in winter, you get the biggest swells, right? so actually, people probably don't know it, but Ireland has some of the biggest waves in the world. Now, on the west coast of Ireland, you have, yeah, really massive 50, 60 foot waves.

Yeah, really all you can get some sort of a, all time surf there. So, so yeah, it's one of one of our better kept secrets.

Chris Adams: I was not expecting to learn how to go totally tubular on this podcast.

Killian Daly: Yeah,

Chris Adams: Wow, that's, yeah, that's...

Killian Daly: It's not, not for the faint of heart, but yeah, I would definitely recommend it.

Chris Adams: Actually, now that you mention that, and now that we talk about, going back to the world of energy, now that people talk about Ireland as, the Saudi Arabia of wind, and it being windy AF, Then I can kind of see where you're coming from with it, actually. It doesn't make a bit more sense. So yeah, thank you for that little segue, actually, Killian.

Okay, so we've started to talk a little bit about energy. And, I know that your, the organization you work for right now is called EnergyTag. But previously, as I understood it, you didn't, you worked in other organizations before. And, you've been working as a kind of buyer of energy, so you know a fair amount about actually sourcing electricity and how to kind of do that in a kind of responsible way.

And I think when I heard you, we spoke about this before, you mentioned that, "yeah, I'm used to buying significant amounts of power" in your kind of previous life. Could I just like, could you maybe talk, provide a bit of a kind of background there, and so we can talk a little bit about context and size, because that might be helpful for us talking about the relative size that tech giants might buy and so on, and how much of that is applicable.

Killian Daly: Yeah, sure. Yeah, so, I've been thinking about energy for a long time, even before my professional career studied energy and electrical engineering since I was 18 years old and did a master's in that, also. And then obviously in my working life as well. I've been basically always in the energy sector.

So before EnergyTag, I was basically overseeing the global electricity portfolio, and the procurement of electricity for a company called Air Liquide, which is basically a large French multinational that produces, liquid air. So, oxygen, nitrogen, all the different parts of air which are, essential, feedstocks into various industries, and they consume a lot of electricity.

So, the portfolio my team oversaw was about 35 to 40 terawatt hours of electricity consumption.

Chris Adams: Okay.

Killian Daly: Yeah, it's a lot, it's more than my home country, Ireland. It's about the same as Google and Microsoft

Chris Adams: put together, yeah. Okay, so, wow. And

Killian Daly: So, it's pretty big stuff. And obviously, when you're working on something like that globally, looking at various electricity markets operating in 80 countries in these huge volumes, I suppose you, kind of learn a lot about what it means to buy power.

Chris Adams: I guess if you're looking at something which is basically as much power as an entire country, then there's going to be like country sized carbon emissions, depending on what you choose to power this from. And I guess that's probably why you, I mean, we, have ways of tracking power. I mean, tracking the carbon emissions from various things like this, I mean, called like the GHG protocol, which is a kind of like the kind of gold standard for talking about some of that stuff.

And this is something that I think you have some exposure to and I remember when you spoke to me, I remember us sitting down one time and you were telling me about There's a thing called scope 1 and there's a thing called scope 2, and that scope 2 was actually a kind of relatively new Idea where this came into this. Can you maybe tell me a little bit like maybe you could explain to someone who is Who's heard of, carbon footprinting, and they know there's a thing called scopes.

Why would anyone care about scope 2 in the first place? And how does it come about in the first place? Because it seems like it's not intuitive for most people when they first, when they start thinking about carbon footprints and stuff like that.

Killian Daly: Yeah. I think the obvious, first thing you need to take into account when you think of like a company's emissions is, well, what are they burning themselves on site? do they have gas boilers burning gas? Are they burning coal to produce electricity? So that's, I think, very intuitive and obvious. But actually that is not the end of the story. And there's actually like a, a very funny anecdote. I put a true anecdote from the legendary Laurent Segalen, who does the Redefining Energy podcast and general energy guru. And he was actually involved in the kind of creation of a lot of the carbon accounting standards that are used today, this Greenhouse Gas Protocol standard, which is basically used by over 90 percent of companies now to report their carbon emissions.

It is the Bible of how carbon accounting works, right? and so 20 years back, he basically was, down in Australia and visiting an aluminum smelter. On site, they were explaining, "this is very low carbon product. we hardly burn any fossil fuels on site. This is incredibly, clean production."

Chris Adams: The aluminium here, right? big chunks of aluminium. Okay, right.

Killian Daly: Aluminum, aluminum smelting. So like one of the, biggest metallic commodities that we have, very energy intensive. and so, he was there on site and just saw these big overhead wires coming in from yonder, from somewhere, right? And he said, hang on, what are the, what are those big cables above? and they were like, "oh, yeah, that's the electricity," obviously driving the smelter because aluminium, it's all about electricity. That's what power is an aluminium production facility. And so he said, well, hang on, where is that coming from?

They're like, "oh, no, don't, don't worry about that. That's not our responsibility." Well, it absolutely is, right? so you need to think about where is that electricity coming from? How is that being produced? And in that case, it was coming from a very large multi gigawatt coal power plant right next door.

Chris Adams: Okay. All right. So I thought you were gonna say, oh, it's maybe a, something clean, like a hydro power station, but no, just a big, fat, dirty, great coal fired power station was the thing generating all the power for it. And that's where

Killian Daly: Absolutely. So, that's kind of the, just a bit of an anecdote is that's why it's so important to think about what we call scope to emissions, the emissions of electricity that I'm consuming, because especially as we electrify the economy, right, more and more emissions are going to become scope 2 emissions.

They're going to be related to someone else either burning fossil fuels to produce electricity and to give to a consumer or ideally, using clean energy sources to generate that electricity without carbon emissions.

we need to think about this kind of properly and do the accounting correctly.

And unfortunately, we don't do the accounting very well today.

Chris Adams: Alright, so previously, before we even had that, there wasn't even this notion of scope 2 in the . , you might have just had direct, and then maybe this kind of bucket of indirect stuff, which is really hard to measure, so you're not going to really try to measure it. And okay, so, I remember actually reading about some of this myself, and I always wondered, like, where do some of these figures come, where do, where does even the notion of a protocol like this come from? And one of the things I realized was, particularly with the GHG one, was that they're like, when I listened to Laurent Segalen speaking about some of this, he was basically saying, yeah, this was essentially like Shell, the oil company, who basically said, "we have a way of tracking our own emissions."

And, why not use that as a starting point for talking about how we do carbon accounting? And then, scope 2 was a new concept. That was one of the things that they were kind of pushing for. But I suppose this kind of speaks to the idea of, who's in those rooms for those working groups to kind of, that is going to totally change the framing of how we talk about some of this.

And I guess that's probably why this, is this a little bit like why you started talking and getting involved with things like EnergyTags so you could take part in those discussions? Because it feels if this is what we're going to use to define how we do this or how we do that just like you have people talking about okay BP had an impact of changing how we think about carbon footprints from, from an individual point of view.

But you do need people involved in that conversation to say, "actually, no, that's possibly not the best way to think about this, and there are other ways to take this into account." I mean, is this why you got involved in the EnergyTag stuff?

Killian Daly: Yeah, it's one of the main reasons, because I used to do, so, work for one of the world's largest electricity consumers. And so I was responsible for calculating all of the electricity emissions for that company, right? Like doing the scope 2. And so I read the Greenhouse Gas Protocol back to front.

That was how the, all the calculations were done. That's what qualified clean and not clean, right? And I remember thinking, "this is an insanely influential document," right? It's kind of in the weeds. It's kind of stayed maybe, to some people, but I was

Chris Adams: of tedium around it, here.

Killian Daly: Yeah. But the more I've gotten involved in things like regulation and conversations like that, that is where, it's in the annexes, it's in the details that the big decisions are made often. So I remember thinking back then, this is insanely influential and some of the ways that we're allowed to claim to consume clean energy are, frankly, disconnected from reality in a way that is just not okay, right?

As in this is far too weak. And definitely, I thought, someday I'd love an opportunity to be able to, say, "hang on, can we,

we fix this please? can we do this differently? Can we start to respect some sort of basic realities here?" So, yeah, it was definitely one of the drivers why I joined EnergyTag, which is obviously like a nonprofit that is, has as its mission to clean up accounting, right? And to clean up the way we think about electricity accounting. So, yeah, obviously it's a great honor, I suppose, to be part of those ongoing discussions in the Greenhouse Gas Protocol update process.

Chris Adams: So, We spoke before about how there, before there was even no scope 2, right? So that was like, the bar was on the floor. Right, and then we introduced the idea that, oh, maybe we should think about the emissions from the electricity. So that was kind of a bit of a leap forward by one person pushing for this, that otherwise wouldn't have been in the standard at all, right?

And I just realized actually now that you mentioned that, we spoke about oil firms being very involved in this and being very organized in this, and I remember people talking about Shell, that's what you use, and how much, and I'm just realising, oh Christ, Shell's in the Green Software Foundation as well.

We should, that's something I didn't really think so much about, but they're also there too. So they are organized. Wow. So let's move on. So maybe we could talk a little bit about scope 2 here. The thing I want to kind of get my head around is I'm like, can you maybe talk me through some examples of where this doesn't, this falls down a little bit, where might be a little, stretching your, you spoke about the physicality, the physical reality. where does it need a bit of work, or need some improvement that you're looking to do, looking to address in EnergyTag, for example?

Killian Daly: Yeah, so basically, one way of doing scope 2 accounting is basically looking at the energy contracts or the electricity supply, contracts that companies have and saying, well, where are you buying your energy from? How are you contracting for your power? Right? And there's a kind of a number of fundamental issues.

One of them is around the temporal correlation, or between when you're consuming electricity and when the electricity you're claiming to consume is being produced. And today, right, we actually allow an annual matching window between production and consumption. And put in simple terms, what that means is that you can be basically solar powered all night long, right. You can take solar energy attributes from the daytime and use them at nighttime, or you could take them from the daytime in March and use them at nighttime in November. At any other time of year. And this just does not make sense, right?

Chris Adams: Not physically how the science works for a start. Maybe if I can just dive into that a little bit in a bit more detail because you've mentioned this idea of certificates for example or like claiming like that and as I understand it if I am running a solar farm right I'm generating two separate things. I'm generating power but I'm generating the kind of greenness so these are two independently sellable things which will sometimes be bundled together. That's how I might buy green energy. But under certain rules, they're not. They can be separated. So it's like the greenness that I'm moving or I'm buying and kind of slapping onto something else to make it green. Is that? And if it's at the same time, it's kind of okay. If it's from totally separate times of day, you do like you mentioned where you're saying this thing running at night runs at solar, is running on the greenness from a solar farm, which is stretching the, well, our imagination, I suppose, and your credulity, I suppose.

Okay, so that's one example of this is something that you wanted to get, wanted to get fixed. Are there any other ones, or things that you'd point people to, because

Killian Daly: I think you know the. The other, the other aspect, I think that's pretty, problematic in today's standards is so we've talked about time and the other big one is space, right? Today we allow consumers to claim to use green energy or clean energy over vast geographical boundaries that really don't respect the physical limits of the grid.

So, for example, the whole U. S. is considered to be one region, right? So you can buy green energy attributes produced in Texas and say that you're using them in New York. So you could be 100 percent power by Texas solar in New York. Or if you're in Europe, Europe is considered of one region. So you have really absurd cases where you can be powered by Icelandic hydro in Germany, and Iceland has never exported any electricity to anyone. There's no cables leaving Iceland. So, that just doesn't make sense. And this has real consequences because what we're trying to do is obviously drive consumers to buy green energy. If they're doing it in this way, then they're kind of, in some cases, pretending to buy green energy rather than actually going and buying green energy and incentivizing more production of green energy and clean flexibility that's needed to integrate that solar and wind, at every hour of the day.

So, that time and space kind of paradigm is maybe a good way of thinking about, some of the fundamental issues here. There are other ones. I don't know how far we want to go into the rabbit hole, but that's two very high level, and hopefully very kind of understandable examples of the problems we have with today's carbon accounting.

Chris Adams: Yeah, I think I understand why that would be something we would address, and so presumably this is the thing that EnergyTag's looking to do now. You're basically saying, well, the current system is asking you to make quite spectacular leaps of faith. And there are certain places where you do want to do leaps of faith and be super creative, but accounting might not be where you want to be super creative or super jumpy. That's not always where you want to have your innovation.

So that's, this is, so you're saying, well, let's actually be, make this more reflective of what's really happening in the world. So that we've got like some kind of solid foundation to be working on. So,

Exactly.

Killian Daly: And just maybe on that point, this is not what we advocate for is not, it's not anything radically new, to be honest, because the way electricity markets work today, the way electricity utilities deliver power to customers, just you know, let's say pure gray electricity on electricity markets.

It is based on fundamental concepts of time matching. Power markets work on a 16, sorry, a 60, 30 or 15 minute, like balancing period. In Australia, it's 5 minutes. In Europe, there's things called bidding zones. So that's the area over which you can buy and sell electricity. And all of this is to kind of capture these fundamental physical limits of the power system.

You have to balance it in real time. And there's only a certain amount of grid capacity. And so you need to realize areas over which it's reasonable to trade power or not. So all we're saying is, make the green energy market much more like the real power market. So we're actually, if anything, trying to make it a bit more common sense,

whereas today, we're, quite detached from some of those basic limits that

Chris Adams: Ah, I see. Okay. So in fact, in some ways, there are some kind of comparisons where you could plausibly make where people there's a push right now for people to talk about treating environmental data with some of the same seriousness as financial data and apply some of the same constraints it sounds like something a little bit like that so if people are going to have basically take into account the physical constraints when they're purchasing the actual power part, they should think about applying their same ideas when they're thinking about the greenness of it as well. You can't kind of cheat, even if it makes it a bit easier, for example.

Killian Daly: Yeah, well, exactly. And, ultimately, what are we trying to do here? Is the purpose so that certain consumers can say that they have no emissions, or is the purpose to set up an incentive system so that when those consumers actually. Do you say they have no emissions that they've gone through all of the challenges of grid decarbonization?

So they've bought renewables. So they've invested in storage. So, fine, you can consume solar power at nighttime if you put it in a battery during the daytime. They're thinking about, demand flexibility. Are they consuming a bit less when there's less wind and sun? They're hard challenges, right?

We need to do a lot more of those type of things, and a proper accounting framework will make sure that in getting to zero that you have to think about and take all of those boxes. Whereas today, you can just be 100 percent solar powered and obviously that's just not going to lead to the grid decarbonization in the real world that we want to see.

Chris Adams: Maybe if you're in space it might work, but mostly no. Okay.

Killian Daly: Mostly, no. Yeah,

Chris Adams: Okay, so we spoke a little bit about why there are some problems with the existing process, and like you, we've spoke a little bit, hinted at some kind of ways you could plausibly fix this. So do you, could you mind just talking me through some of the key things that EnergyTag is pushing for in that case?

Because it doesn't sound like you're trying to do something totally wacky, like, say you're never allowed, sorry, you're, it's not like you're asking for something like a significant change, like you're not allowed to split the greenness from power and or stuff like that. It sounds like you're still working inside the current ways that people are used to buying power and do all that stuff at the moment, right?

Maybe you could tell me about how it's supposed to work on the newer schemes that you're working with.

Killian Daly: Yeah. So basically what we're advocating for is that, if you're gonna claim to use green energy based on how you contract for power, then, well, you have to temporally match, right? So you can only claim to use green energy produced in the same hour as your consumption. Not in the same year, Okay. number 1. Number 2 is we need to think about the deliverability constraints, right,

and this geographical matching issue. And what we're saying is that, for example, in Europe, Europe is not a perfectly interconnected grid. And so you shouldn't be able to claim you're consuming green energy from anywhere else in Europe, you should be doing it, in the same bidding zone or, at least at a

Chris Adams: There needs to be some physical deliverable, physical connection to make it possible. Okay.

Killian Daly: Or fine, you can go across border, but you have to show that actually the power actually did come across border and that you're not violating like fun. You're not importing, 10 times more certificates than you are real power between 2 countries, right? So we need to have those, limits put in place.

And another thing that we think is important is that there needs to be some sort of controls on individual consumers just buying a load of certificates, for example, from very old assets. And I'm totally relying on those to be 100 percent green. For example, if I'm in Germany, right, and I just sign a deal with a hydro power plant, that has existed for 100 years and I'm time matched and I'm also within Germany, spatially matched, and I'm claiming to be 100 percent renewable

Chris Adams: it's not speedytransition if it's a hundred years old, that feels like that's stretching the definition of being an agent of that. Okay.

Killian Daly: that's another thing to kind of, you know, having this 3 pillar framework.

Sometimes we call about, and that is very important. I think for an existing consumer, it is legitimate to claim a certain amount of that existing power, but that must have a limit, right? You can't just be resource shuffling and "well I'm the one who's taking all the green energy" and everyone else is left with the, fossil that needs to be controlled also.

Chris Adams: All right. I think I follow that. So basically, so timely has to be more or less the same time, right? Deliverable, like you need to be able to demonstrate that the power could actually be delivered to that place. So deliverable there. And this other one was like, additional, like we need to transition, so you can't look at something which is 100 years old or 50 years old and say "I'm using that, I'm fine." There is this notion of bringing new supply stream to kind of presumably displace or move us away from our current fossil based default, which is not great from a climate point of view, right?

Killian Daly: Exactly. And I think one way, there's a really, a good friend of mine, who's in the Rocky Mountain Institute, Nathan Iyer, smart guy. We've worked a lot on US federal policy topics, and he actually has a really, good analogy about this stuff. BYOB, right?

So, yeah, of these 3 pillars. So, like, when you're going to a party, you need to bring your beer to the party on time. You can't bring it yesterday, you need to bring it when the party is happening. You need to bring it to the party, not to another party. And it needs to also be your own beer.

You can't just be taking someone else's. And it's it's kind of like a bit simplified, but it's a good analogy, I think for what we're trying to get out here. It's if we get everyone to start like thinking that way and acting on those kind of fundamental principles, obviously, we're going to end up being much more effective in deeply decarbonizing our power systems.

Chris Adams: So, decarbonization of the grid communicated through the power of carbonated beverages, basically. Wow!

Killian Daly: What could be better?

Chris Adams: I think it's, well, it's topical, at least it's still talking about CO2, just on slightly different scales, actually. I quite like that, actually. I might borrow that one myself, actually. Okay. So, there's one thing that you mentioned then. So this notion of, we spoke a little bit before about there's this idea of greenness that could be split, you're still keeping that, so you're not, saying, there's no ban on saying you're not allowed to sell power, that is unbundled from that, there is, that is still a kind of key idea of flexibility, could you maybe, I mean, cause from someone who isn't familiar with it, they might say, "why do we even have this, idea of being able to have separate these in the first place.

Doesn't this make things much more complicated?" I mean, I might be going down into the weeds, but is there a reason for that? is it just because that's how it's such a big change there that, or it's really hard to make that, to get people to shift to a new way of doing things or, what was that, what's the thinking around that part there?

Killian Daly: Well, basically, right, anytime you want to claim or have a contract, whether that be an unbundled or a bundle PPA contract,

Chris Adams: Power Purchase Agreement, right?

Killian Daly: Yeah, a power, like a long term power purchase agreement, for example, right? so anytime you have a contract for a specific type of electricity, you need an accounting mechanism or a tracking mechanism that kind of sits on top of the grid and allocates generation to consumption, because

obviously, the way that the grid actually works, is that electrons are just oscillating around the place. there's not really a methodology to physically trace this individual electron started here and went there, right? And so, much like power markets do, and they have mechanisms for contractually allocating power between different buyers and sellers, as long as it's matched in time and space, that's a fundamental premise of our power markets work, we're basically borrowing that concept, but attaching the greenness attribute,

Chris Adams: Ah,

Killian Daly: and saying "provided that this system, of detaching greenness from the power is respecting temporal and geographical matching requirements, deliverability requirements, sufficiently, then that should be the basis of legitimate green claims and that essentially creates a market mechanism for financing renewables.

If you don't do that, then you cannot have a green power market basically, right? You,= don't have a way of differentiating buyers who are contracted for green power and those who are not doing anything. So, yeah, for example, a few years ago in Air Liquide, we only did this, we didn't look at what contracts we were sourcing.

We just did this location based accounting where you take an average of all the generation in the grid. Which is another way of looking at electricity emissions and a very valid way of doing it. But obviously one disadvantage that has is that it basically leaves all consumers passive.

They have no incentive to do anything in terms of driving electricity decarbonization. So that's why we need these, these mechanisms of essentially having tracking

Chris Adams: systems.

Oh, okay, I see. So, if you, if there's no recognition, if I'm working at a large company, why would I, why would I choose to buy something green if I can't be recognized for me doing something, doing that green step? And, so the downside of the location based approach is that yes, it gives you one single answer, but it takes away this idea that organizations which have honestly massive amounts of resources can influence or speed up a transition.

That's what it seems to be a kind of it's trying to respect that reality or at least acknowledge that this is what we expect of organizations if they're that powerful.

Killian Daly: And one person, I know you've had Olivier Corradi from Electricity Maps on before they've done, some very good blog series on this topic. They're obviously have insanely deep knowledge of grid emissions is really no one better that I've come across.

And they did a very kind of simplified explanation of this stuff. And you have the location based method, which is like maximizing physical accuracy and then you have the market based method, which is trying to maximize incentives and financing. And what this 24/7 accounting framework that we're advocating is basically trying to make those things meet in the middle, right? Today we have a market based system that is too much focused on, I would say, flexibility, making it easy for people to say they're green. and so has led to very valid criticism. And what we're trying to do now is bring that market based mechanism back closer to the physical realities of the grid,

Chris Adams: Oh, I see.

Killian Daly: But keeping the, incentive system, because if you don't have that, then, well, I don't really see the point in even doing the exercise.

Chris Adams: Okay. So there's two things that I wanted to kind of just see if I could maybe dive into a little bit on that then. So it sounds like this whole notion of not having this stuff tied to each other is to reflect the fact that people have all these complicated ways to purchase power in the first place.

So in my world as a cloud, as like someone working as a cloud engineer, right, I might buy computing by the hour, but I might also buy it, in advance for three years, for example, for a lower price, and that, that provides a bit of stability for whoever's running my server, but this kind of, this is an example of me having multiple different ways of being able to buy something, and essentially, some of that unbundling there is actually trying to capture the fact that there is, there are all these complicated ways to arrange to pay for something, and this is one way that we can use to value some of the Flexibility and stuff you said before.

So for example, you spoke about you can't run something on solar power, right? But if you had a battery, you can capture that and then use a battery bit like a time machine to kind of run at night almost right so but therefore you're trying to but that's more expensive than just making some claims.

So you need to have some way to recognize the fact that it takes a battery and a bunch of extra smarts to run something at night from that. That's what you're trying to go for with that, right?

Killian Daly: Yeah, exactly. And again, basing things on how power markets contractual, they have ways of already have contracted with allocating power between generators and consumers. I think the biggest issue with unbundling, so, selling the energy attributes and the power to different people. Actually, I think what the fundamental problem is the lack of time matching and deliverability requirements. That's where unbundling has gone wrong. Because it's, it said, "we're going to take the green attribute from this energy in Norway, and we're going to allow it to be used at any time of year, anywhere in Europe."

That's insane. That's where it starts to get completely insane. I don't have any particular problem with you producing it in one hydro plant, and selling the power into a power pool. and then that being consumed in Norway in the same hour. That's literally how power markets work on a short term power market.

Everyone bids into a common pool. And why not just put the attributes into the same pool and well, they, all have the same properties anyway. So it doesn't make a difference. It's the only way you're ever going to have liquidity, right? so I don't see any fundamental issue with, that.

The fundamental issue is with the annual matching and the

Chris Adams: the physics beyond breaking point, essentially.

Killian Daly: And that's, I think that's why I'm bundling, it's got such a bad name, right? And I think that's actually been fair, but I do think that it's not that bundling around bundling or necessarily the issue is, kind of the

Chris Adams: like those three pillars you mentioned. Okay, gotcha. Thank you for indulging me as I went down that thing, because I didn't know the answer to that, and I've always been wondering. Okay, so, we spoke about this thing called EnergyTag. We've spoke a little bit about how it's supposed to work and how it's basically an improvement on some of the approaches before.

And, maybe we could talk a little bit about who's using it? Is anyone, adopting it? maybe we could go from there, because this sounds like a cool idea, but there are many, cool ideas. That no one is paying attention to. And I suspect that would be quite a demoralizing conversation if that was the case.

So, yeah, I mean, who's using this and where, are there any kind of big name adopters you might point people to or anything like that?

Killian Daly: Yeah, so, yeah, two of the leading ones that kind of come to mind immediately, obviously, especially for software folks like yourselves or Google and Microsoft, they have 24/7 clean energy targets by 2030. Basically, they're committing to buying clean power for every hour, their data centers are consuming electricity, everywhere in, in which they're operating.

So they're two of the most, I would say, advanced, ambitious, corporate climate commitments in terms of scope 2 electricity procurement, at least. And they're obviously two major buyers. And they've been signing some really interesting deals as well. So there's, gigawatts now already of these 24/7 or close to 24/7 PPAs signed, 80, 90 percent firmed, portfolios of renewables, and that's game changing, right?

that's something we've seen emerge in the last few years where traditionally, the way of buying renewables has been "I'm going to buy a solar contract, and I'm going to blend that into whatever I'm buying elsewhere." And that's fine, right? But it's only giving you maybe 20, 20 percent of your electricity on an annual basis.

Now, we're seeing new contract structures that are blending together. Solar, wind, batteries, and getting maybe 80, 90 percent like of a flattened,

Chris Adams: so that's what I mean by firmed then, so firmed is this idea that it's basically it's when you say, so if it's not firmed, it's like I'm gonna buy the same amount totally without thinking about when it's matched, but if it's firmed then I am trying to think, I'm taking the steps necessary so that I can make a much more credible claim that the power I'm using is coming from generation or from stored amounts of power or something like that.

Ah,

Killian Daly: And that's, as I said, there's gigawatts of deals done already to date. Are there people doing this hourly matching stuff? Yes, absolutely. Check out our website. There's 30 projects there, with millions of megawatt hours of hourly matching being done.

So, this is not 40 organizations or something doing it 5 continents. So, This is not rocket science, right? This is literally taking meter data. That's very common, hourly production and gen data. You could do it on an Excel file with three columns if you wanted, and matching those things together and seeing where we're at. So it's absolutely demonstrated and leaders are doing it. Is everyone doing this? Is this now the status quo way of doing it? No, absolutely not. And that's what we work every day to try change, right? so we're still, I would say, relatively in the early days of this transition, but, as far as I'm concerned, it's kind of inevitable for credibility reasons, transparency reasons also for pretty fundamental economic reasons. Companies going out there and committing to buy loads of energy that is unmatched to their consumption profile.

They're leaving themselves open to a lot of risks. So, what if you say, okay, I'm just going to buy a load of solar. That has no connection to how I actually consume electricity. You're leaving yourself open to a lot of volatility that we're seeing electricity markets today. A lot of super high prices in the evening.

For example, when you're, when your solar contract is not delivering you anything, then what do you do? Right? you have all this gas volatility and exposure. So it's not just about decarbonization. It's also about things like electricity price hedging. So there's kind of various, I think, fundamentals that mean that.

We are going to move in this direction.

Chris Adams: okay, so So if I understand that final point that you've basically made is if I want to do this kind of matched thing for example, or if I want to, if I want to be buying some power, one of the advantages of doing like a longer term deal is that there's a degree of stability. So let's say, I don't know, a one country decides to invade another country and then totally make gas prices go through the roof.

I'm somewhat insulated from all that stuff so that it's not gonna massively destroy, it's not gonna destroy the, make impossible to kind of pay my own bills, for example. And like we've seen those of examples of that over the last few years, for example. So there's a bit of insulation from that kind of stuff.

Yeah.

Killian Daly: Exactly. So now we do get into kind of contracting mechanisms here. It's a little bit similar to what basically, if you're committing to a fixed price, for example, for a number of years, if you sign like one of these PPAs and you commit, let's say, to a 10 year fixed price for power. And if you're committing to like a affirmed profile, let's say 90 percent matched,

that has a very significant hedging value. So it means that basically you fixed like a lot of your power price. So no matter what happens, if, there's a massive spike in gas prices and power prices go through the roof. You're protected against that. We actually worked on a really interesting study on this a couple of years back or 18 months ago that said.

With Pexapark, who are like PPA analysts, and they basically showed that like a 10 megawatt consumer in Germany could save over 10 million euro, in the best of cases, and at least millions of euro in a given year by signing these 24/7, or close to 24/7 power purchase agreements with clean electricity assets, because one thing that clean energy has as an advantage in an ever more uncertain world is that the costs are basically known up front. You know how much money you need to build a wind turbine to build a battery up front.

It's all capex heavy. And that means that renewables can basically Give you a fixed price up front where honestly, gas cannot, because, most of their costs are operational. It's about buying the gas when you need it to.

Chris Adams: And there's a constant flow is not

Okay, I guess with the sun, I mean, there's maybe a scenario where, I mean, it's not like there's a Mr burns style blackout of the sun kind of thing, right? if you're relying on something where no one has control over, no one can, kind of blockade the wind or blockade the sun.

That's where some of the stability is coming from, right?

Killian Daly: Yeah, exactly. Right. so you have those things, and you know that those fuel sources basically don't cost anything. Right? so you're all your costs are in construction, materials, all things you basically know, largely upfront, and that does enable you to provide long term contracts, typically way beyond the terms that fossil fuel generators can offer.

And so it can protect you for, the consumers willing to take that long term price risk. It can really offer really significant hedging benefits. not above alternatives.

Chris Adams: Buy that on like the spot market as it were or buying something just like on the regular market. Okay. All right. So, so you mentioned a few large companies doing that stuff and outside of technology, I know that

I think it's the federal government. They've, it sounds like you said one or two things, which are quite interesting.

There is this idea that 100 percent is obviously really, good. Right. And that's what you want to head towards. But given there are some places where aren't, they're not going, they're not shooting for 100 percent straight away, for example, they might be going for 50 percent or 60 percent or something like that.

This is something that is kind of okay to do, or that's okay to start at. Cause I think I heard about the government, the US government had a plan for something about this by 2030 or something.

Killian Daly: Yeah. So basically, what we, we started the conversation talking about accounting. So I think the first thing you need to do is get, the accounting right. So that when you say 50, it means 50 or when you say 100, it means 100 because if you're just saying 100 and it means 50, then well, you're screwed, right?

You have a bad system. So, I think, actually being at 70 percent renewable, but saying that out loud

Chris Adams: 70%. Yeah.

Killian Daly: and addressing the, the basic fact that you're only there that's much better than kind of saying I'm 100 percent renewable on some annualized basis and kind of like misleading people about where you're at with, decarbonization.

Chris Adams: So it's better to be a real 70 than a fake 100, basically, yeah?

Killian Daly: Yeah.

And, so, you have, electricity, like suppliers, for example, who are, there's like Good Energy in the UK, Octopus Energy in the UK, most of the electricity suppliers now in the UK, in fact, are, offering these like hourly tariffs.

And, I don't think any of

Chris Adams: it was only one or two that did that. Whoa. That's

Killian Daly: Now, I think this year it'll become more, a kind of a norm, where they will offer this alongside their a hundred percent renewable tariff. And none of those hourly tariffs are gonna start off being a hundred percent renewable, but it's bringing that extra bit of transparency, which I think is great.

And, the likes of good energy, they're already offering to thousands of customers, right? This is not just the Googles and the Microsofts who have their long term targets on this. This is already being offered to thousands of customers around the world because electricity suppliers are basically taking.

They're doing all the work. They're just giving the consumer the number on some dashboard saying, this is how much matching you have. if you look at the Octopus Energy example, it's quite interesting. They have a tariff called Electric Match for some of their B2B customers and they're actually basically reducing your price of power. when you're more matched, so that's quite cool, yeah, they're charging you less the more that your demand is matched to their generation. Right? And I think that's quite a cool gamification of this. They're saying basically try to consume when there's more wind and sun in the UK, you'll be more matched and we'll cut, we'll cut your rates because obviously it's sort of, it costs them less to deliver that in the first place.

So that's. That's the type of cool mechanism.

Chris Adams: So, I swear, every single time I speak to energy people, and they say, "oh yeah, the price is totally changing." Then I think of one level up, when we're like paying for cloud, and it's the same price all the time. Someone's making a bunch of money off us doing all the kind of carbon aware computing stuff, because if the price is going, low, I would expect to see those numbers go low.

This feels like something we might want to have a conversation about inside the tech industry then, if they are, if there's savings being made here, because it feels like it would be nice if those were passed on, I suppose. So, all right, let's speak, go on,

Killian Daly: I think just very importantly, of the, the more I think one fundamental truth that we're going to see,

it's already the case in some parts of the world, but this is going to be an essential truth of the transition. The more renewables you have, the more volatility you're going to have in power prices. And the more flexible you can be in your consumption. It is going to be very rewarding economically, if you can consume, at the times of day when there's loads of wind and sun, power prices are going to be very low and you're going to get rewarded for that. If you can't, if you can only be base load, then that is going to cost you.

Chris Adams: Ah, okay, alright. Okay. Alright, that's it, that's a useful thing to take into account. And so, we spoke before about, scope 2 and stuff like that, and you spoke about this idea that you're defining this standard. Now, EnergyTag is a standard in its own right, but, as I understand it, it's not like you're stepping outside of this.

You are still engaging with the protocols and all the stuff like that right now, yeah?

Killian Daly: Basically, so yeah,

EnergyTag is a nonprofit. we do a couple of different things. we're obviously focused on this area of electricity accounting, electricity markets and better green energy claims and all that. And so yeah one of the things that we do is we have a voluntary standard for hourly energy tracking because one of the kind of blocking points we have today, is that the way we do this tracking with these energy certificates, it tends to be on a monthly or even an annual basis globally.

And sometimes we don't have the information on the certificates to do this hourly matching. So we're trying to un debottleneck that particular technical issue. Think about how do we track through storage, like doing some novel things there. So we have a standard for that, but that's only one of the building blocks, I would say of this much larger question of, like, how do companies do electricity accounting or how do they do carbon accounting more generally? Our standard is there to work on that specific topic, but actually a lot if not most of what we do today is like working on policy advocacy around the world, working on global standards and basically advocating for those to change because ultimately it's the meta-levers, regulations,

standards. Once they change, then we're just there to help technically put that all together with some voluntary standards as long as they're needed.

But it's not our aim to be the world's next greenhouse gas protocol. That's really not in our wheelhouse. What we want to do is make sure that global standards and regulations are as good as possible.

Chris Adams: Oh, I see. Okay, so that, so if we go for a concrete example of this. So, in Europe, if you want to do a hydrogen project, which is, in some ways, kind of a bit like an AI project in that it's like a building that uses loads and loads and loads of power in one place, right?

Really dense. If you're going to make, green hydrogen, for example, you're taking water, adding loads of electricity to split that, and that's incredibly energy intensive. So you've probably want that, if you want the green hydrogen to be green, probably only use green energy. And one of the things you told me about before was, yes, we won that fight so that any, and if people want to get any of the subsidies from the government to kind of do this green energy thing, they need to have those three pillars style approach, right?

That's what, that's an example of your strategy, yeah?

Killian Daly: Yeah, so this is actually the reason I what really brought me into EnergyTag, it was a Greenhouse Gas Protocol thing, but basically are the key to one of the world's largest hydrogen producers. Right? And so I got put onto this topic a few years ago, which I found incredibly important and fascinating and, maybe not well enough understood.

It's like, when we're going to produce hydrogen using electricity, we need to really make sure that the electricity is squeaky clean, because of the efficiency issues and losses that you just inherently have with electrolysis. And so, just to give a quick example, Jesse Jenkins lab in Princeton University, a guy called Wilson Ricks, who is a rock star of power system modeling, they model this right?

And they show that in the US, if you basically use today's carbon accounting rules, this annual matching stuff, and you built out a hydrogen sector based on those rules, you will have hydrogen that is twice, maybe even three times as bad as today's fossil fuel hydrogen production. and you'd be calling it clean and subsidizing that production. Totally insane, just literally wasting money.

And so it's actually really, important. Billions of dollars of subsidy are going to go into hydrogen in Europe and in the United States. And so we worked a lot with NGOs, advanced companies and other partners to advocate for these strong requirements on green electricity sourcing for hydrogen, both in the US and also in Europe, and we won on both fronts, which has

Chris Adams: Oh, the US way as well!

Killian Daly: Yeah. Yeah. Yeah. Yeah. And it hasn't been, so both of those are legislation in

Chris Adams: place.

They're in! Yay science!

Killian Daly: Yeah, that's the legal way now to qualify for the tax credit in the US. In Europe, there's a phase in period on the hourly part to 2030. So, in 5 years or whatever.

But anyway, projects built now, they have to be designed to comply with that. And so,

Chris Adams: if you know,

it's going to be in the law of five, you're just going to make sure you

Killian Daly: going to start doing it now, right? more or less. yeah, so that's, yeah, obviously, this is kind of like hundreds of millions of tons of CO2 per year on the line between good and bad rules and that, that's kind of a concrete example of, why these things matter. Right? Like accounting sounds boring sometimes. I definitely thought it was boring before I realized like, "Oh my God, I'm working for a huge power consumer and this is changing everything." So yeah, it's definitely super, super important that we get this stuff right.

Right.

Chris Adams: Okay, so we spoke about, it sounds like you've done the work with Air Liquide and you've kind of essentially laid the groundwork to move from a fossil based hydrogen thing to hopefully a greener way of making hydrogen, which ends up being used in all these places. And now it seems like you've got the, okay, you said Google and Microsoft, same power usage as Air Liquide in a single year.

Maybe they might've changed, but back then, there's, so it looks like we're seeing some promising signs. For that over here. So maybe, I mean, if we, want to see that, what do we need to see at a policy level? Do you need to have, government saying, "if you want to have green energy for data centers, you need to be at least as good as the hydrogen, industry."

Is it something like that you need to do? Because what you've described for the hydrogen thing sounds awesome, but I'm not aware of that in the, kind of IT sector yet. That's something that I haven't seen people doing yet.

Killian Daly: That is also coming, right? So hydrogen has just been the first battleground or the first palce, I think.

Interestingly, actually, on the 14th of January, just before the inauguration of Donald Trump, as US president, so the Biden administration issued an executive order, which hasn't yet been rescinded.

Basically on data centers on federal lands and in that they do require these 3 pillars. So they do have a 3 pillar requirement on electricity sourcing, which is very interesting. Right? I think that's quite a good template. And I think, we definitely need to think about, okay, if you're going to start building loads of data centers in Ireland, for example, Ireland, 20 percent 25 percent of electricity consumption in Ireland is from data centers. That's way more than anywhere else in the world in relative terms. Yeah, there's a big conversation at the moment in Ireland about "okay, well, how do we make sure this is clean?" How do we think about

procurement requirements for building a new data center? That's a piece of legislation. That's being written at the moment. And how do we also require these data centers to do reporting of their emissions once they're operational? So, the Irish government, is also putting together a reporting framework for data centers and the energy agency.

So the Sustainable Energy Authority of Ireland, SEAI they published a report a couple of weeks ago saying, yeah, they, you know what, they need to do this hourly reporting based on contracts bought in Ireland. So I think we're seeing already promising signs of legislation coming down the road in other sectors outside of hydrogen.

And I think data centers is, probably an obvious one.

Chris Adams: So people are starting to win. Wow, I didn't realize that. I knew somewhat about that there was an executive order that there was a bit of buzz about. But I didn't realize that, set the precedent. So, yeah, we should do what that massive industry over there is doing because that's now the new baseline that, that's where the bar should be.

We should do that as well, basically.

Killian Daly: Exactly, because that those hydrogen rules, it's actually what it actually is. Well, actually, the whole debate was about is what is clean electricity procurement? What does that mean? What does it mean to use clean electricity? And that has been defined now in hydrogen rules and that can be copy and pasted to any large new load.

Well, if you want it to be clean, we already know the answer. It's in legislation,

Chris Adams: It's how to tell when energy is green,

Killian Daly: MIT, IEA, the who's who of energy experts have all modeled this and they've all found that this is the way to do it. So, there's a template there, right? And it's, if you're going to go against that, it, yeah, well, obviously, then you're, obviously sacrificing the integrity of your accounting schemes.

Chris Adams: Wow! That was, we spoke about how to tell when energy is green, and you've, We seem to be ending on a high, I didn't realise we'd actually got to that. That's really, awesome. You've really made my day for that, Killian. Thank you so much for coming on and diving into the minutiae of carbon accounting for electricity, but also ending it with a slightly less depressing piece of news, which I'll take in this current political climate,

Killian Daly: just to interject before I say goodbye, there's one, one really, it's good to end on a positive note, I suppose, in this mad world we live in.

There was a project announced recently. I think people should go check it out in the Middle East in UAE, where basically for the first time, they're going to deliver basically, around the clock solar power. So 1 gigawatt of solar, all night long because they're basically, building a massive battery and a huge solar farm, and basically all year round is going to deliver, green electricity at under 70 us dollars per megawatt hour, which is extremely competitive.

So, I think solar and storage, what they're going to do together is going to be, is going to change the world. Right? I really think that is going to happen faster than people think. They're going to start to kill gas. So, yeah, I think green energy economics, despite what politicians will want to do with their culture wars,

I think will at the end of the day, hopefully, answer some of the questions we're trying to solve here. So, yeah, thanks so much for having me on. It's been a real pleasure.

Chris Adams: Brilliant, thank you so much for that mate, and may the fossil age end. That's really, that's so, so cool to actually see that, I totally forgot about the Masdar thing, which is the city. Yeah, and we'll share a link to that so people can read about that, because if you care about, I don't know, continued existence on this planet, then yeah, it's probably one to, good one to read about.

Killian, this has been loads of fun, thanks a lot mate, and next time I'm in Brussels I'll let you know, and maybe we can catch up for, have a shoof or something like that. Take care

Killian Daly: Yeah. A hundred percent. Thanks. Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

Backstage: Impact Framework

13 februari 2025 | 28 min

This episode of Backstage focuses on the Impact Framework (IF), a pioneering tool designed to Model, Measure, siMulate, and Monitor the environmental impacts of software. By simplifying the process of calculating and sharing the carbon footprint of software, IF empowers developers to integrate sustainability into their workflows effortlessly. Recently achieving Graduated Project status within the Green Software Foundation, this framework has set a benchmark for sustainable practices in tech. Today, we’re joined by Navveen Balani, Srinivasan Rakhunathan, the project leads and Joseph Cook, the Head of R&D at GSF and Product Owner for Impact Framework, to discuss the journey of the project, its innovative features, and how it’s enabling developers and organizations to make meaningful contributions toward a greener future.

Learn more about our people:

Navveen Balani: LinkedIn
Srini Rakhunathan: LinkedIn
Joseph Cook: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Impact Framework | Green Software Foundation [00:00]
The SCI Open Ontology | Green Software Foundation [04:27]
SCI for AI - Addressing the challenges of measuring Artificial intelligence carbon emissions | Green Software Foundation [06:57]
SCI Guidance [12:07]
CarbonHack [13:03]
Impact Framework Github Page [17:58]
IF Explorer [20:18]
IF Community Google Group [23:42]

Events:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Chris Skipper: Hello, and welcome to Environment Variables, where we bring you the latest news from the world of sustainable software development. I'm the producer of this podcast, Chris Skipper, and today we're excited to bring you another episode of Backstage, where we peel back the curtain at the GSF and explore the stories, challenges and triumphs of the people shaping the future of green software. We're no longer gatekeeping what it takes to set new standards and norms for sustainability in tech.

This episode focuses on the Impact Framework, also known as IF, a pioneering tool designed to model, measure, simulate, and monitor the environmental impacts of software. By simplifying the process of calculating and sharing the carbon footprint of software, IF empowers developers to integrate sustainability into their workflows effortlessly.

Recently achieving graduated project status within the Green Software Foundation, this framework has set a benchmark for sustainable practices in tech. Today, we have audio snippets from Naveen Balani, Srinivasan Rakhunathan, the project leads. And Joseph Cook, the head of R&D at GSF and product owner for Impact Framework, to discuss the journey of the project, its innovative features, and how it's enabling developers and organizers to make meaningful contributions toward a greener future.

And before we dive in, here's a reminder that everything we talk about will be linked in the show notes below this episode. So without further ado, let's dive into the first question about the Impact Framework for Naveen Balani.

Naveen, the Impact Framework has been described as a tool to model, measure, simulate and monitor the environmental impacts of software.

Could you provide a brief overview how this works and the inspiration behind creating such a framework?

Navveen Balani: Thank you, Chris. And thanks to all the listeners for tuning in. Let's first understand the problem we're solving with the Impact Framework. Software runs the world, but its environmental impact is often invisible. Every CPU cycle, every page load, every API call, these all contribute to energy consumption, carbon emissions, and water usage.

Yet, without the right tools, measuring and managing this impact remains a challenge. This is where the Impact Framework comes in. It's an open source tool designed to transform raw system metrics like CPU usage or page views into tangible environmental insights, helping organizations take action. Built on a plugin based architecture, it allows users to integrate, customize, and extend measurement capabilities, ensuring scalability and adaptability.

More importantly, the Impact Framework helps realize the software carbon intensity specification, making sustainability reporting transparent, auditable, and verifiable. Every calculation, assumption, and methodology is documented in a manifest file, ensuring that impact assessments are replicable and open for collaboration.

At its core, the Impact Framework is built on a simple yet powerful idea. If we can observe it, we can measure its impact. And once we can measure it, we can drive real change, reducing emissions, optimizing resource use and building truly sustainable software.

Chris Skipper: What were some of the most significant technical or organizational challenges you faced during the development of the Impact Framework and how did you and the team overcome them?

Navveen Balani: The Impact Framework wasn't just built, it evolved. It was shaped by real world challenges. Lessons learned and the need for a scalable, transparent way to measure software's environmental footprint. The foundation of the Impact Framework was laid through previous projects and ideas. Starting with SCI Open Data, which tackled the lack of reliable emissions data, and SCI Guide, which helped organizations navigate different datasets and methodologies.

Another critical component was the SCI Open Ontology, which defines relationships between architecture components, establishing clear boundaries for calculating measurements. Alongside these foundational efforts, real world use cases from member organizations applying software carbon intensity measurement played a crucial role.

These practical implementations tested SCI in diverse environments, refining methodologies, and ensuring that SCI calculations were not just theoretical, but applicable and scalable across industries, but data alone wasn't enough. We needed to scale measurement across thousands of observations.

Sustainability assessments had to be continuous, automated, and seamlessly integrated into software development. This led to key innovations like aggregation, which enables organizations to condense vast amounts of data into meaningful, structured insights, rolling up emissions data across software components to provide a holistic system wide view.

Technology, however, was just one piece of the puzzle. Adoption was equally critical. To accelerate real world impact, we opened up the Impact Framework to our annual Carbon Hackathon event. Where teams worldwide build projects that pushed its capabilities. This was a turning point, validating its flexibility and refining it through community driven development.

At its core, the Impact Framework is built on transparency. Unlike black box solutions, every input, assumption, and calculation is fully recorded in a manifest file. Making assessments auditable and verifiable. This commitment to openness has been crucial in building trust and driving adoption.

Chris Skipper: Looking ahead, what are the next steps for the Impact Framework? Are there specific new features or partnerships on the roadmap that you're particularly excited about?

Navveen Balani: That's a great question, Chris. Looking ahead, the Impact Framework is entering an exciting new phase with a major focus on expanding measurement capabilities for AI. Right now, we're working on the SCI for AI specification. which extends software carbon intensity to both classical AI and generative AI workloads.

Measuring AI's environmental impact comes with a new level of complexity. AI isn't just another software workload. The environmental footprint varies significantly depending on whether you're training a model from scratch, fine tuning a large language model, or simply using an AI API like ChatGPT or Gemini.

Each scenario has different compute demands. Memory requirements and energy consumption patterns, making standardized measurement both challenging and essential. Through the Impact Framework, we aim to tackle this by developing new plugins and contributions that enable precise measurement of AI related energy use, hardware efficiency, and emissions across training, fine tuning, and inference workloads.

These capabilities will collectively evolve, through community participation with researchers, developers, and organizations, contributing to refining methodologies, expanding data sets, and ensuring that AI measurement remains transparent, auditable, and standardized. This collaborative approach will allow organizations to quantify, compare, and optimize their AI workloads.

Making sustainability a key consideration in AI deployment. Beyond AI, we are also exploring new partnerships to further enhance the Impact Framework's adaptability. Collaboration with cloud providers, software vendors, and sustainability researchers will be crucial in ensuring that the framework evolves alongside industry needs.

Our goal is to make environmental impact measurement not just an option, but a fundamental part of software and AI development at scale.

Chris Skipper: Moving on, we have some questions for Srini. Srini, IF emphasizes composability and the ability to create and use plugins. Could you explain how this innovative approach has enabled more accurate and flexible environmental impact calculations for different types of software environments?

Srini Rakhunathan: Absolutely. The Impact Framework's emphasis on composability and the use of plugins is actually a game changer for different environmental impact calculations. If you notice that the framework is highly modular, making and allowing users to create and integrate various plugins. What it means is you can tailor the framework to fit the specific needs of your software and it doesn't matter what type of software you have, whether it's cloud based, on prem or hybrid.

What is also advantageous is that the plugin ecosystem has a wide range of tasks. For example, it has something around data collection, it can do impact calculation, it can do reporting. It can do also very, very specific tasks like math functions and aggregation functions. What this means, you can mix and match plugins to create a mashed up pipeline that reflects your environment, whether you are running your software on web, cloud, mobile, doesn't really matter. As long as you know what your software boundaries are, you will be able to combine these plugins and create your own, um, pipeline, if you will. And that pipeline will help you, uh, create your calculation pipeline that can either run one time or run as a batch or, you know, run based on certain triggers.

What it also means, and if you notice, there is also manifest files, and we will be talking more about it later in this conversation, is that the manifest files ensures that you have a repeatable way of calculation. I mean, you mash up these different plugins and you create a pipeline and you embed it in a manifest file and it's repeatable.

So what I think is this framework's capability of composability and plug in can help you make very, very accurate impact calculations.

Chris Skipper: How have collaborations with organizations like Accenture and Microsoft, as well as the open source community, contributed to the success of the Impact Framework? Are there any standout moments or partnerships you'd like to highlight?

Srini Rakhunathan: Thanks, Chris. That's a great question. So the cornerstone of the success of Impact Framework has been collaborations. And this has been ongoing from the time this project was conceptualized. Bear in mind that when we, like Naveen, who's there also with us, and I, along with the Joseph and Asim started thinking about the project.

The initial vision of the project was very different. So we started off with something called SCI Guide, where we wanted to collate datasets across the open source community to help calculate emissions from software. And we built the SCI Guide and that transitioned into something called CarbonQL, which is a primitive version of what we see today in the Impact Framework, which is more like how do we make sure that it is easier for users or developers to calculate emissions from software and the learnings that Naveen, Joseph, I and Asim went through to come up with the initial version of Impact Framework and the amount of work that the team has put together to get it to graduation state is amazing and it speaks volumes about the collaborations that has gone ahead into the building of the tool.

One particular highlight I want to call out is every year, GSF organizes what they, what is called the CarbonHack. And in 2024, the CarbonHack focused on getting the open source community to come and build tools.

On top of Impact Framework, either extension of the tool or building content or newer areas where the Impact Framework can be used. And you would be amazed at the amount of contributions that came in and newer use cases that looked at calculating emissions, not just from carbon, but from water and other forms of renewable resources was also identified.

And that's great. That, I believe, was a standout moment for the tool.

Chris Skipper: The IF documentation highlights the use of a manifest file and a CLI tool to calculate environmental impacts. Could you walk us through how these tools work and how they lower the barriers for developers to adopt sustainable practices?

Srini Rakhunathan: Definitely, we can talk about both the CLI tool and the manifest file. These are actually cornerstone capabilities built within the Impact Framework, and they help us to calculate the environmental impacts. What happens is, the manifest file contains a list of of the software's infrastructure boundary encoded as YAML files.

It's in the standard YAML format, and it contains every bit of component that is part of the software, whether it's front end, middle tier, back end, database, API, everything encoded as what's the hardware used, what's the utilization, what's the telemetry involved. So much so that it can be used to give us an input to the Impact Framework CLI tool that calculates emissions.

The use of the file enables transparency and rerunability. That means it can allow anyone to re execute the manifest file and everyone will come up with the same calculations. The second piece that we spoke about, which is a CLI tool, it's a command line tool, which means it can be used to run on any environment.

It processes the manifest file and computes the environmental impacts. So the way it works is developers can pass the path to the manifest file to the CLI tool, and it'll take care of the calculations. The tool has capabilities to do phased execution and that allows efficient and flexible use of the framework.

Chris Skipper: And finally, what lessons have you learned from working on this project that might benefit other teams looking to build tools or frameworks for sustainability in tech?

Srini Rakhunathan: Thanks for asking this question. At an overall level, I would like to respond to this question by focusing on lessons learned from two aspects. The first is the execution model and the second will be the technical design. In the execution model space, this project is a good example of how open source collaboration works.

The team used GitHub extensively, and most of the meetings were asynchronous. And the engineers and the product managers and everyone who worked on the project worked through GitHub. And collaborated extensively using the open source tools available, which is a great model for scale. The second aspect we should look at from an execution model, and which is a success story here, is how the team used customer feedback as inputs to make the product better.

There were constant, if not many sessions with many customers with whom the team worked to engage with them and understand what the requirements are for building a tool that can help them calculate emissions and use that feedback into the process, into the backlog to make the tool better. The second aspect of lessons learned will be on technical design.

And here I would want to call out that. The whole concept of building a plugin ecosystem and make them composable such that it can, you have a, you know, you have a set of plugins that you deliver to the community, like a base framework, and then you allow extensibility. So that's a great model, which can help tools that can use sustainability as a calculation engine.

And then the second piece is, which is also equally important. As you do this. You also make sure that you have extensive and good documentation that can help anyone who's coming on board understand the framework and be able to get on board and run with building a new plugin as soon as possible. The IF code, the GitHub site, if you go there, You will have a link to the docs page.

And if you read through the docs, it's very, very self explanatory and will allow anyone who can come in and who's interested in building a plugin, do that at the fastest possible time. So these are, in my mind, lessons learned both from an execution model and the technical design aspect.

Chris Skipper: Moving on, we now have some questions for Joseph. Joseph, the Impact Framework recently achieved the status of a graduated project under the GSF. What does this milestone mean for the project, and what were some of the key factors that led to its graduation?

Joseph Cook: The Impact Framework graduation was a huge milestone because it represents the moment when the project is considered sufficiently mature that it no longer needs to be incubated and instead it can largely be handed over to the community. We consider the software to be feature rich and stable enough that people can integrate it into their systems, and in order to graduate, the project had to meet a quite stringent set of requirements, including demonstrating that Impact Framework had real world users, and that we had addressed community requests and bug reports, and that we had suitably comprehensive test coverage, and that the documentation and the onboarding materials were all fit for purpose.

Now that milestone has passed, development activity is going to be much more ad hoc and driven by the community, rather than following a development roadmap that's defined by Green Software Foundation. Our efforts at the GSF will now be in driving adoption instead.

Chris Skipper: How does the Impact Framework engage with the broader tech community to encourage adoption? Can you tell us what steps the GSF is taking to include the community as part of the IF development?

Joseph Cook: Impact Framework is used by all kinds of organizations, but it also has a thriving open source community. And most of the discussion with the community happens on GitHub, either through issues or on the discussion board. But we also have a Google group where we share updates and collect feedback. Open source development on Impact Framework is really fundamental.

It's really baked into the very core of the project. Instead of trying to ship Impact Framework with all the built in features to connect to thousands of different services and systems that people want to measure, we instead focused on making it really easy to build plugins, and then encouraged an open source community to develop, where people create their own plugins for all the features that they care about, and share them with each other on our Explorer website, which is like a free marketplace for Impact Framework plugins. This model actually makes the Impact Framework much more robust and much more stable because we have a much greater diversity of voices influencing what Impact Framework can do and what it can connect to. It decentralizes the development of the project without compromising the core software, and it also means that our small development team doesn't shoulder the burden of maintaining a huge code base with lots of different brittle connectors to third party APIs and services.

And going forward, we want to keep this community thriving and see thousands more Impact Framework plugins listed on the Explorer.

Chris Skipper: How do you see the Impact Framework setting new benchmarks for environmental responsibility in tech? Are there specific metrics or practices that you believe will influence industry standards?

Joseph Cook: Impact Framework is a lightweight piece of software for processing what we call manifest files. These are YAML files that follow a simple format that captures the architecture of the system that you're studying. All the observations that you've made about that system and all of the operations that are applied to your data.

I like to refer to these files as executable audits because they mean that you don't just report emissions numbers anymore, you actually show you're working too. And this enables the community to fork and modify your manifests and challenge you. And through iteration, you can come to crowdsourced consensus over your environmental reports. We would love to see this radical transparency become the gold standard for environmental impact reporting for software. Not only that, but manifests can be the basis for experimentation or forecasting, and help decision makers to assess the environmental benefits of implementing some change. Imagine you're challenged about why you chose some specific action.

Your manifests are your evidence. And we think this combination of transparency and reproducibility, composability, and openness is a unique selling point for Impact Framework, and it could transform the way projects and organizations report their emissions and introspect their own operations.

Chris Skipper: For listeners who are interested in getting involved with the Impact Framework, what are the ways they can contribute or support the project? Are there specific skills or areas where the community can make the most impact?

Joseph Cook: If you would like to get involved in Impact Framework, there are many ways to do so. If you're a developer, you can head to the GitHub, where we have plenty of open issues, including some specific good first issues to help people get started. If you want to build plug ins, then you can download our template and use that to bootstrap your way in, and then submit your plug in to the Explorer using a simple typeform on our website.

We always appreciate updates to the documentation too, and if you're interested in integrating Impact Framework into your systems, we'd You can always reach out to research at greensoftware. foundation to discuss it with us directly. We're always happy to help. If you just want to test the water or you have general questions about Impact Framework, you can start discussions on our GitHub discussion board or communicate via our Google group, [email protected].

Chris Skipper: Awesome. So I'd like to thank Naveen, Srini, and Joseph for their contributions to this episode. Before we finish off this episode, I have a few events that need announcing.

Starting us off, we have an event that will be happening today, the date of the publication of this episode, February the 13th, 2025 at 5 p.m. CET in Utrecht, Netherlands. Any Netherlands based listeners, you're invited to a Green Software Community Meetup today from 5pm until 8pm at Werkspoorkathedraal. Join us for a free in person event to kickstart a more sustainable year in tech. You'll hear insightful talks about reducing your software's energy footprint, scaling down for greener computing and building a grassroots digital sustainable movement. This is a great opportunity to connect with like-minded professionals, share ideas, and be part of a growing Dutch community that's dedicated to building a greener tech future. Food and drinks are provided free of charge.

Next up is an event in Brighton in the UK, happening on February the 19th from 6:00 PM to 8:00 PM at Runway East, which features Senior Digital and Sustainability Manager for OVO, Mark Buss, speaking about the challenges with advocating for digital sustainability within his company. The talk will also be live streamed, so we will have a link in the show notes below for that.

Next up for any Spanish listeners, we have the first ever meetup of the Green Software Community in Spain that will be happening online at 6pm On February the 20th, Dia Zero, Comunidad, Meetup, Green Software Foundation, España will be a chance for you to discuss how to collaborate with other people passionate about climate change and green software. And we'll have a link to that in the show notes below too.

Next up down under in Australia on February the 20th at 6pm AEDT in Melbourne, we have Digging Deeper into Digital Sustainability. How to design and build tech solutions. This will be happening at ChargeFox. Katherine Buzza will be talking about the impact that software is having on the world's carbon emissions, and how to align your career in tech with the decarbonized future we can all play a role in creating.

Next up, another UK event on February the 27th at 6pm GMT in London. Practical Advice for Responsible AI will be held in person at the Adaptivist offices. Talks about Green AI with Charles Humble and AI Governance with Jovita Tam. Click the link below to find out more.

And finally, on our events list, we have GSF Oslo will be having its February meetup on the 27th of February at 5pm in person at the Accenture offices from 5 until 8pm. Come along to find out how leveraging data and technology can drive sustainability initiatives and enhance security measures and dive into green AI. Talks from Abhishek Dewangan and Johnny Mauland. Details in the podcast notes below.

So that's the end of this episode about the Impact Framework project at the GSF. I hope you enjoyed the podcast. To listen to more podcasts about the Green Software Foundation, please visit podcast.greensoftware.foundation, and we'll see you on the next episode. Bye for now!

Backstage: Carbon Aware SDK

23 januari 2025 | 13 min

Deep Green Technologies

16 januari 2025 | 45 min

In this episode of Environment Variables, host Chris Adams sits down with Mark Bjornsgaard of Deep Green to explore a transformative approach to data center design and sustainability. Mark shares insights into how Deep Green reimagines traditional data centers by co-locating them in urban areas to provide heat reuse for facilities like swimming pools, district heating systems, and industrial processes. They discuss the challenges of planning and policy, the rise of high-density computing driven by AI, and the potential for data centers to become integral components of community infrastructure. Tune in to learn about the intersection of digital innovation and environmental responsibility, and how new business models can turn waste into opportunity.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Mark Bjornsgaard: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Mark Bjornsgaard on LinkedIn: Dell's OCP Solutions Propel AI Innovation [07:52]
Civo [37:31]
Real Time Cloud | GSF

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Mark Bjornsgaard: The government does need to legislate. There is just not enough structure and there's not enough impetus for people to do the right thing. But the also, and particularly in the UK, what the government needs to do is planning is a huge, huge hurdle. I never really understood that until we'd be working with Deep Green for, you know, building data centers.

It is breathtaking how Kafka-esque the planning system in the UK is. It's just,

It's beyond insane.

I'm your host, Chris Adams. Okay, Mark, a few years back, when people were asked what a data center was, if they knew what one was at all, they might talk about some kind of thing, room, cupboard full of a few machines, maybe in a rack inside a unused room inside a building, for example. But these days, in the 2020s, people are more likely to talk about a warehouse full of hyperscale kind of data servers in a building, which is maybe the size of a football field or larger, for example, the kind of things that are run by massive firms like Google, Microsoft and Amazon, for example.

Now, as I understand it, you work with data centers, too, but they can take a rather different shape and interact rather differently with the built environment. So for those who've never heard of Deep Green, or how the stuff you're doing is different, give a kind of brief introduction to like how your approach to like building data centers is and how that has an impact on how it works with the surrounding area, for example, communities.

Mark Bjornsgaard: Yeah. So, as you say, most data centers are built in the middle of nowhere, and the vast majority are built without heat reuse. So the vast majority simply eject the heat that comes out of the computers. Data centers, we know, two to four percent of the world's electricity supply, and computers themselves are incredibly efficient electric heaters.

So 97 percent of the electrons that go into a computer come out as heat. So you've got us as a species, us in a climate emergency, taking two to four percent of the world's electricity supply, converting it into heat, and then ejecting it into the atmosphere, which 10 years ago, that might have sounded kind of plausible or even sort of necessary.

But in a world, as I said, in a climate emergency, that doesn't look so clever. So the difference between Deep Green and every other data center, most other data centers is we are building the data center where the heat can be reused. So very hard to transport heat, but relatively easy to transport electrons to take the data center to where the heat's required.

So that's what we do. We build smaller data centers, co-locate them where heat's required. Now that might be a laundry, it might be a distillery, it might be food production, it might be antibiotic production, it might be a swimming pool, but more often than not, it's what's called a district heating system.

So these large centralized heat networks that through super insulated pipes supply heat to large areas of different cities. That sort of principle, that district heat systems and heat networks, we're not very good at them in the UK specifically, but we are, the government is certainly planning for us to get a lot best them in the years to come.

So, that's where we're anchored. We, you don't build them in the middle of nowhere, you build them where they're required. There's a further, there's a further caveat and a sort of, a kind of context to this, I suppose, if you'd like. Up until the point where AI started to become part of our everyday lives, those normal data centers aren't on very much.

They're only on 20, 30 percent of the time, and they don't actually generate very good waste heat. So you can certainly forgive the great, the good of the data center industry for not necessarily trying too hard to reuse heat in the old world. But in the world that's coming where we've got these incredibly dense racks of NVIDIA and other chips, where, you know, she utilising a massive, huge amount more energy than previously the datasets had.

That, it's at this point where those are on 70, 80 percent of the time, and they're generating an enormous amount of heat, and the heat's relatively high grade. It's not high grade heat as class within, but it's good low grade heat. So at this point, then the ability to reuse heat becomes a real thing. And that's why we exist.

Chris Adams: Ah, I see. Okay, so there's a couple of things I'd like to unpack if I may. So the first thing you said was, okay, so there used to be data centers if they were going to be built in a kind of hyperscale thing. You're looking for kind of cheap land and then that's why they're often kind of miles away and probably maybe near things like say a grid connection or fiber connection or something like that, all right?

So that was like one of the previous approaches, but the downside of that is that, well, you've, you might have all this heat, but no one's able to use it, so you just vent it into the sky, so it's basically wasted in that way. So the other, another way you could do this is you can actually build these, where they kind of interact more, where they're kind of more complementary to the kind of urban fabric, as it were, and then you can use that.

But the thing that we've seen, one of the reasons that's been stopping that before is that essentially the data centers might have generated some heat, but it wasn't enough heat. So, you said low grade, and when you talk about low grade heat, that's like maybe 40 degrees, 50 degrees? Like, maybe you could expand on that, what that might mean, because I think for people who've never heard of the world of heat reuse, they don't know what high grade heat or low grade heat might be or what some of these uses might be, for example.

Mark Bjornsgaard: Yes. Yeah. No. It's so as you say, low grade heat in industrial settings can be as high as a couple of hundred degrees. So when you say a data center is going to be producing heat at 45, 50, 55 degrees, then that doesn't sound very warm at all. That said, 30 percent of all of the economy, 30 percent of all of the industry can use that very low grade heat.

So for example, a swimming pool very reliably loses a degree of temperature every hour. And it only needs to be 30 degrees. So if you've got, if you're trying to push heat from a, from one side of heat exchanger into another, if you've got kind of pool temperature water at 25 degrees, one side of that's the heat exchanger, and you've got, you know, our heat at 55, the other side, then heat flows the right way.

When it comes to district heating systems and heat networks, the old ones, actually, again, they weren't very, it was quite difficult to plug data centers into them because those old heat networks were quite high heat. They needed heat at 80, 90 degrees. So if you were a data center and you said, I'll give you heat at 35 degrees, it really wasn't that useful. Now, fifth generation district heating systems, the ones that we're building in the UK and the ones that are beginning to be built elsewhere in the world, they can use very much lower temperature heat because the buildings themselves are better insulated. So the whole, the kind of what we think of as ecology, industrial ecology, the kind of ecology starts to, to make sense because lots more offtakers can use this relatively low grade heat,

Chris Adams: Ah, I see. And you also said one other thing about, this is kind of one of the kind of flip sides of massively more dense compute. Here's one thing we've spoken about before. People talk about, okay, there is like worry about data centers, basically, or like AI data centers being massively more dense.

Like the examples, I think I saw you share a link on LinkedIn, which kind of blew my mind. Like, some of these new racks from Dell can have like half a megawatt of

Mark Bjornsgaard: half a megawatt per rack.

Chris Adams: and like, I couldn't really kind of picture what that was. I know it's about 30, it's around 30 times minimum, or around 30, more than 30 times what you might have for an enterprise data center rack.

So like, that's quite a lot of energy there. But like, can you maybe just like, what does half a megawatt even look like for most people, because it's really hard to

Mark Bjornsgaard: it's really, yeah, it is, it's really, it's sort of so vague, it's very hard to get your head around, isn't it? So, I always like to think of it in terms of your boiler on your wall at home. So that's going to be about 10 to 20 kilowatts, right? Your boiler at home. So that one Dell rack is, produces 50 times the amount of heat on the basis that on the basis that 97 percent of the electrons that go into it come out as heat.

That 500 kilowatt rack is producing anywhere between 30, 40, 50 times more heat than the boiler on the wall of your house. And so, an unfathomable, you know, amount of kind of heat. Then if you look at it in the context of a normal data center, if you go into a conventional data center now, you might have rack densities of between 7 and 12 kilowatts a rack.

So when you're talking about densities of again, kind of, you know, 20, 30 times. the density of compute in a single space. Now for us, we love that because we have the opposite problem of every other data center. We're space constrained, not power constrained. So if we can go to a swimming pool and we can heat a very large swimming pool with only two racks of gear, like a megawatt of, that for us is amazing because we spend much less money on building a data center, fencing, security, containers, all the other gubbins, fire suppressant systems, all the other gubbins that you'd have around a data center, when you compress them and you squidge them down, you make them much easier to deploy in the fabric of our communities and society. And then you get this really crazy kind of stats where I was in a data center in Sacramento, a couple of weeks ago, and you got this massive data hall,

it's meant to be one and a half megawatts. It is one and a half megawatts of power, but the whole hall is empty. There are just three or four racks just at the end of the hall because those racks are 130 kilowatts a rack. And so they've built a data center. The physical shell of the data center is built for those rack densities, but they don't need all of that space.

So actually what's going on at the moment in the data center industry is we believe is this sort of giant misallocation of capital where people are building data centers in the old way, when they actually should be building them for the world that's emerging, which is this really high dense, these rack densities that look nothing like conventional data centers.

Chris Adams: So you, okay, that's interesting, and I'd like to come back to some of the things you said there about what the implications of massively more dense compute might actually be. But you also said a few things interesting about this idea of saying, you know, community involvement and things like that.

Because one thing that I've never heard anyone else talk about in the data center industry or even the kind of like tech IT industry talk about was this idea of a, borrowing the idea of a social license to operate. This is an idea that people talk about in say fossil fuels and oil majors and stuff like that.

And you said, well, this is one way that we can actually essentially keep that social license to operate by actually offering a much, much more kind of equitable deal with the communities we're kind of trying to integrate with rather than having this kind of like standoffish approach. Maybe you could like talk a little bit more about that, because I don't really hear people saying that much about data centers.

They usually say, "well, you should be grateful because without us, you wouldn't have your cat pics without and and and..." It does feel like it's kind of missing a huge power of why people might push back against data centers or why they even talk about why they, you know, whatever the deal is when someone comes in and says, "Hey, can we build a bunch of digital infrastructure in your part of the world," for example.

Mark Bjornsgaard: Yeah, I mean, as you say, we talk a lot about a social license to operate because, and we believe that in the future, you will get more and more pushback from communities around having data centers in their backyard,

because you've got these huge sheds which are hogging and clogging transmission grids.

So these transmission grids to be built by public money and then their commercial enterprise, yeah, dumps down there and says, "well, I want 100 megawatts" and then suddenly you realize that half the streets in the area can't put in heat pumps because there's no more grid capacity in the substages or they can't have electric cars. So, we think that social license to operate will be increasingly important in the future. No doubt. But the also the other, I guess the other on the other sort of flip side of this is that datacenters don't really employ anyone, right? I think the datacenter industry is a bit naughty when it says, "oh, you know, we're going to build a datacenter, we're going to employ 4,000 people."

It's like, that's actually not true. You might employ 4,000 people while it's being built, but the reality is once a datacenter is up and running, the number of people who have to be employed in the actual vicinity are very low. But if you build a data center and then you say "I'm going to reuse the heat with a aquaculture park or a distillery or a laundry," suddenly then you then produce genuine net new jobs in a local area.

So not only is the kind of the environmental bit of the social license talk very important, we think increasingly data centers are going to be looked on as having to be good citizens in terms of, you know, employment and doing the right thing with the community and we've already seen a lot of this, right?

We've had moratoriums on data centers in the Netherlands and in Ireland and Singapore. We think we're in this sort of grace period in the transition. In the next 3 to 5 years electrons, then the amount, the number of electrons are going to become very constrained. We're not actually yet in the bottleneck, but in the next three to five years, we're going to start going to that period of time where they just genuinely are not enough electrons to go around.

And we are going to have to make genuine choices about what we do with scarce electrons. And at that point, we believe, that if you're a data center and you're not doing the right thing, then, you know, you're the very least your operations going to be severely curtailed. Stroke, you're going to be in the midst of a full scale culture war, which you just don't want to go anywhere near.

Right?

Chris Adams: Okay, so you said a couple of things which I think might be worth exploring or kind of diving into there because a one of the key things I think I'm getting from you is that, yes, you might be able to kind of force some changes through quickly or you might say like, okay, well, I think one of the key things is that we need this transition itself to be sustainable and if you are able to kind of maybe push through some changes now you'll end up with so much pushback that you won't be able to sustain that state of changing as we end up like essentially moving away from fossil fuels a society based on electrification in many cases.

Mark Bjornsgaard: That's exactly. Yeah, exactly. So, yeah, I think what we see is that we see that. We are energy and software folk and we're venture capitalists by trade. We see, we don't see the data center industry as a, we don't take it as sort of face value. What we see is 70 percent of the UK's total energy budget being the heating of spaces.

So what,

we're looking at from the other end of the telescope, we're saying, well, how could we, how can we best, what's the fastest, quickest way of heating all our shops and offices and factories? And the reality is, the quickest, fastest way of doing that is using computers as electric heaters.

The fact that they happen to be there as data centers is almost, you know, that's kind of just a happy circumstance for us. We're solving what we see as a, as the meta problem, if you like. And just seeing what tools and capabilities we have to be able to solve that problem.

Chris Adams: Okay all right so this is actually one thing that you...

Because I think this is the thing that some of us forget about when we just think about IT like okay there's other transition, other changes that need to take place and before we, before you came on to this, I remember I saw you did a talk about these kind of for the wicked problems related to climate.

And I wonder if you might get a kind of maybe kind of expand on some of that because I think it's quite a useful context to help people who are thinking about their role as a technologist. But, okay, like, why would you even care about heat reuse, and why would you care about anything other than just the efficiency of your code directly, rather than this kind of wider, more systemic view, for example?

Mark Bjornsgaard: Yeah. Of course, we are. We all see our worlds in kind of what's in front of us, and that's completely understandable. As you say, we frame heat reuse and the electrification of heat, as you say, in context of what we think of as four wicked problems. So and these wicked problems make out make up roughly about 50 percent of the entire transition.

So if we solve these four problems, then we will be somewhere around 50% of the challenge of the transition take place and those problems are the heating of, of spaces, so all of our homes and offices, the industrial use of heat, so all industrial processes need to be de decarbonized and kind of electrified, and then we think of, controlled environment agriculture

and what's going on with how we grow stuff, the sustainability movement is rapidly kind of moot, sort of casting its eye across agriculture is realizing that actually how we feed 8 billion people on this planet is actually kind of some like 70 to 80 percent of all of our food is intensively farmed and based on fossil fuels.

And then the fourth wicked problem is carbon sequestration. So how do you, actually sequester carbon out of the atmosphere? That is also a problem around heating. If you take those four wicked problems, they can all be somewhat or completely solved with data center heat, with low grade heat on it. And so we're sitting there saying, well, look, if those datacenters are going to be built anyway, if we already need to spend between 10 and 20 percent of our entire

electricity budget for our country on data centers, then all logic says you build those data centers where you can use the electron twice. The electron can do its funky thing in the data center. We can have all that utility. And then so long as you've done in the right way, like we're doing it, you can just pass on 97 percent of that electron in the form of heat for it to then be used in those four wicked problems. So to us, that is, there's sort of a beautiful, immutable logic there, particularly in a world where you haven't got enough electrons. If you had bountiful, you know, fusion, fission, whichever the good nuclear bit is, if you had a bountiful electricity supply, then you might not be that bothered.

But the reality is in the next 10, 20 years, we're going to be so constrained by the amount of electricity that we have, we're going to have to get really good at being as efficient as we can.

Chris Adams: And I suppose it's actually, I mean, in the I mean, I'm calling you from in Germany, where most of our, almost all of our heating is still coming from basically combustion, burning like gas and stuff like that, for example, which is expensive. And even when you look at the UK gas again is one of the, what was the, I think it's the largest source of heating in the UK by quite a long way.

And these are two things which are volatile and where you're exposed to all kinds of changes in prices and things like that. And this is one thing that we probably do need to move away from. So that seems to be one thing like you're kind of, this is one of the approaches that you're looking at doing here, I suppose.

This is one thing I should ask you about then, because we spoke a little bit about this being a thing that we, that is valued and this is like a shift in how the role that digital infrastructure plays in kind of like the wider societal role. We've also spoken about in the UK, there is this goal to get entirely off, essentially have like some as close to as possible as a fossil free grid by 2030, which basically mean getting rid of a bunch of this heating from burning fossil fuels, right? Now that's a really ambitious goal. And like, as an as someone who grew up in London or grew up in the UK, I'm like, "wow, this is really cool."

This is like, I'm really impressed by that kind of ambition. And it's also one thing we've seen where a number of larger providers have basically said, "well this 2030 goal, it was a nice idea, but the moon has moved," to quote president having Brad Smith at Microsoft saying, "Oh, yeah, we were not pushing for 2030 anymore."

And I kind of feel like if there is this goal of 2030 in the UK, for example, and we have very similar goals in other parts of the world. Like what needs to happen at policy level to actually make this possible for the actual data center or the kind of digital infrastructure there because right now, I'm not aware of the kind of support or how policy kind of values this kind of different way of thinking about the role that digital infrastructure plays.

But we have seen with new government, basically in the UK, they do seem to be very keen on having a massive rollout of infrastructure. So. what's the deal here? Is it gonna be, how do we make, how do we square this circle basically?

Mark Bjornsgaard: It's not,

the declaration of data centers as critical infrastructure isn't quite as good news as it looks. So the, so that is that predicated on regulatory capture and if you declare data centers as critical infrastructure, you can then basically run ride roughshod over any local objections.

So the fact that the labor government announced that isn't necessarily a good thing. It's probably the opposite. In Europe, we've got the EED, we've got the European Energy Directive, I think it is, and by an

Energy Efficiency Directive, which is, which effectively says that certainly in Germany by 2028, you won't be able to build a new data center without reusing 20 percent of the heat. So there is a, there is already a, some sort of regulatory framework out there that's saying "you've got to do the right thing.

You've got to have, you've got to use green electrons. You've got to reuse the heat." So that's good. The reality is, as we all know, governments probably have to use carrot and stick. So you probably have to do a little bit more stick and a little bit more carrot. Those people who are being good citizens and reusing heat should get some brown points and should get some economic benefit from that.

And those who aren't, increasingly should be penalised. I mean, now you'd expect us to say that because obviously we're on what we think of as the right side of history. So I think the short answer is the government does need to legislate. There is just not enough structure and there's not enough impetus for people to do the right thing. But the also, particularly in the UK, what the government needs to do is planning is a huge, huge hurdle.

I never really understood that until we'd be working with DeDeep Greenor, you know, building data centers. It is breathtaking how Kafka-esque the planning system in the UK is. It's just,

it's beyond insane. it's crazy. So you've got regulations like, because you're leased of a council on a district heating system means that you only got that lease because you said you'd use green energy.

If you put a data center within the environment of your district heating system, because we've got generators that kick in if, you know, for redundancy and resiliency, that then means that you're in contravention of your lease. So instead of somebody just going, "yeah, that's a shit idea, let's not do that. Put across through that. That's an unfathomably complicated year long process."

We've had to put one pool we're trying to qualify, we've had to resubmit planning seven times. So this is just, I mean it's beyond rank stupidity, it's just a madness in this country, in the UK at least, around, we hate success in this country. We just hate success. This will be the third business that we develop in the UK and then scale in the US because in this country it is, yeah, we just can't get out of our own way.

It's really sad. And, you know, everyone says, "oh, we'll try and change." It's like, it's very simple. It's like, you either want people to do this or you don't. Do you know what I mean? Like no amount of meetings or nice coffees or platitudes or strongly worded emails. Do you know what I mean? Like, it's very fucking simple.

Can I build a data center or not? If I can't, then I can't. You know what I mean? Like it is, yeah. So this country is, it's very difficult to do here. And I suspect in a lot of Europe it is. So we need government to get out of its own way and clear a path for us.

Chris Adams: So you said a couple of things that I think maybe we could just go into a bit more detail before we move on from there. Because you said one of the things was, things like the, there is one like regulation, the energy efficiency directive, which is It's one of the ideally one of the drivers of transparency for organization for people operating digital infrastructure, like they'll, you know, as a result, you know, for you to comply with this, you need to be able to listen information like the carbon intensity of the power, how much your, you know, how clean the power is, for example, how much of it is coming from, say, fossil fuels, how much water you're using and things like this.

And presumably, like, these are some of the metrics that you might be able to kind of look good on, as it were, or this kind of way of building infrastructure might look a bit better, for example, like, if you're reusing some of the heat, I suppose, then does that have an implication on maybe how much water might be used, for example, and things like that?

Mark Bjornsgaard: Yes. And you've got to be very careful that it's not whack a mole that you don't, you know, you don't drop your PUE, but then you raise your, so you use evaporative cooling, you might drop your PUE or your, the energy use, you know, the Power Utilization Effectiveness of your data center, but then you massively increase the amount of water you use.

So there is a balance. There is a balance to be struck across all of these metrics. That's why there isn't one perfect kind of measure, if you like. Certainly in our case, we don't use any water, so the way that we cool, the way that the direct chip cooling and, the types of cooling we use, we don't use any water and, you know, there really isn't, as far as I understand, and I'm not an expert in terms of a techie expert in this area, but, really using water is a question of just how much margin you're prepared to sacrifice, you know, it is perfectly possible to cool the data center without using any water.

It's just you make a small amount more money on each data center if you use water and people again, the great and good of the data center industry are always be good environmental citizens. They could choose to use no water and just spend a little bit and make a little bit less money. Okay.

Chris Adams: You, ah, so you said something quite interesting there about how So you're using essentially liquid cooling as one way we can, as I understand it, liquid cooling in cars is way more efficient than air cooling in cars, which is why we've moved over. Presumably it's the same kind of idea here. So that's, that would result in a more efficient system that you'd be looking at using here.

Okay. And that, okay. That helps me understand how that might actually fit into heating a swimming pool or something like that. So if you've got an efficient way to move the heat from one place. to another place and like the whole point about you know people use water for heat storage and stuff like that it makes total sense I can see why you'd have like a nice chunky kind of like sink I suppose and if you if these are the things that you're doing then I suppose there's a chance to be more transparent, I suppose, with the kind of figures you're using for this.

So this might be, okay, that's, okay, that's interesting. All right, so if I could, I'd like to ask you a little bit about this AI question, because the approach you're describing here, of having lots and lots of distributed, having series of smaller data centers, like, built into the kind of fabric around us, that seems quite a bit different to the massive, centralized, gigascale data center, kind of paradigm that people talk about so I want to ask like if this is, I've always assumed that you need to have massive centralized data centers to do some of the kind of. AI workload stuff because you need to have these things network with each other. The way you're describing it sounds like that might not be the case.

You know, the things not being in the same building might not be the showstopper that people initially thought it was. Could you maybe talk a little bit about this? Because this suggests like a kind of post cloud way of thinking about computing, for example. And I want to ask, like, do you actually need a data, a mega cluster?

Or is there a, an alternative that you're suggesting here?

Mark Bjornsgaard: The truth is at the moment you need the mega clusters. So we, when we think of training large language models, those need to be done at the moment, those mega clusters need generally need to be all in one place. The trouble is, as data centers grow bigger and bigger, and as you build gigawatt data center campuses, and even larger, when we get, when we think of the trillion dollar cluster, the amount of compute we're going to need to, kind of enable artificial general intelligence, I think we're going to need something like 100 gigawatts of power, right?

100 gigawatt data center, which is, now, when you build, start to build data centers in these sizes, You Actually start to have a distributed problem anyway because you physically can't each sort of node running a version of the model has, it's so far away from the other node. You've got a distribution problem almost by default by size.

If that make if that makes any sense. So we've certainly got to be better at networking the architectures around large language models. And, there isn't very much academic research on this, there is a bit. We're doing a lot of work with NVIDIA and Nokia around this. The Chinese, we think, are doing a lot more work around this than other people, which is in itself interesting as we see a race to AGI emerging. So certainly the networking between data centers is going to become increasingly important. See, in the last six months, you've seen Microsoft spending billions laying massive fiber pipes between its AI data centers because it's trying to use these, you know, even 100 megawatt data center needs to be kind of physically clustered with another 100 megawatt data centers.

But that's also all in the world of training. Now, of course, when that, and that's where the models are learning, and that's great, and that's going to go on. The world that will emerge is obviously mostly going to be inference. So when you think of a world of AI in 10 years time, actually 90 percent is going to be inference, 10 percent is going to be training.

So we are, at DeepGreen, we're not necessarily trying to win the large language model, massive cluster game. What we're building is, the compute substrate for the future, where there will need to be thousands of megawatts of smaller data centers, smaller cluster sizes, much closer to where we all live and work.

So we're, this substrate, this compute substrate will be required in the future.

Chris Adams: Okay. All right. So, so basically, what I think you're saying, or what I'm kind of taking away from that, is that it was almost like a typology of different kinds of digital infrastructure that you might think about. So rather than just being one model, which is inherently better than the other, you probably would need to have different setups, depending on the different kinds of roles that you might actually be having.

And it's, you can kind of see people talking a little bit about this with the whole idea of like edge computing, but it sounds like for certain things you do need, you may, there may be a world where you do have big box Walmart-style out of town data centers doing certain things because, and you just, and you may have to accept that there's, you're not able to use some of the waste heat or you may need to like co locate things to use that and like have some kind of clusters and I guess China's, you can see some examples of people co-locating energy generation with industry and things like that.

But then there's this other kind of like other end of the scale, which is a more distributed thing. And that's something that you're looking, that you're looking at building, like, the kind of data centers that might actually integrate with, say, cities and things where they're closer to where it's actually being used.

But the, you're trying to go for a more kind of integrated approach by making as many of the outputs, the waste outputs, something that can be reused by other people for example because presumably there's a cost to like heating a swimming pool like it's non zero if you need to do that and if you've got the heat coming from what you're using then that's something economically benefit that's something that you might write into like currency benefits agreements and things like that.

Mark Bjornsgaard: Yeah. If you think about some of the inference work use cases that are already emerging, whether that's, you know, you integrating, you interfacing or chatting, maybe your kids are talking to a chatbot and they're trying to learn about they've got some visualization, some rendering visualization, which takes a lot of GPU compute.

That will be, those GPUs will be, it is better that they are co-located, or they're located somewhere closer to where the user is, particularly in the US, where they'll see, or other countries, and not just the US, but, you know, across Europe and other large continents, large land masses, you want the compute to be physically closer to people.

So, you know, where they're living and working. So that that is very important. But of course that world is just emerging. So at the, but that said, there are already a, there's already a lot of refining training. There's already a lot of people who are taking the outputs of the very large language models and then applying their own data to them and then refining, training them.

And then there's a whole bunch of other use cases around medical science and fluid dynamics and all the other stuff that the robots are gonna do for us. That world is now, as we know, emerging fast. That's the world that we're really building for smaller compute clusters, much closer to where people live and work.

And then, as you say, then you start to change the economics about how society works. You know, in the UK, we're spending 1.5 billion pounds heating our swimming pools every year. Really, we shouldn't be spending anywhere near that. Because those, pools should be being heated by recaptured heat. If we allow ourselves to build the data center infrastructure in the right way, the interesting thing about the UK particularly and other countries is that there's lots of fiber in the ground.

So when we first started building a data center, we talked about them following the fiber. Now, data centers don't really need to do that. There's plenty of fiber around. You can pretty much build a data center wherever you like. Now you have to, now people are saying they're following the heat, sorry, the power, but the third generation, the third phase of data center development, we see is people following the heat.

So first of all, you went to where the fiber is, then you went to where the power is. that's the era we're in now, but very quickly you're then now going to build data centers where the heat's required.

Chris Adams: i see where there's presumably like someone who like an offtaker who would use that and then be in favor of something being set up in their neighborhood or in as part of their project, they're getting a bit set up. Okay, so you said one thing that was, I think, quite interesting from there about, okay, there's loads of fiber, there's more fiber than we thought, like all this kind of dark fiber from 20 years ago, the last boom and bust, there's people might reuse some of that.

And some of this has, this could feel a little bit kind of academic or maybe not, it might feel a little bit like, "okay, what's happening in the future?" But As I understand it, some of this stuff is like, what if I'm a, if I'm a developer, I think, "oh, this is kind of cool." I like the idea of actually being able to run infrastructure, run kind of the code or run my applications in somewhere like this, in this kind of environment, because I think it's maybe more interesting.

Or, and if I can have the same convenience and same, the same kind of experience as a developer deploying code, as then why, you know, I might try this out. Is it something that people can use? Like, is there like. I mean, if I'm used to, like, deploying things into, like, virtual computers, I mean, virtual private servers or Kubernetes, is there something like that?

How do I actually try out some of this or use some of this stuff, for example?

Mark Bjornsgaard: Yes, it's because we're we are just a dumb datacenter operator. We are making our capacity of our datacenters available. Then that's the physical space in our datacenters for people like Amazon and Microsoft and Google and loads of other people to come and put their kit in our datacenters. So the minute you put your kids in our datacenter, then it will be doing something useful with the heat.

So as you say that there are a few cloud providers who already partnering with our main partner who have been incredibly supportive to us for years is a platform called Civo. So yeah, again, a UK business paying UK tax. If you as a developer want to run, you want a cloud service that is every bit as good as AWS or Google or Amazon or Azure,

and you want it to be green, then just go to Civo. And then you will be, Civo are using our data centers. So you as a developer, you shouldn't have to make any compromises at all, right? You shouldn't have to worry about any of this stuff. This should all be abstracted away. And in time will be where you can just be assured that when you're running code, it's running in the most environmentally, you know, it's being run in the most sustainable way possible. Now, part of the problem with the large clouds is that their reporting, their ESG reporting, their sustainability reporting is pretty shunky, stroke, complete bullshit. So I think that's part of the problem that I think a lot of cloud services at the moment aren't really taking this very seriously.

And what is certainly very hard as a developer or as an end user of a cloud platform to know how green or not your cloud is. The reality is any cloud platform that's claimed to be green just by using green electrons is ignoring 90 percent of the problem, right? 90 percent of the carbon in a data center is in the kit itself.

The scope, what's called scope three, the carbon that has been used to manufacture the computers themselves. So however much you jump up and down and say, "I'm doing really well because I'm buying green electricity or I'm buying" that's pretty much. I mean, it's not

Chris Adams: 10 percent rather than the other,

Mark Bjornsgaard: exactly so really, as, we all get better at this and as reporting becomes better and as greenwashing gets, people start to come down on greenwashing, as developers, as a whole community, we will have much, much better visibility about how green our clouds really are, but the reality is a green cloud, it comes down to the carbon in the compute and what you're doing, what you're doing to mitigate and reduce and remove that carbon.

Chris Adams: Okay. Alright, so maybe this is one thing that, so, there's one thing, there's one project that we work on in the Green Software Foundation that may be relevant for this. There's one project called the Realtime Cloud Project, where there is an effort to basically work out the carbon intensity for on a kind of per hour for every single region that we have.

If this is something that, I mean, it would be wonderful to have groups like Civo or people like that share something like this. Because the whole effort is to have some standardized data sets, some standardized numbers that you can trust and you can optimize for. And if what you've described is basically saying that yeah, running stuff inside infrastructure here is essentially somewhat fungible compared to running in other infrastructure here.

But if the number, if you're able to kind of reflect that in a lower carbon intensity or lower embodied energy or lower water usage then or any of the any other metrics that are available then that feels like a useful thing to actually allow people to be able to do and it sounds like that is something people can do today rather than having to this being a conversation about 2026 or 2027, by the sounds of things.

Mark Bjornsgaard: Well, to be clear, we're still, we're bringing our capacity online now. So we'd be a year in sort of designing since raising the money from Octopus designing building and now getting shovels in the ground and actually getting our data set the first wave of data centers built. So we've not done, we deliberately not said anything about this because we didn't want to be kind of part of the problem.

We want to be very much part of the solution. Whatever we will be reporting next year will be, you know, we'll be holding our hands up saying this is. This is as good as it gets the moment and we're going to improve it. But I think it's incumbent on all of us to be very transparent about that. I think that's it.

No one's trying to be perfect. No one's going to get kind of shot down for not being perfect. I think it's much more about the attitude you bring to it as a business rather than being, you know, "this is the law and I'm telling you it's like this" when we all know that's not true. But I think it's much better to be more tentative about it and say, "look, we don't know everything, but, you know, we think our scope three is this, and we are removing it using these removals."

And if somebody says, "I don't like those removals, I think they're nonsense." And whilst you say, "well, okay, but we are paying, you know, $250 a tonne for that carbon, so they're not complete bullshit." You know what I mean? I think it's in the, in this next phase, it's all about hopefully not giving each other too hard a time, but actually getting a bit more transparency and a bit more kind of clarity on where we are, because only then can we then start chipping away at it, right?

Chris Adams: Yeah. And like in the UK, we have very, clear targets for the very least like 2030 to get there, for example.

Mark Bjornsgaard: Quite, which is incredibly short

Chris Adams: It's very, it's like, it's almost tomorrow, isn't it? Yeah.

Mark Bjornsgaard: I'm so old that the years pass like days these days, but yeah, five years doesn't feel very long at all, frankly. Yeah.

Chris Adams: I could definitely sympathize with that because we are a non profit focusing on a fossil free internet by 2030. So that is very, acute for us as well. All right, Mark, I've really enjoyed chatting with you. And I've learned a bunch from us, like wonder or wandering through the world of digital infrastructure and stuff, we're just coming to the end of the time.

So I want to ask, like, is, I mean, if you,

are there any projects or things you want to kind of point people's attention to, or people, if people want to find out more about the work you're doing, where should people be looking, for example?

Mark Bjornsgaard: Yeah. If you're a developer, go to Civo. They're amazing people. It's an amazing platform, as I said. And the fastest, quickest way of supporting us is by using Civo. Buying Hewlett Packard Enterprise, Hewlett Packard GreenLake AI. So we're landing whenever you buy HPE kit in the UK and hopefully the US, you will have the option to land it in a Deep Green data center now.

So increasingly, developers and businesses can make green choices just by searching out our partners, you almost certainly never come to us directly. You're going to be consuming cloud services by a third party, but asking your cloud service providers to land that kit in our data center is the fastest, quickest way of helping us.

Yeah.

Chris Adams: Brilliant. Well, in that case, I'll speak to other friends to see if there's a way to filter any kind of like cloud providers for heat swimming pool as one of the kind of like features when I'm looking for my cloud computing in future. Mark, this has been fun. I really enjoyed it. Thank you so much for making the time, especially given like getting hit with COVID last week and everything like that.

So once again, thank you again for this and yeah, this is great. Take care of yourself and have a lovely week. All right, Mark.

Mark Bjornsgaard: Thanks very much for having me. Thank you.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

Finding Signal Amongst the Noise in Carbon Aware Software

9 januari 2025 | 35 min

In this episode of Environment Variables, host Chris Adams is joined by Tammy Sukprasert, a PhD student at the University of Massachusetts Amherst, to dive deep into her research on carbon-aware computing. Tammy explores the concept of shifting computing workloads across time and space to reduce carbon emissions, focusing on the benefits and limitations of this approach. She explains how moving workloads to cleaner regions or delaying them until cleaner energy sources are available can help cut emissions, but also discusses the challenges that come with real-world constraints like server capacity and latency. Together they discuss the findings from her recent papers, including the differences between average and marginal carbon intensity signals and how they impact decision-making. The conversation highlights the complexity of achieving carbon savings and the need for better metrics and strategies in the world of software development.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Thanathorn (Tammy) Sukprasert: LinkedIn | GitHub | Google Scholar

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPTION BELOW:

Tammy Sukprasert: With that one hour job with perfect knowledge of one year, we can reduce the carbon emission of the whole world by 37%.

I'm your host, Chris Adams. Hello, and welcome to Environment Variables. Where we bring you the latest insights and updates from the world of sustainable software development. I'm your host, Chris Adams. One of the oft repeated quotes when people talk about sustainability in software is that if you can't measure it, then you can't manage it.

And when it comes to working out the carbon footprint of a software application, a significant portion of the footprint comes from what we refer to as the carbon intensity of the electricity in use,

i.e., how green it is. And there are various steps you can take to make the same application using the same code, you can make it greener by running it where the grid is greener. So if you were to choose to run it in Iceland, that's one example. Or you can choose to run the grid, run the application at different times when the grid is greener, like when the sun is in the sky and your solar panels are wearing away. But how much greener can they get? And what else could we need to think about when trying to adopt a ways or ideas like this? Enter our guest for this episode today, Tammy Sukprasert, a PhD student at the Laboratory of Advanced Software Systems and Sustainable Computing Lab at the University of Massachusetts Amherst.

Tammy recently authored the paper on the limitations of carbon aware, temporal, and spatial workload shifting in the cloud, which examines how shifting computing workloads across time and space can help cut emissions. Tammy, we're going to spend a bit of time talking about why you chose to work in this field.

But to begin with, can I give you a bit of space to introduce yourself and what you do?

Tammy Sukprasert: Hi, Chris. Thanks for having me here. I'm Tammy Sukprasert. I'm a PhD student from the University of Massachusetts Amherst. I work on cloud and edge computing with a specific focus of decarbonizing computing. I'm currently calling you from Amherst, Massachusetts, and it's nice out here.

Chris Adams: Cool. That's nice. We've had a, it's snowing in Berlin, so I'm a little bit jealous, actually. Hi folks. If you are new to this podcast, my name is Chris Adams. I am the Director of Technology and Policy at the Green Web Foundation. And I'm also the, one of the chairs of the Green Software Foundation Policy Working Groups.

And also, the host of this podcast. Now, before we dive into the conversation with Tammy, if you're listening to this for the first time, here's a quick reminder. We will try to link to all the papers and all the links and all the projects on GitHub in this, and there will be extensive show notes as well as a transcript if there's anything you particularly missed.

And I think that's pretty much it. Tammy, are you sitting comfortable?

Tammy Sukprasert: Yep. Nice.

Chris Adams: In that case, I guess I'll begin. All right. We've linked to this in the show notes, but the paper title, On the Limitations of Carbon Aware Temporal and Spatial Workload Shifting in the Cloud, does kind of give a clue about what this research might actually be about.

But for those who are new to this idea, would you mind bringing listeners up to speed about what workloads are, what workload shifting is, when we talk about carbon aware computing?

Tammy Sukprasert: Sure. So to understand what workload shifting is, we need to have some idea of why we can shift the workload in the first place. So carbon intensity is based on the contributions of the different energy sources in the electric grid, right? So at different point in time, the demand changes. So there is different contribution of different sources.

That's why there's variation in carbon emissions. So there will be a high carbon period and low carbon period. And because of that, instead of running the workload during the high carbon period, you can actually schedule the workload to the lower carbon period or lower carbon region. So some of the workload, you can delay the start time.

The workload could be machine learning or some batch jobs. And instead of running right away when it was dispatched during the high carbon period, you can delay the start time and run it during the low current period. And at the same time, there are also, there is another type of workload that you can move or shift the workload around.

That could be a web request or an inference request. And instead of running your workload at your own region, you can look into other locations that have lower carbon intensity and migrate the in it.

Chris Adams: So if I, so let's say I'm using like maybe a chat bot or like, or I'm using something like maybe chat GPT and I am in, say, Germany, maybe it's dark, it's not very windy and it's not very sunny, for example, and most of the power is coming from coal being burned on the grid, for example, I might, rather than my request being served in Germany at the same time, it could plausibly be, say, forwarded to somewhere else in the world, as long as it's fast enough.

So, it might get forwarded to, say, Denmark, which is super windy instead. And that would mean that it would be slightly greener, for example. That's what you were referring to when you spoke about the inference. And then the other thing you mentioned before was like a machine learning job or like a video encoding thing.

That's something that I might not be seeing myself. But it's something that probably needs to happen within like a few days or something like that. So it's important, but it's not urgent. And because there's a bit of flexibility, I can choose when to do that to minimize the environmental impact of the extra amount of demand being put onto the grid.

Is that what you're, I think that's what you're saying there, right?

Tammy Sukprasert: Right. So it's just basically align your job schedule with low carbon period. Yeah. That's the key idea of the shifting.

Chris Adams: Gotcha. And then, so you spoke about there's one, which is if I'm doing something through time, that's like the temporal thing. Like I either bring it forward or wait till later. And then there's a spatial idea, which is me just moving it somewhere else. It might be happening at the same time, but it might be happening in Denmark, for example, or Iceland rather than in Germany.

Yeah?

Tammy Sukprasert: Yes, that's correct.

Chris Adams: Okay, cool. So, okay. We've got a good idea about what some of this might be. And a question I might ask is like, why is this interesting to you? Like what, how do you end up finding out or even kind of wanting to research this in the first place?

Tammy Sukprasert: Yeah. So there are many works that look into the benefits of reducing carbon reduction based on time shifting or spatial shifting, but it happened in a limited setting. i.e., a small number of regions or specific type of jobs, so people only look into spatial shifting or people only look into temporal shifting, or maybe they only look into a few number of regions but we were wondering, what if we look into both spatial and temporal and with the big picture of the whole world. So instead of looking to into a few regions, we look into 123 regions that we have in our data set and we want to see what is the broad impact of temporal and spatial shifting as a whole.

Chris Adams: I see. Okay. So thanks Tammy. So for this research paper, as I understand it, you decided to see how much, what kind of savings you really can achieve with things like Carbon Aware Computing. And a little bit about what kind of conditions might be necessary for these savings to be possible. So would you mind expanding on some of this?

We can start simple, fast, simple first, and then we can work our way up. So yeah, let's see, what were the first things we started with? And what were the first, what was like the ideal scenario for the savings? And we can go from there.

Tammy Sukprasert: All right. So with the current state of the world, right, the average carbon intensity is about 368 grams per kilowatt hour. And to achieve as much savings as possible in terms of carbon reduction, right, you will want to migrate your workload to Sweden, which is the region with the lowest carbon intensity in our data set. And migrating all the workload to Sweden, you can actually achieve 96 percent carbon reduction for the whole world.

Chris Adams: Okay, so what you're talking about there is you've basically gone from an average figure for carbon intensity of electricity to much, much cleaner electricity. And that's in this kind of ideal scenario, that's what you've essentially done. You've moved all of the computing jobs to the cleanest possible electricity there.

That's what we've done there. This is the ideal scenario. So where do we go from here then, for example, are there other constraints and things we know we need to take into account when doing this?

Tammy Sukprasert: Great. So of course, Sweden cannot take all the workloads in the world, right? So we were like, okay, instead of just moving everything to Sweden, what if we have capacity constraints? So we look into the scenario where every region in the world has an idle capacity of 50%. We're trying to be generous here because we want to understand the impact of the idle capacity on carbon reduction, right? So with every region having 50 percent idle capacity to absorb the job from other regions, instead of achieving, so now no one can actually migrate. So now not everyone can migrate to Sweden, right? Some other regions have to migrate to somewhere else. So, with that, the savings from 96 percent global reduction.

Drops to 51 percent.

Chris Adams: Okay.

Tammy Sukprasert: if not everyone can go to Sweden. Yeah,

Chris Adams: All right. That's still not bad. And when you're talking about capacity, you're referring to the fact that say, maybe there's a, like you've used the word region here, and for region, I think that's like a cloud region, like say AWS West or something like that. That's what you're referring to there.

And there's maybe a certain amount of reserve capacity they have to hold back. And that's what you're referring to there. So the idea that maybe different cloud places, different cloud data centers have a bunch of spare capacity and that's what they'd be using to move everything there, right? So, okay.

Okay. Well we never actually talked about latency constrains

Tammy Sukprasert: as well, right. So let's say for example, a web request, you need some service level objective or SLO to respect, to be respected, right. And so we look into that as well. And with, so now we have capacity constraints. So the scenario gets more and more realistic, right?

So from 96% you added a capacity constraint, and now the saving drops from 96% to 51%. And we also look into a more realistic case where we think about web requests that have some latency constraint, where there's some service level objective that has to be respected. And so on top of the capacity constraints that we have, that we achieve 51%, we added a 50 milliseconds capacity constraint, and that further reduced the carbon savings to 31%. So in the real life scenario, we are really far from the 96% that we want to aim for, right.

Chris Adams: So if I understand that correctly, basically there is a speed, the speed of light is fast, but it's not infinite. And therefore there are certain parts of the world where you definitely need to get a response back in time. And that's why you've introduced this kind of 50 millisecond kind of budget. So it has to be, your ping, your request has to come back in that kind of time budget.

And that basically places a second constraint. And even with these two constraints, this is essentially talking about, okay, these are the carbon emissions that can be reduced. By moving things to the various regions that are available based on the capacity of all these other places, like Sweden and then the next cleanest one and the next cleanest one.

That's what you're referring to there. All right. Okay. I think I understand that part there. And that honestly, 31 percent still sounds pretty good, to be honest. But if we look at the figures for what, 2%, if we're looking at maybe, A hundred million tons of CO2 each year, and 30 percent of that is 300, is 30 million tons.

That's not bad. That's more than at Google, for example. So, okay. That's okay. So that is interesting, then. So this is one of the high level findings you found, assuming you could do this in this kind of decreasingly idealized scenarios. And eventually we get to a point where, okay, this is actually something that you might plausibly try adopting in, or you might be kind of advocating for in certain regions, for example.

Tammy Sukprasert: Right. Yeah. The point that we're trying to make is that as you added more constraints, the gap between the ideal case of 96%. Your achievable goal widens. So that's what we're trying to show in this paper.

Chris Adams: Okay, cool. And when you're talking about the regions here, these are largely the regions that are inside the electricity maps. Was it the electricity maps dataset or was it just the list of all of the regions for the biggest cloud hyperscalers? I wasn't quite sure when we were looking at this, cause there's a list of them, right?

Tammy Sukprasert: Right. So we used a dataset from ElectricityMaps. Shout out to ElectricityMaps. Thank you for the dataset. The dataset has 123 regions worldwide, right? But on the dataset, we group them up, we filtered the regions that overlap with the cloud region, and look at all exclusively the results for the cloud regions.

Chris Adams: Ah, I see. So you created this way to make these comparisons basically by saying, maybe there's one data center, which we see in the cloud, like say Amazon AWS West, which is a lot of people refer to as like Oregon West 1. And because we know that a data set of carbon intensity from electricity map says, yes, this is Oregon.

You've been able to look at the numbers then in that way, right? That's where some of this is referring to.

Tammy Sukprasert: Yeah, so we did a mapping between the electricity map data with the location of the cloud region.

Chris Adams: Okay. All right. So, and that, and when we're looking at those numbers there, so you mentioned this figure of 96%. Was that looking at just location or was that looking at anything to do with time as well? Because I wasn't quite sure about that part there.

Tammy Sukprasert: So the 96 percent is just spatial shifting. So we have a separate result for temporal shifting where everyone in the world, every region in the world can schedule their workload based on one year ahead data. So everyone in the world can schedule their workload if they know about

Chris Adams: perfect forward knowledge. Yeah.

Tammy Sukprasert: yeah, perfect, knowledge for one year ahead.

And with that, we look at the extreme case, the most ideal case where the workload is a unit job, one hour job, to understand what is the best case scenario for temporal shifting, right? So with that one hour job with perfect knowledge of one year, we can reduce the carbon emission of the whole world by 37%.

Chris Adams: That's just temporal, not looking at location as well, right?

Tammy Sukprasert: Yes, so we have the results for temporal shifting that if we give every region a perfect knowledge of their carbon intensity a year ahead to plan their workload, what is going to be the best scheduling scenario for the future? Temporal shifting, right? So with everyone having the perfect knowledge for a year, you can reduce the carbon emission of the whole world by 37%.

Chris Adams: Ah, okay. So you're looking around maybe 30 percent when we were looking at purely locational, and then we're looking at just purely time. It's around, it's relatively similar, basically, but these are relying on. A kind of visibility that people don't really have a lot of the time, but, and, okay. So the next question I'm kind of asked is, it possible to look at time and space for this to get an idea of what the savings might be next from that then?

Tammy Sukprasert: Yeah. So we also look into that in our paper. So if you look at spatial and temporal shifting combined, the result actually shows that spatial shifting dominates the carbon reduction. This is simply because when you move the workload to the lowest region possible in your data set, right, to achieve the savings, that region is already low in carbon intensity, so time shifting doesn't make much of a difference.

Chris Adams: Ah, I see. Okay. So it's, basically the clean regions tend to be clean most of the time anyway, rather than being kind of spiking up and down for example. So that's what it seems like you're suggesting there, right?

Tammy Sukprasert: Right. It still varies, but the variation between the high carbon period and low carbon period is relatively small.

Chris Adams: Okay, well, that kind of makes sense. Cause I mean, now that when you lay out like that, I don't really think about it until you framed it that way, but like Iceland is usually green because it's running on geothermal, which is like pretty standard. Like it's steady. And even when you look at like, say Sweden, for example, there's like a wind and everything like that, but there's lots of hydro and stuff like that.

So again, it's not nearly as spiky as, say, Germany, where we are the land of like wind is, we're land of coal and solar. We have lots of coal, which is high carbon intensity, and lots of solar, which is very, low intensity. And flicking back and forth between these things means that we might have big swings, but on average, it's not particularly low compared to Iceland or Sweden, for example.

Huh.

Tammy Sukprasert: Correct. Yeah.

Chris Adams: Oh, right. Wow. I, that's,

in retrospect, it kind of seems obvious when you, but things are only obvious with when you look at it like that. And one thing you shared with me before we spoke about this was that some of this stuff is actually like, if people wanted to kind of explore some of these calculations, is this online somewhere? Is it like a GitHub repo or something where you can like poke around at some of these things?

Tammy Sukprasert: Yeah. So all the simulations in this paper, it's open source. So please check my lab website, my lab GitHub for the simulations.

Chris Adams: Okay, cool. All right. I think I've got the link here. So that's, this is from, so there's literally a repo called decarbonization potential. That's the one you're referring to here, right? On GitHub.

Tammy Sukprasert: Yes, that's correct.

Chris Adams: Brilliant. Okay. We'll definitely add that in the show notes because people who aren't like frantically exploring this themselves, it's where it's, right there.

Okay. So that was one of the first pieces of research. Essentially that there are some savings that can be made. It's around like the 30 percent mark in a kind of perfect world with location and sort of about the same with temporal. And if I understood it correctly, combining the two doesn't deliver massively more savings than that, right?

It's still never more than half this kind of intervention that you could possibly make, right?

Tammy Sukprasert: Right, yeah, combining the two doesn't give you double the benefits, because the benefits are dominated by spatial migration, but not much of the temporal, if you combine them together.

Chris Adams: Okay. Thank you. I'm really, glad you actually spoke about this because we can now have some of the numbers. To basically talk about the fact that, yeah, we still need to do other things. You can't just like leave your code and make no changes. That might get you some of the way. And if you're looking at Temporal, it'll get you 37 percent of the way in a perfect world.

But you still need to make some other changes if you want to kind of reduce the environmental footprint further. Brilliant. Okay. Thank you for that. So we talked about some of the savings you can get in your previous paper. The fact that there's maybe around the 30 percent figure. And if you can move everything through space, you get around maybe 30 ish percent savings.

If you look at, if you have perfect knowledge forward for the year, then it's maybe slightly higher than 30%, but it's in the same kind of ballpark. And if you were to look at moving all of your computing jobs through time and space, you can't just double this number. It's still going to be a meaning, it's going to be more than 30%, probably less than 50%.

So that's one of the figures that we have. We'll share a link to the GitHub repo for people who are curious about this and want to see if they know what jobs they ran last year, they could see what kind of savings they could have achieved. So that's one thing. And we've spoken so far about some constraints that we have, but there's a few more constraints that we need to take into account.

So for example, so far, we've been talking about how much, how many spare servers we have, like data center capacity inside this. But there are other constraints that we need to also think about, which are a little bit further down the stack, as it were. So there may be a certain limited amount of green energy, at which point when you have more demand than that, you might need to have some other forms of generation come on stream.

And like, this is something that I think you explored in one of the other papers. So maybe we could talk about that. So, okay, this other paper that you spoke about, maybe we can just like, let us know the name and then we'll see where we go from there.

Tammy Sukprasert: Right, so this paper, titled, On the Implications of Choosing Average vs marginal Carbon Intensity Signals on Carbon Aware Optimizations, basically, average vs marginal for carbon aware optimizations, right. So this paper came from the fact that, okay, People have been suggesting, let's shift the workload through time, let's shift the workload to different locations, but we never actually agree on which carbon intensity signal to use for carbon aware optimization, so as the title suggested, there are two types of carbon intensity signals that are mainly used, namely average carbon intensity signal and marginal carbon intensity signal.

So for average carbon intensity signal, just think of it as a snapshot of the grid at that point in time, right? And the way it's calculated is the weighted average of carbon emissions weighted by their production,

Chris Adams: Okay. So if I just check, I just want to start you there. So make sure I keep keeping up with you. So there's two ways you can measure carbon intensity, like how green electricity is. And this first one, this average one is basically saying, well, I've got maybe two coal fired power generators and one wind farm, so therefore I'll apply double the weighting of the coal versus one of the wind farm.

That's kind of what, that's a simplified version, but that's essentially how you work out an average figure, right?

Tammy Sukprasert: Right, right, but marginal carbon intensity signal is different. The way it's calculated is the carbon intensity with respect to the change in demand. So let's say just now you said you have two wind farms and one coal, but the next unit of demand is going to be served by gas generator. So then the marginal carbon intensity signal is the current intensity signal of that of the gas generator.

Chris Adams: I see. Okay. So rather than looking at the average, it's almost like the kind of consequences of me doing a particular thing. That's what we're looking at there, right?

Tammy Sukprasert: That's correct.

Chris Adams: Okay. And this, so now we've got this. I hope if you're listening and you're struggling, this is really hard.

So, thank you for staying with us so far. So this was the general, this is what we were looking into. And, as I understand it, this incentivizes different actions, or if you were looking at this, you might choose to move things to a different region or choose to run a computing job or do something at a different time.

That's been my understanding of this. Is this is what you looked into then?

Tammy Sukprasert: Right, so the paper look into the fact that if you follow one signal as a scheduling signal, you might end up in more carbon emission based on the perspective of the other signal. Yeah, so it turns out like you cannot just follow one signal and hoping that you will do well based on the other signals perspective as well.

Chris Adams: Ah, okay. All right. So this adds another layer of complexity to this then. So if I understand it, I could be following one and that gives me some idea here, but there are certain places where they can be different. They can have different signals. So like some places might be the same, but there are certain parts of the world where I might have quite radically different signals between these two.

That's what I think I'm hearing from that.

Tammy Sukprasert: Right, because the two carbon intensity signals are calculated so differently, so in, within one region, the signals are generally not correlated. So when you schedule for one signal, let's say, for example, I use in the marginal carbon intensity signal as a scheduling signal, right? And I place a workload in this low carbon period based on marginal, but within the same time period, someone else is like, looking from the perspective of the average carbon intensity signal, they'll be like, "Hey, I wouldn't place my workload here because it's high carbon period right now."

So it has some conflicting decision making.

Chris Adams: And, presumably when you looked in the, when you're doing this research, were there particular parts of the world where you see wild spreads between these two places? Like there's some places that it's quite safe, right?

Tammy Sukprasert: So in the paper, we look into, Arizona and Virginia for this kind of conflicting scheduling. So Arizona has fluctuating average carbon intensity signal, but really flat marginal and vice versa for Virginia. So let's just take Arizona, for example. Like if. You want to schedule based on marginal carbon intensity signal, you wouldn't do anything because it's flat.

You can just place a workload wherever you want. But if you want to schedule the workload based on the average signal, you'll be like, I would place my workload at this particular time slot because it had the lowest carbon intensity signal during the day.

Chris Adams: Ah, I see. Okay. So this suggests that you're going to need to be really explicit about which kind of signal you're following. And, there are certain parts of the world where it, you're more exposed to the differences between this, for example. That's what I think I'm hearing there.

Wow. that sounds, yeah. Sustainability in software does not get easy. Okay. So that's one of the things we were looking at here. And, it sounds like that you've spent quite a lot of time looking into this, looking at this whole field then, and, presumably when people are taking their first steps to trying to work out the environmental impact of software, for example, would you suggest, is there like an order of things you might start with this?

Cause this feels like relatively advanced, high level, complicated, calculations here, and is it possible to kind of look at the environmental impact of software without this straight away? Like, can you add this a little bit later, perhaps? Maybe there's like some rules of thumb or some approaches you might suggest as a researcher who has looked into this and tried to understand the environmental footprint of some software and said, "well, okay, you might want to just look at the total amount of energy used or the total amount of resources used first, before you look at, say, this carbon aware stuff. And if you can look at carbon aware, then maybe look at location first" or something like that. Cause this feels like kind of exciting, but this also feels like it gets complicated very, very quickly.

Tammy Sukprasert: So when I started working, on carbon intensity signals, I find that the average carbon intensity signal is easier to understand simply because you just look at the overall picture of the grid and you take the average of the energy sources, right? But for marginal carbon intensity, it was interesting concept for me.

You look into the carbon emission based on the change in demand, but I was having a hard time understanding this because in a practical sense, I feel like it's going to be challenging of understanding which power plant is actually serving my compute workload. Like, it's not transparent enough.

Chris Adams: I see. So there's almost like a counterfactual you're, comparing it against like a, how do you know if someone, I think you, we spoke about this sort of like there's a power stack, right? Like, yes, I've switched off, I've stopped pulling power from the grid, for example, but, how do I know that no one else has pulled power from the grid at the same time?

Is that what you're kind of getting at there?

Tammy Sukprasert: Right. For marginal carbon intensity for me, the concept is actually good. Like, you're responsible for the carbon emission that you triggered, right. But, In, reality, like you don't know which power source is serving your demand and whether in the next time it's to serve by the same force. So for example, like I plug in my laptop only, maybe I could, my laptop maybe is fulfilled by coal, but someone, let's say, Chris, you unplug your lab, right? Maybe now you left the, now your, the demand decreases is my laptop still, my laptop power is still fulfilled by coal? Like I don't have that. So...

Chris Adams: Ah, I see. Okay. Alright. That makes, no, that makes a bit more sense. And I kind of, I think I understand why, I think I follow basically the reasoning between why you might start with one before starting with the other one. Because I think I agree with you on that. I found the average a bit easier for me to get my head around two as well.

And, marginal does sound really cool, but I don't think I'm very confident explaining it to other people. And I think that, I think my experiences seem to echo yours, actually. I'm glad you said that because I did wonder if it was just me and that does make it a bit easier for me too.

I feel a bit better about myself now, actually. Thanks for that, Tammy. Okay. So, this has basically been your day job for the last few months, diving into the world of carbon signals and things like that. Is this some of the continued research you're doing, or are you looking into other fields now beyond software carbon intensity and working out the differences of carbon, working out the, potentials of carbon aware computing here?

Tammy Sukprasert: So I'm still working on carbon aware computing stuff. Currently I'm working on a web service that harnesses renewable energy and I have to think about how we should handle the workload when there is no renewable energy available.

Chris Adams: Okay. All right. So one thing this does seem to suggest is that if we're just looking at carbon in here, that's not showing us the whole picture. And even when we just look at carbon. We end up with quite a, we can end up with like difficult or conflicting signals for this. So it may be that we need to, we might need to expand the way we think about as software engineers, we think about the next layer down and say, like, are there other things we take into account beyond just looking at marginal or looking at average?

Maybe there's something else we need to do or another way of thinking about the grid and how our interactions as software engineers kind of work with it and how that can have an impact there.

Tammy Sukprasert: Right. So I think we need to move beyond the static signal and instead maybe look into other characteristics to take into consideration when doing carbon aware optimization, maybe in future direction, maybe we would agree on some other signal that captures the long term impact of the grid, like average carbon intensity signal and the current, like the instantaneous change in carbon intensity, like marginal. So yeah, apart from optimizing for carbon efficiency as a community, I think everyone should keep in mind about like, we need a better metric to capture this carbon emission.

Chris Adams: Okay. Thank you for that. Tammy, this was a ride for me. Every single time I come to trying to understand the environmental footprint of software, I think I understand that there's a whole nother set for this. And you've really opened my eyes to this. Tammy, if people are interested in this field, are there any other projects or work that you've read about recently that you'd like to draw people's attention to?

Tammy Sukprasert: Yeah, I think you should look at Carbon Scaler. I think that's one of the things I

Chris Adams: Oh.

Tammy Sukprasert: recommend people to check it out.

Chris Adams: Okay, we'll have to share a link to that because that's totally new to me. I've never... I'm not aware of that one actually.

Tammy Sukprasert: So yeah, it's a system that reacts based on the available carbon intensity, and you scale the workload based on that. So you don't have to shift the workload.

Chris Adams: Okay. All right. And if people want to find out more about the work that you're doing, where should people be following? Is there maybe, is there a website or are you on LinkedIn? Like what's the best place for people to direct people's attention if they wanted to follow up and read actually some of the work that you've been publishing and talking about here today?

Tammy Sukprasert: Yeah, so, I'm on LinkedIn. You can search my name up, Tammy Sukprasert, or T Sukprasert for the link, yeah.

Chris Adams: Brilliant. All right. Well, Tammy, thank you so much for giving us some of your time and sharing what you've learned from here. It's been absolutely fascinating. And we now finally have some numbers about what we can achieve with carbon aware computing. At least we have some numbers now to work with. So thank you once, again for this, and I hope you have a lovely week.

Cheers, Tammy.

Tammy Sukprasert: Chris, cheers.

Chris Adams: Hey everyone. Thanks for listening. Just a reminder to follow Environment variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode!

The Cloud and the Climate: Navigating AI-Powered Futures

19 december 2024 | 36 min

Environment Variables host Chris Adams is joined by Jo Lindsay Walton, a senior research fellow at the Sussex Digital Humanities Lab and co-author of the report The Cloud and the Climate: Navigating AI-Powered Futures. They delve into the intersection of climate and AI, exploring the environmental impact of AI technologies and the challenges of decarbonizing the ICT sector. Jo discusses key takeaways from the report, including the importance of understanding AI's direct and indirect impacts, the nuanced roles of big tech companies, and strategies for critically assessing claims of AI-driven sustainability. This insightful conversation highlights the need for interdisciplinary approaches and robust collaboration to navigate the complex relationship between technology and climate action.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Jo Lindsay Walton: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Jo Lindsay Walton: There's this great metaphor that Arvind Narayanan and Sayash Kapoor have in their book, AI Snake Oil. They say, "imagine we just talked about vehicles. We didn't talk about bicycles or cars or buses or trains. And we tried to talk about the climate impact of vehicles." It would be very difficult to do.

And that's essentially what AI discourse does, right?

I'm your host, Chris Adams. Hello, and welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. Like seemingly everyone else in the industry, we've been talking about AI a fair amount recently, and earlier this year, in September, the Sussex Digital Humanities Lab published their report, The Cloud and the Climate, Navigating AI Powered Futures.

It's not a small report, weighing in at around 190 pages, and it has a number of key messages we'll be exploring in this episode. Also, one of the previous guests from back in September, 2023, Jo Walton was one of the authors of this report, and he was a nice enough to make some time to join us today on the pod.

So, Jo, thank you so much for coming onto the pod again. Can I give you a bit of time to introduce yourself and what you do in your day to day for people who missed the last episode back in September?

Jo Lindsay Walton: Hello. Yes. Thanks. Thanks so much for having me. So I'm a senior research fellow in arts, climate, and technology at the Sussex Digital Humanities Lab. My day to day is 90 percent playing with my cat, but I am also part of the Digital Humanities Climate Coalition and the newly launched Climate Acuity Initiative, which does facilitation and CPD training around climate and tech in hopefully fun ways involving storytelling and games and things like that.

And yeah, it's just, it's really nice to be back on this wonderful podcast. I feel like the host of SNL.

Chris Adams: Thanks. So, just before I check, when you say CPD, CPD is continuous professional development. People who want to build understanding of climate into their professional life, right? Is that what it is?

Jo Lindsay Walton: Yeah, that's it. Exactly. And it's, really I guess, part of my work is at the intersection of climate and technology, but I'm not personally super technical. Most of your listeners probably have a lot more technical knowledge than I do. What I am really interested in is communicating around these issues and education as well.

So I'm raising them for all the stakeholders for whom they might be important.

Chris Adams: Brilliant. Okay. And, on the subject of other three letter acronyms, I've just had my cat walking myself, so if it walks across the, when we're recording, please do forgive it. It's just, that's what he does sometimes. Okay. Folks, if you are listening and you've never heard my voice before, I am Chris Adams.

I am the executive director of the Green Web Foundation, which is one of the members of the Green Software Foundation. The Green Web Foundation is a nonprofit based in the Netherlands, focused on reaching an entirely fossil free internet. By 2030, I also work as one of the policy working group chairs inside the policy work in, inside the GSF, as well as being the host on this podcast as well.

All right, then. So just before we dive in, if we speak about a particular paper or a report or a link, we will add these in the show notes. And if you, if there's something you're missing, please do send us an email or get in contact us because we do our best to keep these available and like useful resources for people.

All right, then. Jo, are you sitting comfortably?

Jo Lindsay Walton: Very comfortably. Thank you.

Chris Adams: All right, then. I think I'll begin. All right, then. So before we dive into the report and some of the key takeaways, which we'll be going into more detail, can we talk a little bit about why you decided to put some time into this report in the first place and how this entire project came about, please?

Because I know that you're, you said yourself, you're a researcher and in the Unix School of Media Arts and Humanities at Sussex University. And this came from the Digital Humanities Climate Coalition. Now, most software developers, when they think about AI and reports, it might be something that's within the industry.

So I want to give you a bit of space to talk about why it's interesting or why it's relevant to have people who aren't inside technology, who aren't like practitioners per se, talking about some of this. Because I think there's a different, a couple of perspectives that you might have that is worth.

Making clear for people, for example, or some techniques that humanities people might actually have that, the developers or techies might not be so cognizant of.

Jo Lindsay Walton: Oh Yeah. Absolutely.

So the report, as well as the kind of DHCC toolkit, which is an online resource, these are, they're very much community projects and they have an open source ethos and a part of that is an aspiration to interdisciplinarity. The report itself is a kind of stretch goal or spin off from a small innovate UK project that I was doing with GreenPixie, who are this fabulous cloud carbon data company.

Yeah. And we were basically exploring how to talk to a wider set of stakeholders about the cloud and about the climate. So not just IT people, but also, for example, chief sustainability officers, people who need to know about this stuff. That might not be quite so up to speed on the technical detail.

And over the course of that, it grew apparent to me that there was a gap really for an accessible resource that didn't oversimplify and that really tries to be a bit holistic. Can you really understand one bit of it without understanding the big picture? Can you, I don't know, understand how your little piece of software that you're trying to optimize is going to have an impact without thinking at least a little bit about carbon accounting and the greenhouse gas protocol and carbon offsetting? Can you really understand how green a data center is without understanding a little bit about how energy gets into the grid and then gets into a data center and the kind of energy procurement rules around that?

Chris Adams: Okay. I hadn't actually realized that you've been working with GreenPixie and just for people who are new to that term, GreenPixie is a UK-based SaaS provider of essentially carbon accounting tooling for cloud, just like, so if you're using Amazon's or Microsoft's or Google's cloud carbon calculator, they provide something very similar, but with a more kind of open methodology that allows them to be comparable to each other.

Really nice to know that there's, I didn't realize that you two would be working together on that. And that's cool, actually.

Jo Lindsay Walton: So, I mean, they, particularly that collaboration informed the green ops section of the report. But as you allude to, there is this attempt throughout the report to also bring in DHCC type perspectives, that kind of humanities flavor, really drilling into the details of the cultural factors.

So not just how we communicate things, but also how we imagine things, I guess. Big tech and tech communities don't just have direct impacts. They also shape the way that we imagine the future. So Google is not in the business of building kind of direct air capture, giant reverse hairdryers that are sucking carbon out of the sky. That's not something they do, but they do influence the way we think about technology and climate. And so they also influence the way that we think about things like greenhouse gas removal technologies.

Chris Adams: Although, earlier on this year, we saw that Microsoft patented, actually, some of the use of some particular things around carbon capture in data centers to use some of the waste heat to actually separate captured carbon, so it can be actually stored in other places. So, there's maybe more links than we actually had, yeah, exactly.

Jo Lindsay Walton: That's really interesting.

Chris Adams: Yeah, I'll share, we'll share a

Jo Lindsay Walton: link in the show notes for sure.

Chris Adams: Definitely. All right. Okay. So that gives me a good idea and then provides a bit of context to where this was and for people who are not used to the UK, Innovate UK is one of the government funding agencies that has provided some of the funding for some of this.

So that's where that has come from. All right. And so maybe we should talk a little bit about the report. So there's a number of takeaways. In fact, I counted more than five when I was running through the report. So there was a lot there, right? And there are some things which probably don't need too much attention because we're, because of the listenership.

So for example, we probably won't spend too much time dwelling on one of the takeaways being we're in a climate crisis or the other one bang, yeah, that digital has a physical basis. These are things that we can assume that people have internalized already, right? But there was actually some nuance to this because.

While people do talk about that, the kind of magnitude of the numbers might not be something that people are quite so comfortable about. And also, it's an area of contention in many cases, many places. And as someone who's been looking at a lot of the literature, I figured it might be interesting to have a bit of space to talk about one of the other takeaways, which you shared was basically the ICT sector is not a leading contributor to global warming, but it still must decarbonize rapidly.

Now, I think It'd be useful to unpack some of this because a lot of the time, a lot of the stories do talk about either data centers as like this new monster or new kind of like media baddie, for example. And it seems like there's, you've got a kind of more nuanced take on this and I wanted to give a bit of space for you to allow you to talk about some of that.

Jo Lindsay Walton: Yeah, I mean, coverage of the drivers of global warming is totally out of proportion to what those drivers actually are. We've seen data centers be in the mainstream media quite a lot recently. So I think maybe that's falling victim to that a little bit. Where do most emissions come from? Food production, for example, is absolutely huge.

And we hear a little bit about food miles. But food miles are not a massive part of it. A much bigger determinant of the impact is, "has the food come from a cow or from a nut?" Constructing and heating and lighting homes, road transport, fugitive emissions, fossil fuel companies basically being a little bit sloppy as they extract these fossil fuels and letting them escape.

There's a good, a lovely breakdown on Our World In Data, which is maybe we can put in the show notes as well, although a little bit dated now. ICT? What is the impact of ICT on global warming? Would like to offer a provocation and hope that maybe one of your listeners can, prove me wrong. I think nobody knows.

I think nobody knows ICT's impact on global warming. There's that 2021 Freitag et al. estimate that gets quoted quite a lot, but it's been a very, busy four or five years. I feel like I've lived through the AI singularity. And there's more complexity than that, right? When you factor in secondary and tertiary impacts, both good and bad, from the digital, then you're in the realm of deep uncertainty.

There is unlikely to be any expert consensus. Even so, despite that complexity, it's not controversial that tech needs to decarbonize along with everything else. It's all hands on deck. Everybody's on board with that. All the big companies have these ambitious pledges. What's concerning me a little bit is how that discourse is shifting.

So for example, Microsoft in 2020 sets out its pledge to achieve net zero.

Chris Adams: Moonshot. The zero carbon moonshot you're referring

Jo Lindsay Walton: yeah, yes, And we talk about that term moonshot in, in the report, actually, cause it's a, it's an interesting metaphor. And the moon has, is now said to be

Chris Adams: 5 years further away. Yeah.

Jo Lindsay Walton: The moon has moved five times. So actually I think that's incorrect.

I think the moon. The moon has been vaporized. The moon, as in Neil Stevenson's science fiction novel, Seveneves, the moon no longer exists. The target has already been missed. And that happened this year. "Okay, how is that possible?" you're asking. Does Microsoft have a time machine? How can they fail their net zero pledge of 2030 in 2024?

Well, that's the way that net zero pledges work. They are about cumulative emissions. They're not about a snapshot of emissions at a particular date. They are about the pathway from the date of the pledge 2. 0 staying within a given emissions budget, right? So you could draw a descending line graph and it's about the area under that line, not about the point at which the line intersects the axis.

And to their credit, Microsoft absolutely was transparent about this back in 2020. They showed the linear descent to zero. And by my estimates, that budget was burst sometime this year. maybe now, maybe as we are recording this podcast. And poverty is no effect. The concerning bit is that this isn't being talked about more openly.

It's much more this discourse, as you say, of "okay, now we have AI." In 2020, we didn't know about that, but now we have AI and AI has these sustainability benefits. Okay, so if that's the argument, if that's the implied case for emissions increasing, let's be very clear about that. Are we saying that it is prudent to increase emissions from the tech sector for the next few years?

Are we saying that the tech sector has been doing the right thing emissions-wise for the past few years, because those emissions on a robust methodology are shown to be more than offset by the sustainability benefits that they can provide on an appropriate timescale?

Chris Adams: We'll be touching on that a little bit later, but, alright. Okay. Thank you that I appreciate you providing a bit of extra context on that. And just to check if I understand, you said one or two things about, okay there is, the way you could work out the environmental footprint of the ICT sector when people talk about the direct impact, you said there's like a primary, tertiary, sorry, secondary and tertiary, presumably you're talking about like there is a direct impact, but there's an impact from people, what you enable with that computing and stuff like that. Is that what you're referring to with that primary, secondary and tertiary stuff?

Jo Lindsay Walton: Yeah, absolutely. So you and I are on a, Zoom call now. If we weren't on the Zoom call, I probably would have ridden to you on a giant lump of blazing coal. Or some more carbon intensive mode of transport. And those are very, complex calculations to do. You have rebound effects where, things look like they're providing efficiencies, but those efficiencies are mitigated or more than offset by increased volume, it's complicated stuff.

Chris Adams: Okay, cool. All right. Thank you for providing that extra kind of elucidation or like clarifying that part there. Okay. There's another thing I wanted to give a bit of time for actually was this one. You said, and given that we just spoke about kind of cloud giants and one of the takeaways, which was none of the cloud giants is a monolith.

So this is a bit of a kind of more nuanced take on big tech bad, big tech good that we often see in the discourse, because it's very simple and attractive way to talk about that, but it sounds like you're trying to go for something a bit more sophisticated there, a bit more multidimensional there.

Maybe we could spend a bit of time trying to see what you were trying to get at there or what the report was trying to really get across to people.

Jo Lindsay Walton: Yeah. I think climate invites us to really reflect on our roles in our professional lives and other aspects of our lives. And sometimes to challenge and push back on the parameters that are set for us in those roles. And that may mean that your company is pushing a particular line or your bosses is pushing a particular line, but there is a kind of, there's a practical incentive and there's, frankly, there's a kind of ethical duty to be critical about that and to step outside of the boxes that you're asked to perform in.

And definitely these companies are huge companies. There's a great diversity of knowledge, a great diversity of kind of politics, really, within any particular industry. big tech company, nevermind between tech companies as well. So in the realm of the greenhouse gas protocol and how we do carbon accounting, there's a lot of disagreement within big tech between on the one hand, Amazon and Meta who want one kind of particular set of rules as the greenhouse gas protocol is revised and Google and perhaps Microsoft who would like to see it go another way.

I think we look at this a little bit in the report. We look at a nature article that is largely authored by, Microsoft researchers. And spend a little bit of time in a hopefully good natured roast of the estimate of the carbon impact of AI, which the methodology there just isn't really fit for purpose.

If you drill, really drill, drill, drill baby, drill down into the details, you find that it is based on one back of envelope kind of estimate by Vijay Rakesh, who is really a stock market analyst who said that he expected NVIDIA to deliver a hundred thousand AI servers in 2023.

It's not a sufficient basis for estimating the global impact of AI, but that's hopefully not the main point because the bigger part of this article, which I think speaks to your question about companies not being monoliths and trying to build alliances for progressive and robust climate policy that cut across your loyalty to a particular company.

The proposal of this article is that AI researchers should work more closely with climate, the climate modeling community, and that AI should be integrated into the IPCC's shared socioeconomic pathways and integrated assessment modeling. Which is, I have mixed feelings about that. Like the closer collaboration sounds really great. It does feel like in that particular article, there isn't yet a very deep understanding of how those climate models work. They're not really scenarios. They're more like building blocks for scenarios. And to some extent, they already do build in the possibility of technological change.

So you could go down a rabbit hole as to whether or not AI is already priced into these models or not. I think what it speaks to is a certain kind of nervousness here, like, okay, so we are big tech, we are AI, we're presenting this AI powered future, and we're increasing our emissions, and we're doing this on the basis that we think, we believe, that AI is going to unlock all these fantastic sustainability benefits.

But can somebody please check our working? We recognize that we may have conflict of interests. We need to do this in a more collaborative way. We need to have all kinds of expertise and we need to have more independent voices. I think that's what that article is ultimately calling for.

Chris Adams: Okay. So there was one thing you said that you were getting at there was the idea that cloud giants not being a monolith isn't just within the cloud giant. If you think about it horizontally, like Meta and Amazon having one point of view. And I think you're referring to the emissions first versus the 24/7 kind bond fight about how do you count energy as green?

Because the current process has a few significant issues with basically, there's people trying to work out a new approach and you have two camps. So that's one thing you were talking about. And then there's almost one within each company. Like there are different people who have different drivers inside that. If you just assume that someone's working for say Amazon, that ends up being a very lossy way of talking about, okay, what are they doing? And like, what might, the drivers be, for example?

Jo Lindsay Walton: Absolutely. And some of those disagreements might not be so visible for obvious reasons. People have to be tactful and work in constructive ways with their colleagues. I mean, to respond to that, I think that I, share a bit of alarm about timescales and, solutions being proposed that aren't immediately referred back.

If you've got any kind of plan to do with the climate, check it against IPCC timescales. We were supposed to, in the next six years, cut carbon emissions by more than half, four of which are going to be under a Trump administration. And I would definitely, I would celebrate that kind of all hands on deck approach where everybody's doing everything they can in their role and maybe rethinking their role in creating alliances.

At the same time, I also think we need a little bit of reflection on actually which hands are on deck. Are there problems that aren't owned by anybody, risks that are not being addressed by anybody. And I think that we need a little bit, in the AI space, there has been talk of pauses and kind of moratoriums, not always for the best reasons, but I do think these are really important tools in our toolkit rather than, "okay, we're going to just keep doing what we're doing and, hope to sustainabilize it as quickly as we can,"

actually, saying "maybe we need to pause this and maybe we can pick up where we left off, but we need to pause it while we're gathering more data or we're greening our energy supply or we're building capacity" or whatever it might be. I wrote an article about this in the Fantastic Branch magazine called Pause.

I just realized this morning, I should have called it after Andreas Malm's How to Blow Up a Pipeline, I should have called it How to Blow Up an AI Pipeline. But yeah, I something else for that.

Chris Adams: Yeah. All right. All right. Okay. Thank you for that. Let's move to the next one. Cause you spoke a little bit about AI and, in the report, you actually spend a bit of time talking about sustainability. Basically the sustainability of AI, but also AI for sustainability. Right. And these being two somewhat different things.

Now we talk about sustainability of AI on this podcast quite a lot. So we talk about how to use like more efficient algorithms or how to clean the energy and some of the steps you might take. And obviously the report talks about that, but there's actually something that you speak about in terms of the claims about AI for sustainability goals that you spend some time talking about and like you also raised, like "these are some of the red flags you might be looking for." Could you maybe, are there any like specific messages you might use or anything you draw to people's attention to when they're trying to navigate claims about AI for sustainability and like, "yes, there's a massive energy footprint, but the upside is this, for example, and these are the upsides that we're delivering."

Jo Lindsay Walton: Yeah, absolutely. And all that kind of sustainability of AI stuff is extremely exciting. And, as you say, we, touch on that in the report. AI for sustainability. There's this great metaphor that Arvind Narayanan and Sayash Kapoor have in their book, AI Snake Oil, they say, "imagine we just talked about vehicles.

We didn't talk about bicycles or cars or buses or trains. And we tried to talk about the climate impact of vehicles. It would be very difficult to do." And that's essentially what AI discourse does, right? We don't on a regular basis make these kinds of fine differentiations in public discourse, in journalism, in conversations with friends.

So right before the show, actually, we were talking about acronyms and I tried to come up with an, acronym of the things that you might want to ask when you find a claim that AI is delivering some kind of sustainability benefit. So the first thing to consider is maturity. That might be technology readiness level whatever it may be.

Often there is a claim is inflated. It says something is already happening when actually what we see is that there's been a study that says it might work. It could be rolled out commercially, scaled up in five, 10, 20 years, whatever it might be. So maturity is one. Then additionality. So AI is responsible for delivering this sustainability benefit.

Well, do your best to identify which bit the AI is responsible for. Often an AI sustainability project will involve data collection and analysis, and then some kind of efficiency gains from that. What could have been delivered with the, with more kind of traditional data analytic methods? And then generative or discriminative or some other type of AI.

What kind of AI are we talking about here? These are often conflated. Is it even machine learning at all? Is it something, some cool new thing like, I don't know if it's new, but active inference, for example. And how big is the model and so on? What kind of AI are we looking at? And then finally adaptation versus mitigation.

So these are the two broad categories of climate action that most climate scientists will recognize. And they're interrelated and they overlap in various ways, but mitigation is really about decarbonizing,

Chris Adams: Yeah. Green energy instead of fossil energy. And then the mitigation might be building the seawalls because the sea levels have risen. Stuff like that.

Jo Lindsay Walton: other

Chris Adams: I say, yeah. adaptation is building the seawalls because the sea levels have risen. Mitigation would be switching out of fossil fuels and burning, using greener energy, for example, which

Jo Lindsay Walton: Absolutely.

Chris Adams: the

Jo Lindsay Walton: As can imagine with AI, if the AI has a problematic carbon footprint, but delivers substantial adaptation benefits, that again is a very hard calculation to do. You, can't simply. Subtract one, one from the other. The acronym unfortunately came out as MAGA, which has already been taken. So, I'll keep working on on it.

Chris Adams: Okay. All right. I don't know how far that's going to go. I'll be honest. But. All right then, so, so that's one of the things you're speaking about was this idea that these are two separate things and it's worth being aware that there's, there, there are different ways you can essentially critically engage with some of these claims.

And I think I'm get where you're going with some of that now. And I've realized that I'm basically an Englishman in Germany, speaking to someone, to an Englishman who's also in the UK. And this was a report that came from a UK research unit. And obviously there's a UK research focus on this, but it's also, we're also in a scenario where there is new government in the UK who have very aggressive goals of like decarbonizing the entire grid by 2030.

So we spoke about 2030 target before, and like, this is one where there is a goal to decarbonize the grid by 2030 and reduce nationally carbon emissions by more than 80 percent by 2035. So this is like, in many ways, this is like a similar kind of moonshot thing we have here, but there's also, it's the government is also very, Gung ho right now on the increased deployment of data centers around the UK as one of the kind of drivers for growth, for example. So I wanted to ask you, like, when you look at this, do you see these goals as complementary or compatible or are there any specific areas of attention for the UK that are like for policymakers should be thinking about if they want these goals to be possible, for example?

Because yeah, there's, it sounds like it's, there's probably a lot of nuance to it, and this is something that you've been having to navigate or have to think about.

Jo Lindsay Walton: Yeah. And I mean, I don't know. I really don't know. And I wonder what guests we will need to assemble on your show to solve this question. It's definitely a, it's an interdisciplinary type question, right? We need people who can think about the counterfactuals, the opportunity costs. If data centers are not expanding at this particular rate in the UK, what's happening in that alternative universe?

There's in the report, there's a quite upbeat section lead authored by my colleague, Benjamin Sovacool, which is all about the wonderful things data centers can do to be more efficient and environmentally friendly. And so from a UK perspective, you can see those things going together. Yes, we're going to, we're going to be a leader on net zero.

We're also going to be a leader on data centers. And we're going to do that by having the greenest, the best, the most efficient data centers. Microsoft is shifting from concrete and steel to a special new timber. The new exciting innovations happening all the time. As a thought experiment. If we were building global data center infrastructure from scratch, knowing everything that we know, how would we design it?

Maybe you can get some experts on your show and ask them this. I've heard it said that data centers are these kind of fabulous heat generators that just happen to be able to do computation as well. One of the reviewers of the report said that. And so we should really go in hard on small and medium data centers woven into the fabric of our urban environments.

Anne Currie, who, we did that previous really fun episode about data centers on the moon and various things. Anne has said that a key consideration is that you really don't want to be competing with other local energy needs. So this is a contrasting view. You don't want to be displacing demand into carbon-intensive, generation then claiming that you have these wonderful green credentials.

So then the question is really, where in the world would you locate a data center and the green energy to power that, data center where it otherwise wouldn't be used for, anything else? How will data center expansion in the UK affect data center expansion in the EU or in Trump's America? Who is doing all this?

This is the real question for me. Who is thinking about these things? I mean, I'm here and glimpsing how huge and complicated a question it is. Who is doing this difficult holistic joined up thinking, including thinking through those second and third order effects? Are policymakers in the UK thinking in those terms?

Is SECR reporting going to have any impact? The Environment Agency, they like the detail and the nuance, but their remit has tended to be a bit more narrow. Their budget has been absolutely slashed under the conservatives. Is the onus on civil society to, to work through consultations, local planning authorities on a kind of data center by data center basis?

Is it maybe up to Environment Variables? Maybe it's on you.

Chris Adams: Well, what I can share with you is that we've got someone who's leading one of the distributed data center companies to give their side of the story in a future one, precisely to talk about okay, just how you spoke about the idea of like you mentioned that quota of AI, and imagine if we only spoke about vehicles, I wonder if there's maybe a thing where we talk, there's a similar comparable way of thinking about data centers, right?

Like if we only think about data centers as one thing, rather than being like, there's a typology of this giant, gigantic out of town hyperscale data centers, like gigawatt scale. And there's one at the other end, which are not the same, for example. Maybe there's a need for a kind of different strategy to think about what kinds of data centers make sense in what circumstances.

So like, maybe that you want to have certain kinds of computation. Like you mentioned that word, like inside the urban fabric, and there's certain things where you don't want to have it because you might have a different use for this. This makes me think of actually China. So China does have something along these lines, where in China, there's a really aggressive target to A, get lots and lots of data center, lots and lots of computer computation out of relatively old data centers into much more advanced centralized hyperscale kind of facilities, which are being paired with the kind of energy bases where there are just significant amounts of clean generation being put together there.

So you've got co-locating hyperscale data centers with the kind of generation that you have. So that you have different approaches and maybe there's something that you might see like that in the UK. I, don't know, but, I found maybe it's someone we should speak to. And if you're listening to this podcast and who is thinking about that.

Please do suggest them because we'd like to cover that in a bit more detail. All right then, you've spoken about two of the things that I think we, I'd like to just, if I can, jump into. You mentioned SECR, I don't know, could you maybe expand on who that is or what that is for people who aren't familiar with that acronym?

Jo Lindsay Walton: Oh, Streamlined Energy and Carbon Reporting regulations.

Chris Adams: okay, all right, so that's basically UK government has that data centers above a certain size have to report, basically, right?

Jo Lindsay Walton: Companies, yeah.

Chris Adams: Oh, okay, got it, okay, thank you. All right then, okay, so we've touched on quite a few, we've gone into a number of different areas for this and we're coming up to time. So I guess to ask you, you've spent this time and you've put a labor of love into this report, for example, but that came out in September, in the last, in the kind of subsequent months.

And are there any, is there any kind of, what work is exciting you? What things do you want to, are you looking at, you think, "this is really exciting, I wish more people would, who are interested in sustainable software, I wish they would look at this," for example. What's on your radar these days?

Jo Lindsay Walton: Well, it's been a very kind of busy and strange couple of months. So just even as you say that, it just reminds me how quickly these things move. Basically, I feel like I'm a little bit behind and I need to listen to some podcasts and click on some LinkedIn links and, bring myself up to speed.

I continue to be delighted by the work of the Green Software Foundation. I'm a big fan of your podcast. The GARP Climate Risk podcast is one that I like. Top three podcasts, the other one would be the Bunta Vista podcast, but that's not actually about climate and environment. That's just people getting high and reading news stories.

I'm interested in further collaborative work at a smaller scale with individual kind of companies and organizations. We've been doing a little bit of work with kind of cultural heritage organizations, thinking about their carbon impact. The focus of that work is under the rubric of climate acuity.

Which we've recently launched. It's connected to the DHCC in that we have a workshop that we do called the Digital Sustainability Game. So I'm, excited about continuing to iterate that work with all the constant barrage of developments that happen week by week in this space.

Chris Adams: It's pretty exhausting. I could, I can definitely share that. I struggled to keep up myself and this is pretty much my job.

Jo Lindsay Walton: I think, yeah, I think we do need to take a break every now and then. Pause, moratorium.

Chris Adams: Okay. On that note, we're coming up to time actually. So Jo, thank you so much for coming onto this and providing extra context to the report. If people are curious, where should they be looking if they wanted to read this report themselves?

Jo Lindsay Walton: It will be in the show notes, or if you type in The Cloud and The Climate AI powered or Navigating AI-Powered Futures, I think it should pop up.

Chris Adams: As the first result in pretty all the search engines.

Jo Lindsay Walton: I hope so anyway, otherwise something's very wrong.

Chris Adams: Well, in that case, folks, that's what to look for then. All right then. Well, Jo, thank you so much for coming on to this. This has been really, fun. And let's do this again, maybe next year. Like continue this tradition of every 12 months, we have you come on and tell us what you've been up to.

Jo Lindsay Walton: I would absolutely love that. Thanks so much for having me.

Chris Adams: All right. Thanks, Jo. Have a lovely afternoon. All right. And take care of yourself. Bye! Hey everyone! Thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing.

It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundationon. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

Green Networking with Carlos Pignataro

12 december 2024 | 40 min

In this episode of Environment Variables, Anne Currie welcomes Carlos Pignataro, a leading expert in sustainable network architecture, to explore how networks can balance energy efficiency with performance and resilience. Carlos shares insights from his career at Cisco and beyond, including strategies for reducing emissions through dynamic software principles, energy-aware networking, and leveraging technologies like IoT and Content Delivery Networks (CDNs). They discuss practical applications, the alignment of green practices with business interests, and the role of multidisciplinary collaboration in driving innovation. Tune in for actionable advice and forward-thinking perspectives on making networks greener while enhancing their capabilities.

Learn more about our people:

Anne Currie: LinkedIn | Website
Carlos Pignataro: LinkedIn | Website | GSF Champion | IETF Profile

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

IETF [24:21]
Internet Research Task Force (IRTF) [24:54]
E-Impact Workshop | GitHub
Green Software Practitioner

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Carlos Pignataro: Many of the things that we do with a little bit of spin, hugely benefit the bottom line by cutting down costs, hugely benefiting the world by lowering emissions. I'm going to run this batch job whenever there's renewable. I'm going to turn off the lights and the APL, the access points on the ceiling automatically when there's no presence.

Basic things that make a difference.

I'm your host, Chris Adams.

Anne Currie: Hello, and welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Anne Currie, and today we're joined by Carlos Pignataro, a leading voice in sustainable network architecture. Carlos has over 20 years of experience in technology, innovation, and strategic thinking, holding key roles at Cisco, including head of technology and data for Cisco's engineering sustainability.

Currently, he's a mentor in residence at Duke University New Ventures, adjunct facility member at NC State University. And he is the founder and principal at Blue Fern Consulting. In this episode, we'll dive into Carlos' work on architecting networks for environmental sustainability. He's contributed to cutting edge research on how networks, which are critical to our connected lives, can balance energy efficiency with performance and resilience.

How can the networking industry innovate without compromising the planet? What trade-offs do we have to consider? Things like uptime, speed, against, or as well as, environmental responsibility. Let's explore these questions and more with Carlos. So welcome Carlos. Can you tell us a little bit about yourself?

Carlos Pignataro: Thank you very much, Anne. I am so excited to be here, first of all, and I'm happy to tell you a little bit about myself, even though you covered a lot, that was a very kind intro. I appreciate it. My favorite number is 27. My favorite color is aquamarine, just for completeness. But on a more, on an equally serious note, Important for my introduction is to say that I have passions for technology and for sustainability.

And I look at tech as a means to an end and making things better is one of the ends. I've been involved, as you said, as you mentioned in building networks, computer networks, internet infrastructure, data centers. I have also been involved in other uses of technology, such as technoconservation or conservation technologies, protecting endangered species, such as rhinos in different countries in Africa or in India from poaching using IOT, Internet of Things technology, I've been involved in building data models and information models for recycling and overall circularity. So I try to build my overall experiential breadth through trying different aspects of technology and sustainability is one on which I'm super passionate about.

Anne Currie: Excellent. That's very good to hear. And that's why you're here today. So just a little bit about myself, because I'm not always the host. I'm a guest host for Environment Variables. My name is Anne Currie. I am, I've been in the tech industry for about 30 years. I am one of the co authors of O'Reilly's new book, Building Green Software, which covers a lot of, actually covers a lot of what Carlos and I will be talking about today, cause there is a networking chapter in it.

And I strongly recommend it to everybody who's listening to the podcast. I am also the CEO of learning and development company, Strategically Green. We do a lot of workshops to help people push knowledge and engagement with green subjects within their own businesses. So touch me up on LinkedIn if you're interested in that kind of thing.

So before we start in, we dive into today's content is a reminder that everything we talk about today will be linked to in the show notes below. So feel free to have a look, read through, read as we go, read afterwards, whatever, but the data is all there. So, let's kick off with an introductory question, Carlos.

How did you become a Green Software Foundation champion? What are your goals as a champion?

Carlos Pignataro: Thank you very much. And it's a very meaningful question. When I looked at green software champion, when I was exposed to the acronym, to the three words, actually, and the GSF acronym, I couldn't help but looking at each one of the words and understand that each one had a very, profound meaning for me. Number one, green is something that encapsulates what you and your book and all of us are trying to push our industry and our different industries towards. Software is corner of what we really do day in day out and our expertise where we can actually make a difference is where we live. And in champion, I really always wanted to be a champion.

And I try soccer. I try tennis. In Argentina, I have, which is where I come from, I have a very, below average soccer scores, no, jokes aside, I, champion is such an important word because I look at it as, the first followers, and I am very moved by the way in which we create a movement on what Derek Sivers describes into being the first follower. And having these champions are like reflectors. It seems this is a little bit of networking, BGP reflectors of the message. So we all different have insertions in different parts of the ecosystem, whether it's within corporate, within different software, different repositories, whether it's different standard bodies and being a champion means being a follower and reflector of that larger message and echo it within the all the different places that we live.

So it really aligns very deeply with my passions and frankly, something that I see you as I follow you as well, is like I see the three words very clearly and I don't know how you feel, Anne, but to me it's always a little bit of a work in progress. What we have achieved, I'll play that back to you a little bit, but as a question, but to me it's like we have achieved a lot of awareness and we have achieved a lot of education and we have achieved the realization that there's a lot more and a lot more to actually do as a champion. What is, I don't know,

Anne Currie: I think you're, very right there. We've done a lot. Things are a lot better than they were 10 years ago. Vastly better. And I feel constantly buoyed up by that, but there's just so much to do still. I mean, still, when I talk to most people outside of the immediate movement, folk don't realize that actually there is a lot of good stuff that the tech industry can do to cut carbon emissions.

I speak to folk in tech all the time who are really interested in climate change and doing their bit, and they often focus directly on individual change they can make, like becoming a vegan or something like that, which is fine. But that's as nothing compared to the far more scalable change they can make through their jobs.

Through being a software engineer, there's an enormous change that we can make. There's an enormous improvement that we can make there. And it's nice when people realize, because folk do want to fix things. They do want to make a difference. And a lot of folks don't realize.

Carlos Pignataro: A hundred percent. It's so interesting as you were saying that I was, and picturing myself the concept of bring your whole self to work and the fact that we are individuals with a set of values and we can project them in different personas that we have. And for me, it was actually a part of my professional growth and realization earlier on as an engineer that I could actually bring my values to work as well. And exactly like you said, that's when the difference really compounds. It's, not like I have a nights and weekends green version. I can apply that in my day job, not only nights and weekends. And also what I find super monumental and I applaud what you all do and what we do here is there's so much more resources, so many more resources available for anyone who says, "yeah, I want to learn, and apply it."

Anne Currie: Yes. Oh yeah. The number of resources are really, is really taking off, which is amazing. I like what you're saying about bringing your whole self to work. The good thing about building green software and being more carbon aware at work, if you are a tech company, is that, I mean, a lot of people feel, "Oh, well, it's a bit unprofessional because I'm bringing something to work and actually that's, I'm asking the business to do something that is not in their best interest to do."

When it comes to cutting carbon emissions, it is in the interest of the business to do that. It cuts costs. It makes the business more resilient. It's, it makes it more future proof to the way that the energy system is going. Eventually, folk are going to have to be ready to run on renewables because that is the new form of energy.

I mean, people can do this in a way that is against the interests of the business. If you decide that you're going to rewrite all these systems in Rust, that might well be against the interests of your business. But cutting your energy use, cutting your carbon emissions in half by just being more clever and smart and modern about the way you operate systems so they're more efficient and more secure and more cost effective.

That is in the best interest of your business. That's not going against it. And I think we do need to keep constantly hammering that message home.

Carlos Pignataro: Absolutely. It's a message in my corporate tenure, within big tech. That's one of the key messages that resonated inside the company, resonated with customers, which is good for the world and good for the business and finding these things it's. And that's one that actually I don't think is a marketing tag.

I run numbers on, "shall I do this? Does it make sense or not?" And just like you say, many of the things that we do with a little bit of spin hugely benefit bottom line by cutting down costs, hugely benefiting the world by lowering emissions. I'm going to run this batch job whenever there's renewable. I'm going to turn off the lights and the APL, the access points on the ceiling automatically when there's no presence. Basic things that make a difference.

Anne Currie: Yeah. If you go back to the, I would say almost the inception of the modern thinking about being more efficient in data centers is the work that Google and Sun were doing at the beginning of this century around containers and orchestration, the use of that precursor to Kubernetes, their Borg orchestration system. That wasn't about cutting carbon emissions.

That was about cutting the cost of operating systems and improving the resilience of operating systems. Those two things are completely aligned with carbon reductions. It's the way of being more efficient, being more resilient, being more, usually adopting modern operational practices like auto scaling.

Do all the, deliver all the things. It's

Carlos Pignataro: And I'll tell you one thing, if you don't mind, allow me extrapolate from what you're saying, the idea of actually having software, green software principles, really as a, in our workspace, visible and very aware, is so incredibly important because the virtualization concept that you're talking about, Kubernetes and so on, are things that we apply to fight in the networking world, are things that we apply to any type of SFC software function, virtualization, NFVs, et cetera. And trying to bring a software more dynamic approach of how we think about it, to think that are traditionally more, I have a big router and a big switch and a big antenna, softwarizing, if you will, the thinking and doing that with the principles that you can have within, things like that are at the reach of all of us, like some of the courses on green awareness, building those principles into any type of software practice is such a win/win.

Anne Currie: Yeah, absolutely. There's no reason for people to go, "Oh, I love the idea, but it's actually going to hurt my business." If it hurts your business, you're not doing it right. If you think it's going to hurt your business, don't do it that way. Do it a way that is materially aligned with your business. That will scale better, will deliver more value, and you'll actually make it happen.

So there is no reason why being green should hurt. If it does, you're probably doing it wrong. Stop.

Carlos Pignataro: And I tell you, one of the things that we often think about, and I thought about very much in my CTO roles or in when I do standardization is how can we actually make some of these things codified, reusable, repeatable. Someone can actually learn, someone can actually use, whether it's a standard or whether it's a certification or, and do that in the context of. Not only the architecting of software and networks, but also the operationalization of networks. One small example that I can share is work that we're doing with Alex Clem on sustainable network operations, which is an IEEE, SIG or special interest group. And the interesting thing there is that when we look at the overall life cycle of networking and software, we have the use phase within the life cycle that focuses on the specific operations of the network, operational aspects of networks. There's a lot to gain in terms of managing energy efficiency and carbon awareness and carbon efficiency. So how can, within our different insertion points, within the life cycle, some are earlier, closer to manufacturing networking equipment or designing chips, some are closer to doing architecting networks and actually designing networks and operations, and then plus plusing networks, right? Like updating and upgrading equipment in a more circular, sustainable way. Each one of those areas has a very strong sustainable benefit that we can actually bring. These are the reasons, and honestly, Anne, hearing you talk are the reasons why, going back to your earlier question, why I got really drawn into this green software champion concept.

Anne Currie: It's interesting. It's, I'm going to talk about something you don't, you wouldn't necessarily speak specifically at the moment, but one of the things that interests me, and we mentioned it a lot in Building Green Software. Is that networking is one of the few areas of the tech industry where there's already been a lot of thoughts, which is directly aligned with energy saving.

That's, that networking has a concept of, is it bits per watt or watts per bit?

Carlos Pignataro: Yeah. And the interesting thing there is that there's been a lot of research and there's still ongoing discussion into how much things like energy over a number of throughput, like Watts per bit or, it's over megabit, how useful something like that is because, which I think is actually useful in many areas.

And there's, it's still an ongoing discussion about it in some standard groups, because if you take, for example, a core in a router, piece of equipment of the guts of the internet. And you have it powered up without traffic that is consuming about 80 percent of the power.

Anne Currie: Which is astonishing, isn't it?

Carlos Pignataro: It's crazy. And then with power, 95, right? So the proportionality has a smaller slope.

Anne Currie: So actually let's move on to one of our, some of our discussion points from today. So we're going to be talking about two papers that you were involved with and, co-authored, contributed to the first one is an article entitled Challenges and Opportunities in Management for Green Networking, which explores the environmental impacts of networking technology.

Noting that while, that basically it says networking is great, there's loads of things come as of networking that can cut travel, that can, could significantly contribute to reducing carbon footprints. Networking is an amazing thing. But it also uses a lot of power. So if there's some way that we can get all the wins and reduce some of the losses and some of the waste, that would be fantastically good.

But at the same time, we don't want to lose any of the wins. So do you want to talk a little bit about the article?

Carlos Pignataro: Absolutely. Thank you so much. I'm going to start, Anne, by honing into a couple of keywords that you mentioned, which are important in my mind. And the first one is the one about trade-offs. It's potentially tempting to say, "I'm going to move a part of the system to another part, which resides outside boundary conditions, and therefore things are more green." And it's one of the learnings that I've seen. I've worked in a number of technologies in my career and sustainability, environmental sustainability has been the more new one by far. We're looking at the system with broad boundary conditions. It's not like saying "I'm going to replace all the lights with LEDs" because we have material extraction to actually make the new light bulbs and we have to dispose of the old ones.

So if we look in the broader sense, we've really had trade-offs. Imagine to make it very practical that we have a link between two routers or two devices. If we add more links, we can have more redundancy. And if we add more routers to actually duplicate that and more links between them, we have even more redundancy. And that redundancy or resiliency improves asymptotically. It gets to a point in which I add more. And while my carbon emissions continue going up linearly, there's no benefit to redundancy and in fact, the system can become a little bit more brittle. So one of the concepts that we kind of talked about a little bit, we have a multi goal scenario.

We are optimizing for two different goals at the same time, one of which is resiliency and performance and traditional business metrics, and the other one is for sustainability at the same time. The main area where I feel that makes such a, strong difference is in Moving to automation, moving a lot of these processes to automated processes. It's one in which we can actually get to an optimal point in sustainability while lowering the extra links for redundancy based on the needed traffic at the time or the seasonality of the traffic or the seasonality of the requirement. So that's one of the key pieces of that work. And let me explore, if you don't mind, another area of this paper, which I think is very, relevant and important, which is, which really is how we define terms. It's a Socrates quote that the beginning of wisdom starts with the definition of terms. And one of the things that I found is that in such a multidisciplinary field in which you have people coming only from environmental sciences, people who are only tech, there's a little bit of an impedance mismatch sometimes in the dialogue.

So even simple things like saying, for example, sustainable something versus something for sustainability, right? Do we have sustainable AI, meaning AI systems that are sustainable in themselves? And we call that the footprint, or AI systems that are AI for sustainability, meaning the output of the system can actually help you with sustainable outcomes and we call that the handprint, right?

So the concept of footprint and handprint are not necessarily well understood within the networking and software spaces in my experience.

And I feel there are fundamentally, when I reduce the footprint, you want to improve or enhance or grow the handprint.

Anne Currie: Yeah. Yeah, I agree. It's not a well known phraseology. we use this in Building Green Software. It is a bit counterintuitive in that kind of like footprint, bad hand print good. But it's like, why is the foot so bad if you are in, if you are judging by your, reverse of the football analogy where hand is bad, foot is good, but in green software, we talk about footprint, foot being bad, and handprint, hand being good.

Carlos Pignataro: That's right. In football, you would get a red card and in green software, you're going to get a green card.

Anne Currie: Indeed. Indeed. But I don't terminology because I think it's a tad confusing, but it is well known that is, and that's what we're talking about here, which is fine.

Carlos Pignataro: Thank you. No, for sure. And I'm going to bring back also something that you said before, which is the networking industry have been thinking for quite some time in many sustainability aspects and at different levels, at the cheap level, as I was mentioning, at the power efficiency, there's metrics, so you cannot improve what you cannot measure trackers.

So we have power efficiency for data centers and so on. And that becomes much more and more important because the amount of electricity that gets consumed by networks and particularly data centers these days keeps growing and keeps growing. There's many things that we can learn from the way in which we design protocols.

Part of the work that I've done historically is protocol design within the Internet Engineering Task Force, the IETF. Reading RFCs and things like that. And there's very interesting work on protocols for the Internet of Things. And that is interesting, Anne, because IoT Internet of Things devices, just by their use case, need to minimize power consumption, need to be alive for years and years on battery power.

So protocol definitions evolved so that protocols are a lot less chatty. There's a lot less back and forth and wake up. Devices can actually go to different levels of sleep. The same way that we have sleep levels in the laptops that we're using you and I right now, and on the phones that on the handheld that we have. There's less of that in some of the networking areas, and that is an active area of research and development. How to bring dormant states, less chatty protocols into some of the networking arena.

Anne Currie: That is something that I, really liked in this paper. So stepping back for a minute, there's the IETF and the IRTF, which are both parts of the same organization. The IETF tends to focus on protocols, designing protocols, but actually, and the way it does that is it, talks about a famous, Definition of how, it approaches things, which is rough consensus working code, which is based on the fact that actually it's very hard to tell what's going to work in networking because it's so complicated until you actually get something out there and see whether it works or not.

And IRTF, which is a lot of what these papers are, is about kind of research and thinking and, but it's hard to marry those two things I would imagine in many ways, because the whole point is that you can't necessarily just think through which network protocol is going to work and which are not.

You almost have to suck it and see. But one of the things that you pointed out in your paper, which I liked, was there are already protocols out there that are working, that are achieving what we want to achieve here. There are already protocols for, Internet of Things, which are out there working. And it would be great if people looked at them and said, "What works there and what doesn't work there?"

"How can we apply that learning in other environments?" I think that's what you're saying.

Carlos Pignataro: A million percent. That is actually in a much more eloquent way with a couple of the very important points that we try to convey in the paper. Thank you very much, Anne, for explaining IRTF, IETF, ISTAR, it's, I wasn't sure if we want to, how boring that was, but it was actually very useful. So you put both of them in context.

Anne Currie: Well, what I quite like to do all the time when we're doing it, what we know works in terms of delivering good stuff in the human, in human history is to look at stuff that works in a similar but not identical field and say, let's steal some of those ideas. And something else I liked in your papers was when you talked about CDNs.

And the ideas in CDNs, Content Delivery Networks, if people aren't familiar with them, they are what I think one of the most interesting architectural approaches in networking and in technology above the network as well. And there were tons of ideas that feel that they could be borrowed. And you said the same thing in your, paper.

What was your, what's your thinking on CDNs?

Carlos Pignataro: Yeah, no, it's, exactly that. Thank you for it. Because the main point is exactly that. There's many areas that there's already been deployment, not only research and not only development and not only test and Q& A, but deployment, that can be applied to new things that we need to do today, right? So Spoton on that, IoT is one of them. And because naturally there's not necessarily a very smooth bridge in my experience between research and standardization and actual deployment and running code is a little bit bumpy. So having examples and use cases that work, that we can apply to the problems between code that we have today is critical.

When we look at the most dynamic and complex networks, I really look at CDNs. Because it's a network that is actually focused on delivering the content and in a CDN, it's incredibly critical to number one, replicate content near the receiver, right? So that you don't have to stream from transatlantic, but don't over replicate if there's not a lot of listeners and receivers. So the equations can really, help you to minimize the overall end to end system electricity, consumption, and maximize efficiency just because of. What to replicate, where to replicate it, at what times do we do this when we have a signal that the electricity feed that we have is coming from renewables. It's one of the systems that really gives you the flexibility to implement all of the things that we discuss in a paper. And if you allow me again to extrapolate a little bit more, I frankly think that talking about green and talking about sustainability, we can actually extrapolate further. And look more into what nature does and try to understand and replicate that in some of our systems.

The fact that our laptops hibernate, the first there were animals that's where the word come from, during the winter, they were saving mode, right? It was a bear in saving mode and we have a laptop in saving mode. And next we're going to have a data center cluster in saving mode potentially. And. Many, ways in which if we look at the amount of energy that our brain uses versus an LLM system uses, there's clearly a huge, ginormous opportunity for improvement.

Anne Currie: Yeah. Yeah, that's true. But yes, that actually it's interesting that you mentioned LLMs and AI there, although not, directly, but I'm very interested in networking and LLMs and how those are going to be merged together in a green way. I, as I said, I'm a huge fan of CDNs. So as you said, CDNs use buffering effectively close to users to do two things, to mean that every time somebody that, if you've got somebody in London and they want to look at a huge asset that was served from the US, maybe an episode of Game of Thrones. I still use Game of Thrones, even though it was a bit out of date. It's better to have that episode, one copy of that episode, move over the Atlantic and be cached somewhere local to the user in the UK.

And then they, and then all the, users there just take it over a shorter distance rather than have to take it all the way from the US, as you said. There's another benefit, which is that you can move that giant asset at times when the grid is, the internet is not busy, so it flattens, something that you've said in your papers you talked a lot about was the efficiency of flattening peaks in load, peaks in demand.

It's much better if you could spread demand, find some clever way of spreading demand out so that you don't have peaks that you have to provision for, because that means you need more equipment and more of everything. It's not as resilient and it's more expensive. So, CDNs are fantastic from that perspective, but AI feels like it could potentially benefit, AI inference feels like it could potentially benefit from CDNs as well.

If you could try and cache responses that are common questions so that you weren't having to run everything, sorry, I'm taking you off networking now. I don't know if you've had any thoughts about AI and networking and CDNs or anything like that.

Carlos Pignataro: I work a lot on AI these days and I definitely, have thoughts. I think number one that, and thank you for, the bunny trail towards where the dialogue takes us and things are relevant, right? The way in which we actually train systems today can like immensely be improved and whether that's by some mechanism of incremental caching so that you don't have to relearn everything as you actually tweak the model and things like that, CDN like, absolutely.

And in another way in which I really think about it, and frankly, particularly with a couple of startups that I'm either working on or following or seeing is, do we want a, like the typical army Swiss knife B2C business to consumer that can actually solve everything and we need three cities worth of electricity to train? Or do we define more constrained SLMs instead of LLMs, small language models that are a lot more domain specific and a lot more domain shifted and more B2B potentially, business to business type. And regardless, I think that going back to sustainable X, versus X for sustainability, I always like to do like two by twos or X, Y.

And I think that AI as a broad technology from whatever, from machine learning and computer vision and, has not significantly into going AI for sustainability, we have Google maps today that can actually give me the most sustainable travel, fuel efficient route, and I go in to book my flights and I see carbons of each one of those.

And there's a lot of AI that is applied for sustainability. There has not been enough, or I should say, there's a enormous upside and opportunity for sustainable AI, right?

Right. So in the backwards lingo that we were using. A lot of handprint, not enough reduction of footprint.

And I think a lot of methods that we know from other domains, like CDNs, that we can apply to inference and learning of models. Absolutely. Please let's do.

Anne Currie: Yeah, I mean, in many ways, it feels like it is the lesson to take away from the internet, which is that it's really hard to make things work. You have to, like working code. Working code absolutely is the king. So if you can take someone else's working code and apply it in your situation, that's a great idea.

Carlos Pignataro: Exactly.

Exactly. A million percent. And one thing, Anne, that I'm going to mention now, because I fear that the way in which we're choosing conversational forking paths, I'm going to forget. So, so. Because it's important. It's a call to action. And one of the things that you mentioned, I mentioned, is how much more resources and material exists.

And really my call to action is to go to learn.greensoftware.foundation and start with the green software practitioner curricula. It is such an incredibly well packaged. Set of modules that go through a lot of the fundamentals and demystifies it provides lexicon, it demystifies, it talks about carbon.

And one of the important ways in which I wanna make this actionable is to really encourage any listener to, is super easy: learn.greensoftware.foundation.

Anne Currie: I totally second that. And I really, I apologize, Carlos. We've gone all over the place on our discussion today. So you're quite right to stop me and make sure that's a very important message got through. So, I mean, we are now coming towards the end of it. Is there anything else that you want to tell?

There's tons of interesting stuff in your two papers that are well worth reading. You don't have to feel that it's too, that it's too difficult. The papers are quite accessible. Is there anything that we haven't talked about that you would like to talk about?

Carlos Pignataro: Thank you, Anne first of all, thank you for actually reading the papers and, actually not only reading, but really reading because you actually distilled some of the fundamental principles that we, and I wanted to convey. More than anything, what's needed to drive green software is our full commitment and bringing, like you were talking about, like I was mentioning, our values and whole selves to the one which we call and the one which we do finger to keyboard.

This is a very multidisciplinary nuanced area. And after leading in Cisco technology and data for the engineering sustainability team, we really don't know what we don't know. And for me, learning and think humbly in every conversation is fundamental goal. One of the things that I love about the approach that the GSF is taking with, SCI is that it's data driven, right?

Let's get matrix, let the data as opposed to myths drive the conversation. And to continue to stay together because the ecosystem is multidisciplinary, and we all learn a little bit from each other and reusing and leveraging particularly code and principles that you talk so, so very well in your book and those principles, that code, bring it in our insertion points within the ecosystem, whether it's at the vendor or if we're cloning in a public repository and make some changes, or if we're thinking about the mixed networking protocol or networking operations.

Anne Currie: Yes, very true. So I think that's a very good place to be finishing up because we've pretty much come to the end of our episode. And I have one final question for you, which is where can listeners go to if they want to find out more about you? Obviously links to the papers will be in the show notes, so you can, you should, I strongly recommend you to read them, but where else can people find out about you?

Carlos Pignataro: Hey, thank you very much, Anne. LinkedIn is an easy place to go. And I'm always open to any connection and any messages. My website, you can check out also bluefern.consulting and has my email, has my contact, I really, I don't just say that IQ and you respond, I respond. So, super happy to continue the conversation and continue engaging.

Anne Currie: Excellent. That's very good to hear. I'm sure that lots of our listeners will reach out and talk to you, but certainly they should be reading your papers and they should be connecting to you on LinkedIn or looking at your LinkedIn, following your LinkedIn. And so thank you very much for being on this episode.

It's been a fascinating episode, a deep dive into networking and all the... networking is not so, it's been very interesting because a lot of the concepts are also applied to non-networking software. All the ideas and the overlap in CDNs, which really are the concept I think that's best suited to environmental sustainability and aligning with renewable power in the long run. And a final reminder to all our listeners that the resources for this episode are in the show description below, and you can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables. So see you all soon. Bye for now.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

The Quantum Entanglement of Software Sustainability with Wilco Burggraaf

5 december 2024 | 52 min

In this episode of Environment Variables, host Anne Currie speaks to Wilco Burggraaf, a lead green practitioner and architect at HighTech Innovators, for an engaging discussion on integrating sustainability into software development. Wilco shares his journey into green software, the inspiration behind his innovative workshops, and his efforts to build a vibrant green tech community in the Netherlands. The conversation explores his articles on the Software Carbon Intensity standard, the complexities of balancing micro and macro sustainability goals, and the synergy between FinOps and green software. Tune in for actionable insights and strategies to make greener choices in tech while aligning sustainability with business goals.

Learn more about our people:

Anne Currie: LinkedIn | Website
Wilco Burggraaf: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Green Software - The Netherlands | Meetup [50:48]

Resources:

Software Carbon Intensity (SCI) Specification Project | GSF [41:14]
Green Software Maturity Matrix [46:16]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

Transcript below:

Wilco: At some point I came to the conclusion, like, okay, we can measure a lot of things, we can have all these metrics but at some point the numbers are not going to change outcomes. Decisions do.

I'm your host, Chris Adams.

Anne: Welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host today, Anne Currie. And joining us is Wilco Burggraaf, lead green practitioner and green architect at HighTech Innovators. Wilco brings a wealth of experience in software development, been over 20 years in the industry, and is an active Green Software Foundation champion, and we'll be talking a lot about that today.

So he cares a great deal about integrating sustainable practices directly into the code and architecture of software, helping to make greener choices not only possible, but essential in tech. And in this episode, well, this episode really is the Wilco show. We will be talking about three articles that he's written on LinkedIn

and what they mean and what people should learn from them, what he's learned on his journey in becoming a Green Software Foundation champion, a green software practitioner. So yes, he has a lot of interesting thoughts on integrating software sustainability at the lowest, the deepest level, the lowest micro scale, the code level scale, and the macro scale.

Bizarrely today, we're going to be talking about those in the reverse order, focusing on his articles on the micro scale first, and then moving over to the macro scale, which I, and I'm a big fan of macro scale. So that'll be interesting when we get there. Actually, I'm a fan of all the things, but I'm a big fan of starting at the macro scale.

So yeah. So Wilco's going to be talking about his articles. And he'll also be talking about his experience using the SCI, the Software Carbon Intensity standard and the Impact Framework, because I'm very keen on his thoughts about whether they're useful or not, why they're useful and what they add to the software development process. So, welcome, Wilco. Can you tell us a little bit about yourself?

Wilco: Yeah. Hi, Anne. Yeah. Thanks for having me. Big fan, by the way. I love the book you worked on, Building Green Software. So I'm Wilco. 41, married with no kids and I live in the Netherlands. We have an Airedale Terrier named Iron. And although the country I live in is small with only 18 million inhabitants, I grew up in the South near the coast on a factory plant tied to the coal industry in the eighties. And my dad was a night guard. So we lived in the factory plant and yeah, when you come out of the bed and you smell the stench of chemical processes in the air and when the wash was hanging out to dry, but beautiful weather, but, the coal dust came on the clothes.

Yeah, that was, but yeah, if I look now back on it, that was kind of weird, but that was normal. That was, home for us. So I deeply love nature. I spend a lot of my time on hikes of two or three hours in the forests and the heat lands, and that's only 10 minutes from my home. So, yeah, I love to live here in the South and what we call the nature.

Anne: That's great. That's lovely. And a really interesting backstory that your first,

coal was your nemesis was your laundry's nemesis from a very early age.

Wilco: Yeah. It's always a story that my mom tells people because yeah, a lot of people who didn't experience that can not have a understanding of how it must been.

Anne: That is really interesting. That is a very interesting backstory. So my backstory is not quite so interesting. So my name is Anne Currie. I am, as I mentioned, one of the co-authors of the new O'Reilly book, Building Green Software. And I said that in the last podcast, and I'll say it again, if you care about this kind of stuff, if you're listening to this podcast, Building Green Software from O'Reilly is a really good book to read to get cracking. And it doesn't, it's not particularly techie, it is useful for everybody. So if you're a product manager, if you're a marketing person, you can read that and understand it. And it's a good, place for you to kick off because I think a lot of the changes that we're going to need to make to build green software actually start with product managers, not necessarily with techies, but that's an interesting other point. I'm also the CEO of the learning and development company, Strategically green.

And we do workshops as Wilco also does workshops. We'll be talking a little bit about that later, but we do workshops to kind of get your company started on getting people understanding what it is to be great and kicking off some interest and excitement, as well as helping you build some internal expertise in that.

So if you want to do any of those things, hit me up on LinkedIn. Before we dive in, I want to make a quick reminder that everything we talk about in this podcast today will be linked to in the show notes below the episode. So you can go and you can read it and you can follow along as you listen to the podcast.

So back to you, Wilco. I think the place for us to start is what started this off for you? What kicked it off? What led you to transition into green IT and how has your journey evolved over time?

Wilco: Well, only 10 months ago, it's not even that long ago, I dived into green IT and sustainable coding, starting with no background in green. In IT, of course, with 20 years of experience. And. Now I'm progressing to discuss things with university professors. So it went kind of quickly. And also since March this year, I'm a co-founder and co-host alongside Pini Reznik, I think a familiar person for you, of Green Software Meetups in the Netherlands.

Anne: And you've had a lot of success with meetups in the Netherlands, which is really good. So, what, role do you see your current work playing in the larger mission of sustainability?

Wilco: Well, maybe a fun detail. I work in secondment, for some countries that is not a familiar thing, but it's meaning I'm contracted by various companies. And this year I'm working with the National Databank for Flora and Fauna as a solution architect and together with a fantastic team, we're making hundreds of million biodiversity observations publicly accessible to everyone in the Netherlands.

And that is kind of something really cool. And we're on track to reach our first major release in the new year.

Anne: That is very cool. That's very good. And that quite interests me, links back to something you said on LinkedIn when I was talking about the last environment variables, where I was talking to Stefana Sopco, who also lives in the Netherlands. And you pointed out another Dutchie. Which, you're quite right.

We have a, there's a lot of interest in this in the Netherlands. Do you have a feeling for why that is?

Wilco: I hope I helped a bit with that the last half year. But no, of course, no, that's a, just a joke. But when I started like at the beginning of this year, I was looking on Google, searching for information and information was hard to come by. And at some point I was thinking, yeah, of course, books and podcasts, the GreenIO and Environmental Variables.

That is a place where I find a lot of that information that I needed. But at some point I was like, okay, so maybe I need to talk to people to gather more information. And when I was searching on LinkedIn for people who knew more about green IT and green coding and green software, I found out that there were all these kind of bubbles, yeah, in the Netherlands we call them bubbles, like you have 20, 30 people working on a certain topic.

And I was like, also at the same time, we were thinking, okay, how can we build a community for the meetups? And I was like, yeah, the only thing that I can do is connect to these people and make aware that the other bubbles exist and to keep on doing that. And when I was finding out, and I found the other group and another group and eventually there are, I think, right now, yeah, I think a small 2000 people in the Netherlands busy with this topic.

But a lot of those people are not aware of each other. So you have to think about people working on CSRD and monitoring, people on FinOps, but are really that are interested in sustainability, people who are like, "yeah, we, need to measure not only emissions, but also nitrogen and other things and PFAS," is it how we call it in Netherlands?

So yeah, I don't know if it's because of a trend or because of a lot of people now with CSRD are looking, "okay, how do we need to do this?" But yeah, there's a lot of activity in the Netherlands.

Anne: That's, that is really interesting and there's a lesson there for anybody who wants to grow a community is that you went out and found all the small communities and hook them together. That's an incredibly valuable thing to be doing.

Wilco: And it's also cool that there are also, there's an organization, the National Coalition of Digital Sustainability, and it's a little bit different, the acronym in the Netherlands, but they are already busy with this topic for more than 10 years. And then when I was doing my thing on LinkedIn, and then I found out that there was an other meetup group from a bank and a consultancy company, and then we're already busy with doing meetups in the year before.

And, but they weren't aware of even sometimes other organizations and also like a Green Software Foundation, but there's also of course the Cloud Native Computing Foundation, where you have a sustainability group. And I'm not even talking about things like Climate Action Tech and, that kind of organizations of groups.

Anne: And of course actually trying to link together these groups is incredibly, so we have actually met in person, we met at a Green IO conference in London in September, which was great. And that was very good. That was a very good way of getting a whole load of people in Western Europe basically to all connect together and have a drink and see one another face to face.

Very effective. So..

Wilco: Yeah, it's very inspiring to see other practitioners and also other perspectives from UX to Green Ops to yeah, all the different, because that is something that is so clear. And this is also maybe eventually, if we go to macro, why it's hard to implement is because sustainability, it hits so many fronts within a company or an organization, there are so many roles.

Where if you start thinking about, "okay, what am I, actually doing?" So the impact, what we're doing from boardroom to eventually DevOps teams or members of DevOps teams. And it's cool to see that all those people come then together in such a conference. Yeah.

Anne: Yes, it is true. And I've said this many times before that it's, everybody's being bonded together by having the same goal, which is reducing carbon in the atmosphere. An intrinsic goal that's, you know, it's doing good. It's improving the world. And it does mean that you can share common ground with people you wouldn't necessarily previously have shared much common ground with.

Wilco: Yes.

Anne: So, I mean, you, said you've only been interested in green software for about 10 months and you've certainly done an awful lot during that time. How did you get started? How did you get started organizing meetups particularly?

Wilco: Well, in January, my former boss introduced me to the Green Software Foundation website, and I immediately noticed two things. So CarbonHack24 was on the website, the Hackathon and the company I work for, they really love Hackathons. So I formed a group of volunteers together. And beside that, I will come back to that later.

There was also a lag. I saw the website of meetups in the Netherlands. So I reached out to Asim for advice and he connected me with a group of Green Software Foundation employees and the Green Software members in the Netherlands. And including with Pini Reznik and together, we started planning. And by April, we had our first meetup.

And my team even won the CarbonHack24 Best Contribution, which is crazy if I think about it, which was such an incredible motivator. And each step I took from organizing meetups to winning the hackathon felt like a chance to make a meaningful impact.

Anne: Which is fantastically good. That is good. So, but you didn't stop there, did you? You became a green software champion, which is a new Green Software Foundation kind of a project to build up people who know more and can go out and shout about green software. How did that happen?

Wilco: So by May, after hosting two meetups and writing over 10 articles on green software, I felt certain that this was my calling, right? I felt so much passion and fire. So, I mean, I think through all the content I create and all the conversations, that was kind of clear. And I discovered the Green Software Champion program on the Green Software Web Foundation website.

And I knew it was the right path to amplify my impact because I believe that if you have a recognition of a certain organizations that especially like multinationals and big organizations are like, "Hey. This is something that we maybe need to take more seriously." Not because of me, because of the, "Hey, this is, there's something going on here."

And yeah, forward four months and we've organized five meetups in just over seven months. And with the sixth one on the way on the 22nd of November, with the Green Waves Hackathon at the TU Delft, that's a university in Delft. And I now written over 150 articles on LinkedIn, collaborated with professors to bring green theory into practice.

And we're still, I'm still doing that and have given six talks so far. So, with more plans and each step has deepened my commitment to building a sustainable tech community.

Anne: Which is absolutely fascinating. It is amazing how much you've done in 10 months. So, but what next for Wilco and the green IT community?

Wilco: Yeah, what is next? That is a good question. So what I really try to do is to follow this certain path. So when I started gathering the information, I found kind of out, okay, there is already a decennia of research done and a lot of information, but to some degree, we have a hard time transferring this information to other developers and we are kind of stuck.

So for me personally, I was really invested. "Okay. How can I make this first stepping stone on making this a thing that other people can understand?" And that's why I started to invest what I now call EQUAL,

Energy, Quality, Utilization, and Load. So the idea that you have an application that has a certain algorithm or a certain logic that everyone understands.

So kind of a loop. And if you have this loop, so a few important things in, in, in green coding and green software is, okay, how can we, based on utilization, can we estimate and not exact utilization, but can we estimate the energy and then eventually related to emissions? So if I started understanding, hey, wait, utilization to some degree is like the amount of threads that the CPU is running.

Also, of course, based on the cores. So 50%. You would maybe expect, like, if you have like 12 cores that with 50%, six cores are running, but it is not necessary at the truth because frequencies, of course, can be higher and lower and there are some things going on, but if you start, okay, so the amount of threads, so let's say the 12 cores.

So you can have then most easiest for a clock, 24 threads. So if you have a loop that you can start playing around with two threads, four threads, eight threads, 12 threads. So that's a first parameter you can give to this loop that I placed in the API. Then the next one is the amount of iterations. So do I want to do a small test?

And the funny thing is one line, one normal line of code, because I can make a line of code that just can gas pedal the CPU 200 percent for an hour for one line of code. If you take an average line of code, it's most of the time so insignificant for a CPU that like, if you have a loop that is running within this very small time, then 10 million iterations is quickly over that's very fast.

So my EQUAL starts with 10 million and it goes eventually to in the billions of iterations. And then the third parameter of equal is the use case. So you can place in the iteration just an I++ or just any use case you kind of want. And then what I kind of start doing when the loop is running, I start asynchronically, I start measuring the utilization of the cores in a very high time resolution. Like 10ms, 20ms, 30ms, so very small. And then after the whole loop is done, what I then do is I can place, those samples, I can eventually connect them back again to the traces of the code. And then you can see a certain few things happening here. So what you can see happening is if you will reduce the amount of operations happening on the CPU, yeah, of course you probably,

your, utilization will be lower and your energy use. But there's also another thing is because CPU, how CPU works, that sometimes you will see unexpected behavior. So although you start to play around with these use cases and you think, "Hey, this should be more efficient" and you start rerunning it and then you're seeing things happening like, "Hey, wait a minute, if I run this on 18 threads, this use case works more efficient energy-wise on 12 threads.

How does that make sense?" Well, that is something I tried to figure out, but this is what I place in a demonstration style, because if you demo this and you show us this loop and everybody understands the loop and you show this in the user interface, and then with Prometheus, with eventually graph set that you show to everyone,

then it's makes more clear for, "Oh, okay. Wait a minute. There is beside time efficiency, compute, power, and there's also this third dimension, energy consumption. And it has sometimes another effect than we sometimes expect." And I started, okay, so if I can eventually use this in talks, if I can start using this in eventually a use case for blogs.

And eventually this is also where the workshop that I'm going to give from January is built on. So this is for me is like the future. And then my estimation model that I just created on Prox. Which is kind of built in, of course, in your Linux kernel with just a dumps is now not the most perfect model, but this is the reason why I work where I have contact with, especially University of Groningen to make this model eventually better with socket management of measurements and real kind of measurements.

And yeah,

Anne: So that so the all sounds very, so basically you're working on a tool that helps people measure, at least proxy measure, their carbon emissions through energy use,

then tune it and improve it. And I'm guessing that there's kind of several advantages to that tool. If you work on that tool to deliver the same functionality using less energy, the product, your application will run faster.

As you say, CPU cycles are another proxy for energy use. So is that commonly what happens? It improves the performance of the application?

Wilco: Well, if you say performance in time, well, this is a funny question. So if your focus is on performance in time, sometimes if you say I make my code quicker, it sometimes start using more energy. So, and then the question is, I have this value, "is it okay for a user to wait on it or does it need to have this very fast?"

And there's also a difference between the performance if, and this isn't on the, on the, in the cloud, on a server, almost impossible. If you on A CPU only use a few cores, it has often a very higher CPU frequency, so it probably will be with this exact same code will, be quicker than if the complete CPU has 80% of the cores or a hundred percent of the cores active because the higher, especially with a hundred percent because of heat, the frequency goes down and it kind of becomes slower.

And this is what I say, you can, of course, if you lower the amount of code or operations to the CPU, it will eventually be more efficient. But there is also this thing going on that the CPU has sometimes 20, 30, 40 percent influence based on the state of the CPU it's in. And yeah, your code can have some influence on it, but it's more in a different way.

So how many threads am I spinning up? Or how many things are going on this server that I'm running my code on? And yeah.

Anne: So yeah, I see now why you, and when we go over onto your next, onto your second article, talking about trying to balance these micro line level changes with a more macro perspective. That's yeah, it gets quite complicated and you don't always know what's going to work until you try it. So obviously, you know, the whole point of running this tool will be to make a more energy efficient application within your kind of high level goals of your SLAs. But I'm imagining it's also quite fun, that it's quite a good thing for a hackathon, it's quite a fun thing developers to play with.

Wilco: Yeah. Yeah. That. And also you can just replace, that's what I every time keep saying. The loop is just for demonstration purposes that people understand it. But I use this whole logic in an API and you can just put your own code in there. That's the whole thing that we're, with the workshop, going to do.

So people will build their own API. And then with the same process of asynchronically measure when you run this code, what's happening, because you will see funny things going on when you're waiting or things are connecting to a database or connecting to another API. And based on how things are programmed.

So are you waiting with a loop that is pushing your CPU high? Or are you using smarter mechanics so that there is a drop, but is a drop sometimes something you want because if you want to be very efficient with your resource, you kind of want to maximize it around 80%. Well, I don't want to be come too fast to conclusions yet, because I think we still need to figure out what the patterns are and what are good patterns and bad patterns, but yeah.

Anne: Yes, because as you say, and again, this leads back to the kind of macro micro picture. It's definitely.. In certain circumstances is definitely the right thing to do to if you're waiting on an API call or something to kind of say, "right, I yield all the threads and everything running on the machine to somebody else to use the machine while I'm waiting" so that the machine is still highly utilized whilst I'm waiting for my API call, but that relies on you having a design or an architecture, which might be within your application, but it might be within your operational decisions. You know, are you multi tenant? Is there somebody else or some other company or some other application that is going to be able to pick up and use the machine while you're yielded?

But if you're just waiting around. then that's less good. So then the machine is just going to waste during the time that you're waiting. So you're, right in saying that there's so much context to this.

Wilco: Yeah. Yeah. And, okay, so, okay, I'm really getting excited about this topic. So, because I instantly thinking like, yeah, of course in the cloud, in the server, you maybe they have not full control, but one thing we know the grid is getting fuller, the electricity grid. And one of the things is that some university have research done is how can we optimize the devices that we have better?

And one of the things that we in the Netherlands have, we have a lot of people with solar panels. We're not using optimally the, electricity that is generated by the solar panel. So if you think about the following, so what are the devices that you can easiest, how do you say, charge based on the solar panels, then it's mobile devices or tablets, or maybe a laptop with maybe a good battery lifetime.

So now we're from originally, we always been like, "okay, we need to move logic to the server because it's more secure and you can not manipulate it." But if you start thinking about, "Hey, we need to optimize the devices better." What if we start using a WebAssembly in a better way? So things in your browser or on IOT device, or in this case, then on mobile or tablet, and use that green energy, especially if someone is as smart that like not charging it at night,

but more than the day when the solar panels are active, that is always like an important catch. And of course the solar panels have some embodied carbon, but yeah, but still, so there are so many cool things you can go on in this.

Anne: It is really interesting that renewables, unlike, you know, you aren't going to run your own coal fired power station. You say, nobody apart from Wilco has ever lived on a coal fired power station, but particularly solar, it's a very distributed technology. There are lots of people, I've got solar panels, and when they're running and the sun's shining, I've got more power than I know what to do with.

When I had, when I got it installed, I said to the, chap who was installing it, "what should I do with this? You know, and he said, "Oh, well, you will have, there are times when you are going to have more power than you know what to do with." So make sure that you, it might seem historically, it's always been very inefficient to heat, your water with an immersion heater rather than using, a gas burner or something like that. So it's a very inefficient way to do that, to heat the water. But if you've got free energy and it's just otherwise going to go to waste, heat your water with an immersion heater. He said, "get a swimming pool." Not that I did get a swimming pool, but get a swimming pool and heat that up because some of that has changed the way we need to think about, we still talk about green software very much from the perspective of efficiency and improving efficiency and reducing waste.

But I would say even more importantly than that, it's about doing more when the sun's shining. You know, don't forget efficiency when the sun's shining. You might want to write applications that are very efficient, that operate in totally different ways at night and while the sun's shining.

Yes.

Wilco: Yeah. I would even dare to challenge the following. So. If the sun is very shining in the day, what I see in the Netherlands a lot is if you look to the electrical grid, especially also on windy days, that's, or somehow, and I don't see the relation yet, is that the industry seems to be working harder in the general.

So you still see the gas turbines in the Netherlands emit a lot of emissions. So it's very sunny. And if you go to electricitymaps.com, to the Netherlands. And you look there, you will see then the solar panels, generating a lot of energy. And sometimes of course, also the wind turbines, but that's also the gas turbines.

And that is mainly because there's a higher demand or there's instability on the net. And so you could even start. And that's why I think that carbon-aware is a very complex topic, because are you gonna do a weather forecast and then run, but then find out that maybe the grid was emitting more than you expected?

And lately, the last days, we had a lot of emissions in the Netherlands. Or are you more going to try to indeed optimize the devices we already have that maybe run on green electricity? There's no perfect answer in this, but we need more data, we need more access, but I understand from security standpoint, even with electricity grids, I mean, they want us to give us the information, but they're scared for terrorist attacks or for things that the information that we want to do for good that can be used for bad.

But yeah..

Anne: I mean, and quite often, grids just don't have the information yet. I mean, there is, for carbon awareness, we were a long way from having really good data on that. So I always tend to say, don't start, well, you could pick proxies, perfectly reasonable to pick a proxy because actually the difficult thing is designing systems that can respond.

That's going to take years to do. So you can, in many ways, pick a proxy now, even if it isn't great. Design a system that is responsive to that proxy. And then as that proxy gets better, your system will get better. So you might be going, " actually that proxy's terrible" now, but the difficulty is, well, getting the data is often somebody else's problem.

It is put pressure on, you know, suppliers and energy grids and everything to provide good data to us. But in the meantime, the big job for us, the thing that's going to take us a long time is redesigning our systems to be able to respond to that data. So that's things like thinking about what your graceful downgrade options are for when the grids are very dirty. You might have to move big, having big latency-insensitive tasks that you can move to when the sun's shiny. The Texan grid is doing a lot of good work on that. And I talk, again, I'm sorry, I talk endlessly about large, flexible loads. So the Texan grid is putting out a call to industry for large, flexible loads, which you can run, which are latency intensive, don't matter when they run.

But they can run when the grid is full of solar because Texas is quite rightly putting a whole load of solar panels because it's very hot in Texas, it's very sunny in Texas and there's a lot of desert. So they want something to run on that, solar power. It's very sad at the moment that the people who are really responding to it are the Bitcoin miners, but AI is another potential customer who have large flexible loads.

So very CPU intensive loads.

Wilco: Yeah. Oh, but this is perfect because, okay, you kind of influenced me when we had talked in London. So to think more about, okay, if you can better react to renewables in a more flexible way, because I started thinking about it, especially if you put it in the following perspective. So most front-ends have a lifetime of two, three years.

There is of course always shorter and longer. Back-end systems often have like a longer lifetime span. So if you build something today in 2024, and it runs five to six years, that means that it still runs in 2030 when we have our big first milestone that we should have reached. So if you're not building your software today that it can adapt, that it can be flexible, you have to refactor things in the future, or you're getting in a stuck position because most of the time, especially with big systems, the more you build, the more dependencies you get, the harder it becomes to eventually change things when that foundation is set.

So, yeah, I really like that idea that, although we maybe not have all the right answers now, and maybe the situations are not always perfect, but it doesn't mean that you shouldn't start thinking and implement in this way.

Anne: Absolutely, I think it's going to take ages to do this, it's a completely different way of thinking about it. I mean, there are tools out there that already exist, that can help you get into this way of thinking. So I'm a huge fan of spot instances on AWS and, or Azure or, preemptible instances on GCP, because they're, a kind of mini version of Texas's large flexible loads.

You say it's a small flexible load. You say. What you're saying is "I've got this load, it's flexible, run it. I don't really mind where it runs. I don't have any particular SLA associated with it." And you can use it for, the clouds use it to improve operational efficiency, which is good for green as well. But in the future, I can see those loads absolutely match to what we're going to need in order to shift work forward in time or later in time or forward in time.

Wilco: Yeah, I really believe to be, to some degree as much in control as possible because it's easy to let some other company or SaaS solution fix things for you, but especially from a board perspective, it's a good idea. Like, okay, we're working together with hyperscalers and we're doing things serverless.

And especially if you're a big organization. And then mainly like if we do serverless, they are kind of responsible to fix if the utilization on the background is well organized. And I find this always very interesting because yeah, to some degree that's true. So for consumers, normally, if you go to Azure, you're probably, if you're using serverless, I think they can really optimize it very well.

But if you have a very big multinational where kind of, they already reserve a certain space for you in the data center and you're running serverless. I'm very curious because we don't have the information now. So do they reach indeed that more utilization because you work serverless or do you have virtual machines where often in a while some function comes by and it runs and it's done?

So that's why, and you mentioned spot, that's why I like, and not because I have stocks in them, because it's not possible, I think, but I, that's why I love Kubernetes and Cloud Native thinking so much. That's also why I really like to check out also every time what's happening in the Cloud Native Computing Foundation environment, because course that is where Kubernetes is very active, because I strongly believe that if you are in control as far as possible, you have, not only you can better measure what's happening to some degree, although you're doing some estimates, I think it's also from a security perspective, it's a good idea. And I think also from just the willing, yeah, the willing to be responsible because nine of the time you're also responsible of the value that's running in your cluster.

And I think just outsourcing the company's core values or the product values outsourcing to somewhere else, it's possible, but you're giving them also some control away. Yeah. And this is something that I think a lot about, but not having all the, I think I will never have all the answers, but, yeah..

Anne: I think you're very right to be skeptical about serverless running on prem rather than in the cloud. Cause it does feel to me like, I touched a little bit earlier about multi tenancy. When you're not doing something, what's somebody else doing on the same machine? A lot of these tools like serverless work really well because you're in a multi tenant environment.

So the classic example is with, if you're in the cloud and you're, you might be sharing physical resources with a company that has very different demand profile to you. So if all of your, if you are, say a retailer, I used to be head of IT for an online retailer, then all of your resources are assigned to your demand.

So if there's a peak, then you have provision for your peak. And that means that a lot of the time, so say Christmas might be your peak. You have a provision for Christmas. And then most of the year, the machines are underutilized because you had to provision for Christmas. If you move into a multi tenant environment, one of the root things that the hyperscalers attempt to do is to pair up, or not just pair up, but group users on machines in such a way that they have different demand profiles. So everybody has a correlated demand at Christmas. Maybe if you're a retailer, you might be sharing a machine with a training company. And the retailer is very busy at Christmas and the training company is very quiet at Christmas.

So you kind of, you, rather than needing to provision for the peak, you are getting better utilization of those machines over time. And serverless is a little bit of an example of that. It's, the win is with multi tenancy and multi tenancy is easier in the cloud than it is on prem.

Now, having said that there are some multinationals that are so big and have such complicated systems internally that they are effectively their own multi tenant. I mean, Google is a hyperscaler, but forgetting GCP for the moment and looking at Google's internal tools and applications, they are their own multi tenants, they have enough variability in what individual tools are doing that they can act to keep their machines fully utilized all the time. They're kind of designed for that, but most companies are not at that level and quite haven't quite designed for that yet. But I agree serverless internally on prem for a small enterprise probably doesn't buy you that much.

Wilco: And there's also an other perspective on this. And I heard this in a conversation with a bank and that was also very inspiring for me. So. The fun thing is everything we just talked about can also work in harmony. So if you always have a baseline utilization, you could do that on prem where you know, "okay, with these applications, we always have activity.

So we have a certain utilization." And what they say is what then a big, they have done in control. They know exactly the energy usage and because of adventure, the energy uses also the negative impact and emissions and avoid carbon. But when they have peaks, they overload to the cloud. So they're like, "okay, so if we have, then we go there." And you could kind of also do sometimes maybe do the same if you want to save costs to maybe the devices to some degree that using your application.

So there are ways, and this is complex, but I think that is a way. And also there is a topic that we had in the past and nobody's really talking about anymore because it's complex. It's distributed computing, of course. And that's also another pattern that could, have a play in this. So, yeah..

Anne: Yeah. I mean, distributed computing is very potentially well aligned with this whole thing of demand shifting and shaping and saying, you know, "actually I've got a, thing here. We need to treat it as essentially asynchronous." And of course, asynchronousness is, or asynchronicity, is a really key part of distributed systems, designing distributed systems well.

Completely synchronous distributed systems are often a little bit of a distributed monolith. You don't, you often don't get the same breadth view of

Wilco: a funny thing. I heard, yeah, if you're YouTube, you can indeed see, like serverless monolith, microservice monolith, there are the.. So sometimes you think, "okay, this is a good start, if we design it that way," but if you create so much dependencies and then you're still, on an abstract level, creating a monolith.

Yeah..

Anne: Yeah, except a more difficult one. Because a monolith, the value of monolith is, quite simple and, well, it's, yeah.. Monoliths and microservices and distributor systems all have their place. It's always a matter of choose the right tool for the job. That is efficiency 101 is choose the right tool for the job.

Wilco: Oh yeah. A hundred percent. Yeah.

Anne: So we have chatted so long, we've got hardly any time to actually go over your articles. So do you? Let's see, we were going to talk about three articles. One was about your using the, SCI and the impact framework.

Wilco: The thing that I would like to say on that one is that sometimes maybe people think that I'm only interested in energy and that is absolutely not the case. The only thing that I came to the conclusion, not that you have to do it that way, but to some degree, because we're still also figuring out how to best measure emissions based on the grid.

And the other thing is embodied carbon. So we can do a lot of stuff with lifecycle access data of assessments and hopefully also the correct information we get from our scope three suppliers, but to some degree, if you know, and this is what I always keep saying, and maybe it's, if you know, at what moment, at what location your software was running on what kind of resource type on what kind of hardware, and you log that down, the most important thing is energy you cannot historically get back. It's you compute it and it's gone. So, and we have still with hardware disproportionality, but we have still things to figure out, okay, how do we really measure it? And I really strongly believe if I can help to get it down somehow, we to some degree can historically get back. Okay, we, knew that in that moment in time, even with electricity maps, with historical data, we can get the emissions from that moment, that location.

And also with the information we gathered with past procurement information or supplier information, we can get the embodied carbon right and that kind of stuff. So that is the main reason why I really focus on that E of energy and the SCI, the Software Carbon Intensity, you know.

Anne: The thing I like about the E is it's something that software developers can have effect, you know, it's not like we're, we don't care about the other things. We're just trying to focus on where we can have, make a material impact. And there is somewhere where we can really make an enormous impact. Our ability to change things in other areas is more limited and that's why we started the Green Software Foundation, was to find ways that software engineers can improve things and people related to software development.

So product managers, testers, that kind of thing. It doesn't mean that we don't care. It means that we're looking at where we can have the an impact.

Wilco: Oh and the quantum entanglement. That came from the idea, I'm a big nerd and I, like to know things, a lot of things, and it's also counts for physics. And I'm watching a lot of YouTube videos, also science videos, and a lot of videos every time came back on the concept of, okay, we have Einstein's, real, real,

Anne: Relativity. Yeah. Yeah.

Wilco: Relatable.

I'm still going wrong here, like, like about the, eventually can calculate things going on in a black hole. And we have quantum mechanics with quarks and all cool stuff. And when they try to bring this theories together, they have all these kinds of issues that doesn't really match up. And that's what I've found really interesting because I started before my journey, I had some information from the holistic view perspective.

So from boardroom perspective, from more top, if you think from top to bottom. And they are more interested in compliant data. So they, their data that they're using has to be valid, has to be compliant and also streamlined. So eventually they can report and also use it to create internally policies on it and that kind of stuff.

But if you talk like the software engineers, we're really like, "yeah, we're missing data on calculating the correct energy consumption. If we run software on this CPU or this GPU and we're still figuring things out," but very on the lower level, but I think on a boardroom level, they are probably not that interested in how correct the CPU or GPU was, as long as it is correct enough to make it right for the reporting.

So, and then the other thing where it really aligns to is at some point I came to the conclusion, like, okay, we can measure a lot of things. We can have all these metrics. But at some point the numbers are not going to change outcomes, decisions do. That was kind of a big moment for me. So there has to be at some point the sustainable decision making process going on from bottom to top or the other way around where those worlds connect to each other.

And that's what I really, with this article for the first time, it was also a few months back. That I was thinking, "okay, how can I connect this worlds together?" So what would the steps be? And the main, I think outcome for me from this whole thought experiment was, is that based on certain levels in the organization, you have different requirements for the data you want, different tools you probably want to use, and also different reasons you want to use this. And yeah, that is still a thing that I'm working on. One thing where I think that everything in comes together is a simple concept that if you set up an organization, a big IT resource list, or even resource less in general. And you have this resource list where you have like your mobile phones, your laptops, but also your cloud resource.

We have the infrastructure as code. We could generate those resources. And you have all these resources in this list. And if it's a multiple of the same resources, you just do a count after it. And then you would say to each resource, you would say you're using so much kilowatt hour a year. This is the footprint estimated benchmark or real time.

But also this is the security need because the security people are in the same thing as us. They really want more information. And that would be a great starting point because from that perspective, you can eventually bring that to information that is more on the higher level of the organization and you also can connect it to the really nitty gritty things on the bottom.

But yeah.

Anne: I was very interested in your, quantum versus relativity. I also, I have, my degrees in theoretical physics, but I will say, it kind of, when I read it, I thought, ah, now this, I find this quite an interesting analogy because I was, there's two. I was the fan, I was a particularly fan girl of neither.

I was always a fan girl of classical physics. I like you remember that enterprises operate at the level of classical physics. I think it's actually a really good analogy that enterprises are thinking about, "is my data compliance, you know, are my bills ridiculous, are my system staying up?"

And I think that actually, and only once that's, you know, in the Maslow's hierarchy of needs or the Maturity Matrix that the green software maturity matrix from the Green Software Foundation that Pini and I run together, the first thing you need to do is get your operations right. And that is classical physics.

You know, it's kind of like, are you paying too much? Are you over provisioned? Do you have a whole load of machines that you're not properly monitoring and using anymore? Maybe they are security holes. So you've got security problems there, you've got financial problems there, and you've got waste there as well.

You've got loads of carbon as a result of,

Wilco: yeah. That's the one I forgot on my list. Waste that also needs to be Yeah on there. But yeah..

Anne: I love relativity, I love quantum physics. But for most enterprises, I would say start with classical physics, really just focus on getting your basic ops good, do those thrift a thons. Your last article, which you're not going to have time to go over, but I would strongly recommend people read, is about the alignment of FinOps and green software.

And I would say that the alignment of FinOps and green software is your classical physics. Oh, that's awesome. It works for every single enterprise. Nothing fancy is required. It's totally aligned with the business, with the desires and the goals of your business. No one is going to complain that you saved the money.

Wilco: But I think that is,

Anne: once.

Wilco: yeah, I think that's very important because sometimes, I mean, a lot of people, they say, "why are our companies not starting?" Or "I have this idea, why is no one picking it up?" And I think we also have to be honest to ourself. If you invest money in something, you want it to be, have a certain maturity level and also, especially if we buy something, we want to know that it's going to work and that's, yeah, it's going to have the value that you expect it to be.

And so together, I really believe together that we can, and the whole, or the whole line of field can bring it to a certain maturity level. So with the maturity matrix, that's a different story. Yeah. But because I think that a lot of people always focusing on, "yeah, but if you do this, you have lower of cost," I think the most important thing for a company to be more interested in how can we solve the things we need to do for sustainability

is to make it more frictionless implementation and have it less risk, because I think that if you can do it in a way and it doesn't have to be perfect, but that it's easy to implement and to use, I think companies will start doing more.

I strongly believe in that.

Anne: I do. I think there's a huge, people really care about being green. People do really care about the environment. They don't know that there are changes they can make through their work as software engineers, that will make a huge difference. And if they do think that there are changes, they have a tendency to think those changes are misaligned with their company goals, which quite often they are because people think, "Oh, I'll just rewrite everything in Rust."

And that would generally be misaligned with the company's goals. But going through and making sure that you're not over-provisioned in your data centers and you turn off stuff that's not in use, that you're being cost-minded, that's totally aligned with your business goals. And it's also aligned with being green.

On that, I think we really need to raise awareness of that.

Wilco: Yeah. Yeah. And that's what I really can bring back to the fact, like if you.. And like, Hey, the mention of time efficiency, something has to be fast, but you want to do it in the most low energy consumption, but then the most important thing to, so what is the value, what you're doing is going to bring. And it's something that we struggling in for a while because we really tried with agile, define like the customer value, business value.

But I think those three together, if we figure that one out in sustainability in a better way, we really can make some jumps.

Anne: And with that, I think we'll need to end because we have been talking for ages and it's been absolutely fascinating and I've really enjoyed it. So thank you very much indeed, Wilco, for this. So where can people find you and get involved in your meta communities?

Wilco: Well, mainly I'm active on LinkedIn on my name, Wilco Burggraaf. I try to post every two days, a new content. And yeah, you can also, if you look up meetup.com, you can find the meetup group under the green software meetup Netherlands. And another thing is of course, from January, I'm starting with my workshops.

So if you're located in the Netherlands and you're interested in that, yeah, reach out to me.

Anne: Excellent. Thank you very much for coming on this episode. It's been really fun. And if anybody wants to contact me or chat to me about my, I also do workshops, which are not the same as Wilco's. So we are, I would say we run a complimentary workshops, then you can also take me through LinkedIn. And this is a final reminder that all the, well, we didn't really talk through the resources for this episode, but they are good background resources for our discussion.

They're quite easy and pleasant to read. So have a look at the links to Wilco's posts, follow Wilco on LinkedIn and read his articles. They're very, very good. So thank you very much. And I will see you all in the next episode. Goodbye for now.

Wilco: Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: Powering AI on Atomic Energy

28 november 2024 | 55 min

TWiGS host Anne Currie is joined by Stefana Sopco, Marketing Manager at PortXchange and a passionate climate activist, for an insightful discussion on the intersection of AI, sustainability, and the maritime industry's decarbonization journey. Stefana shares how PortXchange leverages green technology to help ports achieve net-zero emissions through innovations like just-in-time arrival and emissions tracking. They also dive into the challenges posed by AI's growing energy demand and explore nuclear power as a potential ally in the climate fight. Throughout the conversation, Stefana emphasizes the importance of mindful technology use and the urgency of adopting sustainable solutions.

Learn more about our people:

Anne Currie: LinkedIn | Website
Stefana Sopco: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

PortXchange

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIOT BELOW:

Stefana: Is ChatGPT going to decarbonize the maritime sector? No freaking way!

I haven't heard of anyone saying how they plan to use ChatGPT to decarbonize or to help ports account for their emissions.

I'm your host, Chris Adams.

Anne: Hello and welcome to The Week in Green Software, where we bring you the latest news and updates from the world of sustAInable software development. I'm your host today, Anne Currie. And today we're going to be joined by Stefana Sopco, Marketing Manager at PortXchange, a leader in driving ports towards a net zero future.

Stefana's work centers on using digital solutions to reduce emissions in the shipping industry. So a bit, a little bit different from our normal guests who are mostly directly programmers or very programmer related. So Stefana's vision extends beyond ports to broader decarbonization goals across the tech industry and the maritime industry.

She's also passionate about D&I, diversity and inclusion. And hopefully we'll talk a little bit about that today as well. So in this episode, we're going to talk about three articles that are all about AI, and the effects AI and the demand for AI is having on grids, particularly in Europe. We'll be talking a little bit about the power requirements of AI and the, grid capacity requirements.

And we'll be talking a little bit about whether or not it's always going to be a bad thing, or will actually could potentially be a very good thing for grids. And I'm quite a positive person. I think that there are massive advantages that come from the extension of the grid to support AI. So that'll be an interesting thing to talk about.

I'm also really interested in talking to Stefana about this because a lot of the issues around AI are effectively logistical issues. And logistics is something that we don't think enough about in the tech industry. But in the maritime industry, they think about it a very great deal, so I'll be very keen to hear Stefana's thoughts.

So, on that note, welcome Stefana. Can you start by telling us a little bit about yourself?

Stefana: Thank you so much, Anne, and thank you for the invitation. I'm really honored to be here and super nervous, to be honest, because it is a heavy subject, and I really hope I will do it honours. My name is Stefana Sopco. I'm the marketing manager at PortXchange. I've been working in the energy and maritime industry now for five years.

And for the past two years, I've also been a core member of Women in Tech, Netherlands, an NGO striving for diversity and inclusion, as you beautifully mentioned. Here I act as the marketing manager and spokesperson as well for our Dutch chapter. I'm an active member of WISTA NL and for those who don't know, WISTA stands for Women in Shipping and Trading Association.

And like I like to say about myself, I declare myself a climate activist next to my proud feminist tag, which I've learnt to embrace completely, especially in the past couple of years, and one fun thing about me, you will always find me mitigating for a cause. Either diversity in the industries that I work with, or animal rights, animals are my life. And of course, climate change. And next to my professional activity, I try my best to act as a role model for the younger generation.

I mentor and guide young professionals who either want to start their career in marketing and come into these industries that are very intimidating for some, or might want to switch careers. And I also dedicate a big bunch of my time to guiding expats, me being an expat as well in the Netherlands, and I try my best to help them navigate their journey here the way I wished someone would have guided me back when I moved here six years ago.

Anne: That's great. That all sounds very good. Interestingly, this isn't even our first all female Environment Variables podcast, but normally that's because I have one of my two co-authors from Building Green Software, the new O'Reilly book, Sara, or Sarah on with me. So I think this is the first one. I'm going to guess this might be the first one,

all women, that doesn't involve Sara or Sarah, so it'll be an interesting change. So say a little bit about me, because I'm not the normal host. I'm quite a common host, but I'm not the usual host of Environment Variables. My name is Anne Currie. I am, as I've said, one of the co-authors of the new O'Reilly book, Building Green Software.

And if anybody who's listening to this podcast, you really should be reading that book. It's suitable for everyone. You don't have to be a techie, you'll enjoy it. It'll be useful. I'm also the CEO of a learning and development company, Strategically Green. So my focus there, our focus is trying to get companies to build up enthusiasm and internal expertise on how to make these changes happen.

So we do workshops and training, things like that. So if you want to talk to me about any of that, then you can find me on LinkedIn. So what we're gonna be doing today is we're gonna be deep diving into three really interesting articles about AI and their impact on the grid. But before we do that, Stefana, do you want to talk to us a little bit about your work decarbonizing the shipping industry through Green Tech?

'Cause that's something that we have never heard about before, so I think it would be very interesting.

Stefana: Now, I feel even more stressed that I have to do the honors about talking about green tech in shipping. But before we dive into the goodies, I really want to take the chance to thank three of my colleagues, Eugene, Abhishek, and Sjoerd, because they helped me prepare for this session. I was a little bit nervous. I really wanted to put our technology in really a beautiful light and also talk mindfully about technology. And they've been extraordinary. They've been very, very supportive and I wanted to thank them. So maybe I should start by mentioning that PortXchange is a B Corp organization.

For those who don't know, B Corp by definition means impact before profit. And you see now a trend of more and more companies being B Corp certified, which is in direct relation with the Green Software Foundation's principles in a way or another when I looked into it. And in addition to striving towards sustAInability within the maritime sector, we also consider socially responsible and ethical ways for

everything throughout our business, so our entire value chAIn. Our mAIn mission at PortXchange, like you said, is to help port authorities, predominantly, worldwide to achieve net zero emissions. We help them by measuring their emissions throughout the entire port, and not only track them, but also analyze them and report them and take proactive steps to reduce them because that's the ultimate thing that we want in the end.

But I'm really proud to say that our green tech ecosystem is larger than that. We have developed many other products within the shipping ecosystem. So we address organisations beyond port authorities. We have tackled just-in-time arrival, I'm not sure if you are familiar with that, it's a solution that is now quite trendy, but a little bit more difficult to implement because of data sharing fears and thoughts from these organizations. But basically, just-in-time arrival helps limiting the fuel consumption of the vessels, because ships tend to rush through the sea to arrive to the port, like there's no tomorrow, but then, they arrive too early, and what does that mean?

It means that they will stay there, stranded, and the emissions that will go around the proximity of the port population will be higher because they are just sitting there wasting time and emitting more emissions. So our organization acts as a partner in the decarbonization journey of the maritime industry, players like port authorities and shipping companies.

But I really wanted to make a point here is that, as a technology provider ourselves, the computing power that we need also has a carbon footprint of its own. So what do we do to make sure that we are mindful and we remAIn mindful of that? Well, for example, we are extremely careful of using only energy providers that provide the majority of the power sourced through green channels.

We are very transparent with our emissions. We use all kinds of solutions like Goodwings, for example, to make sure that all our traveling is as green as possible. We account for our scope 3 emissions as well. And there's a saying that I really like to attach to PortXchange. How I envision it is business as a force for good.

Anne: That sounds fantastic. So there's loads of things that you've mentioned there, which we could talk about. The changes that you're making within the industry to make shipping more efficient, that's very analogous to something we often talk about in tech around operational efficiency, kind of not using, not being wasteful, not using more resources than you need to, in order to achieve the same goal.

So, and while ships are powered by fossil fuels, then everything that you are wasting is, it results in carbon emissions. So that's really good. But you're also balancing that with thinking about how your tech itself is going to be greener. So yeah, so very interesting things there, which I'm sure we'll focus in a little bit, a bit more on as we go along. But let me zoom in.

I'll get started now with talking about some of the articles that we're going to be talking about So we're going talk about three articles. And to kick us off, the first one is a McKinsey article entitled, The Role of Power in Unlocking the European AI Revolution. So it's an interesting paper and it's all about the fact that at the moment, and it's, there are some very interesting numbers in it.

They are estimated numbers, but they're useful and interesting in and of themselves. So McKinsey estimates that as a result of AI and the desire for AI, data centers in Europe are going to expand from using about 2 percent of Europe's total power consumption to around 5 percent by the end of the decade, which seems quite plausible to me.

And they point out that there are several challenges associated with that for the European power infrastructure. And the article is entirely about Europe, but I think that the issues also apply elsewhere. The first is that we're going to need a lot more green power. I mean, the good thing, the thing that it does say in the article is that the hyperscalers are saying that the power that they're going to want to use to power this new AI revolution will be green.

So that's good. And that was not the case 10 years ago. So I think it's really good that they're now all singing from our hymn sheets on that, they want the power to be green, but they need more green power. They need the ability to get it from where it's generated to the data centers. So they need better infrastructure, better grid infrastructure.

And we need ways to handle the fact that renewable power is not always guaranteed to be avAIlable in the same way that fossil fuels are. So they're gonna need to make changes to, well, we're all gonna need to make changes to handle that. So generally, I thought it was a pretty interesting article.

What was, what were your thoughts?

Stefana: Well, first of all, with any change, we will inevitably feel uncomfortable. But in order to achieve the great outcome that we are all hoping for, we must collectively feel uncomfortable. It should even hurt. And it's okay to hurt because we have to think beyond that. So how does that apply to my world, to the world of maritime and ports?

And do I believe that decarbonization within the port ecosystem is more or less following the same path? Definitely different worlds. Completely different worlds. We are here, within the port world, we are looking at a traditional sector that has been known and popular for having ways of following from back in time.

But we start our presentations at PortXchange whenever we are invited to an event with "ports as battlegrounds for decarbonization." Because if you look in a more simplistic way, ports are the puppeteers, but let's not put a negative connotation there. Let's put a good connotation, as in, it's where the good fight can start for real.

But, as in any other sector, you need proven and robust sustAInability practices. Everything should be backed by verifiable data. I think it's very important here, not verified data, verifiable data. And one thing that is very important and maybe one of the biggest differences between other sectors or other industries is that ports benefit directly from community relations, from public perception, which inevitably leads to public and private partnerships and policy support.

For example, if you look at Norwegian ports, which have declared themselves as the employee of the people. Their ports have managed to not only super digitalize their operations, but they are almost net zero. And it's all purely because of the change in mindset. And I actually talked about this earlier today with someone from Climate Pledge, how your mission and your intrinsically motives can make the difference in this fight for decarbonization. And one good thing to mention is that earlier this month,

European Seaports Association launched their environmental report. In this report, it has been made a point that climate change is our top one priority for ports. So what does that mean? That means that we should also expect full transparency and full accountability because being at the forefront means you have probably the biggest impact in terms of carbon emissions, et cetera, et cetera.

So ports have the authority in my world. Ports have the authority and influence to move the need. And what I would like to mention is that in my world, I keep saying in my world, it's almost funny because while I'm saying it, I'm envisioning how we're moving from a world to another. We are quite dependent on regulations here, and we have plenty of them.

We have the IMO, I'm not sure if you know about IMO, that AIms to achieve net zero in emissions from international shipping by 2050. And, while we are doing that, we also have checkpoints, 2030 and 2040. Do I think that we will start seeing real progress by 2030? No, I don't think that, unfortunately. But then we also have, since 2023, other mandatory regulations, like in energy efficiency existing ship index.

And then we have the Carbon Intensity Indicator. We have the Monitoring, Reporting and Verification from EUMRV. We have the Emissions Trading System. And more recently, we have the CSRD, which is more predominant for Europe. And CSRD is actually one of our allies in green tech for green tech providers, because it somehow pushes these organizations to report, especially their emissions and their impact.

But now there is a question here that remAIns constantly. We have all of these things and the industry is being pushed from the left, from the right, et cetera. But we ask ourselves, how do we use technology to help us be compliant with all of this? And meet the requirements, of course. And we ask, is AI the answer?

Is, I don't know, well, large language models the answer? Is blockchAIn the answer? What is the answer? We don't know yet.

Anne: So your potential, I mean, obviously when you talk about increasing the efficiency of journeys, making sure people get to just-in-time, making sure that people aren't hanging around getting places early, getting places late, going too fast, optimally using their fuel and being as minimally wasteful as possible whilst delivering the service. So operational efficiency. The obvious thing that springs to mind is can AI help with that? Do you think it can, or do you?

Stefana: Yeah, absolutely. So, AI is not the enemy. Like, It can be. It can be the enemy, but not when used for this purpose. We use AI in our technology as well. We have a couple of case studies about that on our website, but we don't need AI for everything, Anne. That's another thing, because sometimes I ask myself, "do I even need AI in this situation?"

Last year I was nominated for an award by Women in AI, Netherlands. It was Responsible AI Leaders. And the nomination really made me think, what if instead of chasing unicorns, we grab the low hanging fruit and start implementing and bringing into operations the technology that is already avAIlable and viable.

Because we have so much, we have great technology already and we have the skills for it. So I would like to see, because I'm working in an industry that moves slower by default, because there are many stakeholders and many people and many other third parties involved, I would like to see what we have finally being implemented.

And meanwhile, we train and we upscale. But focus should be on scaling what's already there, instead of maybe continuously developing new tech that needs to go through the adoption curve anyway. One of our core specializations, sorry English is not my first language, like I mentioned is to provide optimization models.

And we also provide estimates based on data. Our technology might not be the fancy, super large language models, which everybody seems to be either an expert or an advocate or something nowadays, but statistical algorithms built for the goal that we want to help ports decarbonize by emissions monitoring or by just-in-time or more sAIling, et cetera.

It might not sound fancy. A couple of years ago, it sounded super fancy before GenAI popped, but it's doing the job and it's doing the job very well.

So sometimes I'm in doubt. Are we really that stubborn to think that we need AI for everything? Especially in sectors like the maritime sector, where there's still so much legacy that needs to be erased and so many things that need to be relearned,

forgotten and then learned. We are, the industry is still working with traditional professionals that are very good at their job, but in order for them to embrace AI, they need to embrace technology.

Anne: Yeah. Yeah. And in fact, we find the same in just pure tech as well, outside of the maritime industry. And that's. The lowest hanging fruit is usually very simple. It's turning things off that are no longer in use. It's right sizing. It's just minimizing waste in the same way that you're looking at minimizing basic waste.

The basic waste minimization is where you usually get your biggest reduction in carbon emissions. There's a French philosopher once said that, the perfect is the enemy of the good, that the pursuit of a perfect solution makes you step over a very good solution that you already have in your hand.

And that's true of, it sounds like it's true of your industry. It's certain true of our industry. Manual thriftathons, turning things off, right sizing, you don't have to do anything clever to really start cutting your carbon emissions quite significantly. Yeah, no, we're all the same, really.

Stefana: Yeah. And we also have to understand I had quite an aha moment last year. I went back to my home country, Romania, and was right after I was nominated for this award. And my mom is a teacher at a school in a small village. And people came to me, they asked me, okay, what's up? How's your life? All that. And it was a really big boom with AI and all that, and they asked me, so what is your opinion on AI?

What is AI? And I realized that the majority of the population needs a more simplistic understanding of artificial intelligence. It's not the fancy super professional profile that everyone is putting on LinkedIn or on their blogs and all that. And in order to see adoption, day to day adoption in our lives and in our employees' work lives, we need to make it simpler.

We need to... again, do we really need AI for everything? I keep getting, I keep coming back to this.

Anne: It's interesting you say that. So I read it, I think it was an article in The Economist, last week. It was talking about how AI, the companies are really struggling to work out what they're going to do with AI. Because interestingly, the people who are finding uses for AI are more likely to be individuals because making ChatGPT a live service that anybody could access for free really did kind of open it up to people could just go on, play with it, talk to it, use it to rewrite their essays at school, all the things, good or bad, that you can do with ChatGPT. There it's, interestingly, it's an interestingly accessible technology, isn't it? But businesses are really struggling to work out what they're going to do with it.

Stefana: Yeah, and is Chad GPT going to decarbonize the maritime sector? No freaking way. I haven't heard of anyone saying how they plan to use ChatGPT to decarbonize or to help ports account for their emissions. And another thing that I experienced actually in my close circle is that when they are, when ChatGPT is being taken from them, like for example, their companies blocked the ChatGPT server, et cetera.

They don't know how to do their job anymore, or they are panicked that they will not be as efficient as before. And it's just, for me, it's quite concerning because as a responsible AI advocate, it's scary to see that. Last year or two years ago, nobody have ever thought "I cannot do my job if I don't have this little chat to support me."

But now we see more and more. In marketing, I've been quite active in testing all kinds of AI solutions up to the point that I hit burnout because I was so afrAId that I will be left behind and I will not be employable anymore and companies will not see me as the next gen thing anymore, that I worked myself to burnout and I don't want to see that massively in all industries because there's, there are still many good things that we can do before we rush into something that might be a little bit over the top.

Anne: Yeah, it is quite sophisticated. It's still very early days. I'm very ambivalent about AI, but I think there are some amazing things about it. There's some bad things about it, but there are some amazing things about and I really want to see it. So what I'm going to do is talk about the next, because we're chatting a lot, we're running behind. So let's talk about the next thing, because I want to talk about the final one as well.

And the next one, I will put in some of my thoughts in there. So the next article was AI power demand might actually turn out to be good for climate. And it is an article in Energy Connects. And it's talking about what the last article talks about, which is the massively increasing electricity consumption driven by data centers, particularly those supporting AI, but it's, it, has a more positive take on it, which is that data centers are actually a pretty good potential consumer of renewables. We have a lot more ability to shift and shape our work. We're a lot more aware. There's a lot more demand for green, particularly from the hyperscalers, which, as I said before, is great news.

And we want to see it happen more to make sure that it's not just them saying it, but them actually doing it. And one of the things that's coming out of, what's happening at the moment is that AI is quite early in a, massive boost in electricity demand, which we know is coming because the, energy transition is going to drive a massive increase in electricity demand, so we are going to have to get grids ready for that.

And AI is, to a certAIn extent, giving us a little early exposure to that. There's a really interesting stat in the article about

likening the demands to the biggest demand surge in electricity since World War II, which made me think, oh, I didn't realize there had been a massive electricity demand surge in World War II, but there was right at the beginning of World War II, there was a 60 percent increase in demand for, I went and looked this up on Wikipedia after reading this quote in there, 60 percent increase in electricity demand that came from the fact that actually there were a lot of new electrified factories that had to be built, but also an awful lot of Aluminium smelting. They needed a lot more aluminium to build ships, as you know yourself, and planes and aluminium smelting is very energy intensive, very electricity intensive.

So I thought this was, it was quite, it was a really quite interesting article that I would strongly recommend people have a read through. And it did make me think there were a few things in here. Yes, we do need an awful lot of additional, not just the ability to generate power, but we really need to upgrade grids all over Europe to get that power to where it's going to be used.

We are all going to need to learn to love the pylon is what I'm thinking. What were your thoughts on the article?

Stefana: So I think I have a little bit of understanding of where we are heading or where we should be heading but I'm definitely not an expert on this so more from the sideline. But also as a consumer I want to be very mindful. And PortXchange as well as a green tech provider needs to be mindful of that because we have to live by what we preach.

So from what we know right now, nuclear power does not have a large profile in emissions. So that's green, that's a green check.

Anne: Yeah, indeed.

Stefana: What we also know is that nuclear waste can be catastrophic. And it needs to be handled very carefully and also they are expensive as hell. Solar requires a lot of space.

Windmills produce lots of noise pollution. Also super expensive. And you cannot live nearby, so any communities or There's so many considerates there. Hydro requires a lot of space, and it's super expensive, dangerous as well if it breaks, and it has a direct impact on local climate and ecosystem.

And the one that should not be named, or the one that can't be named, it's fossil fuels.

The nastiest in terms of emissions profile, it's what we all want to leave behind. But unfortunately, as you already know, Anne, and all these articles mention it, with so much demand in AI, the question is, do we have enough time to test all these alternatives? Do we have enough bandwidth to keep looking into solutions and all that?

Or the demand is so high that we go drill, baby, drill, codes with codes, or even worse, we go back to coal, which is one of my recurrent nightmares, to be honest. And that's why I keep getting back to the question, do we really need AI for everything? Do you really need to have ChatGPT giving you that image with cats and dogs and whatever?

Because I don't know if you know, but the carbon footprint of Dall-E 2, similar with ChatGPTs, is at 2.2g CO2 emissions per query. Which is, sounds little,

but if you start doing the math, Jesus Christ... And then the CEO of MidJourney, David Holtz, said a while back that image generation requires about 1,000 trillion operations, which suggests a carbon footprint of about 1.9g CO2 emissions using NVIDIA, A100 GPUs, which consumes, only for that, right, for one image, 400 watts of power. For one image. And I, you probably did this, Anne, you went on ChatGPT, you said "create an image," blah blah, and then it gives you an image, and it sucks, it's horrible, it's... "Jesus Christ, what is this?"" And then you say "no, another," blah blah blah blah, and it gives you, and it sucks even more, because now it's something completely different, and you are stuck in that loophole for 15 minutes, and nine images later, you finally have something, but it's still not good enough, so you go on Google and you find an image that, but you just wasted God knows how much emissions, correct?

Anne: Well, yes. But we were at the beginning on AI. I mean, I, they're good and bad and sometimes, and, there's an awful lot of waste associated with AI, but I think we have to kind of say, "look, it's coming and we need to make the best of it." It's, people want it, it's useful. It's really at an early stage.

We will definitely get better at it, and we will get more efficient at it.

Stefana: What we need to get is a little bit more environmentally conscious. So I want it because it's shiny and cute and it helps me. But what do I do to compensate for the carbon footprint? Educate myself first of all, right? And then compensate with that.

Anne: So there was something in the article that wasn't what they meant it for me to take away from the article, but I took away article, from this article, which is that when they talked about the massive increase in electricity demand in World War II, and then I looked on Wikipedia and a lot of it was from aluminium smelting.

So aluminium smelting, really, we still do tons of that because we all want aluminum for everything. It's an incredibly useful metal part of our day to day lives. Nowadays, we don't do, we don't tend to do aluminum smelting off a national grid anymore, because it's a very expensive way of doing it. A lot of aluminum smelting, take it for Europe, takes place in Greenland, because Greenland has 24/7 carbon-free electricity from hydro, from water running off glaciers. It's another source of energy that we don't often think about, but should think about more these days. Water running, melt water running off glaciers is a potentially a really good source of cheap carbon-free electricity.

And so a lot of aluminum smelting moved to Greenland where it's done in a very low carbon way compared to doing it on a grid. And I talk about this with AI, I've talked about this with AI many times before in that, there's a lesson to be learned there. Sometimes you need to take the very electricity-heavy load to where there's a load of low carbon power to do it.

And yeah, I'd love to see. We've had this conversation on Environment Variables loads. I'd love some, a lot more AI going on in Greenland.

Stefana: 100 percent sure that with this increase in demand, especially, we are putting more pressure on the ones that move the needle and can really do a change. And we also need to keep in mind that when we push for more green alternatives, there will be more pressure to invest in that. And with the conscious use of AI,

the same as there are activists that, fight against Shell or fight against the shipping industry because that happens as well. There are also now climate change activists that militate for green or conscious use of data, of technology, of AI, so on and so forth. So every trend creates different ecosystems and different communities. And communities and individuals have the power to push for more things like you mentioned.

The moment we become conscious of how we use them and the moment we become conscious of how they impact the global matters of things and not only us per se.

Anne: Yeah. You're quite right that we have more power than we realize. If we actually shout for things, if we shout for green AI, if we demand that hyperscalers are running off green power, if we demand that the growth, that things are running on smaller models, running on local devices, rather than over egging it and not thinking about these things.

Think green AI all the time. And so these articles have been quite useful because they suggest that the hyperscalers are pushing for green AI. And they're only doing that as well. They're doing that partly because they're not idiots. It's

Stefana: Exactly. That's what I wanted to say.

Anne: That's, we know in the future that, renewable power is where all the growth is.

If you look on our, I always recommend people have a look on our world in data, excellent source of all graphs on anything you might want to have a graph on, but they have some excellent graphs on, global solar power generation, global wind generation, but mostly global solar power generation. It's going up exponentially.

And you kind of think, well, that is the future. Fossil fuels are not going up exponentially. Solar power is going up exponentially, so we need to be ready to use it. And then the hyperscalers know that, but at the same time, they feel the pressure from customers constantly saying, what is your green story?

Stefana: Absolutely.

I put my hopes into that.

Anne: Yeah, absolutely.

Because,

need like you,

Stefana: and there was that ad right at the beginning of generative AI trend, where a construction company said, "Hey, ChatGPT, come and finish this building." It was a brilliant ad. Now, if you think a little bit at the environmental situation that is happening in Europe, and not only in Europe, but also the geopolitical situation that is happening, and one of our closest and one of the saddest examples that I will give right now is what happened in Spain,

Anne: Oh, yeah.

Stefana: And the floods that they've experienced.

And one thing that you will ask now is, how is AI fixing that? How is AI helping those people right now get their lives back together? Is it helping or is it contributing to their catastrophe? And that's a question that I ask myself constantly when I think of mindfulness in using technology and AI and so on and so forth.

If AI can make the world a better place. God, please do it. God, please save us. AI, artificial intelligence, robots, please save us from ourselves. But until they have that power, if they will have it, you never know, we need to consider our Shorter impact on that. The little step.

Anne: Well, actually, I think that is a, that's a really good question to be asking ourselves, it's like AI is just code. It's just technology. It's just another aspect of technology. But for all technology, what are we doing? Are we making the situation worse or are we making the situation better? If we're running systems that are over provisioned and we're just, we're being wasteful, we are making the situation worse.

We're not making situation better and, even if us, the tool that we're developing will help build it, absolutely fantastic. But then go back and make sure that it's not being wasteful. There's really no downside to going back later and just cutting waste, just trying to improve.

That doesn't mean we have to stop using technology. It doesn't mean we have to stop using AI. Use it in a mindful way. All good. In fact, we shouldn't stop using it. All these tools are fantastic and will be required for us almost certain to solve these problems. Quite possibly, AI will help with these problems in the future.

Predicting when storms are going to happen, predicting what the result of that is going to be, telling people where to go and what to do and how to survive it. Actually, AI can really potentially help with that, quite a lot of it, but yeah, but we just need to be mindful, as you say, mindful.

Stefana: And we need to be in balance. And like you said, AI has tremendous capabilities to improve operations too. We saw that AI has finally managed to understand how whales are speaking and what they are saying. We use AI in our methodology to track emissions and to identify the hotspots at ports. But remember what I said at the beginning of our conversation, business for good use.

Use it to make the world a better place, as cliche as it sounds, and if your organization desperately wants to use AI, work with technology specialists, but also with sustainability and environmental specialists in understanding the impact and the long term strategy that you want to follow in order to see feasible results from that.

And I think we are in a position right now where we need to be accountable for our footprint, for our carbon footprint. So whatever we do, track, monitor, analyze, keep an eye on it, because otherwise we're walking blindly without the data that is out there for us to see. Either it's AI, either it's electric cars, either it's whatever, it doesn't matter.

Anything can be used for a good cause.

Anne: That is an excellent point. And I realized we have, we've chatted way too long and we've completely used up our time to talk about the final article, which actually I thought was a bit of a, it was a bit of a weak article. So I'm not too worried. It was a very good article, but it was a, it was about Three Mile Island by somebody who had, it was written in their PhD about three mile island, which, and, it's about anti nuclear and the dangers of nuclear. And it felt a little bit like, well, because something bad happened in the seventies, doesn't necessarily mean... I would rather that we're focusing on what's France's experience of nuclear been for the past, since then?

What have they learned? Is it now a safe technology? How can we use it? We just, because there was a massive accident 40, 50 years ago doesn't mean that it's the wrong thing to do. We've learned a lot of lessons since then. If we'd given up flying in 1979, 'cause there were loads of plane crashes in the seventies, we would have given up flying and we would know, we would, we'd have missed out on the enormous improvements in safety that went on in flying and bizarrely that the improvements in safety that went on in the aviation industry had a huge positive impact on safety in all other industries, including the maritime industry and the nuclear industry. So sometimes you can't just stop. Sometimes you can't just stop. Hey, what did you, think of that article?

Stefana: You mentioned that if we stopped using planes to fly when we know how many accidents there have been, well, what would have happened if we stopped using vessels when Titanic sank?

The titanic example like many aviation example, was a striking point for safety and for many other things, right? It was a breaking point, like we like to say nowadays.

I think Anne, unfortunately, I think the Hollywood, and it will sound weird that I say this, but the Hollywood and the media and all that have romanticized the catastrophe that have been happening at Chernobyl, at Fukushima, and all that. There have been so many collateral parties that have gained a lot from that.

And it was just very easy to constantly. It's very easy to rule by fear, right? And we have been brainwashed. Maybe it's harsh to say that, "oh my God, that's the enemy." But I will look at the facts here that I know from my research. Nuclear is stable. AI needs stability. If we really want output there, and if we really want to use AI for the greater use in any industry, it needs stability.

If I look at nuclear, I look at it more as an ally in the fight for climate change, because, like I said previously, it's emissions profile is close to zero. And of course, yeah, we are still scared because of what I've just said. We have romanticized in a very negative way the incidents that happened in the past.

But like you said, I really hope that we learned from those disasters. I really hope that there are people out there, super smart people, that know how to tackle this. In Romania, we have such a nuclear plant, and it's been staying there, doing nothing, and it has tremendous potential.

If I'm not mistaken, it might, we might have more than one, actually, but I don't want to say things that I'm not sure about here. Everyone was scared when they threatened to bomb the nuclear plant in Ukraine, right? For obvious reasons. But if we look at the facts, nuclear power is, it's emissions profile is very low, close to zero, so it's an ally for climate change, for environmentalists.

And an optimist would say "let's learn from the mistakes that we've done in the past with this alternative and make sure that we don't repeat those mistakes for the greater good." We can.

Yeah, we can't continue, we cannot by any chance continue with fossil fuels and I'm saying this from the bottom of my heart.

At PortXchange we can see the data, we have the data in front of our eyes and the emissions that we see on a daily basis are very concerning, extremely concerning. In the ports ecosystem, especially city ports, Anne, there have been studies that are quite, they are not, the media does not take them, the media does not shine a light on them, but the data shows and the research shows people living in city ports have a 3 percent higher increase, if not up to 7 percent higher increase in lung cancer.

Anne: Gosh.

Stefana: Yes! Fertility rates are heavily impacted in city ports. The closer you live to the port, the more you are affected. There are all kinds of health impacts that have been studied. I can send you a couple. We have quite some good studies in our database.

Anne: Send us that, and Chris will put links to that in the show notes. In fact, all the articles today are in the show notes, just as a reminder.

Stefana: So are there risks? Well, plenty, as with any other alternative. The same risks that we look at nuclear power, there is in solar panels, etc. Risks for everything, but we need to look, what are our ambitions for the future? And AI is very ambitious. And it will require a lot of power. But the Earth doesn't have the resources to power it anymore.

Not the way we've been doing it up until now. So I might sound like a nuclear advocate here, I think I might be, but from the analysis that I've read, it seems like the best solution for us. But of course it needs lots of financial investments because it's super expensive, lots of people to be trained and prepared and all that, and most of our nuclear plants are shut down, so I can only imagine the whole thing that will involve to start them again.

Do we have enough time? That's my worry. Do we have enough time? I don't know.

Anne: It's true. I love nuclear. I really wish we'd started 20 years ago. But, like France. We've done amazingly well on that.

Stefana: Yeah.

Anne: In the Maritime, on the high seas, loads of big stuff driven by nuclear engines. Obviously loads of the aircraft carriers, really heavy stuff driven by nuclear engines, ice breakers, lots of the Russian ice breakers have nuclear engines.

And we've got loads of experience in that. So it's interesting. Are we still about to have the nuclear future that we thought we were going to have back in the eighties? It's, very interesting, but I agree with you that Hollywood had really caused us a major problem there. It's too, it's terrible.

Anyway...

Stefana: sometimes, you sound lunatic when you say that, and I'm often being told, but let's just look at the facts. Why fear, why do you fear nuclear? When you ask a normal person. "Oh, because Chernobyl." Okay. Well, Chernobyl had very corrupt people working there, first of all. The whole system was corrupt, and the incompetency was at a high level, etc. So, yeah, I need to speak from

I wear many hats when I talk to you, but right now I'm wearing my climate activist hat as well as a green tech advocate.

We must, not we should or we need, we must look into alternatives because we are running out of time.

Anne: And on that fantastic note, we will close our podcast for today. We must look at alternatives because we are running out of time. Absolutely the correct thing to be saying. So Stefana, before we go, where should listeners go to find out more about you?

Stefana: You can find me on LinkedIn under my name exactly as it is here, Stefana Sopco. I'm the only Stefana Sopco on LinkedIn. I pride myself with that. So it will be very easy to find me there.

Anne: Excellent. So that's very good. I'm also on LinkedIn, as I mentioned before. I don't know if I'm the only Anne Currie, but I'm sure you'll find me. So thank you very much for being on the episode, Stefana. It's been a really fun episode. I really liked that bit at the end where we talked about nuclear.

That was excellent. A final reminder that all the resources for this episode are in the show description below, and you can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables. Goodbye for now.

Stefana: Thank you, Anne. Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: Wooden Data Centers

21 november 2024 | 56 min

This Week in Green Software, host Chris Adams and Asim Hussain, Executive Director of the GSF discuss the latest developments in sustainable software, exploring topics like Microsoft's innovative use of cross-laminated timber in data centers to reduce embodied carbon, the environmental challenges of generative AI hardware, and the groundbreaking Real Time Cloud dataset. They delve into the impact of new international energy efficiency directives, the interplay between geopolitics and sustainability, and surprising developments in China's approach to sustainable technology. Packed with insights, this episode offers an in-depth look at the intersection of technology and climate action.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Asim Hussain: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Asim Hussain: We do not know how much electricity data centers are using. And the amounts,

and I've spoken to researchers who are doing kind of research for the US department of energy, and they're like, "look, we've just had to... it's guesswork."

I'm your host, Chris Adams.

Hello, and welcome to another episode of This Week in Green Software. And I probably should say welcome to Environment Variables, too, the parent name for this podcast, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. Today, we're doing a quick check in with my friend and Executive Director of the Green Software Foundation, Asim Hussain.

Hi, Asim.

Asim Hussain: Hi, how are you doing Chris?

Chris Adams: And if you haven't heard this kind of format before, we are So, we are pretty much doing a roundup of interesting news or news that has caught our interest in the last few weeks in the world of sustainable software to basically share some of our takes and provide a bit of analysis or the things that we've seen, basically.

So that's pretty much it. We'll be sharing links to all the stories and all of the follow on links that we do from this. And I think that's it. That's pretty much it. Asim, before we start though, I know I've just said you're my friend and you work at the ED, but can I just give you a bit of space to introduce yourself about why you have something to do with Green Software?

Because I've just mentioned Executive Director, Green Software Foundation, but I suspect it might be a bit more useful to share a bit of background about where you came from in terms of professionally and where some of that credentials might be. And then I'll do mine and then we can dive into the story together.

Asim Hussain: Yeah. So Asim Hussain, based in the UK. Background and career throughout all the way to this point has been software engineering and about, six, seven years ago, really started to look at kind of sustainability space. That's when we met Chris, did a bunch of things, then launched the Green Software Foundation, helped co-found it three and a half years ago.

And I describe my job as, very luckily and very boringly being able to focus on answering one question and one question alone, which is "how can we have a future where software has zero harmful environmental impacts?" So I'm very excited to get on this call with Chris. 'Cause I get, I love getting wonky and there's not many people as refined in their thinking as Chris.

So we can really dive deep into some topics.

Chris Adams: That's very nice of you to say, and I assure you, I didn't put him up to that. Folks, if you're new to this podcast, my name is Chris Adams, I am the Director of Technology and Policy at the Green Web Foundation, which is not the same as the Green Software Foundation, but we're fellow travellers, we're a small Dutch non-profit focused on a fossil free, entirely fossil free internet, and we are members of the Green Software Foundation along with companies like Microsoft, GitHub, and Accenture, and so on.

I'm also the host of this podcast, and I actually met Asim in person for the first time ever at an unconference called OMG Climate back in 2019. So the connection is.

Asim Hussain: I love that

Chris Adams: I work at the Green Web Foundation. We publish open source software very similar to how the Green software Foundation publishes some open source software, but we also push out like reports and things like that. And I also work as one of the chairs of the policy working group inside the Green Software Foundation, because it makes sense to be working with other organizations with shared goals here.

If you are new to the format of this, it's literally a news roundup, so it should be, expect, hopefully some light entertainment and some interesting takes on this. We'll share links to all the stories that we discuss, and I think that's pretty much the general plan. Everything we share will be listed in these show notes.

So if you are listening using something like Spotify or YouTube, please remember to go to podcast.greensoftware.foundation to actually find the links that we start talking about. Alright then, I think that's pretty much it. Asim, are you sitting comfortably?

Asim Hussain: Ah, yesss.

Chris Adams: Okay, you look pretty happy, so I'm assuming we're going to go ahead with this.

Alright then, let's look at the first story. Does that sound okay?

Asim Hussain: Yeah. Sounds good.

Chris Adams: All right, the first story is, this is from the register, talking about Microsoft testing wooden data centers. So the headline for this is that Microsoft is starting to move into the world of what you might refer to as CLT, which is cross laminated timber.

This is a new building material that's used for construction in general. But because a significant chunk of Microsoft's emissions, their reported carbon emissions the last year or two actually came from the construction of data centers and the actual pouring of concrete.

This is, in my view, this is actually quite an interesting one because it's one potential lever for talking about the embodied carbon associated with using the kind of software that runs in these data centers. And there are some, honestly, quite impressive savings because the thing about wood compared to concrete is that wood can be a sink of CO2 rather than a source of CO2 emissions.

Asim, I'm pretty sure you've had a look at this and given that you used to work at Microsoft, you might have some kind of perspectives on this one, especially last time we spoke about data centers, we nerded out about the use of mushrooms in data center building materials as well, and you had some thoughts there.

So I'll hand over to you. When you read this, what kind of crossed your mind and was there anything you'd people's attention to on this one?

Asim Hussain: Yeah, I mean, I think one of the things I thought when I was reading it was, I don't think they did as good a job as they should have in communicating what the overall reason for using timber is. So basically this is all about mass timber and why is building things wood even a good thing when we're talking about actually growing trees is a good thing? Surely like chopping the tree down that you've just grown to sequester the carbon is a bad thing. And the rationale behind mass timber is that, that's basically the problem is that, when the durability of a tree I think is considered to be about a hundred years. I can't remember, don't quote me on that, I believe there is an official figure, which is when you grow a tree, like how long have you sequestered that carbon? And I think we've all agreed it's like a hundred years, whereas if you, and what happens, that tree then collapses, it goes to the ground, it decomposes, and it goes back into the atmosphere again.

So it's only sequestered it for a hundred years. Whereas the argument around mass timber is, well, let's take the tree before it's, So, let's take a carbon that's already dead and fallen to the ground and let's turn it into something else. It's a building, it's called mass timber. And through this process, we're effectively delaying the release of that carbon back into the atmosphere.

And I think some of the things I read in the past were like 300 years. So, it kind of goes that way.Now the theory there is it only really makes sense if you're building a building that is going to last for 300 years. I hope this data center is going to last for 300 years. That's a question in my mind. But Yeah, that is ultimately the argument on mass timber.

I think the argument here though is slightly different also because it's a relative difference between concrete versus timber as well, there's that aspect to it also, which as we know, concrete is a significant emitter of emissions, irrespective of the fact that, that tree just got felled. I mean, the relative difference is significant enough that the durability I don't think comes into play that much.

So it's an interesting direction. Just, I don't know if you saw this as well, Chris, but also OCP, Open Compute Project. Did you see they're doing, yeah, they're exploring, is it carbon negative concrete? I can't remember, it was, at the very least, it was less carbon intensive concrete, or it might be the kind of concrete which actually, as it cures, sucks carbon from the atmosphere.

Chris Adams: Yeah, few different ways you can make a cut or you can essentially use concrete or/and in particular, cement, which is a big source of emissions to reduce those ones there, because, for the story we're, linking to right now, when we look, talk about the wooden part, one of the things that's touted is, the use of timber instead of concrete can reduce the emissions associated something like two thirds lower emissions compared to steel, for example. That's one of the reasons why it actually is useful and if you're swapping, say, steel out for wood, there's also a saving around 30% around, is what is cited here. And that's largely because creating steel and creating concrete is incredibly energy intensive, but also just the actual process when you make concrete, for example, if you're using cement. If you're going to create cement when you're taking calcium carbonate, which is essentially limestone, there's a clue in the name, carbonate, you basically, you end up separating the carbon from the rest of it to actually make some of the materials, and that ends up being just emitted as part of the process, even if the cost of energy was entirely free, for example.

If you had entirely zero carbon heat, you'd still have that. So there's a few things that are problematic about using cement, but there is actually a lot of new work going into either creating low carbon cement or even using non different processes like using electrolysis, which isn't even relying on calcium carbonate, but using things like calcium silicate to basically precipitate out the necessary, essentially like compounds you would need it to actually create cement.

There's lots of things that are place right It's really interesting. And Microsoft is actually spending quite a lot of money in this field as well. We can share a few links to this because Microsoft is actually in many ways a complicated player in this role, because there's lots of things they're doing in terms of enabling emissions.

But when it comes to the actual construction of this, yes, they're building masses and masses of infrastructure, like in the order of tens of billions of dollars each quarter. But they're also one of the few organizations that's spending significant amounts on the different clean kinds of cement that we had inside this.

And that's the kind of thing I would respond to when you hand it over to me there.

Asim Hussain: Yeah. No, I think I'd also know that Bill Gates is also investing heavily, kind know he's not, officially leading the organization,

"officially,"

he's not leading the organization anymore, but obviously there must be some influence there as well. So I didn't realize the carbon, so I actually assumed the carbon was from curing of the concrete, but you're saying it's from the construction, it's from actually generating the raw materials.

Chris Adams: Yes, I'm very quickly leaving my domain of expertise, so I'll share a blog post that I wrote when I learned a bit about this, because there's a really interesting podcast by, okay, about as interesting as talking about cement actually gets, I suppose, that actually dives into this by a guy called David Roberts on the Volts podcast, where he does an interview about how this is made, how there are different approaches that you can use for doing this, how you can do it entirely fossil free.

so there are some options that you have here, but there's also, it brings up, raises all these other queries about, okay, if you are going to do this, how do you move cement around? Because in many cases, cement has a very, it sets in 90 minutes, right? So just like electricity is hard to store, things like cement can be very difficult to store.

So there is a whole question about, "okay, if someone wants to purchase this, how do they go about doing that?" And that is how some of this gets a bit more complicated. Complicated. So that's one of the options, and I think we should, if there's sufficient interest, it might be worth actually talking a little bit about reducing the embodied carbon and what levers are available, because this story is about wood, and we've gone down a kind of cementitious kind of sidebar, I suppose.

But there's a lot of interesting stuff happening with wood. For example, you mentioned the OCP, which that's the Open Compute Project. There're actually 3d models of wooden data center designs. There's one chap called Karl Rabe, who has been running the Wooden Data Center company for years, who's been doing this kind of work, and, they, I find it quite interesting and endearing at the same time, there's something pleasing about this, and you realize that this idea of cross laminated mass timber ends up being, it's getting increasing amounts of mindshare because it's one of the ways that you can build relatively large structures for a lower carbon footprint and the, for example, if you're in the UK and you ever go through London King's Cross, Google's shiny new headquarters uses all this kinds of CLT, this cross laminated timber well as one of their ways of reducing it. So it's very much seen as one of the trendy materials that you might actually use because it, aesthetically, is actually quite pleasing.

The thing, the elephant in the room when we talk about using wood for construction is "doesn't it burn down?" We had the Great Fire of London, which kind of redesigned London, as it were, right? And it turns out that the actual tooling that you have, the actual kind of material like CLT, it actually is surprisingly fire resistant.

It smoulders rather than burning in other ways. And this is actually one reason that there's actually something that you might initially worry about, but is actually not such a concern actually. So yeah, there's, I think we've come up with a bunch of links that we should add to this for people who are interested in this, because it's one of the levers.

And we know that this is increasingly, when you have organizations doing a massive build out of infrastructure, if they're going to build new buildings, Then This is one of the tools in the toolbox for reducing the embodied carbon, and Asim, what's the letter in the SCI that we have for this then?

Asim Hussain: I, intensity

Chris Adams: yeah, so it'd be

Asim Hussain: Oh no, it'd be the embodied. Yeah. Sorry. Yeah.

Chris Adams: So that would be one, one lever you have for embodied intensity that is sometimes included.

Asim Hussain: And also I think we should probably like, correct me if I'm wrong, but just cement is, I believe, the last time I looked, 7 percent of global emissions. so yeah. So that's why this is such a.

So why are we talking about cement? It's 7 percent of global emissions, just cement. So it's a finding alternatives is not just an edge case. It's one of the main things that we need to do.

Chris Adams: Absolutely. Yeah. So for context, that's about three to five times people's estimates of the digital sector. So this is one of the big ones that we need to be aware of. And it's essentially one of the most used materials in the world, basically, as well. We'll share some links to that because it's, when you start looking into it, it's surprisingly interesting if you care about reducing the emissions associated with digital, and you accept that a lot of these things take place in buildings, basically.

Alright, thanks for that, Asim! Shall we look at the next story?

Asim Hussain: Yeah, let's do it.

Chris Adams: The next one is, this is about the other part of embodied emissions actually. So this is a paper that was published, that initially published in Nature, but a preprint of it is available also on another website called NetworkDEE.Org. This was actually talking about the e-waste challenges associated with generative AI.

And essentially one of the things that comes up with AI. is that when you're using different kinds of hardware, while you can use existing hardware, like typical CPUs and things like that, a lot of the large providers and a lot of the kind of interest has gone into building new machines, or new kinds of specialized infrastructure, or specialized tooling, like NVIDIA's chips and stuff like that.

And this paper does a kind of material flow analysis of what this might look like if you project this forward over the next 5 to 10 years, for example. And this is actually, in my view, I think it's quite interesting because we often talk about the energy impact of AI and one of the reasons that it's quite difficult to actually even get a figure for the embodied impact of actually creating chips.

And when you move, if you're going to talk about, say, onshoring, moving the construction of chips to, say, the EU or in America, that has knock on implications on where the carbon emissions go and how visible they become, actually. Asim, I'll hand over to you for some

Asim Hussain: I've,

Chris Adams: here, actually.

Asim Hussain: I've not actually.

read this paper, so I can do what I normally do, which is to

to wing, off the cuff. Yeah, it's interesting. A, you're right. The embodied carbon is still like the most stunningly hard thing to get. There's still only a couple of examples everybody kind of interpolates from to apply to everything else.

Sorry, just, so what this article is saying is that there is a faster, or should we say, what's the term in the cloud space? Not turnover, but the length of time.

Chris Adams: life of

Asim Hussain: Yeah. The life, is it saying that the useful life of AI chips is lower?

Chris Adams: I think there's two things at work here. So one of them is just the fact that people are using this particular kind of hardware, which may be more energy intensive. That's one of the issues. And there are basically, I think there's two arguments being made here. One of them is that because the,

because things are moving relatively quickly, a chip doesn't have the same kind of useful life cycle in the same data center, for example, so if you want to keep up, you're incentivized to buy the next chip because it's that much more efficient than the previous chips, right?

So, there's this idea that by doing that, because you have this kind of compressed hardware cycle in order to keep up with other providers, particularly in the kind of given space, then that is actually a driver of creation of much more in the way of electronics, particularly because there isn't much, this isn't a very circular sector right now.

So it's not like those chips are basically being broken down and then remanufactured into new feedstock for super efficient new chips. That's one of the things that is actually being, that's the argument that's being made here. And that has implications here. I think the figures that are used in the abstract is something in the region of 16 millions of cumulative waste by 2030, specifically for the tooling used for generative AI.

So that's like a, that's a non zero figure, that is something significant. I think when you look at this though, one thing it's worth actually, and one thing that I actually thought was, oh, this, if you've got this kind of endless treadmill. Does that mean that these chips are being thrown away?

And I'm actually, when I spoke to some people, I think it was, I actually asked Boris about this. Boris Gormazaychikov, who's the AI lead at Salesforce. Yeah, I asked him about this, and he said that is one of the factors. But one thing we're actually seeing is that given that there's, in the last couple of years, because it was just so hard to actually get hands on any of these chips, even things which were not the latest and greatest, they still were being used.

So it may be, it's not like these things are being thrown away entirely, right? It's not like they're going straight to landfill. But one of the problems we find is that there's almost, there's very little data on the circularity of these chips to see if they are being put to use, or if they are just going to essentially like landfill or not, basically.

This is one of the big problems that we do have and I would love to actually have some meaningful data on this because it's one of the big, it's one of the kind of generational shifts we're seeing in the sector right now. We've never seen so much money spent so quickly in this field with so little data being disclosed about what the knock on effects of this might actually be.

Asim Hussain: Two thoughts popped into my head as I was reading this, well, the abstract and what you said. And from other conversations I've had, especially in some of the places I've worked. The latest chips. Like, why do people want to use, why do Gen AI and AI startups and AI companies want to use the latest chips?

It's because dollar for dollar, it's cheaper for them from their perspective. They, it might cost more to rent those chips, but the AI job gets done sooner, so overall, it's just cheaper for them to use the latest generation of chips. And therefore it's more profitable for organizations.

But I wonder if there's this, almost this, and I don't know if this is because again, there's no data, but I wonder if there's this kind of constant battle then to get the latest chips. Is somebody left holding the bag with kind of slightly older chips, which aren't as profitable anymore, no one wants to use? And then that's just, know, it's not this perfectly free market where everything, all the information flows and the prices of the older chips, maybe it's just a mad battle for everybody to get the latest chips because that's the only thing anybody wants. And we're decommissioning, maybe we'll decommission these older chips sooner than the necessary just because of the incentives in the market.

That's one thought I had, and again, based off of nothing, so there's no data, so take that with the usual pinch of salt. But the other thought I had was just looking at some of the stuff that's happening in the space right now, like moving forward. I've forgotten the name of the project.

I will find it and I will get in the show notes, but there's, NVIDIA is now, they don't sell the boards, they don't sell like graphics cards, they sell entire vertically integrated units that you just like slot into a data center with everything baked in, the cooling baked in all these kinds of, because that's the only way you can get, The true levels of efficiency is if you, it's like the Apple ecosystem.

It's the only way to do it is everything is just owned by one company and it just does everything the way that product does it. so that I think will, A, make things more efficient, but B, potentially make things more complicated when it comes to e-waste, you know what I mean? if there's like a load of graphics cards.

Maybe I was thinking in my head, maybe in the future, everybody's going to have really great gaming experiences. Cause you'll have five year old, everybody's going to have a really amazing graphics card secondhand on their computer. But with these vertically integrated units, maybe that'll be a lot harder to do.

I don't know. I don't know what they look like inside. I don't know if it's a bunch of boards that you just unplug, or whether they're soldered together.

Chris Adams: so there are, you're right. So this is, maybe there's a couple of links that may be relevant for sharing this. So there's actually a really interesting piece by, I think the folks, the AI Now Institute actually spoke about the supply chains associated and where the concentration is for the creation of these chips.

Because when you have one or two extremely dominant players, then they're very much incentivized to not have too much stuff which is interoperable because that forces that, that means they end up being the people who get to set the prices, and we've seen NVIDIA, for example, briefly become the most valued, valuable company in the world, eclipsing both Amazon and Microsoft briefly over the summer before falling down to being in the top five position, for example.

So you do see that, and there is, that is one of the Issues that we do see is that you have this kind of vertical integration increasingly being pushed by this, but you also do see various other kind of pushes for this. For example, we know that like various hyperscalers have been looking to break this reliance on a single provider by having their own chips.

So Amazon has Published there, there's a, they have announced Trainium as an example. Google have been building their own chips, for example. Microsoft, I forget the name that they have for theirs as well, but set that you have and you do have AMD being in the kind of like low, far lower down compared to these other providers as well in terms of like sales and usage, I suppose.

'Cause they do have some options out there as well now, but yes, you're right. It's...

Asim Hussain: But the other argument, mean, you could argue it from that. I mean, I think that the thing to understand is like everything has 15 different arguments for it. So, you vertically alignment could be more efficient, but then you're right, it could be also vendor lock in.

It could be a bunch of these things all at same time. And the arguments for different like Amazon, everybody having their own chips is also pretty decent argument app I'm running on a MacBook, which has its own chip because once you vertically align, the efficiencies increase. And so the argument from Amazon, at least was like, "we build chips that are optimized for the exact workloads that we run.

And that's how we gain efficiencies." So there's multiple things in this, but you're right. That makes it much harder from an e-waste perspective, because then instead of having the same, even just, I don't even know if they're all x86, they must all be x86. actually, no, they're not now, not sure, but anyway, there's more, it's much more complicated when you come into e-waste when there's different protocols and different architectures.

Chris Adams: About this whole idea of the cost of compute going down beneath a certain figure. There's a link that I've added to the show notes from the, actually, the publication, Latent Space, called the GPU Bubble, which does explicitly talk about how the cost of rented GPU compute has changed over the last few months, and what might happen as you have new clouds being available,

and what the introduction of new chips does to the cost of some of the other kinds of compute available, like you mentioned here, that's probably, in my view, I think that's one of the most interesting pieces of analysis about where the cost of this might actually go. And it may be the case that just how, in the 2000s, we end up with loads of dark fiber left around that we're still using now, as a result, when everyone had this massive build out for the initial dot com bubble.

You might see something comparable with AI, where we do have massively fast chips available for all this usage. So that could be the case, but yeah, it's a bit early now, we're not quite sure. I think there's a bunch we've got there. Well, we've added a few links, so if this, any of this is If this has been interesting to you, a listener, then there should be a bonanza of links available for you.

Thanks for that, Asim. Shall we move to the next one then?

Asim Hussain: Yeah. Go on.

Chris Adams: All right. So this one is, this is actually referring to the Green Software Foundation's, one of their projects called Real Time Cloud. This, I think sometime in October, the steering committee inside the Green Software Foundation pretty much merged in or approved the merging in of the yeah, ratified, the most recent work on Realtime Cloud.

Asim, you're in the steering committee, right? So maybe it might be useful for some of this to be explained a little bit to you, or like maybe you might talk about what this process looked like at your end, because I've been joining these meetings, but it's very much been the work of Adrian Cockcroft and Pindy Bhullar really pushing this to essentially create a datacenter-level list of all the carbon intensity figures for the three biggest clouds. So if you run a workload on Microsoft or Google or Amazon, you basically have, in my view, the most complete open data set available and where there has been information about the efficiency of a data center in the public domain, it's included.

Where there's information about water usage, it's also included. And this is, I'm really happy to actually finally see this kind of ratified because there is now something approaching a shared consensus view on what this data set, what this data should look like. And I think this is the thing I would.

Point people to, because I think this is actually used by some of the other software like Impact Framework and some tools like that actually. Maybe Asim, if I hand over to you to let you talk about what the process is when the steering committee inside the GSF ratify this and then maybe talk a little bit about the impact framework thing because there's some follow on things that I might talk about after that but I suspect that might be useful for people who are trying to figure out how they might use some of this information themselves when they're trying to quantify the environmental impact of the services they're operating.

Asim Hussain: Maybe I'll just start off with that. Actually. I'll just start off talking about how the Impact Framework team got involved. I think it was during the hackathon this year and we needed, what we wanted to do was a mechanism where we could help people like compute the carbon emissions of their cloud usage.

And so, the real time cloud was a data set at that point where it gave you regional information. So I forget all of it, but it was, PUE, power usage effectiveness. I think CFE was in there as well. It's.

Chris Adams: Carbon free energy. Yeah.

Asim Hussain: Yeah, actually, it's for some reason, it's dropping from my head, which actual point of that data was needed for the carbon, for the hackathon.

But the impact framework team went, "wow, this is the data set that we need." And so they ended up building, I think all, the only thing we needed to add to actually was the geo location of the data center region so that then you could do good queries with Watttime and Electricity Maps and things like that.

So we ended up building a plugin for Real Time Cloud. We actually then turned into a baked in generic kind of CSV plugin, which was a good call. And so now you can use, now you can use this data set as one of the inputs to a computation, an end computation for your cloud usage and or, other end-to-end usage.

That's how the impact framework team got involved, which is, yeah, I think a really great example of kind of cross pollination of work inside the GSF. And just in terms of like how the process is here, I mean, this is driven through the standards working group. The RTC project is a standards working group project.

There, again, you've been part of it, Chris. It's a consensus-driven process. So the RTC group got together and said, we're happy with this dataset. We're happy with the specification associated with the columns. And that's the way we see it. This is a project which is not just data, but the specification for what the column should mean.

And this is our version of the data that we've published. Now, the goal, ultimately, is to get this into ISO, that if you wanted to disclose a dataset, similar dataset, and you were a different cloud provider, you could. As long as you exposed it with exactly the same columns, with exactly the same meanings. And that's the kind of power of standardization, is getting everybody to talk the same language. Right now I believe, Chris, that it's been ratified, but the process is that we're leaving it out there for six months for public feedback and broader review with an aim that in six months time, we'll take all the feedback, package it all together, and that's what we're going to publish into ISO, which will be another process. ISO is a process where you get reviewed by 175 country bodies, and then you've got to respond to their feedback and make adjustments, and if you have, and once you make those adjustments it gets accepted and you get into ISO.

Chris Adams: And when you say ISO, you're referring to the International Standards Organization, essentially the place where people standardize on stuff, basically, right?

Asim Hussain: One of the main ones, yeah. And one of the things we've noticed, as soon as, I mean, what does it mean? Why, is it important to get into ISO? It's important to get into ISO because what all that everybody's really looking for is, "oh, should I use this? God, is this the right thing to use?

Should I trust this?" And as soon you, the reason why people are really keen to use kind of ISO standards is they know it's going through an incredible consensus and review process, both internally at the GSF and then not only did the RTC team had to agree, then the whole standards work group had to agree, then we send it out to all of our members to see if they agree.

And if they don't, if they don't reject it, then it goes to the steering committee who have this ratification process. And in all honesty, I get very nervous at that process. Cause I'm like, yes, it is their right. And they do ratify things and they've never abused it. But it is kind of... I want people to have the conversation as part of the RTC team, rather than right at the end of that whole journey as a ratification process, but we see is more like ratification typically in the physical world is the actual process of everybody getting together and celebrating and signing a contract. It's actually more of a celebratory thing. It's ratified. So that's the way I think people should view kind of the ratification processes, the celebration from the steering committee that, "yeah, we signed this. We love it. We want it go ahead." So that's that.

Chris Adams: Okay, cool. Thank you for providing the extra context, because I work on one side where we're trying to figure out how to get the data into a dataset like this, but I wasn't so aware of okay, who gets to shoot this dataset down before it gets actually used in various places? And the thing I might share with you that What is probably of interest now is that given that we know there are various laws around the world that are, there's for example, in Europe, we have this law called the Energy Efficiency Directive.

I'll share a link to a post that we have published at the Green Web Foundation called Happy EE Day to those who celebrate. This is specifically about the fact that there's data that's been collected at a data center level, which is actually higher resolution than this kind of regional level that you might see here, which in my view, basically sets out a way for all the other, the providers who aren't just Microsoft, Google and Amazon to have a way of sharing the same kind of figures so that you actually start making comparable, you can start making comparisons between running a workload in one data center with one provider versus another one, for example, so that's some of the work that's going into this right now. And hopefully we'll see some more data come out because we now have this data being published. Or there are laws to get this data published in various places. And there should be a data set coming out, I think, early in 2025 for all of Europe, which is very extensive.

This will be any data center that uses, I think, more than half a megawatt of power. And given that most of the hyperscalers use tens of megawatts, that's going to be thousands upon thousands of data centers. So it'll be interesting to see what actually comes out from this. And what we'll find over the next month or two is what data has actually been shared by providers already, basically, because we are starting to get glimpses of this.

For example, in the Netherlands, we're seeing some submissions come to the public into the light of day. And I believe there's some stuff in Germany coming as well. So we will see, so you get some advance notice, and this might be stuff that can come into this. So we can start creating essentially an open resource of this information, so that if you're a operator of infrastructure, you've got an idea of what can actually measurably reduce some of the emissions on this.

Asim Hussain: And did you say it's definitely going to be a public dataset?

Chris Adams: I'll share a link to, again, a little bit of a kind of like wonkish thing, basically. With the Energy Efficiency Directive in Europe, which is, all of Europe, basically, that's 27 countries.

There's two things. If you operate your data center, you're mandated to publish a bunch of data points, like how much energy have you used? How clean is the power? How much of the power came from certificates versus on site generation and stuff like that? It's quite detailed. And...

Asim Hussain: When you say, sorry to interrupt, but when you say 'published', do you mean like in terms of a CDP where you have to publish to an authority who will then keep it secret? Or,

Chris Adams: make available. Publicly, make publicly available. However, there is a kind of caveat which basically says, "if you're going to do this, you need to do this, except in the case where something is considered a trade secret." So what we've seen is that some providers are sharing this information, and some providers are not sharing this information.

This is the thing that we're seeing now. And

Asim Hussain: So, now we know which data centers MI6 uses.

Chris Adams: Basically, or we're saying which companies are being, when companies talk about being really transparent, we have a way of seeing which one, which companies are walking the walk when it comes to talking about transparency and which ones are not being as transparent as their comms are saying so, basically.

That's one thing we're seeing. However, even if you don't publish the stuff, you still need to report to essentially the European Commission, and they will be creating an aggregated data set that they're publishing next year. So that won't be like "this data set run by Equinix is used this much energy last year," they'll have some figures so that we've got an idea of what how big this problem actually is, because it's 2024, we've been using computers and data centers for decades, but we have no real meaningful numbers at this level of detail. And even half a megawatt, 500 kilowatts of power, that's still quite a, that's a few racks at least, right? So that's still a bunch of data centers which aren't being included.

But this will give us an idea of, for the first time, just how much power is actually being used. And given that in Europe, they have binding legal targets to reduce emissions, they do need to know this because they said we need to halve emissions, more than halve emissions by 2030. And if you don't know how much power you're using and you don't know how clean that power is, it's very hard for you to tell if you're on track or off track. So that's one of the things that will be coming out next year and that's, I'll share a link to me diving into this, because I ended up having to make sense of it myself, and I found it quite difficult to read a bunch of these laws, but I've shared my understanding as I do this, and hopefully that should be useful to other people now, because, yeah, it turns out this data is actually being published, it's just, it's the first time people have done it, just like with other laws, where people are figuring out how to report on sustainability with legal teeth for the first time, and, yeah, it's a bit of a mess in certain countries right now.

Asim Hussain: I think the takeaway for people who might not understand is we do not know how much electricity data centers are using. And the amounts, and I've spoken to researchers who are doing kind of research for the US department of Energy and they're like, "look, we've just had to... It's guesswork. We've sent questionnaires and a bunch of people responded, a bunch of people didn't.

We inferred the rest and there's massive error bars." I mean, that's where we are. We don't even know, we can't project forward. So I think that's a really good thing just to have some solid data to even know where we will be in two, three years time. Yeah.

Chris Adams: And we're going to continue on this thread to come to the next story. So, that's what's happening in Europe, and that's, well, that's what we're seeing right now. But we know that the wind can change in politics all the time. And we've just seen a significant change in America. This is a story from Data Center Dynamics.

That's literally called Trump Won, Now What? And this is a bit of analysis about what we might see as a result with a new administration in the US because in the US we did see some things were looking actually quite positive in the world of AI. There was actually legislation around reporting on AI, just like Asim mentioned.

And there was unparalleled investment in cleaner forms of energy. And now we've got a new set of people who are in power who take more of a dim view on climate change and are unabashedly in favour of fossil fuels. Asim, I'll hand over to you because you said that you've got, there's a few things that caught your eye that to talk about here.

Asim Hussain: I think, I mean, I don't want to get too political and I'll try and keep it more to, I don't know if the facts is the right word. But yes. Okay. Look, when Trump came into power last time, it was quite a challenge for our community. He pulled out the Paris climate agreement. There's a lot of kind of negative rhetoric, which I don't know if he led or he was just, it gave a voice to, about climate change and the work that was being done. I think this time round though, the reason I'm a little bit more positive and it might be, I'll probably preface for this to everybody saying that I'm neither left nor right. I don't identify with a particular side. I identify with the side of the fight that we're fighting.

That's it. And so, like, I see Trump as somebody who is somewhat influenced by the people he hangs around with. and what he watches and what he hears and even though these are two people that I think the left are particularly critical of let's just look at it factually. There's Elon Musk which is part of the, we don't know how he's related to the place but he's going to be somewhat, he has a voice with Trump. And I mean this is a guy who, whether or not he, what his personal beliefs are in this space, he has a solar power company, he has an EV company, and we've seen Trump's rhetoric change to be more pro EV, at least up until, the election completed, so my hope is there's going to be a little bit of influence in that direction.

I mean, one of the things about Elon is he was part of the government committee on sustainability, I think it was. And he dropped in protest of Trump leaving the Paris commitment. So he was very against that whole process, and he is very pro climate, so there is some positivity there. I'm hoping that there'll be some influence to him in this.

This is all what I was thinking before yesterday. I'll tell what happened yesterday. This is what I was thinking before yesterday. That might be more of a positive influence there. And RFK, I think a lot of people know him for one aspect, but maybe people aren't aware of the fact that his entire career is as a environmental lawyer.

So that's what he spent his entire career at. He was a professor at I think Pace University teaching environmental law. He has litigated against many companies on a pollution perspective, including a big case against Monsanto. And he has been a big supporter of not just climate, but kind of environmentalism and caring for our planet for a while.

And I know he's got other opinions which are of a different spectrum as well.

Chris Adams: Who needs vaccines?

Asim Hussain: But let's just focus on what we want to focus on. So my hope was that Trump would listen, be influenced by this space. That was until yesterday when Trump announced some of his first appointments, which now turn out to be quite Warhawk-ish.

And so the whole story of "I'm going to, I'm the peacemaker. I'm going to bring peace to the world" is I think a little bit more up in question, but there's still some hope there. He has the other middle as appointments. So we'll see what happens. That's for me, what I'm looking at right now, is I'm looking to see what actual appointments he makes.

And I think that'll be the judge of what his tenancy is going to be like. And I think we're going to know pretty shortly in the next couple of weeks. Is it all, is what I'm saying nonsense and he just took advantage of these people and then will do whatever he wants now he's in power and listen to the people he used to listen to?

Or can it be swayed somewhat? So I'm probably not as depressed than most people. I'm still a little bit hopeful, but, yeah, that's just my very pragmatic viewpoint on this. That's where I'm thinking.

Chris Adams: Or the environmental impact of digital services. So there's one thing. It looks, given the, you mentioned things like nuclear, for example, you have, and you actually mentioned specifically, Robert F. Kennedy, who's, one of his achievements, essentially, or policy achievements that he's touted before is being able to stop nuclear, or to get nuclear plants closed down.

That is at odds with some of the other work, or the other kind of commensive where there seems to be a lot of interest in nuclear as one of the ways to have a lower carbon form of energy. And I'm actually not sure which way this is going to go, to be honest, because one thing we've seen is that in the current like laws right now, which were passed under the previous administration, or the current administration, things like the Inflation Reduction Act, they work out to be really good for tech companies right now.

So if you, for example, right now, it's not actually that well known but let's say you see these announcements by Google or Microsoft and Amazon talking about "we're going to deploy a bunch of new data centers and they will be powered by renewable, by nuclear power" for example. If you look at how that's being paid you'll see that essentially half the cost of that is coming from the government themselves, right?

So you've got essentially your 50 percent subsidy to the largest companies in the world and you can imagine that there's going to be a lot of pressure to hold on to those particular, that kind of policy set up, because that is worth literally billions and billions to existing tech firms. So that might have some impact on like the environmental impact of the energy that we use, but you also have this thing where you've got someone who's very keen on expanding America's role as the largest producer of fossil gas right now.

So that has a flip side. I'm not quite sure which direction it's going to go in, but it does look like it's going to be bad news for, if nothing else, wind. It doesn't look like it's going to be great. So the idea that One of the knock on effects of this might be

Asim Hussain: Yeah. Elon doesn't own a wind company, so I don't

Chris Adams: but I said one thing that might bring up from a kind of operational point of view is do you end up with a slightly less volatile grid because you've got a bunch of huge amounts of money being thrown at stuff which is either fuel based so it can be dispatched internal all the time or nuclear regardless of your feelings of that which is a kind of a steady thing. That might have some implications of what you end up building as a result Because I could start talking about all the other kind of

all the justice themes around that but we're coming to time and I don't want start on that because...

Asim Hussain: This is a rabbit hole. But yeah, I think probably you and I agree the next couple of weeks are going to be very interesting and we're going to, I think it's going to affect the next four or five years in quite meaningful ways

Chris Adams: And the other thing the other flip side of this might be if you now have this scenario where you have tariffs which massively we've seen stories we'll share a link to some stories inside this where It looks like assume under this kind of tariff regime you would have basically laptops and electronics doubling in price. If you suddenly just doubled the cost of embodied stuff like of hardware that's going to create other drivers of what you actually then choose to do. For example you might hold on to hardware for longer because it's so expensive to replace now It's twice as expensive to replace. So that might change the steps that people end up being incentivized to do under this kind of regime for example. So there's some stuff there and it's also worth bearing in mind that's just America is obviously where one of the largest sectors but it's not the only one in the world. There's also China and there's all kinds of stuff happening in China for example as well. I'm gonna give you the last word actually if that's okay Asim because you were literally in China talking about green software and I wonder if there's any things that caught your eye that you might point people to or if there's any kind of things you saw there?

Asim Hussain: I think you're talking about, I've got a few more minutes, I think you're talking about this, think, I can comment on that, but I think you've got to comment first on the actual, yeah,

Chris Adams: it's a lot of the time when we talk about green software we talk about stuff that's happening either in Europe or North America because that's where the majority of the money is being spent in Western markets like Western markets right but China is also like the second largest source of emissions. Second largest economy in the world right now. There's a huge amount happening there and we tend not to see so much of that but there's actually someone who I've been starting to follow an analyst called David Fishman He's been showing some really interesting stuff because you basically see some in my view quite radical decisions being made in China to actually impact the environmental like consequences of using digital services. There was a thing published in May Basically this Energy Efficiency and Carbon Reduction Action Plan. And the short version of this is basically by the end of 2025 the goal is to have I think 80 percent of all the new data centers that are being created 80 percent of their energy has to be coming from fossil free sources, renewable energy which is, when China gets something done they move incredibly quickly and there's massive human rights consequences as a result of that, but in this one here this is something which is something that a lot of us are sleeping on. Like we talk about say my organization talks about fossil free internet by 2030 and we might see the UK government saying "we can achieve a carbon free grid by 2030." And in Germany there is a target for 80 percent renewable energy by 2030. And then in China for data centers they're talking about, yeah, 80 percent by the end of 2025. The caveats I need to share, this is stuff which is translated from the actual Chinese documentation, so it's going to be hard if you don't speak Chinese and that's why I'm pointing to the analysts who operate in this field and share some of this, but it's just so much more, moving so much faster than we are. And it's something that we don't really have that much visibility on. I think it's worth actually being aware of if we think about the kind of global picture for this because you can come away with the idea that "okay there's an election take place and that means there's never going to there's not going to be anything happening in green software." But yeah, the world is larger than just North America, for example, there's stuff happening in other parts of the world And in many ways they are being more aggressive and more ambitious on climate than we've seen before basically.

Asim Hussain: So So my kind of one addition to this, and I want to acknowledge this is like an anecdotal single experience of a human being who went to China, but I went and I was part of an IEEE conference and there was a specific track, a specific room focused on, sustainable software. And so there were a number of players there who, you know, obviously in China actively involved in Tech sustainability in China, some data center operators, smaller data center operators, we didn't see Alibaba. Actually no, Alibaba were there. So we saw some like data center operators there. And I remember being part of a panel and the various questions. I remember just turning around, I think the, EU AI Act had just come into force or just, yeah, just come into force at that time.

And everybody was talking, in Europe, everybody's talking about the EU AI Act. If I was on panel and I was talking to a bunch of people in the tech sustainability sector, they might not have details of the EU AI Act to speak hand. But if I asked them, "what are some of the policies that the EU is instantiating which is driving some of this momentum?"

They'd be able to rattle off like EU AI Act CSRD, CSDDD, maybe they won't be able to double, triple click into kind of the nuance of it, but they'll be able to rattle it out. I remember being part of this panel and I asked, "what is the policies," I was thinking about the EU AI Act. I said, "what are the policies the Chinese government is enacting, which is driving all of your amazing work that you're all doing?" Nobody could mention one. It was silence across the panel. And so that one anecdotal state, and I did some Googling as well, and I couldn't find much, maybe that's my problem, I'm Googling in English, maybe it's all in Chinese, I will acknowledge that. But I think from the takeaway for me, that is, there does seem to be a disconnect between what is being stated at kind of the state level and what is being understood at the operator level. If this was Europe, and if this was the kind of my world, our world, soon as there's a hint of a regulation in this space, we're all talking about it, every person in this space is like using that as a reason for why you should invest more in the green space and you know this and that and the other, but that I didn't see that over there.

Could be a cultural difference, could be something else, or it could just be that maybe, and again, I don't know, and if

someone's listening to this and actually has better information, I would love for you to reach out and to help educate me, but is, I mean, this could just be something that's said on the state level, which hasn't operationalized down to lower level. I don't know how it works over there, but what needs to happen for this kind of thing to operationalize? So we can have a conversation with Alibaba and all these other cloud players. I mean, what, where's their page on their websites talking about how they're going to help meet this thing?

There's some, there's something missing here and I don't quite know what it is, but it could be cultural. It could be anything. And I want somebody to educate me on this.

Chris Adams: Do you know what, that reminds me of a piece of work we did last year with an organization called Wikirate, where we to track the top ten domains in the world which were, by traffic. Do they have sustainability policies, what are they doing there? And some Chinese providers did come up, and they do actually have something inside that.

I'll share a link to that, because it wasn't particularly easy for us to find this, but that was some information there, and they were some targets that were, they, it's weird. Like the thing you see in China is there's an idea very much of "do more before you say, rather than say before you do it." So inversion of what we often see in the West, for example, are like, this is the thing, like you, China can be referred to as a totalitarian state in many ways, but that means that when you see a policy decision taking place quickly, what do you want?

Do you want green neoliberalism? Do you want green totalitarianism? There's like a whole, you can have all these kinds of.

Asim Hussain: Maybe it's just the way, because of the way Europe's different. So I would expect the conversations, and I want to just, I want to make sure I add to this, the panel I was on and the people I was speaking to were doing amazing stuff. They were

Chris Adams: Was this the ACM one that you were referring to or

Asim Hussain: IEEE

Chris Adams: IEEE. So you were doing the talk for that.

Asim Hussain: There was great, there was some great, even some data center operators.

And I was like, "you need to talk more publicly about the work that you're doing." But yeah, there's some great work happening over there.

Chris Adams: Well, maybe we'll see some stuff come out of COP 29, because COP 28, digital and green digital came on the radar for the first time, really, because when I, when Green Software Foundation sent me to COP 27 in Egypt, it was basically not on the radar in the slightest, no one was really talking about it.

So maybe we'll see something come out of COP 29 as a result from this, because there were some initial movements inside that. All right, Asim, I think we've gone over a little bit on time, but this has been fun to catch up and I hope the kind of takes here, or the links were at least maybe useful for people.

I guess what we'll normally do is we'll just make sure that the things we did speak about, we've got some links to follow up so that people who were curious can follow on from this. And yeah, mate, really lovely catching up again. Let's do this again next month or something like that, all right? Take of yourself,

Asim Hussain: All right. Cheers, Chris.

Chris Adams: Ta ra!

Asim Hussain: Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

Shaping Web Sustainability with the W3C

14 november 2024 | 37 min

In this episode of Environment Variables, host Chris Adams dives into the evolving landscape of sustainable web development with Alexander Dawson and Tzviya Siegman from the World Wide Web Consortium (W3C). Dawson and Siegman discuss the W3C’s efforts to develop Web Sustainability Guidelines (WSG), a comprehensive set of evidence-based practices aimed at reducing the environmental impact of web technologies. They explore the creation and potential impact of these guidelines, especially as global interest grows in embedding sustainable practices within web standards. The episode also covers the challenges of driving adoption across public and private sectors, the role of testability in sustainability guidelines, and future directions for standards that minimize digital carbon footprints. This engaging conversation provides listeners with insights into how W3C’s sustainability initiatives could shape the future of the web.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Tzviya Siegman: LinkedIn | Website
Alexander Dawson: Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

The Week in Green Software: Greening Web Standards at the W3C w/ Alex Dawson and Anne Faubry [04:20]
Web Sustainability Guidelines (WSG) 1.0 [04:56]
GRI [16:36]
RDF-star Working Group [19:26]
General policy framework for the ecodesign of digital services version 2024 [21:34]
Sustainable Web Design Community Group [31:04]
Sustainable Web | Interest Groups | Discover W3C groups | W3C [31:06]
The 2024 Web Almanac [33:40]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Alexander Dawson: We have tried to, even at the point of being a community group,

treated this as seriously as possible and made sure what we've created to be as evidence led as possible.

Chris Adams: Hello, and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host, Chris Adams. Hello, and welcome to Environment Variables, where we bring you the latest news and insights from the world of sustainable software development. I'm your host, Chris Adams. Green software covers a lot of different kinds of software. And today, we're checking in on the work of a group of fellow travelers in the world of web development, and more specifically, the World Wide Web Consortium. If you're not familiar with the World Wide Web Consortium, the W3C, it's effectively the place where people who build the web define how the web works. And where standards like HTML works, how CSS works, and how they're made accessible, and so on. And long before the Green Software Foundation existed, there was the W3C Sustainable Web Design Community Group, chaired by Tim Frick at the agency Mighty Bytes, who was also the author of the O'Reilly book, Designing for Web Sustainability. And in the last few years, there's been a real uptick in activity in this group, with the creation of the W3C Web Sustainability Guidelines and regular monthly community calls to coordinate the work on developing more sustainable web development practices. So what exactly are the Web Sustainability Guidelines and who are they for? If you cared about sustainability in the digital realm, why would you try getting involved in a group like the W3C in the first place? How does this even change the way people build the web and consume web services? And how do we make them more sustainable? Joining us today are two key figures from the W3C to help us explore these themes. Alex Dawson, one of the editors of the Web Sustainability Guidelines at the W3C, and Tzviya Siegman, at the W3C's new sustainability lead. So in alphabetical order, Alex, you win here. So I'm just going to hand over to you to briefly introduce yourself before we hand over to Tzviya. Is that okay?

Alexander Dawson: Yep, that's fine. Hi, everyone. I'm Alex Dawson, and I'm an invited expert in the W3C Sustainable Web Interest Group. And as Chris has mentioned, I'm the editor of the Web Sustainability Guidelines. And I've also recently been involved in co-authoring the upcoming sustainability chapter in this year's Web Almanac.

Chris Adams: Cool. Thanks for that, Alex. Tzviya, can I give you the floor to do the same?

Cool.

Tzviya Siegman: Sure. Thank you for having me. I'm Tzviya Siegman. I'm the sustainability leader at W3C. I recently joined the staff of W3C, but I've been involved here more than 10 years. I also do member engagement in North America. I recently been involved in setting up the interest group in W3C and getting used to being on the staff instead of being a member.

Chris Adams: Thanks, Tzviya. And if you're new to this podcast, my name is Chris Adams. I am the host of this podcast, but I'm also the executive director at the Green Web Foundation. I also work Inside the Green Software Foundation as one of the chairs of the policy working group as well. And I'm one of the maintainers of the library CO2.js, a tool that's used in a number of digital sustainability projects. Before we dive into this conversation, here's a quick reminder that we're going to share a transcript to this and links to all the projects that are just mentioned in our show notes. And, you'll need to look for that on podcast.greensoftware.foundation to find that. So if you're listening to this via Spotify or YouTube, you might not find them. So please remember to look that up there. Alright, so, with that, I think the two of you are sitting comfortably, right? Alex, maybe I'll just hand over to you. So, last time, around about this time last year, I believe you joined us with Anne Faubry, and you spoke to us about the second draft of the Web Sustainability Guidelines. And most recently, I saw you presenting at the Green IO London conference about, again, web sustainability, and specifically the process of making these guidelines for people to follow. So, as I understand it, you're now at, is it draft 8, or possibly draft 9? So, I'd like to ask a little bit about how they get built, but before we do that, could you just give a brief summary of what the Web Sustainability Guidelines are to the uninitiated?

Alexander Dawson: Yeah, absolutely. Essentially, the Web Sustainability Guidelines are an evidence-based set of advisory guidelines. Many people may be aware of the Web Content Accessibility Guidelines as our founding principles of accessibility. And we've kind of followed a similar set of principles, but designing and creating them for the sustainability community.

And they've been in development for about three years now. And obviously, They are to assist those who want to make digital products and services more sustainable through practical techniques. And obviously, the science is developing rapidly as is the sustainability field and of course, legislative compliance targets.

So it's a much needed feature that the web really requires. And, it's a sort of Global growing focus as climate change is becoming a ever more important topic of discussion.

Chris Adams: Cool, thanks. Alright, there's two forlorn questions I might ask. So, first of all, you spoke about the Web Sustainability Guidelines. That's quite a mouthful. Is it okay to call them like the WSGs? Is there a short term that we should use for that?

Alexander Dawson: Yep, WSGs is perfectly fine. We use WSG as our sort acronym.

Chris Adams: Got it, okay. So I use WSG, a bit like the kind of web content accessibility guidelines, or sometimes the WCAG, right? I'll call it. So WCAG and WSG. Got it. Okay. And also, I know that you're at Green.io, the conference, it's like one of the places to be going to if you're going to speak about digital sustainability. How was that? How did you find it?

Alexander Dawson: Yeah, it was amazing, really. I mean, you were there yourself. It was definitely the place to be. It was sort of like the Woodstock of sustainability, pretty much. Anyone and everyone who could be there managed to sort of cram in the place, standing remotely at the back of the conference room. There were sort of all the faces from all these sort of major players that could sort of get there.

Yeah, it was a pretty amazing event. all credit to sort of Gail and getting sort of everybody involved in the conference. It was pretty much the event of the year as far as digital sustainability goes.

Chris Adams: So, you enjoyed that.

Alexander Dawson: Definitely, yeah.

Chris Adams: All right, okay, and Tzviya, if I hand over to you, as I understand it, you work the W3C, and you're the new sustainability lead, and part of your day job is essentially guiding the existing community group that created some of these guidelines. But there's also work to create a more formal, I think they're called interest groups, which is slightly different. We'll explore that a little bit later on. Before we do that, can I just backtrack to talk about how you ended up working as a sustainability lead in the W3C? Because I had no idea there was one until I heard about your involvement in this. And honestly, I was quite pleased because it's been something inside the W3Cs like, architectural guidelines, their TAG, they said the web should be sustainable.

And now they've actually got someone who's talking about this. So that was really nice to hear. So yeah, how'd you get into this? What's the story that this came about?

Tzviya Siegman: I'll give you a short history of my career. I started working in standards back in 2008. I worked for a publishing company, and I was doing the digital side of things, looking at XML, and all of a sudden, ebooks became the thing. And it was, "oh my goodness, Tzviya, we need somebody to work on this." And we created a hundred based on PDF and CR, and I was like, this doesn't really look right, maybe I can fix it. And I did. In order to fix it, you need to read standards. And the whole ebook program came under my leadership, and I was not very high up in the food chain at that point. Eventually, I started working on the standards with an organization called the International Digital Publishing Forum. Because that's where EPUB was at the time. And over the next six years, the IDPF merged with the W3C. And I became the chair of the publishing working group. And then I joined the W3C's advisory board. And then I became chair of the advisory board. And etc. And then I got laid off. And I worked with some people within the W3C to develop this role because I have become interested in sustainability. All along I've been working in accessibility, and the part of technology that's always really interested me is not necessarily creating faster products. That's, time to market was always intimidatingly working with the product managers and things like that. At my job in publishing, I was always frustrated when project issues like accessibility got thrown to the side because there was a new feature game developed. It's nice to have a new feature, but if it's not accessible, if he doesn't have pride, then we were, yeah. And so sustainability is in that area as well. I used to work on what a manager, a former manager of mine called the illities. Accessibility, sustainability, and now I'm still, this is living the dream for me. So, I was able to sculpt this job with the W3C, which is really a dream for me. And the fact that sustainability is mentioned in the W3C's vision, I'll take a little credit for that, because I was one of the people who wrote that. So, just a little credit, because there were a lot of people involved in writing that. But here we are, and we are really excited to be working on sustainability in an interest group, which you mentioned is not quite the same as the working group, because some of the differences now or later that we're working on a specification would, I can explain the different types of specifications, but I'm really excited to be leading the community group into something more formal than just a new file, even though we had a lot of uptake, where we were able to form a policy and an interest group to be through soon.

Thank you.

Chris Adams: I see, okay, so it sounds like this is not your first rodeo. You've been through the process of seeing how standards get developed, and this is why you being involved with this helps, because there may be a bunch of people who have, who might be curious about this and have all the things you kind of wish for, but you've seen how this works, you've seen how the sausage gets made, and that's, some of your role is to help steer some of that through so we end up with something that people actually can use and adopt in their work and things like that, then.

Tzviya Siegman: Yeah, absolutely. The work in W3C, when it's an interest group or a working group, it goes through a pretty rigorous process, and I can talk about that process in more detail in the literature.

Chris Adams: Thanks! Alright, okay, thanks for that, Tzviya. So hopefully we've caught up the audience to where we are today, November 2024. We spoke a little bit about the history. And I teased about the creation of this new group, like an interest group that has some more formal backing within the W3C. So for people who are not au fait with how web standards are formed, or what an interest group is compared to a community group, can you maybe just explain a little bit about why this is a big deal and what it makes possible that wasn't before?

Tzviya, I'll hand over to you, then maybe Alex will come in with some specific examples, because he was the author of the guidelines.

Tzviya Siegman: Yeah, thank you. I'll talk a little bit about what it is that we try to accomplish with standards and what that means in W3C. In W3C, standards require interoperable implementations. Interoperable is achieved by testable implementations where we test things to make sure that they work in at least two instances. Sometimes standards are codified by regulatory bodies. We see this with something like USBC in the EU. We also see this with WCAG, the Web Content Accessibility Guidelines being pointed to by various legislation. There are a lot of standards. Like WCAG, where this is pointed to, and the W3C, we have community groups, so anybody can start a community group, it doesn't have the formal structure of a W3C group, it's people get together, they work on things, and it's the incubation process, we call it, or just people get together because they need

Chris Adams: Okay.

Tzviya Siegman: An interest group is, has the authorization of the W3C's advisory committee. Everybody has to vote on the work. A member of each W3C member organization has the ability to vote yes or no or maybe. Yes, but I would like to see this then changed on the formation of an interest group. And an interest group can write statements. A working group is when the same came with the advisory committee, but they actually have the recommendation track, which is what W3C officially calls standards. The sustainability group is an interest group where they write in statements. Because the Web Sustainability Guidelines include a lot of things that are not necessarily testable. Although there are a lot of testable statements in there, and I'll turn it over to Alex to talk about how we'll proceed with that.

Chris Adams: Okay, thanks for that, Tzviya. So if I just paraphrase that for people, there's almost like an escalating layer of authority or formality that comes through this. And it may be that the bar is somewhat higher with the things that get published by an interest group compared to a community group. And like there's an escalating scale all the way up to having like full blown standards.

And this is like the midway point. And, like, Alex, if that's about right, I'll hand over to you, Alex, and maybe you can chime in on some of it, because, yeah, you are very hands on when writing some of these, perhaps.

Alexander Dawson: Yeah, obviously, we are trying to be as rigorous as possible because throughout the creation process, we have tried to, even at the point of being a community group, treated this as seriously as possible and made sure what we've created to be as evidence led as possible. And we are still in the process of trying to create as much measurability data as we can, which is something that we are going to be increasingly focused upon as an interest group.

So, with regards to testability, which is something that obviously interest groups, having testable implementations, it's not something that we have to be as strict upon, we still want to make sure our work is as rigorous and testable as possible. Because there may come a point in the future that we may want to have that testable compliance state for working what we have created in with legislative conformance and being able to do so may require at some point, obviously in the far future, taking our work from an interest group state to a working group state to get that recommendation status.

But that is something, in the far off distant future, as Veer will probably nod and agree with. But for the moment, obviously, we want to focus on making what we have created as rigorous as possible purely because anyone who is going to utilize our work, we want to have the best quality evidence for anything we are submitting as guidance.

We want to be able to. tell people using our work, this is the best way that you can implement something in your product or service because the evidence shows that this will have the greatest impact in terms of sustainability. And it's something that we've been very much focused upon when creating the Web Sustainability Guidelines from the offset, which is why we've got so many references within our work.

We've got quite a lot of material relating to things like GRI and such. So it's something that we've been very much focused upon from the offset.

Chris Adams: Okay, and by GRI, you're referring to the Global Reporting Initiative here, right?

So that's one of the other groups that talk about, okay, this is how you, set out what counts as sustainable, for example, and they've done a bunch of work that you're referring to with that one there.

Alexander Dawson: Yeah. It's one of the measures that we've been utilizing alongside obviously academic material and many other forms of tooling as well.

Chris Adams: Alright, thanks for that, Alex. So, I'd just like maybe touch on that a little bit more, because one thing that I know that we've seen, like, in France, for example, there are something like guidelines which are kind of mandated for government websites now, and it sounds like you might be looking for something like that on a more international scale that could be adopted in other parts of the world, and I mean, for folks who are new to this, the W3C has, like, a public mailing list, and if you look on there, you'll see, like, there are people from, say, I think, Singapore. The government of Singapore basically said, "hey, we're looking for some standards like this, or something for us to adopt," and I know there are people in other parts of the world who are basically saying, "I want to push for this, but I need some kind of something either like a standard or a set of guidelines I can ask my team to be following," and it sounds like that's the kind of push, because I know the WSGs have been in some kind of usable form now, but it feels like this seems to be a real push for adoption to get larger organizations to use them, who commissioned significant projects. And we spoke about how these might not be standards with a capital S, there might be a kind of guidance for sustainable software, but in terms of providing a reasonable expectation of what you can expect a responsible practitioner to do, that seems to be what they'd be shooting for.

So maybe we could talk a little bit about what adoption looks like here, and maybe we can hear some stories about organizations expressing interest in adopting these and what that process looks like, because, yeah, this is a new field, and I know that we've just spoken about how different places are looking for this. And Alex, I suspect you've been in some of these conversations, so maybe if I hand over to you, and then maybe let Tzviya chime in, because she's seen this happen a few times outside of just sustainability.

Alexander Dawson: Yeah, in terms of adoption, meeting the success criteria is rather open to interpretation at this point. I mean, we're, as I've mentioned, we're aiming to improve measurability data as an interest group, so we can better sort of lead by evidence. Regarding if a particular technique has a greater impact or not and that's sort of on the horizon.

Currently we sort of follow a similar practice to the Web Content Accessibility Guidelines with their accessibility techniques and much of that can be seen with, in our STAR supplement, Sustainable Tooling and Reporting, as that outlines optional techniques that are currently deemed, where appropriate, machine testable, so can be sort of repeated and used to showcase which success criteria can be repeatable and, implementable to meet certain success criteria within the particular guidelines.

Now, obviously, you don't have to follow them, they're not sort of gospel, and we're working to sort of improve them where we can along the pathway. But we've been sort of trying to improve upon this to sort of help people in terms of general adoption. And regarding sort of interest from the public and private sector, we've had quite a lot of interest definitely from various organizations.

I mean, I have been sort of trying to keep track wherever possible, and there have been a number of agencies globally where they mentioned that they have been starting to utilize and implement our work and there have been a lot of government departments who have got in touch who are openly interested in utilizing what we have created, even though it's in draft format, or have interest in actually for tracking our progress or getting on board with what we've created potentially for what could be future compliance, so it's good signs and sort of generally good feeling, if nothing else, that we're potentially on the right track with what we have been creating.

Chris Adams: Okay, all right, so if I understand that correctly, so I spoke about how France has, they've set up some standards themselves already. So like, "all our websites made by the public sector have to meet these standards." In other parts of the world, there might be people saying, "well, we're going to set some of this, and rather than write all this stuff ourselves, we'll refer to these things as like, we'll defer to that, as that's what we're going to hold as our bar that everything needs to meet," for example.

That's what some of this stuff is. It seems to be like the direction that's heading in. Okay, so that gives you some idea of how that might work. But, Tzviya, I am, I'll hand over to you because, obviously, it's a bit more complicated than that. And like, there's, we can ask for certain things, but that's not the same as people actually adopting this stuff for example. Like, if it was, well, we would see a probably more accessible web, for example. So, Tzviya, can I hand over to you to talk a little bit about, like, actually adoption and how this happens, or if there's any things that you've learned over the last few years that you might share that could help here?

Tzviya Siegman: So, adoption is a really complicated topic. Something that I think many people find shocking when they start looking at standards is that we can write standards in an incredibly idealistic way. But we can't actually make anybody use them. It's very frustrating. This is why the implementation tests were important. But even with testing, we can't promise adoption. And that's why it's important for everybody working on this to become evangelists. And I often say that my job is really evangelism, and a lot of people who have worked at, who are working on sustainability come from the world of accessibility because the role is very similar in terms of evangelism. And it's really important to make sure that the set of issues is a high priority. People who are working on browsers, servers, and in their organizations, like you might have an HR department who has a goal of, let's say, reducing carbon emissions, because that's what everybody says, because as Tim Frick likes to say, they have carbon tunnel vision. They'll say they're reducing carbon by 30 percent by 2030, so just pick a number. They don't know what that means. We need to get to the developers who are actually writing the code, and we need to get to the people who are building the servers, and so on. So that They can actually do the work because the people at HR or the CEO or whoever it is who's making these commitments don't have the tools to do this. Our guidelines give us some of the tools and the guidelines are divided into sections so that this touches a lot of different departments in an organization, and we're trying to break them down so that people can pull out the different sections that are relevant to them, but that adoption step is really important. I'll also say that what you were talking about, Chris, where different governments might be able to point to the guidelines, that's really where adoption is important. But it also comes down to being able to educate people, and we're also going to be working on different education materials. I love WCAG, I'm a big fan, but a lot of people are deeply critical of WCAG because it's so hard to understand. So we're trying to make ours a lot easier to understand, and hopefully we'll be able to be really successful with it.

Chris Adams: Okay, thanks for that, Tzviya. So, you, so there's two things that kind of came out of that. So, one of them was about, Alex, you spoke about these things being testable. And presumably, this is some of the stuff that you might see when people are using, say, web page tests, which has, like, Carbon control, or EcoGrid, or some of these things. If you can test for something, then you can provide some guidance as part of the analysis process, or you can automate this and check, like, has this got better or has it got worse? That's what some of the kind of sustainability things presumably would afford, right? And that's what you're kind of referring to with some of that.

And that might be something that you're essentially able to, like, enforce a standard on, or make sure it's meeting the standard that you've set out in some kind of agreement, right?

Alexander Dawson: Yeah, absolutely. If we can provide something that's testable or something that people can utilize, then it's something that toolmakers, people creating products and services, can implement with what they're creating. And that will obviously help us do our job because for people creating tools like, Website Carbon, Ecograder, they can filter that down into their processes so that rather than individuals having to read an extensive specification, with a few clicks of a button, it can provide feedback through reports, potentially tools that can be integrated within IDEs like Visual Studio Code, things that could be integrated within browsers such as the DevTools environments within, say, Google Chrome and Firefox and such. Lighthouse Project is a prime example.

These are things that if it's machine testable, it can be integrated within tooling and that will help take a lot of the friction out of people being able to quickly identify where the problems are occurring, how much of a problem it is, and how to identify what resolution needs to be,

Chris Adams: kind of remedial action you might take. All right, okay, thanks for that. Okay, so that helps summarize that, and Tzviya, you were talking a little bit about, okay, different groups might have different stakeholders, and you were referring to public sector as one group that might have legally binding targets.

For example, we know that in the UK, they have a target to reduce emissions by however much, and in Europe, it's reduced emissions by 55 percent by the end of, wow, in five and a half years time, for example, I mean, I'm speaking to you in the middle of an election, so there's a difference about, but like, up until yesterday, there was a whole thing about the federal government being the single biggest purchaser of green energy, which was forcing a kind of switch to cleaner energy, for example, things like that.

That seems to be what you're referring to, right?

Tzviya Siegman: Yeah, there's public and private, but there's also different stakeholders, such as product managers versus developers. Individuals within an organization.

Chris Adams: Ah, I see, so like when you've got product managers, if they know what to ask for, they're able to, that helps inform that conversation with the team they're working with and say, "can we please agree to meet these standards or have something like that." They don't need to be the specialists themselves to know that if there's a standard or set of guidelines out there, that at least gives a shared set of minimum standards that the work's going to be run at, basically.

Tzviya Siegman: Yeah, absolutely.

Chris Adams: Okay, cool. That provides a bit more guidance about, I guess, the theory of change behind some of this. So, we're just starting to come up to time, and I wondered if we might be able to look forward to, like, what's on the horizon in the world of sustainable web development, and if there's anything that you folks have, that maybe, if I was to ask you, What should be on an interested listener's radar if they were interested in web sustainability? If I hand over to you, Alex, first, and then over to you, Tzviya, is there anything, like, what should people be looking out for between now and the end of the year, for example, or perhaps in Q1 2025? What's on your list, Alex?

Alexander Dawson: I think the thing that's dominating everybody's discussion at the moment is just sort of AI, isn't it, really? Everyone's talking about it, it's causing a lot of discussions around sort of how big an impact it's going to have, and it's pretty much going to be the dominating discussion around, how we're going to sort of wrangle around it in terms of legislation and measurements and tolling and Optimization and such, so I think that's going to be a major sort of domineering factor, especially considering how much infrastructure it's currently absorbing in terms of building new facilities and such, so yeah.

Chris Adams: Okay, cool. And are there any particular projects or things that you're expecting to, any releases by projects you follow, or any big things like that you would direct people to? So, they've been listening to this and they might want to find like a follow on link or follow on project to look into. What would you direct people to if they found something that's interesting but wouldn't want to go somewhere else next?

Alexander Dawson: There's not a particular one especially that I've been following. I mean, if I find any useful links or materials, normally it gets sucked straight into our project into the Web Sustainability Guidelines because we have nicely organized references for any useful material that can help people make decisions and such.

So I always like to point people to the guidelines, to the latest release, because I, nicely guard in those links and keep them up to date with all the latest material relating to sustainability. We've got sort of about 1500 odd links, all nicely categorized for people interested in sustainability.

Yeah, it's probably the largest digital sustainability reference library on the web.

Chris Adams: Blimey. Okay. All right. And, Tzviya, is there anything that you might kind of direct people's attention to? Like, we've spoken all about interest groups and community groups, but like, they still run, right? Like, they happen in the, is it monthly? I mean, how, if someone wants to find out about this, what's the process like to start

getting involved in any of this stuff, for example?

Tzviya Siegman: I'll provide links for both the interest group and the community group in W3C. For the community group, you don't need to be a member. For the interest group, you need to be a member of W3C or be what's called an invitative expert, which means provide a reason that you should be invited and, like, get a review by me and the chairs of the interest group and then show that you'll do the work. Also, something I'll be working on in the next year or so is creating a horizontal review program for the W3C and sustainabilities. The W3C prides itself on doing horizontal review on all standards in the areas of accessibility, privacy, security, and internationalization, meaning, we'll say the example of accessibility, every, there's a review of every single specification to make sure that the specification itself is accessible and will introduce accessibility problems. So I have the job of creating this for sustainability.

Chris Adams: Sorry to interrupt, can I just check, because I'm not sure I'm familiar with the horizontal, is the idea there being that you've got three or four standards, and you want them to be consistent with each other? Is that what horizontal reviews, or is there something else?

Tzviya Siegman: Horizontal review means that, so, let's say you have a new CSS specification. There are dozens of those every year. There's an accessibility checklist for the people in the CSS working group to take a look at. And then we have the opportunity to speak with people who are experts in accessibility to make

Chris Adams: okay.

Tzviya Siegman: accessibility issues onto the web. So I will be working to create a sustainability horizontal review checklist. I don't know yet how I'll be doing this, but that's part of my job as sustainability lead. I'm very excited about this because this has not been done before. So we want to make sure that the W3C is not bringing new sustainability issues to the web.

Chris Adams: Oh, I see. Okay. So if I check if I understand that, and maybe this is one concrete example, I know that there are certain bits of CSS, like there are certain filters that if you run them, you can basically cook that CPU in your laptop. I think Florian at Mozilla, he's one of the people who built the Firefox profiler.

He has this kind of party trick where he basically takes you to a single web page and it triggers one thing which just cooks your machine, because it just runs all the cores really, really hard, basically, just because it's doing a kind of gentle group. It sounds like something a bit like that, like, if someone's going to introduce a feature, they're aware of some of the knock on impacts, or they might have, yeah, okay, I think I understand that now.

Tzviya Siegman: And I'm going to just remind Alex that he's working on the web almanac.

Chris Adams: Yeah,

Tzviya Siegman: And maybe you want to plug that.

Alexander Dawson: Yeah, absolutely. We've got the new edition of the Web Almanac, which will be coming out this month. I've contributed to the sustainability chapter, and anyone who's interested in sustainability will be wanting to check out the results of that. So, yeah.

Chris Adams: Cool, and for people who are not familiar with the Web Almanac, this is essentially a bit like a kind of state of the web. Every year it comes out, tells everyone what's going on. So you've been saying, essentially it's state of sustainability of the web, that's what your chapter's focusing on, right?

Alexander Dawson: Yeah, absolutely.

Chris Adams: Brilliant, well if that's visible, we'll share a link to that in the show notes, but anyone who's curious, Web Almanac seems like a fairly easy term to search in your chosen search engine. Brilliant. Okay, well, I think we've got a good few pointers, and we'll also link to some of the community groups that we spoke about here. I suspect there's a conversation to have about how the GSF should work with the W3C for some of the groups there, because I know there is, we spoke about AI, and there's a GSF, Green Software Foundation Working Group, specifically focused on AI. It may make sense to do something along those lines for web specifically, because it's another clear sector and a clear application of software.

So, just before we go, maybe I should ask, if people want to find out more about you and your work, where should we, where should people be looking? Tzviya, if I start with you, like, where should we go? Is there a LinkedIn page or something like that you'd direct people to?

Tzviya Siegman: Sure, I'll drop my LinkedIn page, it's linkedin.com/Tzviya, I think. And there's also my bio on the W3C staff page, which I don't remember the URL off the top of my head, but I'll find out.

Chris Adams: Okay, cool. And Alex, for you as well?

Alexander Dawson: Mine's pretty straightforward. You can find my details on my website, and my name is my URL, so it's alexanderdawson.com.

Chris Adams: Alright, that seems nice and easy to Google and look up, and Alexander and Dawson are spelt like you would expect them to be spelt as well.

Alexander Dawson: Absolutely.

Chris Adams: Brilliant. Well, folks, thank you so much for giving us a bit of a kind of deep dive into the world of the W3C, and how web standards get made, and how we can actually, how they can be adopted, and wish you best of luck.

Hopefully I'll see you at one of the other community groups, or the meetings every month, and yeah, hope you have a nice week, alright? Take care, folks. Ta ra.

Tzviya Siegman: Thank you.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundationon. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

Backstage: Awesome Green Software

7 november 2024 | 9 min

In this episode of Backstage, we go backstage with the Awesome Green Software (AGS) project, a pioneering initiative from the Green Software Foundation’s Open Source Working Group. Project leads Ahmad Antar and Abid Ali, along with contributor Josh Rauvola, share how AGS is building a centralized platform for developers to access tools, libraries, and frameworks to create greener, more sustainable software. They discuss the innovative tech behind AGS, the project’s progress toward launch, and its potential to make green software development accessible to organizations worldwide.

Learn more about the Awesome Green Software Project:

Learn more about our people:

Ahmad Antar: LinkedIn
Abid Ali: LinkedIn
Josh Rauvola: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Algolia [4:03]
Notion [4:04]
Scaphandre [5:07]
CodeCarbon.io [5:15]

Events:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Chris Skipper: Welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Skipper. On this episode, we have another edition of Backstage. Here, we'll take you behind the scenes at the Green Software Foundation to highlight ongoing projects and share insights from the members directly involved.

Today, we're diving into the Awesome Green Software Project, an exciting initiative curated by the Green Software Foundation's open source working group. Today, project leads Ahmad Antar and Abid Ali have provided us with some informative content and along with some soundbites from contributor Josh Rauvola to give us the lowdown on the collection of tools, libraries, and frameworks designed to help developers create greener, more sustainable software.

So, what makes this amazing, nay awesome, collection of green software resources tick? Let's explore that in this backstage episode. The GSF is all about building software that emits less carbon. Awesome Green Software fulfills that by creating a one stop shop which compiles research, tools, code, libraries, and training from volunteers that will be used for building applications that emit less carbon into our atmosphere.

There aren't too many tools available today that organizations can use to ensure their software is following sustainability principles. But, Awesome Green Software aims to fill that void. The GSF has the Impact Framework and the Software Carbon Intensity Specification, the SCI. And now, Awesome Green Software to provide that support.

As previously mentioned, Awesome Green Software brings research and resources that will be available at one place for organizations to access, hence reducing their efforts and time to find these tools and rather focus their efforts on building impactful software. Awesome Green Software was an idea that the GSF wanted to start for a while.

GSF member Digital Emissions took the initiative and is bringing it to reality. They have made a lot of progress over the last few months. The front end and back end are almost ready, and they are getting ready to start testing soon. Awesome Green Software is still in the developmental phases, but once it's ready to launch, We are expecting many contributors from the GSF and beyond to start contributing impactful code, tools, and utilities.

It will have a great impact on the overall community because currently it is not easy to find such tools in one place. Next, I'd like to play you a soundbite from Josh. We asked him the question, what are some of the key challenges you've encountered while developing the Awesome Green Software project?

And how did you and your team work to overcome them?

Josh Rauvola: Well, one of the main challenges we've faced is consolidating a vast amount of information and resources into a single user friendly platform. There's so much valuable content out there, but it's scattered across various sources. Organizing it in a way that's easily accessible to developers and other people was a significant task.

We also had to ensure that integrating technologies like Notion for content management and Algolia for search, which we can get into later, would provide a seamless user experience. To overcome these challenges, our team focused on building a solid backend and frontend infrastructure. We've been meticulous in our approach, and we're excited to get this project out into the hands of people that truly care about the environment.

Chris Skipper: With such a comprehensive list of tools and resources, the Awesome Green Software Project provides a bird's eye view of everything related to sustainable software out there on the internet. Can you explain some of the methods that you used to compile such a comprehensive list?

Josh Rauvola: Certainly. One innovative approach we're taking is leveraging powerful AI search capabilities through Algolia. This allows users to quickly find the tools and resources they need to build sustainable software. We're also using Notion as a content management system, which simplifies the process, for contributors to add and update resources in the directory.

By combining these technologies with modern front end frameworks, we're creating a responsive and efficient platform that not only compiles existing resources, but also encourages community engagement and contributions. We're all about being as efficient as possible, so we utilized some of the tools provided on our platform to align better with our awesome green software goals.

Chris Skipper: Could you highlight some of the specific tools, technologies, or frameworks included in the Awesome Green Software Project? What are some of your favorites? And how do they support the project's sustainability goals?

Josh Rauvola: That's a great question. The project includes a variety of tools like the Carbon Aware SDK, which helps developers make their applications carbon aware by optimizing energy consumption based on grid carbon intensity. We also feature tools like Scaphandre, a power monitoring agent that helps measure the energy consumption of servers.

Personally, I'm a big fan of the CodeCarbon.io library. It allows developers to track and reduce CO2 emissions from their computing by integrating seamlessly within their code. These tools support our sustainability goals by providing practical solutions that developers can implement to reduce the carbon footprint of their software.

Chris Skipper: The final question we put to Josh was, throughout the development of the Awesome Green Software project, What valuable lessons or insights have you gained that could benefit other organizations aiming to promote sustainable software?

Josh Rauvola: Well, one of the most valuable lessons we've learned is the importance of accessibility and community engagement. Only by the amazing support of the Green Software Foundation and the team at Digital Emissions have we been able to create such a wonderful platform. Creating a centralized platform where resources are shared.

Green software, easily accessible, significantly lowers the barrier for organizations to adopt sustainable practices. By making it easier for developers to find and use these tools, we encourage wider adoption of green software practices. Now all we need to do is get the word out. I'm sure this podcast will go a long way in doing so.

Chris Skipper: Yes, hopefully it does. I'd like to extend a big thank you to Ahmad, Abid, and Josh for sharing their valuable insights on the Awesome Green Software project. It's evident that this initiative goes beyond just providing resources. Thank you. It's about creating a comprehensive platform where developers can easily access tools, libraries, and frameworks to build greener, more sustainable software.

From their use of innovative technologies like Algolia and Notion to the thoughtful selection of carbon tracking tools, Ahmad, Abid, and Josh have shown us how AGS will make a real difference for organizations like yours. Aiming to reduce their software's carbon footprint. Their dedication to making the Awesome Green Software project a one stop shop is inspiring.

And it's exciting to hear about the milestones reached so far. With the front and back end nearly ready, and testing just around the corner, the team's focus on accessibility and community engagement will undoubtedly drive broader adoption of sustainable software practices as more contributors join after the launch.

We can't wait to see how AGS will shape the green software landscape and empower developers worldwide. slide. Now, before we finish this episode of Backstage, we have a few events to read out. Firstly, a free in person event, Green Gaming with Strategies, on November the 9th at 9am CET in Katowice, Poland.

Learn how the gaming industry contributes to and can mitigate its environmental footprint. Next up, we have Implications of AI related data center electricity use and emissions. A hybrid event happening on November the 12th to the 13th in Washington DC. Explore how to map, measure and mitigate the impacts of AI data center electricity usage.

Next up, a virtual event titled Digital Impact Pitch on November the 20th at 6. 30pm CET. This virtual event will help you discover what is the impact of our daily use of digital technologies. And finally, the Green Waves Hackathon closing day on November the 22nd at 8. 30am CET at TU Delft in the Netherlands.

Network, get talks and celebrate the ideas created during the hackathon. This event is your opportunity to experience the most innovative solutions developed by the Green Waves Hackathon participants. Whether you're passionate about sustainable IT or water research, This closing day will showcase the cutting edge ideas that can help reduce environmental impact.

Join us for a day packed with insightful presentations, live demos, and the excitement of the awards ceremony. Links to all the events and the resources mentioned in the show will be in the show notes below this podcast episode. So, we've reached the end of this special backstage episode. I hope you enjoyed the podcast.

To listen to more podcasts about Green Software, please visit podcast. greensoftware.foundation and we'll see you on the next episode. Bye for now!

Remembering Abhishek Gupta: How does AI and ML Impact Climate Change?

17 oktober 2024 | 43 min

In this episode, we honor the memory of Abhishek Gupta, who was an instrumental figure in the Green Software Foundation and a Co-Chair of the Standards Working Group. Abhishek's work was pivotal in the development of the Software Carbon Intensity (SCI) Specification, now adopted globally. His tireless efforts shaped the future of green software, leaving an indelible mark on the industry. As we remember Abhishek, we reflect on his legacy of sustainability, leadership, and friendship, celebrating the remarkable impact he had on both his colleagues and the world. We are airing an old episode that featured Abhishek Gupta, Episode 5 of Environment Variables. Where host Chris Adams is joined by Will Buchanan of Azure ML (Microsoft), Abhishek Gupta; the chair of the Standards Working Group for the Green Software Foundation and Lynn Kaack, assistant professor at the Hertie School in Berlin to discuss how artificial intelligence and machine learning impact climate change. They discuss boundaries, Jevons paradox, the EU AI Act, inferencing and supplying us with a plethora of materials regarding ML and AI and the climate!

In Memoriam: Abhishek Gupta:

Honoring Abhishek Gupta | GSF

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Will Buchanan: LinkedIn
Abhishek Gupta: LinkedIn
Lynn Kaack: LinkedIn / Latest Paper

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Episode resources:

ClimateAction.tech [3:44]
Green Web Foundation [3:49]
Green Software Foundation’s Standards and Innovation Working Group [4:14]
Montreal AI Ethics Institute [4:43]
Hertie School Berlin [5:50]
Aligning Artificial Intelligence with Climate Change Mitigation [6:32]
The IPCC [7:11]
Paper: Green AI | Roy Schwartz, Emma Strubell, Jesse Dodge [8:37]
Project: Pachama [9:33]
Montreal Institute for Learning Algorithms [10:34]
Project: This Climate Does Not Exist [10:48]
Austrian Institute of Technology | Infrared [11:32]
Jevons Paradox [20:19]
The GHG Protocol [23:27]
Legislation: The EU AI Act [25:08]
Paper; Measuring the Carbon Intensity of AI in Cloud Instances | Will Buchanan et al. [30:08]
ONNX Runtime [37:02]
TinyML [37:09]
GitHub: Dynamic Batch Inferencing - Taylor Prewitt & Ji Hoon Kang of UW
GitHub: NVIDIA Triton server on AzureML & Model Analyzer
Green Software Foundation Summit

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Chris Skipper: This week on Environment Variables, we have some sad news.

We have to report the untimely passing of Abhishek Gupta. Abhishek was a key part of the Green Software Foundation. He was the co chair of the Standards Working Group from the GSF's formation to his passing on September 30th of this year. I would like to read out an in memoriam that was posted by Asim Hussain, honoring the great man, his legacy of leadership, his collaboration, and the heart that he put into the Green Software Foundation.

When the Green Software Foundation was formed over three years ago, Abhishek offered to help. The standards working group was looking for a co chair, and he jumped in to help without hesitation. He led the standards working group for over three years, with the full support of everyone involved. Leading a work group isn't about imposing your will on others, It's about finding common ground, nurturing discussions, and searching for the truth.

Abhishek excelled in all of those areas. He leaves a powerful legacy with the Software Carbon Intensity SCI specification. His tireless efforts over the years led to a consensus on the specification, which was later published to ISO in late 2023, and is now being adopted globally.

The impact of the SCI is truly global, with academics and enterprises worldwide adopting it. This widespread adoption is a testament to Abhishek's vision and dedication and his influence will be felt in every software product that bears an SCI score. His legacy is not just a memory, but a living testament to his work.

He has left an unforgettable mark on the community. And will be remembered for his contributions for years to come. Aho brother, I'll see you on the other side, Asim Hussain. To honor Abhishek, we're going to revisit an older episode of Environment Variables from the very start of the podcast, episode five, How Does AI and ML Impact Climate Change?

In this episode, host Chris Adams is joined by Abhishek Gupta, Lynne Kaack, and Will Buchanan to discuss these topics. So, without further ado, here's the episode.

Abhishek Gupta: We're not just doing all of this accounting to produce reports and to, you know, spill ink, but it's to concretely drive change in behavior. And this was coming from folks who are a part of the standards working group, including Will and myself, who are practitioners who are itching to get something that helps us change our behavior, change our team's behaviors when it comes to building greener software.

Asim Hussain: Hello and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode we discuss the latest news and events surrounding green software. On our show you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host Asim Hussain.

Chris Adams: Hello there and welcome to the Environment Variables podcast, the podcast about green software. I'm Chris Adams filling in for Asim Hussain, the regular host while he's on paternity leave with a brand new baby. I met Asim on climateaction.tech, an online community for climate aware techies. And I work for the Green Web Foundation where we work towards a fossil free internet by 2030, as well as working as the co chair for the Green Software Foundation's policy group. Today, we're talking about climate change, AI, and green software. And I'm joined by Lynn, Will, and Abhishek.

Will Buchanan: Thanks for having me. My name is Will. I'm a product manager on the Azure Machine Learning team. also a member of the Green Software Foundation's Standards and Innovation Working Group. Within Microsoft, I foster the Green AI community, which now has a few hundred members, and I'm also a climate activist that's focused on pragmatic solutions to complex environmental issues.

Recently, I shipped energy consumption metrics within Azure Machine Learning, and we are about to publish a paper titled Measuring Carbon Intensity of AI in Cloud Instances, which I think we'll touch on today.

Abhishek Gupta: Well, thanks for having me. I'm Abhishek Gupta. I'm the founder and principal researcher at the Montreal AI Ethics Institute. I also work as a senior responsible AI leader and expert at the Boston Consulting Group, BCG, and I serve as the chair for the Standards Working Group at the Green Software Foundation.

So I've got a few hats on there. Most of my work, as it relates to what we're going to talk about today, It runs at the intersection of responsible AI and green software. In particular, what's of interest to me is looking at how the intersections of social responsibility, the environmental impacts of software systems, in particular AI systems, can be thought about when we're looking to make a positive impact on the world while using technology in a responsible fashion. I also, as a part of the Green Software Foundation, help through the standards working group, come up with the software carbon intensity specification, where we're trying to create an actionable way for developers and consumers of software systems to better assess and mitigate the environmental impacts of their work.

Chris Adams: Okay. And Lynn, last but not least, joining us from Berlin. Thank you very much for joining us.

Lynn Kaack: Yeah thank you so much. I am an assistant professor at a policy school, public policy school, called Hertie School in Berlin. And I am also a co founder and a chair of an organization called Climate Change AI. with Climate Change AI, we facilitate work at the intersection of machine learning and And different kinds of climate domains, focusing on climate change mitigation and adaptation. And in my work, in my research, I am looking at how we can use machine learning as a tool to address different problems related to energy and climate policy. I'm also interested in the policy of AI and climate change. And today, actually, since we're talking about papers, I have a paper coming up. That is called Aligning Artificial Intelligence with Climate Change Mitigation, where we look at the different impacts from machine learning and how they affect greenhouse gas emissions.

Chris Adams: So we actually have some decent deep domain expertise and I'll try to keep this quite accessible, but we might drop into a little bits of like data science nerder on here, but the podcast has done that previously and it turns out to be something that we've got some decent feedback from because there aren't that many podcasts covering this. Okay so let's, uh, get into this topic of green AI and climate change. As we we know, IT is a significant driver of emissions in its own right. When we think about the climate crisis this year, the IPCC, which is the Intergovernmental Panel on climate Change in their big reports, which synthesized literally thousands of papers explicitly called out digital as a thing we, we should be talking about and thinking about.

And if you're a responsible technologist, it seems like a thing that we should be taking into account here. Now, I found it helpful to think about IT IT a little bit like how we think about the shipping industry, partly because they're similar in terms of emissions, which is around between 1 and 3%, depending on what you look at it, but also in that both of these acts like kind of connective tissue for society.

We also think of IT as a kind of force multiplier for The existing forms of activity. So if you use it, which is in line with the recommendations of the science, that's a good thing. But if you use it to do something, which is kind of rejecting some of the science, it might not be such a good thing. And within technology, AI and machine learning in particular is one of the fastest growing sectors and often seen as one of the biggest levers of all. So we're going to highlight some interesting projects we'll start off with. And out of that, we'll probably dive into some specifics about that or some other things you might want to take into account. If you're a technologist wanting to incorporate an awareness of climate into how you work and build greener software, then finally, we'll hopefully leave you with some actionable tips and techniques or projects that you may contribute to or use in your daily practice.

There's another term that we might be touching on here when you're making AI greener, and that's specifically Green AI. Is that the case, Will?

Will Buchanan: Correct. And that actually was coined by researchers a few years ago, Roy Schwartz, Emma Struble, Jesse Dodge. And it's really focused on making the development of the AI system itself more sustainable. And it's to be disambiguated from the term using AI for sustainability.

Chris Adams: Okay, so that's something we'll touch on both today. We'll talk about some of the external impacts and some of the internal impacts. We're going to start with something quite easy first because, well, why not? I'm just going to ask each of the people here to kind of point to maybe one project that they've seen that's using ML in quite an interesting fashion to ideally come up with some kind of measurable win. Well, if there was one project you'd actually look to that you think is kind of embodying these ideas of like green AI something which is really helping us essentially face some of the challenges, maybe you could tell us about what's catching your eye at the moment or what you'd look at.

Will Buchanan: I've been thinking a lot about natural capital recently, and I learned about a startup called Pachama, which combines remote sensing data with machine learning to help measure and monitor the carbon stored in a forest. I think it's really, really valuable because they're providing verification and insurance of carbon credits at scale. They've protected about a million hectares of forest. That's really when you have IoT and remote sensing and machine learning combining to help nature restore itself.

Chris Adams: Okay, cool. So if I understand that, they're using satellites to basically track forests and track deforestation. Is that the idea that they're doing there?

Will Buchanan: Yes, Yes, and and also to verify the amount of carbon that a forest can sequester.

Chris Adams: Okay, cool. All right. I know there's a few other projects related to this. If I just hand over to Abhishek, can you let us know what's caught your eyes recently, and then we'll see what other projects come out of this.

Abhishek Gupta: Yeah, absolutely. I think one of the projects, and I don't know, I mean, what the measurable impact has been so far. In fact, it's, it's something that's come out of MILA, which is, or, you know, called the Montreal Institute for Learning Algorithms, which is Dr. Bengio's lab in, in Montreal. In fact, one of the people who led that project is a part of Climate Change AI as well, who I'm sure Lynn can talk more about too, which is SASA. And she's done this project called This Climate Does Not Exist, which I think was a fascinating use of machine learning to visualize the impact climate change will have on, you know, places around you in a very arresting and visually capturing fashion, which I think when we think about.

What impact climate change is going to have around us. Sometimes it, it feels quite distant because it's a, it's a slow rolling thing that's coming our way. And this puts it in, in, in a way that's quite immediate, quite visually arresting. And I think stirs people to action. I, as I said, I'm not sure what the measurable impact of that has been yet, but I, I certainly feel that those are the kinds of creative users of AI when we want to galvanize people into action around climate change.

Lynn Kaack: I'm happy to also talk about a application, which is also kind of difficult in terms of measuring impact, but I think it's an another interesting component of what AI can do. And this is something that researchers at the Austrian Institute of Technology do on a project called Infrared. And they use machine learning to help design new districts and cities.

And especially at the moment in many countries, a lot of new urban districts are being built and how we build these has a huge impact on energy consumption in cities, both in terms of transportation, but also how buildings are heated or cooled. And by the use of machine learning, they can drastically improve design choices because now they can approximate their very computationally heavy models and run them much faster. which means that they can also have more runs and can try out more design configurations. So this is a rather indirect application, but it has huge implications also on emissions for, for many decades to come.

Chris Adams: So essentially it's using of housing policy as climate policy there, because there's just a huge amount of emissions built into how people live and whether they need to drive everywhere in a car and stuff like that. That's some of the stuff that it's doing and making that part easier?

Lynn Kaack: So it's not really looking at housing policy, but it's looking at how districts are designed. So they take a group of, of houses, like if the new district is to be built and then they simulate the wind flow going through the cities, which are very expensive simulation models. they take the outputs of their model and approximate it with a machine learning model, which makes it much, much faster.

So from hours or days, you go to milliseconds or below seconds for one run. And then you can try out different design configurations and understand better how The built infrastructure affects natural cooling, for example, in cities or walkability, energy impacts generally of the microclimate on, on the built environment.

Chris Adams: Wow, I had no idea that was, it was actually possible. That's really, really cool.

Will Buchanan: That's very cool. That's similar to generative design.

Chris Adams: design. This is a phrase I haven't heard actually, Will. Maybe you could elucidate or share something there, actually.

Will Buchanan: It's similar similar to to some software that Autodesk has built where you can try out many different iterations of a design and come up with optimum solutions. I think it's really cool that you're just consolidating it and running these models more efficiently.

Chris Adams: cool. And that's a bit like following, say, a fitness function saying, I have a chair, or, you know, I want to have something that works like a chair. It needs four legs and a seating pattern. Then it essentially comes up with some of the designs or iterates through some of the possibilities, something like that?

Will Buchanan: Exactly.

Chris Adams: Oh, wow. Okay. That's cool. All right. And so we've spoken about AI, and there's a few exciting, interesting ones that we can add to the show notes and list from, and for people to look into and see how that might relate to what they do, I suppose. I want to ask a little bit about measuring impact from these projects, because quite a few different ways that you can actually measure impact here.

And in many times, it can be quite a difficult thing to kind of pin down. And this is continually a thing that's come up. When I know that. So, people have tried to come up with specs like the Software Carbon Intensity, and I'm sure Abhishek, you've had some experiences here. Will, you've mentioned a little bit about, um, Actually measuring impact internally, and it sounds like you've just had to do a bunch of this work on the ML team right now and expose some of these numbers to people consuming these services in the first place. Could you talk about some of that part a bit, perhaps?

Will Buchanan: Certainly. And so, as I mentioned, we have shipped energy consumption metrics for both training and inference within Azure machine learning. And that's really complex when you think of the infrastructure required to just report that,but that doesn't necessarily account for the additional power that's consumed in the data center, such as the idle power for devices or for the utilization of your servers.

There's so many different factors there. So you really, you could encounter scope creep when you come to your measurement methodology. So it's really necessary to put boundaries around that.

Chris Adams: and when you use the term boundaries here, you're saying, I'm going to measure the environmental impact of the servers, but not the environmental impact of building the building to put the servers in. Is that the idea of when you're referring to a boundary here? Yeah.

Will Buchanan: Yes, that's a great example.

Chris Adams: Okay. All right. I think this is actually something we've come across quite a few times in other places as well, actually, because maybe it's worth asking about this kind of boundary issue that we have had here, because automatically that sounds complicated here.

And I know that Abhishek, you've had some issues at your end as well with defining this style for deciding what's in or out, because I think this is one thing that we've had to explicitly do for the software carbon intensity spec, right?

Abhishek Gupta: Exactly. And, and I think when we talk about boundaries, it's, it's, it's trying to get a sense for what are the actual pieces that are consumed, right? From an operational standpoint, from an embodied emission standpoint, and how you make those, you know, allocations across you know, what, what your system is consuming.

And I use the word system because I think again, when we talk about software, we're not just talking about a specific piece, but we're talking about really everything that it touches, be that, you know, network, be that bandwidth consumption, be that, you know, as, as Will was saying, idle power. Even when we're looking at cloud computing, it becomes even more complicated when you have your pieces of software that are sharing tenancy across the pieces of hardware and how different consumers are perhaps sharing that piece of hardware with you and thinking about whether you've booked the resource ahead of time or not, whether it's hot or cold in terms of its availability and what implications that has.

I mean, there are so many different facets to it. And each of those decisions, what I want to highlight here is that it comes with a trade off, right? So we also don't have any standards in terms of how we should go about measuring that and what should be included, what shouldn't be included. And so the way people report out these numbers today also doesn't really make it actionable for folks who are consuming or who want to consume these reports, these metrics in taking decisions as to, you know, whether something is green or not.

And I think that's one of the places that The Software Carbon Intensity Specification is trying to help folks is to help standardize it first and foremost, but also to make it actionable so that if you're someone who's environmentally conscious, you can make the right choice by being informed about what the actual impacts are.

Chris Adams: Okay, this is question that I'm curious here, because so far we've only been speaking internally about, okay, what is the environmental impact of IT itself, like it's direct emissions. But the assumption that I have here is that there are ways we might talk about the impact that it has on the outside world in terms of activity we're speeding up or accelerating or supporting there.

Is that the only issue that we need to think about, or are there any other things to take into account about like this system boundary part that we've just been talking

Lynn Kaack: Yeah. So the system effects are really important to look at and to consider. Maybe just to give an example, like if you use machine learning in, let's say the oil and gas sector to make a small part of the operation is more energy efficient, that first sight looks like something that could be considered sustainable and green, but you also have to realize that often then you are reducing costs as well, and that might change the way that oil and gas in this particular example is competitive, or the particular company is competitive, and that actually might shift also how much oil and gas we are able to use in the short run, how the price is changed.

So, these indirect effects can actually then have much larger impacts than the immediate effects of such an application. So drawing boundaries is really important and also opening this up to to have a broader system level view and really try to understand how does the technology also change then, then to larger consumption and production patterns. It's important.

Chris Adams: So if I understand that correctly, that's talking almost like the consequences of an intervention that we might make here. So even though we've might have reduced the emissions of say the drilling part by putting a wind turbine on an oil rig, for example, that might change the economics and make people more likely to use the oil, in which many cases they might burn, for example, or stuff like that.

Is that basically what you're saying there?

Lynn Kaack: Essentially what I'm saying is that efficiency improvements in particular, and often they can be done with data science or with machine learning or AI systems. They often come with cost reductions and then those cost reductions do something and change something. And often this is also considered under rebound effects, but it's not only rebound effects.

So there's systemic, the system level impacts that come from more small scale applications that need to be considered.

Will Buchanan: That's such a good good point, and I think I've also heard it called Jevon's paradox, too.

Chris Adams: Yes. Jevons paradox. This is stuff from like the 1800s with steam engines, right? Like my understanding of the Jevons paradox was when people had steam engines and they made steam engines more efficient, this led to people basically burning more coal because it suddenly became more accessible to more people and you end up using them in a greater number of factories.

So there's a kind of rebound, I think, that we need to take into account. This is something I think has been quite difficult to actually capture with existing ways of tracking the environment impact of particular projects. We have like an idea of, say, an attribution based approach and a consequence based approach. And maybe it's worth actually talking about here, about how Some of the complexities we might need to wrestle with when you're designing a system here. I mean, Abhishek, I think this was one of the things that was an early decision with the software carbon intensity part to not try to have an attribution approach versus a marginal approach. And if we're not diving too deeply into jargon here, maybe you might be able to kind of share a bit more information on that part there, because It sounds like it's worth expanding or explaining to people, to the audience, a bit better here.

Abhishek Gupta: Indeed, you know, the reason for making that choice was again, our emphasis on being action oriented, right? So as we had started to develop the software carbon intensity specification, one of the early debates that we had to wrestle with and, and, you know, Will and Will was of course a crucial part of that as well as whether folks who were a part of the standards working group was figuring out how.

For example, the GHG way of going about doing that, you know, accounting doesn't really translate all that well for software systems and how perhaps adopting a slightly different approach would lead to more better. More actionable outcomes for the folks who want to use this ultimately to change behavior because, you know, without getting into specifics of, you know, what marginal is and what consequential approaches are, and, and if we want, I mean, I'm, I'm sure you know, will, would, would be happy to dive into all of those details as would I.

But the, the, the thing that we were seeing was that we're doing all of this great work around, you know, talking about scope one, two, three emissions, et cetera, but it's not really helping to drive behavior change. And that's really the crux of all of this, right, is that we're not just doing all of this accounting to produce reports and to, you know, spill ink, but it's to concretely drive change in behavior.

And that's where we found that adopting a consequential, adopting a marginal approach actually helped make it more actionable. And this was coming from folks who are a part of the standards working group, including Will and myself, who are practitioners, who, who are itching to get something that helps, helps us change our behavior, change our team's behaviors when it comes to building greener software, broadly speaking.

Chris Adams: Okay. So that helps with explaining the difference between a consequential approach and a marginal approach, as in the consequences of me building this thing will mean that this is more likely to happen. And if I understand it, the GHG protocol that you mentioned, which is the greenhouse gas protocol, and this scoped emissions approach, this is the kind of standard way that an organization might report It's kind of climate responsibility, as it were, when, and when you say scoped emissions, that's like scope one, which is burning burning say that's, that's emissions from fossil fuels burned on site or in your car.

For example, scope two is electricity and scope three is your supply chain. If I understand what you're saying, there's like a kind of gap there that doesn't account for the impacts of this, perhaps. I mean, some people who've referred to this as scope zero or scope four, which might be, what are the impacts an organization is happening to. Essentially, we mentioned before, do something around this systemic change, or as Lynn mentioned, like this is changing the price of a particular commodity to make it more likely to be used or less likely to be used. And this is what I understand the SCI is actually trying to do. It's trying to address some of this consequential approach, because the current approach doesn't capture all of the impacts an organization might actually have at the moment. Right?

Will Buchanan: Yeah, that's a good summary. One challenge that I have noticed is that until it's required in reporting structures like the greenhouse gas protocol, then organizations don't have an incentive to really take the action that they need to avoid climate disaster. Something I encounter on a daily basis, and I think broadly we need to bring this into the public discourse.

Chris Adams: I think you're right. I think it's worth actually, Lynn, I think that when I've seen some of the work that you've done, you've done previously, this is something that's come into some of the briefings that I think that you've shared previously with the climate change AI work and some of the policy briefings for governments as well. Is there something that you might be able to add on here?

Lynn Kaack: Yeah, so something that comes to mind is, for example, like a concrete legislation that's currently being developed is the EU AI Act. And that's a place where, for the first time, AI systems are being regulated also that scale. And climate change almost didn't play a role for that regulation in the first draft.

So here it's also really evident that if we don't write in climate change now as a criterion for evaluating AI systems. It will probably be ignored for the next years to come. So the way that legislation works is by classifying certain AI systems as high risk and also just outright banning some other systems, but as high risk systems could, the original legislation stood, weren't really explicitly classified as high risk, even if they had like a huge environmental or climate change impact. And that's that I talked about a lot with policymakers and trying to encourage them to more explicitly make environmental factors and climate change a factor for evaluating systems. So that'd be a very concrete case where making climate change more explicit in the AI context is important also in terms of legislation.

Abhishek Gupta: So there's, there's a lot to be said about the EU AI Act, right? And a ton of ink has been spilled everywhere, I think, as, as, you know, it's, it's, it's, it's called the Brussels effect for a reason, right? Where the, whatever happens in the EU is taken as gospel and sort of spread across the world, which I think, as already Lynn has pointed out there, It's not, it's not perfect, right?

I think one of the things that I've seen being particularly problematic is, is the rigid categorization of what, you know, high risk use cases are and, and whether the EU AI act, as we'll see, hopefully with some, with some you know, revisions that are coming down the pipe is whether we'll have the ability to add new categories and, and not just update subcategories within the existing identified high risk categories.

And I think that's where things like considerations for environmental impacts and really tying that to this you know, societal impacts of AI, where we're talking about bias, privacy, and all the other areas, is going to be particularly important because we need multiple levers to, to try to account for, or to push on getting people to consider the environmental impacts.

And given that there is such a great momentum already in terms of privacy considerations, bias considerations. I think now is the time where we really push hard to make environmental considerations an equally first class citizen when it comes to, you know, thinking about the societal impacts of AI.

Will Buchanan: This is something I'm incredibly passionate about. I think needs to encompass the full scope of harms that are caused by an AI system. So that could be the hidden environmental impacts of either the development or the application. The application could vastly outweigh the good that you're doing, even just expanding oil and gas production by a certain percentage amount. I think it just must account for all of the harms for both ecosystems and people.

Chris Adams: And there's this idea of like a risk thing. Does this categorization actually include this stuff right now? What counts as a high risk use case, for example, when mentioned here?

Lynn Kaack: So I haven't seen the latest iteration. I think that there has been some update on, there's been a lot of feedback on the version that was published in April last year. I haven't seen the latest iteration. I think a lot of things changed in, yeah, in the first version, there was It's high risk systems where when, that affect personal safety, like human rights in the sense of, of personal wellbeing, but the completely overlooked environmental protection aspects aspects of human rights.

Chris Adams: Wow, that's quite a large one, especially when you take into account the human rights. Okay. We've spoken about the external impact, but I am led to believe there is also an internal impact from this as well. Like the AI has, has some direct impact that we might want to talk about as well. As I understand it, we spoke about 2 to 3 percent of emissions here, but if we know there's an external impact, why would we care about any of the internal impacts of AI we might be doing or why you might want to care about the internal impacts of AI as well, for example, like the direct emissions.

Will Buchanan: So by direct emissions, you're talking about, let's say, the scope, too, of the operational cost of the model.

Chris Adams: Yeah, there'll be things that we have, there's an external impact or there is a, we use this phrase scope 4, for example, to talk about all the other things that induce in the world. But there is a, a kind of stuff which happens inside the system boundary that we've spoken about. And presumably that's something we should be caring about as well, right? So there'll be steps that we can take to make the, the use of AI, particularly like say the, the models more efficient and more effective and more, all these parts here, this is something that we should be looking at as well, presumably right?

Will Buchanan: Totally. And so in our paper, which is going to be published, I think on Monday, we've calculated the emissions of several different models. And one of them was a 6 billion parameter transformer model. And the operational carbon footprint was equivalent to about a rail car of coal. And that's just for training. So it's really imperative that we address this and provide transparency this

Lynn Kaack: Is that Is that for developing a model or for training at once? I mean, is that with grid search, architecture search?

Will Buchanan: For for a single training run. So it does not account for sweeps or deployments.

Chris Adams: All right, so there's a, there's some language that we haven't heard for here, so, but maybe it might be worth asking, maybe Will, could you maybe talk about, just briefly, you said a rail car full of coal, and I don't actually know what that is, I mean, in metric terms, what does that look like?

Will Buchanan: A hundred million grams. I don't have the conversion handy, but we took the U.S EPA greenhouse gas equivalencies. And I should add the methodology that we applied was the Green Software Foundation's SCI. So we calculated the energy consumed by the model and multiplied it by the carbon intensity of the grid that powers that data center.

Chris Adams: Okay, cool, and that was per training run? So that wasn't the, in the, the equation of the entire model, is that correct?

Will Buchanan: Correct.

Abhishek Gupta: That's the other interesting part as well, right? When you're thinking about the life cycle is, or life cycle of the model, I should say, because life cycle has multiple meanings here, which is that once that model is out there, what are the inference costs, right? And are we, if this is something that's going to be used you know, hundreds, thousands, tens of thousands of times, if it's something, you know, if it's, if it's a large model that's, you know, now being used as a pre trained model and is going to be fine tuned on by, by other folks downstream, are we able to then, you know, talk about amortization of that cost across all of those use cases?

And again, I think what becomes interesting and, and is how do we account for that stuff as well, right? Because we, we don't have complete visibility on that as well. And, and I know Lynn's nodding here because her paper that's, I think coming out, being released in an hour and a half, actually the embargo gets lifted on our paper, actually talks about some of those system level impacts.

So maybe, maybe Lynn, you want to chime in and talk a little bit about that as well?

Lynn Kaack: Yeah, thank you so much. Exactly. So I think what's a crucial number that we're currently still missing is not what is emitted from a single model in a well known setting, but what is emitted overall from applying machine learning? So what are the usage patterns and practice? Like how often do people develop models from scratch?

How often do they train or retrain them? People I mean, of course, organizations and typically large organizations and companies. And how do they perform inference on how much data, how frequently? And there are some numbers out there from Facebook and Google and in their large scale applications actually inference outweighs their training and development costs in terms of greenhouse gas emissions.

So inference might become a bigger share depending on the application. So we really need to understand better how machine learning is being used in practice also to understand the direct emissions that come from it.

Chris Adams: An inference is a use of a model once it's in the wild. Is that what an inference is in this case? So there's an environment, so you could think of the making part, and then there is the usage part from the inference, right? So is that how that part works?

Lynn Kaack: Exactly. So if you use a model on a data point, we call that inference. So you feed in the data and it gives you a result. Then training means you sort of train a single configuration of the model once on your training data set, and then development is what I refer to as if you search over different configurations of the model.

So there are lots of hyperparameters that you can adjust to achieve better performance. And if new models are being developed, then there's an extensive search over those hyperparameters and architecture configurations that then of course gets really energy intensive because we are training the model thousands of times essentially.

Will Buchanan: Yeah, one of the figures that really resonated with me, I think Nvidia posted on their blog that inferencing for about 80 to 90 percent of the carbon cost of a model. think Lynn on one of your papers, it was, Amazon had also claimed around 90 percent. So these are really non trivial costs, and I'm not aware of any framework to measure this.

Lynn Kaack: Yeah. So that Amazon number just to be clear is costs or monetary costs that came from a talk, but there are numbers now published by Google and Facebook, but they look at some applications of theirs where inference outweighs training in terms of energy consumption. They're not exact numbers. It's not entirely clear which applications those are, but there is some data at least that shows that.

And I think it just highly depends on the application that you're looking at. And sometimes, you know, you build a model and then you do inference once and you have the data set that you, and then in other types, you build a model and then you apply it billion times a day. So, of course, that can then add up to a lot more energy consumption.

Chris Adams: Wow, I didn't realize that was actually an issue, because most of the numbers I've seen have been focusing on the training part. So, Will, I think this is something we spoke about before, that training, there's a kind of trend in the use, in the energy use from training already. Is this something, because I've seen figures from OpenAI, but my assumption was that basically computers are generally getting more efficient. About twice as efficient every two years, or so with like Moore's Law or Kumi's Law or things like that. But if you're seeing an uptick in usage here, is, does that mean that they're staying about the same? Or is there, is there a trend that we should be taking into account there?

Will Buchanan: So I think the computational costs of training have been doubling every 3.4 months or so, and so I think the trend is only accelerating. The models are just getting larger and larger, and you've got, I think, GPT 3 is one of the largest ones around at this point. We might challenge Moore's Law.

Chris Adams: Okay. So if Moore's Law is doubling every once every two years, I mean, what is the impact of doubling every 3.4 months? I mean, over a few years, what does that work out to be? Because I don't think I could do the exponential numbers, the exponential math, but it sounds like it's, it sounds like a pretty big number, basically, if something is doubling on every three or four months, right?

Will Buchanan: I also don't have the math handy, but I think it's important to note here, and Abhishek was talking about this earlier, models are very flexible, so you can train them once and then provide some fine tuning or transfer learning approach on top of them, and then repurpose these models for a number of different applications. And then you can even compress them, let's say using ONNX Runtime. You can be very efficient. You can really amortize the cost of that model.

Abhishek Gupta: So yeah, just building on Will's point, there's a lot of work on quantizing the weights of a trained network, applying distillation approaches, using teacher student model approaches that actually helps to shrink down the model quite a bit, especially with the whole push for TinyML, trying to shrink down models so that they can be deployed on edge devices has been something that's helped to manage to a great extent, the computational impacts.

One of the other things that I wanted to highlight as, as, you know, Will was talking about more models getting larger is, there's this almost fetish in the world today to continuously scale and keep pushing for ever larger models and in chasing SOTA, as they would say, so chasing state of the art, you know, which is great for academic publications where you get to show, "Hey, I improved state of the art performance on this benchmark data set by 0.5 percent or whatever," right? And in performance, I think what's being ignored is that that has a tremendous, tremendous computational cost. In fact, one of the hidden costs that I think doesn't get talked about enough is there's this statistic out there that, you know, 90 percent of the models don't make it into production.

And that kind of relates to things like, you know, neural architecture search and, you know, hyper parameter tuning, where you're constantly trying to refine a model to achieve better performance. A lot of that actually goes to waste because that stuff doesn't make it into production. So it's actually not even used.

And so there's a whole bunch of computational expenditure that is done that actually never sees the light of day, never becomes useful. That obviously has environmental impacts, right? Because of the operational and embodied carbon impacts, but none of that actually gets talked about, reported, documented anywhere because, well, who wants to know that, hey, I trained, you know, 73 different, you know, combinations to get to where I'm at.

You just talk about the final results.

Chris Adams: Okay, let's say that if you don't want to go down one of those rabbit holes, what should you be using or where would you start if you wanted to start applying some of these ideas about greener AI in your work on a daily basis? Does anyone have anything that they would lead with, for example?

Will Buchanan: Bigger is not always better. Sometimes you really should choose the right tool for the job. We've had some really great graduate student projects from the University of Washington's Information School, and they built some case studies and samples around green AI. As an example, a project led by Daniel Chin was comparing a sparse or a dense model of a green AI model to a dense model for an anomaly detection setting.

And they found that using sparse, meaning less trees and a shallow, smaller depth per tree, random forest would save a massive amount of carbon and provide the equivalent accuracy. So I think it saved about 98 percent in terms of the monetary cost and energy.

Chris Adams: Okay, wow, that's bigger than I was expecting. What would you say to people if they're in production, they're trying to do something now?

Lynn Kaack: I think a big goal should be to not only develop more energy efficient machine learning models, but then also ensure that those are actually being used. And surprisingly, even sometimes within the same company, certain model developments are not being passed on to other parts of the company. So really trying to develop standard models that then are also being used in practice is important.

So interoperability of energy efficient machine learning models.

Chris Adams: If people, someone does want to look at this stuff, and they do want to apply some of these ideas, you spoke a little bit about using some other models. would you suggest people look if they wanted to operationalize some of the kind of wins or some of the better ways to make green software greener, for example?

I realize you've got a paper coming out and you work on this day to day. So yeah, what would you point us to?

Lynn Kaack: So, I mean, as I understand, there's a lot of ongoing research in the machine learning community for more energy efficient machine learning. So I don't have any names on top of my head in terms of workshops or community resources where one can see what are the most energy efficient model types. For a specific application.

I know that there are some very comprehensive papers also that summarize all the different research approaches that are being taken, but I would encourage if you are looking for using like a deep learning model of some kind, just inform yourself quickly if there's also a leaner version of it. So much of the like widely used models like BERT, for example, smaller versions that can almost do the same thing.

And maybe your performance doesn't suffer much. If you're using a much lighter model.

Chris Adams: Okay, so light up models and looking around what we have there. And Will, is there a paper or a source you might point to?

Will Buchanan: I was actually going to talk about the Carbon Aware paper that we're about to publish. I think that's a slightly different track.

Chris Adams: That's up next week, right? So that'll be the 13th or 14th of June. That's when that'll be visible, correct?

Will Buchanan: Exactly.

Chris Adams: Okay, cool. All right, then. There's a load more that we could dive into. We've got copious, copious, copious show notes here. So what I'm gonna do is I'm gonna say thank you everyone for coming in and sharing your wisdom and your experiences with us, and hopefully we'll have more conversations about green software in future. Thank you folks.

find out more about the Green Software Foundation, please visit greensoftware. foundation. Thanks again and see you in the next episode.

Backstage: TOSS Project

10 oktober 2024 | 14 min

In this episode of Environment Variables, host Chris Skipper introduces "Backstage," a new segment offering behind-the-scenes insights into the Green Software Foundation's key projects. The episode highlights the "Transforming Organizations for Sustainable Software" (TOSS) project, led by Pindy Bhullar, which aims to embed sustainable practices across all aspects of business operations. The TOSS framework, divided into four pillars—strategy, implementation, operations, and compliance—provides organizations with a decision tree approach to drive sustainability in software development and IT. With collaborative efforts, workshops, and real-time feedback, the project is set to release version 1.0, with the ultimate goal of reducing carbon emissions across the industry.

Learn more about our people:

Chris Skipper: LinkedIn | Website
Pindy Bhullar: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Events:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Chris Skipper: Welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Skipper. Today, we're launching an exciting new segment:

Backstage. Here, we'll take you behind the scenes at the Green Software Foundation to highlight ongoing projects and share insights from the members directly involved.

In the first episode of Backstage, you'll hear about innovative projects shaping the future of green software and the hard work behind it all. It's easy to overlook the complexity and dedication required to drive sustainability in the tech industry. But today, you'll hear more about the passion, thoughtfulness, and expertise that fuels these initiatives.

From collaborative projects to cutting edge research on sustainable computing, the Green Software Foundation is the conversation you need to be part of, and we're bringing you Deep Inside. Join us as we look into the stories, challenges, and triumphs of the people making a real impact on our environment through software.

We're no longer gatekeeping what it takes to set new standards and norms to green software. So, what does it take to inject sustainability into every aspect of software development? How are organizations and passionate climate activists in software turning ambitious goals into practical, actionable results?

Let's find out on this special Backstage episode. For this initial episode, we'll be looking at the Transforming Organizations for Sustainable Software project, better known as the TOSS project. Leading this project is the inimitable Pindy Bhullar, who provided us with some soundbites. This initiative is really important in the Green Software Foundation's broader mission to reduce carbon emissions through software.

It's all about providing changemakers in organizations with the tools, methodologies, and frameworks they need to embed green software practices and drive real, measurable change. Firstly, we asked Pindy to give us an overview of the TOSS project. It's goals and why it's so crucial for the foundation.

Pindy Bhullar: Transforming Organizations for Sustainable Software is the acronym for TOSS. Businesses will be able to utilize the TOSS framework as a guide to lay the groundwork for managing change and also improving software operations in the future. Software practices within organizations can be integrated with sustainability in a cohesive and agile manner, rather than addressing green software practices in an isolated approach.

For a company to fully benefit from sustainable transformation of their software development processes, we need to review all aspects of technology. The TOSS framework is designed to be embedded across multiple aspects of its business operations. Dividing the task framework along four pillars has allowed for simultaneous top down and bottom up reinforcement of sustainable practices, as well as the integration of new tools, processes, and regulations that emerge over time.

The four pillars include aim to foster a dynamic foundation for companies to understand where to act now, to adjust later, and expand with an organizational sustainable software transformation. The four pillars are strategy, implementation, operational, compliance, and regulations. And within each of the pillars, we have designed a decision tree that will be constructed to guide organizations in transforming their software journey.

The Green Software Foundation has created the specification to calculate carbon emissions from software, the SCI, and also developed the Impact Framework to enable firms to calculate their emissions. But in order for firms to transform their technology operations, we need to embed best practices across the wider organization.

Chris Skipper: It's clear that by integrating sustainability across multiple levels of business operations, the TOSS project is setting up organizations for long term success in their green transformation journey. Now, let's talk about the impact this project can have. Reducing the carbon footprint in software development is a huge focus for many companies today.

So, how exactly does the TOSS project help achieve that, and what kind of sustainability improvements can organizations expect by adopting this framework?

Pindy Bhullar: The TOSS project views sustainability in software development and IT operations as a holistic approach, and the primary goal is not to view software and IT as an isolated function, but integrate the carbon footprint into the functions across the organization.

The framework will help organizations to ask the questions and adapt their responses based on their needs, but also connect strategy, implementation, operations, and compliance and regulations.

Chris Skipper: It's insightful to see how the TOSS framework fosters a more integrated approach, making sustainability a core part of a company's strategy, operations, and compliance.

Now, as with any large scale initiative, developing a framework like TOSS must come with its challenges. We asked Pindy to share some of the key obstacles the team has encountered while bringing this project to life, and how have they navigated these challenges to keep the project moving forward.

Pindy Bhullar: We're all busy in our day jobs, so just being able to focus on the TOSS framework has been a challenge.

We've been having regular calls since the project inception January this year, which has enabled us to keep the discussions moving. We have overcome obstacles by building a version 0.5 of the framework and then asking volunteers from the GSF member organizations to join a series of workshops to review, question, challenge, and brainstorm. It enabled us to develop the framework further by incorporating all the great feedback.

Chris Skipper: It's great to hear how the team has managed to build the toss framework through collaborative efforts and feedback from GSF member organizations. It sounds like the workshops have played a crucial role in refining the framework.

Now, Pindy mentioned that the TOSS project incorporates a decision tree strategy. We asked Pindy to walk us through how this approach is helping organizations embed green software practices more effectively. How does this decision tree guide organizations in making sustainable choices throughout their software development and IT operations?

Pindy Bhullar: We use the decision tree approach for the TOSS project for a number of reasons, particularly in contexts where decisions involve multiple options and varying outcomes. It's also a great way to visually represent complex decisions into manageable parts, and it helps decision makers visualize the entire process in a structured, analogical format.

It also enables the evaluation of multiple potential paths and their respective outcomes. So each decision point can have different branches, each representing a different course of action. One of the benefits of decision trees is the ability to easily update or expand as new information becomes available.

This flexibility is critical in the space of green IT and green software, as this topic is constantly evolving.

Chris Skipper: It's clear that the decision tree approach adds flexibility and adaptability, making it a perfect fit for an evolving field like green IT and software. Now, let's dive into the specific tools and technologies that are supporting the TOSS project.

We asked Pindy to highlight some of the key tools or frameworks being used and explain how they align with the project's sustainability goals. How do they contribute to helping organizations achieve lower emissions and more sustainable software practices?

Pindy Bhullar: To build the framework of decision trees, we are using mirrorboards.

Initially, it helped us to brainstorm and use the ideation tools to create mind maps to generate and organize our ideas in an interactive way. We also benefited from the real time collaboration and, more importantly, the ability to create the visual charts. The tool is enabling us to progress with the TOSS framework and build the sustainable software decision trees.

Chris Skipper: Now, let's explore the collaborative side of this initiative. We asked Pindy if there have been any significant partnerships or collaborations with other organizations or teams that have enhanced the success of the TOSS project. How have these partnerships contributed to its development and progression?

Pindy Bhullar: There have been excellent collaborations during this project. The workshop sessions were critical in partnering with organizations because it enabled us to further build the TOSS decision trees. It also helped us to make faster progress, while allowing the group to discuss new ideas, make revisions, and also to bring new insights.

Chris Skipper: It's clear that collaboration has played a crucial role in pushing the TOSS project forward, with the workshops really driving the development of those decision trees. Now, let's shift focus to the progress the TOSS project has made over time. We asked Pindy to walk us through some of the significant achievements or milestones the project has achieved so far, and how has it evolved since its inception earlier this year?

Pindy Bhullar: At the start of the project, we created a business case which outlined the TOSS project and what we wanted to achieve. Then we moved into a phase of brainstorming and mind mapping, really trying to capture key points of how we wanted to develop the framework and decision trees. We started then to break down the functions that would be involved in organizations and we created the four pillars that I mentioned earlier.

And out of that, then we started to get into the detail and start to create the decision trees. We have progressed since the workshops, and now we're working towards version 1.0 of the decision trees. We have redeveloped the strategy and implementation pillar, and we're still working on the operational and compliance and regulation pillars.

From the initial ideas, we've added further content and details to the framework, and it's constantly evolving.

Chris Skipper: It's truly fascinating to hear how the TOSS project has progressed. From the initial business case to the detailed decision trees, and the continual refinement of the strategy and implementation pillars.

Now, as we look to the future, we'd love to know more about what's on the horizon for TOSS. What are the next steps in the project, and what long term impact will it have on the widespread adoption of green software practices across the industry?

Pindy Bhullar: Over the course of the next few months, we want to complete version 1. 0 of the four decision trees. Also, create the guidance documentation that will sit alongside it. And in addition, create a white paper documenting the framework. We want the long term impact of the framework to help organizations implement tools, methods for calculating their emissions, processes, ensuring that we link the strategic level of the organization all the way through to software development and operations.

Chris Skipper: I'd like to thank Pindy for sharing such valuable insights into the TOSS project. It's clear that this framework is not only about integrating sustainable practices, but transforming the entire approach organizations take to software operations. From her detailed explanation of the four pillars, strategy, implementation, operations, and compliance, it's evident that TOSS provides a holistic solution that embeds sustainability at every level.

The collaborative workshops and iterative feedback process she described have played a crucial role in refining the decision trees, and it's exciting to hear that version 1.0 is on the horizon. As the project evolves, it's clear TOSS will have a lasting impact, empowering organizations to seamlessly incorporate green software practices and meet their carbon reduction goals.

We truly look forward to seeing how the framework will continue to grow and shape the future of sustainable software. Now, we've come to the end of the episode, but before we finish off, we have a few events to read out. Firstly, we have Sustainable AI, Myth vs Reality, Gain Insights and Actionable Strategies on the Intersection of IT and Environmental Responsibility at this in-person event hosted by Sustainable IT Netherlands at Accenture in Amsterdam on October 15th, starting at 6pm Central European Standard Time.

Next up, we have CNCF Sustainability Week, Cloud Native for Green Software. Share and explore ideas, patterns, best practices, and tools in the context of the green software principles. This hybrid event is hosted by Green Software development, Karlsruhe at Blue Hands in Karlsruhe on October the 15th at 6:30 PM CEST.

And finally, we have Code Green London October Meetup. Learn about the latest in green software and network with London's green tech community. Also on October the 15th at 7:00 PM BST. This in-person event at the Red Badger in London will feature past environment variables guest Tereze Gail. Who will be talking about Salesforce's sustainable AI approach.

Hear how the world's largest enterprise software company is balancing AI innovation with environmental responsibility. So, we've reached the end of this special backstage episode. I hope you enjoyed the podcast. To listen to more episodes about Green Software, please visit podcast.greensoftware.foundation and we'll see you on the next episode. Bye for now.

The Week in Green Software: The Sustainable Data Paradox

3 oktober 2024 | 43 min

This Week in Green Software, the affable Anne Currie is joined by Sara Bergman, Senior Software Engineer at Microsoft and co-author of Building Green Software. Together, they dive into the complexities of sustainable data in relation to AI and cloud computing. They explore the environmental impact of managing and storing vast quantities of data, and question the feasibility of making these processes more eco-friendly. The discussion touches on cloud providers' carbon reporting, the importance of using AI responsibly, and how businesses can optimize their cloud use to minimize their environmental footprint. Tune in for an insightful conversation on balancing technological advancements with sustainability in the age of AI.

Learn more about our people:

Anne Currie: LinkedIn | Website
Sara Bergman: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Resources:

Building Green Software [Book] [02:04]
Greening Digital and the Rebound Effect | Environment Variables [11:36]
GitHub - Green-Software-Foundation/real-time-cloud [21:35]
Green Software Maturity Matrix [21:55]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Sara Bergman: What data do we need to take in order to take meaningful action? Like, what is the level that, of course, yeah, if I could get minute by minute, like there's tons of stuff we could do and correlations we could draw, but what is the level of data that we would need to start taking meaningful action? And I think that could unlock a lot of good things.

I'm your host, Chris Adams.

Anne Currie: Hello, and my name is Anne and welcome to

The Week in Green Software. So this week, you won't be hearing the usual dulcet tones of Chris Adams. I'll be joining you, Anne Currie, and we'll be delving into the tricky, but interesting world of sustainable data, whether it's possible to store and manage huge quantities of data, which we will need now, particularly for AI, in a way that's eco friendly.

Or is that impossible? Well, look, I'm going to leap through to the end and say, we have to do it. And therefore it is not impossible. It just has to be done. We'll find a way of doing it. And there are ways of doing it, which we'll be talking about today. we'll also be exploring why big cloud providers' carbon reporting isn't really telling us the full story.

Or, well, we are not interpreting it in the way that it was designed to be interpreted. And we need to be more careful about that. We're reading more into it than is true. And we need to be much, much more explicit about what the carbon reports from people like AWS and Azure actually mean.

We'll also be talking about data centers and AI. And that's something that my guest today is an expert in. And talking about my guest, joining me today is my co-author on Building Green Software, the book on what we need to do to make the tech industry green, and also Environment Variables regular, the lovely Sara Bergman.

So Sara, do you want to introduce yourself?

Sara Bergman: Yeah, hi. Thanks for having me on again. Always lovely to be here. I'm so excited to have a chat with you, Anne. Yeah, my name is Sara Bergman. I am a senior software engineer at Watttime. Microsoft, author of Building Green Software. And like here, I'm always asked, what have you been up to recently? Nowadays, I'm like, what have I been up to recently?

I did a fun thing, though. I had the talk for the Norwegian, because I live in Norway, and the Norwegian tax authorities about green software. And that was really fun. I love it when Because they are actually very far ahead in their journey, they're one of the most innovative companies when it comes to IT, so it was really fun to come out and have a chat with them.

Anne Currie: That's great. And I should introduce myself as well. My name is Anne Currie and I am co-author of O'Reilly's new Building Green Software along with Sara and our other co-author, Sara Hsu. And I also do a lot of training. So I've been busy at the moment doing loads of training courses. So, workshops on building green software and also an experts training course, which is all quite good.

So, if you want to get involved in any of that stuff, you can follow me on LinkedIn. So, as usual, today, we will be talking about a couple of interesting articles, publications that have come out over the past week around green software, all things green, and as usual, all the links to the articles will be in the show notes, so you'll be able to read them yourselves afterwards.

But I'll give you a little bit of a summary about what they say. So the first article we'll be talking about today is from the Green Web Foundation. And it was written by our normal Environment Variables host, Chris Adams. So that's where he is, or that's what he's been up to today in his work at the Green Web Foundation.

And he wrote it with his colleague, Hannah Smith. And the report is all around AI's environmental impacts. And it's got some interesting figures in there. Basically, AI uses a load of electricity and at the moment, as we don't yet have a completely green grid, that means that a lot more CO2 is being emitted into the atmosphere as a result of the fact that we're training a lot of models, doing a lot of inference.

So it is an interesting report and it's, I'm going to kind of summarize, they have some actionable things, some questions for you to ask yourself at the end of the report. So I will go over those now and then Sara and I can discuss them. The first is that you should always question your use of AI.

That's kind of part of using the right tool for the job. Is AI the right tool for what you're doing? Is it overkill? Could you use a spreadsheet? That's If you are using AI and you decide it is the right tool for the job, are you using it properly? Are you using it well? Are you using the right AI tools for the job?

And the third is to try and get ideas of your footprint, of the footprints of the work that you're doing with the AI, so you actually have an idea about what impact it is having and what you're going to need to do about it in the future. So Sara, this is kind of your area. Did you enjoy the report? What did you think?

Sara Bergman: I did. It's a very long report. It's very well written. Obviously, I mean, with Chris and Hannah, you're going to get something that's well written, of course. So no surprises there. No, it's good. And I think also for people who are maybe newer to the field of green software and green AI in general, there was a lot of good background to like really help understand the intricacies of this area. And something that I particularly find interesting in the shift we are now is that they talk about different phases where your emissions kind of stem from, it's like manufacturing, training and inference. And now, like you said, we talk a lot about inferencing, like using AI, that the use phase is what we talk about.

But back when I started, sounds like I'm really old, only four years ago, not that long ago, but when I first started talking about green AI. Yeah, a lot of, a few people, not so many were talking about green AI. A lot of people were researching, but not so many people were like discussing it. And then it was a lot of focus on the training.

There was a lot of great research being done on how to minimize the impact on training. I think in the research community, that's maybe the easiest, not the easiest thing, but a good first thing to research, right? And now we're seeing more focus on. on the production side, not like inferencing. And I think that shift has been very interesting to follow.

Anne Currie: But yeah, it is fascinating, isn't it? Cause there's loads, and we've had this conversation offline, because obviously we a book together and therefore we talk together quite a lot. But we've had this conversation a lot in that it feels like there's a load of stuff to be learned about inference.

So how you actually get the answers back as a user for models from the world of things like CDNs, how do you get fast answers and easy answers to things all over the world from data that is not necessarily by default, wasn't created close to the user who's querying it?

So yeah, there's, loads of prior art there to learn from it. It's a really interesting field.

Sara Bergman: It is a very interesting field and I think an additional like spiciness of this is that now a lot of people equate AI to large language models. That's not all of the AI that exists in the world, it's just the right now the most popular and what we think maybe will be the most interesting use case, or for like productivity and stuff.

And for that we require a lot of specialized hardware, like GPUs or TPUs on CPUs, et cetera, which is another thing to kind of reason over. And, maybe if you're very good at handling your production workloads on a CPU, like this might be a new area for you where you, we need to learn as an industry. I thought that was an interesting part of the article.

Anne Currie: Yes. Yeah, it is. It is all very interesting, isn't it? it's like something that all three of us say in the book a lot, is that there's a load of clever stuff going on in the tech industry, and it's usually, there's a desire to make things efficient because it's cheaper. Otherwise, everything gets very expensive.

So if you use things as they were intended to be used and use the right tools for the job, usually that's significantly more energy efficient and therefore greener. And the same is very much true of AI, isn't it? If you use the right tools, the right hardware, the right models,

Sara Bergman: Yeah, exactly. Exactly. So that and then I think that ties into right, that was it number two on their like, use AI responsibly, which was interesting. I also like the first question that they had. Now I'm scrolling and losing it was it think about when you should use AI? Was that the phrasing?

Yeah, question your use, know when to use it. Like, do you have a favorite use case?

Anne Currie: Well, I have to say, I do love generative AI.

Sara Bergman: Yeah.

Anne Currie: it's, an interesting one, cause quite often... well, it's amazing how often it comes up where people think that it's kind of like you're either doing AI or you're green. That's it. Those are the two, two options. And they cannot be the two options.

It can't be either don't do AI and therefore it's green or do AI and therefore it's not green. They have be brought together because if there's one thing that we know, two things we know are coming in the future for humanity. One is climate change. The other is AI and use of AI and AI systems.

They cannot be mutually exclusive. They have to be brought together. I mean, and AI is just computing. It's the same kind of things that we do when we talk about in Building Green Software for all computing just needs to be applied to AI. It's not a, it's not a new thing on its own, but

Sara Bergman: and I agree. And I think that's also so interesting with this, like, larger discourse, as you say, I think not only that it exists in other areas of life, where there are some people who would be like, "no, we should just stop doing this, like, stop advancing technology, and then everything will solve itself," but that doesn't work. We cannot and does not want to go back to like a farmer society where everyone grows your own food. Like if you're into growing your own food, that's fantastic. I'm happy for you. I grow radishes on my balcony and I enjoy that thoroughly. But there's so much advantage and good for humanity and our planet that comes from technology as well.

So we need to do both. We must do both. Like, it's non negotiable. So it's more like how and when? And 'when' is hopefully starting now.

Anne Currie: Cause the trouble is, if you have that conversation, if you say, "Oh, well, I want to do AI, therefore I can't do green." People will choose AI because the businesses, if they don't, if they don't try these new tools and services as they come available, they will go out of business. So, if you say, "Oh, well, if you're going to be green, you can't use AI," then what you're telling people is don't be green.

And it's crazy. You're just making the wrong arguments. But anyway.

Sara Bergman: I fully agree for sure. And I think it was something we were in a different, or like this podcast together, but another episode and you said something that I've been carrying with me for so long. Sometimes people say something and it just get like something clicks. And what you said, and I'm going to maybe paraphrase, but things that are limited are sort of less concerning. Like, for example, if you build a super efficient washing machine, I still have only so much washing up to do. There is an end to how much I will wash it. Like this is in response to Jevons paradox, of course. And the same goes for this. Like, yeah, AI will make us more efficient, more productive.

Okay. But we'll then just do more. It's like, yeah, but the working day is sort of eight hours and we are only so many people on the planet. There is a limit to when we'll be done, whereas for crypto, for example, which always comes up when we talk about this, of course, there's no limit. I will never say I have enough money, I'm done mining. But there is a limit where I say, thank you, co-pilot for GitHub, I don't need more code suggestions.

I'm happy with my feature now, sort of.

Anne Currie: Oh, yeah. That's an issue, because, I always tend to think that one of the, one of the things about AI is that, that we're not,, we haven't yet reached the limits of what we want to do with it. But, you're right. There's, it's nowhere near. I mean, we've already come up with the thing that is the most limitless.

You've heard the use of CPU and it's CPU use that's really the problem that's, that uses a lot of electricity, is crypto. It is literally boundless how much we might want to do with it. So we've kind of already created the worst 30.

Sara Bergman: yes.

Anne Currie: AI is so

Sara Bergman: not it.

Anne Currie: it's a comparison. At least there's some people,

I don't know, we shouldn't be controversial because there are a whole load of people who say, well, there's loads of benefits of cryptos, but I personally do not, I'm not a crypto bro.

Sara Bergman: Same. No, I did write my master's thesis on blockchain. I think blockchain is a cool invention. Like, it's a cool technology. And there are use cases that are, of course, bounded. And then there are use cases that are unbounded. And I think people can use their own head to figure out what's what.

Anne Currie: We've now got ourselves into enough trouble talking about good and bad uses of technology. And to a certain extent, it's very hard to say what's a good or bad use of technology. I tend in my head to think it was a bad use of technology, I, not everyone agrees with me, but we will move on to the next article, which is one, now...

So this one, cloud provider Greenhouse Gas reporting isn't enough. The case for product level accountability. Which is, again, the link is included in the show notes. Now I know, because this talks about Azure, Sara, you, might be in a position where I'm saying things to you and you have to say, you may say that "Anne, but I cannot possibly comment."

So in which case you'll just have to leave me to talk, but that's fine. Cause I can talk endlessly about this stuff because I find it really, I think this is absolutely fascinating and there's loads of stuff to learn from this article. So the article is all about, it's highlighting the limits of cloud providers and data, the carbon footprint reports.

And it emphasizes the need for companies to track emissions at product, at data product level. And it argues that these broader reports mask the true environmental costs of specific processes like data and that things would be better if you broke the emissions down to product level. And I think that's great, but I don't think that's actually the problem.

I think it's well worth reading the article, but I think it skips over, I think, what is the more significant problem that we're having, that we're seeing, because I'm out there talking to people a lot and I know this is a significant problem at the moment, that the cloud providers' reports are doing something, a lot of the new cloud providers' reports, particularly the recent AWS releases are doing something which I can't really shout at them for because I demanded it myself some years ago and I am somewhat hoist by my own petard for this.

So many years ago, me and a fellow, a collaborator called Paul Johnston ran a, back in 2018, we ran a campaign called Sustainable Servers by 2024. And what we were campaigning for was that all of the cloud providers would commit, and this was some time ago, would commit to being carbon neutral by 2024.

And it was really aimed entirely at AWS because Google and Azure were already carbon neutral at that point and AWS was not. So, we ran this big campaign and we had petitions and all kinds of things for saying that they should be carbon neutral, collateral and tradeful. And of course, carbon neutrality is quite a limited demand.

It's basically saying, "I want you to be carbon neutral. I want you to know how much your carbon emissions are. And then buy offsets that match the carbon emissions." Now, we all know that offsets are good in the past, but they're not time matched. They don't mean that the carbon savings that you've created are time matched with the carbon emissions of your systems.

So your systems can still be throwing off carbon dioxide into the atmosphere whilst you are carbon neutral. So it was a useful, a useful measure in 2018, it was still useful. Now in 2024, it's actually not as good as it could, it's not enough. it's the bare bones. It's the least we can ask for.

But AWS have done it on time, 2024, and they are producing these reports that they're giving to AWS customers that say, "you're carbon neutral, your systems are net, not producing any carbon." But the problem is that a report that says, and that's great, but it is nowhere near enough.

It's not enough. It's great information for your finance team because your finance team don't want that data. They need that data, particularly if you're going to be, if you're in the EU or you're selling into the EU. Because the EU is now demanding carbon taxes, and carbon taxes are another form of offsetting, really.

You pay for the carbon you're emitting into the atmosphere. So if AWS produced a report saying, look, there's no net carbon you're producing into the atmosphere at the moment, which they are generally at the moment for most AWS systems, it seems. That is really a report for your finance team, so they don't have to, so they can go, "oh, that's great.

The offsetting has already been done for me." That is not a report for your tech team because they're not saying that no carbon is actually being produced by your systems. Your systems are still producing loads of carbon and you still need to adjust your systems. to reduce carbon, which you can do, but the report is kind of giving you the impression that you don't need to.

So is that something you can comment on or?

Sara Bergman: Yeah, this gets me going, right? Cause I'm all about enabling like people with feet on the ground, hands on the code is what I say. So I do agree a lot of the reporting that comes out, it's much better than it used to be. This difference, and maybe now we're slipping into the next paper a little bit, but there, of course, difference between market-based and location-based reporting in the Greenhouse Gas Protocol.

That is a feature, a fundamental feature of the Greenhouse Gas Protocol, which every single industry makes use of. Now, not only,

Anne Currie: I'm going to interrupt you at this point, just because this is something that I realized when talking to people. One of the issues is that people don't understand the difference between market-based and location-based. What does that mean? So market-based, I think that the word that people really understand is offset.

If you're market-based, it's saying you're still putting carbon into the atmosphere, but we are offsetting it, which was great 10 years ago, nowhere near good enough nowadays. location-based means actually, it's about reducing, it's about not putting carbon into the atmosphere. So that's what we want. Market offsets are a step to getting there, but they are not there.

So sorry, go on.

Sara Bergman: yeah, no, but that's, good. I also like it's like the electrons you pay for versus the electrons you use, if you want to get really down to the socket level. So yeah, I think that's something to talk about. Also, something that I kept thinking about after reading this article, and I just want to read the comment from it because I think the comment was amazing.

And the comment is, "conceptually, I agree. More data is better. However, I've never met anyone advocating for product level data who has actual operational experiences of running shared service platforms and therefore would understand the complexity of delivering these metrics. Because yes, if we're talking about, we could have one report for the finance team, that's fine, but should there be an additional one for, the people who write the code, who maintain the service? And then the question becomes, okay, but what data do we need to take in order to take meaningful action? Like, what is the level that, of course, yeah, if I could get minute by minute, like, there's tons of stuff we could do, correlations we could draw, but what is the level of data that We would need to start taking meaningful action? And I think that could unlock a lot of good things.

Maybe we don't need the world. Do you know what I mean?

Anne Currie: Yeah. I know what you mean. I totally know what you mean. I think we should step down our emissions. Well, having said that, within the Green Software Foundation, there is a project, the Real Time Metrics Project. And that is looking forward to when we can have second, millisecond by millisecond data about, so we can tune applications to, to get there.

But then we also have another project, which I run, so their real time metrics project is run by Adrian Cockcroft and I run a project called the Maturity Matrix Project, which, comes out of our book. So it's, the penultimate chapter in our book. And the Maturity Matrix Project is start, is, it says that all that real time stuff is really quite advanced.

It's way further advanced than pretty much anybody in the world currently is. What we actually need at the moment is quite simple stuff, like "just turn off machines." Turn off machines when you're not using them. And you can, you don't need real time metrics for that. The other thing that we can do that doesn't require real time metrics is, so one of the things that kind of annoys me about the new AWS Cloud Footprint report, which is, it's fine, it's not a lie, it's just giving you offsets.

It's just telling you what your offsets are. It's totally fine. But people are misusing it and misreading it as thinking it's about carbon emissions. There's another thing that AWS have said, which I really like, which is that, "look, we all do a load of work."

It's called the shared responsibility model. "We will take responsibility for the sustainability of the cloud, if you take responsibility for the sustainability in the cloud. Now" I like it, except that it's also very confusing. What they are saying is that "we will build tools that can be aligned with actual location-based zero carbon operations.

We will build those tools." Those tools in the book, we call them green platforms. They're things like serverless, spot instances, really clever instance types. You see, it kind of aligned with their whole modernization strategy, how you get into the cloud and use it well, as it was intended to be used.

So they're saying that, "but it's your responsibility to use those tools. We can build those tools, but if you don't use them, you won't be green. If you just sit in dedicated instances, you will, there's nothing we can do about it. You will never be green. We'll offset you," which is why, the reports say, "well, we'll pay for your bad behavior.

But it is still bad behavior." You'll get to carbon neutral, but you'll never get to carbon zero. And I think that those messages, which are quite complicated, can both exist, but they need to be quite clearly communicated. And at the moment, I don't think we are being so clearly communicated. What's your thoughts?

Sara Bergman: Yeah, this is also, I think this is also something I thought a lot about reading this article. Like, where is the line? Because many cloud providers, they do have clearly green ambition. They are financially incentivized in many ways to be more energy efficient, use less resources of course, because all of those things cost money for any cloud provider, even if that cloud provider is your local on-prem.

So that is one thing, they wanna make cheaper stuff that are often greener, almost always greener, but it is your responsibility to get on them. And how do you do that? Now, all three of the cloud providers do have architected frameworks that have a dedicated sustainability section.

And there for some scenarios, it's actually incredibly good. Like it's very detailed. So you can go and just like, "Oh, my scenario fits into this." I would like those sections to be longer for all of the big three and for the smaller ones, maybe to include more. But I also now maybe skipping randomly ahead, back and forth, something, a mental image that I think helps because sometimes I think the key message is getting lost.

"Okay. But the cloud is green. Why do I need to take action?" and that is a tricky conversation to have because Yeah, if you're selling the cloud, of course, you don't want to say "no, but it's not green." It's like, it just becomes a bit messy. So a mental model I like to use, which I heard from another wonderful woman here in Norway, which is if you have a car and you have a ski box, because right, we have lots of skis here.

So they can think of the car as the cloud and it's the responsibility of the car manufacturer to make that as efficient and green as possible. And they do, right? And but you, as a user, you choose how do I transport my skis in the car? Do I transport them inside the car? Yeah, maybe they that would be greener right because then there's no additional wind catch of the ski box on top, but maybe they don't fit, so you put them on the ski box on the roof. Fine, you accept that additional wind cost and thus increase the energy. But, once ski season is over, and you don't need the skis anymore, what does eeveryone do? You've remove the ski box. I don't see anyone riding around with the ski box in May, just because, right?

And the same goes for the cloud. So yeah, the cloud has lots of work, a lot of stuff in it to make it greener. But if you choose to use a ski box, that's fine. But once the ski season is over, remove the ski box. So once you're not using your test environment, remove it, shut it down.

Anne Currie: Yeah. Yeah. I like the analogy. It's very Scandinavian.

Sara Bergman: It's very Scandinavian, I'm sorry, but I am very Scandinavian, so that's what you get. I guess the same applies if you have a surfing board, I suppose you put them on the roof as well,

or a canoe,

Anne Currie: A canoe!

Sara Bergman: yeah, or a kayak, yeah. They have longer seasons though, but, so yeah, to try to translate, I feel like I was very, poor Chris who has to edit this, all my ramblings.

But yeah, as a cloud user, you have a responsibility and I think there could be two reports of showing this is what we as a cloud provider, I think all cloud providers are pretty good at this. This is what we took responsibility for. That's awesome. So much better than it was like 10, 5 years ago.

But then also how do we enable people using the cloud to take green actions? Because they want to take green actions, we want them to take green actions, like how do we enable that at the same time? Yeah.

Anne Currie: Yeah. I mean, it's, kind of like "is the cloud green?" The trouble is the, answer is as always in tech, it depends. Oh yeah. It's like, it depends. Are you using it as it was intended to be used? And if you, and the, one of the reasons why you sometimes see Gartner reports and things saying that the cloud is really poorly used,

it's because there's two stages in moving into the cloud. One, is you just kind of lift and shift. For most people, they lift and shift. Although you might argue that's actually never a really good way to go into the cloud. I've written books about this. It's not, but it's, it is a way, a common way for people to go into the cloud.

But once you've lifted and shifted, you have to then actually use the cloud, as it were, you see, you have to go beyond that. If you stop at lift and shift, it's really ungreen. It's worse than being on prem', because in the cloud, it's so easy to over provision. So you have to actually, you have to adopt the cloud, it's called cloud native and it's not really cloud native, but then you have to adopt the cloud way of doing things because otherwise you will over provision and you will then make more carbon than you did on prem', because it's harder to over provision on prem'.

Sara Bergman: Yeah, And that's a journey. And that's something I think all cloud providers are pretty good at supporting, right? That's their bread and butter in many ways.

Anne Currie: It is, they just want people to do it, but still it's hard because it's really so hard to lift and shift into the cloud that once people get there, they go, "thank God for that." And they don't want to go and look at what happens next, which is kind of one of the issues of lift and shift as opposed to kind of just slicing up and moving bits and bits one by one into the cloud, is that

these big projects, they're so painful that once you've done them, you never want to, I've done so many big projects in the past, and you always think you're going to go on and do the next stage, but you're so destroyed by that project. You just think, "Oh my God, we're all burned out. We don't even want to think about it again."

It's, yeah. Yes.

Sara Bergman: We need to glamorize DevOps or operations. It needs to be something there, more and more afterworks, I don't know. Something to make them more glamorous.

Anne Currie: These big projects are awful. I never want to do one again. But anyway. Yeah. And they're, ungreen because you then go on to the next bit where you start to actually refine and improve. So yeah, it's, we need to step back and think about how we're doing that. So, which is, I think a lot, I was signing books at a conference about kind of, DevOps and, CICD and team topologies and moving faster releases, last week.

And I think that's really green because if you can't move, you can't adopt these better tools and services that remain.

Sara Bergman: Yeah, exactly. Yeah, I also said someone, or heard someone who said like, what we need now is like performance engineering, we need to go back to basics in many senses, right? And because a lot of these learnings are not new. It's the same when people are like, "ooh, if I," now I'm going on a tangent here, but if people want to be greener, they're like, "oh, I should just rewrite, like, make my code more efficient."

And like, by all means do. But if you haven't had like a reason to do that yet, I don't think you're going to convince your management chain that sustainability is going to be the reason why you go implement it because high performance computing is not new at all. Like we've known about all these things for a very long time.

So if you haven't done it yet, I mean, kudos to you if sustainability is the thing that makes you implement it. Kudos to you. I just think it's going to be a tough sell.

Anne Currie: I've mentioned a few times on environment variables, but it always interests me, is on that note of efficient code is expensive and we don't do it anymore, sara now works on Microsoft Exchange and 25 years ago, I also worked on Microsoft Exchange, but in Microsoft Exchange 25 years ago, we had to have everything written in C because the hardware was, it was just not possible to do it using anything other than the most optimized C code, which meant that it was really expensive to do things.

It took a long time. Yeah. The world has changed. we've got better hardware and we use it. We use it to go, to not have to write everything in C.

Sara Bergman: Exactly. And we see the same on the mobile industry, right? They have more apps that do more things now because their phones can handle it. They couldn't when the first smart modes came out and just, again, going back to what we talked about earlier, I don't think we can stop technology progressing in order to be greener.

We need it to progress and be green at the same time.

Anne Currie: Yeah. Yeah. We need it all.

Sara Bergman: People are going to call us greedy.

Anne Currie: We want it all. So we're going to zip onto our final article link today, which is,

Is Sustainable Data Storage a Paradox? So it's a piece in TechRadar by Jon Howes about environmental challenges posed by cloud storage and AI, and the rising energy demands of data centers that are associated with AI and storage. So thoughts? Is green data storage possible? Sarah?

Sara Bergman: To quote you, it has to be. No, but I, okay. So there are many things here, but I think also he said that cloud is the least wasteful storage solution because it is highly optimized, blah, blah, blah, blah. So I don't know if the paradox is. Could there be other ways? Or if the paradox is, I don't know, it was a catchy title, and I'm not sure I understood what the paradox was.

But also, what I really kept thinking about, is storage really what we should be concerned about here? I (naively, maybe) would guess that compute would be much more resource intensive than the actual, like, store at rest. Then, of course, training the model, you would need to access the storage.

But again, I would be, in these days, more concerned about inference. And then, again, naively maybe, I would expect the CPU and GPU usage to be what we're worried about, not our disks.

Anne Currie: Yeah, I tend to agree. I would say with data, the opportunity there is there's quite a lot of low hanging fruit. And it's a really, good example of the shared responsibility model as well, that it's up to you, to us as users to not be wasteful in our use of data. And by waste, I don't mean storing more than we need, because I think that telling people to throw the stuff away in the world of AI is just crazy talk.

No one's going to do it. You're going to keep everything, right? We just need to see if the general AI can make use of it. The wastage is having it in a medium that's where it can be accessed more quickly than it ever needs to be accessed. So that the longer, if you say, look, I'm going to access this once a year, you can stick it all on tape and it's practically free in terms of carbon.

You can save as long as you like, there's this, that kind of like never underestimate the bandwidth of a station wagon full of tapes driving down the freeway. Tapes are pretty good. You can put a load of stuff on tapes and most data, especially data for AI, it doesn't matter if it takes you a couple of hours or even a couple of days to get it back.

You can, so you don't need to keep it on sSD, where it's just really much more carbon intensive than tape. So it's just, it's quite easy. You just need to think about it and not store things in the wrong medium. Use the tools that are best suited to the job.

Sara Bergman: yeah, be diligent about hot, warm, and cold storage. Like, what do you need, when, and how much? And yeah, also people say, "should I delete all my photos and emails?" I'm like, "no, you want them, right?" You're going to look at your photos again, you're going to maybe search or read your emails again. So don't remove them.

Maybe you know, unsubscribe to the email you never look at. That's like an option if you always get an email from your build pipeline, but you never look at it. Yeah, maybe don't need to get those emails even. Again, low hanging fruits exists. But yeah, storage mediums are not energy intensive if they're cold.

Anne Currie: Yes. Yeah. Don't keep it hot when it can be cold. Cold is the new hot. It's the new hot.

Sara Bergman: and the new green.

Anne Currie: And the new green, yeah. Right, so yeah, we've talked through all of the things. So is there anything else you want to say, coming out of what we've talked about today?

Sara Bergman: Oh, I don't think so. I think it was an interesting discussion. We are, what I love about this industry is that we're constantly on our learning curve. New things always come up, we have to adapt and adjust, and I get to put on my engineering thinking hat. And I love that. I think, we should be excited about that.

Like we have opportunities to be green, "ooh, I have opportunities to learn, opportunities to explore," not like, "ooh, this dreadful thing I have to go do." It can be fun. I mean, yeah, make it glamorous.

Anne Currie: Well, on that incredibly positive note, I will, sign us off. That's fantastic. So before I sign off today, just a reminder, all the links are in the show notes, so have a read through the articles yourself. And to tell you that this week, it is the Green Software Foundation Global Summit event.

So there are live events in quite a few cities. There's one on the first in London. There are ones in Munich and Hamburg, and there's one in Dublin, there's Berlin and Singapore. And, do feel free, I can't actually make, it's a real shame because I do really love to meet people in person.

And I think flying is one of those things where we, flying is an incredibly valuable thing. I don't tell people not to fly, but for me, the reason to not fly all the way over to London if you're in the US, for example, to go and meet people in person is, that you should set up your own local meetups, which are in person, which people can get to and from.

And because having people locally that you can talk with is incredibly useful. So set up your own summits if you can't make any of the summits that are out there.

Sara Bergman: Or your own meetup, if a summit seems a bit much, then set up your own GSF meetup.

Anne Currie: Yeah, that is an excellent idea. You'll meet a load of lovely, like-minded people, and you can chat away about saving the world, which is always a good thing to chat about. So one of the nice things about the green community is we're all aligned on trying to make positive change without turning everything off and going back to the stage.

So we've come to the end of our podcast and oh, it looks like Chris, our editor, has given me a final fun question for you, which is, if you were to design an AI that had zero environmental mental impact for a totally non-serious purpose, what would it do?

Sara Bergman: I'm embarrassed to say that I thought so much about this, because my mind went, ooh, what would I do? And then I'm like, would I optimize my life? Would I make something silly? Would I make something useful? And then I just, there were so many options that I don't know. But some of the things that like, popped up was like, organize all my photos.

I have a small child. I take a lot of photos of him. I would like for there to be some nice organization going on there. That would be very nice. Also, an AI that designed cute nails that I could do at home with like the stuff I have. Because I am not artsy, but I like fun nails. So yeah, like, like, if I am bad at doing my nails and I have four colors, what can I do that's cute?

Anne Currie: And of course, all of those things could be done zero impact because the, key thing about them is that they are not latency insensitive. You could say, well, actually, I'll wait to run my models and do my inference till the sun's shining and the wind's blowing and there's excess electricity on the grid.

The, I don't, I would, to make it so it's there, I wouldn't say, well, it's in the middle of the night and it's still night, but I'm really desperate to get my fancy nails done.

Sara Bergman: Yeah.

Anne Currie: Just wait. Just say, well, I'll, find out tomorrow morning what the AI is storming. It's all about making things less on demand, so we align with when the sun's shining and the wind's blowing.

Sara Bergman: Thanks for taking me back to technical.

Anne Currie: So thank you very much for being on. It's been, as always, delightful to talk to you.

Sara Bergman: Likewise. I had a blast. Thank you so much.

Anne Currie: It's nice. And we don't say, obviously, while we're writing the book, we talked together all the time, but now that the book's finished, then we talk together less often. So for all listeners today, links are in the show notes. And if you haven't read Building Green Software from O'Reilly, you are missing a trick because that really, it's a fun read.

Everybody seems to be enjoying it.

Sara Bergman: Yeah. I think the top thing people are like, "Oh, it's fun." I'm like, "Yeah, it's funny. It's good."

Anne Currie: So if you haven't read it, crazy! Go out and read it, or join one of my training courses, which are also fun. So thank you very much. And I'm sure we will both be back on the Environment Variables sometime in the not-too-distant future. So goodbye from me.

Sara Bergman: Bye.

Anne Currie: Cheerio.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode!

Electricity Maps

26 september 2024 | 50 min

Host Chris Adams sits down with CEO Olivier Corradi and Tech Lead Íngrid Munné Collado of Electricity Maps, a company that leverages data to enable decarbonization of electricity grids. They discuss the complexities of carbon intensity data, the role of accurate forecasting in renewable energy, and how this data helps optimize electricity usage for sustainability. Olivier explains the origins of Electricity Maps, their goal of providing real-time carbon footprint insights, and their theory of change—targeting citizens, corporations, and institutions to create a greener future. Tune in to hear about the history, frontiers, and future of this engaging field.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Olivier Corradi: LinkedIn | Website | X
Íngrid Munné Collado: LinkedIn | Website | X

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Our Road to Impact: How we contribute to fixing climate change [04:56]
[1812.06679] Real-Time Carbon Accounting Method for the European Electricity Markets [20:24]
SCI Specification Achieves ISO Standard Status | GSF [21:24]
Electricity Maps | Client Story: Monta [29:52]
How to save costs and emissions with a flexible electricity load? [33:32]
How the prices can change by location | X [34:28]
Another example in Texas | X
Marginal vs average: which one to use for real-time decisions? | Electricity Maps [39:21]
eCO₂grid | 50Hertz [41:58]
Investigating Apple's Clean Grid Forecast

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Olivier Corradi: The dream is to get to street-level granularity, and this is why flow tracing, by the way, is so important, because as you increase the resolution of what you're looking at, then obviously if I'm looking at just a city, most of the electricity is actually produced outside of the city, and that's also why, by the way, the whole forecasting challenge has just become massive. Try to forecast how clean the electricity is going to be at every location on the planet. Obviously, you need really good renewable forecasts as well. And that's what civilization needs if it wants to rely significantly on these renewables that are intermittent.

I'm your host, Chris Adams. Hello, and welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. We often talk about carbon intensity, or how clean electricity is on this podcast, as one of the levers for making software more sustainable.

To know this, you need to get the data from somewhere, in a form that's easy to understand and consume. So today we're sitting down with two people who live and breathe this data from Electricity Maps. Electricity Maps is a company at the forefront of enabling a data-driven decarbonization of electricity and today we're joined by both Íngrid and Olivier from the firm.

Hey folks, how's it going?

Íngrid Munné Collado: Hi.

Olivier Corradi: Hi, Chris.

Chris Adams: Cool. Folks, we'll talk about the origins of Electricity Maps in a second, but before I do, I just want to give you a bit of space to introduce the two of you, actually. We normally go by surname first, so I think Olivier Corradi, you're ahead of Íngrid, so can I give you some space to introduce yourself first, and then we'll do the same thing for Íngrid, alright?

Olivier Corradi: Of course, and thank you so much for the invitation and for inviting us to the show. So, my name is Olivier, I'm originally a machine learning engineer and data scientist. My background has been academic, I've worked as well in the research industry with IBM Research and Simulating Electricity Grids. And I would say the most relevant thing that I'm doing now is Electricity Maps that I founded eight years ago.

Chris Adams: Okay, thanks for that, Olivier. And Íngrid, over to you, I suppose.

Íngrid Munné Collado: Hi. Hi, Chris. I'm Íngrid. I'm the Tech Lead at Electricity Maps and I'm very happy to be here. More specifically, I'm part of the grid forecast team where we built, as we have said, forecast models for renewable energy sources and other signals such as carbon intensity. I've been recently involved in replacing and improving the weather forecast data that we use at Electricity Maps to achieve better forecasts.

And yeah, outside of work, when I'm not working, I'm either going to CrossFit where I'm trying to learn how to do a pull up, which I don't know, or knitting and just making sweaters. So quite, quite contrast.

Chris Adams: Cool, thank you for telling me. I'm glad you mentioned weather, actually. I mean, as an English person, we talk about weather the same way that we breathe, I suppose, and as I understand it, you folks are both calling from Denmark, right? We're a little bit chilly today than it otherwise would be, right?

Íngrid Munné Collado: Yeah, it is quite cold for September. I think it's quite usual, but it would be a bit on the colder side.

Chris Adams: Yeah, same here. We, it's, Berlin has, someone has switched off the summer, so we now had to, like, I've cycled in with my coat for the first time today, and did not enjoy that. Okay, so I should introduce myself, actually, folks, if you're new to this podcast. So my name is Chris. I work at the Green Web Foundation as the executive director there.

I also work in the Green Software Foundation as one of the co chairs of the policy working group where we Basically, work on policy to see about coming laws and things that we might actually want to respond to or help members understand what the implications of might actually be. The other thing I'll share with you is that we try our best to have quite helpful show notes for this.

So we're going to mention various projects and papers along, and if you are viewing this in Spotify or YouTube podcasts or something, you might not see them. So be sure to look up podcast. greensoftware.foundation to see the full links and the transcript. Okay, you two folks sitting comfortably?

Íngrid Munné Collado: Yeah.

Olivier Corradi: We are, sure.

Chris Adams: Brilliant, okay, then I'll begin.

Okay, first question I'll put to you, Olivier, because I've actually been a fan of Electricity Maps for quite a while, and when I first heard about Electricity Maps back in 2017, it wasn't Electricity Maps, it was electricitymap.org. And this is something that is reserved for non profit organizations. So it's 2024 now.

And I now know that the same kind of cool map that I started talking about at conferences and so on, is now available under electricitymaps.com. And as someone working in a non profit, I end up talking and thinking about theories of change all the time. So there was a really interesting post I saw on the blog, on the Electricity Maps blog post, particularly talking about this theory of inaction.

And it's rare to see start ups talk about this kind of stuff with a fluency, really. So, Olivier, can you tell me a little bit about basically, what this is, and how this informs you spending all the time on Electricity Maps, because I know there's a few other things that you've worked on before.

Olivier Corradi: Yeah, I'm happy that you actually give me the opportunity to articulate this. So in the early days when Electricity Map was started in 2016, there was this question that remained for a couple of years afterwards, which was, "should we actually be an NGO or should it be a for profit company?" And honestly, the conclusion is that it depends on how you want to impact the world and it depends on how you want to be funded and so on.

But the way that we've been looking at things is basically to say when a company will pay you for something you're delivering, that represents a change that is happening in the world that is sufficiently valuable that someone will pay you to do it. And people are taking you seriously when they're implementing it.

So this was a little bit what tipped it over towards more of the, let's say, for profit world. But in order to make sure that we never lost track of the actual impact we want to have, we started having like this framework that we stole from someone else, honestly, and adapted afterwards. But first of all the vision we have here is to imagine a world where we have low carbon electricity that's delivered everywhere across the world, every hour of the year.

And this was really why Electricity Maps was created, out of almost a frustration of folks looking at the electricity grid from a yearly perspective, instead of looking at it hour by hour. And of course it made sense before we had renewables. Now that we, the wind doesn't always blow and the sun doesn't always shine, we need that.

And the angle we're taking here is data-driven decarbonization, as you mentioned, because we're a mixture of folks who come from academia, who know how to build machine learning systems and tech. So this is our angle. So with that vision in mind, providing clean electricity every hour of the day, we realized there's three pressure points that we need to apply if we want to create that transformation.

The first one is, we identify them as citizens. Second one, corporations. And the third one, institutions. And there is this framework that, that depicts a triangle of inaction where each of these corners are pointing at the other as an excuse to not change. For example, citizens are saying, "well I actually would love to be greener, but the government is just not putting the right incentives for me to be cleaner," like the public infrastructure is not helping. Another piece of it is the citizens are pointing at corporations and saying, "well, I, too bad, like, flying is actually cheaper than taking the train. Like, I don't have a clean offering here."

And then the corporations on the other side will say, "well, but everyone wants to fly, right? I mean, the citizens are not ready for that change." They'll say, "well, I'll just keep operating like I am." Corporations will point at institutions as well and say, "well, the right tax schemes are not here for us to make flying cheaper, actually, so we'll just continue what we're doing."

And then institutions will be, "we don't have" like the, let's say, "social acceptance from the citizens to actually, exactly, to put these green tariffs. And on the other hand we're just a small organization and institution, sorry," and you have corporations like Microsoft, OpenAI, Google, that are just like, more powerful than governments.

And so you can basically map these out in a small triangle. And we said, okay, so how are we going to impact this? And then most of our initiatives can actually be mapped to this triangle, where on the citizen side, the app that we have of Electricity Maps is creating this awareness, creating the debates as well to make sure we have a factual understanding and can challenge our politicians in the right way.

There's the piece on corporations where we are working with them in order to ensure that green offerings can actually hit the ground, like that we have electric vehicles that can use electricity at the optimal time, all these things. And it's a win because we get a financial cut of this and we can grow our company and grow the impact.

And then finally, on the institutional side of things, we're trying to make sure that the right carbon accounting methodologies and so on are being pushed. We basically want to live in a world where we have an accounting that represents what's physically happening in the world. And one example is we have a data portal where our historical data can be used by any company who wants to do granular carbon accounting.

Again, moving away from this world where you're doing it on a yearly basis to an hourly basis. So this is the framework we've put in place and the way we're articulating internally as well how we allocate resources and prioritize and make sure that what we do leads to impact ultimately.

Chris Adams: Cool. Thank you for that. I have one question following up from this, because when I was doing a bit of research, I looked up this triangle of inaction, and I think I might need one of your help, one of your help in pronouncing the person's name. Is Pierre, is it Peyretou?

Olivier Corradi: Not too bad. Yeah. Well done.

Chris Adams: Okay, so for people who are listening, what we'll do is we'll share a link to some of the theory behind it.

He's actually got an online course. I think there is, there's a French school of business which does actually have some online resources to understand this kind of theory. So if you're Curious and you want to think about, okay, how does this work? Or if you'd like to see an actual triangle rather than have us describe it, follow the link and you'll see some of it there.

Okay, thanks for that, Olivier. Okay, so we've spoken a little bit about data, and you mentioned about data being in the open, Olivier, and I can actually confirm that when we spoke to Toby before, Toby is one, sorry, Tony is one of the colleagues at Literacy Maps. We did an interview with him around about November last year, actually, and he was talking all about this, Open Data Portal.

And back then we were like, "oh, sweet, there's all this data being published." And I'll be honest, you're a startup. I was like, "are they really going to publish the next year?" And then January came around and it actually did get published. I was like, "wow, this almost never happens." So I was really pleased to actually see a startup and a company follow through with making some of this data available because it's so, so, so useful and this is so hard to find otherwise.

I can speak as someone in a non profit who's been trying to find this data. It's so, so handy to have some of that available now. So yeah, thank you on that one. Okay, so the question, I'll leave some space because I think there was a response coming up there actually, Olivier.

Olivier Corradi: Oh, thank you so much for the shout out. You know, it's important for us as well to feel like what we're doing leads to direct impact and that we have folks that are demanding this data. And sometimes these feedback loops are not always present. So I appreciate you giving us the shout out.

Chris Adams: Yeah, so if anyone is listening, this is open data, so it's something that you're able to build on and we'll share a link to the actual website that makes it very clear how you can use this information. So if you're trying to build something and you're looking at historical stuff, it's totally there.

Have at it. Okay, so we spoke a little bit about carbon intensity at different parts of the world and how you need to think about it on an hourly basis rather than just an annual basis. And we might talk about why when you talk about green energy saying green energy with certificates that came from solar and then saying, using that to make usage at night be counted as green might be conceptually a little bit challenging, we might say.

So that's some of the stuff we spoke about. Now, the thing I want to ask a little bit about, and Íngrid, if I can hand this over to you. When you're working as an engineer and you start thinking about carbon intensity. It's when you use, say, APIs, you just see a single number that goes up and down.

And like, it's fun. And it's very easy to underestimate just how much work can go into this and all the complexities around this. Because as I understand it, it's more than just like, looking at what the power will look at looking at generation from a single plant. Íngrid, can I just give you a chance to actually talk about what goes into sharing some of these current and historical figures? Because I know there's some forecasting work, and we'll talk about that later, but we've just spoken about open data and some of the historical stuff, and some of the context there might be useful for people who are considering downloading some of this or messing around with it themselves.

Íngrid Munné Collado: Yeah, absolutely. And I would say that without this amazing work of collecting data, any forecast, like forecast would not be possible at Electricity Maps because we need this data to be able to produce our forecast. So let's take a look at how we collect data. This is currently done by one of our teams at Electricity Maps, the Grid Modeling and Methodology team.

And our work starts by first trying to get as much real-time data and historical data from, that is available publicly. And here we're talking about governmental institutions, transmission system operators, that for people without electrical engineering background, those are the folks that manage the electricity grid at high voltage, and that make sure that demand and production is actually matches at every single second.

So we try to collect this data and now we have data for 228 zones. And when we think about it, we might say, okay, we just collect CSVs and everything is standardized and nice, but the data is really messy. So we might get data from like in PDF files, in TXT files. We have sometimes, I can tell you a bit of an experience.

When I joined Electricity Maps, I had to build the parsers for Japan. And I'm originally from Spain, so I know a lot about electrical system in, in Spain, but when I had to dig into the Japanese electrical system, I didn't know that they have eight different organizations that collect data and each and single one of them is in a different format.

And sometimes we even had to do some image recognition to get the power of the nuclear power plants that they have in Japan. So data is very messy. It contains outliers. We have missing values, wrong values. Some production modes might be missing as well, so there's a lot of work that we have to do behind the scenes to make sure that we collect this data every hour or even in lower resolutions to make sure that we can use this data and process it.

So once we have this data in the raw format, then we process it to make sure that we don't have these outliers or missing values. And that we have a complete power breakdown. That means that if Spain, for example, has nuclear, gas, coal, and like in total eight production modes, that the data we get has eight production modes because otherwise the carbon intensity values that we might show on the app or on the API, they are not going to make any sense.

So once we have this done, then we can actually go into maybe one of the core pipelines we have at Electricity Maps. That is the flow tracing pipeline. And Tony did an amazing job in the previous episode where he explained how this works. But for those who don't know, the flow tracing pipeline makes sure that we are able to trace back the origin of electricity.

Because if I plug my laptop here in Denmark, the electricity that I'm consuming is not only the one that is being produced in Denmark, but that one that is being produced in Denmark all the exchanges that happen between Denmark and the neighboring countries. So, we have this pipeline that makes sure that we can take into account all the neighboring countries and what happens at every single hour of the day to make sure that we can know the exact mix at a given hour of the day.

That's actually when we get the origin of electricity, the power breakdown, and then we can translate this number into carbon intensity by using emission factors that we update recurrently and that this actually helps us understand one, what's the carbon footprint of one kilowatt hour that we consume at a given country.

So emission factors are different based on the source we use and based on the country we are at. And that's what you see on the map.

Chris Adams: Wow, okay, so that's, let me just run through some of that then. Just to make sure I'm understanding it. So you're essentially getting a bunch of data. You're having to do, clean some of it up. Like, that's quite common. And most of us might be used to like, working with maybe text or CSVs. But I think you said that you're essentially like doing OCR, like optical character recognition in some on like, gIFs or pictures rather than actually having to read an actual number. Okay, and then once you've got that, you've got an idea of what the generation might be, but then you then need to do a bit of, like, working out where data is, where energy is being traded across borders, essentially, because in some ways the grid does have all this stuff, you just can't look at the production, for example. Because, like, I mean, the UK has, like, new connectors to other countries all the time, and Germany uses loads and loads of France's nuclear power, for example.

So there's all the stuff like that, and then once you've got this idea that, okay, there's probably this much coming from these places, you then need to think about, okay, well, what is the carbon intensity of power from a coal fired power station, or a new coal fired power station versus an old one, and stuff like that.

So there's all that other depth as well. Okay, and that all goes into a single number.

Íngrid Munné Collado: Exactly.

Chris Adams: Alright, I am kind of, so, I see why people do this now, because when I've looked at this before, I've looked at numbers saying, "oh yeah, it's just like generation," but no, there is, generation and production are two totally different things, and you do need to take into account some of this if you want a meaningful number.

Íngrid Munné Collado: And if I can say something just to give you a rough number, a country like France, it has interconnectors like with eight different countries. So imagine if you only consider the production or generation in France without considering all the interconnectors around, you might get a completely different picture.

Chris Adams: Ah, okay, alright, thanks for providing extra context because, yeah, I can see how complicated this gets very quickly in that case. Alright, okay, one thing that you mentioned before was that there are all these complicated scrapers, and when I looked at Electricity Maps a few years ago, I was surprised by there's quite a lot of it which is open source, so like, don't believe me?

Look at the repo, for example. There's some of this out there so you can see just how messy the data might actually be or possibly contribute if there's a gap, right?

Íngrid Munné Collado: Yeah, actually, for example, when I had to work in the Japanese parsers, I got help from people in Japan. That they actually would point me to the right CSV file. Oh, and at some point the CSV changed the URL where it was stored. And it was not until someone, a contributor, helped me out and said, "Hey, check this link instead."

So we have, so all our parsers are open source and they are hosted on GitHub, on the contrib repo. And we are extremely happy about all the contributors that help us out, like be able to get this 228 zones that we have now, on the map.

Chris Adams: Okay, go yeah, Olivier, I was going to ask if you anything you wanted to add, because I believe you did a bit work on some of the academic underpinnings for some of this.

Olivier Corradi: I just wanted to add as well that if we tie it back a little bit to the triangle of inaction, what is happening here and why this is so exciting is that we're basically enabling some of the citizens to also act on climate change by being able to contribute with their unique expertise, which helps us.

There's no way Electricity Maps with, we're a bit more than 20 people now based in Copenhagen. There's no way we would have been able to cover the world if not for the help of all these wonderful contributors that have helped us all along, and it gives them also a way to contribute something meaningful where maybe before they didn't have that opportunity.

Chris Adams: Cool, and Olivier, we spoke a little bit about flow tracing, and I understand this is the you published a paper about this years and years ago, and that was one of the things to basically, the methodology that you're talking about, yes, there's some proprietary code, but the general approach that's being taken, it's in the public for people to understand, so they can challenge it and interrogate it and say, well, yeah, this is, I disagree with this thing and this is why I think this might need to be changed in the future, right?

Olivier Corradi: Absolutely and our philosophy has always been if there's something where we think we can move faster with the help of others, then let's open it up because then we can allow for contributions and so on. If it's something where we know that if we open it up, it's going to take us a lot of efforts to handle the contributions, then we don't open it up, and that's why some of the internal pipelines that Íngrid described are proprietary, because it's just faster for us to change a couple of things and not worry about what will happen if we open it up. But in general, we always try to be open, because we are trying to create a global consensus on how we account for things, so that can only be achieved through openness.

Chris Adams: Okay, you're singing from the same hymn sheet as us, like, we, the Green Software Foundation has a big thing about Open, the software carbon intensity metric is Open. If you want to pay to download it, you can buy it from the ISO for 63 Swiss francs. I don't know why you would, but that's available for people, and me, working at the Green Web Foundation, we use open as a lever, so we publish almost everything we can, either under open source or under open licenses. So, Íngrid, if I may, can I just come back to you, because we spoke a little bit about historical stuff, and you mentioned that forecasting is now a big thing, because the thing about renewable energy that we kind of alluded to before is that it changes over time. The sun goes up, sometimes the wind blows, storms move around, stuff like that.

So, can you just tell me a little bit about why this idea of forecasting is maybe more of a focus for you folks now and maybe explain a little bit about how, I guess, the sausage gets made and how that can be difficult.

Íngrid Munné Collado: Yeah, of course. So, as Olivier mentioned in the beginning, we want our data to be as actionable as possible. And now imagine that we are software developers and we run cloud jobs. And at the same time, we want to be aware of our carbon footprint. And we know that our cloud jobs might last five, six hours.

So, real-time data and historical data does not really help us achieve, like, be aware of what's the carbon footprint of our cloud job in the future, like, when is the right time to schedule this? And EV chargers might also think the same, like, when is the right time to charge? So, we saw that real-time and historical data would not help us achieve that.

Then we realized, okay, then we need to provide them with forecasts, and we know that we are providing a global API with data from all over the world for all bidding zones. Then, if we, let's go back to this software developer who wants to schedule a cloud job, because I think that's going to help the audience understand.

I can choose a data center in Sweden, but I can also choose a data center in Texas. So we can, we have the power of choosing which data center we want to run our cloud job. But at the same time, how do I know which one is the best? I don't have any idea. And at the same time, okay, but is it better to run my cloud job at 9 in the morning or 9 in the evening?

And then that's when we realized, okay, forecasts can actually solve that, but we need to provide it globally for all the zones in the world. So that's why we decided to focus on forecasting carbon intensity for the next 24 hours. That means, like, day-ahead. So we run the pipeline every hour and we provide forecast for the next 24 hours.

But then the next challenge came up. When we realized, "wait a second, is carbon intensity a metric that everybody can understand?" If I tell you that your cloud job used, I don't know, 200 grams of CO2, you don't, it's very difficult to relate. But if I tell you that if you schedule your cloud job at 9 am,

you're gonna run on 90 percent of renewable energy share. Instead of running 9 pm and then the renewable energy share is going to be 10%, it's very easy to understand and say, "oh, you know what, I'm going to do it at 9 in the morning because the renewable energy share is 90%." That's when we realized, okay, we can't just focus on carbon intensity forecasts, we need to do that with renewable energy forecasts and more specifically, wind and solar power forecasts.

Because that increases user engagement and the actionability of our forecasts.

Chris Adams: Ah, okay, that's quite a subtle change then, so, and I understand that because intuitively it's something that I just become so, kind of, you take it for granted, right? If someone is coming to this new, yes, explaining carbon intensity is conceptually quite complicated, but how much is running on clean energy is quite a bit easier to understand very quickly.

Íngrid Munné Collado: Exactly. if I can add on that, then when we said, let's start on renewable energy forecasts, it's a problem that has been not solved maybe because research evolves very fast, but it's a problem that has been laying around for years. I've worked on that field for many years. Because energy is traded on this day-ahead market, so people need this forecast to be able to trade energy in day-ahead markets or intra-day markets.

So when I started working at Electricity Maps on that field, I came from my previous job, and I came in, completely biased and thinking that this challenge would be easy to solve, because in my previous job, we had to do renewable energy forecast of specific assets. And then I would have the perfect setup for a machine learning engineer.

That means I knew the location of the wind turbines. I knew all the data regarding the wind turbines, like the blades, the installed capacity, if it was under maintenance or not. I also had access to multiple weather forecast data and I only had to build models for Denmark or the UK or Netherlands, so it was very focused, but when I joined Electricity Maps, the problem is completely different.

Here we focus on building a renewable energy forecast at country level or, bidding zone level without knowing the location of the assets, without knowing what's installed capacity, if the turbines are under maintenance or not, and that makes the problem very difficult because we know that renewable energy forecast is very linked to weather.

And if you don't know the exact location, then that's another challenge. So, I think in the grid forecasting, we did a really good job on finding a solution that generalizes well enough, and that allows us to generate this forecast for wind and solar for all zones in the world.

Chris Adams: Ah, okay, thanks for providing that extra context. So, I'm just going to check if I understand some of the terminology you used there. You said like a day-ahead market and an intraday market. So, so basically, as I understand it, if you maybe run a wind farm, for example, the day-ahead, you're going to say, "well, I reckon we can sell this many kilowatt hours or megawatt hours of power tomorrow," and that's what you'll make a bid in, and that's, there's consequences for either underbidding or overbidding for that kind of number, so that's why you'd care.

And the intraday is a bit like the kind of shorter term thing, so, you might make one big bit but then you might say there's a little bit of flexibility or you say well okay I, okay cool. And you mentioned this term bidding zone. Now bidding zone is a little bit like a country but it's not always a country.

So like America has all these different bidding zones because it has different grids and there's, that's the kind of, when you folks have mentioned the word zone that's kind of what you're referring to. It's not quite a country but it's more related to like, is there a kind of a unit of carbon intensity for a particular grid region, right?

Íngrid Munné Collado: Exactly. I mean, if we focus in Spain, for example, Spain, Portugal, France, Germany, a zone is the same as a country. But if we look at Sweden, it's split into four zones, into four bidding zones, and Norway into three. So if you check out our data, then you can compare between zone and country.

Chris Adams: Okay, and within a given country you can have radically different carbon intensity, and we might talk a little bit about some of that a little bit later then. Okay, cool.

Olivier Corradi: The dream is to get to street level granularity. And this is why flow tracing, by the way, is so important, because as you increase the resolution of what you're looking at, then obviously, if I'm looking at just a city, most of the electricity is actually produced outside of the city. And that's also why, by the way, the whole forecasting challenge just becomes massive.

Try to forecast how clean the electricity is going to be at every location on the planet. Obviously, you need really good renewable forecasts as well, and that's what civilization needs if it wants to rely significantly on these renewables that are intermittent.

Chris Adams: Wow, okay, I didn't realize street level was the dream for this, blimey. Okay, so we spoke a little bit about software and cloud jobs and stuff like that, but it's also worth just briefly touching, like, this is used outside of the cloud world. And I think one of you mentioned EVs as one of the examples here.

Could we just briefly touch on that? Because that might be one of the things which is concrete that lots of people might experience or might know someone who might have something which is, like, impacting them. Because yeah, EVs are becoming more popular now and it's probably one of the biggest new large uses of electricity in most people, in houses for example.

So yeah, Íngrid, maybe you could talk a little bit about that and then we can move on to some of the other questions.

Íngrid Munné Collado: Yeah, of course. So one of our clients is Monta. And we have a very great success story with them. Monta, for those who don't know them, is a global operating platform for EV. And one of the solutions is the smart charging feature, where they offer users to shift their charging according to the carbon intensity of the grid, the share of renewables, and that's by using our forecasts, or the price of the electricity grid.

And users are completely empowered to choose and to prioritize which signal they want to follow, if it's mostly price or low carbon or high share of renewables, and by doing that, those users took action in decarbonizing the grid. And the results are quite impressive because 70, 000 charging sessions were optimized for low carbon or high renewable share, and there was a 48 percent growth in user engagement, according to Monta, and that means that they optimized for low carbon charging, and in this process, 200 grams of CO2 were avoided for each charge on average.

So these are the numbers and this is the feedback that we're always willing to get because when we create, when we produce this forecast and we spend so much time building these forecast models is because we really want to know that people use our data and that they really use our data for this use cases.

So we were really happy to know that.

Chris Adams: Ah, thanks. Okay, so you mentioned one thing that I think was quite interesting. You mentioned, like, the cost being a thing that might change at different times of day and, broadly speaking, this is because, as I understand it, when energy is really when there's lots of green energy, renewable energy on the grid, it will be relatively cheap and somewhat green.

So, in the UK, for example, we have something like this Octopus Energy. I'm not in the UK, I'm in Germany, but I grew up in the UK. So, Octopus Energy is one company that's been doing a bunch of stuff like this about having agile and intelligent tariffs. So there's essentially a financial reason as well as a kind of basically an environmental reason for doing some of this.

And I think what I've heard in the UK, for example, I believe on an interview recently I heard just by doing some of this, essentially, when controlling some EVs, for example, I think the figure was something like 1.2 gigawatts of demand they had control over. Now that's basically the size of a nuclear power station in many cases, so that's like a significant amount of flexibility on the grid that would otherwise have to come from burning loads of fossil fuels.

Right? Okay, so maybe I can allow us to talk a little bit more about the cost thing because I know that when we talk about this, when we talk about green software, you can make an argument that yeah, you should do it because it's good for the planet, but there's also a real cost fact which comes into this that I think is actually growing and maybe this is something that I understand that you folks have been looking into as well as one of the ways to address more this triangle of inaction and align some of the incentives for more kind of grid complementary activity perhaps.

Íngrid Munné Collado: Yeah. So one of the issues we face is that we know that we want, we are implementing climate action and we want people to use our data for, in order to decarbonize the grid. But sometimes the entry barrier can be quite high when we just go there with, "hey, you have to save CO2." But sometimes it might be easier for users to understand that, "hey, if you look at the electricity prices, that you have also seen that they are not constant, you might be able to save money and CO2."

And that's very, that's a very good entry point as we mentioned in the Triangle of Inaction to increase user awareness because the main, for people it's very easy to understand price and how much they are going to pay at the end of the month and if they are going to save money and therefore it's also nice for us to get to them by I'm Price and then explain that, "hey, by doing that, you might also have an impact on CO2," but we have also a blog post where we mentioned that just optimizing for price doesn't mean that you're also saving CO2.

It's you need to implement some what we call co-optimization, where you want to either prioritize one or the other. And at the same time, maybe you might be able to save more CO2 if you optimize for higher renewable share, and at the same time, you can also save money by doing that. So that's why we think that price is a problem worth solving and that we also need to explore that area.

Chris Adams: Ah, I see. And if I understand it, and if we follow the kind of path from before, we spoke about, like, a lot of us might be used to just paying a single price for power, but the price might change depending on where you are geographically as well, and that has some implications too, like, this is one thing that I guess, this is probably the newer world we're moving into, in that, yes, there might be, like, night time tariffs, but it feels like there's a lot more dynamism, both temporally and spatially for some of this.

Íngrid Munné Collado: Absolutely. And we saw that when the war in Ukraine started, that as soon as we saw that there were less gas resources, prices skyrocketed. And for example, in Spain, where there was a price cap

The situation was solved also because Spain has other sources of getting natural gas. But countries like Denmark, Germany, they really struggled with that. And the prices really doubled for some months. And it's not only political conflicts. It's also availability of the power plants. I it was two weeks ago where Finland had some unexpected maintenance in one of the, of their, of the largest nuclear power plants.

And that cost their head prices to double. And also in Texas, we have seen extreme weather events that causes prices to change from

Chris Adams: More than double,

Íngrid Munné Collado: yeah, more than double from negative prices at some point because there, there's one area in Texas that has a lot of renewables and another area in Texas that doesn't rely on renewables and due to those extreme weather events, they have to turn off some of the power plants and they, then this affects the prices.

And something I want to mention, I'm an electrical engineering at heart. So I want to talk about the power of interconnectors, and we might think that just by increasing renewables, this is going to make prices to be cheaper, but the problem is not going to be solved, and it's not going to make prices cheaper if we don't have nice interconnectors that make these flows between countries possible.

That's why prices are just changing so much because there's so much happening and going on right now, both on installing new energy sources, weather events that we're seeing, unfortunately, due to climate change and also the lack of interconnections at the moment.

Chris Adams: Okay, so you mentioned two things that I think are interesting there. So one thing was this idea of, okay, the price can change massively. Like the Texas example, I think if we look at, like, the cost of power, right, it might be between 20 to maybe 60 US dollars per megawatt hour in Texas. For example, you're one, you mentioned, like, there's a bunch of wind in the panhandle in the kind of northwest, and then there's around Houston, there's loads and loads of demand where people use all the power, right?

And I think the figures I've heard were something in the region of negative two and a half thousand dollars at one point, and then 30 minutes drive away, the cost is three and a half thousand positive. So like, you got a almost four, five, thousand dollar swing in the pricing here. And one of the reasons is just because it's a bit like network connectivity.

You know, the pipe isn't big enough. And this is one thing that we have to kind of work around, essentially, thinking about this.

Íngrid Munné Collado: And the same issue is happening in Germany, and like, Germany doesn't have these price mechanisms that we have in the US, but in Germany, there are also, like, huge problems due to transmission capacity, because the, most of the solar power production is, takes place in one part of the country, while the demand is concentrated in the completely other opposite.

And they are now experiencing with curtailments and extreme negative prices because they don't have this transmission capacity. They also don't have storage capacity, and there are no market mechanisms that control how we increase this solar power production in the grid and how we just export this to other areas.

Chris Adams: Ah, okay. All right, thanks for the extra context. For people who are still with us, we'll share a link to the previous episode, because we spoke with chap Philipp Wiesner, who was building a bunch of this work to simulate these kind of grids to give an idea of what the pricing might actually be with different services.

And if you have data centers, how adding some storage might actually change the cost and the carbon of running various software services. So we'll add a link for that. Olivier, while you're here, I want to ask one question, if I may, about basically, optimising for carbon intensity, because that's what lots of engineers are kind of trained to do somewhat.

And I, inside the Green Software Foundation, there is a kind of standard called Software Carbon Intensity. It's the thing that you're supposed to, or that you might optimise for, essentially. And it basically lets you use two different ways of thinking about the carbon intensity of electricity.

And so one of these is called a average carbon intensity and another one is called marginal intensity and they are slightly different. And it can be quite counterintuitive when you're first coming to this because it can give you somewhat different answers or incentivize different kinds of actions.

For someone who's coming to this for the first time, can you provide a little bit of like background on how to navigate some of this and how to think about some of this? Because it's something that. I think a lot of people come to and they scratch their head quite a lot because it can be a bit confusing having two numbers which can suggest you do totally different things sometimes.

Olivier Corradi: Yeah, well, we might get into some of the weeds, so I'll try to keep the, sort of, the discussion a little bit high level so we don't get too technical. But I think the way that I'm trying to explain this, so, on the average side of things, the first thing that I will say, actually, is that the word average can be a little bit misleading, because it sounds like, like you're taking an average over a period of time, which you're not.

Actually, what we're doing here is computing what we call the flow trace signal. So it's like taking the production locally, looking at what's imported, and then concluding on what is the constitution of the electricity that I'm getting at. So it's a representation of all the power plants on the system, basically.

So that's what we'll refer to as average. And on the marginal side of things, the story goes that if you're plugging an electric vehicle to charge at a particular time, then it's not all the power plants in that system that will ramp up to give you that additional electricity. It's the one that's called the marginal.

And a loose way to define it is to say it's the cheapest power plant that still has capacity to ramp up, to produce more. So, in theory, that marginal concept makes a lot of sense, because we're saying that seems to reflect what is physically happening on the system. If you actually go a little bit more into the details, and that's where the differences start to pop up quite starkly.

Well, if you are plugging, for example, your iPhone on this, on, to charge right now. Then it's not this marginal power plant that will ramp up. In all likelihood, nothing will happen on the grid. It's just the frequency of the system that will change a little bit. But you're not shoveling a little bit more coal in the coal power plant to burn off a little bit more, right?

So, emission factor would be zero. Like, no impact on the grid, you could say. If you go the other way, and you say, now I plug, imagine a metaphorical plug on a data center that's using like 90 percent of the consumption in the grid, Then you can't have only that small power plant that has spare capacity, it's just not going to be enough to ramp up this data center.

You would need all the power plants on the grid to ramp up in this what if scenario where the data center didn't exist. And then the last example is, if you have an electric vehicle and you're plugging it in, then the electricity will be delivered instantaneously to you. And that's just not a change that a coal power plant or a gas turbine can react on.

And that starts to create some additional complexity when you think about, okay, but can this be predicted? Can the plugging in of my electric vehicle, could that have been predicted by the market? And if it's already predicted, then it's part of the business as usual scenario. It's not marginal anymore.

So when you actually try to go down to all the details, it becomes hyper complicated. And we've tried really hard to talk to all the power system engineers, the electrical engineers and the TSOs in Europe as well. And one of the most fruitful discussions we had was with 50 Hertz that concluded that they don't think you can actually reliably identify or verify what the marginal power plant is because depending on what market you're looking at, if it's day-ahead, intraday, real-time, depending on the magnitude of the changes, depending on all the interconnectivity, the marginal power plant is just a concept that philosophically makes sense, but from a data perspective, it's just hard to measure, and that's also, I believe, why most of the regulations recently on the hydrogen regulation in both the US and in Europe that is documenting what is it, what signal should you be using, whatwhen you want to prove that your hydrogen is clean, then they are settling on an average signal. So if you take a step back from all of this, you're having an argument that philosophically for me makes sense. Of course, you want to make sure that the short term impact of what you're doing is minimized.

When you start looking at the data, you have a signal that is difficult to audit. And we've been working on this for eight years now, six years, sorry, creating a marginal signal and trying to verify it. And I've just seen enough that it can be manipulated in many ways. And that's why it's a little bit difficult.

So to get back and conclude a little bit and to answer your first question, which is how would we navigate this thing? I think the advice we're generally giving is get both signals, plot the data next to each other, and depend on the use cases, ask yourself "Is user acceptance important?" Are you going to show this in an interface?

Because if you are going to show it in the dashboard of an electric vehicle, in a country like France, which is majority nuclear, or in Ontario, lots of decarbonized electricity already, it's going to be a tough sell to tell them, look, your EV is being charged on gas, which is on the margin. So this user acceptance is important.

The second piece is auditability. Are you thinking of being generally directionally correct? Or do you need an auditor to prove that the data is correct from a scope 2, scope 3 perspective? And these are typically things they're going to inform depending on if you're willing to have something that is philosophically more accurate, but in practice more wobbly or if you want to have something that's just simple to explain and sort of abides by the regulation.

Chris Adams: Ah, okay, all right, so if I try to kind of summarize it, it sounds like the marginal thing it's conceptually attractive and kind of fun and it might give me some and in very ways it's basically gives me this idea that I can make relatively small changes rather than some of the systemic changes that might be needed for some of this.

But from an audit, from an auditing point of view, because you're comparing some of this to essentially a counterfactual which might not exist, it's actually very difficult to say, "well, yeah, I definitely made this impact and if it weren't for me doing this these people wouldn't have switched this stuff on," for example.

So you kind of need to have a degree of kind of clairvoyant level of information for this to actually really check this. Okay, cool. Speaking of clairvoyant, I've just realized the reason I've used this term is it makes me think of, there is a paper by a research student, I think her name, Tammy, I'm so sorry if I pronounce her name incorrectly, Thanathorn Sukprasert, she wrote all about using the SCI and using these different signals and how they, when they agree and when they do not agree, and if you are curious about this as someone listening to this, we'll share a link to that paper because it's a really good paper, she's also doing another one which is all about, okay, what are the possible savings from carbon-aware computing?

And that is also a really fun paper to read, and maybe, Tammy, if you're free, we'd love to have you on the podcast to talk about some of that, because there's quite a lot of fun to read there. Okay, cool. Thank you for that, Olivier. I realize that was quite detailed, but it, I appreciate you talking about the fact that in other sectors, there's stuff we can look at, because, as I understand it, this whole shift towards clean or green hydrogen is somewhat comparable to data centers in the fact that you have a very concentrated amount of energy being drawn in one place. So, in the same way a data center might be tens or hundreds of megawatts, you might see something similar with, like, creating hydrogen.

Is that the idea behind some of that?

Olivier Corradi: Absolutely, and if if the way that I'm thinking about this if you really want to simplify it, if we take a step back, the opportunity we have ahead of us is large, abundant renewables that are going to be the cheapest and the fastest way for us to expand the system. I mean, nuclear is great as well, and hydro and so on, but it's just, it takes more time, and it's a little bit more expensive when you put it in directly.

And so, if we put all this renewable in the system, we better well make sure that every flexible appliance out there is aligning their consumption to the time at which the renewables are creating that electricity, because else we're just hindering their deployment. You know, it's less batteries that we need.

And I think that's generally also the thinking behind the hydrogen. We want to make sure we're electrolyzing the hydrogen at the times where all these abundant renewables are producing electricity. And so that's the simplest heuristic and the fastest way for me to explain the systemic change we're undergoing and also why we're focusing on these renewable power forecasts.

There's a lot of value there. And by the way, why they also typically align with the price of electricity.

Chris Adams: Ah, okay, thank you. I do hope we still have some people with us, because that was a bit of a deep dive. But as sometimes when we're coming to new fields, working in software engineering, you do need to kind of engage with the details sometimes. Folks, I think we're coming up to the time that we had allocated for this.

And this has been loads and loads of fun. I've really enjoyed this. If people do want to know more about what either of you are doing, can I just give you a bit of space to talk about, like, Follow me on either Twitter, LinkedIn, Mastodon, whatever, like yourselves personally. And then if there's anything you would direct people's attention to, yeah please do, and then I think we'll wrap up actually, so maybe Íngrid if I hand over to you and then you Olivier, we'll wrap up with you okay?

Íngrid Munné Collado: Absolutely, so if people want to know what I do, they can check my LinkedIn, and it's Íngrid Munne, M U N N E, and I'm also on X or Twitter with the same name. So looking forward to that.

Chris Adams: Thank you, Íngrid. And Olivier, for you.

Olivier Corradi: Likewise, like Íngrid, you can reach me on LinkedIn and Twitter. If you search for my name, you'll find me. And I think in terms of resources to watch, we try to publish blog posts that are going deep into the topic, that are thoughtful. We don't publish a lot, but when we do, it's like We, we at least try to have serious research there, so check out our website, check out our blog, and we have a couple of guides on carbon accounting as well if you want to go deeper on the topic, and a few videos on YouTube if you want to nerd out more on some of the things we just discussed.

Chris Adams: Okay, we'll add a few links for all of those and the thing I'll also share is that in the Green Software Foundation, there are ongoing discussions in the standards working groups to discuss all this stuff. So you can see all the kind of back and forth around this, so if you do want to engage with this and possibly join to actually take part in that conversation, there is that available.

So you can see what discussions have come before and how people arrive at deciding which carbon signal to be following. Alright, I think that takes us to our time. Folks, I really enjoyed this trip, and we'll make sure that everything, or as many things as we remember, are in the show notes for people who want to continue this quest to learn more about building more sustainable software. Thanks a lot, folks, and have a lovely time in Denmark. Ta ra!

Íngrid Munné Collado: Thank you. Bye!

Olivier Corradi: Bye-bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

The Week in Green Software: Obscuring AI's Real Carbon Footprint

19 september 2024 | 44 min

Host Chris Adams is joined by Asim Hussain to dive into The Week in Green Software, exploring the environmental impacts of artificial intelligence and how the growing adoption of AI technology affects carbon emissions, as well as the growing complexities in the measurement and reduction of these. They discuss a primer on AI's direct environmental footprint, regulatory trends in Europe and the US, and the complexities surrounding the renewable energy credits tech companies use to offset emissions. The conversation touches on real-time cloud data initiatives, carbon accounting in AI, and the future challenges of balancing sustainability with technological innovation.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Asim Hussain: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

The Environmental Impacts of AI -- Primer | Hugging Face[03:12]
How Tech Companies Are Obscuring AI's Real Carbon Footprint | Bloomberg [22:25]
AI analysed 1,500 policies to cut emissions. These ones worked | Nature [32:48]

Events:

Green IO Conference 2024 London Dates and City Locations - Sustainable Tech Partner for Green IT Service Providers

Resources:

Does the EU AI Act really call for tracking inference as well as training in AI models? | Chris Adams [12:21]
Simon Willison on openai [14:15]
EnergyStarAI (AI Energy Star Project) | Hugging Face [16:12]
Meta-Llama-3.1-8B-Instruct · Hugging Face [21:28]
Jevons paradox and greening software—why increasing efficiency makes sense | ASIM.DEV [21:51]
Olivier Corradi [27:43]
Real-Time Cloud | GSF [28:41]
GitHub - Green-Software-Foundation/real-time-cloud
Reviewing the evidence we accept for Green hosting verification | Green Web Foundation [31:06]
The Week in Green Software: Modeling Carbon Aware Software | TWiGS with Iegor Riepin [37:18]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Asim Hussain: Three, four years ago, everybody treated all carbon offsets the same. They didn't realize there was nuance between them. Now that's changed. Everybody needs to now pay attention to the same thing in terms of renewable energy. If you do not pay attention to the fact that there is a lot of variability in a lot of this stuff, it's all going to get tarnished with the same brush in the future and any renewable energy claim is not going to be trusted.

I'm your host, Chris Adams.

Hello, and welcome to the Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. This is our news roundup show in the Green Software Foundation podcast. So we aren't doing a domain expert news roundup. Deep dive where we go into a deep narrow subject, but rather we're taking a more broader view.

So we'll try to add some context and commentary to the stories that have been shared with us that we discuss with our guests. With me today is my friend, colleague and mushroom enthusiast Asim Hussain of the Green Software Foundation. Asim, it's really good to see you again. How are your holidays?

Asim Hussain: Yeah, it's been quite a long time off and just gently dipping my toes back into the swing of things. Glad to be on the show. It's a nice gentle introduction back into the world of green software. So, glad to be here again, Chris. Asim Hussain, I'm the executive director of the Green Software Foundation.

And yeah, I spend my day probably similar to you thinking about how to advocate for green software. What do we need? What do we want? What are the questions that need to be answered and what are the levers that we need to pull to get action taken in this space?

Chris Adams: Reduce the environmental impact of digital services. Yeah. Okay, cool. Thanks. I should introduce myself as well. Hello, folks. My name is Chris Adams. I am the executive director of the Green Web Foundation. That's a smaller, a different organization. We are a Dutch non profit focused around reaching a fossil free internet by 2030.

And we do that using open as a lever. So we do loads and loads of stuff with open source, open culture, and things like that. As a quick reminder, we're going to share all the stories that we have and any projects or things that come up, we'll add to the show notes. So if you want to continue your quest to learn more about green software and how to reduce the environmental impact of digital services yourself, just look at podcast.greensoftware.foundation to see this. If you're looking in Spotify or some of the other podcast tools, you might not be able to see the links. So please do go to the website to see the show notes and you will be rewarded with diligently prepared links and helpful notes, and a transcript as well.

Alright then, so, I guess, Asim, I've introduced our News Roundup format, we kind of know what we're going to do, we've done this a few times, I assume you're sitting comfortably, shall we begin?

Asim Hussain: Yeah. Yes. Yeah. Let's go for it.

Chris Adams: Alright, okay, so the first story I see here is actually a post from, from Sasha Luccioni, Bruna Trevelin, and Margaret Mitchell, Hugging Face the Environmental Impacts of AI - A Primer.

So, this is one thing we had shared, and, Asim, I'm going to ask you, you've had a chance to look at this, what made you think this was actually worth discussing, and what would you draw people's attention to if they'd heard about this? And how would you, like, persuade people this is worth a read, for example?

Asim Hussain: Well, I suppose if you're new to the space and I think there's a lot of people out there who, for whom are kind of surprised to find out that AI has an environmental impact. So this is, I mean, A lot of this stuff, obviously we've been talking about different components of it over the last couple of years, but I think it's a really good, it's actually a great summary of the different components of what makes environmental impacts of software.

It's also got information there about what are some of the legislation coming down the pipeline? What are the, some of the actions that you can do? And some some things there. So I think it's a really good kind of primer for people. I think it's the title, is the title Primer? It

Chris Adams: Yes,

Asim Hussain: Yeah. So it does what it says on the tin. And I think it's probably could do as a really great introductory piece of information. It's got some great links there as well.

Chris Adams: Yeah, the thing that it might be worth just focusing on briefly, or one thing that leapt out at me when I looked at this is that it talks about the direct environmental impacts of AI specifically. So rather than talking about AI for good, or like "isn't it great that you could use AI to, say, make it easier to deploy renewable energy, or do this, or do that?" Right?

They're talking about, "no, there's an environmental impact that you still need to address regardless of whether you use something for good or for bad," and it seems to be focusing primarily on that kind of stuff. So as a responsible engineer, these are the things to think about. These are the kind of, what you might call an impact criteria, like there's carbon emissions, there's water, there's other things like that.

And generally this is from one of the kind of most trusted hands-on group who are like at the coalface for all this stuff. I mean, maybe, is coalface the correct term?

Asim Hussain: Yeah. I always use that term as well. And then I'm like, actually, that's a

Chris Adams: Maybe if you're in, maybe if you're in US AWS East, it's coalface and then if it's on AWS west, it's hydro face.

I dunno.

Asim Hussain: I use coalface and then if I don't, then I sometimes use front lines and I'm like, "actually, that's not a good term as well." We need a term which doesn't have a war Metaphor

Chris Adams: or industrial revolution connotations

Asim Hussain: Metaphor.

But anyway, it's at the cutting edge!

Chris Adams: Yes.

Asim Hussain: Cutting edge. There you go.

Chris Adams: Okay. That, that, that will do for me. Alright then. Okay. That's the first thing. I think on a following from this, when I was reading this. I quite like that it actually touched on some of the regulatory drivers that you have, because it's very common for people to talk about the AI Act, because that's probably the first piece of legislation, but it also calls out stuff taking place in Spain, and taking place in the US, and it shows that there's a kind of growing, I guess, regulatory trend to basically say, "well, If you're going to have this piece of technology in society, then we need to have a data informed discussion about what impacts it might actually have."

So we can talk about, okay, where is it responsible to deploy this? But also just like, okay, how do we actually mitigate this? Because, I don't know, it seemed like cars are useful. And the fact that cars are useful doesn't mean that we don't talk about seatbelts, right? You still have to talk about them being safer, regardless of how useful they are, basically.

So it's really nice to have a trusted organization sharing some information like this in a relatively roundabout in a, in my view, quite concise fashion. But if you look at the set of footnotes, wow, there's so much stuff that you can dive into if you wanted to kind of go down that rabbit hole, basically.

Asim Hussain: Yeah it's a very well researched, almost a state of the art paper. And also say like, I think it's also good to know, because people don't, yes, there's regulation coming down the pipeline, and some of that stuff is more mature than other regulation. But I think when you're working inside large enterprise organizations, this is the kind of stuff that gets people to pay attention.

You can be talking for ages about, "hey, look, there's the carbon impact, it's having a... we should be looking at our AI usage" and sometimes that can land and sometimes that can't, but a regulation or the threat of a future regulation is something that I've seen personally open a lot of doors. It doesn't kind of complete the internal sale of, "we need to invest in mitigating the impacts of our AI use." But it's certainly, I've seen it kind of open up a lot more doors because regulation is something that a lot of organizations pay significant attention to, and it's also something that they will, Invest time before the regulation comes out to look at it and put effort into it.

So I, I look at that, right. And it's kind of, it's great to see those regulations that I wasn't aware of. So I'll be obviously using this as it's a great primer for me as well, but this is a really good way of capturing people's attention. So then you can have that more refined conversation about, well, how are you using your AI?

Let's have a conversation about it. How is it, how is future regulation going to affect you? So it's a really good way of opening that door. And if you're inside an organization and you are a little bit concerned about the consumption of AI that you're having, I think for me personally, like pointing out the regulation that's coming down the pipeline does open a lot of doors, enables some conversations with leaders.

Chris Adams: This actually might be a kind of somewhat appropriate time to mention some of the kind of policy stuff we're doing, because I, so in addition to doing the podcast here, I help co-chair the policy working group. And we've, I think we've, we, we're likely to be getting quite a I'm going to put a policy radar out to see precisely this kind of stuff coming up.

Because, yeah, if you didn't know, I mean, okay, today is the 6th of September, and, you've seen this whole energy efficiency directive thing in Europe, right? So, in nine days time, every datacenter that uses more than 500 kilowatts of power draw has to start reporting and posting in public all of their absolute energy use, absolute water use, the amount of energy coming from renewable energy, how much of it comes from the kind of credits that you might buy which are unbundled, how much is coming from a power purchase agreement, so the kind of green energy that you've purchased directly.

There's all this new stuff. There are some caveats around this. So not every single organization will have to, we'll probably publish, but we basically have a regulation that's saying, "look, this has to happen now in nine days at the time of this recording. So when this goes out, it probably will have already happened.

And like this gives you an idea that if you didn't know this is happening, then you probably will, we do need to know this because this is written into the law in lots and lots of countries now. Well, all across Europe, for example, and I suspect this is the kind of thing we might see in other parts of the world because when you look at the figures and look at the data that people are currently basing policy on, it's really hard to figure out what the environmental impact of, say, data centers might be or what the growth is going to be.

And if you want to plan for a grid or plan for hitting some climate targets, this is the kind of stuff you need to actually be knowing about.

Asim Hussain: But it's also useful because we've been having these conversations for a while about, I think we spoke on the podcast a couple of times in the past and when we were developing the SCI specification, it came up a lot like do you include the data? How much of the data center do you include? But the biggest problem was, is that you don't know.

If you decided to put into the specification, you've got to include certain, the concrete or whatever it is that goes into data center. If that data is not public, then what's the point of putting into specification? That's why these regulations, that specific regulation is so interesting. I'm interested to see what actually happens in nine days time and the quality of data that comes out.

The conversations I've had in the past, because when this first started being discussed, I was chatting to a lot of, not data center operators, but people that worked with data center operators. And obviously it kicked up a storm and everybody's like, well, I need this data. How do I get this data?

What is the minimum level of information I can provide to like meet? And that's where it gets really interesting for me. What is the minimum level of information I need to provide to meet the regulation? And I think in nine days time, we're going to find out what is the minimum level of information that people have figured out that they

Chris Adams: Yeah, they can get away with,

Asim Hussain: that they can get away with? Because when I had that conversations, I don't know where it landed now, to be honest with you, but they were like, is it at the building level? Is it at the rack level? You know, it's, and it was like, it's at the building level is where I was left at. So I think the more and more this regulation comes along and it kind of surfaces this data to us, then we can then use that data to make more informed choices, hopefully not from a consumer level.

I think it should be from a, not from an end user level, but from the people who use data centers and make make different choices.

Chris Adams: I think you're right. Okay, what we'll do, we'll share some, a couple of links to this, because this is something we've discussed in a few places, there's one or two working groups where this has come up, in particular, because there's also, on the, just, just before we move on from this, there's a whole, there's a current kind of, in my view, an interesting discussion going on about, okay, with this, in Europe at least, with this AI act, yes, it says that you need to talk, you need to disclose the training data, the energy used for training a model, right?

But it's not totally clear if you need to also track the inference, right? So if you think about the training part, and I've shared a link to a blog post where I've basically highlighted the bits of the law that make me think that you might need to track inference, or at least disclose some information about likely inference because you can think of like the training part as like the energy going into making a car and then the inference figures as a bit like the car's mile per gallon for that model, for example. And well yeah, well it's not totally clear yet and we've seen the law passed and we'll figure out yeah.

Asim Hussain: Cause I assume the only, I'll be honest with you. I assumed included inference. The only time I did it was when I just like read your article and I was like, oh, hang on. That's and that's where these things get very interesting to me. I mean when I was in working in enterprise like organizations and I've one of the things that was always interesting to me was whenever I asked questions or got meetings together, like "let's talk about, we've got some questions about the, how do we calculate this figure to meet with this specification?" And there's almost always legal got involved and lawyers got involved. And I was always kind of, I'm like, "I don't need to speak to a lawyer. I need to speak to an engineer."

Why am I speaking to a lawyer? Because it's all about, "let me read this text. What can we interpret from this text? What do we need to give?" So I assume just because everybody just, we know inference is where most of the emissions are these days, I just assumed it was that, but you've now actually read the text and gone, the text has, is interpretable.

Chris Adams: Well, yeah, because you think about like when this initiative was written ages ago, like a few years ago, it's gone through this massive kind of gestation process, right? And a couple of years ago, when we hadn't really got to this point where AI is being deployed in the same way, like, was it November 2023?

That was like 100 million users. OpenAI had gone from zero to 100 million users in five months. And maybe last week? We should share a link to Simon Willison's blog post because he wrote about this quite eloquently. He's like, well, OpenAI have just mentioned that they're now at 200 million weekly users. So that's like doubled in a single year. So we've gone from, so inference is now a significant part of the story in a way that it wasn't previously, basically. So it may be that the law, when it was written two years ago and began that process through it, it might not have been such a concern. And this is the thing that we're, this is why it seems a bit unclear, and I think we'll probably end up with a test case that will set a precedent for people to figure out what they should be sharing or what you might need to share if you're building new foundational models in future.

Asim Hussain: I mean, I can tell you that, that preChatGPT kind of announcing, it was very well known that inference was significant, as in way more than training for any, like, I can't reveal too much, but you know, it was known very much that that was the case and a lot of effort had been put into mitigating, not from a carbon emissions perspective, just from a cost and energy, but just all of that stuff.

So it's kind of known, it might not have been in the zeitgeist, it might not have been in the kind of the public discourse because it's so much easier to talk about this big training runner. Maybe there's just more public data about that because inference in a way, if you think about it, is going to be pretty private.

Because that's inference is basically telling everybody the business end of your where you're making money from, and they'll probably keep that pretty private. So yeah, maybe it just wasn't well known, but it was true and well known, I think, to anybody in the kind of the AI space that inference was a pretty big deal prior to this.

But yeah, it makes sense. Yeah, these acts take a long time. So yeah, a couple of years ago, all we were talking about was training. It was a good, it was a good headline to discuss. Yeah.

Chris Adams: Useful insight from the inside tracker team, alright. So there's one thing you mentioned actually, we're just on the subject of inference. There is this, in my view, really interesting project right now, EnergyStarAI, which is a project which is, you see a few names associated with, so, Sasha Luccioni, Sara Hooker, Régis Pierrard, Emma Strubel, Yacine Jernite, I think, Carole-Jean Wu, and one of our own at the Green Software Foundation, Boris Gamazaychikov.

He's at Salesforce and he's been one of the people who's writing publicly about a bunch of this stuff and also about like, quite, the small models as well as large models. And it's, I was really pleased to see his name actually. So. Hi Boris, if you're listening to this. This is a really good story to look at, because this essentially is talking about inference and saying, "well, let's find ways to make this visible for people" two years later now, basically, and say, "well, let's see if we can find ways to introduce some of the incentives to go for more efficient inference," the same way we've done with Energy Star in other kind of industries, for example.

Asim Hussain: There's really interesting, I'll try and get her on the podcast, actually, she's from IBM, her name's slipping me, so I'm not as good as names with you, you can just rattle them off, but I'll make sure to put it in. Because yeah, there's Energy Star for AI, but we're also, there's conversations inside the foundation as well now, kind of looking at SCI and how do you apply SCI to AI and kind of, there's a lot of overlap with a lot of this work as well.

But what's interesting is there's this real question about what to do when it comes to inference and training. Like if you were to report that. How do you report that for a model? And the point that was raised, and I thought it was so, because I never thought of it before, which is, if you've got a foundational model, you've done the training, you've done a big training run for a foundational model, and you're now then running inference on that, when you report, let's say, I don't know how any, I think Energy Star is just going to be like a good bad kind of label, where SCI is more like a score, like a carbon per prompt or something like that.

How do you apportion the training into the, do you include the training every time you call it? Because if you do, there's a really interesting thing that happens, which is these foundational models, like if you're using an open source model and you just, that costs 10 000 to run. Do you include that in your AI solution and then just say like, "Oh, I'm three, three grammars per prompt?"

Then somebody else uses that foundational model. Do you then divide that by two and say, well, now you're, you

Chris Adams: talking about double counting kind of question, right?

Asim Hussain: And then like, if that is how the measurement eventually lands, then if you're an unscrupulous organization, all that you would do is try and get as many people to use this foundational model as possible to then dilute your numbers.

And so I think one of the, one of my little bugbears, and it comes up quite often, is the assumption that something that works in the physical realm will work in the digital realm. And one of the things I try and educate people as much as possible is that stop trying to take something that has worked in the physical realm and apply it to digital because there's so many ways it just doesn't work.

If you're thinking about training like a scope 3 embodied carbon physical device thing. You can't divide a chip in two and I give you half and you, it's like, that chip's yours, but you can do that in the digital realm. So there's this whole supply chain accounting aspects of digital emissions, which, it just needs to be thought of differently in the world that we're in. And if you don't think about it differently, you can then have, I call them unintended, we used to call them gaming, like when people were developing the SCI, like one of the conversations we'd have on the calls is, how is somebody going to game?

How is somebody going to take SCI,

Chris Adams: Carbon Intensity here, right? Yeah, okay.

Asim Hussain: how is somebody going to game the side? That's kind of a lot of the conversations we had at the start. Cause obviously everybody was like, we want to make sure if we develop a standard, people aren't going to then

Chris Adams: Abuse it,

Asim Hussain: it. And therefore the standard has no respect in the world.

And so like a lot of how, like a lot of how I kind of work with the standards projects here is I'm a bit of an annoying devil's advocate. I love it actually because I kind of walk in and go "here are 10 ways I can hack this standard to present a better score without actually doing anything." And so I think that's some of the things we need to think of as we think of SCI for AI, as you think of Energy Star, as you think of these other things is yes, there's this happy path that everybody's a good actor,

it will work and it will give you the right signal, but we need to think about the non-happy part, where people might not even necessarily be bad actors. It's just death by a thousand cuts. You're working in an organization, you've got a deadline, you've got a bonus you have to meet, there's a customer that you're going to have to get or you lose your business.

And so you're just death by a thousand cuts. So yeah, we have to be very careful as we explore like SCI for AI and Energy Star and anything really in this space, which is talking about measuring emissions. Because if you don't think through those unintended consequences, that's a problem. And that's one of the ones I have is like, is if you're including training, how do you apportion that?

You might not actually want to include training. You might actually want a separate. You actually, you might actually only want to measure the inference, because that gives a truer figure.

Chris Adams: So to cap this off, I'm going to, as we move to the next story, I'm going to link, share two links which might be useful for this. So the first is the link to the Meta Llama 3.1 8B, their model card. They literally say "the methodology used to determine energy use and greenhouse gas emissions can be found here."

They've linked to it and they said "since Meta is openly releasing these models, the training energy use and greenhouse gas emissions will not be incurred by others." So they're basically saying you don't need to count that part. That gives an example. We'll share a link to that. The other thing we'll share a link to for the show notes is actually the Asim, for this software, carbon, yeah, this is

Asim Hussain: Yeah, this is my unintended.

Chris Adams: This is a good example because this is where you've basically said, this is me, like, red teaming this approach, and these are the ways I can basically, in bad faith, try and engage with this example.

And this will probably be useful for people who are looking at this, to get an idea of, like, how some of these standards or some of these conventions are being developed. Alright, shall we move to the next story? Because it kind of does relate a little bit to, basically, tweaking numbers to present a view of the world, basically.

So this story is now, how tech companies are obscuring AI's real carbon footprint. This is a story from Bloomberg, I believe. Asim, do you want to introduce this one?

Asim Hussain: So yeah, so I thought, well, the reason this kind of popped up on my radar was, I forgot what I posted on LinkedIn, but I was posting, I started posting a bunch of information on LinkedIn about the use of, the use of RECs and the effectively like kind of renewable claims that organization makes and how it's in a really frustrating way, it actually puts us in what we're doing in kind of competition with this very important energy transition, because the argument I'm making is, look, you can either do two things and then we'll talk about AI.

Let's keep it to AI. You can either make your AI model more efficient, so it consumes less energy. Or, you can do absolutely nothing, and just buy offsets, energy offsets, RECs, whatever you want to call them, to mitigate, theoretically, your energy offsets, your energy consumption. And that's kind of like being this "Are we friends?

Are we not friends?" How do we like, we want to support the energy transition, but at the same time, like we really want to advocate for more energy efficiency. So, and I think one of the things we've spoken about is that there's, when you do make these renewable energy claims, like one of the things that you do with all types of offsets to kind of avoid a greenwashing claim, you have to have that additionality component to your offset, which for the audience means that how do you, if I'm saying this thing is offsetting your emissions.

What is it a litmus test to say that is a true statement and is basically, are you actually adding? So for a renewable energy credit, it's like, if you weren't about this renewable energy, would that thing have happened?

Chris Adams: So you're talking about the counterfactual here, right? So you're trying to compare something against. This is, you see a load of this in the hydrogen circle, in the hydrogen, in the world of hydrogen, because Just like datacenters, hydrogen electrolysis, like the electrolyzers use loads and loads of energy, right, and one way that you can do that is just by plugging them into the grid, right, and there's various people doing various things to say, well, I'm just going to buy a bunch of, say, renewable energy credits, right, and that's going to make that count as green, and there's, that's, in some ways, that's kind of somewhat problematic because, essentially,

Asim Hussain: coal to make hydrogen.

Chris Adams: Yeah, that's not exactly what all, you're, in many cases, you're a, you're burning coal to make hydrogen, so the actual net, it's a net loss in climate terms. But also, the, there's been a big fight in the kind of hydrogen world of, to have like this notion of three pillars. Where you basically, if you're going to have something, if you're going to count something as green hydrogen, then you need to be talking about new infrastructure being added to the grid to provide that new supply.

You can't just use, you can't just take from the existing stock of green supply and then count that as green. And this is one of the things that we've seen, like, I don't, Amazon made the news, I think a few weeks, a while, because they basically acquired a data center from a company called Talon, I believe, where they're right next to a nuclear power station, right?

So this, that you, there are some people saying, "oh, this is great, isn't it good that Amazon's using a bunch of clean power," but then you've got to think about, well, okay, who was that clean power going to before? Was it going to the grid? Like, there's a whole discussion there about this. Yeah, so there's a whole set of things to be talking about and this is why this is such a kind of fraught area, basically.

Asim Hussain: I mean, but I think the way to bring it back to something that people understand is when we talk about carbon offsets, I think now it's more understood that it's kind of like you, you have carbon removal offset. So you can plant a tree. Right? And then you planted the tree, that tree will grow, and there's issues there.

That tree will grow and also suck carbon from the atmosphere. And you can say that's a carbon credit of planting a tree. Or, there's carbon avoidance offsets. And there's many various, and there's actually very good variants of carbon avoidance offsets. But there is a variant of a carbon avoidance offset where I've got a tree.

And you pay me not to cut it down. And so where's the additionality? If I'm actually planting a tree, it's happening. I'm planting a tree. I'm adding additional kind of capacity in carbon removal. And in the renewable energy markets it's exactly the same. You can have renewable energy. Which if you buy means a renewable power plant is going to get built and you can have renewable energy which is just kind of sold and if you buy it or you don't buy it there's no change nothing's going to happen there's no more new renewable plants going to get built. Only one of them has that additionality component and so, therefore, only one of them should really be used in any kind of renewable energy claims.

But both of them are allowed in terms of renewable energy claims. So in terms of what this article is talking about, when they're saying "tech companies obscuring AI's real carbon footprint," they're actually talking about companies using what's called those unbundled RECs, which is those RECs which do not have that additionality component.

And then use buying them and then saying, "well, that's mitigating my environmental impact." And what the article is talking about is really, you should be looking at higher quality, Renewable Energy Credits, ones with more additionality components to it. And I think that's really interesting. There's actually also really, Olivier Corradi from, don't know if I'm pronouncing the second name correctly.

Electricity maps. Yeah, he, when I was talking about, he shared a really interesting article he'd written a year ago, which I thought was interesting as well. I'll share that here if you've got it. Yeah. He's actually advocating for like a more nuanced approach to looking at renewable energy in that there's additionality, then there's additionality, and then there's additionality.

There's like different levels of additionality. There's like, "this definitely 100 percent would never have been built unless you bought this renewable energy credit." And there's other ones like, "we may not have been able to build it, but we had some funding from here and there." So there's kind of different levels of additionality here as well, which I thought was really interesting also.

I'd never thought of additionality more than just a binary yes/no. And he was saying it's actually more of a score for a renewable energy credit. But

Chris Adams: Yeah, there's totally a continuum there. So the thing I might share for people who are looking for something actionable to work with here is basically the numbers that you often see reported by technology firms. There's all this, there's all this nuance hidden behind it. And there's one project called the Real-Time Cloud Project inside the Green Software Foundation, which essentially is a data set of the largest three providers.

So that's Amazon, Google, and Microsoft. And they've got the figures shown in both the kind of location-based figure, which is the closest thing you might think to, like, the physical location, the physical impact on the ground. They also talk about some of the market-based figures, which is what lots of firms like to use, like market-based on an annual basis.

But they also provide a few other details and a few other ways of talking about it, because some firms are now talking about hourly, basically hourly green energy versus annual green energy, with the idea being that you, it's a way to try and avoid making claims about saying, "I'm running a data center at night with certificates coming from a solar farm." This is inherently a little bit silly. So they address that stuff. So there's a, there's now, I think it's in the final stage of what's referred to as consistency review, where every member in the GSF is able to just say, "Hey, I object to this, or I'm not sure about this." And then, yeah, there'll be an open data data set for every single region from the three largest providers, which make up more than two thirds of the entire cloud market.

So you have some meaningful numbers that have come from the actual big providers themselves that you can actually, that we can work with.

Asim Hussain: And I think, like, I think basically my, I think the point I'm generally raising out, out there with another kind of, one of the reasons this article was very interesting to me, and especially the work that Realtime Cloud is actually interesting. Three, four years ago, everybody treated all carbon offsets the same.

They didn't realize there was nuance between them. Now that's changed. Everybody needs to now pay attention to the same thing in terms of renewable energy. If you do not pay attention to the fact that there is a lot of variability in a lot of this stuff, it's all going to get tarnished with the same brush in the future and any renewable energy claim is not going to be trusted.

So I was, I'm kind of a guiding and advising organizations to pay very close attention to kind of the type of renewable energy that you're buying. And be aware that because of podcasts like ourselves, there's generally, it's a Bloomberg article talking about this right now. It is now becoming very aware in the minds of a lot of people who care about this space, who listen to our podcasts, who are paying attention, that there is nuance here.

They're paying attention. And so as an organization, you need to pay attention to this as well.

Chris Adams: Cool. Asim, I'm just going to add this one thing because I realised I should have mentioned this. So I work in an organisation where we do track some of this stuff. We track the transition of the internet away from fossil fuels to greener energy. And, I've shared a link for the show notes. Because we're basically reviewing our own evidence that we accept for green hosting.

And we've linked to a couple of papers. And specific reports, which dive into this a bit more, which have kind of also prompted us to start looking at this. So, organizations like the Science Based Targets Initiative, we mentioned there. We talk about some of the other things that we, some of the nuances around RECs.

And yeah, this is, this will be something we'll be doing. So we're going to be essentially figuring out how to do this ourselves in the open over the next few months. So, Yeah, I guess it applies to small firms as well as large firms.

Asim Hussain: Yeah, yes, absolutely.

Chris Adams: All right, should we move to the next story? So, this is a story.

Researchers analysed 1 500 climate policies to find what works. And these are the lessons for Australia. I think this is the link you shared with me, Asim. There's a very kind of Australian centric kind of point of view, which, as someone born in a small mining town in Australia called Prospect, because what else would you name a mining town other than Prospect, because it's full... you? Yeah,

Asim Hussain: I didn't know you were born in Australia.

Chris Adams: Ah yeah, born in Australia, small mining town.

Yeah, I was literally born in a mining town called Prospect, and it's near One Tree Hill. Can you guess how many trees are on that hill? And it's next to Dry Creek. Can you guess the conditions of that river? Yeah, it's descriptive rather than creative, is the term I've heard people in Germany who do similar things talk about, actually.

Okay, so you shared this story, maybe you can introduce this one here, because I think it's quite relevant in this discussion, what we were just talking about in the previous two stories, actually.

Asim Hussain: Hmm. So I think it was just really interesting. It was an analysis of like 1500 climate policies and just really looking at what worked and what didn't work. And I thought it was interesting because we spoke a lot about, we've spoken a lot about things like carbon levies and things like that on this podcast.

But what I found interesting about this article was they, again, brought nuance to the discussion and saying, "actually there's different, different policies seem to work for different types of organizations and also combinations of policies seem to work better than individual policies." So a couple of interesting ones.

So one of the highlights I got, so some of the, in developed economies, some of the most successful cases were the results of two or more policies working together. So that could be like a ban or something, as well as like a carbon tax, kind of pulling those things together. Oh, for instance, like a great example they had here was like, for instance, example, a fuel efficiency mandate for vehicles combined with subsidies for developing like charging stations and things like that. So then you've kind of got the pressure on both sides. And another thing that was really interesting was cause we spoke about kind of carbon levies and pricing was particularly effective policy in sectors dominated by profit orientated companies, such as electricity and industry.

So I just think it was really interesting to kind of think through it from that perspective.

Chris Adams: So there's a really nice example, there's a few really good examples. Good concrete examples of this to make this, like, something you can, like, get your hands around. So in America, right, we've seen the Inflation Reduction Act. So that, in many ways, are kind of it's all carrot, no stick. So the idea is that there's massive amounts of subsidies for building out, like, for, like, EVs or building out new, kind of, battery gigafactories, all this stuff like that, or things which are essentially make deployment of renewables much, much more profitable than they otherwise would be, because they're gonna, because you have all these kind of subsidies saying, well, we're going to give you a production tax levy.

So for every unit of energy you produce, you'll be able to get, you'll be, you'll get a credit that you can actually apply. So your project over the entire length of it will end up being slightly more profitable. That, there's stuff like that, that you see, which is coming from one end. But we've also seen, In America, the EPA, the Environmental Protection Agency, they've got, they've now come in with a stick, or they're coming in with a stick now, to basically say, well, we're going to have to regulate carbon emissions.

And this now means that it's going to be all new kind of coal fired power stations or gas fired power stations, there'll be all these restrictions on how you should, how you can use them. And this is particular, the reason I raise this in America specifically is we were talking a little bit about AI before, right?

Now these regulations, I don't think that many technologists are aware of right now. They basically say if you're going to run a gas-fired power station, you need to fit loads of carbon capture and storage onto it, which is, broadly speaking, if it does work, it's not something that's really used in large amounts right now.

And what you currently see right now is you see lots of utility companies basically saying, "Oh, the only way we can possibly meet demand for AI is to build all this gas right now." And the problem with that is that ends up locking in all kinds of emissions. Because once you've built something, you have this incentive to kind of try and get your return back on building this in the first place.

And this feels like," I don't think people have realized just how much of a stick this is going to be, because as far as I can tell, all the laws from the EPA basically say, look, you can't build gas like this, and you can't actually do this." So we're going to have, we've got like this case of massive build out of AI coming up against all these regulatory forces as well.

And it's going to be quite a significant fight in the next 6 to 12 months, I think, because yeah, this is, we've now had the honeymoon period of all carrot. Like you said in this piece, and now we're coming up to the stick, which is the other part, to kind of make sure that you can, make sure the significant part of the US grid is going to be decarbonized by, I think it's the mid 2030s, basically, is what they're doing, that they're aiming for with this.

But we have the same thing in the UK as well, like, UK right now, we've got a target for, the UK has agreed to try and decarbonize the grid entirely by 2030, which is great for us as an organization because we, we want a fossil free internet by 2030. So we're like, "oh, thank God the UK is doing this." The UK government, one of the big kind of manifesto policies from Labour coming in, who've just won the election is "we're going to have a clean grid, entirely clean grid by 2030."

So five years, basically five years time, they're going to get rid of all the fossil, almost all the fossil gas generation, right?

Asim Hussain: How are they going to do that?

Chris Adams: That's what we'll find out. But the thing I found out when I spoke to some people who, basically, this is actually all based on some modeling using a piece of software that we interviewed a chap called Iegor Riepin, he was talking about this in one of the episodes, we'll share a link.

That software was, basically, these kind of things were put together by some analysts on our laptop saying, well, this is what you can do. There's a report from Ember Climate where they, the report is called Escape from Gas, I think, or A Path Out of Gas. And this was one of the things that was written in 2022, when gas was super expensive, to say, "well, this is one thing you could plausibly do for this."

And yeah, when, the thing about policy, people reach for what's there. This turned into one of the things that one of the parties led on, and now we're going to see if we do see a fossil free grid and fossil free internet in the UK by 2030. Because, yeah, it's fascinating. I'm so, this is the most exciting, most excited I've been about UK politics in a very long time.

Asim Hussain: I don't know. I might dampen it for you. I'm just not, I'm just not very, I'm just, there's a lot of manifestos that come out from governance when they join and there's a lot of disappointment in the years later when they, when it doesn't manifest, when their manifesto doesn't manifest.

Chris Adams: So this is the final thing that might come in, might be relevant. So the modeling that was used for this as the basis to say, "yeah, we can do this." This one thing that ends up being, so I'll share a post to it, which I end up doing a bit of research and speaking to some of the people about this. It's actually very conservative, more conservative than the National Grid's own

estimates about, specifically in our industry, demand size and batteries. So, these are the two big things that we're likely to see a massive increase in.

Asim Hussain: That's what gas is used for more like this is specifically to get rid of gas.

Right.

Chris Adams: yeah, so the,

Asim Hussain: peaker plants and then therefore you can do a little demand, demand responsible.

Chris Adams: Partly that, the thing that they said is like, the, their plans basically are relatively conservative about the ability for demand side of reduction, making your, you know, Carbon Aware in stuff like that, right? And there is another thing that we've seen is that the UK government is actually being quite gung ho about deployment or deploying all these new data centers.

So I'm kind of thinking, is there a chance to actually say, "well, okay, if you're going to have this deployment of all these data centers, and you know that one thing you're going to need to have is a much more responsive grid, is there a path for all this kind of carbon aware infrastructure to actually serve some of the roles that you wouldn't have to typically rely on peaker gas plants to actually fit, to like kind of fit?"

There's a bunch of stuff there and I think we'll learn basically because, yeah, this has been a really ambitious goal and you've also got this other idea to like bring in something which, can be quite flexible, but only if you incentivize infrastructure to be flexible, because for the most part, we don't see an economic incentive passed down to the consumers of infrastructure to be using this right now.

So, yeah, maybe this is a help of one piece. Yeah.

Asim Hussain: Yeah, I mean, I think there's been, there was some really good work done kind of several years ago, and this will be really good because one of the things I've seen is that the, all the positive moves I saw kind of three years ago regarding new data center rollouts, hydrogen fuel cells, kind of building kind of a much more advanced data center seems to have gone back a little bit.

And yeah, You're right, I think data centers could lead the way in terms of demand response. I'm not even talking about compute demand response. You can just take batteries, you can fill data centers with batteries and then they can store and then they can do their own sucking from the grid when it's clean and powering their own infrastructure when it's dirty.

You know, there's, there's other solutions, which doesn't even necessarily need kind of a software,

Chris Adams: Yeah, exactly. I mean, this is one thing that we've seen in Ireland. There's precedent in Ireland where people have said, "if you're going to be connected to the grid, you need to be prepared to be kind of complementary or sensitive to the needs of the grid for this." So, I think there's actually room for this, and it will be really nice.

I think that this feels like, given such an ambitious target, it does feel like a role where you could actually tell a good story about green Software, and be part of the solution as opposed to part of the problem, because a lot of the discussions around like rolling out of digital infrastructure is basically saying we can't possibly meet this demand.

But if we accept that demand is dynamic, then there is a chance to actually fit this in, and that feels like definitely worth going for, particularly to kind of maintain this kind of social license for operation, particularly for technology firms.

Asim Hussain: And I think a lot of what you've just said over the last couple of minutes runs very counter to what we were saying before about, I mean, everything you just described, this is all related to that whole idea of additionality. It's all about how do we transition, truly transition the grid to be fossil free?

And you need solutions like this. Not necessarily buying unbundled RECs, but you need to actually, like, think through, "well, how do I how do I be a better citizen in the grid infrastructure, do demand response, be sensitive, not demand energy when everybody needs it and therefore we have to spin up a gas power plant or something like that."

So these are the kind of things you need to actually transition the grid.

Chris Adams: Asim, I think we might have fallen down a bit of a grid rabbit hole,

Asim Hussain: Yeah, we've done it again, haven't we?

Chris Adams: Yeah, so, we're gonna have to move on, I think we've come up to time, but Asim, it's really nice to see you again, I'm glad you had a nice holiday, and I guess we've got a bunch of new things to do this quarter, right? With various projects we have inside the Green Software Foundation, and in the other member organizations related to it.

Alright dude, it's Friday, so have a lovely weekend, and for those listening, we'll put all the links to everything we've discussed in there, and if there's something you didn't see, Please do let us know, and we'll make a point of adding it. Alright, thanks a lot folks.

Asim Hussain: Thanks, Chris. Bye.

Chris Adams: See you around soon. Bye! Hey everyone, thanks for listening!

Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

Making Testbeds for Carbon Aware Computing

12 september 2024 | 48 min

Host Chris Adams is joined by special guest Philipp Wiesner, a research associate and PhD student at TU Berlin, to discuss how computing systems can better align energy consumption with clean energy availability. Contributing to Project Vessim, Philipp explains how researchers are now able to model different energy consumption scenarios, from solar and wind power integration to the complexities of modern grids despite the scarcity of available testing environments. They discuss federated learning and its role in carbon-aware designs, along with challenges in tracking real energy savings. Tune in to learn about the future of carbon-aware computing and the tools being developed to help software become more sustainable.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Philipp Wiesner: LinkedIn |GitHub | Portfolio

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

1st International Workshop on Low Carbon Computing | (December 3) [41:26]
Code Green: The Intersection of Software Engineering and Sustainability | (September 12 at 11:00 am AEST · Virtual) [44:41]
Doing for Sustainability, What Open Source Did for Software | (September 16 at 11:20 CEST · Vienna) [45:02]
Engineering a Greener Future for IT | (September 18 at 5:00 pm BST · London) [45:24]
Collaborating On Digital Sustainability | (September 18 at 6:00 pm BST · Brighton) [45:33]
Green IO London | (September 19th - 9:00-18:00 - London) [45:45]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Philipp Wiesner: Optimizing energy usage for when the sun is shining and the wind is blowing, of course it makes sense. But the big problem is that these green energy certificates, these guarantees of origin certificates, are traded independently of the physical flow of energy.

I'm your host, Chris Adams. Hello, and welcome to Environment Variables, where we bring you the latest news and insights from the world of sustainable software development. I'm your host, Chris Adams. Today, we're diving into the world of carbon-aware computing, how computing systems can align the energy consumption with the availability of clean energy.

Natural cycles govern all kinds of things we do. We mostly go to bed at night, and the food we eat is often influenced by what season we're in, and so on. This also shows up in how energy is available to us. You can see how day follows night in the output from a solar panel. And broadly speaking, it tends to be windier around winter than it is in the summer.

Even when we burn fossil fuels, we're basically digging up energy stored millions of years ago. As we move away from fossil fuels though, where we get to use the sunlight that was conveniently stored for us, we move towards a system based around harvesting energy, like we harvest food. The way we consume energy needs to adapt as well.

And that's the idea behind carbon-aware computing. But how do we know if our carbon-aware designs are actually better for the environment? With us today is Philipp Wiesner, a research associate and fourth year PhD student at TU Berlin, Technical University Berlin, whose work focuses on carbon-aware computing.

Philipp is deeply involved in research that connects the dots between computing power and renewable energy, and today will unpack the work he's been doing, including his work on Project Vessim, a testbed for carbon-aware computing. So let's dive in. Philipp, thanks so much for coming back, and for those who haven't heard the last episode when you joined us in 2022, can I give you a bit of space to introduce yourself?

Philipp Wiesner: Yeah, so hey Chris, and thanks a lot for having me back. I'm really excited. Yeah, so I'm Philipp, I'm a PhD student, as you just said, and a research associate at TU Berlin. I'm in the research group for distributed and operating systems. And I've been doing research on carbon-aware or in general, sustainable computing since the start of my PhD or my master's thesis in 2020.

And yeah, at the start of my PhD, I was mainly looking at the basics. So that's the first podcast we had two years ago. The field was still very young and we were looking at the potential of delaying the execution of workloads to make use of cleaner energy. Later, we also looked at migrating workloads from this to different data centers and so on.

And it was all based on carbon intensity. And then from then on later, we moved on from only optimizing for carbon intensity to really better understanding how electric grids work and rather optimizing for curtailed energy, excess energy. So, for green energy that would otherwise go to waste.

Chris Adams: Cool. Thank you. Okay. So you mentioned we spoke in 2022, two years ago, and the field has moved. And while we'll share a link to that podcast, where we go into a lot of detail, can I just give you a bit of time to just provide a brief summary of the paper that you spoke about then? I think it was called Let's Wait Awhile.

That was one of the projects that you're working on, and that was one thing related to the cloud. If we can talk about that'll be a nice scene setter. Then we can talk about the new work that we, honestly, I'm really excited about in all kind of nerdy glee. So yeah, Wait Awhile. What's the deal with that one?

Philipp Wiesner: Yeah, sure. So the idea in Let's Wait a While was just that in different regions of the world and at different times we have different carbon intensities. So carbon intensity is a metric that describes how dirty or how clean is the energy that you draw from the public grid in terms of emissions per kilowatt hour.

And at the time we were looking just, if you have a single data center and you might have some flexible workloads. So workloads where you do not really care if they happen now or tonight or tomorrow morning, how much potential do we have in shifting them, delaying them, computing them a bit earlier, maybe dissecting them in different parts and computing these times in different parts and so on?

And it was like a bit of a, yeah, it was a basic work at the time.

Chris Adams: And there were some, I think, when I first read about it, there were some quite, the savings there were around the low tens of percent for, compared to what the other ones would be. It was something like that, I think, from there. So that was one thing that caught my attention while we ended up speaking about it last time.

This new work that you presented at Hot Carbon, Vessim, as I understand it, this is something which is more general and applicable to all kinds of grids rather than being focused on one or two. Can you tell me a little bit about why you worked on this and what the general idea for creating a testbed for hardware computing was in the first place?

Philipp Wiesner: Yeah, sure. So the thing is that in carbon-aware computing, it's really common to simulate. The thing is that we're often talking about highly distributed systems, often globally distributed systems. And also these approaches often only take effect if you like simulate them or like run them for weeks or months because we need different seasons.

We need different, yeah. So the thing is that in Let's Wait a While and also in basically pretty much all the papers I wrote so far, we have to simulate systems. And in pretty much all the literature so far, people simulate systems because it's just extremely expensive to actually build and operate such systems.

Only big players can do it. And also results are not reproducible. So no one can really check if the stuff that you're reporting is correct.

Chris Adams: So let me just check. So basically, there is no way to... essentially there's no counterfactual, there's no way to see if my, like, I might say "yeah I've made the savings" but without showing your workings it's quite difficult to stand by them or even challenge them actually, that's one of the things that seems to be a bit complicated with that, right?

Philipp Wiesner: Exactly, a general problem in research, to make stuff reproducible. But here we have the big problem that everyone is implementing their own simulation. So everyone is doing it by themselves and just iterating over a big time series for carbon intensity is maybe not that hard. But the moment you have more complex systems and you're working with forecasts and so on, there's a lot of things you can do wrong.

And it's quite a heavy development effort often to do things right. So, just over the years we collected a lot of experience and eventually came up with the simulator. So, Vessim is basically a big co simulation suite where you can plug and play your own simulators for renewable power generation, carbon intensity of the grid, energy storages, and so on.

And you can connect them, you can connect existing simulators, and then test your real system, real applications, as if they were interacting with this simulated energy system.

Chris Adams: Ah, I see, so let me just check if I'm So one of the things you mentioned before is that we tend to look at the grid as like, just one thing that might be changing. And even our interactions are influencing the grid as well, so there's, that's one thing you need to take into account. But you also mentioned just then, that while most of us think of electricity as just power coming out of the wall, there's actually, it's more helpful to think about it in terms of, well, there's maybe on site clean energy, which is maybe very, clean, but isn't always available, and there's maybe on site batteries where the carbon intensities might be influenced by what has gone into it.

Like if it's clean energy that's gone in, it's going to be very clean. If it's dirty grid energy, it's going to be quite dirty. And then there's the grid. So there's multiple different, kind of, ways energy might go into the system that you use for your computing, for example, and you need to account for some of those things.

And then in addition to that, different parts of the world might have totally different grids. So Poland, for example, right now, loads and loads of coal, but comparing that to Iceland, which is, or France, which is like full of nukes, it's going to be very different carbon intensity. So you can't make the same assumptions, and you'll need to essentially do something to recreate the conditions that you might be in so that you can make somewhat more robust and intelligible statements about the savings you might achieve by making any kind of carbon-aware designs.

Philipp Wiesner: Yes, exactly. So yeah, as you said, like most of the existing research is looking at carbon intensity is looking at the global grid. And the problem with this metric is in general, that it's oversimplifying how complex grids actually are. So energy markets are really complex, grids are really complex. So if we only optimize for carbon intensity, which is a relevant metric in the future, we will run into several problems.

Because it's just hiding a lot of underlying complexities. And, as you said, we can maybe avoid these complexities if we're looking more local at microgrids. So if we assume, I don't know, a data center provider is putting solar panels on their roof or like having a direct power purchase agreement with some close by wind park, then we have a lot more direct connection of who's producing and who's consuming, and Vessim is exactly designed for this, to simulate such systems so that eventually you can optimize really for the energy that you're producing or storing.

Chris Adams: I see, okay, so you're still looking at this and we're looking at more than just a grid, basically. We're saying, we're focusing on the carbon intensity because the power can be coming from different places, like the one I'm immediately thinking of, actually. So in the news, Elon Musk's talking about this new data center where he's, where they bought 100 000 Nvidia H100 GPUs using 100 megawatts of power.

And that's so much power that they can't get that from the grid. So what they've done is, it's not clean energy they've used. They've used, say, like, a bunch of mobile gas generators, plugged into the data center that's come in. So we spoke about on site clean energy, but there's actually all kinds of energy that might be going in, and we're now seeing something like this in this kind of world of centralized AI, where you do see organizations looking to things other than just the grid to power things.

So we might need to take into account a whole basically the system seems to be getting more dynamic and being able to simulate this seems to be one of the ideas that Vessim is intended to kind of support so that we can essentially show our working and understand what kind of changes we might be making, or under these conditions, this is what a carbon-aware design might achieve, for example.

Philipp Wiesner: Yeah, exactly. Yeah. As you're saying, it's unfortunately becoming more and more popular to again, build gas turbines or even small nuclear power plants right into data centers to like come up with this insane amounts of energy that they need by now. We of course, like rather focusing on renewable sources because they're also a lot more difficult to schedule.

But yeah, this is one of the use cases of Wessim. So you can like explore how does it affect my data center if I cover the roof with solar panels, build a wind park nearby. How does it affect my data center if I invest into a lot of energy storage? How can I optimize my data center to make best use of these local resources?

And then there's also a lot of like systems questions, like how do I optimize? How do I make applications aware of locally produced energy? How do I make them aware of the state of charge of local batteries? What are they allowed to know? Like there's also security questions involved. And all of this we want to make explorable, so to say,

by providing this very general purpose, extendable simulation framework.

Chris Adams: Okay, cool. That sounds, okay, this sounds lots and lots of fun, and I remember getting quite excited when I read the paper from Hot Carbon. So you've shared this, but as I understand it, some of this has actually been used for some production work, or this isn't just like a wacky paper, there's actually people, there are examples of this being used.

I know that your paper spoke about the use of "Okay, here's how we've been running this on a kind of test system in a Raspberry Pi. Here's how it works with AWS cloud instances." So there are various examples of it being used. But there is a new project that you told me about beforehand called FedZero.

Can you tell me a little bit more about that? Because that seems to be applying some of these ideas in the real world. So yeah, maybe I'll just give you the chance to plug that one, because yeah, it was a fun read.

Philipp Wiesner: No, sure. Thanks a lot. Yeah, exactly. So we're trying to, deploy Vessim in a lot of different projects now. One of the main ideas behind Vessim is also to be have like hardware and software in-the-loop integration. So you can also plug in real devices, real energy generators, and so on.

In FedZero, we didn't do that. FedZero is also fully simulated. But it's a new, it's a scheduler basically we proposed for federated learning systems. This is,

federated learning is some type of distributed machine learning training. So you have a bunch of different clients and they locally train a machine learning model.

So they locally consume energy and the scheduler basically decides when, who is training. So it's like distributed batch processing in the end. And the entire evaluation of the system was done with Vessim. We have, for example, a lot of data sets that are coming with Vessim. They've been provided, for example, by Solcast.

It's a big solar data provider. And the entire idea behind the system is to only use excess energy. So only use curtailed energy and spare computational resources to perform a distributed machine learning training to make it effectively (an exclamation mark) zero carbon! Because you don't use any energy that would have been used by anyone else, basically.

Chris Adams: Oh, I see. Okay, so there's two things that I'd like to unpack there. So could you maybe just talk a little bit about the relationship to, I think I was confusing this with Flower, perhaps. So Flower is one of the wider projects which has something to do with federated machine learning. you maybe just talk on that because I think, for people who have listened to this before, they might not be aware of Flower or the even idea what federated machine learning might be. Because I just spoke about a centralized massive data center, but that's not the only way you can do machine learning and AI training by the sounds of things.

Philipp Wiesner: No, sure. We can also go a bit more into detail. So in federated learning, we assume that our data is distributed. So it's a different clients, a very common example are hospitals that, for example, want to train a common model for skin cancer detection whatsoever, but they cannot share their private patient data.

So we need to somehow train on their data without them releasing this data. And what's happening and Flower is one of the frameworks that lets you do this, that lets you perform federated learning as we distribute the model to the different hospitals, they locally train on the data and send it back.

And we try to schedule this local training in a way that it's only using green energy. Yeah, that's the main idea.

Chris Adams: Ah, I see. So, these would be devices which might already exist, for example, you don't have to have the same kind of emphasis on having a bunch of new stuff that you're using all the time to kind of sweat the asset and make the most of it, so you could be using maybe older technology or a more kind of mixed set of computing, rather than 100 000 H100s, for example, in one straight place, for example.

Philipp Wiesner: Yes, that's true. That's true in one sense. On the other hand, we always have to see what we're optimizing for. And if you're optimizing for something like curtailed energy, like for example, in the north of Germany, we have a lot of wind parks. But the German grid is not always strong enough to get all this energy to the south where like the big consumers are, the big industries.

So often times we have an excess of energy in the north. Now if you have hundreds of kilowatt hours of excess energy in the north, it doesn't get you anywhere to turn on a bunch of iPhones or old computers to make use of this energy. So usually with our approaches we target big consumers. We target the A100s and H100s to, because they actually make a difference.

Like turning them on or off actually makes a difference to the grid. A few thousand iPhones, not so much.

Chris Adams: Okay, so you still would be using some of this kind of relatively powerful hardware, if it represents a significant power draw on the grid, for example. And, is there anything special about where these might be placed? Because I've, seen examples and I've seen people talking about the idea of like literally putting this stuff directly into, say, the towers of wind turbines.

I know this is one thing that Windnode in Germany were doing, for example, because that's basically, where the power's not being used, it's one of the most fastest ways to actually

Philipp Wiesner: yeah, clearly, yeah, that's a super important topic, like placement of data centers. Just what I said, of course, you like place these data centers again, and if we talk about Germany in like Frankfurt region or in the south of Germany, it doesn't get you anywhere because you still need to transfer the energy there.

So you would actually need to build data centers very close to where energy is produced. However, like the big problem that we're facing right now with all this LLM and GPU stuff. is that there's actually not a lot of flexibility. So, FedZero for example assumes that we have some spare resources and some excess energy, and maybe the excess energy is fair, we do have that but we do not have spare resources.

Like, whenever I keep telling people since the last six months about this project they're like, "no, there are no spare resources." Like, GPUs are used all the time, we have a GPU shortage so, carbon-aware computing on GPUs is not really working at the moment.

Chris Adams: Yeah, I'm glad you mentioned this because this was one thing I've always been struggling to make sense of because when you've got a graphics card which is like 40k and you've got only a certain amount of years to use it, for example, the cost of having it off is actually quite high. So therefore, being able to simulate what the savings might be from some carbon air computing might, would actually be meaningful.

Or, if you're not going to switch off the power, then you might need to find another way to, like, not draw power from the grid. So that kind of implies either an increased amount of storage that you might need, so that rather than switching things off, you just reduce the power you're drawing from the grid by pulling it from somewhere else, for example.

So that's another kind of way you might flex your system to... still reduce the emissions, because it may be the case that you're able to load up on clean energy when there is an excess in other places, whilst keeping that thing still running, for example.

Philipp Wiesner: Yeah, exactly. No, you're completely right. And that was different, a few years ago when you, like, were mostly talking about cloud computing on CPUs, when we were talking about carbon-aware computing, because there you usually have a big buffer, because cloud providers want to have this image of infinite scalability, so they usually have spare resources at hand.

So, yeah, like this entire load shifting stuff was somewhat reasonable. Now with GPUs, I mean, you're right, you would need to plug them off, but no one's doing that. Like, if you're buying a GPU, you're using it.

Chris Adams: Using as much as you possibly can to make it, to make the per hour cost as low as possible, because you're, I see, okay, so, this is one thing I'm really glad you allowed us to dive into this because one of the things that I would initially think about, like there's some trade offs you might be making, and I might, before I spoke to you, one of my assumptions was that by using a federated model, you might have a lower embodied carbon because you're able to use a wider set of computing, which might already exist, for example, but it sounds like, because some of these chips are, they use more power, but per instruction, they're much, much more efficient per instruction. So, you're probably not going to save that much from the total embodied carbon of your fleet, for example, by reusing existing stuff in some cases. Is that what it would be like?

Philipp Wiesner: Yeah, there's different studies on whether highly distributed training or computation in general is actually more or less efficient. It often really depends, like usually the centralized, like you won't get anything more efficient than an H100. Energy per instructions. It's super, super efficient, but then of course, if you put them all in one place, you need to cool more and so on.

So often it can be more efficient to distribute it. there's no clear answer to that. I would say it really depends on the system, whether federated distributed system is more energy efficient or a local one, but of course in distributed system, you might have more flexibility for scheduling and waiting and yeah.

Chris Adams: Ah, I see. And I'm just going to move on to another question in a second, but while we talk about that, I just wanted to speak about maybe this idea of more distributed federated learning. A lot of the time you see people talking about saying, "I need a cluster of machines in one place because they need to talk to each other."

And the closer they are, the lower the latency, the faster the responses might be. And. When you're using a more distributed version, a distributed model, if you're not doing, like, even if you're not thinking about machine learning, you still need to think about latency, how long your response comes back.

How, like, what are the trade offs that you have there when you're looking at a kind of federated model, like with Flower, or providing some of these kinds of more distributed, carbon-aware style approaches, for example? Like, how do you address that?

Philipp Wiesner: So in general, in federated learning, we differentiate between cross device and cross silo federated learning. And cross device is what you were mostly referring to as like you have a lot of iPhones, a very distributed small system. They're often connected via mobile network. So latencies are usually quite high.

And there you really need to consider this latency trade off. In cross silo federated learning, it's a bit less important because you assume that each of these clients is, as I said, like a small data center or a big data center, they are connected via fiber. So, latencies are usually very low and not super relevant compared to how long a training takes.

So like sending models back and forth, model weights is not really relevant compared to how much time you need to compute actually.

Chris Adams: Ah, thanks for explaining that, because there are various companies who are talking about having a distributed model, or saying, you need to do, you need to have, maybe, infrastructure that is complementary to the grid. And I was never able how to, how you get around the issue of speed of light, right?

Because they're going to be very, different if, even if it's just opposite ends of the country, that's going to, depending on the size of your country, that's going to be somewhat meaningful there. Okay. All right then, so we spoke a little bit about the use of Vessim, we spoke about why it exists, and like kind of this idea that you can simulate various kind of configurations of grids or configurations of power coming into the system.

I imagine there'll be some people who are listening thinking, "this sounds cool, how do I actually use this on a project when I'm designing something in a kind of carbon-aware way, or how do I actually, what am I looking for?" Could you maybe talk me through? The actual process of using Vessim specifically.

We can start with something simple like a machine learning job first and then we might talk about some other configurations like perhaps a CMS because that's what lots of people might be working on for example.

Philipp Wiesner: I mean, the entire idea of Vessim is that you do not really notice that it's a simulator. It should behave as like a proper energy system. So you can, on the one hand, you have your normal applications. It doesn't matter if it's a machine learning job or if it's a CMS or if it's anything else. You just have your applications that are running.

And then all you need to do is you somehow need to get their power usage into the simulation. You can either do that by modeling. You can, I don't know, look at the CPU usage of your application and somehow estimate how much power it would use. Or you actually let it run on a machine, plug a proper power meter, and then periodically feed this information into the simulation.

And the simulation itself you configure as you wish. So we have a lot of READMEs. You can plug and play custom simulators. We have some implemented, we're right now implementing the ones from the National Renewable Energy Laboratory. They have lots of simulators on everything from solar to wind to geothermal and whatsoever.

So you can plug your microgrid together. Also provided with energy storage. We are releasing Vessim 0.9 this week, I think, which then comes with a few more energy storage solutions. So you can really plug this together as you wish. And then you can write your own API basically to access this data center. The new applications, as in normal carbon-aware computing, can just query a REST API or something and really ask "what's the state of charge of my battery, how much energy is produced right now, what's the forecast for carbon intensity," all of this.

Chris Adams: Okay so I'm going to break that down a bit because there's a couple of quite interesting concepts that you spoke about. So first of all I feel a bit embarrassed by not mentioning this, we just assumed it, of course, was open source, but yes, this is entirely open source, anyone can access it, if they search for Vessim, V E S S I M, they'll likely find the project, because there aren't that many other things called Vessim right now, is that the case?

Philipp Wiesner: I hope so, because we were searching for a searchable name.

Chris Adams: We'll also share links to the actual project on GitHub as well. It's written primarily in Python as I understand it, so if you know Python or know someone who works in Python, and given that Python's basically the most popular programming language on Earth, or by various metrics, is one of them, then the chances are fairly high that you can start doing this.

Are there, like, presets for different grids? So in the same way that you might have a preset for, I'm viewing a website on a mobile phone, or I'm viewing a website on a desktop, you said a little bit about NREL, the National Renewable Energy Laboratory. They've said, "well, this is what the, this is how a geothermal plant might work, or this is how a wind turbine might work and this is solar panel might work."

And that gives you an idea of saying, "well, this is somewhat similar to my grid." And then there's kind of, are there some presets or things so people can come up with some approximation of what it might be in their part of the world?

Philipp Wiesner: Not yet publicly. We're working on this and we will publish it soon. But right now there's no presets, but Vessim comes with some data sets that were provided. One was provided by Watttime on carbon intensity. One was provided by Solcast on solar data. And you're right, Vessim itself is in Python, but we're like building on a library called Mosaik.

Been developed in Oldenburg, also in Germany, and this is basically the entire co-simulation system behind Vessim. And they allow you to basically plug and play any type of simulator that can be written in Java, or it can also be a real system. You can also just take a real battery and measure that and just feed this information.

So it's really very flexible in that sense.

Chris Adams: Ah, cool, I'm glad you mentioned about the power use thing, because that's the thing I was going to speak to you about, because you said that you can draw power from a physical system, so you might time series of power usage, but one thing that we've spoken about previously with other guests on this podcast is that if you know what kind of compute usage, how much of your compute of a server is, you can make some I'm a bit worried about using the word inference when we've just spoken about AI.

But like, you can, there'll be some idea of what the likely power might be, just because machines tend to use more power at 100 percent utilization versus maybe 50 percent utilization. There's various curves and things that give you some numbers. And I believe there's an organization, Green Coding Solutions, who share some of this.

But also, I think this is under the hood what, other projects like Kepler also have some reference to. So, there are different ways that exist right now to give you some numbers. Even if you don't have something plugged into a system, for example. Maybe, you don't have access to, a power meter like you're deploying in a cloud, for example.

Philipp Wiesner: Exactly. Yeah. And this is cool that you mentioned that because this is one of the, it's a master thesis that we are running right now with one of the students who has also been developing Vessim for the last two years. And we are also using Kepler and trying to really calibrate Vessim for real systems.

And of course, these calibrations can be published. So we can really see, because there's this big difference between software-based power metering and like actual hardware power metering. In software-based power metering that's like included in Intel processors and also NVIDIA graphics cards, they just estimate how much power do I draw right now.

But it comes with a certain error. And what we're doing is like we're benchmarking these machines, using the integrated software-based power metering that actually everyone has access to but also like plug a real power meter and see how big is the difference under different conditions and so on and try to, yeah, make Vessim a bit easier to configure for bigger systems. Because this is the entire idea, maybe you have access to one GPU that you can benchmark but then you want to explore, "okay, how does my system behave if I have 200 GPUs?" And you can scale up in the simulation.

Chris Adams: Okay, so when we, so when someone is starting with this, like, we're going to talk about actually running the software, but we're just setting the scene to figure out, okay, what does the setup look like? Because I think most developers might be familiar with the idea of running a test to see if it works as intended.

There's this whole new world outside of your computer system that we need to talk about that I think you might need to kind of, that people need to understand before they start thinking about what, how the system is supposed to be designed. So we spoke a little bit about the fact that Vessim is using a existing tool called Mosaik, which designed for simulating grids.

So someone's done all that work already to figure out how to simulate a grid and you're using, you're relying on some of that. So there's a bunch of work that's gone into that already that you can draw upon. And we spoke a little bit about, there are ways that you can get readings on power, so even if you don't have something plugged in, you can still get some maybe low precision but still useful numbers to help inform some of your, like, design choices.

And then, now you've got that part there. You said a little bit about maybe things being in the, software being in-the-loop here, and I think while I've spoken excitedly about the notion of ecovisors and things on previous podcasts, I think that's what you're referring to here, like if the software itself knows how much, how full a battery is, or how green the power is, it might change in certain ways or it might respond to certain things.

Could you talk a little bit about that part? Because I think the software in-the-loop part is the thing that most people who've been taking an existing project and maybe trying to extend it to be carbon-aware. That's where they might be spending their most times either thinking about how to test something or figuring out how to make and adapt their existing system so they can start coming up with some metrics.

Philipp Wiesner: Yeah, It's interesting that you mentioned Ecovisor because it was a big inspiration for this project. Because I think they spent tens of thousands of euros on this one solar array simulator and a testbed. And we wanted to, like, explore similar ideas, but we didn't have the funding at the time to just buy such a machine.

So I had to write a simulator, more or less. So this is how it all started. Sorry, what was the question?

Chris Adams: in-the-loop, software in-the-loop,

Philipp Wiesner: Ah, software-in-the-loop, sure. Yeah,

Chris Adams: We've simulated all this stuff, we know roughly how much power we're using, so the system, the software now has the data going into it. So what happens next? That's the thing.

Philipp Wiesner: Yes. So the idea is we need to run the simulation in real time, right? In wall clock time. So one second of simulated time needs to be the same as one second of real time. And if we manage this, then we can simply from the simulation, host any kind of API. We can just spin up a REST API that gives us access to the system.

And in Vessim, users can define these APIs themselves. So in Ecovisor, for example, they already came up with an API. I don't know, "/battery" and then you can query the battery state of charge, but something like this. Vessim lets you explore this. How do we abstract the complicated energy system and how do we like expose relevant information for the computing system?

And then you can just run a real normal application running on your computer, running in a docker container whatsoever. And it can send requests to this energy system API. And best case, if we did everything correctly, it behaves exactly as if you were doing a request to an actual energy system. So your application doesn't know that it's working with the simulation.

Chris Adams: Okay, so I'll try and make this a bit concrete. So when I'm working on a project, in an existing kind of carbon-aware system, I might make an API call to like, I might ask electricity maps, "Hey, what's the local grid intensity where I am right now?" For example, and like, it's a little bit like that kind of request I'd be making, but I might be making it to Vessim to say, "what's the grid intensity?

What's the state of my charge or stuff like that?"

Philipp Wiesner: Exactly. You just write a normal application that is running on a system, just that the system is power metered, so Vessim can simulate how it would be behaving if it was plugged to batteries, for example, and then it can talk to this simulated energy system.

Chris Adams: Gotcha. Okay. And we're using Vessim as just the kind of, simulator to share this stuff and it may be that in the real world, like I might be, speaking to the electricity maps API or something like that.

Philipp Wiesner: Absolutely, would be, yeah, absolutely.

Chris Adams: Okay.

Philipp Wiesner: A plug and play thingy, yeah.

Chris Adams: All right, that part, I think I understand. So then at that point, I'm then able to basically say, well, let's imagine electricity maps has given me this figure for carbon intensity, or let's imagine my, the state of my battery is high rather than low, I then might make some changes and say, well, "When the energy is low, don't do this stuff, or just delay it for maybe half an hour to check it again."

So that's the kind of stuff. So maybe you could talk a little bit about, now that we're at that point, how does that manifest into me seeing if that has actually reduced carbon emissions over a period of time, for example. Maybe we could talk about that part, because I think that's the final thing, now that we've set up our system, we've connected to, we've figured out how to get data coming into my system to tell me what the carbon intensity might be, or what batteries might be, or what the power from a wind farm might be. So yeah, maybe you can talk about that. So over a given time, has this actually helped, for example?

Philipp Wiesner: Yeah, And this is actually what makes simulations so powerful, because you can just execute a bunch of different scenarios. You can execute your scenario one time without any carbon-aware strategies. You can execute it a second time with the carbon-aware strategies. You can simulate as if you were executing it in summer or winter, or you can simulate power outages and whatsoever.

Because it's all really fast and really cheap to do. And then of course Vessim is recording all these metrics. Like for every second it's recording how much power has been drawn, produced, how much emissions were there. So afterwards you can analyze these results and compare the different runs.

Chris Adams: Ah, okay. I think I'm understanding this now and I

kind of figure out how I might build some of this now if I was to be building maybe a batch job, for example. I can kind of maybe imagine how this might work for maybe a content management system or like a CMS or WordPress or something like that. So let's say I would do, would I be looking at like, this would be my traffic the last month, run it through and then compare it to what the carbon intensity might have been.

Is that how I might do this?

for

Philipp Wiesner: Yeah, absolutely. Absolutely. And you can also use it in testing pipelines, for example, in continuous integration pipelines. You can just simulate a bunch of different scenarios for your carbon-aware application. Some of them might be power outages or something like really edge cases that you do not want to happen, but you can really test cheaply that your behavior, that your system or your application would behave as expected

Chris Adams: Ah,

Philipp Wiesner: In a reproducible manner.

Chris Adams: So you've raised something quite interesting for me here because you, when you spoke about the idea of like things, power outages or stuff like that, this kind of implies that if I've got a system here, I could see to what degree I might make a change to the software, or I might just say, well, given I have control over my infrastructure, I'm running a data center, what if I was to just put a battery inside that data center?

How would that impact the carbon footprint of using this system every month, for example? Would that, could that be something that I could simulate with that?

Philipp Wiesner: Yes, absolutely. Absolutely. And if you have a model that is able to, like a battery model, a simulator that is able to simulate battery aging, for example, you can also simulate what's happening over the course of the next three years with my batteries and so on.

Chris Adams: Ah, okay, so it's actually two way So therefore, I can look at, I can, look at, say, making interventions at my code level, and I can interventions at the kind of deployment scenario level as well, actually. Okay, that is something I hadn't actually realised before then. Okay, so we've dived into this quite a bit and I hope there are at least some people following us now.

Can we just take a step back to say because we have just been assuming that carbon emissions or carbon intensity is absolutely the figure we should be looking for and that is our North Star and you mentioned before that maybe that isn't or there's a bit more to it than that because we can't just look at this number, see if it's going down, and say, "Oh yeah, we're making progress."

Maybe you could outline on some of that, because it's a lot of nuance here that people often miss, I think.

Philipp Wiesner: Ah Yeah, sure sure sure. So carbon intensity is a metric that defines how clean or dirty the energy in the public grid is. That means if it's low, you can use it if it's high, you should maybe rather not. And the problem here is that it's hiding a lot of the underlying complexities of how grids work and how energy markets work.

So for example, one big problem that we have in energy markets is, like in carbon accounting in general, is this double spending of guarantees of origin certificates. So for example, you can easily buy certificates produced in Iceland or something. They produce almost all the energy through renewable sources, hydropower, geothermal power.

You can buy their energy and then tell everyone you're 100 percent renewable energy. But at the same time, of course, any data center that is located in Iceland will tell you that they're 100 percent renewable, which is also true. and in fact, there's not even a power line between Iceland and mainland Europe, and still you can buy their certificates.

so we have a lot of double accounting and of course, everyone is coming up with new methodologies of what's good and what's the right thing to do. And of course, everyone's also just doing what's in their favor. And this is the big problem with carbon intensity. So of course you can say now "carbon intensity in Germany is really low.

I should consume energy." But unless you have a green energy contract, you're for sure only getting the dirty fraction, because all the green energy is already sold.

Chris Adams: So let me just check if I can stop you there for a second. So what you spoke about just then was there are kind of two approaches or like there's one one approach of looking at how kind of green power is might be the existing kind of system that you see in most developed energy markets right now where there's like a market-based system where you essentially can purchase renewable energy certificates and apply that to kind of brown energy to then say it's green.

And there may be reasons why we have ended up with this scenario. As you said, there are two problems there when we talk about, say, the idea of buying green power certificates from Iceland and then applying them to dirty power in, say, Poland or Germany to say it's green. There's like a, there's one issue which is that there's literally no way to connect, there's no connection, like we definitely are not using geothermal that's generated in Germany to get there. There's no physical connection so it's not deliverable. And the other issue is that if Iceland was to do that, they would then have to stop saying their power is green because they've already sold on their greenness. And in many cases they're not doing that. They're saying, "oh, we're still green even though we've solved the greenness."

So, there is a kind of problem about the, yes, let's assume that the market, the concept was conceptually sound. There is a whole issue about people kind of cheating a bit and saying no, it's still green when they've sold the greenness on it already, basically.

Yes,

Philipp Wiesner: and exactly in this sense, optimizing for carbon intensity is cheating, because you're doing exactly this. You're like saying, "I'm using low carbon energy," but all the green, as you said, has already been sold. At the same time, it makes sense. The metric also makes sense. Optimizing energy usage for when the sun is shining and the wind is blowing, of course it makes sense.

But the big problem is that these green energy certificates, these guarantees of origin certificates, are traded independently of the physical flow of energy.

Chris Adams: Ah, okay. Okay, so there is, there's a whole discussion about this, like, where we work at the Green Web Foundation, we're kind of borrowing some of the ideas we've seen in the hydrogen sector, where they talk about if you want to say energy is green, it needs to be, I believe it is, timely, deliverable and additional.

So you need to be new power, like the, you mentioned this example of nuclear being used in other data centers, where essentially you're reallocating power that would have been going into the grid. So everyone has slightly greener power, but you say, no, that all has to go over the data center. Then that has the impact of essentially making the power that everyone else is using be seen as slightly less clean because of this accounting system, and that's one of the unfortunate artifacts of that. So that's one thing you mentioned about that. And I think the system that you might be talking about was the idea of like a location-based system, which is much more about like the physical environmental footprint of the power you're using rather than, say, the market-based approach, right?

Philipp Wiesner: Exactly. So we need a system that is somehow time based and location-based, a certificate system, then this all would make a lot more sense. But even then, grids are very complex and it's too easy to boil it all down to a single number and say at Germany, for an EPM we have this number of carbon intensity.

It would maybe make more sense if we, like, break down everything into smaller pieces, again, go more local. I have, for sure, Northern Germany separated from the South and stuff like this. But still, it makes sense, and this is what, for example, Vessim is good for, to really think about and understand how is energy actually flowing, how is the energy system actually working, because it's a bit more complex.

Chris Adams: Ah, I see. Okay, so for people who want to go down this weird accounting rabbit hole that we just kind of skirted around, I will share a link to a journalist called Hanno Böck who's dived into the world of Iceland, and in particular about this whole idea of selling green energy twice like this, because if you are looking into this, it's, as we end up with a much more decarbonised grid, this is going to come up again and again.

And right now, we do see the systems being used, like the GHG protocol, the people who actually define these standards. They're going through a process of reform and basically saying, "well, these are the changes we need to make and this is why this is possibly not fit for the purposes that it was 10 years ago," for example.

So there is a chance to make interventions and talk about this, if you want to look into this as well as looking into the other things we spoke about with Vessim. Okay, so we've spoken about this idea of setting up a kind of testbed that we've established that this is kind of open source, that anyone can pick up and start playing with right now, like the code is open source on GitHub, and as I can see, documented well with like, videos and stuff like that.

We're just coming to time, so I just want to give a bit of space for some of the future events that are coming up, because I understand that, well, part of your job, as you told me, is like, I'm the publicity chair for this all about low carbon computing, so I should probably give you a bit of space to talk about that as well.

Philipp Wiesner: Yeah, that's super nice. Thank you to just take two minutes. So we're organizing this workshop. It's the first workshop for low carbon computing, which will take place on the 3rd of December in Glasgow, and It's mainly an initiative, a local initiative from some folks at the University of Glasgow, but it's an international conference, it's an international workshop I mean.

And we expect submissions from all over the world, but mainly probably the European community, I guess. And yeah, welcome everyone to submit your work. So we have a very broad scope on really anything related to sustainable computing, from energy efficiency, carbon efficiency, embodied carbon and so on, but also topics that are often a bit underrepresented in computer science like circular economy, sufficiency, demand reduction, and all of this. exactly. So this is all very welcome. And for regular talks, all you need to submit is like a two page extended abstract by the 24th of September. Although we're like currently discussing maybe a tiny deadline extension and all accepted authors, all accepted papers will afterwards be invited to submit like a full workshop paper of up to eight pages if you're interested, which will be published in our post proceeding.

So yeah, feel free to have a look. I think we will also put a link in the

Chris Adams: a link in the show notes, for sure, absolutely. So if I understand this correctly, is this a little bit like, possibly, a kind of like a sister conference to Hot Carbon, but where it's a little bit colder, and you're more likely to kind of find a burrito served at Legal Jack's.

Is the name of a restaurant in Glasgow that eaten at before.

Philipp Wiesner: Yes, I guess like it's the first version. Let's see where it's going. But maybe we can, if it's going well, we can maybe establish it as like a European alternative to Hot Carbon or something, which will, I guess, take place in the US in the upcoming years.

Chris Adams: I see. And there's one thing I wanted to just check because when I looked at the kind of call for papers, the set of topics, it's a little bit more wide than just like let's focus on efficiency. There seems to be a kind of systemic angle that I hadn't seen quite in the same way, for example. So there are discussions about like you said,

okay, looking at the demand side as well as just the supply side, for example. And like, I think I've even seen the word frugal computing used inside here. Like, okay, this is almost like somewhat possibly like the limits conference or maybe some of those ideas somewhat comparable to inside this.

Philipp Wiesner: Yeah,

guess, yeah, I guess it's somewhere in between that, in between Hot Carbon and limits, maybe.

Chris Adams: cool. All right. Well, I am now thinking it might be nice to figure out a way to get a series of trains to get up to Glasgow, because, I really like that city.

Philipp Wiesner: Yeah, 18 the last times.

Chris Adams: Okay, cool. So that's one thing that's taking place. So the deadline was the 24th of September. Easy to apply to, there is, and that'll be taking place in December.

Philipp Wiesner: Just one last thing, it's a hybrid event. It's a hybrid event, so anyone is welcome to participate, and also authors are of course welcome to present remotely.

Chris Adams: Okay, so I don't need to get on that if I really don't want to, because, well, I might want to see people, but it takes quite a long time

Philipp Wiesner: only a one day event, so you need to consider if it's worth two days of travel.

Chris Adams: Alright, thanks for that Philipp, I appreciate that and we'll share a link to that so people know when the CFP is open how, well actually when it closes more specifically because I think it is open right now. Okay, cool. So I'll just touch on some other events that we have inside here and then I think we'll wrap up if that's okay with you.

So these are some of the other events that our producer Chris Skipper has shared. So there's an event called CodeGreen, the intersection of software engineering and sustainability. This is on September the 12th at 11am AEST. This has Sara Bergman from Microsoft Norway, and one of the, one of the authors of the book, Building Green Software, and also a regular on our podcast.

She's presenting, speaking with Michelle Sandford. And there's also an event in Vienna, in Austria. This has Asim Hussain, our Executive Director of the Green Software Foundation. He's talking about the Impact Framework, this radically transparent tool for sharing the environmental impact of your tooling, and that's on September the 16th, and I think that is in person, but then maybe we'll share a link for people to see about if they can get involved remotely.

There's also another event in the UK, Engineering a Greener Future for IT, that's on September the 18th at 5pm British Standard Time in London. And there is also another event in Brighton as well, so the UK is quite heavily represented right now, Collaborating on Digital Sustainability, that's September the 18th.

I'm going to use this platform while I still can. I'll share a link to Green IO London as well, which is the conference that I'll be keynoting at with a number of luminaries. We'll share a link to that because I, it's very front of my mind right now and I'll be there for the day. So if you listen to this and you fancy coming along, we'll share a link where you can get a free ticket to join at the conference.

And yeah, that's us. I think we've come to the time. We'll just actually make sure that all the things we've discussed, like I mentioned before, we always try to get as many of these show notes full of links as possible, because we cover a lot of ground on these free wheeling conversations, so that will be visible.

The thing you might need to do though, if you are listening via Spotify or the Apple platform, you might need to look up podcast.greensoftware.foundation, then follow the link to this podcast to see all the links, because I don't think they're made available if you just are listening to it in Spotify, for example.

And we do put quite a lot of time into doing that. And Philipp, I just want to say thank you so much for helping unpack what can be an extremely complicated topic and putting the work in to make a open source project that makes it easy for people to figure out what to do with carbon-aware computing and hopefully guide themselves to slightly more sustainable software.

Philipp Wiesner: Yeah, no, thanks a lot for having me. Yeah, had a lot of fun. It was a pleasure. Thank you.

Chris Adams: Okay, final question I should ask, if people do want to find out about you, where should people look? Are you on LinkedIn? Are you still using X slash Twitter slash what's the best way to follow your work these days?

Philipp Wiesner: Probably just Google me, I guess there's maybe one or 2, 3, 4 Philipp Wiesners that are more important than me, but if you just look at TU Berlin or Philipp Wiesner carbon-aware where I should pop up and then you can see.

Chris Adams: Cool, well your name will be in the show notes and the show title, so if anyone is curious about how to spell Philipp Wiesner and not sure how many P's they should be using, then that'll be inside there. All right, Philipp, really lovely chatting to you, mate, and thank you so much for coming along, and yeah, have a lovely week.

Philipp Wiesner: Thank you. You too. Thanks a lot.

Chris Adams: Cheers, bye!

Philipp Wiesner: Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

Academic Forefronts

5 september 2024 | 51 min

This week we are joined by two PhD researchers, Silke Kaiser and Chiara Fusar Bassini, from the Hertie School in Berlin. With host Chris Adams they discuss their use of data science and machine learning and how they are using them to tackle some of today’s most pressing environmental challenges. Silke shares insights into her research on predicting cycling traffic in cities to better inform urban planning and promote sustainable transport, while Chiara discusses her work on analyzing European energy data to support the renewable energy transition. Together, they explore the intersection of technology, data, and policy, highlighting the importance of data-driven decision-making in achieving sustainability goals.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Silke Kaiser: LinkedIn | Website
Chiara Fusar Bassini: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Silke Kaiser: I like the term of fighting fire with fire. You know, you're trying to make it better, but you're making it maybe even worse. But I think if we make some smart choices along the way, I rather like to compare it to the idea of fighting a forest fire with a controlled burn. What I'm trying to say is that there are different approaches that we can actually also reduce the emissions caused by AI.

Chris Adams: Hello, and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host Chris Adams. Hello and welcome to Environment Variables, where we bring you the latest updates and news in the world of sustainable software development. I'm your host Chris Adams. It's important to understand that when we talk about sustainability and technology, it's easy to mix up sustainability of software with sustainability through software. Sustainability of software development is about understanding the direct impacts of technology and doing as much as we can to reduce it without delivering a worse experience for people using the software. Here, we care about the impact of code, like making it more efficient or making sure the energy we use is cleaner and coming from the cleanest possible sources. So when we talk about green software or green IT, this is what we're talking about. Sustainability through software development, this occurs through the application of software to solve a specific sustainability problem or provide us with insights that we didn't have previously to help us meet some of our sustainability goals. To make this really concrete, you can use green software and talk all day long about the sustainability of software whilst helping people drill for oil and gas. Now you can do that, but it's really not a good idea if you wanna hit any kind of societal climate goals. And if you're listening to this podcast, I think you probably don't wanna do that either. So while we usually cover the sustainability of software. It can also be helpful to look up from our keyboards sometimes to talk about the effects of software, the effects it can have for helping us reach our climate goals. So in this episode, we'll be diving deep into the work of PhD candidates who are pushing the boundaries of what's possible in sustainability ad software. In this episode, we're joined by two researchers from Berlin Institutions, the Berlin School of Economics and the Berlin Hertie School of Governance. So what insights can we gain from their research? How are they using technology to address some of the challenges around sustainability today? Let's find out. So, first of all, Silke, can I just give you folks a bit of time to introduce yourselves? I'll hand over to you, Silke, and then hand over to you, Chiara, to introduce and give you the floor. So, yeah, Silke, thank you very much for joining. The floor is yours.

Silke Kaiser: Thank you very much, Chris. I'm Silke Kaiser. I'm a PhD researcher at the Berlin School of Economics and at the Hertie School Berlin. I'm excited to be here today. My research focuses on the analysis of sustainable transport data, with particular emphasis on cycling data. I utilize various tools from machine learning, data science, and spatial statistics to explore this field.

Maybe more on a personal note, outside of academia, I would say I'm a bit of an auto enthusiast. Just earlier this week, I came back from a vacation in France, but I'm happy to be back and to join you for this episode today.

Chris Adams: I'll be Back just in time for the weather to be nice in Berlin, right?

Silke Kaiser: Exactly.

Chris Adams: Cool. Thank you for that, Silke. And, Chiara, can I do the same to give you some space to introduce yourself as well?

Chiara Fusar Bassini: Yes, thank you for that. My name is Chiara Fusar Bassini. I'm a PhD researcher at Hertie school in Berlin. I'm very excited to be here for my very first podcast. My own research focuses on the analysis of European energy markets. And I use data science and machine learning to analyze time series of dispatch of single power plants and try to look at how power plant dispatch has changed in the context of an evolving energy system. Beside being an academic, I'm also a rather lousy actor in an amateur theater group here in Berlin.

Chris Adams: Cool, thank you. So you said you're working in a theatre. Are there any particular roles you play or anything like that? Because I think you may be the first actor who's come onto the podcast actually, Kiara. Chiara.

Chiara Fusar Bassini: Well, I have a tendency to take on the roles of either mad people or police men or like with gender changes. So anything in between is business, usually like mad people it has been, like the latest role I've had was a police officer. And before that I was a mentally ill person.

Chris Adams: Wow. I did not, I was not expecting that. Okay. All right. Well, welcome onto the show. And I guess that maybe we'll see some, productions in future. Folks, if you're new to the podcast, my name is Chris Adams. I am the executive director of the Green Web Foundation. That's not the same as the Green Software Foundation.

It's one of the members of the Green Software Foundation. I'm also one of the chairs of the Policy Working Group and also one of the hosts of this podcast. Okay, so before we talk in depth about your research, I just want to check for people who are also listening to this, you've been doing this research under the supervision of Professor Lynn Kaack, I believe, is that the case?

Chiara Fusar Bassini: Precisely. We've been working with her for a few years now, each of us.

Chris Adams: Brilliant. Okay, so what we've, the reason I'm sharing this for listeners is that we did an episode 5, where we spoke to Lynn and another person, Will, oh his name has changed, I think it's Will Alpine now, talking all about climate change and AI, two years ago. So if you enjoyed this, I would suggest looking at that to learn a little bit more.

And that might provide some extra context for this discussion. All right, then. Are you two folks sitting comfortably? Happy to go ahead with this?

Silke Kaiser: I think it's good to go and share insights on our research.

Chris Adams: Good stuff. Thank you. For anyone who is listening as well, the thing I'll just share is that we will share show notes with links to all the projects and papers that come up to this. So if any of this is interesting to you, then yes, we, you can continue your quest for more knowledge and insight outside of this podcast. All right. Silke. Let's start with you. Your research focuses on predicting cycling traffic in cities using data from bike sharing systems. And this is something I believe you worked with Lynn and another researcher, Nadja Klein, on from this. Could you maybe just explain a little bit about how this actually helps people when they're trying to design how people move around in cities like, say, Berlin or Paris or things like that?

Silke Kaiser: Yes, I'd love to. So what we generally see when we think about transport in cities is that public space in cities is limited. Whether it be in Europe, the USA, or any other place. Generally, then when we think about how we want to redistribute the space among different mode shares in city, we see that there often tends to be a heated debate.

And especially as we work towards promoting more sustainable modes of transport and therefore reducing the CO2 footprint of our cities, conflicts often arise. And the question then remains, how do we actually want to prioritize these different modes of transport and allocate the space and also financial resources among them?

So, for example, take Paris as an example. I've lived in the city for several years during my studies. And what you can see in the city that in the past few years, they made a lot of changes to prioritize cyclists, which has improved the uptake of cycling, but which has also led to quite heated debates.

The same we can see here in Berlin, the city that we're currently both located in, is that we had a re-election last year here in February, and a lot about the debate actually hinged in part on the choice between prioritizing cycling and individualized motorized transport. So what I do try to do in my research is actually to provide more data to this debate, because what I see as the main challenge in these kinds of debates is that we don't have accurate data on how much cycling traffic we actually have in cities.

Chris Adams: Ah, okay. So maybe just kind of dive into there. So it's basically, we don't have the data to really have a data informed discussion, basically. That's one of the things that is the challenge here. And for context, so the three of us live in berlin. We saw, basically, the new government and the new mayor come into power on a very kind of pro car platform, basically.

So this is what you're referring to, right?

Silke Kaiser: Precisely. So, answering maybe the first part of your question, so for example, in Berlin, we have only 40 locations where we count cyclists. I think in Paris, it's around 53, in New York, it's 41. And then in Berlin, for example, we have around eight times more locations where we count motorized traffic. So we just have much more data and much more information on motorized traffic than we do have on cycling.

And then yes, in Berlin, actually, there was quite a bit of a heated debate pretty much between let's say the inner city, which was more pro cycling and in the suburbs, which were more likely pro cars. The government switched from a green to a more conservative government, which actually decided to suspend actually just this month quite some, projects, long distance commuter paths, and both bicycle parking houses in cities.

But that's just more really on the political side of the debate. And then what I really see as the main challenge is that just this data and the information is missing on where we actually would need the infrastructure most, given how much we want to prioritize cyclists.

Chris Adams: Okay, so maybe I should ask, where is the data coming from then, for this?

Silke Kaiser: So, we do have these 40 counting stations in Berlin, that's the case study that I'm looking at, and what we then figured is, well, we don't have that much precise data on cycling, but we do have an abundance of other data on cycling. So, for example, we do have, as you mentioned, the bike sharing data. We do have as well data from Strava.

That's an app popular to record yourself while doing sports. We do have data on infrastructure, we do have data on weather, we do have data on socio economic factors. And we figured, well, why don't we use all this available data to actually extrapolate from these few isolated locations to actually obtain city wide estimates.

And that's what I did in this research with Lynn Kaack and Nadja Klein. So what, precisely we do is that we train various machine learning algorithms to use all this kind of data in combination with the cycling counting station data that we have to obtain citywide estimates. And what we actually found is that only using this data is a bit tricky.

It provides us estimates for completely new locations 32%, which is you know, rather good in comparison to having no data at all, but still 32 percent is an error I take seriously. And what we then simulated, continuing with this research was, what if we make some sample counts for new locations? So for example, if I want to estimate the traffic in front of your house, Chris, the cycling volume in front of your house, we figured that if we would maybe put you, someone else, on an automated machine there to count the cyclists, and we were to count the traffic for 10 days.

And then combine it with our models, we're able to get estimates with an error of only 17%, so really rather low for complete new locations. And this gives us a good estimate of how much cycling traffic we actually have in every single street of a city.

Chris Adams: Ah, I see. Okay, so you're using the machine learning model, basically, to make the extrapolations when there's, there might not be so much data, give you more accurate so you can say, "well, I'm more confident that this many people are trying to get around using an active, non gas burning form of transportation," for example, right?

Silke Kaiser: Precisely,

Chris Adams: Ah, okay. All right. Thank you for clearing that up. And I understand. And just, we'll come back to this a little bit later, but you said you're using ML. not the same as generative AI or something like that. That's a different, there's, a whole flavor of different things you might be using there, right?

Silke Kaiser: precisely. I mean, there's many different models out there. In this paper, we used rather simple machine learning algorithms, nothing comparable to maybe what most people think of when they think about chat GPT or whichever generative AI you might think about. Those are really rather simple models making usage of the data we have.

And those models So, I just tend to have, I sometimes, you know, sometimes I think about, I mean, you can target these problems with very complicated algorithms, but sometimes using rather simple algorithms might just be sufficient. And that's what we actually found in this research.

Chris Adams: Okay, alright, thanks. When I was doing a bit of research, I realized that you also wrote a piece around, I think in a publication called Catalyse, where you're talking about active transport and why it's important for greenhouse gas emissions, because you've spoken about like mode shift, which I assume means basically moving from a Being in a car to moving and maybe a active transport, which people use like scooters, bikes, stuff like that.

That's what I think you're referring to there. Could you maybe talk a little bit about how this research actually contributes to like the adoption of cycling as one of the potential modes, because you spoke about it in Paris and I went to Paris and was being pretty transformational when I was there compared to being a few years back.

And I also bike around Berlin too. So I have a vested interest in learning this. Maybe you could like lay out why that's actually benefiting and how active transport helps. Basically, us meet our climate goals.

Silke Kaiser: Absolutely. So, when I, Referring to other research that I've read, research that I haven't done myself, but it is out there and it's been cited a lot, is that we do see that cycling has numerous benefits. It benefits your individual health. If you cycle, it's good for your physical health and then all the sicknesses or illnesses related to insufficient physical health.

We can also see that if you cycle, it's also good for me because then generally we see a reduction in noise and air pollution in cities. So it really benefits the public health, the broader public. And then yes, absolutely. I mean, I did read the IPCC report, which is a report on climate change and it comes out, the last one came out in 2023.

And what they found is that actually 15 percent of net global greenhouse gas emissions are related to transport. This of course includes all kinds of transport, but also, one of them is urban transport. And then switching, there are many levers how to tackle this, right? And within general, as in climate change, there's no one solution fits all, but switching from motorized traffic to cycling is one of those means to actually reduce those greenhouse gas emissions.

And coming a bit back to my research as well, what we find also in research and science is that all you can think about, you know, talk to your friends and family and gather some anecdotal evidence, you'll probably find that one of the biggest deterrents that keeps people from cycling is that they actually are afraid because there's not enough cycling infrastructure.

They're afraid of accidents. And that's a relevant fear. We do see many accidents in cities right now, mainly between motorized traffic and cycling, but also all kinds of other accidents. And what we do can do in cities is, to actually promote cycling, is to build more attractive infrastructure for cyclists.

This can include bicycle lanes, a better design of roundabouts. And all this attracts people to cycle, but, it actually also reduces the risk of serious and fatal accidents. So what I really try to do with my research is that again, if we have these heated debates in cities, how we want to distribute space among cyclists, cars, delivery trucks, etc.

I'm trying to provide data on where we actually have how many cyclists in which street, so that when policy makers or transport planners come around, they can use my data and actually make fact driven decisions when and where infrastructure benefits the most, the greatest number of people. And that's how I hope my research can contribute

Chris Adams: Ah, I see. Okay. So that's really helpful. And I think there are some kind of comparisons I can make, which make to maybe help me understand if, and some of the listeners. So you know how a couple of years ago, in the middle of the pandemic and COVID, one way to reduce the number of COVID cases was just to reduce the number of people taking tests, right? You know, that's not necessarily the best way to solve that, know, and it feels like

we've got a similar situation. We've got a data asymmetry problem here it looks like you're doing some work to address for that. I mean, Also, you've spoken about, as I understand it, there are various parts of, like, our economy which are easier to decarbonize than other ones.

Like if, for example, in Germany and in America, transport's the biggest, one of the places where we've seen not so much progress on reduction, on carbon emission reductions compared to things like the energy sector and stuff, which is decarbonizing relatively quickly. So this is what some of this is a reference to. Okay, so what we'll do is we'll share some links to Catalyst, I'm sorry, Catalyse, the paper there, and also some of the papers that you have. So we spoke about Paris, and we've spoke about Berlin, where we both live. Are there any other places you would point people to as examples of, okay, this is what good might look like, and this is one place which actually has quite good data to show where you've actually seen quite effective policymaking to kind of change the environment to make it easier to cycle? Because, yeah, not everyone wants to become a MAMIL, like middle aged man in Lycra wearing the helmets and everything like that.

Silke Kaiser: So, I mean, there are definitely some cities that you know that are popular for cycling, for example, just earlier this spring, I did a research day for some months in Copenhagen. And obviously Copenhagen is a bit of a dream for cyclists, right? I'm not the first one to mention this. And then there are other cities, Amsterdam, you name it, but generally I do have to admit that in my research, I haven't really come across cities that do have much better data.

I would say it's a grasping problem across different cities that data is missing. Copenhagen and Amsterdam have taken political decisions to prioritize cycling, but I do have to admit that I didn't, I haven't come so much across that they've made this as a data-driven decision, but this was more of a political decision.

Chris Adams: Ah, okay. And there's one thing I'll just ask before, Chiara, I'm, okay, I am a closet energy, well, not very closet energy nerd. I'm totally gonna, looking forward to talk about that. But Silke, I was just going to ask you, so you mentioned use of Strava and you mentioned the use of, Okay. It's useful to have these new sources of data, but there's also a question about the provenance of that data and like the circumstances under which it's collected.

So for example, we've seen Strava used in lots of other places and if you're using Strava, you tend to be a bit richer, a bit younger, a bit healthier than most people. Maybe you could talk a little bit about some of that, because there are various sources available to inform these policy sessions, and like, Strava is one.

But like, where else, like, assuming you had, you were suddenly queen of the world, where would you wish you could get some of the data from to kind of inform this in future?

Silke Kaiser: So you're absolutely right, Strava definitely is quite biased. It's the data, for example, for Berlin, I definitely know that they're all male, they're mainly male, young, and they do tend to do very sporty biking in comparison, for example, what I probably do to commute work. So it is true that some of the data we use is biased and we're trying to balance this off with the other data sources that we're having.

We're also taking socioeconomic factors into account because obviously we do not want to have, infrastructure is meant to be there for everyone and not for privileged or less privileged people. It's, meant to be equal for everyone. But then obviously I thought about a lot, well, how could we actually improve the data availability in cities?

And I definitely see two levers that we have. Well, first of all, we can place more cycling counting stations. That is a bit challenging because, for example, we have so many kilometers of roads and it's hard to track them all. There are cheaper options than the ones that we're currently employing. So this might be our one option.

Some of them are then using cameras, for example, that are just much cheaper to put them out there. And then the other question, and that's something I'm also looking forward to, to answer this question is because I'll be looking at this in my future research, is that actually how can we place the sensors that we have better across a city? Because currently look it up again for your city.

You'll probably find a similar image is that we do tend to place these censuses as very busy and scenic roads. And the question is actually, can we maybe place them at more diverse spots within a city? And if yes, how can we choose those streets to place those sensors at to actually get a more comprehensive image of the cycling traffic and then also of all kinds of socioeconomic areas and a more equal data image.

Chris Adams: Cool. Thank you for that, Silke. All right. We'll come back to you a little bit later about some of the specific techniques that we were using, because we spoke a little bit about ML and there's a lot more we might dive into there. Chiara, if it's okay, can I ask you a little bit about your research analyzing European energy data?

Because you didn't hint a little bit about how this can affect renewable energy transition, and one of the, one of the things that Germany has in particular is a target to have 80 percent of the grid running on renewables by in, wow, in five years. So that's not much time, and we've also spoken about on the grid, we've spoken about things like time shifting and location shifting as a kind of carbon-aware software, particularly in changing how data centers, like, fit into the grid, I suppose, or the energy they use. Can you maybe talk a little bit about some of the challenges you've found actually working with this data hands on? Because we At best, most of us developers, we might use it in a really nice, pretty fashion from electricity map or Watttime or in an SDK, but it sounds like you're pretty much at the front end having to figure out how the sausage gets made. So yeah, if I can ask you, maybe you could tell us a little bit more about how this data comes about and what are some of the challenges.

Chiara Fusar Bassini: I mean, if you've been using Electricity Maps, you probably have been using an application that in the back uses NSOE data. So you, in Europe, we are rather lucky because. There have been two regulation that have been released in 2011 and 2013, which forced in a way transmission operators to publish a variety of a time series of energy data from the grid and from the markets in an effort to increase transparency. And we have a number of data in a central, on a central repository, which is called the Transparency Platform. We have load data, we have generation data, we have transmission data about the grids, we have balancing data, balancing markets, but also a lot of information on individual power plants. This data is overall extremely useful, but unfortunately it's not Always accurate and it's not always complete and not all the data is not always published in a timely manner.

That very much depends on the type of data, the country itself. We are still very much better off than other markets where there's no data at all. But it's still an issue of like how good and like what the data quality actually is. Because you mentioned time and location shifting. To do time and location shifting, most likely you will be working with aggregated data.

For example, load data, load data, load forecast data. And. One could analyze, for example, a load to decide whether to shift more energy consuming activities at night or at moment where there are off peak time windows. And on the other hand, one could look at aggregated renewable generation data to try to relocate some more of energy consuming activity to time of the day where the grid is actually greener and there are a lot, there's a lot of interest in academia, but also in, in the industry sector to provide us information, to have an estimate of carbon intensity. there are a number of startups out there. You mentioned Electricity Maps, but also academics have come up with top, bottom up, top down and bottom up approaches to compute really at hourly or quarter hourly level, these carbon intensity estimates. The trick here is that you are working with, the aggregated data and aggregate the quality of aggregated data and the timeliness of aggregated data is rather high, the situation is a little bit different when you move to a more geographically granular, like a higher geographical granularity.

Chris Adams: Okay, so from Germany going to like Berlin or Germany going to another part like Frankfurt, for example, something like that. Yeah?

Chiara Fusar Bassini: Rather, when you're looking from, the aggregated generation to the generation of individual power plants, because in that sense, you might be interested to know which power plants are actually generating right now. And you might be interested to know which areas are generating more solar, for example, which areas are generating more wind. Unfortunately, we don't have data on all power plants. Which would be rather impossible in terms of like amount and extent of the data, but we have only data for power plants that are at least 100 megawatt. It's mostly conventional power plants. So, for example, we have no individual information or very little information on wind farms, for example, because Some of them are not big enough to qualify for this criteria. And also this data get published with a significant delay of four days so that you can't really use it to

do anything operational. It's also not conceived for that. And we can or we cannot use the data to properly, like we cannot do it for, we use it more for analysis than for forecasting, but nonetheless, we can use this data to understand a lot on individual power plant data, why they decide to dispatch on how they are dispatched, and especially in the context of conventional power plants, how their dispatch has changed over time because of political reasons, but also because of the increase in cycling of fossil power plants, because they have to adapt to the renewable energy generation, to more renewable energy generation.

Chris Adams: Can I just quickly stop? I just want to check I understand some of the terms you've used for listeners who might not be familiar with load, cycling, some of these things here. So when you talk about loads, you talk about energy, like basically that's power draw, what people are trying to draw from the grid.

You mentioned that. And then you also mentioned, I think, like cycling. So that's like basically scaling down a power station in response to there being loads of wind on the grid or stuff like that. Maybe is that about right?

Chiara Fusar Bassini: Yes. Thanks for asking. So to clarify, so load means the demand and in the past, I mean, demand, especially from, industry, but also from household, it has been rather predictable. And the way we faced demand, or we satisfied demand, because well, in an energy system, demand and supply need to be equalized at any time. In the past, the most of the baseload, so the main bulk of consumption, have been satisfied using traditional fossil fuel technologies, so called dispatchable, because you can, decide when and how to switch them on and off.

Chris Adams: Ah, okay.

Chiara Fusar Bassini: But the thing is, as more and more renewables enter the grid, they cannot be dispatched whenever, they can only be dispatched, or they can't choose when to dispatch a

Chris Adams: Yeah, control the sun and the wind. We can just respond. Yeah.

Chiara Fusar Bassini: They just respond to external weather factors, right? But that also implies that we still have these conventional power plants that are

dispatchable, but we now have to operate them with increased flexibility. So they have to be able to ramp up and ramp down as the load is more and more satisfied by renewable energy sources.

Sources when they are there. And we always make the assumption that say conventional power plants are a hundred percent flexible, but that's not actually the case. For example, some power plants, when they are turned on, they have to generate a minimal capacity. And if the demand for that capacity is not there, that might be an issue, or they might have some minimal times to be switched on and switch off.

So there, there is a plenty of interesting question that arise from the increase of renewables. Like how will conventional power plants cope with more renewables in the grid?

Chris Adams: Ah, I see. Okay. So one thing you're saying, she's like, yes, it's not like a computer. You can't turn it on straight away, like in milliseconds. And so that's one thing you mentioned and the fact they need to do that more is another issue. And if I understand it, what you described was quite a physical process.

It's like, we're not using bits, we're using atoms, like burning coal, things are expanding and contracting. Like, is there a risk that, you know, the, a big power plant could be damaged some more, or does that introduce any wear and tear when people need to scale something back? Because I can imagine someone saying, "hey, you're making me change how I do things, and therefore you're introducing some risk into this.

That's not what this was designed for in the first place."

Chiara Fusar Bassini: Yes, that's actually a very interesting question. What I mentioned, cycling, meaning that you operate conventional power plants more flexibly, has some consequences on the lifetime of power plants, especially if you keep on turning it on and turning it off. There are some wear and tear indeed for thermal power plants, wear and tear consequences, some of some power plants may not even be able to do so because they have some agreements with O&M managers that tell them, you know, "you can do that, but then you'll have to pay more because you will have, we'll have to do more main maintenance." And also, there are a number of obstacles that arise, especially for older power plants that have not been conceived with this flexibility option in mind, but rather to satisfy baseload.

Chris Adams: Ah, okay, thanks for clarifying that. And you're essentially doing some of the research to see how you might predict some of this better to either reduce or basically accommodate some of these changes that we might have when we've got a much more dynamic grid that is influenced by the sun shining and the wind and all the things like that, right? So maybe if we can talk a little bit about some of the techniques being used to track this and reduce the amount of, maybe, reserved capacity that needs to be done, or reduce the amount of wear and tear that might be imposed on the kind of entire system full of all different power generation. You said you spoke a little bit about using machine learning, and We spoke to Silke.

Silke mentioned that she's using some ML models, which are not like generative AI. That's a very, it's a different kind of AI. Could you maybe talk a little bit about like how that gets used, because you hinted at it, and like what some of the barriers are for using some of that, because it, that sounded quite enticing and interesting.

Chiara Fusar Bassini: Yes, before, before that, I might, I want to add on this. So there is some parallel research being done, especially like at engineering department of a lot of engineers trying to use machine learning to efficiently operate conventional power plants to reduce this wear and tear of wear and tear problems.

And in general, like damages from cycling while still satisfying a change in demand. What I'm doing is and Rather different analysis of historical usage of power plants. So to see how power plants are, have actually been operated so far in the markets, how they're, how they, how flexible they actually are.

Because sometimes we assume that they're 100%, again, we assume how they're 100 percent flexible, but how flexible are power plants that we already have in the grid? And also how available are there in cases of outages, for example, how, like. What's the percentage of time in the year that they actually could provide electricity, for example? And in terms of techniques, well, it's a lot of time series data, so most time series apt methods can be used here. It very much depends on the ultimate task, but one of the major obstacles I encounter is that this high granularity data is by far not as good as the aggregated data, especially, for example, an availability of power plants that has to be reported in a rather accurate way, but then is not one to one translatable to time series format because it's published as market messages, meaning that the data that we have is not in a format that makes it directly usable for a researcher. So there there are a number of obstacles that are really determined by the data quality rather than by the task itself.

Chris Adams: Ah, okay, so it comes down to the data a lot of the time then, basically, yeah?

Chiara Fusar Bassini: Again, like as Silke said, sometimes it's really just a matter of the data that you have, like the research that you can do is going to be determined by the quality of the data that you have.

Chris Adams: Okay, we'll touch on that a little bit later, but I guess the, that does make me think about, particularly in Germany and countries where we've seen very rapid changes. Like, Germany, there's, you know, there's a massive craze of balcony solar, for example, or we've seen loads of battery coming onto the grid, or even Pakistan. We've seen, like, a third of the power, the new power introduced this year, was come from rooftop solar, and each one of those is individually less than 100 megawatts. That's an enormous chunk of power. So there's all this new stuff that we're not, don't necessarily have access to the data for to actually figure out, okay, how will the grid work and how can we make sensible predictions on this? That's useful to know. Brilliant.

Okay. So we're speaking a little bit about the upsides and how, where some of the potential might be. We do speak about Green Software, about reducing the environmental impact of some of this, and obviously when we're doing some of this work, I've asked a little bit about the kind of models you might be using, partly because there's a question, whenever we start using technology to help us meet climate goals, it's when some of that energy is still coming from burning fossil fuels, for example, there's trade offs to be made. Does anyone want to go first, talking about how we think about these trade offs? Because as practitioners, I imagine you're at the coalface, but you're also working with some of the people who think about this every single day. And like, if you're working with Lynn, and like, Lynn was one of the founders of Climate Change AI, I reckon she probably has some reckons and you've probably had some conversations about this, right?

Silke Kaiser: Absolutely. Actually, just, I think just the last group meeting we had, we just discussed about precisely this topic, because it obviously, it is a question that keeps coming, a question that we do want to answer. And it is also like, it's in our minds, right? Because if we want to do something positive for the climate, and then actually, the net result might be negative, because our models consume that much energy.

This definitely is a topic that we think about a lot, I would say. I see Chiara nodding. I think she's agreeing with me, but, and I can see that maybe to the outside world, often this can seem a bit like, I like the term of fighting fire with fire, you know, you, you're trying to make it better, but you're making it maybe even worse, but I think if we make some smart choices along the way, I rather like to compare it to the idea of fighting a forest fire with a controlled burn. So right, that we do try, for example, in, in the models that I was employing, I did partially check how much, how big the energy usage was.

I was using simpler models, as I mentioned earlier. So the energy consumption wasn't that high, but I think it's good for us and for everyone out there using, similar models. To track your energy consumption and there are very nice packages and libraries out there, tools, all kinds of things, open source, freely available that are very good in, in managing or in measuring the energy consumption you have.

And then of course there are a whole bunch of other approaches that you can take. Right? I mean, you mentioned it's an issue if it comes from fossil fueled energy, but obviously you know, you can think about, I know that there are a lot of like. Service and data science centers, for example, out there in Iceland, where you tend to have more natural cooling, where a lot of the energy being produced is renewable.

I'm not saying that at all perfect, but what I'm trying to say is that there are different approaches that we can actually also reduce the emissions caused by AI.

Chris Adams: Ok, so there was one thing about the actual technique, like, AI is not a monolith. There's all different approaches within this, in some ways, not particularly helpful term, like the use of relatively small machine learning models, which are relatively simple, that's going to have a totally different footprint to the model used to generate SOAR, like a video or something like that.

And that's something that we probably would benefit from having a better kind of intuition off as practitioners, for example, and you spoke a little about the carbon intensity. So there's two, two things that you have there. And you mentioned some software that you have. And you said that, Chiara, if I can kind of give the floor to you, because I think you mentioned you, you've spoken about some of this before about, yeah, there are some tools and I use them as well. Can I ask you a little bit about when you've been thinking, I mean, how do you think about these trade offs? Or is it a trade off? Or is there another set of dimensions you might be thinking on rather than like forest fires and controlled burns, for example?

Chiara Fusar Bassini: Yeah, I think, there are two things that need to be thought through when using AI. Number one is like, how do you develop your model, and then what do you use your model for? So how do you develop your model? That's similar to what Silke said, for example, doing emission emission tracking while developing the code and while training the code. And at the moment, I think AI is missing some embedded indicator of the social environmental cost of the training. So. We kind of think of performance metrics such as accuracy, such as like classic cross entropy losses and so on, and we think only about precision. But sometimes we need to be a little bit more critical of whether an increase of accuracy of 0,1 percent is worth an increase in the training time of two hours or an increase in the size of the model of 25%. These are like actual numbers and scientists have coined for that the term green AI, meaning Okay, can you know, can we, in a way, embed this measure of the size of the model within the loss that we are trying to minimize in the training of our model? There is another, a good example, for example, is the Bloom model that is an alternative large language model to GPT. It is similar in size, but it required Like the CO2 emissions of the model are 20 times lower than GPT 3. And this has been made possible by, first of all, in smart usage of the training and also tracking of the carbon intensity of the grid. It system was trained, the model was trained mainly in France, which is, which has runs predominantly on nuclear power. So in like carb, much more carbon neutral system. So there are a number of things that one can consider while training their model. But also another thing that is very important, and I think that we sometimes don't really think through, is what are we using AI for? And in that sense, there is currently no standard assessment in place. Like, is this application really worth using AI? AI is, by its nature, ethically neutral. It can be used for anything from targeted advertisement that will have probably a negative impact on environment to detecting wildfires. So very positive impact. I think policymakers in that sense can make a great deal to really make a difference and start, for example, by providing a classification of which user cases are positive for environment and which are negative. It sounds, it may sound like science fiction, but it has already been done in the European AI Act in looking at the perspective of risks, like which application have a high risk and hence should be more controlled and which other have lower risk. And I think a similar classification would be also very useful for environmental purposes.

Chris Adams: I'm really glad you mentioned that because I ended up reading through the AI act for research recently. And the idea of the risk that is, you're right, there doesn't seem to that much be that much reference to the use of AI for, let's say, you know, increasing the extraction of fossil fuels, right, versus that.

That's, there doesn't seem to be much to mention about that, but there is some information about the transparency around training. And now that we've looked at it a bit closer, so within the Green Software Foundation, there is a group called the Real Time Carbon Group. We've been looking into some of the specific implications of this, and it looks like the AI Act, it also, it looks like it's probably going to suggest not just understanding the training, but also the cost of inference, like the use of the model rather than just the training of the model. If I can just quickly, you've, you, mentioned there are tools out there, and Silke also mentioned there are tools out there. If I did want to measure some of this, and if I did think there was some legislation coming for this, what tools are there available for me to measure the direct impact? So at least I know what the trade off might be.

So we understand that the carbon footprint of decarbonizing transport, like Silke mentioned, that's going to be, you know, positive, but quite, but there's ways of calculating that, but for us as practitioners, are there any software or any tools you might recommend that are kind of common in the field now? Either goes. I'm happy to, whoever's more comfortable talking about this.

Chiara Fusar Bassini: I'm thinking CodeCarbon is more probably a standard used by many scientists. I know there are more applications that might have a higher granularity, but I guess that's a

Chris Adams: That's the one that you folks have used, right? Okay, I hear CodeCarbon used a lot, and I, as I understand it, that's the one that's been used for the Bloom model when they wrote a paper about that. That's what I'm not sure Facebook have actually explained this because when I was looking at LLlama's model, so AI models have model cards, which basically, which I think, various responsible practitioners now say, "this model took this much carbon, or they had this much energy gone in to kind of create it," for example, if you go to the existing Llama 3.1 model card on Hugging Face, and you try to follow a link to the actual methodology, It's not actually explaining how it works. there's now a bug. I filed a bug to ask out, ask, well, how did you work these figures out? Because these feel like it's quite important, especially because when you look at the numbers, it's significantly larger than Bloom, basically. That's, so, so what you're referring to is CodeCarbon. That's one tool that people can use that will give you some idea that is in use in a few places already that's relatively safe to start off with. Great. Okay. And we spoke a little bit about some tools. So if someone is, wants to take their first steps, they might look at this.

And there are various projects I'm aware of to make it a bit easier to understand the impact of one versus another. I believe there's one Energy Star AI or Energy Star, AI Energy Star or something like that. There's one person who I've spoken to who's involved with it. Boris. I'm so sorry I can't pronounce your surname, but I do know you're the AI lead, the AI sustainability lead at Salesforce. Boris G is one of the people who's been writing about this. He's not the only author, but he's the person I know, and we'll share a link to that as well, because that's the first thing I've seen of a useful, like, A kind of nice idea to give you an idea of what the inference, the usage as well as the training might actually be. If you were to look at this, we've still got this issue of data or having access to data like, and Silke, I ask you, if you were like queen of everywhere for a moment, how would I change it for here, right? Let's say that you want to be responsible AI practitioners, like what are the things that we need to see in the next, in the coming years to make it possible to be like responsible practitioners so that when we do use AI, we're using it in the kind of greenest possible fashion. Silke, I asked you first about Queen of Everywhere, so maybe you go first and I'll hand over to Chiara.

Silke Kaiser: Well, that's a very good question. I definitely say, as in general, with all kinds of, you see, in more technical approaches, we do need reproducibility and traceability of what we do in our research. I mean, just as you mentioned with the Llama, I think it's important that other people are also able to understand what we did, what was the energy consumption of what did, how can they, how can we check the things that we've done and, see if, we did it properly, if it took a right approach?

And then obviously, I think this is a bit less related to, the topic that I'm or that Chiara was working on, but also in the longterm, we do need to think about ethical concerns coming down to this. And then again, I think just, really, transparency. So I really think that transparency is a good way to address this.

What take do you have on this Chiara? I think one of my major takes also from what Silke mentioned, and I'm really glad you mentioned, is the fact that when we were talking about policy making is that very often policy making is not data-driven. One problem is that we don't have the data and it can be addressed partly by regulators asking for those data, right? But another issue is also that we don't really do data-driven assessment of the policy that we implement then. And I came across very recently a paper that tries to systematically evaluate policies.

And having been implemented in the last 25 years, this very recent paper has been published like a few days ago.

And I thought it was very interesting to well, once again, the results of the paper is there's no one size fits all and some countries depending on their level of development might need different policies. And we have to keep that in mind that we can't use the same policies for a developed country whose energy consumption, for example, is no longer linearly dependent on its GDP from a different, from a developed country or a developing country that has very different issues. But I think in general, this approach of doing data-driven policymaking and science-driven policymaking is something that would really, we would really need in this space.

Chiara Fusar Bassini: I think that's something I can really agree on. I often feel that as a scientist, we feel like we're trying to really produce clear results, objective results. And then often we feel there's maybe a bit of a lack between the two. The research that we do and how much this is sometimes uptaken, by policy.

And obviously we hope that because we do really put so much effort into this and always try to be objective. We hope that this will eventually be more used more and more in the policy sector.

Chris Adams: You've touched on a really interesting point, and I can think of some examples that just occurred to me. So, we had an interview with, oh, Vlad Kor, his first name is Vlad, I'm gonna mispronounce his surname, but we spoke a little about, all about the rebound effect, and Vlad Coraoma, that's it, Vlad Coraoma had this lovely post actually on LinkedIn talking about the curse of potentialities, potential itis, which is basically talking about, we have all these kinds of really exciting projects, but whether people follow through to check whether the actual gains materialised, or the benefits materialised, there's much less effort put into that.

And we've been seeing, like you said, Chiara, from the last 25, we've been seeing predictions for things that would happen in 2020 or 2030. And 2020 is in the past now, we can check if this is, if these actually delivered, but a lot of the time we do not see that. And in our field, specifically as kind of cloud providers, or people who might be consuming services, there's some, there's a really, I'm thinking of a really good example. Microsoft has a whole thing about pushing for AI and everything like that. And we know that, as you mentioned, AI can be used for good, and can be used for bad, or used for Climate aware things, which are really helpful and things are not so good. And we've even seen like people who are workers really pushing for this. I'll share a link to an article in Grist where, written by Maddie Stone, where she talks about some of the sustainable connected community inside Microsoft, speaking to some of the management there. There's a guy called Darrell Willis. He's the vice president of energy. And they spoke and said, "hi, we are pushing for," you know, "can we please have a conversation about what we're using AI for inside our company, because we're one of the largest companies in the world and we're one of the leaders in various industries," right?

And there was a commitment to say, we're going to produce, as the management said, "we're going to start releasing information about, okay, how much of our use of AI is coming from the fossil part of the industry versus the renewable part of the industry?" And this feels like a really important data point if we're going to be looking at tens of billions of dollars used on AI.

I mean We know that it's an accelerant. If it's an accelerant of fossil fuel extraction and burning, that's a very different story to using tens of billions of dollars for renewable energy, for example. And if we've seen commitments at a management level, then it would be nice to see these. As we understand, these commitments were made, these were shared inside the team, but we don't have this, and we'll share a link to the specific terms, because actually, I'll just share the quote with you, because I think it's one thing that, if you're an employee of a cloud firm, or a customer of a cloud firm, it's the kind of thing you might want to know about, so on the call, "Darrell Willis, committed to providing employees with updates on net zero requirements as Microsoft continued to implement these energy principles. Committed to providing a breakdown of energy divisions revenue across six different sectors from oil, gas extraction, to all zero, low to zero carbon energy. So sharing this information internally." Now this feels like a thing that employees probably should be aware of or asking for. Also feels like something that if you're an investor of Microsoft or a customer, you might want to know.

Because there's an impact inside your supply chain thinking about this. And if you're choosing one provider because they have really strong GSG credentials, this may make you view it somewhat differently. We'll share the links because it seems to be the best concrete example I can think of at significant scale that we might be talking about. And I'll get down on my soapbox because that's just the thing that really leapt out when I, when you spoke about that. So we coming up to time, and we've spoken about the different uses of AI, sustainability of software, as well as some of the Things you might want to use or be aware of as a practitioner. If people do want to find out about the work that the two of you are doing, where should people be looking? So Silke, if people are interested in your work, is there a LinkedIn page or is there a website that you direct people's attention to?

Silke Kaiser: I normally try to direct people to my personal webpage, which is silkekeiser.github.io. Or you also, you can also find me on X or on LinkedIn. And I'm always happy to share news on my research as well as the articles that are out there. And I'd be happy to, if people were to look at those pieces of information.

Chris Adams: Cool, thank you. Alright, and Chiara, if I just hand over for you?

Chiara Fusar Bassini: I've seen Silke's website and you guys should really see it.

It's a very nice animation. I don't have myself a website, but I'm very active on LinkedIn. You can find me under Chiara Fosar Fusar Bassini.

Chris Adams: Chiara F U S A R, we'll put it in the link, we'll add it in show notes. So, Chiara Fusar Bassini. Brilliant. Thank you, folks. This has been lots and lots of fun. I've learned a lot from this, and this has been a really nice chat. Hopefully, we'll cross paths sometime in Berlin, but otherwise, thanks again for coming on, and have a lovely week.

Silke Kaiser: Thank you very much for having us.

Chris Adams: Ta ra! Hey everyone, thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

Green Software with Gaël Amongst the Whales

22 augusti 2024 | 67 min

TWiGS host Chris Adams is joined once again by Gaël Duez to discuss the latest news in green software around AI. They discuss insights from recent reports by Google, Meta, and Amazon, as well as looking at the implementation of the GSF’s Software Carbon Intensity metric. Similarly, the conversation touches on the distribution of renewable energies and the use of different means of measuring carbon in reporting, and how this can affect the behavior of consumers and organizations alike. Tune in for an enlightening discussion on the latest in green software.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Gaël Duez: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Green IO London 2024 [1:01:15]
CNCF Cloud Native Sustainability Week 2024 [1:05:48]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Gaël Duez: We cannot fully rely on companies agreeing on how they should measure their own environmental impact, even if they are well meaning, with tons of great people trying to do the right things, etc. It's not a black and white world out there, but there is a question at some point of financial pressure, shareholder pressures in many of these companies.

They're just stronger than the entire stakeholder's pressure.

And the ups and downs of consulting remotely on digital sustainability from a small island off the coast of East Africa. It was a fun chat. And when I asked him if he'd be up for coming on again to review a few of the latest stories around green software, he basically said, "yes, Chris, but I can't do Friday because I promised my daughter we'd pop out to go whale watching." As one does on a tropical island, I guess. So here we are, recording on a Monday morning instead. Gaël, thanks so much for coming back on. Can I give you a bit of space to introduce yourself and what you're up to these days? Also, how are the whales?

Gaël Duez: Oh Thanks, Chris. It's great to be back again on Environment Variables. And the whales were there, like several made quite a show, including a mother and her little newborn. Well, little meaning four tons. So it's always impressive to see there's a 20 tons mammals jump in the air like this and cherry on top of the cake, actually, we saw a lot of dolphins and two turtles.

So it was a perfect trip.

Chris Adams: Wow, I'm jealous. Okay. Yeah?

Gaël Duez: Actually, to connect a bit more on our sustainability topic, I'm also relieved that the best practices for whale watching are more and more enforced, such as, minimum distance to approach them, turning off the engine when they come in, or direction, et cetera.

And, I also say, on top of the pleasure of, watching them and being on the boat, etc. It's a very positive sign that we can enjoy nature without destroying it. And that's pretty cool.

Chris Adams: Cool. So it sounds like there might be, hopefully, when your daughter's taking, maybe some of her kids for whale watching, there's a chance to continue that, by the sounds of things.

Gaël Duez: Yeah, I'd like to focus on whale and dolphins rather than coral reefs, which are something that really is puzzling me, but yeah, we might, hopefully. And to answer your other question, when I'm not whale watching in Reunion, I try to be useful in the tech community, by advocating for more sustainable ways of designing, coding, hosting, and even considering the use of technology itself. And my main tools remain the Green IO podcast, as you mentioned it, as well as the Green IO conferences, which I started to organize in several cities, but I'm sure we'll get back to this point later. And besides, as you already mentioned, besides as volunteering activities, I do public speaking and consulting on systemic strategy for mostly for tech companies, for both paying the bills, but also to have an impact.

Chris Adams: Cool. Thank you for that. I will also share a link to the previous episode if you're for people who are listening, so they can get an idea of some of the other things that Gaël has been working on and been discussing previously. If you're new to this podcast, I should probably introduce myself as well. As I mentioned, my name is Chris Adams. I am the executive director of the Green Web Foundation, which is a Dutch nonprofit based around reaching a fossil free internet by 2030. And I also work as one of the policy chairs in the green Software Foundation, particularly in the policy working group. Okay, and before we dive into this podcast fully, just a quick reminder, everything we refer to, every story, we'll share in the show notes and there will be a transcript as well for you to kind of search through and look into later. Okay, so as per usual with this week in Green Software, We run through some of the latest stories or projects that have caught our eyes over the last few weeks. And Gaël, I should probably ask you, are you sitting comfortably, ready to go?

Gaël Duez: Perfect. Everything is fine.

Chris Adams: Brilliant. Okay. Let's start with the first story then. So this is the first one that kind of came up on my radar. There's a website called Green Coding AI. And we've spoken about the environmental impact of AI on this show multiple times.

We've also spoken about there is different models that you can actually use to maybe ask the same question and get something back. This one is actually a project from, the Berlin based group, Green Coding Solutions. And what they've actually done is put together a service, running on their own hardware, where you can basically try various, models to prompt them to, like, try out, say, Llama 3, one of the big versions, maybe a smaller version of some of these to see, to ask a question and see what kind of responses get back. And one of the things that's particularly interesting in my point, from my perspective is A, the transparency, but also it gives you a software carbon intensity score for every single inference when you ask a question. So you can start comparing the utility of a large model versus a small model. Gaël, I think you might've had a chance to play with some of this already.

When you looked at it, what, were there any things that sprung to mind for you when you first had a bit of a kind of kick the tires and mess around with it at all?

Gaël Duez: Well, what I was pleasantly surprised with is, as you mentioned, the use of the software carbon intensity. I, as far as I know, this is the first time that I've seen it being used for this kind of tools. I know that what is very interesting is that we can see now, not blossoming yet, but several initiatives have popped up with that kind of approach and I need to shout out and at least I need to give a big kudos to Arne and his team because once again it's very thoroughly done for what they've done in Berlin.

What I also like is the work that has been done with another initiative, and I hope that they will start discussing, with each other, which is, it used to be a data for good, project, but now it's a, an, an association on their own called Gen AI Impact. Yeah. So the association is called Gen AI Impact and they created this tool Ecologits, which is kind of just kind of the same approach.

With a very strong focus on transparency, how they calculate everything and all the code is accessible, et cetera. And there is also obviously Code Carbon by Jürgen Fais, which was kind of the trailblazer, if I recall. And I think it says something positive about the trend in the AI industry, that they need to tool up to Assess more and more precisely what is the carbon impact and hopefully soon the water impact.

And that's still, I know that it's not feasible at the moment because of the lack of transparency around water, but we will discuss water later in this episode. I'm pretty sure regarding the other news that, that we have, but it, is very important. I cannot count how many times I've got potential customers or heads of IT telling me what are the tools we can use to measure the footprint of AI.

And Also to use them to push back a little on the AI hype. everything should be AI'ed. I don't know about it. With a pinch of AI everywhere, but it comes with a cost. Now, going back to your question. So I really love the multiplicity of tools that we see at the moment. And now going back to more specifically in Green Coding AI, I think the way they, try to cover all the angles when it comes to energy, having both the CPU joules, the GPU joules, but also the temperature.

It's very interesting because the temperature, it connects a lot with a discussion I had with Professor Lee. One of the big experts on water cooling, especially in data center in tropical island or in tropical climate, sorry, not island. I'm the one on a tropical island, but there are a lot of tropical climate without tropical island, starting with Virginia in US that we tend to forget.

And he was telling me how important it is to understand that there are thresholds when you use a chip. And there's a threshold that impact massively the energy consumption. So it's not the same to have a chip being used around 25 degrees Celsius, for instance, and 30 degrees Celsius. It's just a linear progression in energy consumption versus, temperature.

And I was really delighted to see the temperature being put as a metrics, because that's where we are getting more and more in details and understanding that it's not that easy, it's not that linear, and we need to investigate things in a more systemic view. And the temperature for many people operating data centers, is absolutely pivotal when it comes to anticipating energy consumption.

So I think, yeah, I was very pleased with this CPU temperature metric that they added. I have no idea how they made to calculate it. So that's something I'm going to investigate, but I like the idea of putting this kind of metrics on the table as well.

Chris Adams: So, thank you for that, Gaël. So, what I can share with you, there's a chap, one of the chaps working at Green Coding Solutions, Didi Hoffman, was one of the people doing a significant amount of the work on this, and I've used some of their tooling, they've made open source before, and they basically pull some of the temperature figures from the, from basically the CPU itself, I mean from the system that they use, because for example, most machines can tell you how hot or cold they are, for example.

And that's for pulling some of those numbers actually from. The thing that I might share with you that I think is interesting here is, normally, Green Coding Solutions, they tend to open source as much of the work as possible. And they have one of the nicer tools out there for looking at the environmental impact of pretty much any kind of service, for example.

And the thing that So the, I did actually ask Diddy about this and said, "Diddy, this looks really cool. Are you planning on open sourcing?" He says, "yes. The only thing that has stopped me open sourcing it so far is tidying up because it is a bit of a mess." But the thing we can point you to in the meantime is a paper that he worked with, I think Professor Verena Majuntke from HTW Berlin. They submitted a paper specifically for, to Hot Carbon talking about, essentially being able to optimise different models for optimize the use of AI services specifically for particular tasks. And that paper is really interesting in my view, because they say, well, given these particular tasks, some tasks are amenable to running on small models.

And some tasks are better for large models, for example. And they were basically demonstrating how you can pretty much use a system, which looks at the prompt and then will kind of route it to the appropriately sized model to reduce the environmental impact of this. But one of the things they said was that they can look at the energy used when making an inference like, asking for a response back, but it's still a massive challenge to get any numbers from the embodied energy in making some software. And this is one of the key things that we might, maybe wasn't so obvious when we first spoke about this, that right now there are kind of three broad life cycle stages in an AI project. There's a training part where you create a model. There's an inference part where you have the use of the model. And then there's the kind of embodied carbon around this. And this is one of the first projects we've seen that really makes the figures for the inference part quite visible and quite easy to work at, and one of the key differences between this and the Ecologits project that you just mentioned before, was that this is actually running on the software itself, so they have access to the hardware, they're running the hardware themselves, and they, actually document how they do this. The, when I spoke to Samuel Rince at GenAI, I asked him, "how does your thing work?" Because that's really cool, I really would love to see some figures for the inference prices. Well, what we do is we have to make some educated guesses and estimates based on how big a model might be, how much memory might be allocated for it, and how long it's run and how big the response is.

So they're essentially annotating responses that come back from Open AI or from Mistral or from Claude or anything like that to give you some numbers for this. And this kind of does beg the question, if we see green AI instrumenting their own physical hardware, and if we know that the best we can see elsewhere is us having to instrument things ourselves. Based on guesses, why is it such a challenge to find services that provide these numbers as part of how they work? This is still a challenge in 2024. And I think these projects here demonstrate that yes, there is demand and there's also ways to do it. So if these aren't being exposed, they really should, because it becomes much easier to be a responsible technologist if you have access to the figures about the environmental impact of what you're using AI, for example. So yeah, those are the things I'd share, and I'll add some links to that to follow on from this.

Gaël Duez: I wasn't aware that they were using their own materials, and that, that's a massive change indeed, because no more educated guess, as you said, and the ability to measure yourself. Of course, it means that they create their own assumption when it comes to the hardware setup, but Yeah, I really love the idea.

And I have a question for you because I wasn't able to unpack yet all the publications from Hot Carbon. So how was this article aligned or not aligned with the last one from Sasha Luchoni when she was actually testing? So once again, on the inference side, because we know that this is where most of the impact comes from, when she was categorizing the use of a different LLM

in front of different use case and, she was, for instance raising the alarm that for speech recognition text or other, I would say, basic AI solution, using large LLM was a complete waste. And that's something that really struck me that how important it is to pick the right model for the right task and not using LLM, especially GPT 3 or GPT 4 for everything.

So I was wondering, do you know how well that was aligned or not? Their own findings.

Chris Adams: Okay, so it's been a couple of months since I read the paper, or about one month, because the actual paper was released, I think, on the 11th of July or

something. The key thing was this paper from Verena and Didi, the people at the University HTW Berlin,

and, Green Coding Solutions. Their one, while they do talk a little bit about classification of tasks, they don't go into the same detail that the paper from Sacha Luchoni, which was, that was the one that we saw in MIT Technology Review, which very much said, generating an image has the same kind of energy demand as charging a phone, for example.

So they don't provide that same kind of breakdown because their primary focus was working with text. However, there is something we should share a link to, which is some recent work in April where the, where Sasha Luchoni and other people, other luminaries at Hugging Face spent some time working on what you might refer to as like Energy Star for AI.

This idea that for different tasks, you do have models and so you can start making some more informed choices about your choice of model when you're doing some of this stuff. So we should share links to some of these things because

there is quite a lot of work taking place that I think not everyone is aware of right now and it's a really useful kind of jump off point for this.

So yeah, useful questions Gaël, thank you.

Gaël Duez: It connects well with what you've mentioned, that there's a rating should be provided by the one making these models or operating these models. And it's great to see all those initiatives, but eventually, and I think this is a nice connection to the next stories, but when you operate something, you need to be transparent about the environmental footprint of what you do.

Chris Adams: So this is actually, you're right, that's a very good point, Gaël, and maybe just the final thing I'll share with this before we move on to the next story is that for training, we've seen pretty much the industry start to settle on the use of CodeCarbon, which is an open source Python project that you can use to essentially work out the energy use for any piece of Python code, but specifically it's used for quantifying the energy usage for training.

So, this is what's used for some of the papers by, from Hugging Face. I believe this was also the tool that was used for the most recent information from Meta when they shared the carbon footprint of Llama 3, their big model as well. So you do have some existing tools that are out there. One thing I've actually tried looking at recently was I realized that you can use these tools and there are tools like GitHub and GitHub Spaces which allow you to run a virtual machine, leave and run, like some inference locally, for example, to try and like test something out, for example. And, what I found was that I was preparing a workshop for, to deliver at DjangoCon in June earlier on this year to help people, like, figure out, okay, what's the environment impact of maybe using an AI service or figure out what some of the kind of service side environmental footprints might be. I found that, there's a competitor to GitHub spaces called Gitpod. They use a slightly more up-to-date version of Ubuntu, which basically means that if you're running a virtual machine inside this, you can actually use Code Carbon and get numbers back. But when I try to use this with GitHub spaces, because they're using a slightly older version of the underlying operating Linux that's used for this, you can't get the same numbers back. And I think this is important or worth being aware of because there's a recent release from GitHub, I think. I'm not sure if it's totally available for everyone yet, but there are now some tools specifically for using inference in GitHub spaces specifically. So it'll be really lovely, and I know I'm kind of nerd sniping the GitHub team here, if they could expose some of these numbers, because the tooling totally exists now, and the bar is so low that even just having something like CodeCarbon, which is useful, but has some, when you look into the details, has a few kind of compromises and few issues. That would still be massively more useful than what we have available to right now. So yeah, there's there's definitely worth, there's definitely tools out there that organizations can use to make it easier for consumers like yourself or me to use these in a more kind of responsible fashion. All right. Thanks, Gaël. Should we look at the next story now?

Gaël Duez: Let's do it.

Chris Adams: All right. The next one. This is a story from a former VP of Sustainability Architecture, I believe, at Amazon, Adrian Cockcroft. he's written a piece for The New Stack and the title is, How did Amazon, Azure and Google perform in 2023 Sustainability? So this is a piece by Adrian Cockcroft where he's basically read through the three sustainability reports. And as someone who actually does have a significant amount of context working on the platform side as well as, since leaving Amazon looking at the tools out there and trying to collate a kind of like useful data set with the Green Software Foundation Real Time Cloud framework, he's been able to say well these things which are easy to use this is where some of the data is helpful this is where there are real challenges and he's it's really useful to get to have someone who, in my view is very much seen as like a kind of real kind of trusted message of saying look,

these bits are okay, this is where the bar is really low and we probably should be expecting quite a bit more given the amount of resources available to people. And yeah, I wanted to just check, is there anything that caught your eye or that really leapt out at you when you were looking at this? I really like this piece and I'm really glad it's actually out in the public domain.

Gaël Duez: Yeah, I love Adrian's work. I love his optimism. I'm not, I would be a bit more cautious is when he, well, especially regarding Amazon, but I guess we're all biased at some point. And what caught my eyes was we discuss a lot numbers that are not that easily assessed and separated. The very big first issue that I had is when we talk about Amazon, it's not the main, the same thing that when we talk about Google or Meta, because, or Microsoft, because there is this big on premise brick and mortar, as we used to say, chunk of Amazon's carbon footprint.

And what is strictly related to AWS should be extracted. And that's not the case with all the numbers.

Chris Adams: You're referring to AWS and

Amazon the retailer, like there's

been separate business, that's what you're talking about here. fact they're not breaking down makes it harder to understand, right?

Gaël Duez: It's, very hard because it's a bit like if we were discussing the, Fusabi report of Google plus Walmart or Sainsbury's or whatever, or Carrefour, and, I'm always very concerned about how globalized the data are.

And I've, got some cave hats with what Adrian said. I won't challenge the numbers or his analysis because I think it was well done and the trend are there. Like AWS, Amazon is doing a bit better and Google and Microsoft are slipping up as the, as he mentioned, but, focusing a bit more on Amazon, for instance, there are a few things that I'm a bit concerned about.

So the first one is that they gather everything in this big Amazon and what actually as techies, we would like to understand better is AWS on its own. I think it's big enough to have its own sustainability report. The second is that they continuously provide numbers on the market-based approach, especially for energy. And I think that there are now countless examples where it's not really how Sustainably is done. Sustainably is as much a global matter than a local matter.

And I'd like just to take the example of Ireland. So if you run as many European techies, your instance on AWS, there are a great deal of chance that by default, you will be using the Ireland region. And when you log in the dashboard or your Sustainably dashboard, exactly as Adrian mentioned, you will see that everything is fine.

You're zero percent, you're carbon neutral, everything has been offsetted, and ciao, bye bye, well done, you can, do business as usual. Now the reality of the Ireland electricity grid is that one year ago, in 2023, the amount of electricity consumed by data center equals residential urban residentials.

It means that every houses, every buildings in Ireland consumes now less, a bit less electricity than data centers. So it has put a tremendous pressure on the electricity grid and the Irish electricity grid is not the cleanest or sorry, the lowest carbon on Earth at all. So, technically speaking, when we add resources, when we add instances on AWS Irish region, we are adding pressure and pushing the Ireland electricity providers to emit, to produce more energy, which is kind of high intensive energy.

And now you've got this market based approach, which has is it's prone. I'm not like, it's not black and white here, but saying, okay, but we invested energy elsewhere. And we show it either by a, power purchasing agreement or AAC. And so that's all good on the market because everything shall be offset.

But the local realities matters and that's even more true for water. But let's put that aside for the moment. So as long as they don't at least try to localize a bit more the carbon emission and the related energy carbon emission. I think it will be always very hard to say, okay, the trend is okay, the trend is not okay.

So that's my first issue. My second issue is that, and I think it was our dear friend from SDIA, Mike Schultz, who once said, one of the most precious resource on earth today is renewable energy. Because of course it's growing, but we don't have that much. And we should always question how much we allocate to which use.

And by having this 100 percent focused on offset or net zero approach, that is the one from Amazon, Google, etc. We cannot leave the elephant in the room, which is, but what are the absolute numbers? And when the absolute number are getting higher and higher, almost from a logarithmic perspective, it's almost exponential, not fully, but almost, we should question ourselves, but where is the limit?

Because we do know that in systemic and in environmental ecology, there are, there is always a limit to how much a system can grow. So, that's my two big issues. It's not localized enough. And it doesn't talk about absolute values. It only talks about the potential of things being offset or being carbon neutral.

And we need to think more about when we slow down or even we reduce our energy consumption. That's not on the table at all. So yes, of course, there are a lot of progress being made. They buy a lot of renewable energy, but is it the best use what we can do about renewable energy? And what are the trends? I know I could speak for ages about it, but sorry, Chris, and I didn't even mention water.

Chris Adams: So just, so let's just check if I understand the key things you're referring to there. So one of them is, just to be clear, we're talking about the Republic of Ireland, as in the island of Ireland here inside Western Europe.

That's we're talking about here. And if I understand what you're referring to here, there is one of the big things, big parts of this story this year is that Amazon has made a big song and dance about saying "yes, we are now 100 percent renewable powered for all of our infrastructure." And what you're, what it sounds like you're saying is that The physical reality in Ireland doesn't necessarily match this claim because it may be that the kind of the way people are substantiating this green claim is that they're basing this on credits like renewable energy credits and while these may be kind of considered kosher or like considered like legitimate in like maybe a trade electricity trading market kind of sense, the fact that we don't see the actual location based figures for these data centers brings up all kinds of questions. And also, the, there is also questions about, are renewable energy credits the correct way to actually basically back up any claims around the use of green energy, particularly when we know that the underlying grid, there may be more power being used than it actually, than renewable energy is actually generated in Ireland itself for this, right?

Gaël Duez: Absolutely. And then to, give them credits, they use less and less, renewable energy credits, which are highly questionable tools, and they use more and more PPA, which are Purchasing Power Agreement, where actually they commit to add new renewable energy via partnership, long term partnership with electricity producer.

So it actually increases the amount of renewable energy available for everyone on the grid. So I'm not saying that everything is bad or everything is great. My question is if you, for instance, just staying within Europe, invest in northern Germany in a wind farm to produce that amount of gigawatt of renewable energy, that's great.

That's necessarily, that's something that is very useful for the German market and for German users, but it will not offset the fact that there are still gas and even, correct me if I'm wrong, coal based port plant in Ireland, and that the use, the rise of energy use in this part of the world will emit more greenhouse gases.

So, once again, it's the incredible ability of humankind to tell itself stories, which has made us what we are today, has also a dark side, which is it's not because we decide that we create a fancy story called the market or the energy market, et cetera, that it is completely disconnected of the physical reality of thing, as you mentioned.

And the physical reality of thing is, it's great to add more renewable energy to the grid every day. Anywhere on earth, because anywhere on earth, we need more renewable energy, but it cannot really offset or compensate the fact that if we put some stress on electricity grid somewhere, it will add the emission of greenhouse gases and, eventually everywhere around the world, because I think it's any like carbon molecule that take 15 days to do a round trip.

So, it's a global challenge.

Chris Adams: If I understand what you're saying, basically, the instruments being used do not fully capture the physical realities of what's taking place. And while there may be progress, we probably need more progress in order to actually face the challenges that are being kind of set out by the actual, the real science that we're seeing. I'll share a couple of little points around Ireland specifically before we move on to this. So Ireland is actually one of the few countries where Green and IT claims around green energy have actually been challenged by the Advertising Standards Agency, specifically saying if you're a green energy firm and you're saying you're using green energy. We've, there have basically been cases where the Advertising Standards Agency in Ireland has said, you can't make these claims in Ireland if you're using just renewable energy credits as the basis for making this claim. So that's one thing we've seen. And that has interesting implications for technology firms that are using these green energies if they're substantiating their kind of claims around green energy by using these certificates.

If you've already had a ruling saying, "nah you're not allowed to do that." The other thing that surprised me, when I was looking into this, because the Renewable Energy 100 is a ranking of the top of a large number of firms who are significant investors in renewable energy. They actually don't accept the use of these kinds of renewable energy credits if they're not physically deliverable.

And one of the challenges you see in Ireland is that there's a limited amount of capacity to move the kind of like green energy that might be generated elsewhere in the world to there for this. So that's one of the challenges that you see. And we'll share links to both of those two things because for people who are kind of wonkish and want to get down to some of the bottom of this, they're really, I think they provide some interesting background to this.

We'll also share a link to the real time, to the Green Software Foundation Real Time Cloud dataset where there's been a bunch of work into trying to find some location based figures for this stuff so you can come up with some more accurate numbers than what we're seeing here. And I think, okay, I'll leave the last word with you, then we'll move on to the next story.

Go for it.

Gaël Duez: There are two things that I'd like to give credits to Adrian in his article. It's like stressing how much there are two sides of the story. And there's reports that focus a lot on sustainably of the cloud. And that's definitely what Amazon, Google, et cetera, are trying to do. But there is also this question of sustainably in the cloud, which is how as a user I can do or not a better job mitigating, reducing my carbon emissions, my water consumption, et cetera.

And he's right to say that not significant, no significant progress has been made on Amazon side and on AWS side story. And they are still infant phase at Google and even at Microsoft when it comes to transparency. And as a CTO, as an software engineer. And when you look at these dashboards and you see that everything is fine, everything is offsetted, you've reached carbon neutrality, it doesn't empower you to do the right things, which is optimizing, reducing your carbon emissions, your water consumption, etc. So that part, empowering consumers is still lagging of what we should expect from these tech behemoths. And my last comment is that I was very pleased that he mentioned and he reviewed, thoroughly the water consumption because for water and that my message about global versus local, it doesn't really matter.

It doesn't really make any sense to analyze the water consumption in terms of global consumption. It's water is a local matter. And it's really region per region, even data center per data centers. How much water comes in? How much water comes out? And in which state? Is it reusable, not reusable? Is it a closed loop or not?

In most of the data centers, including the one from the hyperscalers, are far from a closed loop. I know that Google has experienced once and they told quite a lot about it and it makes total sense. But we need more. precise and localized information on water. And that's a massive challenge as well. We focus a lot on carbon, but water is the next big issue that we need to pay attention to.

Chris Adams: Alright, water, that's the next horizon. I'm going to park that because we'll come back to it a little bit later. The next story is actually from the Financial Times. This is talking about Big Tech's bid to rewrite the rules on net zero. Now, at the time of this going out, it may be that the really nice looking piece may be hard for people to see, but no, the link does seem to work actually still, thankfully. The Financial Times has a really interesting piece, basically talking about the large technology firms that we often see coming up again and again. And this is a bit of a deep dive into some of the things you just referred to about like location based carbon footprints for electricity, because that's one of the key drivers of emissions for our use of digital services, and the market based approaches. And this pretty much dives deeply into something of a bun fight that's taking place where you have two kind of schools of thought where there's one set of companies like, to an extent, Microsoft and Google are pushing for this notion of 24/7 renewable energy and are having a quite kind of tight accounting process. And then you have another approach being largely put forward by Meta and Amazon talking about their kind of emissions first approach saying, no, what we should be looking at is decarbonizing the entire grid, not so much looking at our carbon footprint. And there's a couple of things that are really interesting inside this.

There's a few nice interactive graphics for you to see how people make green claims around energy usage. But one thing that I think is actually really stark is this set of charts showing the difference when you try looking at these figures. So, if you were to look at, say, the carbon footprints from, say, Microsoft, you can see, like, from 2018 to, like, now, you've got a figure of maybe, you see one chart showing the market based footprint, which is, pretty close to zero for Microsoft and close to zero for Meta and likewise for Apple. And then you see the location based figure for Microsoft. It's something in the region of like 8 million tons or zero tons, for example. And likewise with Meta, you're seeing 4 million tons versus zero tons. And Amazon's got the same issue where you're looking at like 15 million tons of location based carbon footprint from using electricity versus 3 million tons from using this.

So you, this really gives an idea of how these two different perspectives end up changing how you might report on this and how you might think about the environmental impact of using some of these tools. And like, to an extent, there is, there are reasons why you have a market based approach because, these come out of the fact that people who are inside large firms are looking for ways to be recognized for the investments they're making so they can justify this internally.

So there is a role that some of these play, but it often, it obviously gets quite a bit more complicated than that, especially because this is the year that the Greenhouse Gas Protocol, the kind of gold standard for reporting, is currently being overhauled to rethink how you should report this stuff and how you should be allowed to talk about energy being green or not green in this context. So Gaël, is there anything that kind of leapt out at you when you had looked through this? Because I would love more people to see this. I think it's a really fascinating story.

Gaël Duez: I think I've already commented it in advance when I was referring to the struggle between market based and local based approach. And once again, I think we should stress how important it is to understand that the way we build things in our mind and in our society as humans is one thing, and the physical reality of the world is another thing.

And when you add energy on a grid, wherever, et cetera, you have no clue on how it will be used, even if it will be used, because when you create PPA, it's potential energy to be used. You create new capacities, whether those capacities will be used or not remains a challenge. Obviously, they will be used, but not necessarily 100%, etc.

And I think the right approach is clean up your own mess. Everyone should start with this. So, I'm fine with having part of the sustainability report explaining what has been done and what could be the approach of market based, but the truth is local based approach. And when you see these figures, they're actually very consistent.

Yes, they're increasing massively their investments in data centers to fuel the AI boom. Their entire business model is based on infinite growth. The numbers go up, that's pretty, pretty logical. And what I've just kind of, when I read this piece of news, it also connected a lot with the crisis at the SBTI, the Science Based Target Initiative, that happened this year, when there was very strong push to allow more offset techniques to be recognized as science based,

Chris Adams: Ah, you're referring to the Scope 3 thing. The push people being able to use offsets in their supply not just electricity, as a way to kind of decarbonize that without having to necessarily make some of the changes to like reform the supply chain. Is that what you're referring to here?

Gaël Duez: Absolutely. Thanks for making it much clearer than I was about to say. So, I think the struggle is everywhere because we see that the low hanging fruits, most of them has been already taken care of in this big corporation and they're entering the bumpy road where you've got harder choice to do. And when you face this kind of choices, well, either you do the right things and you go back to the physical reality of our world, or you try to change a bit the narrative or change a bit the rules, and I think this is exactly what we've been seeing at the science based target initiative where, some companies were obviously not able at all to meet the decarbonizing plan that they proposed just a few years ago, and they were trying to change a bit the rules.

And that should really question ourselves when it comes to transparency and acknowledge that even the most well intentioned CEO, the most well intentioned C suite, they cannot really do the right things without a bit of external help, whether it comes from pressure from activists or governments or UN, you name it, but we cannot fully rely on companies agreeing on how they should measure their own environmental impact.

Even if they are well meaning with tons of great people trying to do the right things, et cetera. It's not a black and white world out there. But there is a question at some point of financial pressure, shareholder pressures in many of these companies. They're just stronger than the entire stakeholders pressure.

Chris Adams: I think I know what you're referring to here, and what I'll do is we'll share a link to some of the stories about the SBTI, the Science Based Targets Initiative, the whole kind of bun fight that took place there, that ended up resulting in the then CEO stepping down from the organization, partly because there was so much pushback against this idea that you might be able to use offsets for looking at your supply chain emissions is in this same fashion. So we'll definitely add a link to that, thanks for that Gaël. The other thing that we maybe might maybe you must referring to inside this is there are some really nice quotes. Is that what you're talking about here does actually there's some relevance back to the Financial Times piece here, about when you look at the decarbonizing energy in this particular space. One of the approaches being used, which is described as Emissions First, for example, this is the one that's being largely put forward by Meta and Amazon. And one of the arguments for this is that we should be able to be optimized for absolute impact, not necessarily our own carbon footprint.

So given a hundred euros or a hundred dollars, we should be able to deploy that in the most effective place globally, wherever it might be. So one of the arguments being made is that we, if we want a data center in say, North Virginia, we should be allowed to basically purchase the right to claim the emission reductions in somewhere like India, which has a very coal heavy grid, and then kind of count that against our own emissions. And on one level, yes, you are reducing the environmental impact, you can make the argument, but it also means that some of the cheapest possible emissions in India are now being bought by one of the richest, some of the richest companies in the world. And this is very similar to the dynamic you were referring to with offsets, where, essentially, if you have a kind of, climbing scale in terms of how difficult it will be to reduce emissions over time. If you grab all the low hanging fruit, that raises all kinds of equity issues about what's left over for people who don't necessarily have the same resources available to them as Facebook or Meta or Amazon, for example, when looking at this. So, this is one of the challenges you do, find. But I'd really urge people to look over the FT piece and what we can do is we'll share a link to an archive link for this if the one, if it's behind a payroll, because it's a really fascinating piece and it's really worth looking into. Okay, Asim, sorry, Gaël, should we move on to the next story here?

Gaël Duez: Yes. I'm honored that you thought was... I

Chris Adams: Bit of a Monday morning moment here. Okay, so this is actually a story speaking about AWS again, by moving an AI workload to AWS, you can reduce the carbon footprint by up to 99%. That's the argument being laid out inside this. And this is a piece from The Stack, which is not the same as The New Stack, who covered our previous story.

It's another organization. This one basically takes apart some of these questions, or at least dissects some of these claims and say, well, what's the basis for this? Gaël, I wanted to kind of give you a bit of space to talk about this because I know that you've read this and I know that this is something that is... and you often advise firms who are actually trying to figure out how do I reduce the environmental impact of the services I'm using. So when you read this, what leaps out at you and what kind of things are the kind of most salient points would you say?

Gaël Duez: Well, I think my main reaction was, "oh no, yet another one."

Okay. So there is the SCI piece, which is interesting, but let's go back to basics about it. And I encourage listeners to have a look at two great resources. The first one is an article from the Boavizta Association investigating the claims made by several hyperscalers that when you migrate from your services, from a traditionally hosted perspective to a hyperscaler perspective, you save 60, 70, 80, 90 percent of carbon emissions.

And the second source is actually a man, and I was very pleased to see him being quoted and interviewed in the article, which is, he's Mark Butcher. He has been very vocal about hyperscaler claims, how they measure things, the scope three not being that well taken into consideration. Mark works at Positive Cloud and he works with a lot of clients across the UK on these topics.

And why I'm mentioning this too is that hyperscalers, when they say, basically "move your instances more, move your computing power to our facilities, because we are so much more efficient than the other data centers or the other hosting solutions," they might be right, but the others are a role, a world on their own.

And this is really the question of the baseline. Yes. If I run two or three servers in my office building, in a small room with a very old air conditioning, yes, I might reach a PUE of 2, 2.2, maybe 3, whatever, et cetera. But most of the clients, they don't start with these baselines. They start with servers and instances being managed in already quite professional and by seasoned providers, okay, and I won't provide any name here, who have already reached a decent PUE and the gain is much smaller.

So that's. On, from which baseline do you start?

Chris Adams: I see, like the example given here is like a really inefficient thing. If you compare really an inefficient setup compared

to this idealized, going to give you a disproportionately large saving. That's

Gaël Duez: And this is when you study the case studies, which have been provided by hyperscalers. it's literally my grandma managing a data center. Worst practices possible that are accounted as a baseline. And this is not the truth. Many data centers already reach pretty decent or, pretty good actually,

power efficiency. So where you start from matters. And that's my, point number one, my point number two, and we go all the way back to this local versus market based approach. But I'm sorry to say I'm, I, feel a bit like I'm, rambling here. But once again, if, even if, okay, I run the worst possible data center on Earth, I've got

maybe a PUE of two, maybe three, et cetera, but I run it in a very low carbon place, say France, if you account, for, the nuclear, a share of energy, say Scotland, for instance, and I recall Mark wrote something about it, Mark Butcher wrote something about comparing the energy intensity of North Scotland and Ireland.

And once again, just by migrating your instances there. You might, let's say, divide by two, your energy consumption, thanks to better energy efficiency by AWS, because they're very active in Ireland, but you can do the same math with Google or Azure, but then you start operating in a country which is, and that's the case between, for instance, Northern Scotland and Ireland, six times more energy carbon intensive when it comes to electricity production.

Chris Adams: So let me check if I understand, well, I think with the point you're getting at, you're basically saying, yes, you may, the infrastructure may be more efficient, but if the local energy is dirtier, it doesn't necessarily matter that it's six times more efficient. If, say, the infrastructure is twice as efficient, but the energy is six times worse, then it's still, you're not coming out ahead. That's what I think you seem to be making the argument there

Gaël Duez: I actually, I was actually, I was not the one making the argument. Mark did it and

Chris Adams: but that's, that's you're saying. Okay.

Gaël Duez: but that's exactly what I'm saying. And so this question of where do we start from the baseline is super important. Then there is a specificity of the AI itself. And we should always remember that most of the time by AI people, and I think this article is a bit misleading here as well, they think generative AI, which is part of AI on its own.

We start from so high. It's still very infant. The LLM are not that old. And of course you can decrease very significantly the energy consumption and the emissions of your AI model, because we are, we've just started to do so, but what will be the trend for, I would say, everyone rather than AWS is an open question.

Obviously, they're doing a better job reducing and optimizing everything, but it's fair to assume that other actors as well are doing the same. So I'm always very concerned when we take a very new algorithm, I would say, or a new part of the AI industry and say, "Oh, look, we're going to reduce by that, that much."

But of course, it's like with cryptocurrency. They've started at such an inefficient way that they made a great progress. Still, they consume a lot of energy. So you see that's my point. So I will be always super cautious with this kind of stuff. And then comes the good part of it, which is using the, as the software carbon intensity, but maybe Chris, you want to elaborate a bit more on

Chris Adams: So, it's nice to see an international standard being used in reports like this, but in order for a standard to be used, you need to follow the standard. And one of the key things that the standard actually does is it, basically, when I've looked through this, I look through the report, I can't see any numbers for any of this.

So, it's like you have a car saying, well, this car is twice as efficient as this other car. If you don't have any numbers about, like, the miles per gallon or the equivalent like that, it's very hard to be able to trust that number, for example, or trust any of these claims here. So you have an issue about, well, there's a lack of underlying numbers. Also, the thing we see is that a significant part of this is based around the environmental impact of the energy itself. Again, we don't have the numbers for this, but in particular, one of the key things, one of the key claims being used in the report was we're going to take into account these market based figures here.

Now, the Software Carbon Intensity specs explicitly says we don't use these inside this. So, you've essentially got people using this term, Software Carbon Intensity, without actually following any particular nuances of this. And this, It makes it very difficult for me to recommend this report for anyone else because it essentially is going against how this is intended to be used. And the firm that was working with Amazon, they are very involved inside the Green Software Foundation. This really needs to be a thing that we can't do if we want to see this to be adopted and respected because this essentially, in my view, undermines a huge amount of work that's gone into developing a standard here because this makes me trust the Software Carbon Intensity less after reading this report and seeing people cite it, because it's being used incorrectly. So that's one of the things I would actually raise and something that does need to be addressed. Like, there are mechanisms that the GSF has to say, please do not use it in this way. It's misleading and it undermines some of the work we have. And I think that's something that will need to happen because, yeah, I cannot recommend anyone looking at this report or even recommend using or referring to this standard like this because it's an incorrect use of the standard. So yeah, that's my take on it.

I'll move on from this because that was a bit of a downer, but it's really important if you want people to trust this in my view.

Gaël Duez: And I think also that there is a way to protect the SCI and the Green Software Foundation tools, which is using the Impact Framework manifest,

Chris Adams: Yeah, I'm glad you mentioned this, Yeah.

So there is a mechanism. The whole thing around the Impact Framework was specifically set up to say, "hi, you're going to make a claim? Make this transparent. Show you're working inside this." And that's like, there's a huge amount of work that's gone into providing this, and if you're not going to share any of the numbers or share the basis, and there are now lots and lots of really helpful case studies demonstrating how to do this, like we'll share links to this to show this is, these are the correct ways to use this. When you have it being used in a way which is so unhelpful, it's, really problematic, and you can see why people are going to struggle, and why, you can see why people end up essentially dismissing so much of these, efforts as greenwash, when people aren't sharing the underlying numbers for this stuff.

So yeah, that's like the framing I would take, and I would really like to see this addressed, because it's going to be, it's going to be a real challenge going forward, in my view.

Gaël Duez: Transparency, transparency, that's all that matters with this kind of claims.

Chris Adams: So, this probably takes us to the last story I think we have time for. This is one actually, this isn't so much a story, but more of a kind of discussion about some of this. So we've been talking about how it's real, really difficult to actually get some numbers from this, and how the way people report this is also a real challenge.

The thing I kind of want to share with you, and I realize I won't have time to talk about this while I, this is partly what I'll be talking about at some events in September, is the fact that we have, we Some regulation, which is forcing some of this. So in Europe, all across Europe, there is a law called the energy efficiency directive, and basically any data center that uses more than 500 kilowatts of power, which is, that's not a small data.

It's not a tiny data center, but pretty much every single hyperscale you imagine would have to do this. There is now a law, which basically says, every data center and any organization operating a data center has to make publicly available, Information like the name of the data center, who runs it, how large it is, how much power is used. It talks about the amount of energy used, the water used, all this stuff. So we do actually have laws which are kind of forcing some of this now. The, there is one caveat in that where companies consider this information to be a trade secret, they don't necessarily need to publish this information into the public domain, but where companies are not saying this is a trade secret, we now, as technology professionals, can ask and say, this information should be in the public domain if you're in, if you're in Europe. And the thing that I might share as a follow on from this is that for companies that are not sharing this information, they are now mandated to report to a centralized database with the idea that some of this information will be shared in an aggregated format.

So for the first time, we can actually get some meaningful numbers that come out of this. So, companies that are prepared to be transparent, you can ask for this stuff. Companies that are then saying, "we're not going to make this transparent because it's a trade secret." There is still a mechanism by which they will need to report so that we can finally have some data informed policy around this. Because one thing that's come up again and again in this discussion has been that we don't have access to this information. And there is so much pressure or there's so much, there's so many incentives to construct a message which makes you look good that it becomes very difficult to trust a lot of the statements around green software that come from lots and lots of large firms right now.

So yeah, this is, I'll share a link to the issue where this has been discussed inside the real time cloud working group inside the Green Software Foundation. But Gaël, I wanted to just check, as someone who's not in Europe, what's your take on this? Is there anything that kind of caught your eye when you were looking at this?

Gaël Duez: Well, the first thing is that it is required to disclose numbers in both relative, but also absolute numbers. And that's very important. It's not only PUE. It's also how much energy did you consume overall? And that really connects well with what we've discussed before. I also believe that this is something that is

pivotal for country with weakest electricity grid to consider. It's always claimed that for instance, Kenya, my neighboring country of La Reunion island, has a very strong policy of attracting data centers. They want to become a computing power.

Chris Adams: Iceland of Africa, because have more geothermal than

Gaël Duez: Absolutely. And I reckon that this kind of disclosure will also have them a lot anticipate what is needed for them to prepare the electricity grid to this kind of increase in electricity consumption caused by data centers.

And also making sure that what has been dealt and agreed when they do this big hyperscale deal are actually what is provided, that the energy is there, that the water is there, and I believe that local populations, which are often caught in between, like, "oh, it will create a lot of jobs, but when you do the mass, not that much," so it's not that an obvious investment to say, okay, we will welcome a lot of data centers in our country.

It might be, but it's not like a big investment. A clear win or the case, they will have the ability to scrutinize, how, what are the impacts and environmental impacts. And I must admit that, if you look a bit at the history, in 1982 in Europe, a European directive created the Seveso listing, and the Seveso made it compulsory for every state in Europe to list what are the facilities, industrial, agricultural, mostly industrial, which can create significant environmental risk.

And we were talking about chemical industry, et cetera, et cetera. And to some extent, data centers, they impact a lot of the environment. It's just that they will not blow away like a chemical industry, but on the long run, they've got a lot of impact on their environment and it makes a lot of sense to, at least for the bigger now, the biggest data centers, to be able to provide

environmental information, in a comprehensive way, a comparable way, and to make sure that we monitor the environmental impacts of these big facilities. we're talking about facilities that are built on hectares of land. There's not like the small, tiny warehouse that we might still think of.

They're like massive industrial facilities. So having open and transparent reporting seems to be quite straightforward.

Chris Adams: So maybe we might see some of these ideas adopted in other parts of the world, especially because on the underlying data for this. We'll share a link to this. We've been doing some research ourselves in the Green Web Foundation. A lot of these data points are based on the EU code of conduct, which is a public document for people. So it may be the case that you might see some of these data points being reported in other parts of the world as well to set a precedent. So they actually have the data to make data informed decisions about how, about the role digitalization plays in society and

how the impacts are actually shared around this.

Thank you for the link about Seveso as well. I didn't know about the Seveso directive. That's totally new to me. What I might share is a link to some of the work that the Green Web Foundation has been doing. We have a fellowship and one of our fellows, Samantha Nidwalana, she's based in South Africa and the Netherlands, and she spent a bunch of time looking through, trying to get some numbers in South Africa for, to basically explore okay, what's the environmental impact of data centers inside this?

And we'll share some links to her challenges in this because she's been trying to find these numbers and it's been a real, it's been a real challenge in many cases actually. And it does give you some idea about like where some of this might go, but also hopefully stories like this and seeing some of these laws being passed might help set a precedent so that we can have more transparency in other parts of the world as well. Okay, I realize we're coming up to time. Let's do a quick just run through. We've got some events coming up. I know that I'll be catching a train in a few weeks time to go to London, and I think I'll see you in person for the first time for quite some time actually. This is an event called Green IO Conference. Maybe you could just briefly touch on that before we move on to the next set of events and then close out for the day.

Gaël Duez: Yeah. Well, absolutely. I will be pleased to see you again for the second time. The first time was in Berlin, if I recall, and I think it was for being interviewed on the GreenEye And what. I launched with my partners RPI Days last year, was a series of global conferences called Green IO, not very original, sorry about this.

And the idea was to have also on site events. We can see a lot of hybrid and a lot of online events when it comes to digital sustainability and they're great. We can give kudos to the Green Web Foundation, the Green Software Foundation, CNCF as well. They're doing a lot of these events. Now, what I also realized is that if you look at other specialties in our IT industry, let's take cybersecurity, for instance, accessibility, design, or you name it.

When you work in a city with a significant enough workforce in cybersecurity, for instance, let's say Barcelona, Berlin, Paris, New York, Singapore, Beijing, you have at least 2, 3, 4, sometimes 5 different on site conferences when you know that you will meet your peers. Today, if you are an IT sustainability specialist or green IT folks or whatever, like the dude who anyone reached out to because, "oh, it's about green, et cetera.

He's a person, she's a person we'd like to talk to." You've got basically nowhere to go except for one conference in Paris and one in Belgium now in Brussels this year, where you know that the topic will be a hundred percent your specialty. So don't get me wrong. We can give kudos to a lot of tech conferences like QCon, like even Reinvent AWS, you've got to sustainably track, that's fine, but having a 100 percent green IT focus or IT sustainably focus conference, where you know that basically this is the place to be to meet all your peers, most of the cities around, they don't have it. So that was really the idea of creating this Green IO.

Even since Singapore, London, and Paris this year, we are like 90, 90 percent sure that we will add New York and Munich next year also. And the idea is really, it's just It's a place that has been created for the local communities to do what they want with it. And so you, this year in London, it's a bit like go back to the basics and let's talk step by step.

What about low carbon infrastructure? What about design and eco design? What about beyond, understanding the organizational challenges and HR challenges that when you want to rule out more sustainable approach, but that's a one day conference. It's two days in Paris this year, but in London, it will be one day.

And we expect to have like several hundreds of people joining. And I know also that we will have a great keynote speaker that some of the listeners might known named Chris Adams. So I'm really delighted that you agreed to join Chris. Thanks a lot for this.

Chris Adams: Cool, thanks, so that's the event. It's the 19th of September, taking

place in Bishopsgate in London. And thank you, I am indeed keynoting in the morning, so I hope my trains are on time. But there's a number of people from the Green Software Foundation, for example. I can see Sara Hsu, who is one of the people leading the Green Software patterns project in the Green Software Foundation. There's some people from the W3C on working on web sustainability guidelines. There's also, I'll, share a link to Therese Gale who is working at Salesforce. She'll be talking a little bit about some of the experience, some of her experiences as well. So there's a number of people who

Gaël Duez: Mark Butcher will

Chris Adams: and Mark, yes, the, yes, Mark Butcher of Positive Cloud.

He's been one of the people who's been really instrumental in some of the most recent work inside the UK government, put together a kind of Digital Sustainability Alliance. There's a bunch of things there. So I'm actually looking forward to this Gaël, and I want to just give people a heads up that it's taking place. Gaël, thanks so much for coming in for this and telling us the stories about Wales, And we're watching and all that sharing all your insights elsewhere. Gaël, if people do want to follow up and see what you're up to, what's the best way to find you for future work?

Gaël Duez: Well I would say on LinkedIn and I'm pretty easy to find Gaël Duez. There is not that many. And more generally, if you are interested in what we do in Green IO, it's greenio.tech. That's the website. You've got access to the podcast, the conferences you've got link to my own websites. But I think greenio.tech is the best place to start.

Chris Adams: Brilliant. Well, thank you for that. We'll share some links to all of those websites and to this event coming up. And, for people who aren't able to go to Green IO, the Cloud Native Computing Foundation, the TAG ENV essentially the green part of the Cloud Native Computing Foundation.

They have a series of remote events taking place in October. We'll show a link to that so that if you still want some, to get an events fixed, that's where to look. Alright, Gaël, thank you once again for making the time and I hope you have a lovely week. Okay. Take care of yourself, mate.

Gaël Duez: Thank you, that was great being there. Bye, have a nice week as well.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundationon. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

A Greener Internet that Sleeps More

25 juli 2024 | 60 min

Host Chris Adams and guest Romain Jacob delve into the often-overlooked energy demands of networking infrastructure to discover A Greener Internet that Sleeps More. While AI and data centers usually dominate the conversation, networking still consumes significant power, comparable to the energy usage of entire countries. They discuss innovative practices to make the internet greener, such as putting networks to sleep during low usage periods and extending the life of hardware. Romain talks about his recent Hypnos paper, which won Best Paper at HotCarbon 2024. He shares his team’s award-winning research on how energy demand for networking kit powering the internet can be reduced by simply by powering down links when not in use.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Romain Jacob: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

SCION Architecture [11:30]
Environmental Impacts of Internet Technology (eimpact) [17:15]
Why we should be intentional about the mental models we use for thinking when we think about digital sustainability | Chris Adams [18:30]
A Sleep Study for ISP Networks: Evaluating Link Sleeping on Real World Data | Romain Jacob, Lukas Röllin and Laurent Vanbever [18:59]
Network energy use not directly proportional to data volume: The power model approach for more reliable network energy consumption calculations | David Mytton [38:55]
Co2.js - The Issue | The Green Web Foundation [42:57]
Rethinking Allocation in High-Baseload Systems: A Demand-Proportional Network Electricity Intensity Metric — University of Bristol | Daniel Schien [43:53]
Introducing Web Sustainability Guidelines | 2023 | Blog | W3C [49:31]
Greening of Streaming [52:16]
Network Power Zoo | ETH Zurich [54:46]

Other source material:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Romain Jacob: We used to consider that energy is cheap. Energy is there. We don't need to worry too much about it. So it's just simpler to plug the thing in, assume energy is there. You can draw power as much as you want, whenever you want, for as much as you want. And it's time to get away from that.

I'm your host, Chris Adams.

Hello, and welcome to another edition of Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. Back in episode 10 of this podcast in September 2022, we did a deep dive into the subject of green networking, because while a lot of the time people talk about the energy demands of AI and data centers, in 2024, in absolute terms, the amount of power consumed by networking was still larger.

Back then in 2022, the best figures, when we looked at this, came from the AI, which put the energy usage of data networks at around 250 terawatt hours per year. So that's about the same as all of Spain's energy usage in 2023, so that's not nothing. Now, it's a few years later, 2024, and the best figures from the same agency, the IEA, now give us a range of between 260 and 360 terawatt hours, which could be anything up to a jump of 50 percent in three years now.

Now because of much of this power is coming from fossil fuels, this is a real problem, climate wise. So what can we do about this? With me to explore this once again, is my friend Romain Jacob, who helped guide us through the subject in 2022, along with Dr. Yves Schouler at Intel at the time.

His team's recent research won the Best Paper Award at HotCarbon, the conference that has fast become a fixture on the green IT and digital sustainability circuit. So he seemed a good person to ask about this. Romain, thank you so much for joining me for this podcast. Can I give you the floor to introduce yourself before we revisit the world of green networks?

Romain Jacob: Chris, welcome. I'm very happy to be back on the podcast to talk a little bit more about this. Hello, I'm Romain, I'm a postdoctoral researcher at ETH Zurich in Switzerland. I've been working in sustainability for two to three years now, more or less full time, as much as full time research happens in academia. And yeah, I was, I had the pleasure to present some of our technical work at HotCarbon and I'm sure we're going to deep dive into a bit more in the podcast.

Chris Adams: Okay. Thank you, Romain. And for people who are new to this podcast, my name is Chris Adams. I am the executive director of the Green Web Foundation. We're a Dutch nonprofit focused on reaching a fossil free internet by 2030. I also work as one of the policy chairs inside the larger Green Software Foundation, which is why I'm on this podcast.

Alright, if you're also new to this podcast, every single project and paper that we mention today we'll be posting a link to in the show notes. So for people who are on a quest to learn more about reducing the environmental impact of software engineering, you can use these for your own practices and your own research.

Okay, Romain, are you sitting comfortably?

Romain Jacob: I am.

Chris Adams: Okay, great. Then I'll begin. Okay, Romain, when we last spoke, we covered a range of approaches that people are using right now to rein in the environmental impact of networking. And but before we spend too much time, I wanted to kind of see if you could help set the scene to help folks develop a mental model for thinking about, say, networking versus data centers, because I touched on this a little bit, but it might not be obvious to most people.

So maybe if you could just provide the high level, then we can touch on some of these Differences between the two and why you might care, or how you might think about these differently. So yeah, let's go from there.

Romain Jacob: Yes. Sounds good. It's true, you mentioned in your introduction that we sometimes get numbers that oppose or compare data center corporate footprint or usage and the ones from networks. But now, what does network really mean? It's not really clear what we mean by that. Because there is networking in data centers and, you know, the rest of networks are not completely detached from it.

But at a high level, you have a set of networks that are meant to provide internet access to individuals and to other networks. What this means is that you have companies that are specialized to just make your laptop, your phone, or other appliances you have, being able to talk through the internet. And typically when we refer to networks, without further details, this is the type of network you're talking about.

And data centers, on the other hand, are something that is in the scale of IT fairly recent, where we have this idea of if we centralize in, in one physical location, a lot of powerful resource machines that have a lot of compute, that have a lot of storage available, then we can use that as a remote computer.

And just offload tasks to those data centers and just only get the results back. In today's ecosystem, data centers are a very core element of the internet. The internet today would not really work without data centers. Or at least a lot of the applications we use over the internet only work thanks to data centers.

So from a networking perspective, in a data center, you also have, you need to exchange information and bits and packets between those different machines that live inside the data center, but the way the network looks like is very different from the cellular network that is providing mobile connectivity to your smartphone.

Chris Adams: Ah, I see.

Romain Jacob: So those are very different type of networks, and they have different, very different type of way they are designed, way they are operated, and how they are used. Typically, in a data center, you tend to have a quite high usage of the data center network because you have a lot of exchange and interaction between the different machines that live inside the data center, whereas in the networks that provide internet access, so the networks that are managed by entities we call ISP, for Internet Service Providers.

Chris Adams: Mm hmm.

Romain Jacob: Tends to be much less utilized. There is a lot more capacity. in those network that what is really demanded by the end user.

Chris Adams: Okay.

Romain Jacob: That's a fundamental difference between, between the two networks and something that we try to leverage in our research.

Chris Adams: Okay, so if I just check if I understand this, so you said that you might have networks which might be like, say the ISPs and things, they are individually not that high themselves, but because there's so many of them and because they're so diffuse, in aggregate, this can work out to be a very large figure, for example, and, and, that also speaks a little bit about, I guess, how you might power some of these.

So, like, when we think about a large hyperscale data center, that's something in the region of maybe, if you're looking at a large one, which is maybe the high tens to maybe low hundreds of megawatts, that's maybe thousands of homes. That's a lot of power in one place, whereas with a network, you don't have quite so much, but it's because it's distributed, you might have to have different approaches to managing that.

So, for example, you might and, you might have to take different strategies to either decarbonize that or deal with some of that, some of that load that you actually have. Okay, and I guess one of the questions I might have to ask you about this is then, when you have this, split between the maybe it's just worth talking a little bit about the different kinds of networks that you have here.

So for example, as I understand it, there's maybe an ISP I connect to, maybe my ISP, but then they need to connect to some other cable. And if I'm going across, maybe if I'm connecting to a server across in the Atlantic, then I'm going through some like backbone or something like that, maybe you could talk a little bit about the different layers there, and what some of the, and what the kind of, how much they might make, or if there's any differences in how those ones need to be powered, for example, how they're used?

Romain Jacob: Yeah, totally. So the, the name internet stands for an interconnection of networks. So the internet by name and in private space is a network of networks. Right? So when you connect to the internet. What it means is that you connect to another machine somewhere on the planet that has also access to the same global network.

But these global networks, the connection between those two endpoints has to go through, most of the time, several different networks. And so, typically, if we look at the internet infrastructure, there are several ways of representing this, but one division that we usually use is You have Core, IAP, so the one that kind of sit more in the middle and they provide transit for many, many different interconnections.

Then you have networks that are qualified to be more belonging to the metro area. So this is where it's getting closer to the user, but it's not yet the network that provides direct connectivity to, let's say, your phone or your laptop. And then you have the edge network. And the edge network is really there to provide what is called the last mile connectivity key to the end user.

And those categories exist and were proposed to Helsinki because this share of the network has different characteristics. The core tends to look a bit more like the data center. Like, it's more dense mesh, so there are more interconnections between the different points in that network. And the utilization tends to be higher and kind of constant because it's a global network.

Whereas, the closer you go to the end user, the more you're going to see filtration, because, for example, while users are awake during certain hours during the day, and this is where they tend to use their machine. You will see peak of usage during, you know, TV show primetimes in the evening, but much less at 5 a.m., where most people are deep and not using their phone or their laptops. And so, those networks look differently in terms of What they are used for and how they are built and designed, because we try to adapt the design for the particular use case.

Chris Adams: Okay, thank you for that. So it's a little bit, if you squint, it's a little bit like how you might have motorways and then A roads and B roads and maybe smaller roads, for example.

Romain Jacob: Very like that. It's very much like that.

Chris Adams: All right. Okay. So that's actually quite helpful. And when I think about other kind of systems, I think a little bit about say, like electricity networks, which have, you know, big fat transmission ones which send lots of things, but then you have like the smaller distribution networks which are, so it's somewhat comparable. Okay.

Romain Jacob: It very similar in principle.

Chris Adams: Okay, so this is helpful for developing a mental model about some of this. Alright, okay, so last episode when we spoke about the different techniques people spoke about, you, we spoke about things like carbon away networking, different protocol designs like I think SCION, which was one of the, projects proposed.

And you kind of coined this phrase that an internet of the future needs to grow old and sleep more. And I really, I found this kind of quite entertaining and it stuck with me. But for people who are new to this, maybe you could just unpack what you meant by that because not everyone has read the paper or seen the talk. And I think it's quite helpful for thinking about this subject in general, actually.

Romain Jacob: Yeah, sure. So, so two years ago in the first edition of the HotCarbon Workshop, we, we outlined this vision of what could be relevant to work on in sustainable networking area. And the two ideas that emerged were essentially captured by this growing old and fleeting more aspect. So what do we mean by that?

Growing old is essentially the idea that we tend to be using the hardware we buy not long enough. So if we take an end user perspective, we tend to change phones every couple years. Numbers are changing about this, but we can debate whether this is a good thing or a bad thing. In the networking area, so for the hardware that operators buy to make up the network, so devices that we call routers and switches, it tends to be a bit of the same thing.

Devices were changed, the standard used to be every three years. So in three years, the entire infrastructure would be renewed. So you would buy new hardware to get higher speed or better energy and so on. And there are various reasons for doing that. We can detail it if you're interested afterwards. But it has a very significant cost, financial cost, but also in terms of carbon cost.

Because one people need to understand is that every time you manufacture a product. Not just for networking, but for any product, there is a carbon footprint associated to it. This is where we typically refer to the embodied carbon footprint. And so this embodied carbon is a one time pay, but if you buy more often, well, you pay this price more often.

Now, there's a bit of a tricky thing, which is that you, you can argue that If I buy a new device, a new phone or a new router that is 10 times more energy efficient, then over time, I will then do more saving that would compensate for the embodied cost. The problem is that you, it's very hard to estimate how much you would save and how much is the embodied footprint.

It's a very firmware, a break even point where you're doing this upgrade, buying this new hardware, start paying off from a carbon perspective, but it's not necessarily clear ahead of time when that happened. Generally speaking though, what was pretty clear to us is that we could and we probably should be using the hardware longer.

And this is what we meant by the grow old idea.

Chris Adams: Ah, okay. Thanks for that.

Romain Jacob: Now, to describe more, this is kind of simpler. This is the idea that is very common in other fields like what we know as embedded systems or the Internet of Things. Think about devices that run on battery, to make things simple, that are more or less small but run on battery. And because they need to run on battery, for decades, engineers have been trying to optimize the energy efficiency of those devices.

And the most efficient way of doing this is essentially turning off everything you don't need when you don't need it. So if you think about your phone, your, the screen of your phone is off, I don't know, maybe 90 percent of the time. And this is to save the power drawn by your screen, which is by far the most expensive or power hungry element in a smartphone.

And we do this in order to save on power, on the average power and so on energy at the end of the day. And we are arguing that in the networking world, this is not done too much. And it should probably be done more. So now I need to be quite precise here when I talk about the networking world, I'm talking about the wired networking domain.

In the mobile domain, so in cellular communication that connect to your phone, or also in Wi Fi and so on, the idea of sleeping is already used quite a lot. But in the wired domain, it did not transfer too much. And so the reason why it did not transfer is because we used to come to the idea that energy is cheap. Energy is there. We don't need to worry too much about it. So, it's just simpler to plug the thing in, assume energy is there. You can draw power as much as you want, whenever you want, for as much as you want. And it's time to get away from that.

Chris Adams: Okay, alright, so if I just play that back to you to make sure I understand it correctly. So, the growing old part is essentially a reference to the embodied energy that goes into making various kinds of hardware. So, like, when we're looking at a laptop, around 80 percent of the carbon footprint, it comes from the manufacturing, compared to the running of this.

And, if I keep that laptop for a short period of time, It's a great proportion of the life cycle, lifetime emissions, for example. And we see the same thing in data centers as well. So, for example, Facebook and company, you know, some companies and hyperscalers, they might have had this three year period that you spoke about before, but in the 2010s, we saw figures anecdotally, but not published ever.

It was like 20 years ago. But sometimes these would go down to as much as as little as 18 months for some service because they wanted to get the maximum usage of kind of compute for the power they're using for example. So they had incentives to change like that. So that's what that part is a reference to.

And I think on the eImpact mailing list, where I've seen a lot of the discussion. I will share a link to this in the show notes of this really cool 3D chart showing how, where the break even points are that you mentioned about that. And the sleeping part seems to be this reference that, in many ways, networks are often designed for kind of maximum amount of usage, not necessarily what the average usage might be, similar to how, say, the electricity grid in America, for example, is designed currently designed for everyone to be using aircon at the same time, when normally it's maybe 40 percent utilization.

So there's all this kind of headroom, which doesn't need to be accounted for. And we, and it's a bit like service, you know, we have, as software engineers, we're taught generally to size for the maximum output because the loss of business is supposed to be worse than the cost of having that extra capacity.

But in 2024, there are new approaches that could be taken. And we do things like serverless and scaling things down. And these ideas are - they've been slower to be adopted in the networking field essentially, right?

Romain Jacob: Yeah, that's kind of the, that's kind of the idea.

Chris Adams: Brilliant! Okay, that is good. We'll share a link to the paper because it's quite a fun read and I really helped, it stuck with me ever since I saw you speaking about that.

Okay, so we've kind of set some of the scene so far. We've got some nice mental models for thinking about this. We referred to the energy and the embodied part and I guess the thing we didn't mention too much was that the growing old thing is going to be, you know, more of an issue over time because while we're getting better at decarbonising the electricity of the internet, we're not doing such a good job of decarbonising the extremely energy intensive process of making electronics right now. So we're only, this is only going to become more acute over time.

So, maybe I can allow you to just talk a little bit about this Hypnos paper, because as I understand it, it was an extension of some of this vision going forward, and I know that it wasn't, you weren't presenting yourself, but I know it was your team who were presenting it at HotCarbon, so maybe you can talk a little bit about that, and maybe say who was presenting, or some of those things there, because, yeah, I enjoyed reading this, it was quite fun, it, similar way, I enjoyed it as well, basically.

Romain Jacob: So Hypnos is a, a recent proposal that we've made to essentially try to quantify this sleeping principle. So in two and a half years ago, we said, okay, we could look at the embedded aspect, we could look at the operational aspect and how to improve them. And we thought back then that the one way was to just try to apply those principles of heaping to wired networks.

And so together with some math students from ETH, we started looking into this and say, okay, in theory, we know how to do this. Let's try. You know, let's try for real, let's take some hardware, let's design a prototype, protocol that would just put some things to sleep and see what happens. What was surprising to us was that the, the theory of how you would do the sleeping in a wired network and how much you would expect to save by doing that was old.

It was the first papers go back 2008 or so, so there's been a while and back then, people were saying, okay, assuming we have hardware that, that allows us to do everything we want, then we could implement seeding in this way and then we would save so much. So they knew that the proposals that were made, they were making back then were not readily applicable.

And so we felt like 15 years later, it's kind of interesting to see where are we today? Like how can we do things? And the key element, key there, was how quickly you can turn on something. You turn it off, you can always turn off something, you take some time, you save some power, okay? But then, eventually, if you need it back on, you want it to react quickly.

You know, I talked about the screen of your phone before. It's always off, and it's fine, because as soon as you press the button, and you touch with your finger, the screen lights up, right? It feels instant, right? So it needs to happen quickly to be usable. Except that in networks, it's not like that, it's, I mean, not, not, at least not today.

So if, if you think about a link that connects two routers, this was the first, the first thing that we started considering. Okay, let's put that to sleep. It's essentially the smallest unit in a network that you could put to sleep.

Right.

Chris Adams: A bit like a lane in a, like a multi-lane in a multi-lane car road.

Romain Jacob: Yeah. I would think about the road network, like, turning off a port or turning off a link in a network would be like. Cutting one road in your network. You know, like here in the city, you have many different ways to go to different end points and you would just say, okay, this street is closed. So you can't use it.

Chris Adams: Ah, I see. Okay.

Romain Jacob: That's kind of like the simplest thing one can do from an networking perspective, except that to turn the thing back on, to reopen the street would take multiple seconds.

Chris Adams: Mm.

Romain Jacob: And it doesn't sound like much, but in the networking area, multiple seconds is a lot of time because a lot of traffic can be sent during this time.

And. If you make things short and not too technical, it's way too long.

Chris Adams: So we're looking, you want milliseconds, which are like thousandth of a second, and if something, it takes two or three, it's two or 3000 times slower than you'd like it to be basically.

Romain Jacob: Exactly. So, without getting too nerdy and too technical, the problem is, we can't do what we're suggesting in the literature because we cannot sleep at short timescales as we were planned. And so we're like, okay, so is it over or can we still do something? And so what we were thinking is maybe we cannot sleep, you know, at millisecond time scales, but we can still leverage the fact that networks, some networks, are a lot more used during the day than during the night.

So we, we have a lot of patterns that are daily or hourly that we can leverage to say, okay, well, we have a predictable variation in the average use of the network. And so when we reach the value to declare night time. Then maybe there are some things we can share. And so we, we try to implement a protocol that we do do.

We say, okay, let's do the simplest thing possible and see how well it works. And Hypnos is essentially the outcome of that. So in essence, it's a very simple tentacle that looks at all of the roads, so all of the links in the network and how much they are used. And then we start turning off the, the unused one.

Until we reach some kind of like stopping condition that we say, okay, now it's enough. Like the rest we really need to keep it. At a high level, this is what we do. So one, one challenge was to get actual data to test it.

Chris Adams: Mm-Hmm.

Romain Jacob: Because if you stimulate a network and you stimulate the utilization of your network, you can make things as pretty or as, as ugly as you want, you know, depending on how you look at things.

And so it was, what was really missing from the literature was precise case study that says, okay, here is the data from a given ISP. Here's what the network looks like, and here's what the utilization looks like. In this network, what can we do? So, there has been a long, very long effort to actually get this data.

And then, the Hypnos paper is essentially say, okay, we have the protocol, we have the theory, now we have the data. Let's match the two things together and see where that takes us. And, we looked at two internet service providers that, that belong to the access part. So, those are, networks that are very close to the end user, where you would expect more of D&I fluctuation.

And we do see that. What we were a bit surprised to confirm is that those networks are effectively underutilized. You want to dare a guess what's the average utilization in those networks?

Chris Adams: I literally couldn't, I have no idea what the number might actually be to be honest.

Romain Jacob: Guess!

Chris Adams: Okay is it like Okay, so I said the national grid was about 40%. Is it like 40, 50%? Like, that's, like, not-

Romain Jacob: Four, four zero?

Chris Adams: Yeah, four zero is like what I, is what national, electricity grid is. So maybe it's like, something like that, maybe?

That's my guess.

Romain Jacob: Now you're an order of magnitude too high. So we are talking a couple of percent.

Chris Adams: Oh, wow! Okay, and the whole point about the internet is that if you don't have one route, you can still route other ways. So you've got all these under, you've got all these things which people are currently on that almost no one is using ever, basically, at like 2%. Okay, alright.

Romain Jacob: I need to modulate this, right? Okay. For the couple of networks that we got, we managed to get access to the data, right? So I'm not claiming this is the general number. I would love to know, if you have data, please let us know.

Chris Adams: Mm

Romain Jacob: But for the networks we could get access to, this is the type of numbers you would see. An average utilization of a couple of percent. And again, going back to what we were saying before, in a data center, things would be different.

Chris Adams: Mm hmm.

Romain Jacob: I actually don't know because I was there a little in data center networking, but I would expect things to be more in the 40 50 percent kind of like what you were mentioning before.

But in an end to end service provider network, the underutilization is extreme. There are various reasons for that, but it tends to be the case.

Chris Adams: That's really interesting, because when you look at data centers, so like, I can tell you about the service that I run, or that our organization runs, the Green Web Foundation, so we run a checking service that gets around between 5 and 10 million, like, checks every day, right? So that's maybe In the order of like 400 million per month, for example, something like that.

It's, a relatively high number, for example, and even when we have that, we've got around 50 percent, we, we did, we started working out the environmental impact of our own systems recently, and that's with us with utilization around 50 percent for our systems and in cloud typically you'll see cloud providers saying oh we're really good we're 30 or 40 percent like the highest I've seen is Facebook's most recent stuff about XFaaS and they say oh yeah we can achieve utilization of as high as 60 odd percent right but for lots of data centers the kind of old Data centers would have been in the low digits.

And you've had this whole wave of people saying, well, let's move to the cloud by making much better use of a smaller number one. So it sounds like the same kind of ideas of massive underutilization and therefore huge amounts of essentially hardware, you know, it seems like it's somewhat similar in the networking field as well.

And there's maybe scope for reductions in that field as well. Okay.

Romain Jacob: Exactly. And so, I want to make it clear, like, it's not happening this way because operators are idiots, right? It's just, there are a number of reasons why you have such underutilization. One what I would say is probably the main one from a decision point of view is that you want to provide high performance for number of connections in your network.

So to reach from point A to point B, you want to make sure that you want to have the lowest delay typically. And that requires to have a direct line. Do you have other concerns that are that things do fail in networks. Link failures happen and they can be quite drastic. You operate like a physical infrastructure in a country where people leave and work, you have incidents, fibers get cut, and those are things that take a long time to fix and so on.

So you want to have some resiliency in your network. So that if some part of the network goes down, you can still reroute the traffic the other way around and still have enough capacity to serve that traffic. So you have some names that are not used by default intentionally. So you get to a 2 percent or a couple of percent average.

But you don't want to be at 50 percent because if you are at 50 percent and something really goes down, then you may run into a situation where you don't have enough capacity left to run your business. So the point is, we have such a high underutilization and something that I don't think we explained so far is in network equipment, so routers and switches, you have very little proportionality.

What I mean by that is that the amount of power that is drawn by a router. It's essentially, from at a height, it's not exactly true, but at a high level.

Chris Adams: Hmm.

Romain Jacob: The amount of power drawn is almost independent or varies very little if you send no traffic at all or if you send at 100%.

Chris Adams: Okay.

Romain Jacob: So, what it means is that if you have a router that you use at 1 percent of its total capacity, you pay almost 100 percent of the power.

Chris Adams: It's almost like one person in that plane going back and forwards, for example. Like, if I'm going to fly, there's, you know, if I'm going to fly somewhere and I'm the only person, it's going to be the same footprint as if that plane was entirely full, for example.

Romain Jacob: Kind of, yes. It was the same kind of idea. And this is why for us, investigating this clipping was kind of interesting because we know we have such method underutilization, although I probably would not have guessed it was that low, and it wastes a lot. Third, we are essentially operating most of those links at the worst efficiency point possible. And so we try to remedy this and it goes one step in that direction.

Chris Adams: I see. And I think one thing that you mentioned before was this idea, you can power these things down and you know there's very, because there are alternative routes through the network at any time, it may be the case that even if you do have these things powered off in response to upticks in demand, just like with, say, national grids, people might. You know, switch on batteries to, or feed power into the grid from a battery or possibly a peak of gas plant.

You have, you still have the option of switching these route, these links back on when there is mass a, a, a big peak in power, for example.

Romain Jacob: Exactly. And it's the same with the kitting protocol we proposed, right? So, it still makes sense to have those redundant links deployed, you know, those fibers laid out. And then you may say, yeah, but we've paid all this effort to actually install this and it's there. Why should I not turn it on? Well, because it consumes energy whether you use it or not. That's for one. And second, It's good to have it in case you need it. But you can turn it off so that you save energy. If you can turn it up quickly, right, then it goes back to what I was saying before. The turning of quickly part is still problematic today.

Chris Adams: Hmm.

Romain Jacob: So orders of magnitude that the time it takes to actually do this would be in 10, 10 seconds, roughly a few seconds, let's say, up to more, a minute or so.

So then what, what do you have to wait? Yeah. The benefit you gain by turning links off in terms of energy versus the time you may have to wait until you go back to a good state in your network in case you have some failures in your network. And of course you need to multiply that risk by the likelihood of getting such link failures.

So if, let's say, if you have a doomsday event that, you know, will just kill the network error, but that happens one every hundred year. Maybe you can be fine having a day to day management policy that says, okay, to manage this doomsday event, we will need an hour long, but in all the rest of the time, it will be fine and we'll save energy every single day.

You know, you, you have weighed the pros and cons of a strategy in terms of performance and in terms of energy usage.

Chris Adams: And presumably, one thing that you've entioned is because you mentioned that you often have these regular, kind of, predictable cycles, like, most people don't, you know, fewer people use the internet when they're asleep than when they're awake, for example. Like, it sounds really silly, but like, yeah, you're going to see these predictable patterns.

Some of the work with the Hypnos paper was basically, essentially taking some of these things into account. So you can say, well, you need to have this buffer, but we don't have to have the buffer massive you don't need to have every single car in the world engine on idling just in case you need to use it you can turn off some of these car engines for it so that was the kind of idea behind this.

Okay neat so I've used this car model a few times but it suggests that actually think about how, the amount of energy usage and how it scales with how we use the Internet. It might not be the correct mental model. And I just want to kind of run this by you, because this is one thing that I've been thinking about recently is that a lot of us tend to instinctively reached to a kind of car and driving and burning fuel model, because that's how a lot of expo experience costs of energy a lot of the time, right?

But it feels almost like if you've got this thing, it may be a different model might be like, I don't know, like bike lanes where there's a matter of time that you need to build something. You might need to light, make sure a bike lane is well lit, for example, if you do use, but the amount of people using the bike lane that isn't the big driver of emissions in this, for example, maybe something like that.

Romain Jacob: Yeah, I was about to say, I think the analogy is not wrong per se, it's just a matter of the trade off between the, the infrastructure cost and the driving cost, let's say. So let's assume you, you're using a, a mean of transport, whatever that may be. That has a cost X per kilometer, but then you need light and you need, I don't know, cooling or if you're using something that works on under like, I don't know, superconductive environment, then you need extreme cooling. And so the cost for the environment gets very high. I think the superconducting thing is actually a, a, a pretty, a, a much, very much closer analogy to how the way network works.

And you need to spend a lot of power, or to draw a lot of power, just to get the infrastructure on. But once the infrastructure is on, once you get your superconductive environment, then traversing this environment is very cheap.

Chris Adams: Ah, okay.

Romain Jacob: And networks are a bit like that today, right? So turning the wires on costs a lot, but when it's on, sending the bits through the wire, it's pretty cheap.

Chris Adams: Ah, I see, and if I understand it, when I've spoke to other people who know more about networking than me, they've basically told me that at some levels, even when you're not sending any data, there is a signal being sent that basically says, I'm not sending data, I'm not sending data, I'm not sending data, just to make sure so that you've got that connection so that when you do send some data, there's a fast response time.

So, just because we aren't perceiving something doesn't mean there isn't energy use taking place for example. So there's maybe some leakiness in the models that we might instinctively just use or intuitively try applying when we're trying to figure out, okay, how do I make something more sustainable for example?

Romain Jacob: Yeah, this is very true. And it also, it gets a bit more detailed than that. It also depends on the type of physical layer you use for sending your information. In networks today, you have, I think, I guess we could differentiate between three main types of physical layer. One is the electrical communication, so you send an electrical signal through a power rail.

You have optical communication, so essentially using light that you modulate in some way. And then there's everything that is kind of wireless and radio wave communication. So I'll leave the wireless part out because I know less about it and it's a very complicated bee. But if you compare electrical to optical, things work kind of differently.

In the electrical environment, you, you can, you have essentially a physical connection between the two points that try to talk to each other. And so, when the physical connection is there, you may send messages as you were saying before, like, I'm not sending, I'm not sending, I'm not sending, but you can do this, for example, once it be, I don't know, 30 seconds or so.

It will be enough to, yeah, keep the connection alive. Whereas if you use optical, it's different. Because if you use optical communication, the line does not exist. The line between the two exists only because you have a laser that is sending some photons from one end to the other. So, where it's different is that a laser is an access component.

You need to, to, to send energy to create this link between the two ends.

Chris Adams: Mm.

Romain Jacob: And so, now, it's not that every 30 seconds you need to say, I'm not sending anything. It's like, all the time, you need to have this laser on so that the two endpoints know they are connected to, to each other.

Chris Adams: Ah, okay.

Romain Jacob: And it's actually one of the reasons why the early ideas about tweaking are not so much in use today is because they don't work nicely with optical communication.

And optical communications are the de facto standard in networks today, in the, in the core of the internet and in data centers as well. For reasons that we don't have time to detail, optical is the primary means of communication. And it is by design the laser needs to be on for the link for the communication to exist.

Whether you send data or not.

Chris Adams: Alright, okay, thank you for elucidating this part here. So it sounds like the models we might use a lot of the time, as lots of, when you're working with digital sustainability, it's very common to look at a kind of figure per gigabyte sent, for example, and like in some cases It's better than having nothing, for example, but there is a lot of extra nuance here.

And, there is, we have seen some new papers, I think there was one paper by David Mytton, who, that we'll share a link to, he's been speaking, he's, he shared one recently about the fact that, there are other approaches you might take, for example, for this. There, if you could, just brief, it'd be really nice to just touch on some of that, if we could, and then just, and then to add some extra nuance, realize that, like, It's not that there is no proportion, because there is something you need to do.

Maybe we could just talk a little bit about some of the things that David Mytton's been proposing as an alternative way to figure out a number here, because I'm mainly sharing this for developers who get access to these numbers, and they want to make a number go up or down, and it's useful to understand what goes into these models so that you are incentivizing the correct interventions essentially.

Romain Jacob: Yeah, of course. So that's actually a very important point, I think. You will often find if you look in the, on the web or anywhere. Figures are in energy per bit, or energy per X, or energy per web search, or energy per email sort of thing. Whatever we can think about whether computing such numbers make sense, what, what is very important to understand is that those numbers were derived in an attributional way.

That means that you take the total power cost of a system. And then you divide by the number of bits that were transmitted. If you take a network, you take the sum of the energy consumption of all the routers and all the links and all the calling and all of everything. And then you look at the total amount of traffic you've sent over your reporting interval, like a year.

And you take one, you divide by the other and ta da, you get energy per bit. That is interesting. That is interesting to get an idea of how much, how much energy you spend for the useful work you've done in that network. But it should not be interpreted as, this is the cost for a single bit, because if you do this, then, and that would be a different type of reasoning that we call consequential reasoning.

If you do this, you would then draw the conclusion, the wrong conclusion, that if I have a network that has, I don't know, a hundred kilowatt hour, if I send a hundred gigabit more, I will use ten kilowatt hour of-

Chris Adams: It increases entire system by that rather than my share of this, for example.

Romain Jacob: Exactly. Except that it's not true. It's not true because the total number in watt, in energy per bit accounts encapsulate all the infrastructure costs. And those infrastructure costs are constant, they are independent of the amount of traffic. And so this summary statistics is useful in order to track the evolution of how, how much is used, how much is your network used over time.

But it's not good to predict the effect of sending more or less traffic. And it's a subtle thing that if you overlook this, you can make the very wrong statement and make bad decisions. And so this is what kind of like these papers you refer to try to highlight and explain. And say that you need to have a finer view on the, the energy per unit that you're interested in.

It's a bit more subtle than that. People should read the paper. It's a great paper. It's very accessible. It's not too technical, I think. And it's great for people that are interested in this area to get a good primer on the challenge of computing the energy efficiency of a network. I think it's really a great piece.

Chris Adams: Okay, cool. So basically the, I think one of the implications of that is that let's say I'm designing a website, for example, if I make the website maybe half the size, it doesn't necessarily mean I have half the carbon footprint of it because some of the models we use and they're popular, they are an improvement on having nothing, but there's extra nuance that we might actually have.

I say this as someone who works in an organization where we have a library, we have one, we have a software library called CO2.js. We have a transfer based model for this because this is one of the ones that's most common that is just like one of the defaults. We also have like an issue open specifically about this paper because there is, when you're starting out, you will often reach for some of these things for this.

And while there's benefit and there's some value in actually having some of these models to help you work out, it's also worth understanding that there is extra nuance to this. And they can end up with slightly different incentives for this. This is something we'll talk about carbon aware as well because again, different ways you measure the carbon intensity of electricity can create different incentives as well. So, like, this is one thing we'll be, I guess, we'll be developing over time, but the thing that I just, if we may, I'm just going to touch on this other thing before we move on to kind of wrap up on this.

This can give the impression that there is no proportionality between using digital tools, and, like roll out of extra infrastructure and if you said there's no link that would be an oversimplification as well and I think we're gonna one of the previous guests Daniel Schien he came on he spoke about some of this and maybe you might paraphrase some of this because I think his this perspective is also very helpful and kind of illustrates why we need to be doing coming up with better models to represent this stuff.

Romain Jacob: Yes, exactly. Thanks for bringing that up. That is also extremely important. So I've said before, power is kind of constant. It doesn't depend so much on how much you send. Two, two things to keep in mind. First, there is some correlation. So if you do send more traffic, there will be an increase in power and so you will consume more energy. That is true. And the work that I'm doing and fuel make that even more so in the future. So what we are trying to do is essentially say. We tried to find ways of reducing the power draw when you're under low utilization. And if we were successful in doing that, by sleeping and by other methods, then it will create a stronger correlation between traffic and power.

Right? So, and this is actually good, right? For energy efficient theories, the closer you are to proportionality, the better. So if we are successful, then the correlation will increase. And then sending more bits or, or having smaller website will have a stronger impact, in energy consumption in carbon footprint.

So that's one aspect.

Chris Adams: Mhmm.

Romain Jacob: The second aspect was also extremely important. You mentioned already the, the work of Daniel Schien that is great about this is to think about the internet in, in, in a different timescale. If you look at one point in time right now, the network is the static element. There's so many nodes in the network.

There's so many networks and therefore today, if I send more traffic; be low impact. However, if you put a longer timescale and you look at a one year, six month, or ten year horizon, what happens is that when people spend more traffic, you see the utilization of the links going up, and that will have a future consequence of incentivizing people to deploy new links, to increase the capacity of the network.

Chris Adams: Hmm.

Romain Jacob: Which means that over a year, so over time, as you send more traffic, you create more demand. As you create more demand, you will create more offer. That means scaling up your network, and every time you scale up, almost away, you will increase the energy consumption of the network, right? So, you will further increase the infrastructure costs.

So, as you send more traffic, as you watch more Netflix today, It does not consume more energy, not so much, but it will incentivize the network to be scaled up, and that will consume more energy. So, there is a good reason to advocate for what is known as digital sobriety, to be, to try to use less of the network or to make a more sensible use of the network, because if we use it more, It will incentivize future increase of the digital network size and therefore future increase of the energy consumption.

Chris Adams: I see. Okay. So basically, if you like set these norms of all this extra use, even though you're not making these changes in the meantime, on a kind of large, on a multiple, multi year timescale that people make investments, like infrastructure investments on, they would then respond to make sure that they've got that kind of headroom available over time.

And I think that's actually some of the work that Daniel has been doing to model, because there is no way that we can do that. We're going to deploy new infrastructure where there being zero carbon footprint, even if everything is green. So there is a, there's an impact there that we need to be mindful of.

That's some of the work that he's referring to there. We'll share a link to that paper as well, because I'm not quite sure how Okay, I know that we can't model that in co2.js, for example, in our library, but he's, this is literally the cutting edge work that I think he's been doing. And he's, last time he was on, he was hiring for some researchers to find out, okay, how do you represent this stuff?

Because when we think about large organizations of the scale of Amazon or Microsoft who are spending literally tens of billions of dollars each year, then you do need to think about these kind of multi year, decade style infrastructure kind of investment scale. Okay. All right. So, we've gone really into the details, then we've spoken at the kind of macroeconomic level now.

I wonder if I can just bring this back to the kind of frame for developers who are like, oh, this sounds really cool. How do I use some of this? Or what would I do? Like, if you wanted to have an internet that was able to kind of sleep more and could grow old, are there any ideas you might use? Like how might it change how you build, for example?

Are there any kind of sensibilities you might take into account? Because as I understand it, some of the things with Hypnos were primarily designed to say, this is how you can do this without forcing people to make too many changes at the end user level. But there may be things that as a practitioner you might make things more conducive to or something like that, for example.

Romain Jacob: Yes, definitely. So I think it goes back to the question we had just before about Daniel's work, about looking at the longer time scale perspective. I think as an end user, as a software designer, I think thinking about sobriety is something that everybody should be doing. It's not just for sustainability.

I think one very recent Environment Variables podcast was about the alignment between the sustainable practices and the financial operations, and in many cases, those two things align. In a similar mindset, if you think about web design, this, if you look at the system, the WebW3C sustainability guideline, they align pretty much almost perfectly with the accessibility guideline and with the performance optimization guideline. Why? Because a smaller website will also load faster and, you know, get people faster to what they want. This is the content they really want to consume. So, there's general value into being as, modest in your demand from the system or from the network as, as possible and for the compute as well.

It's the same thing. Today, it does not yet translate into net benefits. At the network level, but it might in the future. And, you know, somebody has to stop. So you need to, I need the efforts of all sides in order to, you know, make that work. In the networking domain, I know there have been some people studying this from a theoretical point of view where you would say, the end user could be able to say, I want to, I want to place a phone call.

But I'm willing to wait for, I don't know, 20 seconds or 30 seconds before my call is being played. And if you have an ecosystem of users of that network, where the sufficiently large share of users are so called delay tolerant, then you can optimize your network in order to save in resources. So saving energy and ultimately in reducing your carbon footprint.

In the more traditional networking domain, one could envision something like that. There is no work in this area, as far as I know. One way you can think about this would be, the incentive would be pricing. That you, you, you could say, okay, I'm, I'm winning. So not so much to wait the most, most likely to cap the bandwidth I can get out of the network.

But you would, if you were to say, I'm winning to get at most, I don't know, 100 megabits per second in high utilization times, then you would get a discount on your internet deal. I think that's sizable, that's possible. One quick working thing that could happen. But that would be more like a global effect, like it's between the user and the internet service provider.

If you're a software developer, If you think about how your application could be built in such a way, I honestly don't know. I think it's extremely easy today, technically speaking, to have any sort of flagging where your application can say, I'm data intolerant, I can wait, I will not use more than an egg.

One can do that, that's easy. Your network can get that information. The tricky bit is, how would the network then use that information to route traffic in a way that would save energy? That is much trickier.

Chris Adams: Ah, okay, so that sounds like a possible route that people might choose to go. Because I think, I, I know, for example, there is some work in the world of streaming, where there was a notion of a, I think it was the gold button, that was put together by the Greening of Streaming group. They were basically saying, look, most of the time, I, if I'm looking at television from across the room, I can't really tell if it's 8K or 4K.

So, allow me to, you know, have a default which lets me kind of, reduce the resolution or the quality so that when there's lots of people trying to use something, we can see the amount of data reduce somewhat. And that reduces the amount of kind of extra peak capacity people might need, for example. These are some ways to kind of make use of the existing capacity that lives inside the entire network to kind of smooth off that peak as it were, for example.

That's some of the stuff that we might be looking at. So it's very, in some ways, it might be, kind of providing hints to when you send things over the network. And I think there's actually some work that we've seen from some existing tools. I know that Facebook's serverless platform does precisely this.

And there is also some work in Intel, we'll share some links to this, where when you have a computing workload, you can basically say, well, I'm not worried about when this gets delivered, for example, or I have a degree of, as long as it happens before this time, it's okay. And this does provide the information for people running these systems to essentially, like, move things around to avoid having to increase the total capacity, for example.

So you make better use of the existing capacity you have before you have to buy new capacity or deploy new wires or anything like that. That seems to be what you're kind of suggesting.

Romain Jacob: In the cloud computing world, that does exist for real. Yeah, for sure.

Chris Adams: Okay, so there's a possible path for future, future research. And maybe this is the thing I, this is what we can kind of wrap up on. So we've spoke and we've done a dive into sleeping and getting old. Right. But that's not the only tool available to us. Are there any kind of papers or projects or things that you would direct people's attention to that you think is really exciting but may not necessarily be in your field that you think is worth, that you're excited about?

For example, because you spend a lot more time thinking about networks than I do, and I'm pretty sure there's some things you might say that that, that other people listening here might, might enjoy following, for example.

Romain Jacob: Yeah. Yeah. So I think two things come to mind. The first is that today I talked about protocol adaptation, about, you know, we, we would put things to sleep and save energy and so on. But one, one key problem we have, we as practitioners in this field and researchers and operators is the lack of visibility in the power data.

It's actually really hard to get a good understanding about how much power is going to be drawn by a given router, depending on the amount of traffic and depending on how this thing is configured and so on. So, this became very clear at the beginning where we started working on this. And so, a big part of my research has been to try to, to develop tools for building datasets to aggregate such power information in a way that people can contribute to and then use in their own research, do their own analysis.

And try to do some predictions about, okay, now, if I were to buy this device, for example, it will cost me so much in embodied carbon footprint that this is how much I could hope to save. Because I know how this device typically operates and how much it consumes. And so, this led to a dataset on a platform project we call now the Network Power Zoo.

Which is actually a reference to another very well known networking data set that was the Network Topology Zoo. It's kind of like a historical reference to that. But it is really a zoo in a sense that it's very broad. From devices that look the same can consume from, I don't know, two, three times more magnitude power.

Whereas it seems there's the same number of ports, the same part of the number of connections, but it can change drastically. So, it's still a work in progress. I mean, the database is in building and we're starting pouring data in. And very soon we'll do an open call for anyone to contribute their own data sources into this database.

So that people can have access to richer power related data.

Chris Adams: For a data informed discussion. Yeah.

Romain Jacob: So that's one thing that is very active for us. It's still very much related to networking, but that's one thing that is not related to protocols so much. More generally, what I think has become clear and clear to me is that if we want to address the sustainability problems in networking, what we need are not really networking researchers, because a lot has to do with the hardware design, and the hardware architecture, and writing good software for optimizing the hell out of the hardware we get.

And those are just not the typical expertise that you find in networking people. So, networking people are a bit at a protocol, but, you know, they don't know as much how the hardware is built and designed. Maybe I should not make such generalities, but it's definitely true for myself. So, we've been poking more and more people from the computer architecture area, from the hardware design to collaborate with us and say, okay, look, we have those sorts of needs.

This is mainstream in embedded systems for 20 years. We still don't have it yet in routers, it's hightime, we need it now. If anyone is working in this area and is interested, please reach out to me, I'll be happy to chat.

Chris Adams: Brilliant, thank you for that. Okay, and I suspect there's some, I'm just looking through our notes. We actually have been in touch on the E Impact mailing list, which is one project, but I think it's by the IETF. That's one of the things that we can share a link to where there's often quite a lot of, if you want to go into the networking, that's probably one of the deeper ones I've found.

Okay, great. I think, is there anything else? I should, we're just coming to time, so I just want to check. This has been really fascinating, and I've learned a huge amount from this. So, if people are interested in the work that you're doing or they want to learn more about this, where should people look to find more about this?

Like, we'll share a link to the paper that you worked on, for example. But beyond that though, where do we find out what's going on with Romain Jacob and his team of research, researchers?

Romain Jacob: So, nowadays the best place to find me would be on LinkedIn. So, we'll add a link to my profile, but my name is not that common, usually I'm findable on LinkedIn at least. Yeah, that's the best place for you to reach out, I'm quite reactive there.

Chris Adams: Okay, cool. So I'll share the link there. I'll also share the link to HotCarbon, which had the paper that you had. If there's a, and if there's a link with ETH, the research institution you're part of, I'll add a link to that as well. Brilliant! Well, this has been really enlightening for me and hopefully other people who've been listening along with this.

Thanks once again for being so generous with your time, Romain, and, yeah, have a lovely holiday over the summer, okay? Take care of yourself.

Romain Jacob: Thanks, bye bye.

Chris Adams: bye! Hey everyone, thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser.

Thanks again and see you in the next episode!

Preview: CXO Bytes

11 juli 2024 | 12 min

In this episode we preview the first episode of CXO Bytes. Join host Sanjay Podder as he talks to leaders in technology, sustainability, and AI in their pursuit of a sustainable future through green software. Joined by Dr. Ong Chen Hui, Assistant CEO of Singapore's Infocomm Media Development Authority (IMDA), the discussion focuses on Singapore's comprehensive approach to digital sustainability. Dr. Ong highlights IMDA's efforts to drive green software adoption across various sectors, emphasizing the importance of efficiency in data centers and the broader ICT ecosystem. Listen to the full episode via the links below.

Listen to CXO Bytes:

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Chris Skipper: Hello, and welcome to Environment Variables. This is Chris Skipper, the podcast producer, bringing you a very special episode of Environment Variables. Today, I'm excited to share with you a preview from our new podcast series, CXO Bytes. Hosted by Sanjay Poddar, Chairperson of the Green Software Foundation.

CXObytes dives deep into the intersection of innovation and sustainability within the tech industry. In each episode, Sanjay will be joined by industry leaders from the C Suite to explore practical strategies for greening software while driving enterprise growth. So without further ado, let's get started.

Let's listen in on a sneak peek of the first episode.

Sanjay Podder: Hello and welcome to CXO Bytes, a podcast brought to you by the Green Software Foundation and dedicated to supporting chiefs of information, technology, sustainability, and AI as they aim to shape a sustainable future through green software. We will uncover the strategies and a big green move that's helped drive results for business and for the planet.

I am your host, Sanjay Podder.

Hello everyone. Welcome to CXO Bytes. This is our inaugural podcast on how do you use green software for building a sustainable future. This is a new podcast series and the whole idea behind it is, you know, embracing a culture of green software, it needs to come from the top. And we therefore want to talk with decision makers, with business leaders, with leaders who are running nation states like Singapore, for example, at C level.

You know, how are they driving this culture change when it comes to digital sustainability and green software, for example?

Today I am super excited to invite Dr. Ong. She is the Assistant CEO of IMDA, which is the Infocomm Media Development Authority of Singapore. And we are going to chat on how IMDA is championing digital sustainability as well as green software. Welcome, Dr. Ong.

Dr. Ong Chen Hui: Thank you for having me on your inaugural podcast on green software.

Sanjay Podder: And you know, I had my own selfish reason for inviting you because while the Green Software Foundation has been interacting with many, many large businesses across the world, IMDA and Singapore GovTech, these are two members of Green Software Foundation who represent the government, right?

And we all know the very important role that government will play in sustainability in general. So I wanted to understand from you, you know, how are you looking into this space? So we will talk a lot about that. The other aspect is probably to begin with, for our audience, a perspective on what is IMDA.

You know, what is your specific remit, what you are trying to do in Singapore, if you can give us, you know, a few insights into that.

Dr. Ong Chen Hui: Okay, so here in Singapore, of course, climate change is actually something that is a bit of a existential thing for us, us being a small nation state and we're also an island, to us, climate change and the associated rising sea level is a matter of concern. Right? So, as a result, we have put in a green plan that states our, sustainability goals by the time we reach 2050. And this is actually a whole government effort. So, I don't think it is a case where it's one ministry or one agency that's responsible for the whole world. It is about the whole of government working together in order to make sure that we meet the goals of our Green Plan.

Now, what are some of the things that we are doing? Many things, for example, the National Environment Agency is actually rolling out some of the regulations. We have things like e-waste management, for example. Just now you mentioned GovTech, which is our sister agency. GovTech is also rolling out green procurement when they're actually procuring software solutions. Within IMDA, we are responsible for some of the industry development. We're also what we call a sectoral lead of the ICT sector. So, our own green strategy, comprised broadly of three different strokes. The first is about greening ourselves as an organization.

The second is really about greening the sector that we are responsible for, that we are leading. So, in that case, there will be things like the telecommunications sector, the media sector. And the third thing we want to do is to enable our ICT solution providers to provide green solutions to the broader economy so that we can scale the adoption, we can ease the friction out there in the ecosystem.

So essentially, that's greening ourselves, greening the sector, as the lead. And the third is really to kind of provide solutions through the ecosystem so that the wider community can actually benefit.

Sanjay Podder: Now this is really a full 360 degree kind of approach and it is phenomenal. And, I was, I was wondering, you know, and you mentioned briefly on Singapore being an island state. I was thinking, why digital sustainability? What will happen if Singapore decides not to do it, for example, right? Do you have a point of view, say, because, you know, there are many different levers of, sustainability, you know, I can understand the larger sustainability, but what is the importance of digital sustainability?

Do you think it's an important enough lever or maybe you can look at nature biodiversity or something else, right? So specifically for digital sustainability. What is it that triggers IMDA that this is a important initiative? And I'm, I'm seeing this is my second year in Asia Tech that, you know, this is something you give a lot of importance to.

Bringing in leaders from various organizations. Doing deep deliberation. I also remember last year, you brought out your new data center standards, I think increasing the temperature by one degree that has an implication. If you could throw a little bit more light on digital sustainability in particular,

Dr. Ong Chen Hui: Mm hmm.

Sanjay Podder: why do you feel that's a very important lever for a country like Singapore and maybe for many other countries around the world?

Dr. Ong Chen Hui: Yeah. Well, I think you're actually exactly right that when we are trying to drive sustainability, actually there are many different strokes. Some of it includes looking at energy sources and all that, which actually is also very important for Singapore because we are small. We do, have to look at, different kinds of energy sources and how we can potentially actually import some of them, right?

Now, when it comes to digital sustainability, actually our journey, I would say started many years ago. Maybe more than a decade ago, when we started looking at, some of the research work within the research community about, making sure that our data centers, can operate more efficiently in the tropical climate.

Now, data centers, comprise of almost a fifth, of the ICT carbon emissions. And because they are such a huge component of the carbon emissions, of course, their efficiency has always been top of the mind. Now in the tropical climate like ours, a large part of the energy sometimes is attributed to the cooling systems, right?

The air conditioning that's actually needed to bring the temperatures down. So as you rightly pointed out, what we found out is that actually if you were to increase the temperature by one degree, that can lead to a savings of between two to five percent off. Carbon emissions. So, and that as a result, we have been investing in research within our academia, funding some of the innovation projects with our ITC players, in order to look at what actually works and what doesn't.

Because I think in Singapore, regulations always need to be balanced with innovation. So that have kind of, led to what happened last year, which was that we released the first, standards for tropical data sensors. But we wanted to go a lot more, right, because some of those standards, around cooling and all that, that's kind of like looking at how efficient the radiators are in a car.

But we also need to look at how efficient the engines are. And the reality is that, if you look at the trends of ICT usage of software applications. I mean, so much of our lives, whether it is watching videos, watching TikTok, right, our education, around all that, most of this have moved to become, to be enabled by digital technologies.

And when we look at the consumption of, data centers and the kind of workload in it, it is increasing year by year. Now, with the explosion of AI, we know that the trend is probably that there will be more consumption of digital technologies. And those are the engines that sits withinssb the data centers.

And we need to make them efficient. And as a result of that, we have decided that we need to also get onto this journey of greening the software stack. And greening the software stack means a few things. The first is, of course, I think this is still a fairly nascent area. How do we make software more measurable, so that there's a basis of comparison, so that we can identify hot spots that I think is important.

The second part that I think is important is also, given all the trends today, GPUs, CPUs all needing to work together, how do you make them work efficiently? How do you process data efficiently? How do you make sure that the networks and the interconnects within the data centers are efficient.

I think all of these are worthy problems, to look at. Some of it will rightfully stay, still in the research stage. So we'll be funding, research programs, called the Green Computing Funding Initiative around it. But at the same time, we also think that there are some practices that may be a bit more mature already, and we should encourage companies to actually innovate on top of it.

So we're also conducting green software trials.

Chris Skipper: Hey, everyone. I hope you enjoyed that preview from CXO Bytes. If you want to listen to the rest of the episode, please go over to the CXO Bytes page on wherever you find your podcasts. Just search for CXO Bytes and enjoy the rest of this insightful conversation between Sanjay and Dr. Ong Chen Hui of the IMDA.

And to listen to more episodes of Environment Variables, Please visit podcast. greensoftware. foundation. Bye for now!

The Week in Green Software: AI’s Power Problem

4 juli 2024 | 53 min

In this episode of Environment Variables, host Chris Adams is joined by Asim Hussain to dive into the complexities of AI's growing energy demands and its environmental impact. They discuss innovative approaches to sustainability, such as using fungi to manage building waste in data centers and the potential for greener materials and practices. The conversation also covers software optimizations to reduce AI's carbon footprint, emphasizing that energy inefficiency cannot be outsourced. They highlight the importance of integrated sustainable practices in tech development, particularly in the face of increasing AI power consumption projections.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Asim Hussain: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Asim Hussain: And now there's a big hoorah in our space because like AI's now gone and

our automation has now gotten the point where even our jobs are now we're all really like nervous and upset and but you know this has been the pressure that we've been applying to the rest of the world with all the industry for decades and decades and it's just now coming to us and affecting us so you know we don't really have a leg to stand on I'd say.

I'm your host, Chris Adams.

Hello, and welcome to Environment Variables; This Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. Today, in our news roundup, we're diving into some of the pressing issues at the intersection of AI and sustainability.

So, with AI rapidly advancing, the energy demands of training and running these models are also seen to be skyrocketing. Posing significant challenges for the environment. We'll also be touching on some legislation for promoting sustainable business practices amongst AI companies and the potential for a green levy to drive investment for greener eco-friendly technologies.

We'll also be talking about some of the latest papers that have been published for people trying to understand and get grips on cloud carbon emissions. And finally, we'll touch on some of the exciting events in the green software community, including conferences, workshops, and masterclasses aimed at fostering sustainable development practices.

Joining me today for today's news roundup is my longtime friend, Asim Hussain of the Green Software Foundation. Asim, for people who've never listened to the podcast before, can I give you the floor to introduce yourself and some of your background?

Asim Hussain: Yeah, sure. So my name's Asim Hussain. I am the executive director of the Green Software Foundation. And yeah, I've been at the intersection of sustainability and software. I've been very lucky to be thinking about the same question about the intersection of sustainability and software for quite a few years now. And yeah, I've been well, mate. I've been, I've just come back from a vacation, which has been long time coming. I do have a little bit of a, cold, so that's what would explain the slightly nasally, annoying nasally sound the audience members are going to have to experience for this podcast.

Chris Adams: So this is with your hot beverage and Nurofen chaser

Asim Hussain: hot beverage. Yeah, that's what I like to do. That's how I like to start every podcast episode is a coffee and a Nurofen, every conversation with Chris Adams has to have a, both a coffee and a Nurofen.

Chris Adams: Wow, that's, one thing to take away with me. All right, folks, I should just briefly introduce myself before we dive in. I am the executive director of the Green Web Foundation. It's a small Dutch nonprofit based, well, in the Netherlands, where we are working towards an entirely fossil-free internet by 2030.

I also am one of the hosts of this podcast here. As well as one of the organizers of ClimateAction.tech, which is an online community, where actually me and Asim first met online before he basically headed off to set up the GSF in its current state. Alright, Asim, are you sitting comfortably?

Asim Hussain: I'm standing uncomfortably.

Chris Adams: uncomfortably at your swanky standing desk.

That's good enough for me. Should we start with news then?

Asim Hussain: Yeah, let's go for

Chris Adams: Alright, okay, so Asim, I was thinking of you when I saw this paper, this story. The first one is a story about mushrooms eating building waste in data centers. So this is a link to the Data Center Dynamics website talking about specifically the use of these, of essentially building-waste-eating mushrooms at the Meta data center and other ones.

And the general spiel of this is that there are now a number of companies which are, essentially deploying fungi, various kinds of fungi, to deal with all the building waste that ends up being created when you might kind of read, when you're essentially demolishing a building or creating a new one.

And, it essentially takes all this waste, and the fungi are able to Essentially, deal with the toxins, and then create something like, kind of, fungi-style bricks that can then be used as a kind of circular building material going forward. And Asim, given that you're our kind of resident mushroom fan, I wanted to just like, see what you thought about this, or if you had any particular immediate like, hot takes or things when you saw this one?

Asim Hussain: No, that's all. it's a great application. In fact, it's not an uncommon application of, you know, what people are applying, you know, fungi in this technology for. It's actually one of the, one of the, one of the, one of the very exciting kind of broader sustainability solutions in this space that there is.

I mean, there's a couple of different types of fungi. You're going to have to pause me at some point. There's a couple of different types of fungi, but there's one particular type was kind of saprophytic, which is effectively what Fungi, the purpose it has in kind this life that we lead is it basically, it's the thing that destroys things that have died. And if it didn't exist, then we'd be basically living on top of this massive mountain of logs that aren't decomposing. So that's one of the things that they're really, really good at. And it's been, it's, there's been a lot in many different spaces, been a lot of active research, a lot of startups, a lot of organizations exploring how to use Fungi to decompose things that they don't normally decompose. And it's actually quite an interesting technique. Because even people, what's so fascinating about the the fungi space is it's, driven in large part by citizen scientists, which is one of the things I love about it. And there's a lot of citizen scientists out there who are doing things like trying to find a strain of fungi, which can decompose certain types of plastics and you would literally do this. You would literally grab a selection of these plastics. You put them in a, you know, you can go online. It's as simple as this. You put it in a blender, you blend up a plastic in your home blender. And then you just have like, as you can probably see behind me, I have lots of jars out there, which have like different strains of fungi, and you would just put that with other material in the jars.

And then you'd collect lots of different strains. Like. Every single kind of mycelium is like of a different strain of it of the same one. So you can have like millions and millions of different types of strains. You go into the forest and you see a type of mushroom you've never seen before. And you're like, ah, maybe that will absorb this plastic. And so there's a lot of interest in this trying to like find those strains of fungi that can kind of absorb and transform, you know, different materials. Obviously certain fungi only in the forest, they only work on certain types of trees. They have like a relationship with them, but you can actually find strains of fungi, which do different things. And kind of the interesting thing about it is about turning them into, bricks like that as well. There's actually organizations out there trying to replace packing material for boxes. But what you do is you basically, you create like, so you take some time. It's not like a foam that you stick in and like 30 seconds later, there's a thing you basically have to Have the thing you want to pack in a box Inside the box you put like a substrate which can be the thing that mycelium grows on you then like almost like impress inside this substrate the shape of

Chris Adams: like the

Asim Hussain: book or something the mold and then you inject it with the mycelium and you put it in basically an oven for like a month. And it comes back out and then you basically spray it off and like the actual mycelium has grown into the shape of the thing. And then you've got something which you can put in the packing crate. And then at the end of the day, you just, it's a mycelium. You just break up and put it into your garden and you, and it decomposes.

There's lots of like wonderful stuff. There's a, great guy called Paul Stamets, who's quite a character. But he's done a lot of, he's done some great TED talks in the use of like fungi and mycelium in, kind of waste and getting rid of waste even getting rid of oil and there's a lot of kind of very active stuff in this space.

Chris Adams: Can I stop you one second there, mate? Cause, cause there's, so you just said Stamets, right? So Stamets hit, so in real life, there's someone called Stamets and in the whole, and that's, so the whole, when they had this whole Star Trek discovery thing, where there was this guy called Stamets who was using like the mycelial internet space thing, that's a direct reference to this dude, basically.

Asim Hussain: That's a direct reference to Stamets, which is kind of, which joins two of my biggest nerd bubbles together in the most beautiful way, but yeah, there's the engineer, and in Star Trek Discovery, how they, instantly, there's a new type of drive, and they instantly can, move to one part of the universe to another, and it's called a spore drive, and you need to kind of enter this kind of psychedelic realm to connect transmission, so yeah.

Chris Adams: My word, Asim, I was not expecting us to dive down that myco rabbit hole for that, but that was a lot of fun. Thank you very much. So you basically said. By doing this, so in this case of packaging, this basically removes the need for like, say, fossil-based expanded polystyrene in packaging, and in the case of materials here for buildings, you would use that instead of having to get a bunch of virgin materials, for example.

This would be like a circular, that's the approach that they'd be using here, right?

Asim Hussain: So what, so the specific approach they're using in this particular article is, I believe it's more about, it looks like they're using, basically trying to get rid of the drywall that they have inside the data centers. So I don't know if it's a particularly from a decommissioning a data perspective or renovating a data center perspective, but they're ending up with a lot of material, which typically you would just dump in a waste fill, but now they've basically got a form of mycelium, which can eat drywall and generate something that's, decomposable, maybe edible.

Yeah

Chris Adams: All right, cool. And so we, this is, mentioning, referencing the metadata center in, I think, Tennessee. But we've also seen Microsoft, as far as I'm aware, Microsoft has also been a bit of, it's been dipping their toes into this field as well. And one of the reasons why you might care about this is that, well, last year, Microsoft's reported emissions, when they released their sustainability report, it was like, up 30 percent and a significant chunk of that came from buildout of data centers.

So we are now starting to think a lot more about the embodied energy in the facilities that are created so that we can actually have data centers, so we can actually use compute, a lot of the kind of compute power available to us, or even some of the AI power, or the kind of sources of AI and stuff like that, because you need, they need to be in a building somewhere to get this stuff built.

And like, this kind of made me wonder actually, Asim, surely I imagine some of this might show up in an SCI score, a Software Carbon Intensity score, if you're purchasing cloud from a certain place. If you've seen a massive buildup of data centers, surely that might have an impact on the embodied carbon for the compute you might be using, right?

Asim Hussain: It could do, it depends on what your, cause in the SCI, there's kind of two components. One is the software boundary, which is "what are you going to include and not include in that score?" I remember us having quite a few conversations in the early days of the SCIs, which "should you include the, like the concrete that was used to pour the floor of the data center?" And I'm not too sure we really, I don't, think anything particularly made its way into the specification, but it has to be, if you're measuring something, it has to be something which drives a choice or a behavior. So, you know, I suppose what I'm going to with this is, if there was a data center, which was particularly built like a zero carbon, maybe built with mycelium or something.

I don't know, but like, if there was a data center, it was particularly built with that choice, then maybe it is something you want to include in the score, because then that can drive an action of choosing one option over another. But if every single data center is effectively built exactly the same way, the discussions we were having was, well, that's just overhead of adding a sec, effectively a coefficient, which wouldn't really drive a decision-making factor. So, I suppose as what I'm going to this is excitingly, if there are data centers that are being built, which are going to have

vastly different embodied carbon profiles, and then you, and then if that was included in an SCI score... and I think as we move forward with SCI, because one of the things that's happening... SCI on its most, we, launched kind of the version one and now 1,1 of the SCI and it's, and it was, it's very bare, very basic. It's designed to be built upon. And so now what the teams are having conversations around is like, if you were to apply SCI to AI, specifically, what does that include? And one of those questions

they want to answer is what,

Chris Adams: building,

Asim Hussain: yeah, like what you include so, so when someone reports an AI score, we can actually start getting to like some apples to apples comparisons. So that's, a conversation that, you know, it's interesting. Maybe we should bring that, bring it up again is do you include the embodied of the data center? But then you also get into the headache of "my God, it's hard enough trying to figure out what the embodied of a chip is."

Chris Adams: Yeah.

Asim Hussain: And now going to ask to figure out like what the embodied, so there's has to be some practicality aspect to this as well.

You know, you know, we have to, there should at least be some models. I don't know.

Chris Adams: So there are models that, as I understand it, there are models for working out the, essentially like the carbon emissions for a kilo of concrete, for example, or stuff like that. Some of these exist. And,

Asim Hussain: exists from a

LCA,

Chris Adams: and there are companies, and we know there's like, in Europe at least, I know there's a company called LeafCloud.

That are explicitly using, are doing, making reused, or, they're reusing heat, but also a very specific kind of data center, which isn't like a very large outta town thing. They have like essentially shipping containers put into places like say greenhouses, where the heat is being reused, for example, and where they're not having to build a whole bunch new buildings. There's also, I think in Switzerland, there's one company, because we maintain, where I work, we maintain a directory of green data centers. And one of them is a, they basically reused a old factory building with a waterwheel that used to be kind of like a clothing factory, and now it's a data center.

So they've essentially reused the whole building shell. They haven't built a load of stuff as a result. So this is one place where this might show up, but in order to do this, you need to have access to the numbers for this. And that's still a bit of a challenge because, yeah, we don't have the, we don't have easy access to these numbers, and like you do say, it's a challenge just thinking about chips, let alone expanding the boundary to the actual buildings instead.

Asim Hussain: Yeah. I mean, at least you have some information when you're running software, like what, you know, you can, now that we've done, a lot of that workings out so you can figure out, you know, perhaps it's this chip, but I think given the secrecy around data centers, I don't know, I think there's going to be,

I don't know.

Chris Adams: It's gonna be a challenge, because while we have this practice of, essentially, water usage and electricity use, so many things being under NDA, it'll be very hard to come up with some numbers without using, like, a basic number. Okay, alright. We have totally gone past talking about mushrooms and data centers into all these other things, but I guess this is part of the whole thing about sustainability and technology.

It's all interconnected.

Asim Hussain: It's all interconnected.

Chris Adams: Shall we go to the next story?

Asim Hussain: Yeah. Let's go for it.

Chris Adams: So this is a piece from Bloomberg, actually, so this is the topic of this is Solutions for AI's Energy Inefficiency Can't Be Outsourced, and this is an opinion column from Bloomberg talking about this projected demand some of the figures which are pretty, pretty impressive, they basically say, in the US at least, it's, there's a projection saying that AI, the growing demands of energy are like, is projected to make up around 8% of the US' power consumption, electricity consumption up from 3% in 2022.

Now these numbers seem a little bit high and they are citing a kind of this arms race of different kinds of organizations, essentially building out these massive data centers but also buying loads and loads of chips, but it does talk about some of the approaches that we're seeing now to kind of rein in some of this growth.

So one of the things was this idea of one-bit architecture, which is essentially, I'm not going to pretend to understand it. And I'm not sure if you are similarly informed in this one, but

Asim Hussain: I'm going to definitely pretend to understand it.

Chris Adams: In that case, I'll hand over for you to confidently bluff it around, just like a ChatGPT would actually, Asim, the floor is yours.

Asim Hussain: I'm going to, I'm asking GPT. No, I'm guessing, and this is, I haven't really, I've, seen it, but because it's one bit and a bit can only be one or zero, I'm guessing what this is, that, you know, instead of like pumping in a number between one and 256 as one of the inputs to a node in a model. Maybe you could just try one or zero. And then output one or zero and then see if that actually still gives you some pretty reasonable results and from what it looks like it might do and you know for those of us I kind of it's like there are extreme inefficiencies you can do when you're working at the bit level in terms of programming and computation and instructions on the chip and things like that because it's so much lower level than the architecture.

I presume that's what it is, which actually to me is. It's really exciting from a, from the level of, this is a software architectural solution, which is effectively, I think what we've been advocating for, a large part of the time, which is, you know, we do need to I hate to use the word code because I don't want people to dive down the, you know, building more. I, well, AI is one of the, one of the few areas I would say where actual code efficiency is extremely important. But yeah, this is kind of, it's interesting. Now that there has been a pressure applied to optimize, the optimization has happened.

Chris Adams: Yeah, that's, so basically, Asim, I think you're about right. Now, when I, remember when I skimmed over this paper before, one of the key ideas was, the one bit approach was, essentially, when you, would be able to, you would use this to encode the difference between different parts of, like a dataset rather than showing absolute numbers.

And one of the things that this allowed you to do was allowed you to just use addition rather than multiplication in some cases. Now, I'm not an AI specialist and I'm not a hardware specialist, but the general idea was by representing things in a more somewhat simple fashion here, you avoided having to make some of the expensive calculations that you would otherwise need to do.

And this basically reduced the energy that you might need to run some of these calculations. So this was like one example. I was quite impressed to see this inside Blumberg because it was a very quite new research, but also really, technical and actually quite promising. So yeah.

Asim Hussain: It's, it, the title is interesting though, isn't it? Cause it's not like, it's not like software making, it's kind of talking about energy inefficiency can't be outsourced.

Chris Adams: Yeah,

Asim Hussain: I just thought it was an interesting, it's like no one really knows or cares or thinks about the software side of the, this whole equation.

To me, this is just like a software optimization. So you would just say like software can be optimized to reduce any AI's energy footprint. It's not, not expressed in that way. It's interesting. Kind of expressed

Chris Adams: there's a couple of things that I think are also really interesting about this piece, in my view, was that it talks about the kind of economics around some of this, and basically the idea of outsourcing this is essentially how we have a bit of a tendency in the technology industry to say, "well, we're just going to have, like, We realize that data centers use loads and loads of power, so what we're going to have to do is just somehow get loads more power."

And so you basically have people talking about, oh, obviously the solution is to deploy loads and loads of nuclear, for example, right? Never mind that these take between minimum 10 years to get built, right? So, what are we going to do in the meantime? A lot of the time it's likely to be coming from things like gas, if you're going to be using something like that.

So that's an issue there. But it's also worth thinking a little bit about these figures that were mentioned in this story. We've seen numbers like 8 percent of the USA's energy consumption by 2030. It's worth bearing in mind that these numbers are often coming from the utility providers in various states, all right?

So like, say, in, say, Virginia, I think it might be, I forget the name of the actual monopoly provider, but there's only one provider over there. And basically I think that's Pacific General. I think that's actually on the, other coast, basically.

Asim Hussain: We've played this game before where you can name all

Chris Adams: was Caiso last time, which was, which is, that's the Independent Systems Operator, which is not the energy company.

The energy companies are somewhat different because it's a, because in many, because they have very specific, if you're in a state where you've got a single provider, they are allowed The only, the reason you only have a single provider is that you have basically had that state agree to have a, what's referred to as a natural monopoly.

So, they basically, the agreement is, we will give you a guaranteed 10 percent net profit plus for your organization, alright? But you need to basically, yeah, as long as you agree to share your plans, For the new infrastructure you're going to build over the next few years, but also you need to justify this in each of these cases.

And when you think about this, if you're going to get a 10 percent net profit from that for any of the energy you, get. Now, what you, if you want to increase your profits, what you need to do is you need to say, "well, I have loads more demand coming. I need to like double my expected demand to double my profits inside this."

Asim Hussain: Oh, I see.

Chris Adams: is one of the things, because this is a lot of the existing providers, they're used to saying, "well, we've got all this extra demand. What we need to do, we need to build a bunch of new gas, fired power stations. And 'cause we know we're gonna make a 10% guaranteed profit on all the infrastructure we, build.

That's basically, you know, we are incentivized to say it's gonna be really, high" basically. So. You, it's, really worth looking at a paper by one, one gentleman, John Kumi, who's actually, who's spoken a lot about this, because 20 years ago, we had the similar thing when you had people in the coal industry saying, "well, coal was what powers the internet, so you need to have more coal fired power plants if you want more internet."

We have a very similar thing happening. In this case as well actually. So it's worth bearing in mind that yes, we do see these kind of apocalyptic forecasts for energy, but you also see that when you do have constraints on this because it's so difficult to build, then we do end up with a renewed interest in energy efficiency.

And even at the kind of like energy level, right, there are different ways that you can basically meet demand. You can meet demand by adding new supply, but you can also meet demand by investing in energy efficiency. And that's, and this is very much what it looks like, so a lot of the ideas you might see at the energy sector, I think, are at least applicable, or at least relevant in what we talk about with cloud, because essentially you're looking at a kind of commodity that you pay for on an hourly basis, or something like that.

Asim Hussain: Well, that's kind of one of the... All I see is there's, a significant amount. There's not a significant, there's a fixed amount of investment and focus that organizations can put into something. And if you present them with an option, either put all this engineering effort to make something more efficient, which costs 10 or buy renewable energy, which costs five and then, well, I'll choose the five one. So I think that's kind of, that's. That's why kind of investment goes in one way or the other. Whereas I suppose what's happening now is that energy is now, we're reaching the point where energy is, and I'm just throwing out numbers here for energy is now costing 12, but the development still costs 10.

So like, well, maybe we'll put some money into development

Chris Adams: Yeah.

Asim Hussain: So that's kind of, and that's interesting that's why you want things like we were talking about levies and money last time. That's kind of why you want to change that balance of it a little bit. And that's also significantly why in the SCI specification, the decision was made not to include any energy offsets or anything like that.

Just because if you gave somebody the option of spending 1 instead of spending 10, they would spend 1. And we want people to spend the 10 to actually make things more

Chris Adams: Yeah, to address the consumption issue, rather than just think about the intensity.

Asim Hussain: Exactly, yeah. But this is exa I'm really excited. And the other thing I was thinking about as you were talking was I was just remembering about my time at Intel. I think this maybe links a little bit to the Nvidia's statement as well. So, I think I might have, I always love telling people this story because I just think it's such a cool word and it's such, it tickles my sci-fi bone so much. But there's a statement they used a lot, which was dark silicon. Have I told you this? I

Chris Adams: No, you haven't.

Asim Hussain: people. There's a dark silicon. And I was like, "Oh, that sounds good. What's the, what's dark silicon?" And what dark silicon is when they're kind of looking at a chip and they put load on it.

And the key thing with a chip is how much can it expel heat and still function at that level. So looking at heat on a chip. And so when they're running a certain software on a chip, they'll put like a, what you call the heat detecting camera on. And you know how they look like this. It's very red, it's very red on the bits that are hot.

And it kind of looks

Chris Adams: Ah, okay, yeah, look at a thermal house,

Asim Hussain: Thermal imaging. yeah. Yeah. Even though the black might not be like ice

cold, quite hot, but like relatively it's cold. And so the things that they would be really like thinking through is like, how come this software, how come half the chip is black?

Like, why aren't you using the rest of the chip? Like you've maxed out the chip, but half the chip is black. And really what it kind of, you know, what it goes back down to is that, you know, we called it, I think I might call it the silicon gap, which is the gap between what engineers are building and what silicon manufacturers are enabling on their chips. And there's this disconnect between, you know, they're, all building, "why aren't you using this, these more advanced chip sets that are more efficient? Why are you using this stuff on this side of the chip?" And so I think that's something that we need to get down and tighten that gap to use this infrastructure more efficiently, I think over the years, from a developer's perspective, it's always been about time to market.

How do we beat our competition? It's never around, how do we use this chip more efficiently? And so I think that one bit architecture is, it sounds like an example of that. It sounds like an example of, we want to leverage the instruction set on this chip to be as efficient as possible. We need to change. Fundamentally how we're architecting and even thinking through algorithmically this code to take advantage of that. And that's, I think it's also like this, other area, which is completely, we're just ignoring, you know, there is this dark silicon and honestly, the silicon manufacturers are like, why are developers not, I don't understand, we put so much energy and time into like

building

Chris Adams: we're only using a of it, right?

Asim Hussain: a The percentage of it, and there's that also, and I'm going to, I'm rambling on for a second, but just one more, one more point. I thought it was really interesting. One of our organization's Entity Data, they did a really great report. It was kind of two years ago now, I think. I don't think we really circulated.

I don't think we circulated that well. And it was, they just, they just looked at Java, you know, Java, like still, there's still a lot of very antiquated Java applications running out there in the world. And they just said, what is the energy difference if we just upgraded, not the code, but the JVM, the underlying JVM.

And that's all they did was I think they upgraded, I cannot remember. I'll find a link to the article and the paper. It was like several steps up. But they were like, "look, most apps are still running on whatever the JVM was they were built with like 10 years ago. And it was a seven, it was a 60 or 70% energy efficiency improvement.

It was unbelievable. The energy efficiency improvement just from grading the JVM. And that was

Chris Adams: Ah, okay.

Asim Hussain: That, and if you think about what that means, what happens was the chips evolve to have different instruction sets. The JVMs, only the modern JVMs are built to the new one. And so if you're running on the old ones, it's just using the old instructions.

So you're not really leveraging the infrastructure the same way, which is why like recompiling software with like, you know, the latest version of the compiler against the latest version of the chip. It's really important. And it again, that was Intel's, when I was at the time, that was their big push.

They were like, "use the latest bits, use stuff that's compiled now using the latest optimizations." 'Cause they saw a lot of people were still just kind of compiling, leaving that binary, letting it run for like four or five years. And then, that's it. And yeah, I'm going to stop ranting now.

Chris Adams: No, that's actually, I didn't realize, I was somewhat aware of things like the JVM, there's like hotspot or different kinds of flavors of the Java Virtual Machine to run this code, and it's somewhat similar to like in PHP land, like when a new version of PHP came out, because it made much better use of the underlying code, the underlying hardware, you saw a massive increase in performance.

And like, you kind of see something a bit like that with Python as well, with the whole global interpreter lock. Like, I can have a piece of Python that'll be running, and it won't be able to use all the other cores in my machine, in my computer, right? So, rather than lighting up the rest of the silicon, it's got just, it's, most of my computer is dark, basically, in that same kind of approach.

All right, yeah. Cool, alright, so that's like one of the approaches that we have, and this is one thing that you could plausibly do. I've shared a link to a blog post that I've, I was trying to explore this to find a way to explain it, to basically explain the fact that you can reduce the emissions associated with code without actually changing the code, by thinking about what options you might have in terms of, like you said here, like you change the VM or something like that, or change when you run it, or anything like this.

And I'll share a link to that, because I've kind of framed it in terms of If, there are three, three things you can change, basically. You can change the time of running something, which is kind of speaks to carbon awareness. You can change the speed, the amount of compute, computation you might be using, the number of cores you might be using something, or you might change the place, like where in the world you choose to run this for the carbon intensity of the underlying energy.

So I'll add that to the show notes because it might be another nice helpful addition to this. Alright, okay, that was quite, that was fun. Shall we look at the next story?

Asim Hussain: Go on. Yeah.

Chris Adams: Okay, the next story is, this is actually from The Guardian, and this is talking about the balancing some of the incentives of the kind of profits that are projected to come from deploying AI with something like a green levy on these profits, basically.

And this is actually, came from that, Left leaning organization, the International Monetary Fund,

Asim Hussain: Oh, wonderful. Great. Great to see them in this space.

Chris Adams: And basically what they are, the argument from the IMF is basically saying, well, you've got all these very profitable AI firms, and they, and we know there's both the social and environmental impact that's taking place here. So what you should actually have is some kind of green AI win for tax, essentially, that will be used to fund some of the, sustainability initiatives.

And like, to be honest, I have a lot of sympathy for this because what we've seen from the largest providers in the last like year is that given the choice between investing in efficiency or investing in more capacity and building loads more data centers, we've seen all the big providers go for building more capacity and like emphasize profits rather than the environmental impact here.

So it looks like we're not the only people thinking about this. The IMF is thinking about this as well. And they're saying, about this. You need something a levee to do this. Do you have any kind of particular thoughts on this? First of all, Asim, because I was really surprised to see this come up from the IMF of all organizations.

Asim Hussain: Yes, I thought it was a very important point. I love the fact because the use of the word levy, and I remember us having the conversation last podcast about like use of the term levy rather than

tax instead of, explaining it, but then in the article actually uses the word tax, like all over the place. Yeah, I think it's really important. Like one of the things I'd say is that like, why is everybody so excited about AI in the first place? In my most cynical moments as as a software engineer, I would say, Our purpose in life is to either find solutions that help people waste more of their time or get rid of jobs and automation.

If you think about kind of why we have been like one of the most highly paid sectors for quite a significant amount of time, it's because building automations, yes, you could argue and helps you deliver kind of projects faster, but it also helps you to do more with fewer employees. It decreases the.

You know, the earnings potential, the, a lot of this stuff from that perspective, and now there's a big hoorah in our space. Cause like AI is now gone and

our automation has now gone the point where even our jobs are now we're all really like nervous and upset. And, but, you know, this has been the pressure that we've been applying to the rest of the world with all the industry for decades and decades.

And it's just now coming, to us and affecting us. So, you know, we, don't really have a leg to stand on, I'd say. We also. There's nothing we can do. It's happening is the only other thing I'd say. There's no, you know, we have to just accept it and move forward. But yeah, I really liked the idea of that because like, let me put it another way.

Like there's this huge kind of bro down AI tech bro showdown with like, well, I think Sam Altman posted something a year ago now, which deeply disturbed me. And he said he wouldn't be surprised, I think within the next couple of years, if there's a one person unicorn startup, which is a single, like, billion dollar organization run by one human being. And I was thinking to myself, like, that might actually be true. And I would, you know, I do, there is a chance that would be true. might happen in the future, but how do I feel about that? What is the human impact to that? I mean, what is the green impact to that? So I'm now going beyond green because I think that there's like this AI is going to make a few people and organizations immense amounts of power and wealth. How do we have ways to, redistribute all of that and to kind of add a level of fairness to, the rest of society? Is it okay? And so from a green level, absolutely. But I'd also argue from a societal level as well, like, like, you know, like when we talked, you know, how about this? When we spoke about the green transition, it was impossible for us to have like a proper conversation about the green transition without having a real conversation and talking about how we're going to transition the people who are employed in the fossil fuel industry over to other areas. I don't see having, I don't see us having that conversation here as well.

Like it's just ignored. And so I think that's something that we need to have is like, is if you're, if you want to have. The opportunity to get this much power and money and wealth, I think it should come with a certain amount of social responsibility to, you know, be a green levy in terms of the green ones, but I think it actually should be broader than that. It should be, you know,

Chris Adams: So address some of the. Some of the inevitable costs that might be incurred upon society to provide to, like, ease that transition. Okay,

Asim Hussain: yeah.

Chris Adams: I was not expecting you to go there, Asim, but

Asim Hussain: Well, I feeling it.

Chris Adams: Bang yeah, that's usually me kind of jumping up and down, actually. All right, okay, that

Asim Hussain: How do you feel about it?

Chris Adams: I think, so you said this idea, like, I feel uncomfortable about a billion dollar startup with a single person.

And I, okay, how many, we're not that far from it, I think, because if, actually, no, we've been, so if you think, how many people work for WhatsApp?

Asim Hussain: WhatsApp! I was WhatsApp, I think it 11

Chris Adams: was purchased for $12.6 billion by Facebook a few years back, right? So that's not that far away on a kind of per person basis, but that's not a single person.

But you've got to realize that like, you know, if that was probably a lot better for the people who own shares in that than the people who are working for this. And we have seen multiple cases. We've seen cases like when a company has to choose between keeping on staff to work on something and getting rid of them, and then spending multiples of the staff's wages on buying their own shares to kind of in increase the cost, increase the share price. We've seen the decision that people have been taking, and there is the, I think that this is a thing that needs to be addressed. 'Cause the current, if we're going to assume that, if we're gonna accept that digital is gonna be the, this thing which is just as important to our lives as access to water or you to or energy or anything like that then you probably want to have a discussion about okay well how are the dividends shared well how is the upside shared in an equitable fashion so that we don't end up with people rushing outside with guillotines or in the very least right like it's not good for social cohesion basically so that's my the view that i might actually have on some of this

Asim Hussain: No, that's a, really, I think that's something that people, I've been, I don't know, should I say this?

Chris Adams: I'm going to stop you because it's coming up to 40 we've got one story, so we can, talk about societal accounting, all to carbon accounting, this new paper, which I think is really interesting and we both were nerding out about it before this call. So this is a new paper from Google, Carbon Accounting in the Cloud, a methodology for allocating emissions across data center users.

So, We can totally talk about the societal aspects of cloud computing here. But this one here is really interesting because this looks like one of the most interesting papers about how you apportion responsibility for your use of cloud services when, using these things. And for the longest time, we've had a real struggle because we haven't had access to any of these numbers.

And this paper really. Lays out a bunch of really interesting ideas with lots of really helpful diagrams, and it dies into how inside Google, people allocate carbon emissions for both internal use, but also for cloud customers. It's a really, fun paper, but it is also quite a significant piece of work.

Read. Like, me and Asim were quite excited about it, but we realized this is almost like a kind of book club kind of paper to read through, basically.

Asim Hussain: Yeah. Yeah. Yeah.

Chris Adams: So, Asim, I'll hand over to you because, there's a couple of things I'd like to draw attention to, but I suspect there's some things that caught your eye as well, or at least maybe you might have some context about why the two of us might be so excited about this.

Asim Hussain: Well, I just think, I think it's really exciting. Well, I get really excited when any organization does such a deep, thorough analysis of their emissions. The thing I'm kind of like going through as I go through this is like, part of me is like, I'm going to try and like represent some of this stuff as an impact framework manifest File because I can read an impact framework manifest file and I can compare it and I can look at it and I can know what's going on. Like one of the first things that the paper outlines, the approach that Google developed to quantify location based emissions of its individual products. And now I'm now like, I now need to dive into this paper to understand, well, yes, I'm understanding the, my definition of location based. However, I'm also seeing references to CFE, which to me doesn't factor as

Chris Adams: CFE being carbon-free energy

Asim Hussain: carbon-free energy So like, there's like a lot of nuance to this stuff. and yeah, I I would probably, as an experiment for me to try and understand the paper, I might try and represent some of this stuff as a manifest file. 'cause for me that's quite useful as a way of, learning something in, in a way, in such a way where I can compare and contrast it to other methods and methodologies as well.

But it looks very, interesting, very exciting. And when Google first worked on its carbon dashboard as well. They were the only organization to have done it the slightly different way, which is kind of like bottom up, like from products and services up towards the top, and now IBM has actually done actually to give IBM some credit as well. And they've got great white paper as well. If you're not ready, I'll send it over to you, Chris. they've done another approach is kind of bottom up as well. And so they're the only two organizations that have gone bottom up. The other ones have gone top down and I'm always excited when people go bottom up, because then you get the data with the granularity, the level of products and services that people actually use, which they drives.

Which is what you need to drive emissions reductions. So that's it. I haven't read, I haven't read almost any of this So I'm interested to get Chris's, if you've got any time, I don't know. What are your hot

Chris Adams: So, the thing that really caught my eye from this is that it talks about some of the ideas that Google has been doing that other organizations aren't so open about. So, for example, you have like a given amount of capacity that might be available inside this. Now, what Google have done previously is they've kind of said, well, We know that we've got a certain amount of, kind of, energy that's coming from green sources and we've said that we want to have a percentage of our energy that is always running at, say, 100 percent carbon free, for example, right, and they, Google have an approach where they say we count something as carbon free if it's matched at the time of use and if it's I think it's on the same grid.

So it's not literally a solar power, but a set of panels on the data center. It might be a wind farm that's where you could plausibly deliver the power to that place. And they use this to kind of represent the amount of clean capacity as something which they call a virtual capacity curve, because it changes over the time of day, basically.

So they talk about in this paper, there's some production, there's some production loads, like that always have to run and always have to respond very, quickly. And there's things where they've got a Bit of freedom in how they move it around and I think this is actually quite interesting because they talk about where they have some flexibility inside this and they talk about how they account for a bunch of that because it's the first time I've seen a paper, A, talk about this, but also talk about the fact that there's like, a set amount of kind of idle power then there's amount of power that will kind of ramp up and down based on the amount of use you're introducing they do they speak about a bunch of really interesting things inside this and the thing that i think there's like there's a couple of figures which i really find like really quite fascinating actually and the fact that there's like one thing like If you are at all interested in, like, Sankey diagrams, they've got this really cool Sankey diagram of saying, well, this is all the kind of power that goes into running machines, running the overhead.

This is how it gets proportioned across all the different services. And this is how these end up being allocated to both our internal use, but also cloud customers and stuff like that. It's a really, fun read. And I'm probably going to spend like, I think, an afternoon or maybe the weekend making some more notes on this.

'cause there is a bunch of stuff which is beyond my can, like some of the equations are. I, don't have the, I don't have the ability to kind of make sense of those. I am looking forward to reading this nonetheless, because it's really nice to see something like this, not least because by putting this into the public domain, it's now raised the bar for some of the other providers to be more transparent about this.

Because if you're looking at, say, Amazon, you're looking at Amazon's calculator, you don't have scope 3 emissions, so that could be up to 99 percent of your emissions not accounted for in the numbers. So if they look suspiciously good on the Amazon dashboard, maybe they are suspiciously good. But also you look at the resolution.

This is something where they providing information at both the location and a product line value. So let's say I'm using Cloud Run or one particular kind of storage. I can see it at that kind of resolution. And that's that kind of location. In some other providers, you might see Europe and then compute.

So, there is nowhere near that kind of resolution. So, people talking about this is how we do it. This is how it's possible. This is what you should be expecting from other providers. I think it's really, good. And they also do mention the fact that they're using high time resolution. So, they say, "We're using data from electricity maps to help us work out these hourly curves, so that we know at what times of day, what the kind of carbon intensity for the power might be, so that we know that we've got this much kind of green compute that we can plausibly use," and in a defensible and transparent way, say, "yes, this really is running on renewable energy, according to the way that we talk about this."

And like, they do refer to like, they, you don't need to agree with the approach that they use in order to at least understand where they're coming from, because there's plus points and minus points with using the approach here versus having something totally location based like we've spoke about before.

So, yeah, that's kind of my initial thoughts when I see this, actually.

Asim Hussain: We need to get this to the point where it's kind of easily absorbable and understandable by folks in such a way that they can actually, because you know, otherwise it's just, oh, right. There's a great paper we'll use, you know, but I think you need to understand the nuance of a lot of this stuff.

I don't

Chris Adams: Well, there's two things, but what I think, so... you know, there's a, the real time cloud working group that Adrian Corkcroft is

Asim Hussain: Oh yeah. Yeah.

Chris Adams: He's been really pushing on a bunch of this stuff. This feels like one of the abs, absolutely worth sharing to that group because they've been doing a really good job of actually collating this data so it can be used.

And like, a scene like that was the thing that kind of fed into the impact framework stuff, right? So there is a path for this. And like, There's a job to decode some of this and make it easier for lay people because, yeah, like, Asim, we've been talking about this for literally five, six, five years to get to, for us to understand why it was exciting.

But yeah, you do need a job to actually make this easier for people new to the field because there's lots of developers who are kind of coming into this kind of sustainable software field.

Asim Hussain: you know, we should also have and this, we're now talking about GSF work. I know I've actually got to drop like very soon, but this is interesting because as I just asked you that question, I think an answer came to my head as well, which is like standardization. So, you know, as we like to, even as a real time cloud project is kind of evolving, like one of the things it's trying to figure out and work through is like, like you mentioned CFE and your definition of CFE.

And I can tell you right now, I've heard different definitions of CFE from other organizations, which don't. Count the, it to be, it doesn't have to be the grid. It could be anywhere. So I think some of this stuff, it might be interesting to have conversation with Google, like other ways to standardize some of the terminology, the methodology, the equations to this, as soon as you can create a standard, maybe something we can push into ISO or something like that. That kind of also in a way also forces the, not only simplifies everything for everybody, cause they're like, "well I don't really understand what the standard is, but I can see that it's got wide adoption and it's a standard and competitors have got together and agreed on this standard." So I don't really, you know, and there seems to be a wide body of people who support it. I don't need to look at this. This equation is so thick. I'm staring at this equ

Woo!

Chris Adams: it's quite a man,

Asim Hussain: It's a quite, a yeah, it's quite a lot. Selecting it has like 43 components in the selection.

But...

Chris Adams: the thing that we can maybe talk about is that there are standards. We don't need to be doing all this work ourselves. Like energy TAG is one standard that is essentially written into European law and American laws around hydrogen now, like hydrogen production. So there's things that, you know, we wouldn't be starting from zero.

We could be using some of that stuff, but you're right.

Asim Hussain: We how we push

Chris Adams: Yeah. I need to also. Thank you. Trying to rush us through for the last few bits, because there's a few things inside this that, as hosts, I need to be doing, and I sadly can't talk about the marginal carbon intensity of baked bread, because we had a really lovely follow from Dr Daniel Schien, who responded about the low carbon bread and high carbon bread thing. So maybe we'll, actually, we should commit right now to do an article about the, about how green the energy is, and how you talk about that, because this is what we just spoke about. Right now, Dr schien raised a number of really good points and linked it and shared some really helpful resources with us.

Okay.

Asim Hussain: we should get a proper conversation together with, if we can, like with maybe EM and some

Chris Adams: Yeah, we have some people inside the organization who've been doing that. Alright, let's look at events. So, stuff that's coming up. HotCarbon is a workshop on sustainable computer systems. This is happening on July the 9th. It's free to attend virtually. You can turn up in person if you're in California.

It's really, good. And I, there's, they have videos online and really fascinating papers. It's really worth reading. It's like absolute cutting edge stuff. There's also a, the IEEE. They have a cloud conference on the 7th through to the 13th. This is in Shenzhen, China. And for the first time I've seen, there appears to be something like a sustainable AI track, or sustainable computing. You are? Oh,

Asim Hussain: to be there. Yeah. I'm going to be there. Yeah. Yeah. I'm going to be there. I was invited by, well, the, anyway. So yeah, I'm going to be, I'm going to be, I'm going to be over there. I'm going to talk about sustainable AI in the cloud. There's going to be a whole track, several panels, discussion topics.

And

Chris Adams: Oh, wow! Cool!

Asim Hussain: You know, I don't think we particularly speak too much to, I'm a big believer in that this is a global challenge and a global issue. And yeah, most of our conversations happen in the Western world. So one of the things I'm personally trying to do is to try and

Chris Adams: Bring the other 1,5 1 point something billion people into that. Okay, great. Okay, so other two things we had. There is a, so outside of programming, this was shared with us. The masterclass on becoming a sustainable UX designer. This is one thing that's been led, I believe, by Thorsten. Oh, Thorsten, I'm afraid I forgot your name.

Thorsten Jonas. He's been helping organizing Sustainable UX as a community, which has the unfortunate short name of SUX, S U X, but it's, this is a virtual masterclass you can join. We'll share a link to that inside the show notes that's taking place, and he has a number of these coming up. And there's also Asim, I think this is yours, right?

The Software Measurement Landscape Workshop on the 9th of July? That's something that you've got, you're looking at?

Asim Hussain: That's not, well, no, not, it's yeah, it's not something I'm particularly involved with myself, but the green, the, we have a meetup group in Brighton in the UK, organized by an organization called Root & Branch, which do a lot of green software, the kind of green software startup, and they're doing a workshop on software measurement.

Chris Adams: Folks, ah, so Cardamon were the people behind, sorry, Root & Branch were behind the Rust based software tool called Cardamon? I think Cardamon, it's monitor for some of this carbon related things. Okay, that makes sense now, putting two and two together. Okay. And we have one last wrap up before I need to go, actually.

So, this is the, this week in Green Tech, where it's basically you and me, and occasional guests talking about the news. We have a, there's a new podcast, or a new podcast related to the Green Software stuff called CXO Bytes, and this is, I believe, Sanjay Poddar, the chair of the Green Software Foundation.

He'll be talking a little bit about some of this and speaking at the, I guess, CIO level, something like that. That's what we have. So, there's a first episode coming out in July, for this and, yeah, I figure we should tell people about that because it's something that's coming up too. Alright.

Asim Hussain: That will be, that's a separate, just everybody knows that's a separate podcast, so you'll have to find it and put it in the show, no late. We have to find it and subscribe to it separately. And as Chris mentions, like Sanjay, I'll be speaking more to the c-suite level of an organization and like one of his personal beliefs, and I do subscribe to this as well, if you want change in an organization, it needs to come from both directions or it will just stall in the middle. So he's very keen

Chris Adams: Getting by on it at the management level, at the very top, yeah? The big cheeses,

Asim Hussain: do is big cheeses and we do the witherbread whatever. That's

Chris Adams: Or the mushrooms, if you prefer, the

Asim Hussain: right. There you go. Wither, wither, yeah.

Chris Adams: breaking down barriers to adoption, like you break down dying logs, right? Okay. But Asim, really, enjoyed this, mate, I'd really love to see you again, and get well soon. I hope the Nurofen and coffee thing doesn't have to be a daily thing, and that you get better over the weekend, alright?

Take care of yourself, mate.

Asim Hussain: Wonderful. You too.

Chris Adams: Okay, ta ra!

Asim Hussain: Catch you there.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

The Week in Green Software: FinOps, GreenOps and the Cloud

20 juni 2024 | 45 min

On this episode of TWiGS, host Anne Currie is joined by Navveen Balani of Accenture and fellow GSF member. This conversation navigates the landscapes of, and intersections between GreenOps, DevOps, and FinOps, as well as the vital role of Infrastructure as Code in marrying financial and ecological efficiencies in cloud operations. Lastly, they tackle the intersection of cybersecurity and AI development, emphasizing the need for green software principles to fortify AI systems while minimizing energy use.

Learn more about our people:

Anne Currie: LinkedIn | GitHub | Website
Navveen Balani: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

Cloud Native Computing Foundation [14:36]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Navveen Balani: Definitely, I would say there is some synergy between security and green software and certain, I would say, features of green software principles can also be applied to security domain, right, to make it more energy-efficient.

I'm your host, Chris Adams.

Anne Currie: Welcome to another edition of the Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. Today, I'm your host, Anne Currie. So you're not quite hearing the usual dulcet tones of Chris Adams. You will have to do with me instead. But as usual, we'll be talking about the world of green software and what's, what's going on at the moment.

And today we're going to talk a little bit about how being green matches with FinOps, which I think is very true. And I think that's a really important part of the story. So we'll be talking a little bit about that. We'll also be talking a little bit about code and code efficiency, which I'm much more...

You have to be very careful about code efficiency. So that's, that's the part that we need to be really careful about. What's the context there when we talk about code efficiency. And finally, we'll talk about the intersection of cybersecurity and AI developments, and we'll be talking a little bit about GreenOps.

And there is a new Green Software Foundation project, GreenOps project, which is aimed to look about how we can embrace AI and use AI without totally throwing all our green principles out of the window. And I think that is absolutely doable, but we need to think about it. We need to go in, in a very knowing way.

So as I said, I am your host today, Anne Currie. But first I'm going to introduce our guest, Navveen. Navveen, do you want to, do you want to introduce yourself?

Navveen Balani: Thank you Anne. Hi everyone. I'm Navveen Balani. I'm the Managing Director and Chief Technologist for the Techno and Sustainability Innovation Group at Accenture, working at the intersection of technology and sustainability. I'm also the co chair of the Standards Working Group and the Impact Engine Framework at the Green Software Foundation.

I'm a Google Cloud Certified Fellow, a LinkedIn Top Voice, and author of several reading books. Very glad to be part of this podcast.

Anne Currie: Thank you very much, Navveen. It's very good to have you. So just a bit of context for me. My name is Anne Currie. I am one of the co-chairs of the community group of the Green Software Foundation. And I am also one of the authors of the new O'Reilly book, Building Green Software. So that's, fills you in a little bit on my background.

So before we dive in to the articles this week that we're going to be talking about, it's just a reminder that everything we talk about will be linked in the show notes at the bottom of the episode. So you can, you can read the articles that we're talking about. You don't just have to rely on us telling you what was in the article.

So, well, let's, let's move first to the first article from today, which was in Computer Weekly. And it was called Green Coding. It was, it was a, basically a puff piece by a company called CloudBolt who do a look at code efficiency, cloud frequency. So taking it beyond FinOps, beyond dollars, pounds, and pennies.

So actually, but it was a very good article, I thought. I was very pleased to see it in Computer Weekly. So it was fundamentally, it was about how GreenOps and FinOps are very aligned. They're very combined. And, and I can, I'm in complete agreement on that. It's a good article. It doesn't say, tell you anything that you probably, that you won't know alreby knowingwing that FinOps and GreenOps are quite aligned and they're all aligned through the fact that in the end, being green, a lot of being green, not all of being green, but a lot of being green is about cutting down on how many machines you, and how much electricity you are using to run your systems, which, which generally speaking cuts down on the cost.

So cost is somewhat of a proxy measure. It's not a perfect proxy measure, but it's somewhat of a proxy measure. So the question that Chris Skipper, our excellent editor has left me and Navveen to discuss is about considering the role of infrastructure as code in enhancing cloud efficiency. How can developers ensure that their infrastructure as code implementations are aligned with sustainability practices to reduce both costs and environmental impact?

So Navveen. What are your thoughts on that subject?

There's,

Navveen Balani: I think that's a great question. Yeah, agreed. Developers need to embed sustainability as part of the infrastructure as co implementation. And the frameworks that I suggest developers apply is based on our Software Carbon Intensity specification from our Green Software Foundation which also recently received the ISO standard recognition.

So, for those who does not know what SCI is, SCI is a specification to measure the carbon emission of any software application and it promotes three key levers. Writing energy-efficient code, using less hardware for same amount of work, and making applications carbon-aware. And if you apply this strategy to infrastructure and the code, first you start with writing energy-efficient code.

So developers can focus on optimizing resource utilization by right sizing resources and implementing auto scaling. This means allocating only what's necessary for each workload, and also adjusting dynamically based on the demand. Second strategy is around using less hardware for same amount of work.

This basically involves automating resource management like automating the shutdown of non-essential resources during off hours and starting them during peak times. Even conserve energy and cut cost. Also, tagging and monitoring resources usage helps identifying optimization opportunities and eliminates waste.

You can also go with serverless architectures in your ISE code, particularly, it's quite effective as they scale with demand and eliminate, let's say, any provisional requirements. And finally, the third strategy is how do you make applications more carbon-aware. And that's where, as part of your infrastructure code, you can say that I want to deploy a particular workload in a clean free region so you can basically take, apply strategies like region shifting and time shifting and selecting cloud regions which are running on renewables and also maybe deploying workloads or scheduling jobs when the carbon intensity is low.

So all of these strategies can be definitely applied and designed as part of your infrastructure code.

Anne Currie: That's a, that's a very thorough answer and there's loads to unpick in there. Lots of different things. And some things I think at the moment are very aligned with FinOps and cutting your costs. And some things are not aligned yet, but are almost certainly going to become aligned in the future. So for example, you talked about operational efficiency and automation, which is interesting.

Operational efficiency, if you use for your machines. And if you use less electricity, your bill goes down. So that's all good. So in that respect, your FinOps and GreenOps are really well aligned. You know, fewer machines, less stuff, less carbon goes into the atmosphere and it's all fantastically good. And in that respect, I would say that That, that FinOps, that, that, that, you know, your, your hosting bill is a really good proxy metric for your carbon emission.

But of course it's, it's, it's almost stupid, it's so obvious to say it, but it can't just use your cloud bill totally blindly as a, as a guide. Just, you know, what I, when I used to do start, startups in my youth, quite often Azure or, AWS would give you loads of free credits. And then of course, but that doesn't mean that it's carbon free.

So, so there are times when, you know, you just, but you just need to use your head, don't you? Sometimes, obviously, you've been given a discount, but it's not green. It's just a discount.

Navveen Balani: Yeah, yeah, I think that that's, that's a good point. Because if you look at cloud, right? Cloud has infinite resources, right? So, it doesn't mean that, so you have to use it responsibly, right? So, you can bake in energy efficiency and sustainability, right? So, definitely, you have to look at how can you lower the carbon emissions.

And now, and there are also dashboards available from cloud vendors, right? Which at least gives you, some approximation, right? How, what, what is your carbon footprint of your application?

Anne Currie: Yeah, yeah, yeah. I mean, yes, they do provide really good tools and, and, and it's, it's a cut cost off really is the best possible metric you could potentially use, but it's where you've got the tools, it's where you've got the data. So, you know, it's, it's quite good from that perspective. Sometimes you just have to take what's good enough.

And, but so something else you mentioned is automation and obviously the, the really good operations is all automated these days. It's auto scaling and it's using the you know, not just in the cloud, but on prem as well, but actually I, in the. You can just do a lot of stuff manually, you don't have to leap straight, if you, if automation is too scary and you know, it's too much of a leap, just going through and turning off machines at the weekends, even manually, identifying machines that are over, that are over provisioned can actually, bizarrely, I think, I suspect in, well, and in fact, I've seen It might be the biggest carbon reduction you ever do, it's the simplest thing and the least techie thing.

So what do you think about, even for automation?

Navveen Balani: Yeah, totally agree. I would say even just turning off machines. Yeah. I mean, that's just manually, right? Would definitely save also the cost also, as well as the carbon emission also, right? Man, especially if you've turned on GPUs, that would affect. So, yeah, But actually, I think, also, I think if you look at the infrastructure, right, it's, it's, if you, if you break down the infrastructure in two parts, right, production and non-production environment, you can definitely have a lot of savings on non-production environment, right, because it doesn't need to be on always.

Production definitely needs to be on 24 by 7, but you are definitely have a lot of improvements that you can done on your non-production environment, dev environments, people, right, and I've seen customers having more non-production environments, right, than production ones.

Anne Currie: Indeed. And it's so ungreen. It's there's, there was another interesting thing that you talked about in your when you were talking about the SCI, which is carbon-awareness, which of course we know is the most, it's actually the most, it That's the code efficiency and operational efficiency are all good for kind of short term mitigating the harms, but in, but to actually take full advantage of that, you know, the soaring production of cheap energy for renewables, we need to demand shift to when the sun is shining or the wind's blowing.

But the interesting thing about that is that although that's by far the most interesting thing in being green, I would say, it's the bit that at the moment doesn't really save you any money because most countries don't yet have dynamic pricing. So dynamic pricing, so, so what's your thoughts? What are your thoughts on dynamic pricing and when it's coming?

So dynamic pricing is basically when the price of electricity changes through the day, depending on how expensive it was to produce, which usually means, you know, at times when the sun's shining and the wind's blowing, the power is cheaper than others. And that's, that's now very common in certain countries like Spain, but very uncommon in other countries.

Navveen Balani: I think definitely that's a good concept that will promote more sustainability. Typically, if you look at cloud providers, like let's be, to give you an example of Google cloud, it at least gives you now, if you're deploying something on a region, it tells you that it's, it's a low carbon region and hopefully in future you will have the dynamic parsing also, right? It will also give you the time when you should run the workloads. And there are a lot of workloads we doesn't need to run 24 by 7, like batch jobs. We get a lot of emails, right? Of all those millions of emails, right? You'll be sending, right? For promotions. All those can be run on time where, where there is least intensity.

And definitely if you have, if you have a cost, if the cloud provider gives you a cost that this is a good time window, and this is less costly and then all, all, all the activity which doesn't are not critical enough can definitely take care of the dynamic pricing. So I assume in future, I mean, we can see the trend, right?

Maybe where you have, I mean, it's all about data. If you have the data from the grids available to the providers in future, then we can definitely definitely tap on it.

Anne Currie: Yeah. Yeah. I agree. It's, it's a bit of a shame. It's, it's where there's the hole in the, in the alignment of green ops and fin ops at the moment. That's, that is incredibly green to demand shift, but you don't necessarily get money off for doing that. You know, moving to that green region is incredibly green.

But it doesn't necessarily save you money, but it will do once dynamic pricing comes in.

Navveen Balani: I also, I think if I also feel if regulations also are there, right, with regulations around carbon emissions reporting. Especially, I think, the EUA Act just talked about reporting the carbon emission, but it's a dog bone mitigation. But at some point of time, I would say, when you have a reporting mechanism also, and everybody have to comply for it, I see a lot of these trends coming, right, a new innovative way, right, to lower the carbon emission, right?

So I think regulations at some point of time will also enable, right, a lot of these, I say, innovations, right, to come up.

Anne Currie: Yeah, I agree. Oh, I meant something I meant to mention that, that's aligned with what you were saying earlier about automation. There's the CNCF, the Cloud Native Computing Foundation. So another of our, of the Linux foundations out there. They describe, and it's, GreenOps equals Fin, GreenOps equals FinOps plus GitOps.

So basically they're saying GreenOps is automated FinOps, which is an interesting one because it feels to me like they're really saying there that GreenOps is good FinOps. Uh, and oddly enough, FinOps often say, well, FinOps is just good ops. GreenOps is just good ops. It's, it's which is interesting, which I think that people often don't really appreciate. No, sorry, I'm taking the final word there and I need to, I will leave the final word to you, Navveen, on FinOps and GreenOps.

Navveen Balani: I'd like to end with what you said, right? So it's, we started with DevOps where you decide to automate something, then you, then the FinOps came, right? So we, because cloud resources were getting expensive, right? And now we have GreenOps, right? So all, we have to look at it in holistically, right? Across DevOps, FinOps, GreenOps, right?

And ensure you take care of both the cost and carbon, right? And keep it under control.

Anne Currie: Yes, totally agree. Totally agree. Right. So we're going to move on to the next. So that's the first article we're talking about there, I would say is extremely uncontroversial. Operational efficiency is like a total win all around. The next article, which is about why you should switch to green coding for a net zero feature, which is a LinkedIn article from the CEO of CSM Technologies, I think is vastly more controversial. Not because it's wrong that you should write a more code efficient, more efficient code. But because I think that there's a lot of context around it. So, so he's written a lovely article, links, as I say, links in the show notes, saying we should all be, be coding more efficiently, which is, which is nice.

But, and it ends with the, with the line, we should change, change, save the world one line of code at a time, which I find massively controversial because I think that when it comes to code efficiency, a lot of business, it's just not the right thing for a lot of businesses to do. It's too expensive. What they should be doing is putting pressure on suppliers to write their code efficient, you know, write the scaled code efficiency.

I think that it can really waste time going down, people going down that rabbit hole and their, their bosses were very right to say, "don't, I'm not going to do it" because it would put you out of business if you rewrote all your systems in Rust or C. So I think that is, it's, it's, it's an article that's true, but only true in certain contexts and not in others.

So Navveen, what's your, what's your thinking on it?

Navveen Balani: So I would say we need to look at this holistically, particularly around green software, focusing on, let's say... I would say three dimensions we have to take. First is having developer training and implementation of green software. Second, we need management buy-in. And third, I would say, it's the culture shift that needs to happen.

And if you look at green software, right, when we when we started, when you all started with the foundation, three years back, green software was a very relatively new area. So we need to provide training and certification in this area so that developers are aware of, right, how they embed sustainability in their day to day work.

Apart from, I would say, the training, a developer needs to have accessible tools, right. Now we have the SCI specification, the Impact Framework, Carbon-Aware SDK, right, which, and there are a lot of other open source tools also available now, which can make it more actionable and developers can actually embed them as part of their DevOps process.

And once the measurement is done apart from code, right, it's also the optimization piece we talked about, right in the infrastructure as code earlier also, right, how you take it all together, right, and try to optimize not just the code, right, but also the resources which are running the hardware, the resources which are powering those applications.

And as I would say green software practices gain traction, I would say securing management buy-in also is essential, right, for widespread production. For instance, highlighting the business benefits is crucial. For instance, implementing green coding, right, can lead to also significant cost savings by, let's say, reducing energy consumption and optimizing utilization.

And as you mentioned, right, it's not just Our footprint, but the scope, the footprint of our suppliers to ensure they also follow the same standard methodology. And that's where I think it comes to the culture change, right? That we all need to go through, particularly for, for green software. And we need to look at how we can embed green software, right?

Going forward in all our work, right? Similar to, let's say, similar, we do it for security, right? So when we, when we had security at 10 years back, right? Now we have security by default, right? We don't talk about that application needs to be secure. We assume application is secure by default. Similarly, if we embed green software, not just code, but across all, all the layers, then we can ensure maybe over the next four, five years where all applications, the new applications that we build, right?

Have a green software principles baked in. So I would say, I mean, it's basically a holistic approach that would be required, right, to, right, from enabling the development community, the tech community where the foundation also, like, for foundation, like, Green Software Foundation plays a critical role.

The management needs to buy in, and also the culture change that needs to happen, right? And the culture change also needs to happen, I would say, at the universities and schools, right, where they can start educating green software early on, right? Similar to the way we have learned object oriented programming, right?

That's by default. We have learned over the, I mean, over the last I would say decade, few decades, right? If we have green software, same as object oriented programming concept, then I think whatever application we build in future, right, we'll have green software baked in.

Anne Currie: Yeah, I, I may agree up to a point. I agree with you, but I think we, I, I'm a big believer in separating out two types of developers. Obviously there are loads of different types of developers in the world, but so, but two types of backend developers, so front end developers, this is almost a separate thing.

But backend developers, you've got people who just working in an enterprise and the code that they're producing is not going to ever be deployed. You know, it's, it's the code that they are writing is never going to be run by billions of people in their own data centers. And then you've got people who are writing platforms and the whole purpose of the platform is to try and get some billions of people to, or at least millions of people to write this code, to run this code.

And those people, they absolutely need to write efficient code. When I think everybody needs to, should be, could, should get used to getting out their performance profiler and just making sure there are no egregious performance problems with their code. Because performance problems are your code's slower and you're burning a load of carbon and it's total waste.

So all again, very aligned with the business. You want your systems to run fast. Your customers want your systems to run fast. So, so having a decently performance system is good. But beyond that, you probably don't want to be writing code yourself, which is massively efficient because that takes a long time, but you do want the platforms you're running on to have done, to have made that investment.

So it's, it's kind of like, there's a lot of context here, isn't it? Are you writing code for mass use or are you writing code that is not really for mass use, which is, which is interesting. I think that's a subtlety that we, like, for example, lovely article though this was, did not point out that difference.

Navveen Balani: That's a good point. So, yeah, especially applications, right? Package applications which will be used, let's say, by millions of developers worldwide or users worldwide definitely needs to make that in right to ensure that for instance simply like all the large language models a good example right all all generative applications will be used by millions of applications millions of developers so how can you make the AI more efficient right both on the user side who is creating let's say the prompts right to Create in an efficient way so that the round trip is reduced and secondly on the backend side, right?

How do you have a low cost efficient energy-efficient model? That's why you also seeing a lot of LLM models are now Talking about the small language models more energy-efficient more compact, right? There is a trend where I would say right where Organizations are now looking at energy efficiency also, right, as part of the applications or whatever work they have been doing.

Anne Currie: Yeah. And of course that's all driven by cost. It's taking us back to our previous thing about cost and cost and green are very aligned. And the good thing about, well, the bad thing about AI is very costly. The good thing is it's driving quite a lot of efficiency improvements. So, I mean, I talk about this every time I'm on here, that, that Python has got a lot more efficient because, because of AI, that they've rewritten all of the code, core, core libraries in Rust. And of course that's, that's a perfect example of they are the kind of people you want to be writing super efficient code. They can save the world one line of code at a time because so many people run Python. But you want to be getting that out of your platform and not having to do it yourself as a, as a Python user.

You don't want to have to change to Rust yourself. You want to be able to get the value of Rust whilst still using Python. But yes, yeah. So, so that is all very interesting stuff, but yeah, very nuanced. It's all at every degree of it. It's what my, to my mind is what makes green interesting, is it's not simple.

It's not trivial. You have to step back and you go, "where am I? What am I doing? You know, what's, where, what's, how do I fit into this? Where does mine, where is my effort best applied?" I mean, you're obviously part of the SCI, which covers all of the things, you know, operational efficiency and code efficiency and demand shifting and shaping.

What's your interest? What do you like the most out of those things?

Navveen Balani: I would say from a, I think from a developer standpoint, right, depending on your roles, right, so SCI, I would say is more inclusive in terms of, depending on roles, right, whether you are a developer, architect, data scientist, right. All are various parts to play, right, to reduce the carbon emission and make applications more energy-efficient.

So, depending on your role, for instance, if you are building, or you are a developer writing code, right, then you can really focus on energy efficiency. And it's not, as you mentioned, right, it's not just moving towards a C or C++ language, right, which is more efficient, but it's, so you have to basically look in the context of the work you are doing and trying to optimize it, right.

So you have to do that trade off as a developer and how, what libraries access to make it more efficient. Second, I would say the whole hardware optimization, I think in terms of where all the DevOps, cloud, cloud architects comes in. How there are various custom chipset from various vendors, right? How can you best utilize from an infrastructure point of view?

And third, I would say is more strategic in nature, right, in terms of how do you bake in the whole carbon away computing concept, because that's new. You need to have data providers, you need to tie up with various licenses which are actually costlier, right, if you look at getting the real time data, right, from various providers.

So how do you bake that in in the application to more of a strategy kind of work and thinking? So in that way, I would say it's, I mean, depending on your role and context, right? I mean, whether from developers to architect for data scientists, right? Each can find definitely a value for, in SCI and then try to reduce their scope of work.

Anne Currie: Yeah. It's so, it's interesting. When I first heard about the SCI, I was a bit, I was a bit dubious about its value, but I have completely changed my mind on that as time goes, especially because I teach, I teach people who are green software and, and, and. One of the things that often comes up is people wanting to be able to do like for like measurements.

And I think that I originally thought the SCI was about a standard that you could compare between applications. And that was where the value would lie. And I was a bit dubious that we could realistically do it. And, and now I've realized that, that I like that the SCI has stayed fairly woolly and loose.

It's more conceptual than it is a specific implementation. And I like that because really it means that companies could choose their SCI score, they can choose how to define their SCI score for their applications and choose what's appropriate. What it's going to, what it's going to, what the denominator is going to be.

So it's per user or per transaction or per, so everything, something that's specific to them, and then it's essentially for like, for like, so you can say, well, last year it was this, and the next year is this, and you can average over time so that you're, you, you know, saying you don't say, well, I'm comparing a sunny day with a non sunny day, or, you know, all those kinds of things.

What I like about the SCI is it's, is it's very kind of conceptual high level nature that forces people to think, "well, actually, how do these things apply to my system?" You've got to use your head. You can't just, you know, you can't just follow it. You can't follow it blindly because it doesn't make any sense if you do that.

You have to say, how does this apply? Forces you to think, which I like.

Navveen Balani: Yeah, I think that's a good concept. Particularly if you look at SCI, right? It's for an application, so you have better control rather than giving you the carbon emission of all the applications. Right. Which typically is given by various cloud providers given an application, then you have a better control, as you mentioned, right?

You can define your own boundary and architecture and calculate the SCS core, right? And the intent is to. Basically, as you deploy new versions, right, the intent is the SCI, you should look at how you can reduce the SCI score, right? We can't achieve zero, but definitely across releases, right? How can you make it have a lower SCI score?

And the point you made about the comparison also, right? So you're comparing your application versus your previous application that you have deployed. It's not about creating two applications from two different organizations, right? We're not there yet. It's about currently using this methodology, right, for your own application and trying ways to reduce it.

Anne Currie: Yeah, which makes all the sense to me. So many years ago in my, when I was more youthful, I used to work in retail and in retail, like for like comparisons are very important. You want to be able to say, well, this year we've made more money than last year, but you can't just say this year we made more money than last year.

You can, but it's not all that useful. What they actually want to do is say, it's per thing. So in retail it's often the kind of per square foot of retail space, balanced for kind of like, well, how expensive was that retail space? You know, so you're not comparing it and say, well, this year we made more money on the same on, on the same amount of floor space, but, you know, it was in London versus it was in the middle of, you know, of the desert. It's, it's kind of like you, you've got to, you've gotta come up with your own, like, for like measure so that you can say, well, is our business improving or is our business getting worse? And the SCI is exactly the same. It's, it's the concept of like, for, like, it's for you to check Ron North's for you to check against other people.

So yeah, it's yeah, I, I, I, I've been completely won over to the SCI. I was highly dubious to start with. Right.

Navveen Balani: Good example. Yeah, that's a very good example of a retail. I'll also use that.

Anne Currie: Right. So now we'll go to the final, the final thing we're going to talk about today, which is, we just touched it from a slightly, again, a slightly more holistic perspective. So this was an article in a Silicon Republic and it was a Q&A with the chief security officer of a, an AI company. It was talking about.

Basically, her premise, and I totally agree with it, is that security, cyber security is very aligned with being green. It wasn't, it's not, it's a bit thin as an article, it doesn't give you an awful lot of information, but I think basically, yeah, that there's the, the idea that, that I think we should be discussing is, is security and, and green aligned?

And you, you, Navveen, you've talked a little bit about that, about in terms building security in is, is like, it is, we've, we've learned to do that and we should learn to do green things, build green things in, in, in the same way. But separately to that, are there security benefits to being green? What do you think?

Navveen Balani: I would say, yeah, definitely there's synergies between security and green principles. So I like to again give that example of SCI, right, if you want to break down the methodology into three parts, right, making applications more energy-efficient, right. So, if you look at the security algorithms, right, how can you use, how can you basically optimize the security algorithms to, let's say, use fewer computation resources.

Particularly, if you look at the security stack, right, they also have evolved from various encryption and cryptography software, right, and now I think you have various key ciphers available across different dimensions. So, they're already following this, I would say, backtest, right, of making encryption security, right, more performance and more easier to adapt. So in that case, I would say it's more aligned towards the algorithms that they're using are more efficient, right? As compared to what it might be, let's say five years, 10 years down the line for, for security protocols. And similarly, I would also say new strategies can also be applied for security scanning.

For instance, vulnerability scanning is one commonly used, right, to identify any threats, maybe in cloud or maybe in desktops and other systems, right, that can actually run, take the advantage of running it on a time, right, where the carbon intensity is low. So in that way, it can apply certain green software principles.

To run all those scans where the carbon intensity is low and also save on the carbon emissions. So, definitely I would say there is some synergy between security and green software. And certain, I would say, features of green software principles can also be applied to secure the domain, right, to make it more energy-efficient.

Yeah,

Anne Currie: And so, so something that actually I'm, I, well, you did mention a little bit earlier is, is security is the perfect example of where hardware, using the right hardware for the job massively cuts your emissions. So dedicate the right kind of chips that are designed for for encryption are just so much more efficient than using general purpose or CPUs for, for that, for that.

So, yeah, we wouldn't be able to do what we do these days if it wasn't for dedicated chips. Oh, and so, and oddly enough, this does also map to some of the stuff that we said at the beginning about manual ops and manual FinOps, that's, it's amazing how many systems are through machines that are kind of like, nobody's, everyone's kind of forgotten about them.

They're not keeping them patched. They don't really do anything useful anymore. Those are your backdoor. Those are the ways that people break in and they're just wasting electricity. So even in building green software one of my co-authors, she, she brought up the fact that, that it's interesting that security, well, that are very much an example of a waste of electricity.

They, they're, they're wasting. So, so something like a denial of service attack, the whole purpose is to burn your electricity in your systems and burn your system so that your systems don't have anything, any time to do the thing that they're designed to do. You know, the thing that has value for you, instead, it's just burning your systems up burning electricity, running up all your bills to do something which is bad for you.

So having a secure system that, and things like applying the latest patches so that you're less, less exposed to denial of service attacks is green because denial of service attacks are very ungreen, they are very dirty. Same, it's, it is quite interesting, isn't it, from that?

Navveen Balani: Yeah, that's why I think the provisioning the right hardware, virtual machines, especially, right, provide various cloud providers. Right. They all, they now provide a managed services, right, to detect various denial attacks. And I assume, right, the underlying hardware that they are using, which is, which should be used by millions of applications, right,

would be definitely more sustainable, more energy-efficient, right. And, and more scalable.

Anne Currie: Yeah. Yeah. Securities are really interesting in that, that's it. So FinOps is just really a, a, a pretty much, except with, for, you know, your free Azure credits or whatever, is pretty much a group, the direct proxy measurement for carbon emissions. And, and likely to become more so in the future when we get dynamic pricing.

But security is not a direct proxy measurement is just, it's just that there are a lot of, you know, best practice in ops is also best practice in secure ops is also best practice in green ops. You know, they're, they're, they're kind of like, you can't use the number of hacks you have, then the number of attacks you, you fall foul to as a proxy measure, you, well, maybe you could, I think that would be a bit complicated.

Proxy measures for carbon is how many times your data gets stolen, aligned rather than proxy. It's interesting. I mean, so we just, we've talked about those, those three things, but have you run across anything interesting at the moment that's, that you think that our listeners should, should hear about? And

Navveen Balani: So, yeah, I would say for the, on the Green Software Foundations, specifically, right, we are working on green AI and so we are trying to look at how we can extend SCI to take I mean, how we can do SCI measurements for large language models generative AI models. So this is something we are actively, I would say, working towards from the foundation perspective.

And if you look at from an SCI perspective, right, we want to have various extensions to SCI, for instance. How do you do SCI for web applications, backend applications? And make it more easy to measure, right, different parts of the code and make it easily available to developers so that developers can measure their, their part of the overall carbon footprint, right so we can make it more accessible.

So that's, that's one thing I think we had from a foundation perspective, looking at how we can make it SCMO extensible to various other use cases.

Anne Currie: That's, that's very interesting. That is the, yeah. 'cause obviously Ai, AI is the, is the workload of the, it's on everybody's lips at the moment. It is... and so I, I saw very interesting charts from all the, the Economist, I think the, the other week that, that said, you know, that's, that showed the enormous amounts of power that was currently being used on ai, but still less than the amounts of power being used on Bitcoin.

So it is just worth reminding. And of course, bitcoin is very aligned with our last conversation about security and what people wanting and people who are attacking you wanting to run up your energy bills, because quite often what they want to do is mine Bitcoin on your, on your machines that you're not properly watching.

So security sweeps are a pretty good way of identifying machines that are burning power totally unnecessarily. From political... I, I quite liked to keep my eyes not just on the Al news, but the really good news news last week from a political rather than a technical perspective was the world's just got its first climate science trained president in the new female president of Mexico is a, is a climate scientist by, by trade and training.

So, I, I, I would be very interested to see what affects that out on the country. Any other interesting political news that's good news, do you think?

Navveen Balani: No, I think I've yet to catch up on,

Anne Currie: Well, I'm quite nosy, so I keep my, I keep my eyes open on all things. I would say there's, there's loads, actually, there's, there's a lot of good news going on at the moment. Texas is now a massive solar producing state. It's, so, yeah, there, there is, it's, The world is changing in a positive way. I, I like to, to, to keep reminding everybody that we are not doomers at the Green Software Foundation.

We, we are doing this because we believe it will have an effect.

Navveen Balani: I totally, I would say, especially with the foundation, right? It's all our collective journey that we've gone through. I mean, we started three years back, we didn't have, we didn't have any specification tools, right? Three years down the line, we have the first specification tool. Software carbon intensity specification, which is now an ISO standard.

We have various tools now, carbon SDK, impact engine framework. And I, I know that 1230 projects already in the pipeline on the foundation, right, which will make the world, I would say, a better place, right. In terms of sustainability, right. For all the work that we do. So, yeah, it's basically a shared responsibility, right.

Climate change is basically a shared responsibility, right. And from our perspective, developers, all we can do is contribute, right, by using the three SCI principles, which I talked about, right? Which I again repeat is either write better energy-efficient code, use hardware wisely, and make applications carbon-aware.

Anne Currie: And of course, actually, not just write it, cause you might not be the one who's writing, it's more important that you use it. So as I'll, I'll constantly, cause Python is such a good, good example of the moment, at the moment, because of their big, big revising stuff. If you people who upgrade the latest versions of Python that are much more, more efficient will be saving a lot compared to people who don't upgrade.

And that's, those kinds of things are the kind of things that will immediately be unearthed by running the SCI like for like. I mean, a really big change might be on your like for like is that you upgraded to more recent versions of a particular library or a particular set of tools that you're using that are more, more efficient.

SCI isn't just about what code you write, it's about what code you use. And that is almost certainly going to be where you get the biggest, biggest value, the biggest return. Anyway, sorry, now I'm doing my, I'm trying to take the last word again, and I'm going to leave the last word to you, Navveen.

Navveen Balani: So yeah, very happy to be part of this podcast. Enjoyed this conversation talking about, I think, three different aspects. I would, I would say. And to all the viewers is, yeah, thank you for listening in and have a good day.

Anne Currie: Navveen, awesome. Thank you very much for coming on this podcast. And a final reminder that all the resources are in the show description below, and you can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables. And see you all soon at some point, if they ever let me back in again as a guest host.

Good bye.

Navveen Balani: Goodbye. Thank you, everyone.

To find out more about the Green Software Foundation, please visit greensoftware.foundationon. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: Carbon Hack 24 Recap

13 juni 2024 | 64 min

TWiGS host Chris Adams is joined by Asim Hussain the executive director of the GSF to talk about the recent hackathon hosted by the GSF : Carbon Hack 24. Asim goes through some of his favourite projects that featured work with the Impact Framework including some surprising choices! They also cover some interesting news from the world of cloud service providers and the new CSDDD developments. Asim also talks about how mushrooms are out and bread is in!

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Tom Greenwood: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News & Topics:

GreenOps, FinOps, and the Sustainable Cloud | Forrester [6:03]
EU: New CSDDD compromise finally accepted by Member States | Linklaters [30:02]
Carbon Hack 24 Recap: Asim’s Favourites: [50:12]
- Facebook Open Sourcing the code behind its Power and Water Efficiency Dashboards [51:40]
- Grasp [52:22]
- Amazon Packages for Delivery [56:08]
- Kubernetes Focused Project [56:40]
- GreenerMeet: Assessing Energy Usage & Carbon Emissions in Zoom Conferences [57:04]

Resources:

The Baking Forecast | ShouldIbake.com [2:57]
Hetzner [17:09]
DigitalOcean [17:32]
BackMarket [18:04]
Sesame Open Hardware [20:32]
DjangoCon [23:18]
Green Metrics Tool [23:57]
How does AI and ML Impact Climate Change? | Environment Variables Ep 4 [36:49]
Green Digital Coalition [44:15]
In-depth Q&A: How does climate change drive human migration? [54:32]
All the winners of Carbon Hack 24

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Asim Hussain: And I'm just sitting there thinking, we are old. These are the kids of the future, you know, maybe gave me so much hope, like these are 16 year old kids, and they were doing some incredibly advanced green software measurement, reporting, zoom, understanding curtailment, understanding like how to measure this stuff.

And I was like, ooh, we've got a, we've got a good future ahead of us.

I'm your host, Chris Adams. Hello, and welcome to Environment Variables, the podcast where we talk about sustainable software development. I'm your host, Chris Adams. This week, our format is This Week in Green Software, where rather than doing a deep dive into some of the specifics of green software, we're going to look at some of the news stories that have been making the rounds and provide a bit of commentary and analysis on this. And joining me today is my good friend Asim Hussain, the executive director of the Green Software Foundation. Asim, I'll hand over to you to introduce yourself a bit more than what I've just shared so far, if that's okay with you.

Asim Hussain: Yeah, of course. Yeah. Thanks. I really love these, having these podcast episodes with you and talking through this material. Asim Hussain, Executive Director of the Green Software Foundation and been, you know, lucky enough to be at the intersection of sustainability and software for quite a few years. So based in the UK and excited to be on the podcast again.

Chris Adams: Thank you, Asim. And when I last spoke to you on the podcast, the recurring thing is your history of growing all kinds of delicious mushrooms. How are the myco friends doing?

Asim Hussain: Not good, not good. I've lost, my green thumb or my blue thumb. Maybe if that's how you would, talk about it. But yeah, no, I've lost my, I've lost the thumb. I now bake bread is, the thing that I do, which is equally disappointing, to be honest with you. I just love doing stuff, which is like very, the people, you put a lot of effort in and then you get kind of like middling results is kind of my, seems to be my hobbies of, the day, hobbies du jour.

Yeah.

Chris Adams: Okay, so you've been culturing um, what's it, butter, sourdough and things like that?

Asim Hussain: Yeah, sourdough right now, yeah. I do have a sourdough starter that I, was gonna say keep refreshed, but I probably should say like, keep on the verge of starvation all the time. But yeah, no, it's actually quite fun. I make, I bake bread every weekend and I give it a go. Yeah.

Chris Adams: you for sharing that. Presumably, this is going to be carbon aware bread that you're using, that you're baking when the energy is green, because in the UK, you have the baking forecast. Are you familiar with the baking forecast?

Asim Hussain: I am not aware of the baking forecast.

Chris Adams: Ah, dude, this is so cool.

Asim Hussain: Now that you've said it, I can't believe I'm not using the baking forecast. Because it is, a lot of, that's the thing my wife questions me on. She's like, does the financials of this work out? And I'm like, I don't know, but there's healthier bread than the stuff you get in the supermarket.

Chris Adams: So for those who are curious, and I promise we will talk about green software, this is a kind of segue towards some of the ideas we might talk about later. In the UK, there is a website called the Baking Forecast, named after the shipping forecast, and the idea behind it is to look at the carbon intensity of the electricity that might be going into the oven when you're doing any kind of heating.

And this means that you can then decide to time your baking to be zero or very low carbon bread by making sure you bake when there's lots of renewable energies on the grid. Or, alternatively, if you are not paying attention, you can end up with very, high carbon bread by baking when there's lots of fossil fuels on the grid.

Now fortunately, sunnier, we've got some more solar and wind coming in, it's not like the depths of winter, so it's probably greener breaking today than it was in November or December. But yeah, this is one thing that we should, that I think is, only the UK has this right now. And we'll have to share a link in the show notes because it's a really, cool project.

Asim Hussain: That's cool. You could also end up with really high carbon bread if you leave it in the oven for too long.

Chris Adams: Oh yeah,

Asim Hussain: wait, wait. Oh my God. Where is it? Where is it? I don't know which one it is. You do it.

Chris Adams: I think I know what you're after.

Asim Hussain: Oh, there we go. Oh no.

Chris Adams: This is what you're after?

Asim Hussain: That was it.

Chris Adams: Yes, it, yeah, as, you, Asim has discovered the sound effects on the, on Riverside, the podcast platform we're using. So-

Asim Hussain: You told me about it.

Chris Adams: Yeah, I do apologize. So if we have, we might be like two children at the front of an aircraft playing with this, flip, flicking all the switches, but we'll try our best to not make listening to this too insufferable.

Okay, so we've, mentioned about the baking forecast as a way to talk about carbon aware use of electricity and things like that. Shall we look at some of the stories together, Asim?

Asim Hussain: Yeah, let's do it.

Chris Adams: Oh, if you're new to this podcast, my name is Chris Adams, or if I'm in trouble, Christopher Adams. I am the executive director of the Green Web Foundation, a small Dutch non profit focused on reaching a fossil free internet by 2030.

I'm also one of the chairs of the policy working group inside the Green Software Foundation. The other thing I'll share is that at the end of this show, we'll be showing a link with all the show notes, and it's usually written in Markdown, and we do accept pull requests, so if there's a thing that we've linked to that you think is incorrect, or if there's things you'd like to add, then it's open source and you can do that.

And I think that's the main thing I'll share. The stuff we talk about will be linked. And I guess we should probably find something to talk about then, shouldn't we?

Asim Hussain: Yeah. Something that's not bread related.

Chris Adams: Alright then, well in that case, shall we move on to, I want to make a joke about biochar, but let's just move on from, yeah, alright, so let's, focus on the first story, which is GreenOps, FinOps, and the Sustainable Cloud.

This is from, uh, I think it's actually originally from Forrester, a big analyst firm, who have now kind of woken up to this idea that If you're tracking cost, in many cases you might be tracking carbon. And this is the main thing they're talking about here. There's a few kind of key takeaways and key points inside this.

And this basically does talk about things like, say, you can reduce emissions through kind of looking at kind of existing cost bills, for example. There's different steps you can take. They talk a little bit about some of this stuff. And they also provide a very, high level idea that Right now, we do have inconsistency across the large cloud providers, so Google and Microsoft, they might report all of the emissions, or all of the emissions according to the kind of GHG protocol, which is like the gold standard of reporting, giving you scope 1, 2, and 3, where the majority of the emissions for what we do is probably in scope 3, and talking about how Amazon is doing this, and they've recently, we now know on the grapevine that they've started to rehire people.

But at present, if you look at the dashboards or the calculators of Amazon, you're only going to look at scope one and two, which on average means that you're missing a significant chunk of the picture. So that basically means if you dashboards, and you think, oh, my numbers are really low, that might not be the full story.

All right, Asim, I trust you had a chance to look over it. So are there any particular takeaways or hot takes you had on this one that you might share with us?

Asim Hussain: Well, I don't know if I had any particular hot takes is stuff I've heard before and kind of stuff I've heard mentioned kind of in quite a few times over the past few years. And what I was actually interested in is perhaps just having a more. I'd actually love to hear your thoughts and maybe we can like walk through some of the ideas and some of the question marks I have are just about the whole premise of this link to cost and carbon and like, there's a few things I think of when I think about that.

One of the things I think of is, well, isn't, the cost of things the reason why we're in this problem in the first place? You know, how can costs, how can money both be the cause of the problem and the solution to the problem at the same time? And then, like, you know, you know, yes, there is this argument that kind of reducing cost reduces carbon, there's this kind of correlation.

I've spoken about it extensively, in my past, but kind of I'm trying to refine my thinking in this space. We know that there's somewhat of a correlation there. But we also know there's things you can do which can dramatically reduce carbon, which don't affect your bottom, the cost at all in the slightest.

So what is the, how big of an opportunity is that and is therefore focusing on cost kind of distracting us from things that we can do, which can dramatically reduce our carbon that they've got nothing to do with the cost. And another thing just kind of pops up in my mind as well, which is I remember I was speaking to just the energy purchasing department at Microsoft years and years ago. And I actually asked them, can I quote, quote you on this.

Chris Adams: And this is because you worked at Microsoft.

Asim Hussain: I used to work at Microsoft. Yeah. Yeah. And a rule of thumb that they gave me was, which was the, this is pre AI, was that the cost of running a data center is only 10 percent energy. The actual dollar cost, 10 percent energy, 50 percent depreciation of assets.

So you bought some chips, and they're gonna be worth zero in five years, so they're depreciating of assets. And everything else was like, you know, people and everything else was number one. And so like, when you think about that, certain parts like that emits carbon into the atmosphere, all of those-

Chris Adams: The creation of, oh, each of these, right?

Asim Hussain: The energy emits carbon, but it's only 10%.

You know, the human beings running around emit carbon, the actual physical infrastructure like emit carbon when it's created. And then what happens in an organization is you normalize that into a dollar value, which you then charge for services that run off there. But does that dollar value, like represent, you know, accurately, you know, like the carbon emissions put it that way like there's probably a lot of nuance here yes. Reducing your cost reduces emissions, but maybe reducing costs In this way has unbelievably emissions reductions and reducing, you know, reducing $10 in another area, eh, probably doesn't have that much impact at all.

So I'm kind of a little concerned with how I don't really think the link is kind of that well understood and I'm worried a little bit that if we focus too much on cost reduction, we might be able to reduce costs in ways that have no impact on carbon emissions reductions. That's kind of like where I'm thinking about.

Chris Adams: Mm.

Asim Hussain: I was just wondering what your thoughts were, actually.

Chris Adams: So, I think there's two things. I'll share a question to that, but you said two things which I think are really interesting there. First of all, the cost of energy as, like, that's a part of what the data center might be, right? And you also said, like, depreciation of assets being one of the big costs.

And that, I'm assuming that basically means, like, I've got a server, it's going to last over, say, three or four years, and because of that I split the cost over four years, and that's how I, that's, you know, it's-

Asim Hussain: Yeah. Accounting wise that let me see exactly how they figure out accounting wise. Yeah.

Chris Adams: All right, so there are two things. Most of these discussions seem to be about energy usage, with the idea being that, I think you're right, there's totally mismatch between these two things, because we are, in many ways, we don't see any of this stuff, we don't have the visibility in a bunch of these things.

And one thing that we've, one thing that I guess we've seen a number of large companies start talking about now is saying, well, we are looking to extend the life of our servers, and we're doing this as a way, as a green measure, right? That sounds great, yeah? But if you extend the life of your servers, you've gone from 50 percent depreciation to maybe 30 percent or 40 percent depreciation, right?

I mean, from a financial point of view, that basically, I mean, Amazon did this. It put like billions onto their, on to their balance sheet because they now look way more profitable than they were before because they don't have to kind of write it off. So, yes, there are some things being shared for this but whether how that actually factors into some of the calculations people use.

I mean, I don't know. There was an announcement there. We haven't seen any changes whatsoever in any of the kind of cloud dashboards, so you would have expected that to make a difference if you did have, if we did see that, and both Amazon has done this, Google has done this, and Microsoft had done this, right?

So, I would argue that looking at calculators, there's a gap here between what we're seeing here and what we are being taught to optimize for. And there's also another question. Yeah.

Asim Hussain: Yeah, no, it, and there's also, I remember just as you were saying that, I remember this, and this is years, this is information is years old, but I was chatting to the, uh, the Xbox team at Microsoft and like one of the things, I mean, I still, no, I don't have an Xbox. I finally got rid of it. That initial estimation

Chris Adams: You sorely miss. Your love to Xbox, that is. Yeah.

Asim Hussain: Yeah, love.

Chris Adams: Oh no, you don't work for them now. You're okay. You can say anything.

Asim Hussain: I'm okay. I can, get rid of it, I think. I think. Yeah, you wouldn't get, you wouldn't get a free Xbox if you went to Microsoft, and I always thought they should give you one. But, it was, like, they assumed that it would only last, like, four, three or four years. I can't, I'm, don't quote, no one quote me on this.

But the actual lifespan of kind of these consoles turned out to be a lot longer. And people, I don't know how old Xbox is now, like, it must be like six, seven years. But it's still, like, pretty, you know, still pretty active. The decision was to kind of still keep, from an accounting perspective, still keep it kind of three, four years, even though the reality was it was being used for longer.

So those decisions are being made as well. But it's interesting that's part of the, yeah, I suppose that's part of the story we don't really, The point I'm coming from is this, is a $1 reduction in your costs, how do you link that to an actual carbon emission? Because coming back to my point of like the energy versus the embodied, like how much of the emissions is linked to the energy versus the embodied and therefore how much of your cost is linked to reducing the energy versus reducing the embodied?

Does that make sense? It's

Chris Adams: This comes to basically disclosure of cost structure, which is a bit of a kind of like jargony term but basically without having visibility into how this stuff is paid for because in many cases we have very proprietary services, it's really hard to basically, be sure, like, is this really happening?

And even when we talk about things like, say, cloud, moving to cloud compute, moving to, say, serverless, all these other tools, which are kind of higher up the stack, right? There is, it's not like the profit margins for cloud at the very bottom, like paying for compute per, on a per hour basis, will be different from things higher up.

And we don't have much in the way of visibility into any of that stuff. I mean, I, as I understand it. The higher up the stack you go, the more value you add. A bit like if you're buying refined, if you're buying oil, and then you're selling, say, refined kerosene or something like that, it's going to be a different price, right?

So you can think of, I think we have maybe some mental models we can apply for some of this. And this is actually possibly one way that you can say, well, if we have If we're running a bunch of companies, and we have a set amount of resources, and we cannot use more resources, like more inputs, there are ways to still continue to increase the amount of kind of value and profits being made by shifting people to kind of go higher up the stack to get them to purchase, maybe say, a serverless thing, or something like, something which allows you to kind of cram more stuff onto the same infrastructure.

But the problem is, In order for you to do that, you need to not build more infrastructure. And what we've just seen is the opposite of that happening right now. And I think this is what, I think you're somewhat right. It's easy to talk about cloud costs because it's, you've already got the numbers there and you don't have to, yeah.

Asim Hussain: It's like one of the most obviously disclosed, like, facts about your cloud usage is how much it's going to cost you.

Chris Adams: Yeah. And like, if you're in an organization, it's probably easier to start optimizing a number that you already have than to pay for someone to, or even pay for a solution for someone to do this stuff. Because I've invested like. We've, uh, okay, I promise we'll get back to the other stories a little bit later on, all right?

But inside the Green Web Foundation, we've been looking to, like, audit all of our own services recently, or just, like, work out our own kind of, look at our own annual emissions. And the thing we found is that, let's say we've, uh, the majority of our impact, because we're a quite small organization, comes from, not necessarily from the services we're paying for on an hourly basis, like, we use Hetzner, and we don't, Which we're spending maybe, I guess, maybe a thousand euros a year on Hetzner in total?

Asim Hussain: What's Hetzner?

Chris Adams: Hetzner is a German cloud computing service. So, about half of all of Mastodon run on Hetzner service because they're so cheap. So much cheaper, and in Europe at least. They've been one of the greener ones to use, because they build all their own software. And if you look at their prices, they're typically, like, you know how DigitalOcean will sometimes be cheaper than, say, the big cloud providers?

They are cheaper than DigitalOcean, once again. So as a result, we're a cash strapped non profit, we're going to go for the cheapest green, option we can find. And the cheapest green option that We can find that has cloud like APIs and is used and works quite well. So we've been using those ones, and as a result, cloud makes a small amount of our kind of digital estate.

What ends up being a larger thing is the infrastructure we use, like the laptops and the monitors we have inside our own houses, inside our own offices. But, even if we buy those from circular providers, so my laptop is from BackMarket, which is like a kind of second hand provider, and likewise, Hannah does the same thing with hers, and all this.

Hannah's my, she's my colleague who's been leading on this work here. We've done this, but under most of the ways there's no guidance on, okay, is it someone else's carbon? Do you know, has someone else banked it so therefore it's free for you because they've depreciated the carbon there? There isn't much in the way of guidance or real stuff that you can adopt.

Like, you can might, you might be able to sight a paper, but that's not the same as there being, say, guidance that you might use for your own reporting to say, this is why we've chosen these numbers here, for example. And like, this is what, these are some of the problems. There isn't this recognized way to account for circular versus buying new stuff, for example, and this does speak to this idea of, well, okay, how do you estimate this stuff and work out what is the most effective thing, which, you know, which levers should you be pulling if you want to reduce the emissions from software or digital services in general?

Asim Hussain: So that would be like an example of, well, yeah, cause I've even had these conversations before we're talking about like, what'd you do? I think with all these like hyperscalers, like once they actually reach end of life, they're not broken.

Chris Adams: Mm

Asim Hussain: the, equation that they're running is like, we've got per square foot of infrastructure, of real estate, what is the profitability? And at a certain point, it becomes more profitable to take out this old server, which is perfectly fine.

Chris Adams: Yeah, usable. Working.

Asim Hussain: But you can just put something else, which is more profitable per square foot for that given situation. So that, I mean, people think these things are breaking and then no, they're not, they're just perfectly fine and working.

They might be more inefficient. Then we've had like conversations in the past with, I've had people going, well, if I was to take that server. End of life server and build my own data center. Could I argue that is zero? Could I argue that it has zero embodied emissions? Is that like a reasonable statement to argue?

And, I don't know. I don't know what the answer to that question is.

Chris Adams: The question, the answer is likely to be about, oh God, this is so nerdy, mate. I'm so sorry. But like, what is the depreciation that we've just spoke about now? You said half of the cost in a data center is a depreciation. What's, how do you split that? How do you kind of, Amortize the carbon over these years and then how do you then share that with someone else?

So there's a company called Open, I think they're called Open Sesame or Sesame. They basically build data centers or like racks and stuff out of all of the, uh, Yeah, decommissioned stuff from Facebook and all these open compute projects, because all the companies, because you know how these servers are designed, and know how they're specced, it's actually quite common to buy these, and these tend to be cheaper than what you might have elsewhere.

And in many cases, it's not like, It's not like they rust, right? They really, they're, kept in kind of quite good con condition and they are working in a lot of cases, but the, they make the argument that, well, this is circular, so this is gonna be greener what you have, what you've been using elsewhere, but whether you are able, but yeah, it's, a number question.

It's an open question about how you account for that stuff inside, if you're purchasing, say, computing from these folks compared to people from somewhere else, because a lot of the time, order for in you to do this, you probably need organizations to be really transparent about where they got their servers from, and that's not always what companies are incentivized to do, basically, right now.

Asim Hussain: but then we get to that same argument of like, at the end of the day, that's going to be sold to somebody as like a dollar cost, like dollar per hour of a CPU. and if we just look at cost as a proxy for carbon, then you could argue there that will, they might have to end up selling that service at a rate that is actually quite comparable to a brand new server.

Chris Adams: Yeah.

Asim Hussain: This is where I think there's this kind of, I don't know if it's a breakdown or there just needs to be a lot more thought put into it. It's almost like there's a, that you pay a dollar value. There's a black box. And then at the other end, we know there's some sort of carbon emissions. And I think there's a lot of nuance there.

And maybe what I'm saying is that actually cost isn't a great proxy for carbon. Because really, maybe that is, I'm kind of refining my thinking as I'm talking to you. Maybe cost isn't a great proxy for carbon. Or maybe you put it this way. Three, four, five years ago I was advocating for it because we had nothing else.

Chris Adams: Mm.

Asim Hussain: There were so few other options, but I'd actually argue now, we have the capacity and capability, even if a cloud provider, somebody themselves don't even know what the carbon emissions are. There are models that now exist where you can estimate from what you can see about. But the infrastructure that you're using and make your own assumptions and make your own judgments as to what the carbon emissions are.

So I think I suppose where I'm thinking now is we can actually move beyond costs. I suppose that's where I'm thinking.

Chris Adams: Well, the thing I might share with you, and if you forgive the plug, so I've just been travelling all around Western Europe by train, and I literally started on the 31st of May, and I arrived at this conference called DjangoCon, which is a conference all around DjangoCon, Django, which is a very popular framework that initially powered, say, Instagram, and like, part of the NHS website, all this stuff here, yeah.

Asim Hussain: I used to be a Django developer. That was like my job. Yeah. My job, my paid for profession for quite a few years was a Django developer.

Chris Adams: Wow, I did not know that. Okay, no, that's a, okay, that's a story I'm going to use then, because I think we can bring this back to the cross thing for a second. Anyway, I did a talk, I, there was very short notice, I was going to deliver a workshop about how to, basically, how to use some of the tools available to kind of green your Django stack.

Right, so we were using a tool called Green Metrics Tool, which as Ana, one of the, like, I guess the founder of this, the guy who, the lead maintainer, he says the name is descriptive more than creative, but you kind of know what it's for, right? I quite like that. Anyway, that does give you figures for energy usage and embodied carbon and it does provide these and even provides like SCI figures. So there's now like a workshop and a deck and I will show a link to it to see how you can take an existing project and come up with SCI figures for various user journeys and things that you do have. Asim, I didn't have time to talk about the Impact Framework, and I, once I know enough about the Impact Framework, I'd be really up for using it.

But one of the key things I had to do was I was doing this workshop and then literally the day before I arrived, we had someone with a visa problem, so they couldn't do their talk. So basically the organizers asked me, said, Chris, are you, do you have a talk ready or are you, would you be up for filling this spot?

We have a 30 minute slot to talk about this. And I basically had 24 hours to take the workshop, turn it into a presentation. And the thing, the reason I'm talking about this is that I presented a kind of taxonomy of ways to think about these tools that you have. And I described things in terms of usage based and cost based specifically for this reason.

So you would use cost based tools to do your first round of disclosure or to work out like a baseline because until you, and until you've been able to demonstrate any value from doing this, it's going to be really hard to argue for like weeks of developer time to come up with some numbers. Right?

Whereas this at least gives you a number that you do have. And then I spoke about how you would use usage based tools to start figuring out, okay, well, what changes can I make to reduce some of this stuff? And yeah, I'll share the link to the deck, because it's, it allows us to create the, you know, we need, like, taxonomies to realize that we're not conflating these ideas too much.

Because a lot of the time, you're doing different jobs, and you're asking, you're being asked by different people. So, like, for example, the whole kind of Software Carbon Intensity thing, initially, it was, as I understand it was created partly because the sustainability, like the head of sustainability asks for some numbers, and if you're going to report along these numbers, you don't really see much in the way of incentives to show how you're changing in future, right?

it doesn't incentivize changes at the engineering level.

Asim Hussain: Yeah, yeah, yeah.

Chris Adams: right? So therefore you need a, you know, one of the thrusts would say, well, let's come up with a metric to show that we are planning forward. And we're taking some steps so that if you're an engineer, you have an idea of what's, what, you can do rather than just outsourcing all to energy procurement, because that-

Asim Hussain: Yeah,

Chris Adams: isn't necessarily the full solution, for example.

Asim Hussain: I mean, the way, I mean, I really love your taxonomy of cost based versus user based. I think that, yeah, I like that actually. And user based being more, perhaps more, more granular and more action. Cause it kind of gives you more, more specific things you can do. Whereas cost is like this really high level thing.

Just to bring it back to what you said about the SCI, the, one of the things, And I really, I didn't coin this, uh, but when it was coined, I loved it, but they described it as measurement for reporting versus measurement for action. And that's, I think, an interesting way of looking at kind of the different measures, uh, in this space.

And, you know, we talk about regulations. We need to talk about regulation. We want regulations a lot, but once you start talking about regulations, the measure that you have is a measurement for reporting and the, behavior it triggers is what, number can I disclose that I can defend? What can I defend? Whereas a measurement for action is kind of what number do I need to drive action change. And that's when we think about the SCI, it's a measurement for action. I don't know whether, where, it will need to land to be something, you know, to be, you know, To be, you know, defensible.

Like, I don't know why I'm gonna go off on a massive tangent right now, but-

Chris Adams: Go with it, let's go with it.

Asim Hussain: Go with it. Let's go with it. We know Krav Maga. You know Krav Maga?

Chris Adams: Wow, yeah, the Israeli martial art you're talking about?

Asim Hussain: The Israeli martial art. Like, I've always, there's actually a place nearby and I've always wanted to learn.

Chris Adams: I did not expect you to go in that direction, let's go with

Asim Hussain: Well, but let's just go with it.

Because here's an interesting thing about it. And one of the things that the founder of Krav Maga said, this is will never be allowed to be done in like a competitive environment. There will never be the world championship of Krav Maga. There will never be like this, you know, global Krav Maga, blah, blah, blah.

Because by, by that's when you start bringing in rules and the actual nature of the sport changes. So it doesn't become good at what it's supposed to do.

Chris Adams: Ah, yeah, so for people who are not familiar, Krav Maga is, it's like, oh, it's like punch him in the love spuds and all the stuff like that. It does all the kinds of things which you wouldn't do in a, kind of formalized mode, because they are, they're basically weapons of war, rather than actually weapons of art, right?

So it's more like, there's more martial than the art part, perhaps.

Asim Hussain: I mean if I'm gonna learn a martial art, it's gonna be there so I can like actually get out of trouble, so I don't mind like poking somebody in the eye or kicking them where they shouldn't be kicked.

But like, but I think that might, that's just one of my thinking and I'm, and I, this is just my thinking and this is a consensus based organization, so I just want to really be clear that I'm just expressing kind of some thoughts have and they might change in the future as well. You know, can you have both things?

Can you have a measure which is good for action and good for reporting? Or does, you know, as soon as you bring a regulation into a measure, does it then transform the nature of it, so it, you know, it doesn't become good as an action driving measure and it can only serve as a , as a measure for reporting.

That's kind of some of the thinking we've gone, I don't know how we got to this.

Chris Adams: No, this is a graceful segue into the next story, Asim, don't worry. All

Asim Hussain: But yeah, that, that's some of the thinking I had.

Chris Adams: I don't have a, I won't try to directly answer that, but what I will do is name check the new story, the next story we have, which is from Linklaters, the law firm, and they talk about, so this is going to be a bit of jargon.

Asim Hussain: Oh, yeah.

Chris Adams: I'm afraid, so this is a, this was a new story from actually a few weeks back, say, the new CSDDD compromise finally accepted by member states.

That is impenetrable to most people who are programmers, but the short version is that there is a piece of law. called, we already have a piece of law called the Corporate Sustainability Reporting Directive, which is a bit like, which essentially compels organizations of a certain size to disclose their carbon emissions on an annual basis, just like they disclose their financial things and their financial figures.

Different parts of the European Union are now, okay, for want subclassing the directive and turning it into their own law, right? So, France has been the first of the countries in Europe to actually do this. And Egalité, CSAD, yeah, like, they have a really hardcore version of the law, where if you are a corporate director, and you don't disclose, and you take steps to block disclosure, You have jail, you know, there are jail time sentences and stuff like that now.

There's some really, hardcore things, right? And that's like the disclosure thing. But the thing that's interesting is that this plays nicely with this new law, which is the Corporate Sustainable, Corporate, I think it's Corporate Supply Chain Due Diligence Directive. And this introduced some new demands.

So you basically need to like, as you'd imagine, Have some due diligence in your supply chain, but they also need to demonstrate how this fits into some of the laws or some of the kind of longer term goals of being a company active in, like Europe, where Europe has set targets of saying, well, you need to reduce emissions by, you know, we aim as a kind of union to reduce emissions by 55 percent and who knows, maybe even 90 percent by 2040.

And as a result, you know, you now have to publish a climate transition plan, which shows what your steps are to actually help get there. And, uh, this I think is interesting because this now means that you have basically a mechanism for accountability each year, I mean, kind of anyway, but you've also got something to show that you're looking forward.

So you've got forward looking and backward looking. And I think this is quite helpful for the conversations around, say, a consequential model like The Software Carbon Intensity thing, plus some of the kind of somewhat backward looking models that you might have with the GHG. And I think the GHG protocol, and this is how I now think about this, you will say that you might use the SCI to plan forward and say, well, am I going to do this versus that?

And that allows me to talk about, it's not just my organization, like, uh, are there places where impact is taking place where I don't have to put it on my books, but I know it's a significant problem, right? Because the current state of the law doesn't actually incentivize you to do that. And, uh, particularly if you're like, say, a media company, for example, where, you know, most of the impact is in a router in someone's home, or in someone's end user device, like their big ass television, you might not try and tell someone.

Hey, can you please stop using your big ass television and use a, tablet, right? Yeah, you might do that, but like, it's, that's going to be a hard thing for you to kind of argue is inside your kind of organizational boundary, for example, for, reporting. So this is why having two ways of looking at this is actually quite helpful.

And this is how I've expressed the difference between an SCI and the GHG protocol. The fact that they are complimentary to each other, but very but different.

Asim Hussain: Yeah, and I think that's, I never meant to throw shade on any kind of measurement for reporting, like it's all, needed. it's just different audiences. It's just different audiences, different needs, different, other things. So would you say the CSDDD is, because I just think the way you just described it, there's, it's, different to how some of the ways I've heard it described, but the, does it go beyond the organizational boundary?

So the climate transition plan that you have to expose, is it how it supports the EU's aggregate goal, or is it just how you, your organization plays, how its emissions has to reduce? Because if, it's an aggregate goal, then the kind of debate that you're just describing, which is like, I'm making my TV more efficient, but who cares? It's not part of my or something. That's not part of my effectively changing Help showing how you're helping the climate transition by things outside of your value chain.

Chris Adams: I need to stress, I'm not a lawyer. All right. Okay. And uh, I think I see, I think you mentioned that there might be someone who is a lawyer who does look at this, that I think it would be really nice to have someone who, with that deeper domain expertise to kind of talk about this 'cause. I'm mindful that because I have an English accent, it sounds like I know what I, that I am confident and, uh, about what I'm talking about when I might not be that well informed.

Asim Hussain: In a lot of you're very well informed in a lot of areas. But anyway, yeah.

Chris Adams: Alright. So yeah, I'm definitely not a lawyer in this con in, this context. Uh, but the, I guess the key, you know, your question is do you use 'em differently and would you be, I mean, is that the question that you had? Like, uh, uh, is that, maybe you could just repeat that to make sure I understand the question.

Asim Hussain: Well, I think it's interesting. I think it's, I think, because a lot of these questions are kind of like insular to an organization. Like what are you doing? What's your organization? How is your organization reducing its own like emissions? And then the conversation is basically coached around, "well, that's not my problem. That's not me. I, even though you could do something materially to reduce those emissions, it's just not part of my value chain. So I'm not going to be focused in on it." Is it, I presume because it's a business reporting directive. It is kind of very much like how, like you've, made, how are you as an organization going to your emissions in line with the EU's target by 2040?

Chris Adams: So, okay. That's, yeah. I think I understand your question now. So, uh, as I'm aware so far. The CSDDD is basically, they're still all focused very much on your own organization, right? So it, I think most of the supply chain due diligence part is about your upstream, right? Not necessarily so much stuff that you have downstream, for example.

So, you know, you might talk about who you purchase things from, but it might not be about to what extent am I incentivizing or am I accelerating the production of fossil fuels, for example, right? So right now, let's say, when we saw this, when we saw, I think Will Alpine, you know, he was a guest before, and so, so, so, so, so he mentioned this a while ago, and I think way, way back in the first or second episode.

He spoke about, okay, we need to be doing, you know, we, need to be using, we need to be responsible for our use of AI. Yes, we need to think about the green part, but we also think about these parts elsewhere. He's been like really, he's been leading on a bunch of this stuff and talking about this. Yeah.

Yeah. So, the yes.

Asim Hussain: Scope 4, he got me onto the idea of scope 4 emissions, which is like Very interesting.

Chris Adams: Yeah, so he, so like, he's been talking about this in quite kind of clear ways for quite some time. So, the scope 4 stuff that you've just mentioned, don't know to what extent that's actually included in this right now. But if you're talking about your plan, it's likely that if you're going to have a transition plan, you probably would have very good reasons to talk about what your transition plan looks like because you're helping transition the entire economy.

So it may be that you might refer to this, but this is where I'm outside of my kind of comfort zone, basically.

Asim Hussain: We can, lobby right now for what we'd want. I mean, what the ideal solution I would say is not lobby. That's the wrong word. We don't, do lobbying, but you know, you know, There is this overall aggregate goal that we need to achieve. And that's it. I love this. It was Henry Richardson that said that.

I remember it so specifically. I know the person that said it. And I know in what meeting they said it. was such a great statement. We were talking about the SCI. And we were talking about this whole idea of double counting. You know, double counting, when it comes to reportings and all this stuff, it's so significant because you're an organization.

You don't want to, you know, You don't want to be responsible for somebody else's emissions. You don't want double counting. But he said, actually. As a humanity, we want double counting because what does double counting mean? Double counting means that there's more than one person incentivized to reduce that emissions, right?

That's, the advantage somewhat of double counting. And actually, if you think of scope 3 in a way is double, counting across the board because somebody's scope 1 and 2 is somebody else's scope 3. But I actually completely forgot where I was gonna go with this.

Chris Adams: I'll try and rescue Asim before we move to the next story. So, we were talking a little bit about, and bring it back to digital services, right? The, so, I can't name the organization that we did some of this work for, but we were talking about a large media company, we were talking with a large media company, and we were helping them understand this, because some of the jargon, I'm afraid, that people use for this is like an attributional approach, which is like, are they my emissions? Versus a consequential approach, which is, do these emissions, you know, is this activity going to increase or reduce emissions? And when you're looking at media, for example, essentially, if you look at the attributional part and the GSG protocol, what you have to report on, right, you are incentivized to care about the data center more than anything else, right?

Because that's kind of inside your boundary at the moment, or what. Most organizations report as their boundary, right? Not every organization. But if you were to look at the consequential approach, you'd be looking at, yeah, all the things mentioned before. And like, the, this is quite a significant thing because for consuming digital media for consumers, the data center's making, Less than 10 percent, right?

It's tiny compared to the, well, the energy use at least of the, yeah, and uh, if you look at the energy use, it's going to be, you know, more than two thirds is coming from all the stuff at the end users, which is outside of your, you know, it's not on your book, so why would you care right now?

Asim Hussain: Well, it's an externality in a way, isn't it? Yeah.

Chris Adams: So this is, this is why I think it's interesting to see these two things have some interplay, and we, I think we need some wording for this right now.

Like, the one I'm trying to popularize is, you have climate disclosure, Which is one of my missions. And then there's climate response. What was my plan to reduce this? Right? Climate disclosure, you want your missions to be small. Your climate response, you want it to be really big. really, like, massive and ambitious and, like, loud and everything like that.

And that's what I think you can get behind.

Asim Hussain: But I think that's what Scope 4, that's why I think Scope 4 is a really interesting concept. It is, I've forgotten how it's defined exactly, but you know, things like your response, like things that you are doing outside of anything to do with making money or products, like that is what's counted in Scope 4.

But also it's like the impact you're having, the impact your, business and the work that you're doing is having on the fundamental problem of climate change. So, i. e. like, if you're enabling something which helps people to find another oil reserve, that's a massive scope for. That would like overshadow anything that you're producing in this world.

But scope for could also be negative, because if you're doing lots of work that actually sucks out carbon from the atmosphere. You can actually have a, like a, scope four, which a note, scope one, two, three, can't be negative, like, unless you're not doing, unless you're just not breathing or just sitting very still, but like, it can't be, but scope four could potentially be positive or negative and it shows your, like, how you, really like impacting the world, like.

Chris Adams: Yeah, so the thing we could possibly link to, I'm going to add to the show notes the episode we did, the podcast episode, which was all about AI and ML impacting climate change way, way, back in the day, I think June 13th. Oh wow, it's literally two years ago.

Asim Hussain: Is it two years?

Chris Adams: Yeah, we're recording on June 12th.

Asim Hussain: About AI?!.

Chris Adams: No, we've spoken about it before, but this is the, this is one I'm really, actually, I really loved this one. This was recorded on June the 13th, and this included, it was Abhishek Gupta, who's now, like, inside CAT, inside Climate Action Tech, which is a community that we're both part of. He's being funded on a micro grant to come up with some sustainable AI principles, but Will, Will Alpine, he, well, Will Buchanan is what he, before he got married, he was called Will Buchanan.

He mentioned a bunch of this stuff. And we also have Lynn Kaack, who was one of the co founders of Climate Change AI, which is another one of the really, in my view, really fantastic organizations who have real authority in talking about that. They mentioned a bunch of this stuff, so if you want to know a bit more about AI, look at these things from two years ago, because a lot of that still stands, and it's very, it's, I, learned so much from that one, basically.

So yes.

Asim Hussain: I get pinged by organizations all the time going, "oh, we want to talk about, or media, we want to talk about AI and, oh, AI and sustainability is suddenly a problem." And I'm like, we've been talking about it for a long time, like well before chat GPT came on the market as AI being a, hot topic of conversation in sustainability circles.

Chris Adams: Well, maybe the thing we should also link to in then Asim is, A, there's, so the GSF has actually started to endorse some existing legislation in America, to make this a bit clearer. But also, in Europe, I think something like a million and a half or a million euros, there was a tender a few years ago, to basically start researching some of this stuff that you might refer to as Scope 4, basically.

Like Scope 4 is, it's an idea, but it's not, there's no standard around it right now. So there was some work there to start looking at. This to see how you might quantify the positive and negative impacts of various steps. Now, because you've got, this was funded by the Green Digital Coalition explicitly to kind of measure the positive impacts, it's not surprising that every single intervention is only positive.

So it's not like there's a methodology for talking, oh, if we use AI to extract oil and gas, does this mean that we've, you know, there is nothing like that. And I would be very happy to see something like that because those climate emergency, see all this stuff here, right?

Asim Hussain: Well, that's one of the reasons I was thinking about this recently is kind of where we're starting to, well, we've somewhat measured our emissions. I'm starting to like, now that we're growing, right.

Chris Adams: This is the GSF you're about?

Asim Hussain: GSF. Yeah. And I suppose the green web is the same, isn't it? Like if you were to employ just one more person, your emissions would jump up significantly.

And so like, how do you meaning, meaningfully have a plan? Or what do you even say when you're like a very small organization that's just growing? Like, I mean, forget if you're

Chris Adams: If you hire five people, you've doubled your

Asim Hussain: have like you've doubled your emissions and you've set yourself a target to reduce that, what does that really mean?

And I think for, especially for organizations like you and me, like Green Web and Green Software, like if there was a scope for, our scope for is significant. Because the activities that we do as an organization, I don't mean scope 4, our negative scope 4's are significant. Because the act, us existing, arguably, hopefully, reduce the overall emissions of the world.

And so in a way you could, you know, it'd be interesting to look at organizations from their scope 4, because then you're investing in kind of a different component. There could be like startups out there who aren't like in the green software space, but have a negative, if you have a negative scope 4.

The more you grow, the more emissions get reduced from the atmosphere. It's interesting.

Chris Adams: I'm I'll be honest, dude, I'm a little bit I'm sceptical of this scope 4 idea because it very much, when I've seen people talk about it, they use it as a way to either say well this is why it's okay for us to deal with these really damaging things elsewhere, right, it's used to basically draw attention away from other things, or I just don't think like It's not, it's never going to balance out.

It's okay for things to be like, orthogonal, in two different, moving along two different axes, right? And I think it, that makes more sense. What you can do instead, is actually be responsible about the impact that you do actually have. And the, there's a, an approach used by the New Climate Institute, who I think, which I think is really interesting.

What they do, they basically say, look, We're a small non profit, our job is to be at, say, COP 28, 29, 26 or something and essentially work with policy makers to set the rules that end up impacting, influencing ginormous organizations that are way, larger than us. And what they do instead, they basically say, well, we have to fly there, we have to do something like this, we're going to impose an internal cost of maybe 100 euros per ton that we put into a kind of, I forget the term they use, Let's just call it a carbon war chest, right?

A climate war chest. And then they use that to fund other systemically effective things. And, like, this feels like a much more sensible way to do it. And this dude, like, dude, like, you know how the GSF, where the GSF came from, right? The funding for the GSF came from the internal carbon levy inside Microsoft.

You told me this before.

Asim Hussain: Well, actually, no,

Chris Adams: Oh,

Asim Hussain: I didn't, get there. No, there is an internal carbon levy in Microsoft, which is then used, to fund green measures but the actual, like, technically the money from this actually came from a different bucket.

Chris Adams: Ah, dude, I, ah, that would have been such a good story.

Asim Hussain: It's unfortunately not a good story, yeah.

Will Alpine, a lot of his work he did at Microsoft was actually funded from that war chest. He knew how to navigate that space and get money from that war chest.

Chris Adams: So this feels like a more sensible way to do it, right? The idea of saying, well, you're not trying to say, you're not trying to get back into eco heaven, you're being an adult about the emissions, and about that there is an impact being caused, and then you're talking about, well, what is the way that I can do it, which doesn't give you my sugar rush of saying, I'm carbon neutral again, but says, well, we have a societal goal to get to, so let's think about it in that sense.

I'm, this was, I mean, Microsoft is using an internal carbon price. It's not the same as a carbon tax, but it does revolve like, it's a discussion about how you allocate time and money to projects, rather.

Asim Hussain: It's a tax because different departments have to pay different amounts depending on how much carbon they've been deemed to have omitted. So it is kind of like a tax.

Chris Adams: Yes, this is something we can-

We don't want to use the word tax? Oh, is this another topic?

It's not so much that, it's, I mean, it's more that, yeah, you might not want to use the word tax in various places, because certain people are in favor of tax, certain people are not in favor of tax, right? And like, there are loads of ways where people allocate funds to various things inside this.

I mean, like, there's a reason inside Microsoft they use it, they called it a levy, right?

Asim Hussain: Oh yeah.

Chris Adams: But like, we, you know, there are loads of places where we allocate a percentage of some funds to funding something else. So this is why I've stayed away from using the term tax inside this, because in the way I see it, governments get to levy taxes, they get to like raise taxes and stuff like that, but organizations don't necessarily, and there are loads of cases where if you're inside an organization, you're going to allocate this much to kind of keeping your staff happy.

Is that a staff happiness tax? Or if I'm paying for cloud, right, and it's a chunk of my, let's say I'm building a digital service and I'm paying Amazon a chunk, or Microsoft, is that a Microsoft tax? I'm not sure it is. It's, I'm paying for something, right? Yeah.

Asim Hussain: Good point. Now that you've said the word tax like five times, every time you said it, I get a-

Chris Adams: You, sir, have said the word tax. I have not used the word tax at all.

Asim Hussain: Have you not? Okay. Well, I think it, I think anyway.

Chris Adams: I talk about internal, I talk about internal carbon pricing, because this is internalizing the costs that are otherwise being shifted onto society, and that seems a kind of economic and kind of grown up way to think about this.

Asim Hussain: Yeah.

Chris Adams: Yeah, uh, yeah, I'm not using that word either for very good reason, because there's another thing. Anyway, we've totally gone off script. This has been fun, though. Let's, should we go back to where we are? So we were going to talk about CarbonHack, some of the aftermath and learning points from that before we wrap up.

So are there any particular projects? So maybe it might be worth you just briefly summarizing what CarbonHack actually was, and then maybe if we talk about one or two projects. So, two projects that really caught your eye, then we can wrap up after that because we've this, has been, this was going to be a short, one and we're running about 50 minutes already, basically.

Asim Hussain: That's okay. People must have some long journeys ahead of them. Summer holidays coming along. Yeah, no, it was really good. So CarbonHack this year, we focused on Impact Framework, but more specifically focused around measuring, like how do we actually measure software, so it's really exciting. So, I'm really excited to be here where we can actually talk about this.

So, really quickly, Impact Framework is a framework that we've been building here to a very low level. You can measure, print, kind of most things with it, but also provide that evidence of the measurement in this impact manifest file. So there's a couple of, we had a couple of prizes. We had the ones I was very excited about.

I was inspired by all of them, to be honest with you. But the one I was really keen to see what people It was we just threw it out there. We were like, well, let's just see how people respond to this one. And it was beyond carbon. Because we've talked about carbon all the way through this episode.

But, you know, as we know, the sustainability challenge we had to have ahead of us is actually far bigger than just carbon. And we just left it fairly open. Kind of anything that kind of measures the environmental impacts beyond carbon was effectively it. And there was some really interesting submissions.

Like the one that won was, as we might guess, like water, like they, they measured the impact of water. And it also triggered some like really interesting conversations around, you know, the, I know there's been some research in this space, but like, at what level are you measuring it? Are you measuring the water impact?

Like at the data center? Kind of like primary. Primary and secondary, primary water and secondary water kind of was like the concept and it was just exciting. But now, what that means now is that there's now a plugin for Impact Framework where you can just, there's actually two, but you can drop it in and just say like, I just want to estimate my water impacts of my workload.

You can just drop it in and do it. Another, team was from a team called Grasp and I loved it because, you know, one of the things that plugin measures. Actually does a couple of different beyond carbon categories, but the three main ones are death.

Chris Adams: Okay.

Asim Hussain: Premature death, social cost of carbon, plus a dollar value on like the carbon emissions and displacement, which is kind of interesting as well.

Like how many people and all this stuff is like one of the, you have to submit like citations and research to prove and evidence kind of where these numbers are coming from. But yeah, you know, like social cost of carbon, we know like what is the cost of carbon as we take it down through the generations and, you know, all this other stuff

Chris Adams: So that's the harm caused by, on society, by the people, right? And like, this is what the, that was 50 under Barack Obama's time in America, went down to 1.

Asim Hussain: No, I think it's two.

Chris Adams: or 2, and then that's gone up to 190 now. No, I think it's actually back to 50. But there, I think the, it's now shifted to a very, a much, Like, fourfold increase, or nearly a fourfold increase in the US, for example.

But this is like, and in Europe it's around 100 right now, that's the figure.

Asim Hussain: I don't, yeah. I don't believe they use what Biden decided and wrote on an exec. It's it was more, more from research.

Chris Adams: Oh, I'm not saying that's like, that's not Biden saying this, that's like, the group who was allocated to do this work, these are the things they've recommended for this. And that's how they've, that's what the number looks like is being mentioned. I'll share a link to that as well, actually, because that's actually This is, really helpful for the internal carbon pricing stuff we discussed before.

If you're going to think about, okay, how do I create a carbon war chest to fund the work like Will Alpine did, or founding an organization, or doing any kind of systemic work. Yes.

Asim Hussain: Yeah, but I think that's kind of like where it's interesting. cause you know, sometimes I do feel that. We sanitize this conversation too much. Like it's too sanitized. Like we talk about carbon, we talk about this, talk about that. But actually, the reason why we're here is this is there's a lot of human suffering that is happening and will increasingly happen because of the work that we're doing and the work. There is, that value in this whole, whole question.

Are you trying to like add a cost to carbon? How many people will likely be displaced because of rising sea levels

Chris Adams: Always like climate migration being informed.

Asim Hussain: Climate migration. Yeah. so like, and of course there's also research which talks about what is the increased death rate, you know, You have to look at kind of, not just carbon emissions, but air quality and things like that.

So there's these, so these are the impacts. I mean, I, and I know it's kind of, it can be hard for some people to swallow, but it is, you know, something I think about a lot. I actually have a big skull next to me. It's like, I'm very stoic.

Chris Adams: You are full of surprises. Wow, you've got kind of Shakespearean kind of Wow.

Asim Hussain: A little

Chris Adams: Alas, poor Yorick. Wow. I did not see that coming at all.

Asim Hussain: Well, you, it's a stoicism thing.

It's like, if you, contemplate death a lot, you, live life more. But those are two, two, two, two really good plugins there. But there's just been, it's hard to like really pick up on the rest. There's some really good work done to using Impact Framework, not as like, not building a plugin for it, but just leveraging it to examine what the emissions of machine learning is.

And the important thing about that is, is the output of it is this kind of very standardized manifest file, which anybody can read. You're

Chris Adams: Like an auditable or repeatable

Asim Hussain: audible, readable, and you're just reading this manifest file. And as soon as you look at this manifest file, if you know now how to read these manifest files, you can go, Okay.

So, straight away, I know you did this, you used that coefficient. I get the kind of measurement that you are. I can, that's an interesting number you generated. So that's one of the things we want to drive from it. And the best, yeah, there's some great work from, I've forgotten the name of the team now.

But the, they did work on the, trying to figure out the, using impact frame in a very In a usual way, which is like in logistics, can you drive, can you create a manifest file which represents the emissions of like a fleet of cars, like delivering Amazon packages all over? Can there be a manifest file for that?

And, uh, I think the winner was they built, it's called green. Oh, no, I want to mention two other teams and they both got the same name, which is very confusing in the hackathon. Find a called green and meet green. Oh, I forgot the actual name of the team. It was green something, but they, uh, they built Kubernetes.

So an interface into Kubernetes and the reason why it won, it was just like Kubernetes is so ubiquitous on the internet. And now if you've got, and it wasn't using, there was a separate one for Kepler, but this is very specifically for Kubernetes, independence of Kepler, taking the observations from there.

And you can compute the, environmental impact of that. And I think, actually, I shouldn't have mentioned it, that my favorite solution was a team called Greener Meet. My absolute, I all, I'm sorry, all of the rest of you, you all weren't my favorite solution. My favorite solution was a team called Greener Meet.

Chris Adams: Green Me, as in, like, Greener You, Greener Me, or Greenami, like Konami?

Asim Hussain: Greener Meet, as in having a meeting.

Chris Adams: Oh, Greener Meet, as in zoom, not like.

Asim Hussain: Zoom. And it was exactly

Chris Adams: Okay.

Asim Hussain: And we had an under 18s prize, which we really spent a long time with lawyers trying to figure out, can we do an under 18s prize? But we managed to get one together. And it was a team of 16 year olds from school in California.

And they created, they used the Impact Framework to actually measure The environmental impact of Zoom. So they literally like they, they got Python libraries. They, ran Zoom calls. They could get the utilization. They compared it to Watttime and they even found they were like, they even switched on and off features of Zoom to show like how that changed the emissions profile of your call.

And even I loved it. They found a curtailment that they were having a meeting and cause they're in CAISO. They

Chris Adams: CAISO being Californian Independent Systems Operator.

Asim Hussain: I knew you'd know that.

Chris Adams: And that's basically the grid operators in California you're talking about.

Asim Hussain: And California has a lot of renewables, as you know, so they had a moment in the meeting where they curtailed, which means that the emissions are effectively zero for the call now because they're factoring that in.

It was really good work, and we tried to give them the prize on the call. They, we were delayed. They had to go to class. So we couldn't give the kids the prize, and if we can, maybe we can bring them on this podcast. I'd love to speak to them. I just, it gave me such, honestly, you watch this, And I'm sorry to the rest of the prize winners, but I think this was the best video.

It was the best presented. And I'm just sitting there thinking, these are the, now we're going to, now we're going to, " the kids of today!" - we are old. These are the kids of the future. You know, maybe it gave me so much hope, like these are 16 year old kids and they were doing some incredibly advanced green software, measurement, reporting, zoom, understanding curtailment, understanding like how to measure this stuff.

And I was like, ooh, we've got a, we've got a good future ahead of us. I think, I don't think, yeah.

Chris Adams: Asim, this sounds really cool, but I have to ask, right, why are we relying on a bunch of literal children to do this work when Zoom is a multi billion dollar publicly traded company that does have to report on its own figures like this, and they are doing, Zoom doesn't expose these numbers themselves, right?

Like they will tell you how much you've saved, but you can't see the figures from this. We're relying on children to do the work of essentially publicly traded companies here, right? Right. This feels like a, we, I think this is really, cool. But this has an open question about what they would do maybe Zoom or they might join an organi organization like the GSF and say, well, this is how we do measure this stuff. And, uh, 'cause if not, I think we might need to refer to the work of children until that happens. Because this is just like, I mean, if, this embarrasses them to actually being more transparent about how this is being used, I think, we should be doing it because it's kind of bonkers that you are relying, like, it's nice to have like youth of today, but one of these groups has literally access to billions of dollars of capital and one of these groups is doing it in their spare time and couldn't even accept the thing because they had to leave for class. Yeah.

Asim Hussain: For class. Yeah. Yeah. I mean, that's a fair point. But at the same time, I mean, I want to counter your point for a second as well. I mean, I'm, you know, I have multiple views. Like, do you really want to be waiting all the time? Like, we're just waiting all the time. We're just sitting and waiting and waiting and waiting, for everybody to be transparent about all of their emissions all the time.

If we can just take that power back into our hands, And we why not wait for people to be transparent, but actually like, look, this is a model that I've created. I think this is the emissions profile of your product. I'm going to publish it. If you feel somewhat different, that's the beauty of open source.

If you feel somewhat different, create a pull request and be, as transparent as me. I think in a way we, it is good. That we're leading the way, it is good that the kids are leading that way in the future because it kind of brings power back into our hands, if that makes sense.

Chris Adams: I think what it sounds like you're suggesting is in the absence of information, it might be worth using the age old internet trick of ask a question on Stack Overflow, answer it with an incorrect answer, then hope that people will actually come up with better answers, because that's the tried and tested way to get a good answer on Stack Overflow.

You answer it incorrectly, then someone goes, no, you're getting it wrong. Now I'm doing it, because they're so incensed that someone is wrong on the internet. So this basically, I think what we can use that, we can harness this natural resource and we can say, we're only going to use the worker of these children, and we're for talking about Zoom calls until someone comes in with actual better figures and ideally the organizations with full access to this, like maybe Zoom themselves, or maybe a competitor of Zoom.

So if you'd like to do this, I'm [email protected], no, I'm very happy to have this conversation because this does feel like it's necessary. Anyway, oh, blimey, I've just seen the time. Asim, we've, gone a bit over, but this has been really, fun and,

Asim Hussain: Yeah. It has been!

Chris Adams: I, this is, We've got another one of these in a few weeks time.

So, Asim, really lovely chatting with you, nice to catch up, and sorry to hear about the mushrooms, but I guess the Mushroom Kingdom's loss is the Bread Kingdom's gain. So, I should probably wrap up now. Folks, if you've made it this far, we will be sharing links Pretty much everything we've discussed and which you can find at podcast.greensoftware.foundation. And thanks again for listening. My name is Chris Adams and this is Asim Hussain of the Green Web Foundation and Green Software Foundation respectively. All right. Thanks, Asim, mate. Take care of yourself. Okay. Bye.

Asim Hussain: Cheers. See you later.

Chris Adams: Hey, everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again, and see you in the next episode!

The Week in Green Software: Building Green Software 2

30 maj 2024 | 49 min

In this episode of the TWiGS, host Chris Adams welcomes back Anne Currie, Sara Bergman, and Sarah Hsu, authors of the book Building Green Software. They dive into the latest updates and hot topics at the intersection of sustainability and software engineering. The discussion highlights the importance of making software and hardware more efficient and explores cutting-edge topics like serverless computing on Kubernetes with WebAssembly, the circular economy for electronics, and the potential for dynamic pricing in cloud services based on renewable energy availability. The episode emphasizes the ongoing energy transition and the need for innovative solutions to reduce the carbon footprint of technology.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Anne Currie: LinkedIn | Website
Sarah Hsu: LinkedIn
Sara Bergman: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

https://ndcoslo.com/ [47:55]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Anne Currie: That is the nature of renewables, is that they are variably available, and we just have to, we have to take advantage of that, not fight against it, and not constantly wish we were in the fossil fuel age.

our roundup of what's happening and hot at the intersection of sustainability and software engineering. So Anne, Sara, Sarah, I know that you've been on the podcast before, but I wanted to just provide a bit of space for people who are new to this to let you introduce yourselves. Anne, it's okay if I give this floor to you before we run through the usual roster?

Anne Currie: Yeah, thanks Chris. So, my name is Anne Currie. I am the CEO of a training company, Strategically Green, and I have been in the tech industry for pretty much 30 years now, and I am one of the co-chairs of the Green Software Foundation Community Group.

Chris Adams: Cool. Thank you, Anne. All right. Sara, is it okay if I hand over the floor to you next?

Sara Bergman: Sure. Thanks for having me back. I always enjoy being on the podcast. So my name is Sara Bergman. I am a senior software engineer at Microsoft, where I work with the Microsoft 365 products, which is very fun and exciting. I'm also one of the co-authors of the book Building Green Software. And other things that are new in my life is I'm recently back at work after maternity leave.

So I'm still figuring it out, you know, what is life now?

Chris Adams: Well, welcome back and congratulations on, yeah, the new instances of Sara, I suppose. Yeah.

Okay. And moving on to other Sarah. Sarah Hsu, if I give space to you to introduce yourself.

Sarah Hsu: Hi everyone. It's nice to be back. So my name is LSara Hsu. I am an SRE working for a financial institution. I'm also a project lead for the green software course for the GSF and similar to Anne and Sara, we recently just published a book by O'Reilly called Building Green Software. Very excited to be back.

Chris Adams: Cool. Nice to see you again. And Sarah Hsu, we met in person for the first time when you were in Berlin delivering a keynote for, was it CamudaCon, the conference there?

Sarah Hsu: Yes. A Process Orchestration Conference, which is exactly what we need, right? Because their motto is automation. And automation is the foundation of modern software systems. So...

Chris Adams: thank you for that, Sara. I was also pleased for us to both realize that we had arms and legs and were more than just a YouTube square, basically, or a square in a Zoom call. So yes, that was lovely as well. Okay, for folks, if you've never been listening to this podcast before, my name is Chris Adams. I am the executive director of the Green Web Foundation.

It's a Dutch nonprofit focused on reaching a fossil-free internet by 2030. I also work as one of the policy chairs of the Green Software Foundation Policy working group as well. And before we dive into some of the stories, that's the format of this show, we'll be sharing links to the show notes and all the stories that we do discuss.

So there'll be on GitHub and Markdown, what else, for you to look through. Alright, so the format we normally do is do a round up of news stories, but you three, because I have the pleasure of speaking to people who've written a book about green software, we figured we'd make it a little bit more interesting for listeners and touch on some of the topics in the book, so we've got a bit of a kind of nice way in to cover some of that content. Does that sound okay to you folks?

Anne Currie: Great.

Chris Adams: Alright then, so the first story that we have is actually about a new project called SpinKube. So this is hyper efficient serverless on Kubernetes powered by WebAssembly. Now a few podcasts ago before, we did a whole story all about Wasm, WebAssembly, and why it's an interesting piece of software. But, previously, back then, it was only available on Nomad, which is a similar scheduling tool, but not the same as Kubernetes. And this story, as I understand it, is basically the idea of providing some of these kinds of tools for Kubernetes, the most popular scheduling tool for this. This basically means that if you thought that was a cool idea, you've got access to it yourselves. And Anne, I think I might hand over to you for this, because this really touches on some of the things you spoke about in your book about operational efficiency versus coding efficiency. Maybe I will hand over to you here.

Anne Currie: It really does. So Wasm is something that I find absolutely fascinating. It's both, as you say, both from, at the moment, the operational efficiency perspective, because they're really focused on how you can bin pack as much work onto each machine as possible. So you can reach that magic 50 percent utilization number and then get even higher, which is very hard to do without a tool to help you.

Very hard to do. But beyond that, what also is, is they have ambitions even further, which is that they want to start actually optimizing the code that you write in WebAssembly as well, so that under your feet, it will become more efficient. And we all know that writing efficient code is really, really hard.

There's a, there's a huge impact on developer productivity. And we, we cover this quite a lot in the operational efficiency and the code efficiency chapters of our book. And one of the senior engineers, Danielle Lancashire at Fermion, who's behind all this Wasm, one of the, one of the groups behind all this movement in Wasm came and talked to me, talked for me at a conference in London a few months ago, and she was actually talking about code efficiency, the code efficiency improvements that come from Wasm.

Now they're a bit down the road. The moment they're focusing on operational efficiency, but I really like to see a platform with a vision for, because in the end, we're all going to have to run on much, much less power at times when the sun isn't shining and the wind isn't blowing. That is the nature of renewables, is that they are variably available and we just have to, we have to take advantage of that, not fight against it and not constantly wish we were in the fossil fuel age.

And so I really like to see a platform which is thinking about operational efficiency now and also thinking about code efficiency down the road. So I was, I like Wasm from that perspective. I think they're doing a lot of nice work.

Chris Adams: Yeah, we have some nods here. I just realized when I spoke about this story, I dived straight in without really explaining why some people find this actually exciting in the first place. So when we spoke about things like, say, serverless, and one of the, as I understand it, one of the key things behind this tool is it's like a very fast version of a serverless platform that spins up and down quite quickly.

So we've had previous generations where you might need to have a bit of a wait before a piece of software can start running before you can really use it. And this is like one of the key things that Wasm has made available. And that allows for, like you said, time, it basically means that there's less wasted time, but it also provides, opens the door for newer, more efficient platforms like this. Okay, before we move on to the next story, Sarah, Sara, is there anything that you want to add to this?

Anne Currie: Actually, before we move on, I'm going to just hark back to something that Sarah said just before this, which is that the benefits of modern ops and automation is that a big part of automation, of modern automation is the ability to spin up quickly because things like auto replacing, just having stuff sat around waiting and for you to fail over to with autoscaling, which is vastly more operationally efficient, relies on things being able to spin up fast.

If you can't have fast instantiation, you can't have a lot of these modern automation, this modern progress in automation. So yeah, it is a really good thing.

Sara Bergman: Also, I think one thing that has been said before and can always be said again is the benefit of the sort of platform that makes it easier for people because people want to be green, but sometimes it's a lot of work and anything that can make it less work to be greener is a great thing and should be celebrated.

And on the ops side of things, not every software developer or software person is like, highly interested in that. There are some people who love it and like go all in, and there's some who like, "I just want to write my code and like deploy it in some way to my users." So therefore things like this is so important to, to help bridge that gap in a green way.

Sarah Hsu: Yeah, yeah. I a hundred percent agree. Like software engineers, we're inherently lazy people, right? Like none of like, not speaking for you guys, I'm not a security expert, but I know how important a secure application is, right? I'm waiting for this tooling framework and best practice from the security people. So I think that's the gap we really trying to fill, that "how do we make everyone else have that ability to be green at their fingertip without having all the knowledge of like being really green?" And like, yeah, I feel like we can talk about operation efficiency to... I mean, to, 'til tomorrow, yeah, because it really is the lowest hanging fruit and people don't realize how many things we're already doing or have like knock on green benefits, like exactly what Anne said about reliability and resiliency.

Yeah. They all like come down to like automation and how do you utilize automation? Right. Anyway, I think we should stop. We can always come back to it. Let's get through all the other stuff. And then we come back to talk about Green Ops.

Chris Adams: Okay, all duly noted. All right, so we've spoken a little bit about operational efficiency, which is the running of servers. Spoke a bit about coding efficiency, which is what a lot of people might reach for first, or people wouldn't typically think about. And, but there's other ways you might talk about this.

So let's move to the next story. Sara, I might actually ask you about this one. So this one is a story from Grist magazine, which basically is talking, it covers the staggering quantities of like, Transition metals that we're currently throwing out when we could be mining them. And while we'll share a link to this in the show notes, one of the key things that the four of us can see right now is a chart showing some of the minerals that we hear associated with a transition away from fossil fuels to greener forms of power. There was a report by the UN called the Global E-Waste Monitor and they've shared some of these stats for the first time. And one of the key things is basically that we are throwing out something in the region of 62 million tons of electronics. And when you look at the actual mineral content of that, in some ways it's actually comparable to the demand for new kinds of metals that we have.

So the charts that we'll link to here, things like, say, copper, it's not that, it's quite similar to each other. Some things like cobalt or neodymium. These are, we're basically looking at the amount of minerals that could be circular. Some cases are not the same. So things like lithium, for example, we still, there's still a lot of demand and there's not nearly enough that's in circulation. But this feels like this provides an interesting flip side to the whole discussion around what we do with our stuff. And I think the term that I saw in the book, which was hardware efficiency. So I kind of wanted to like, see what you folks thought about this, in particular Sara, because this is one thing we spoke about in a previous podcast, like, this feels like there's more than just "hold on to your kit," for example, there's maybe a chance to talk about things in a more circular fashion.

Sara Bergman: Absolutely. And I think the key word there is circularity. E-waste is the fastest growing waste stream in the world, which is saying a lot. And it is growing for several reasons. Like we consume more, we have more devices, like more devices are smart devices. My, like kitchen fan has an app. I don't know why it needs an app, but you know, there, there are an increasing amount of what can be classified as electronic, not what can be classified, but what is electronic.

I see the same with like toys and things. So that's one thing. And also we crave the newer and the newer things and things have shorter and shorter Life cycles or the use phase is short and shorter. So it's, it is fast growing and before, in a time before time...

no, but when it was not the case that it was so fast at growing, we could maybe afford to mine everything.

But now when the consumption is so high and the throwout rate is so high, we really need to start thinking in a circular way. Because, in the end, there is only so much metal in the ground and our earth is finite. It's not something that grows back, at least not very quickly. So having this thinking, I think it's going to be, it needs to be a game changer.

And I know several countries have adopted what's called the right to repair, which is basically that you

Chris Adams: is big thing in Europe.

Sara Bergman: Yeah, it's a big thing in Europe. China has, no sorry, not China, India has a similar legislation. New York has a similar legislation. And that's really good. And that's about what you said, holding onto your kit longer.

But that's really only one side of it. I think the right to recycle should be equally hotly debated, perhaps even more hotly debated. Because we, the hardware industry has been forced to make hardware kind of hard to recycle because we are demanding better, yet smaller devices, or bigger screens, but thinner, lighter hardware.

And that's really complicated to produce. And it makes it really hard to recycle. Also in this article that you linked, let me find the number. But, what they said, the percentage of ethically recycled waste was also staggeringly low. It's like a business as usual case was only 20%. It's, well, you're all going to have to go to the show notes the article for yourself to find the percentage.

What I'm meaning to say is that this is incredibly important and something that is an additional dimension to the conversation beyond holding on to your hardware. It's also the reuse and recycle part of the story.

Anne Currie: It's a sales feature as well. It's recycling and it's a sales feature. I just bought a new phone, new, it was an old Pixel 7a, because my previous one had run out of security patches, the classic way that you have to throw away a working piece of kit. So I was quite annoyed because I was having to throw away a perfectly working piece of kit.

So I bought 7a. But the 7a sent me all the stuff to do a, to send my old phone back, and I got 25 percent off and I thought, "well, actually, that was pretty good." I actually felt quite good about that in the end. I thought, well, it's going to be recycled. I got 25 percent off. All right. That was, it was, it felt like a feature that they were offering me.

Chris Adams: Well, this is, maybe this touches one of the key ideas that, if you're moving away from linear economy, once you've, dug these minerals up once. One of the ideas is that once they're in the kind of sphere that we are in, right, you don't need to have, it's not that going into the atmosphere like say fossil fuels for example.

You can have these things circular. Now, this is one thing that we don't have quite the policy support for yet, but this is one thing that we could definitely be seeing more of in future, and this feels like the direction we might be heading towards if things work out, basically.

Sarah Hsu: Yeah. I also think it's an age old thing, right? We should find the thing that fits your purpose. It's similar to say, we want to find the VM instance that fits my purpose. I feel like people shouldn't want the latest and the greatest. Like my laptop is not that great, but I don't do that much on my laptop.

So like, I think people also need to start having that mindset. Like, "Oh, I don't need like the fastest, like the most, you know, cores, power laptop, choose to go on Zoom call, do a little bit of coding." And yeah, I think that's another bit of the things that also really fit into operational efficiency, you know, where you want to find things that fits your purpose.

And I think that applies to hardware as well.

Chris Adams: I agree, actually. All right. Thank you for that, Sara. All right. So, we're going to move to the next story now. This is a story from Carbon Brief. And, Anne, I'm going to hand over to you for this one because this is one thing we discussed before. So, this is a story from Carbon Brief who basically mentioned that earlier on this month, no, last month, actually, now.

The UK's electricity grid operated for a whole hour with just two, almost no power coming from fossil fuels. Now this is a record low for the country and this hopefully is a sign of things to come. So Anne, I know that we discussed this here, but maybe you could talk, touch on like why you found this interesting and what it's kind of telling us as software engineers perhaps.

Anne Currie: Yeah, I mean, obviously this is an incredibly positive story, and it's actually the direction that almost everybody's going in. If you haven't had a play around with electricitymaps.com to have a look at how green all the grids of the world are, it's absolutely fascinating. You'll learn a lot. And if you step back in time, it shows you what the carbon intensity, the average carbon intensity is of every grid in the world that they can get data on, which is a surprisingly large number of grids.

You'll see that if you, and then you can go back in time and step forward and you'll see that everything is becoming greener and greener. The energy transition is happening. You know, it's not just something that will be happening in the future. It is happening, but it won't be easy. Oh, I saw an excellent Uruguay is apparently completely a hundred percent green for eight solid months this, this year.

So there are, but that is because, but the interesting thing on this is that every country does it in a different way and every country being the green intensity of their grid, it varies over time in different ways. It depends what you're using to generate the power. So places like Uruguay have loads of solar and wind, which is fantastic, but they also have tons and tons of hydro and the hydro is used to smooth out the times when the wind isn't blowing and the sun isn't shining.

In the UK, we don't have so much in the way of stored, so it's great if it's windy, and if it's sunny, but mostly in the UK, so if it's windy, we do an awful lot of offshore wind, rather, not onshore wind, but offshore wind.

Sara Bergman: I thought going to say, we don't do a lot of sun here.

Anne Currie: We do not do a lot of sunny here. It's actually, it's not as bad as, I've got solar panels, and through the summer, it's actually, well, through half the year, that will cover all my household usage.

It's oddly much better than you'd think, but nonetheless, in the UK, it's mostly wind. But there's not so much in the way of storage. So it's, the UK is an absolutely excellent example of we're going to have to get good at using power that's variably available and cheap when it's available and either demand shift or demand shape so that we are using the renewable power and using less of non renewable stuff.

So it is absolutely fascinating. It's a great story. It's a really good move forward, but not all grids are identically green. They all have difference. Different ways of doing it and results.

Chris Adams: Alright, I've got a pop quiz for you then, because you're talking about the fact that the grids change, and sometimes they're going to be greener because there's more clean generation on the grid versus other times, right? We know in the UK, for example, there's groups like Octopus Energy, who basically will change the cost of your power, depending on the time of day.

In some cases, they even pay you to use this. And we see this in other parts of the world. When do you reckon we're going to start seeing cloud providers do this in the net? Because we know it happens, and we see this stuff, but none of these signals are passed on to software engineers yet.

Anne Currie: No, and every time I'm speaking to somebody from a cloud provider, I ask about it. Normally the people are quite green people and they want it to happen, but I think it'll be a way off. Now, it's, it's interesting. I spoke to somebody who was working for a cloud provider, but was a software engineer who used to be in the insurance industry.

And he was saying to me, "Oh, well, you know, I can totally see it being to the advantage of the cloud providers to start doing dynamic tariffs, time of use tariffs, because then they have another product they can sell." So the people who can't do it, they will sell them insurance with a price cap. And obviously they'll charge a load of money for that.

If you know, but, and then that is a product. Now, well, is that what they're gonna do? They'd be quite sensible to do it. It's, you know, the cloud providers are very good at making money and it won't be cheap. So I would love to see it, but I think it's going to be years before it happens.

Chris Adams: Okay, so you're thinking three years at least for you, Anne, yeah? Okay. Any other takers for anything faster?

Sara Bergman: Yeah, no, I think, well, working for a cloud provider, I just want to say, I don't know. But what I think all, at least the major cloud providers are fairly open with their use of PPAs, so Power Purchase Agreements, as a way of, yeah, meeting the green energy needs, because they have data centers sort of where they need to have data centers and then use PPAs to, to handle their Scope 2 emissions.

And I think because all of them at least are so seemingly tied to those agreements, it is a contradiction between those and billing a customer for actual usage. So I think it could be like a complicated thing for them is what I suspect. But I also think it's something that customers would really appreciate.

So I'm hoping it will be sooner, but no, I'm with Anne.

Chris Adams: Okay. Years away. And Sarah, you've got something to say, it looks like.

Sarah Hsu: Oh, I was just going to say, probably not in the near future, but we just need one person, one cloud provider to do it and everyone else be like, "right, we need to do it too."

Chris Adams: Do you know what? I had a conversation from people outside of the big three, the big cloud providers. So, in Texas, there's a company called Build AI. They've been doing some work to basically, they'll provide you computing, but at certain peak times, you don't have access to it. And as a result, they're able to have much lower costs for this stuff.

So I'll share a link to that. And there's a company called Saluna, also in Texas as well. And what they do is they speak to generators, people who run like clean energy. And they'll basically say, we will give you a floor price under which you will never, you know, we'll always get give you something like that.

And using that, they are able to provide these kind of services. So we are seeing this start to develop, just not with the big providers. And we'll, I guess the next question will be, at what point do these new providers get bought by the big providers to protect the margins? Because that might be the logical thing to do if you have these kind of funds. All right, let's move on from that, because that was a, we spent a bunch of time talking about carbon awareness, and there's a few other stories that we have up here.

Anne Currie: But carbon awareness is the most important thing.

Chris Adams: It's definitely

Sara Bergman: And operational efficiency.

Chris Adams: Alright, ok, so let's move to the next story then. So, we have another story here, this is actually probably the nerdiest story we have here. This one is actually a link to an issue in the OpenTelemetry repository. Basically, there is a standard called OpenTelemetry which is designed to make it easy to understand what the, I guess it's to make Infrastructure Observable, and Sara, I might need some help from you on why observability is important.

But this one is basically put forward by, I think, one of the people who's inside the Green Software Foundation to start agreeing some sustainability metrics to expose in all the kind of tooling that we currently do have. This was really interesting because this feels like, A, this is something that I saw discussed in the book, but also for people who are not familiar with OpenTelemetry, OpenOps, or even Observability in general, is anyone who might want to go, like, enlighten us or at least give us some points about why this might be interesting from a Green Software perspective? Sarah, I might hand over to you because you wrote part of this chapter for the book, I believe.

Sarah Hsu: Yes. So I guess observability was born out of necessity because things are so complicated now. Microservices is made out of our world. And sometimes one requests have to travel like the entirety of a street of like a hundred houses before they actually reach a part of their journey. And it's really impossible to figure out what and where things have gone wrong, right? Metrics is for when you know something is going to go wrong and then you set up a metric to monitor that. But then in this unknown world, it's really hard to figure out who is going to break, for example, like, oh, I can't, I forgot that guy's name, but someone from Honeycomb.

You guys know how much I love Honeycomb. He gave a really good example. Like if you're like a, like an iOS developer and you support like 10 different phones in 10 different countries, and suddenly one version of the phone in this country is going to break. How do you know to monitor that using metric?

So I think that's a really good example, like why we need observability and observability borrowed its thinking from control theory, which is like, we are trying to understand the internal state of a system by just looking at its outputs. And outputs here are like telemetry. And telemetry, you've got logging, you've got metrics, you've got traces.

And so basically OpenTelemetry is, it's like a set of framework. It's about the creation and then the management. All of those telemetry, it's actually not a backend. So it's not like Prometheus. It's not Jaeger. It's not like Grafana cloud. It's just a convention, which is really good. I remember when we were all in QCon a few years ago and Daniel was talking, Daniel from Skyscanner was talking about their OpenTelemetry migration plan or something.

They basically went from like 300 different components down to 150 because they used OpenTelemetry as a standard, as an auto collector. And then that's the way, become one stop shop for all the telemetry. It's like, you don't need developers sending three different telemetries to three different backend systems.

And then one big thing, everyone's being on call here. One big thing we find difficult is how do I context link everything? Why I need to basically, "oh, this logs happen at 1:21 PM. Right. Let me go find the traces that also happen. But what if time shift, right?" And sometimes like Something is in a different time zone.

So anyway, that's a massive rant about why we need observability. So it's, I know, sorry, it's my job guys. And then I guess one big thing we talk about in the book is that green software needs to be ready with observability. Like we need to be with it. Right. So for example, in this complicated microservices world, we want to be able to know which component, which process.

It's emitting the most carbon. Right. And that's where we want to be. And we need OpenTelemetry. We need people like OpenTelemetry to help us get there. So it's like absolutely amazing to see so many people are standing up and then it's like, Hey, we should add this like semantic thing.

Chris Adams: I'm really excited about seeing this because I've used Honeycomb to understand what was broken about some applications I've been running before. And I was always a little bit wary about saying, well, okay, I'm not sure about, I want to be totally tied to one provider. And this here seems, and we have seen some providers who have started to make some CO2 figures for this. There's, confusingly, a, so a company called Sentry that do provide this and they even propose like a HTTP header for CO2 per request, right? But to see this at a kind of standard level, this feels like it might make it easier for a larger set of providers to come up with and at least make it easier to kind of see some of this because I think this is something that came out the book was that we, you need to be, you need to observe this, but it's often quite difficult to get the underlying numbers from some providers, and this is something that we need a bit, we need some more work with, or we need some progress on, basically.

Sarah Hsu: Yeah. And I guess one big thing about OpenTelemetry is that it, because it is just a framework and it's vendor neutral, I think sometimes people forget how important staying vendor neutral is. So yeah, I think that's why it's so important that we locked in with OpenTelemetry now, because it is going to be the solutions for this observability space going forward.

Sara Bergman: Absolutely agree, because I think sometimes when people talk about software, they think of a specific type of software that runs in the public cloud. But that is not the entirety of software that exists out there. There's so much software who runs on different places and nowhere near the cloud. And that software is equally, equally important.

So, having something that fits more than just the one most popular scenario is incredibly important. And, and I just want to say that, I think this is important because I don't think anyone should get away, quote unquote, from doing sustainability work because, oh, I don't know my number. That should not be an excuse.

Anne Currie: No, I totally agree. But, but even if it's impossible to get your number, there's still so much you can do without the numbers. as well. And yeah, it's amazing how people go, "Oh, I can't get the numbers to get" well, just work on your operational efficiency.

Sarah Hsu: Yeah. Yeah. And like one of the biggest takeaway I gave at the CamundaCon in Berlin was that you can think of BinOps as a really natural evolution of DevOps and FinOps. FinOps is the optimization with money. We basically need to do the same for sustainability. And there are so many things we already can do and yeah, people should really pack themselves on their back because they didn't realize the transition is going to be much smoother than they thought.

Chris Adams: Okay, I'm glad you mentioned FinOps actually, Sarah, because this is talking about OpenTelemetry, some ways to expose some of the figures into this. As I understand it, there is a is it focus, which is the standard that the FinOps groups are pushing for trying to come up with like standardized cost, cloud cost figures, because I understand, as I understand it, there are some people pushing to put some CO2 figures in those as well, so that you'd be able to get some of these ideas from not just billing, but also from operational figures.

So, because in some cases, one view will give you a slightly different view than the other, for example. We've got this, and we're just moving to the last of the stories that we have today. One of the largest providers has published their sustainability report. Microsoft published their 2024 environmental sustainability report in the last week or so, and there's a lot in it, actually.

So, they're one of the large providers, and they have various commitments about getting to net zero by a certain times, but there's actually quite an interesting amount of data for the nerds inside this. Sara, I might hand over to you, because I suspect that you've been poring over this in quite a lot of detail, actually.

Sara Bergman: Thank you. Yeah, I think I love it when this report comes out every year, because even though I work for Microsoft, there's no way I can keep up with everything that's Like, it's just too big. So I always learn so much. But I think for me, I'm a measurement geek, in case that wasn't obvious already.

So some of the things I thought was super interesting was firstly, that the PUE of the data centers was published. I don't think Microsoft has ever published that number, but they said that this year, the design rating of the new data center is 1,12. And I think that's pretty impressive. That's pretty close to 1, which is, you know, the dream where all the energy that goes in goes to actual compute and to know other resources.

Chris Adams: I'm glad you mentioned this, because I wanted you to actually just, for folks who are not familiar with what PUE means, maybe you could just, like, expand on that a little bit, and say why people, why you might even care about that number being high or low, for example.

Sara Bergman: Yeah, absolutely. So P U E stands for power usage effectiveness. Or if, yeah, effectiveness. Yeah. I always mix up the E words. Power usage effectiveness. And I think in the sort of early days of green software, this was the number people spent a lot of time focusing on. It is a number that's higher than 1.

So it measures all the power that goes into your data centers and how much goes to actual compute. If, if all the energy or all the power that goes into your data center goes to meaningful compute operations, your PUE = 1. Only half of what goes into the data center goes to actual useful compute.

Then your PUE number is 2.

Chris Adams: I see.

Sara Bergman: There's been a race to get to one, where you waste basically as little power as possible.

Chris Adams: And when you say little power, you're talking about, like, keeping the machine from overheating and melting, or stuff like that, right? Like keeping cool enough?

Sara Bergman: Yeah, cooling is one thing that consumes a lot of extra power. It's probably the biggest consumption of power, but they're like, they're lighting in the data center also, that's also consumes power. So anything that's in there that doesn't contribute to compute, go into this rate.

Chris Adams: Do servers need to see? I mean, they don't need to be in

the, like, it's not like they have

Sara Bergman: there people who work there.

Chris Adams: Okay, that's fair enough. Okay, so this is, so in this report, they're publishing this information, which has traditionally been a thing that you do not see, all right?

That's one of the key things from this.

Sara Bergman: exactly. And I thought that was interesting. And in the sort of early days of green software, this was a number of people obsessed a bit about because I think there was like this, I don't want to say miscomprehension, but this notion, at least, that if you just got the PUE to 1, all the problems would be solved, like, magically, everything would be so efficient that we would not have to care about everything else.

People were, like, hyper focused on data center design. And I'm not saying it doesn't matter. Of course it matters. Of course it's important. But it's an area where, over time, we paid down massive knowledge and spend time on. And in the end, that's only one side, because if whatever compute operations you run are vastly inefficient in themselves, it doesn't matter if all the power going in goes to compute if the compute is wasteful in its matter.

So it is one part of the puzzle, but it is not the most important. But I'm glad to see it getting so close to one. I think it was very impressive.

Chris Adams: Yeah, this is interesting to see this because there's a project called Realtime Cloud which is run inside the Green Software Foundation and one of the key things that people are looking to do is figure out the PUE for all the different regions from all the different cloud providers right now. And they've got a carbon intensity figures for each hour of compute. But now they're able to use, because there's this information published at a data center level or actually a region level, you've got a level of transparency that you don't see from the other two so far. I think. Google might be sharing some of this, but we still have a kind of like patchy spreadsheet listing this stuff.

What I'll do is I'll share a link to the spreadsheet for this, because this is one of the things that once you have this, that should allow you to start being able to kind of optimise what the carbon intensity and of the compute you're using based on these kinds of figures here. This doesn't touch everything though, and I wanted to just leave the floor open. And Sarah, is there anything that caught your eyes on this, because there is more to this report than just PUE, for example.

Sarah Hsu: Don't want to be the bearer of bad news, though, although I remember hearing this from Anne, so I'm just their messenger. I'm not the bearer. I remember you and Anne mentioned that because PUE has got a lot of branding around it at the moment, and people know that it's like the sort of like the efficiency factor for a data center.

So like people then don't want PUE to go up, even though they need cooling. So they end up using water to do the cooling. So they actually trying to get PUE to as little as possible, closer to one as possible as Sara mentioned, but then they compensate that with using water and water is another like sources that's just as sacred.

Is that how you say that word? Yeah. But anyway, I will like, and talk about this because I remember I heard this from you.

Anne Currie: Yeah, no, as you say, we do fetishize PUE a little bit. It's a measurement and everybody loves to meet it. But the other thing is it's not a carbon aware measurement because it's just a flat number. So it doesn't say, "Oh, well, actually, you know, this is what we did when the sun, when the carbon intensity of the grid was high.

And this is what we did when the carbon intensity of the grid was low." And actually you want, we need behavior to change. Between those two, two times. So it's nice, you know, it tells us something, but we need to make sure we, with all things, you know, it's context specific. We, it doesn't provide a lot of context and we might need more context.

Well, we will need more context in the future.

Chris Adams: Okay, so better than nothing. So two cheers rather than three cheers. So good to see this. But as Sara mentioned, I believe the technical term is impact transfer when you go from one factor like energy to impacting water, for example. And in some places, you may be, you may have data centers sited in regions of water stress or pulling power from like local aquifers and things like that. If that's used for drinking water or agriculture or things like that, that's not really ideal. But there's a big discussion about what happens with this water. In some cases, it might be, I think the term is withdrawn and waste and consumed. So you have water that might be taken in and some of it might be discharged, given back.

But if it's at a different temperature, that might not be ideal for fish. And stuff like that. Or if it's consumed and like just evaporated away, that's not ideal either. So there is a whole bunch here. And maybe this is one of the things we need, we should be hoping to see, or we should be looking to see more of. There's, okay, if we've got data center level PUEs, should we be thinking about water of usage effectiveness and other things like that?

Sarah Hsu: I think there's one, I don't know what to call them, but I know they're called Scaleway. And I think you did a thing for them that they, I think now they have like a landing page where you can see the PUE and the WUE. So I think people are start doing it, but again, like much smaller scale provider instead of like cloud providers,

Chris Adams: I'm really glad you mentioned this actually, Sarah. So, the thing you're referring to, we'll share a link to this. So, Scaleway, they're basically one of the largest cloud providers in France, and they also operate in a number of other countries, but they basically do have these dashboards. But the thing that's, the thing about these dashboards, they were initially created by by Facebook in 2014. So there is nothing stopping every data center exposing these numbers from a technical point of view. And like, if you're looking at a policy level, what they have there is totally something that could be done. It's something that we could be seeing if people chose to be disclosing some of that. And we might see some of this come out as a result of new laws that may be landing or that have already landed in some countries.

For example, Germany has this. But there are more on the horizon, actually. Anything to add from there, perhaps, Anne, you're about to just say something, yeah.

Anne Currie: I am. So, so Scaleway, I think it's kind of related to stuff we talked about earlier. I was on this podcast with Scaleway where I learned a lot of stuff about what they were up to. And there's some interesting things that they're doing totally away from PUE and WUE reporting. So, in Europe, there's a group called the SDIA, which is an acronym.

I can never remember what it stands for.

Chris Adams: Sustainable Digital Infrastructure Alliance,

Anne Currie: Excellent. And their raison d'etre is to try and get European data, non hypercloud data centers to start learning from the American hyperclouds and actually delivering some of those services. And one of the, one of the things they're pushing for is actually starting to orchestrate and offer services on your data centers that will provide operational efficiency and code efficiency in the same way that the hyperscalers do. And that generally, you know, your bog standard data centers don't. And that's a really, that's a really bad thing about bog standard data centers.

They don't do this because it's very hard to do. If you can get your data center to provide it, then that makes life much easier. And interesting, on the podcast that I was on with Scaleway, they had a, one of their engineers on who was working on their serverless offering. So a bit like Wasm we were talking about earlier, services that improve operational efficiency and code efficiency provided through your data center. That's what we want. That's where the cloud is good with being, is potentially green, but only if you adopt them. And all these things, if you just lift and shift into the cloud, if you lift and shift into good data centers like Scaleway, you don't get any benefit really, or don't get much benefit.

You need to be moving over to efficient services.

Sara Bergman: Yeah. And I think that was another thing that Microsoft mentioned in the sustainability report that they are working on several ways to increase resource utilization. And just to mention too, they have these power aware workload allocations, and they're also like smartly allocating CPU cores for internal workloads.

Because don't forget Microsoft is also a software provider, not just a cloud provider. And I thought that was really good that they called out because sometimes when we say things like this, yeah, the hyperscalers are good at this, people question it, like, "oh, is it really worth their time?" So I thought it was good to see them calling out specific things that they're doing that can also hopefully inspire others to do the same with their data centers.

Chris Adams: So here's one question I have there about this report. So, yes, it's very good that you see some transparency here. And yes, there are definitely cases where moving to the cloud can be more, can be greater. But we've seen, both Microsoft and Google and Amazon, their emissions climb. Year on year. And this year, we're seeing emissions 30 percent higher this year than last year, which is not going to make it easy to get to net zero.

And this is one of the key things that Microsoft themselves have been talking about, saying, yes, this is a real challenge for us, and a significant chunk of our emissions are in our supply chain. Scope 3 is the largest source. So this is one thing that I think that it's, it feels like when we're talking about this, green software doesn't really have that much to say about right now in terms of the actual creation of data centers. And this is something that I kind of want to open the floor up, like how do we actually deal with this fact that things can be more efficient, but still growing in absolute terms? Is this something that we can be doing or do we need to be having discussions about absolute resource limits, for example, and things like this, or how are we going to get here? I'm going to hand to Anne because you've got something, but you're about to say something I think.

Anne Currie: I am always about to say something. I think this is a really, the whole, the whole thing of, you know, the degrowth, people shouldn't be allowed to, we shouldn't ban people from doing things, it's a very unpopular argument and it is why people, you know, are not, even though everybody now believes in climate change, they don't want to have to give up all the things they want in life.

Now, efficiency can often really deliver you the same standard or less. I mean, it's not all, I mean, Jevons paradox, we all know about Jevons paradox. The Jevons paradox is you, you, things become more efficient and where there's untapped demand, where people really wanted to use them and they couldn't previously because they were too expensive and now they can.

That's a, that's actually, that's a big improvement in life. But it's not a guaranteed lock in. Everything that you make more efficient doesn't always result in overall there being more usage. So we use a lot less electricity to run household appliances now than we used to. They've become more efficient and there is only a certain number of times that you can wash the dishes with the dishwasher or wash, you know.

At some point, demand does become

Chris Adams: there's an upper limit to how often I want to vacuum my room, for example. Yeah,

Anne Currie: me, it's quite a low number of times I want to vacuum my room. I'm more worried about things like Bitcoin, where there is no upper limit. I mean, so what worries me about things like Bitcoin is there is no, literally no upper limit to how much you might want to do it.

There is no, there's no point at which you go, "Oh, do you know, I'm happy now, I've just done all the things I wanted it to do." So with those things, where there's no upper limit, I think. And they potentially don't have a great deal of benefit for most of society. We might want to go down the China route and start banning them.

But for things where eventually our upper limit will be reached and they are providing societal value, I'm minded to let them grow a bit, at least. I don't know. I don't know what everybody else thinks.

Chris Adams: No, this is actually, this is, you're right, this is a complicated question to deal with, and this is one of the things that we, one of the things that's probably bigger than green software that we need to be aware of when we talk about this, because if we make things more efficient, then we've got to figure out, okay, what's the flip side of this, like you mentioned with Jevon's paradox? We do have an episode with Vlad Kouraoume, where we dive deeply into this if you're interested in listening, and we'll share a link to that. There's a number of other things we might want to talk about for this, but what we'll do is we'll share a link to the report so that there's plenty for us to be discussing. All right, I think that takes us to the stories that we do actually have. We're coming up to time. Folks, I want to say thank you so much for kind of coming onto this. If people were curious about these terms that we've been using, like operation efficiency, coding efficiency, things like that, this is what's outlined in the book that you three have been working on.

Is that correct?

Anne Currie: It is indeed. Yes. Building Green Software, the new book from O'Reilly.

Chris Adams: Anne, there's one question I want to ask because we spoke about this a few months ago. So this is currently available right now, and you can get it via O'Reilly, but there was a discussion about this actually going. Available into the commons eventually. Is there one thing we could just touch on some of that, because this is a really cool thing about this that I was really excited to hear about when you first shared this with me, Anne.

Anne Currie: Like, I feel terrible about this because this is, we just haven't had time to do it. 'Cause we're crazily busy. We've got to take the code, but we, we do have a license. We negotiated, O'Reilly very kindly allowed us to make it available under the O'Reilly Creative Commons license, which is a, a kind of read only license, but totally that's, that's still absolutely fine.

That's great. And so we just need to do it, but we haven't got around to doing it yet. Cause we have to do some tidying up and actually publish the thing, but it will at some point be available under a Creative Commons license.

Chris Adams: Cool, alright, brilliant. Well, thank you for that, Anne. Well, folks, this takes us to time. If people do want to find out about, if they're interested in what you have to say, or what you've been looking at, folks, are there any maybe just do a quick whip round of where people should be looking. Anne, if I hand it over to you first, then to the Saras. Anne, if someone has listened to you and they want to find out more about what you're up to, is there a website you direct people to or a network or anything?

Anne Currie: If you want to find out more about me, then LinkedIn is where I tend to hang out these days, and I'm very happy to chat to people and answer questions. And/or our website strategically.green will give us an idea, and I do an awful lot of public training as well, so you can always sign up for that.

Chris Adams: Brilliant. Thank you for that, Anne. And I'm going to go for Sarah with an H, if that's okay. So where should people, if they want to learn about your things, or maybe hear about your talk, where should, where should they be looking?

Sarah Hsu: Similarly, LinkedIn, I call LinkedIn the grownup version of Instagram,

Chris Adams: Yeah.

Sarah Hsu: but yeah, LinkedIn and yeah, I do fair bit of public speaking. So if you're ever catch us or catch one of us, don't forget to come get the book signed and because it's such fun, who can get to collect all three signatures. Because we are never in the same place once.

Anne Currie: We're never in the same

Chris Adams: Okay, sara, and over to you. If people want to find out some of the work you're working on or things, where would you direct people's attention to?

Sara Bergman: Yeah, I'm also on grown up Instagram, aka LinkedIn,

Chris Adams: Yeah.

Sara Bergman: but follow me, don't send me a friend request because I'm terrible, or what's it called, a contact request maybe, because I've not been good at accepting or rejecting people, so now it's an uncomfortable list and I just ignore it, it's a red flag with me.

So yeah, you can follow me and that that's the best way probably. And yeah, I'm speaking at NDC Oslo in about, yeah, in June. So if you have the book, come and get it signed. It will be lovely to meet folks. I also have a few copies that I can hand out if anyone catches me. h

Chris Adams: Ohere., exciting. Thank you for sharing that little one. All right, then. Well, folks, it sounds like I guess we'll see if people want to follow what you're up to. Millennial Twitter, LinkedIn is the place to go to. All right. What we'll do, if you have any of this interesting, folks, we are going to, and if you're listening to this for the first time, we'll be sharing the show notes with all the links to the projects that we've mentioned here, along with some of the other episodes where we touch on some of the things like Jevon's paradox, or some of the finer points of serverless. All right, folks, this has been fun. Thank you so much for giving us the time and,yeah, have a lovely day, folks. See you around, all right?

Anne Currie: Cheerio everybody. Bye

Sarah Hsu: Bye guys. Lovely seeing you

Sara Bergman: Thank you. Bye.

Chris Adams: Hey everyone, thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundationon. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

Environment Variables Year Two Roundup

23 maj 2024 | 31 min

Join us for a special episode of Environment Variables as we celebrate over a year two years of bringing you the best insights on Green Software! In this episode, we explore the key insights and voices that have contributed to the weaving of sustainability through our conversations this year. Tune in for a refresher on the most interesting discussions on the progress, challenges, and future of green software development.

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Episodes:

The Week in Green Software: AWS & Scope 3 Emissions Data | Ep 27
Fact Check: Colleen Josephson, Miguel Ponce de Leon & AI Optimization of the Environmental Impact of Software | Ep 29
The State of Green Software Survey with Tamara Kneese | Ep32
We Answer Your Questions Part 2 | Ep 39
Sci-Fi Fantasies with Anne Currie and Jo-Lindsay Walton | Ep 42
The Week in Green Software: New Research Horizons | Ep 47
Decarbonize Software 2023: Recap | Ep 53
The Week in Green Software: Google, Grids & Green Software | Ep 55
BETA Impact Framework | Ep 58
AI Legislation | Ep 63

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Chris Skipper: Hello and welcome to this special Year 2 Roundup episode of Environment Variables. I'm Chris Skipper, the producer behind the scenes. As we mark the second anniversary of this podcast, it's a perfect moment to reflect on the journey we've undertaken with the Green Software Foundation over the last year, and how that has been encapsulated through our episodes.

From its inception, Environment Variables has aimed to be more than just a podcast. It's a platform for advocacy and education on sustainable software practices. Over the past year, We've seen the Green Software Foundation grow and evolve, and we've been right there to document and discuss each milestone.

This podcast has not only followed the foundation's developments, but also mirrored the broader shifts towards sustainability and tech, bringing these insights right to your ears. Today, rather than revisiting our top episodes, we will explore how the themes of sustainability have woven through our discussions.

Highlighting key insights and the impactful voices that have contributed to this dialogue. You'll hear about the progress, the challenges, and what lies ahead for green software development. For your listening pleasure, as always, links to each of the episodes will be down in the show notes below. Or, if you want to listen to all of the episodes of Environment Variables, you can head to podcast.greensoftware.foundation, preferably after this episode, please, to hear them all. So, without further ado, let's dive into the collective journey of the past year with Environment Variables.

Chris Skipper: To kick us off, let's start off with an episode of one of our segments entitled the week in green software, or as we like to call it twigs,

This segment delivers a concentrated blast of the latest in green software news, keeping our listeners informed and engaged with current trends and advancements in the field.

In fact, it's a nice touch that this episode covers AWS and Scope3 emissions data, as the very first episode of Environment Variables way back in April of 2022 also covered Amazon's Customer Carbon Footprint tool. In this snippet, our host Chris Adams discusses a fundamental concept in environmental accountability, the Greenhouse Gas Protocol, which is the de facto standard for measuring the carbon footprint of any organization or activity.

Chris breaks down this somewhat complex subject using a relatable analogy involving coffee, explaining the three scopes of emissions. He uses everyday examples to illustrate Scope 1, Scope 2, and Scope 3 emissions, making it accessible for both newcomers and seasoned professionals, which has often become the role of twigs.

This approach not only simplifies the understanding of these emissions, but also highlights the significant impact of Scope 3 emissions, which often constitute the majority of an organization's environmental impact. Let's listen to Chris explain this in more detail.

Chris Adams: So I'll just quickly, for those who are new to the subject or folks who have never heard of GHG, the greenhouse gas protocol, essentially, this is a way, the kind of de facto standard for measuring the carbon footprint of any organization or any activity. And you typically split it into three kind of buckets of emissions.

And because we're nerds or developers and drink coffee, we can use hot beverages as the mechanism for understanding the difference between scope one, two, and three. You can think of scope one, which is from combusting fossil fuel. That's a bit like turning on gas to heat up water so you can have a nice cup of coffee, right?

Scope two, it's like turning on an electric kettle. So someone is setting fire to something to heat up some water somewhere to generate electricity so that you can heat up a kettle. So it's all the emissions associated.

Scope 3 is a bit like walking into a Starbucks or a Third Wave coffee shop and then buying a cup of coffee. So you're not involved in actually farming beans or burning anything, but there is definitely a supply chain associated so that you can have coffee. So these are the three kinds of scopes. And typically, Scope 1 and 2 are quite the common ones that organizations tend to report on, but for these, for lots and lots of organizations, scope 3 can make up 80 percent plus of the environmental impact.

And this is why we've been talking about it as being quite a big deal because if you do not have 80 percent of your reported numbers, they may look somewhat different to the other providers.

Chris Skipper: In the past year, Environment Variables has also introduced a new segment called Fact Check. This segment focuses on verifying claims and clarifying misconceptions related to the environmental impact of technology, providing our listeners with accurate and actionable information. In episode 29, Host Chris Adams discusses the role of artificial intelligence in optimizing the environmental impact of software.

He is joined by experts Colleen Josephson and Miguel Ponce de Leon from VMware, who delve into the intricacies of sustainability in virtualization and networking, as well as VMware's internal efforts towards decarbonization. In the snippet we're about to hear, Colleen Josephson highlights the challenges associated with training AI models in the telecommunications sector, a field where VMware has significant expertise.

She points out the substantial energy required to train these models and raises critical questions about the energy savings versus the environmental cost of AI, emphasizing the importance of evaluating the longevity and efficiency of AI models before committing resources. Let's listen to Colleen Josephson.

Explain these complex trade offs.

Colleen Josephson: Training is a very expensive process nowadays. Whether it's in the telco space, because, again, VMware, as you hinted, we have a long history of virtualization and cloud, but that also has become very relevant to telecommunications. We need to train the models that we use to make these decisions to try and save energy.

And the process of having so much data and training it, it can be really power consuming. So I think one of the things that stands out to me is, what is your anticipated energy savings? savings. Once you've deployed this model, how long do you anticipate that this model will be good for? And do you need to retrain it?

All of those you want to have some idea of so that you can calculate whether or not it was worth the energy to train this model in the first place.

Chris Skipper: Continuing the insightful discussion on the role of technology in environmental sustainability, we turn to Miguel Ponce de Leon, who shares his experience working on a collaborative project in Ireland. This initiative, a partnership between VMware and a local grid utility hosting a data center, focuses on integrating measurements from renewable energy sources directly into data center operations.

This effort not only enhances the understanding of energy usage within the data center. but also promotes actionable strategies for optimizing energy efficiency. Let's hear Miguel elaborate on how this collaboration helps data center operators become better environmental stewards by enabling precise monitoring and proactive management of resources.

Miguel Ponce de Leon: So one of the things I can mention is that we are working with a grid utility in Ireland and with that grid utility that also hosts a data center if that so happens. We're also working with an accelerator program, a program that is helping startups to look at how you can not just link but actually be able to take the correct measurements from the green sources, the wind farm locations, and the usage within the data center for its performance.

So, again, here it's about leveraging not just the research, we'd say, that would come from research performing organizations or from the offices of CTO of VMware, but also looking at start ups and start ups within the space and being able to link this. And that is helping the utilities. understand what type of usage, and imagine it's a utility that has their own data center.

So it's helping them be a good citizen even within their own environment, but being able to measure it and then being able to take action on it, right? Because that's the important thing is, okay, you've got your baseline, but what can I change about what I'm delivering within that data center? Even down to the containers, how can I move my clusters and pods over and maybe consolidate some of the pods?

We're even moving some of that research as well to look even with the pods. It's been available. How many of the CPUs are they using within the cluster? So again, it's about being able to help data center owners being good citizens around that space.

Chris Skipper: This next snippet comes from episode 32, which was all about the State of Green Software Survey, with lead researcher at the GSF, Tamara Kneese. The State of Green Software Survey serves as a political resource for the Green Software Foundation, offering crucial insights into the involving landscape of sustainable software practices.

By highlighting trends like the carbon footprint of crypto mining and the need for stringent regulations for generative AI, the report informs and influences stakeholders across the tech industry, from developers to policy makers.

In this snippet, Tamara emphasizes the report's role in enhancing the visibility of green software initiatives.

Tamara Kneese: So one of the main goals was really to raise the profile of green software. And I was really interested to see the percentage of developers who actually had some degree of awareness of green software. And so it makes sense that a lot of the people who filled out our survey already were somewhat aware of it.

and already interested. Although there were a number of people who replied in the comments that this was the very first time they'd been exposed to green software. And so by putting out this public report that can be taken up by the press, that can be taken up by policy makers, that can be taken up by academic researchers, it is a way of really getting the word out about green software.

Thinking about the reporting Court as a mechanism for evangelizing green software is really part of what we wanted to do. And we also wanted to understand after knowing that 92% of developers who surveyed said that they were concerned about climate change and wanna do something about it. So what do they need to actually make that happen?

What resources, tools. and other forms of support do they need to take action. And another key element of this is reaching out to ICT industry leaders to the C suite who really want to know how and why they should make green software part of their organization. And really trying to emphasize the business case for green software from their perspective was another really key part of this survey.

Chris Skipper: As we continue to look back on the previous year's journey, our next segment brings you insights from the Green Software Foundation's HotCarbon event, which took place on World Environment Day, June 5th, 2023. In episode 39, host Chris Adams, alongside Executive Director of the Green Software Foundation, Asim Hussain, Delved deep into a mailbag session, addressing questions that remained from this engaging live virtual event.

Their discussion begins with a look at the challenges of quantifying energy consumed by various computer components in the software carbon intensity specification, and transitions into real world applications of measuring SCI and CI/CD pipelines. The efficiency of GPUs and innovative uses of data center cooling water.

In our first snippet, Asim highlights a memorable talk from the previous year's Hot Carbon event, emphasizing the critical role of green software in managing the burgeoning demands on cloud infrastructure without the need to expand physical resources. Let's hear Asim explain this further.

Asim Hussain: It shows how important the work that we're talking about is. It's like, actually, it's one of the really great talks from last year's HotCarbon, which I loved, which was, I've forgotten, I've got to apologize, I'm not going to remember which one it was, but it was talking about how projecting forward kind of compute growth and how green software was a way of being able to handle the additional usage and load of the cloud without actually having to build more servers.

Because fundamentally we are constrained at the rate with which we can actually increase the cloud. But the growth is growing significantly as well. So like being more efficient actually allows you to, to deal with growth. So I think that sounds like what you're describing. So you have to be green, you have to use green software if you want a realistic chance of generative AI being as ubiquitous as you want it to be.

Chris Skipper: In this second snippet from Asim, he emphasizes the importance of broadening the dialogue beyond carbon to include other critical resources like water, acknowledging that managing environmental impacts often involves navigating trade offs between different sustainability goals. Let's listen to that now.

Asim Hussain: There might be situations where it's mutually the opposite, being more carbon efficient might actually make you more water intensive. Like for instance, doing things that reduce carbon emissions might require more water consumption, which is why I think it's exciting that we're actually are starting to have this conversation right now because I think we're so focused on carbon.

And we're optimizing for carbon, but actually the landscape is much more complicated. It's much more of a surface where you're trying to minimize the environmental impacts of your choices, and you might have to make trade offs versus one versus the other. If there's a water scarcity right now, you might have to increase your carbon emissions.

I'm excited that this is where the conversation is evolving to, because once we add water to the mix, we can add other things.

Chris Skipper: As we continue our exploration, our next episode takes a fascinating leap and dared to go where no podcast, or at least this podcast, had gone before, into the realm of science fiction and its role in envisioning a green future. In episode 42, host Anne Currie is joined by Joe Lindsay Walton, a research fellow in Arts, Climate and Technology at the University of Sussex.

Together, they delve into the imaginative horizons that science fiction opens for tackling climate change, the practical application of these ideas to green software, and the impact of speculative futures on our environmental strategies. In this snippet, Joe Lindsay Walton questions who really holds the reins in crafting a global strategy against climate change.

Let's listen to Joe's interesting take.

Jo Lindsay Walton: Are these orbital data storage facilities, are they going to Out compete the earthbound data centers that are using the dirty energy. Who actually holds the big picture of global strategy here, of addressing the urgent issue of climate change? Is it the conference of parties? Kind of, but they're mired in all these geopolitical rivalries.

Is it the scientists? The IPC? Yes, but they're constrained by the remit of political neutrality and face challenges around communication. Is it the finance and markets? They're waking up to something. They're trying to incorporate climate into these risk management methodologies that they don't really play all that nicely with.

Is it science fiction? Yes, we're drawing in a really interdisciplinary way. We've talked about Kim Stanley Robinson throwing everything at climate change, but it is ultimately a story. I'm not really sure who does hold the big picture. And if I was to try and summarize it in a crude way, it seems that we're hoping to adjust the rules of the game.

We haven't even adjusted them yet, but we're hoping to adjust the rules of the game so that Goods and services and enterprises and value chains and industries and sectors and whole communities and regions that are incompatible with a broadly livable planet are going to be destroyed in the Schumpeterian whirlwind of creative destruction, will crash and burn.

And I think there's a lot of emphasis. on the creation side of that, building data centers on the moon or in orbit, but not enough imaginative, creative, realistic thinking about the destruction side of it. There's this expectation that enterprises are going to snitch on themselves. Oh, we've tested for impairment.

We're reporting against this particular standard. All our assets are stranded. We're just going to shut up shop. Goodbye. So I think I would be interested in more science fictional Thinking about the potential pain of switching from carbon intensive activities to the sustainable ones. Not just the focus on the kind of shiny new possibilities, but also the focus on what it's like to shut up shop.

Chris Skipper: Environment Variables also dared to break boundaries on the academic front. Our next snippet comes from episode 47, titled, New Research Horizons, which takes us to the forefront of innovation in green software. Host Chris Adams is joined by Dr. Daniel Sheehan from the University of Bristol to delve into the evolving landscape of digital sustainability.

Their discussion spans from the implications of historical studies to the transformative potential of recent research, offering listeners insights into how new findings are reshaping our approach to sustainability in technology. In this snippet, Daniel discusses the nuances of energy consumption across different media delivery platforms, emphasizing the complexities in interpreting data that could potentially mislead consumers about the energy impacts of their viewing choices.

Daniel Schien: The academic publication that I'm sure you will link to it in the show notes. There's a graph that compares the energy intensity of those four different modes of delivery of television from the BC. So IP, cable, satellite, and terrestrial. And even though they, they differ. So between, if I remember correctly, between 60 watt hours to 180 watt hours, that's in the year 2016, there's a potential step in the interpretation of those results that consumers might take, that needs more support.

If you see this graph, you might I think as a consumer, if I change from streaming to watching something via terrestrial broadcast, I am going to save 100 watt hours per viewer hour. However, that would not be a correct interpretation because all of those delivery modes, they are provided by an infrastructure that is inherently inelastic in its energy consumption.

Chris Skipper: Jumping to November and Decarbonize Software 2023, an essential event for the Green Software Foundation took place. This annual gathering is crucial for those passionate about reducing the environmental impact of software. Bringing together experts, practitioners, and innovators worldwide to exchange insights, breakthroughs, and strategies for sustainable software development.

With COP28 kicking off on the 30th of November in Dubai, we wanted to highlight projects which are driving momentum towards a low emission and climate resilient world. At COP28, global leaders discussed how to reduce global greenhouse gas emissions by 43 percent by 2030.

In sync with this ambitious goal, Decarbonize Software served as a vibrant forum to promote collective action and shared knowledge. in our pursuit of greener software practices at the forefront of climate action. In episode 53, I was joined by Sophie Trinder and Adam Jackson of the GSF to give a rundown of what happened.

In this next snippet, Sophie gives a rundown of one talk from the event titled Responsible AI, a fireside chat in which Jesse Mccrosky, Head of Sustainability and Social Change at ThoughtWorks, talked about integrating real time environmental impact metrics into our everyday software tools.

Sophie Trinder: Jesse painted this hypothetical picture. What if there was a carbon counter or there was a water gallons clock at the top of ChatGPT? Would you use it differently? Would you only use it when it was essential? Maybe you would write shorter prompts? Would you ask it to write its answers briefer? Similarly, he painted another picture where, what if a software developer's screen went red?

If the developer made a decision that, yeah, might make them save some time, might make them slightly more efficient, but what if it came with a huge carbon cost? So their screen went bright red when there was a carbon cost associated with the code that they were writing. I think it was just a really interesting way to Start thinking about being more transparent with that data so that we can be more conscious with our decision making.

Chris Skipper: Continuing our dive into the impactful insights from Decarbonize Software 2023, next we'll hear from Adam Jackson. In this snippet, Adam eloquently draws parallels between well crafted software and green software, emphasizing the necessity of built in quality throughout the entire software lifecycle.

Let's listen to how Adam articulates the holistic approach needed to integrate sustainability into software development.

Adam Jackson: I often tell folks when they ask me what green software actually is, that there's a lot of parallels with well crafted software. So well crafted software like green software focuses on built in quality throughout the complete software life cycles and architecture, development, operations, and getting all of the stakeholders together to take responsibility.

And we need the same thing for green software. So at the GSF, we often talk about software practitioners. And this is a broad term that goes beyond developers alone. It includes product managers, uh, program managers, designers, UX, testers, IT operations. And what takes the time, and GSF does have some materials that can help here, is developing the best practice, the processes, and the learnings that bring all of these people together.

into a really holistic software life cycle and I think that's why we're really keen to get organizations to hear and listen to others even if they're competitors. What's the best practice that we can all share that's gonna deliver green outcomes?

Chris Skipper: Transitioning now to our next insightful episode of The Week in Green Software, we delve into the pioneering sustainability strategies at one of the tech world's giants and GSF member, Google. In episode 55, Chris Adams is joined by Savannah Goodman from Google.

Who shares the ambitious climate goals set by the company. Google aims to achieve net zero emissions by 2030 and to operate on 24 7 carbon free energy by the same year. Savannah explains the complexity of moving from annual global matching to local hourly matching of energy use. illustrating the innovative approaches Google is deploying to meet these goals.

This episode not only highlights Google's efforts, but also discusses broader implications for the tech industry's push towards a more sustainable future. Let's hear from Savannah on how Google is tackling these ambitious targets.

Savannah Goodman: Google has two main climate goals. One of them is to be net zero by 2030. The other is to be running on 24 7 carbon free energy by 2030. And just to clarify too, 24 7 carbon free energy is much more complex. And I'm going to talk to you about how we've been able to make this a little bit more complex than the annual matching schemes that have been most common to date, because we're essentially moving from global annual matching to local hourly matching.

And so you can imagine how, especially over a global system, how complex that is. And there's no playbook, but we see these goals as a way to actually help scale new global solutions that drive broader system wide decarbonization because we're actually aligning our own goals with what the grid

Needs.

Chris Skipper: Next, we focus on episode 58 that dives deep into the Green Software Foundation's impact framework, probably the biggest highlight of the previous year. Host Asim Hussain is joined by Srini Rakanathan and Naveen Balani. who are at the forefront of developing this transformative tool. They discuss the challenges and solutions in creating a framework versatile enough to assess the environmental impact of software across various platforms, from large cloud providers to personal devices.

This conversation sheds light on the critical need for standardized models that can adapt to different environments without compromising on maintenance and adoption. Listen in as Srini shares Insights into the decision to implement model plugins for more effective integration and broader applicability in green software efforts.

Srini Rakhunathan: With the original concept that we had, where we wanted to cover all, we would have had to build multiple flavors of the impact framework. And that would have caused issues in maintenance, that would have caused issues in adoption. I think the standardization of a model plug in was more important. a decision that we took once we realized that one model is not going to cut the cake for all of us and you had different models depending on whether you are hosting it on AWS or Azure or GCP or your laptop or even your mobile devices.

But if someone wants to just look at the raw emissions from the software, agnostic of the hardware, you could do that only if you have a very thin measurement tools.

Chris Skipper: Finally, let's finish off on another episode of Twigs. In episode 63, we focus In episode 63, Chris Adams, Asim Hussain, and Anne Currie delve into the evolving landscape of AI legislation and its implications for green software. This first snippet from Asim highlights the accelerating integration of AI technologies and their significant energy demands, which could dramatically reshape global emission profiles by 2040.

Asim Hussain: The growth in AI has been significant. We all know on this call that the previous dirty secret of data centers where most of those servers were idle, in a future of AI, those chips are not going to be idle. They're going to be running at a hundred percent. So like, I think we've spoken on this call previously about, you know, various previous reports that talked about, you know, given the current trajectory by 2040, the tech sector will be like 14 percent of global emissions.

I wonder if anybody's doing any analysis. To revisit, well, now, given what we now know about the complete AI will take over everything. It is taking over everything right now. How does that look now? Where will we be in 2040 with the current growth in AI? Will tech be half of all emissions? And will we just be sitting there, you know, being carried around by robots and being fed by little tubes like that robot show?

But I think, I strongly suspect that they factor that in, and I wonder if it's an underestimate.

Chris Skipper: Our final snippet comes from Anne as she contemplates the future of green software practices.

Anne Currie: All discussion we're doing around this is great, fantastically great, but I always think about taking it back to my Maturity Matrix projects on the GSF Maturity Matrix project. You need to do different things at different times, so this information is useful to you in different things at different times of your journey.

So for most of us at the moment, we're not doing anything at all. Just, you know, you really don't need that much data. You know, your scope 1, scope 2, scope 3 are not immediately critical to most people who just need to turn off the machines they're not using anymore, do a bit of right sizing, have a think about what they're up to.

To start to think about, are they in green regions? What are the future, what platforms are they choosing? Are they likely to be aligned with the green future? To start with, where most of us are at the moment, you don't need a great deal of data to, to really make a huge difference. So it's absolutely fantastic that they're doing this work and that every, that we're doing this work.

Everybody's doing this work, but we don't want to lose sight of the fact that. For almost everybody at the moment, we don't even need data. We just need to start to, to use best operational practices.

Chris Skipper: What does the future hold for green software? Is it all doom and gloom? Absolutely not. The GSF is actively working on expanding its initiatives, pushing for widespread adoption of sustainable practices across the tech industry.

You can expect this podcast to bring you all the significant updates showcasing both the challenges and the innovative solutions that are making a real difference. Thank you for joining us on this look back on the last year of Environment Variables. Make sure to follow us on your preferred podcast platform to stay updated on all future episodes.

And don't forget to check out the show notes for links to further information discussed today. A final reminder. You can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables. See you all in the next episode. Bye for now!

Greening Software Procurement

9 maj 2024 | 52 min

Joining Chris Adams today is Mike Gifford, an accessibility and open web veteran, to look at the drivers adopting digital sustainability in the industry, learn from the field of accessibility and inclusive design how we can further sustainable software development. Mike tells us about the wins from the accessibility movement that we can learn from in this engaging episode.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Mike Gifford: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Mike Gifford: Yeah. So that's an interesting approach to structure content so that you're hopefully trying to go off and encourage your, the organizations to push themselves to better understand their users and be able to build a website that meet the needs of their users.

I'm your host, Chris Adams. Hello, and welcome to the Green Software Foundation podcast, Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. When we look at the drivers adopting digital sustainability, in industry. As much as we want to think people will build greener digital services out of the goodness of their own hearts, it's worth bearing in mind that in a commercial environment, other concerns frequently take precedence.

And if we want to see wider adoption of more sustainable practices in digital, a lot of the time we have to be somewhat strategic, learning from other fields, for example, and being thoughtful about how we spend the social and political capital we might have available to us. So where should we be looking for people who have done this before?

One commonly cited field is the field of accessibility and inclusive design, where over the last 20 years, we've seen accessibility move from a field where lone, heroic actions were the norm, to one where, in growing parts of the world, accessibility is a non-negotiable, mandatory requirement in new projects with the full force of the law behind it now.

So, are there wins from the accessibility movement that we can learn from to apply for digital sustainability? And where in the history of accessibility online should we be looking? With us today, we have Mike Gifford, an accessibility and open web veteran, to help us navigate these questions. Mike, thanks very much for joining.

Can I give you the floor to introduce yourself today?

Mike Gifford: Absolutely. I'm Mike Gifford. I'm a senior strategist at Civic Actions. I'm excited to be here. And this is, I'm definitely an active listener and I've really enjoyed all the. Discussions that have taken place from other leaders in this field. I'm also a Drupal core accessibility maintainer. Drupal runs about a million websites around the world.

And I've been trying to improve accessibility in that field. And increasingly trying to do work in sustainability there as well. At Civic Actions, I lead the accessibility practice area. I've also been involved in contributing to the W3C's sustainable web community groups, draft web sustainability guidelines.

That's a bit of a mouth mouthful, but it's essentially just a group of people within a standards body called the W3C that is trying to go off and create a set of best practices around sustainability, and I've also spearheaded the development of a tool called OpenACR. Which is a new way to try and help organize vendor claims around digital accessibility.

Chris Adams: Brilliant. Thank you for that. And for folks who are listening to this for the first time, we do have a podcast episode specifically with the W3C, some of the other members, we'll link to that in the show notes. If you're new to this podcast, my name is Chris Adams, as I mentioned before. I work at the Green Web Foundation. Foundation, where we are a Dutch nonprofit focused on reaching an entirely fossil-free internet by 2030. I also am one of the chairs of the Green Software Foundation's Policy Working Group, and one of the maintainers of a software library called CO2.js, which you can probably guess what that does from the name.

It makes it easy to work out the environmental impact of various digital services. Mike, before we start, I should ask, because when I have spoken to you before, I've generally assumed you're in Canada, but you mentioned you might not be, so first of all, where are you calling from, and how's the weather?

Mike Gifford: So I'm calling from outside of Carcassonne, France, and I've been here for just over five months and I've got another four months to go. My wife wanted to have a sabbatical between careers, and this was a good place for us to learn French and eat cheese, stumble across castles and explore the culture of Europe.

So that's why we're here. And the weather here generally has been really good, especially compared to Ottawa, but right now it really has been a cold, nasty week this week. And so it's not being very good and it'll actually be warmer in Ottawa later this week than it is here in, in, you know, outside of Carcassonne.

It's a lot of climate chaos is going on around the world. We're seeing it all over the place. It's inconsistent weather.

Chris Adams: Wow, I did not, realize that. I generally would assume that Ottawa and Carcassonne, I mean, they're more or less the same latitude, right? But they are definitely not the same temperature most of the time, and Ottawa's definitely colder normally, right?

Mike Gifford: Yes, absolutely. We can get down to minus 20, minus 30 Celsius in the winter. And in the summer we can get up to plus 30, sometimes 35 Celsius, but it's a, it's definitely, it gets much colder. It has a much heavier impact from the Arctic weather.

Chris Adams: Wow, so I guess, thank you Jetstream for the clement weather in Western Europe. Okay. Long may it stay, I suppose. Yeah. All right. So I think you're sitting comfortably. For folks who are listening, we have show notes, so we're going to mention various projects and we'll do our best to list every single project and website that we do.

So if you are listening along and you want to learn more about one of the things we have, please do check out the show notes at podcast.greensoftware.foundation for more. All right, then, Mike, sitting comfortably, I guess should we start, yeah? Okay, so I did this introduction talking a little bit about accessibility, but I didn't do a particularly good job of explaining what accessibility might be.

And I'm wondering, before we talk about sustainability, maybe you could just expand a little bit about what people tend to mean when they talk about accessibility in this context, because I think there are a few misconceptions that people tend to bring when they first hear about accessibility.

What is accessibility, and what are the kind of myths that we might want to think about that we should probably dispel?

Mike Gifford: First of all, people assume that accessibility is just for the blind and deaf community. Like this is who we're trying to go off and to meet. And there's just a small number of people who comply with that. Why do we need to really worry with this? This is just a, you know, a government overreach additional administration, but that's, it's actually not the case.

If you look at permanent disabilities, most countries in the Western world, a quarter of the population has some form of disability. That could be, and that's a permanent disability, so that could be somebody who's got low vision or no vision, people who have hearing problems, mobility problems, could be cognitive problems, people who have dyslexia, color blindness.

It's a whole range of different aspects. And I tend to think of it like just part of a human experience. Is it that, that when we're born, we have certain limitations and we grow up and we learn how to do things. And we have, you know, we have abilities that we have at this time. But they change over time, you know, when I have allergies, my eyesight isn't as good because I have blurry eyes because I've got allergies.

There's also situations where as we age, like as soon as you're over 40, your eyesight begins to degrade. Probably most people listening to this podcast have glasses. That's a form of assistive technology. If you didn't have glasses, you probably would, would have much more difficulty reading a book or reading a monitor or your phone for that matter.

There's also issues of, you know, temporary and situational disabilities. So if you're, if you're injured, for example, where you're carrying a baby or you're in a loud environment, it may be more difficult for you to go off and to hear the other information that's going on around you. The other thing is looking at people who might be in a place where there isn't a lot of internet, like where if you're using an older device, these are also places where there's limited elements people can access.

And if you're in, in, in rural Canada, you can have a lot of difficulty going off and getting the bandwidth go off and to download a web page. And that's not a disability as such, but it is a way that is people are being disabled because the assumption is that people who are using the technology are going to have super fast

Chris Adams: infinite bandwidth, newest iPhone, everything like that.

Mike Gifford: If you don't have the latest device and the highest bandwidth, then you're going to have a lot of challenges. And so many people outside of urban areas don't. And even inside of urban areas, there's places where you can get, there's dead zones in cities where there, or that the things are wrong or inconsistent.

So a lot of our assumptions are incorrect. And I like to think about it also in terms of just, ultimately, you know, the planet is going to be fine with, you know, with or without us, the planet will survive. It comes down to us as a species and as a species, as Western civilization or civilization, how do we want to organize ourselves to survive?

And the brass tacks is that we have to work on a very fragile planet and we have to work with the fact that we are fragile species and that we need to go off and think about it not us at the height of our abilities,

Chris Adams: The whole gamut,

Mike Gifford: abilities throughout the whole gamut of our lives, right? Hopefully we'll all grow old.

Hopefully we'll all be able to explore and experience worlds where we, we do have different abilities because we've managed to live to 80, 90, a hundred years old, but that will be a different experience for all of us.

Chris Adams: I see, okay, alright, that's actually quite an expansive and interesting way of framing it. I haven't actually thought about it in that way, and the idea of saying, well, accessibility is actually a kind of... the fact that you've shifted it somewhat to the situation you're in is actually something that I think that's that's quite helpful framing. Okay, so we've got a rough idea that accessibility isn't just about helping a small number of blind people It's actually a significant part of the population in many cases probably larger than 25 percent when you include the different situational aspects someone might actually be in.

Okay, and it sounds like you know, did you ever see the London Olympics opening where they've got Tim Berners Lee or like Timball to his friends saying, where he has this whole thing, like this is for everyone. He's talking about the web and the internet. There's a thing that it seems very much that you're kind of, there's a reference to that, I suppose.

So these sound like good things, right? But we know that historically, when people are working on digital projects, accessibility hasn't had the same kind of priority that say, shipping a new feature might be, or making sure a particular date is hit by a team working on something. So let's say that, as society, the things you've described sound like good things to be aiming for and wanting to include.

How has maybe the accessibility community made sure these do get prioritized? Because they don't immediately all sound like they would have an immediate short term kind of benefit, for example, or that you might explain to on an earnings call, for example. Maybe you could, say, lay some of that out there and we can explore some of that area as well.

Mike Gifford: You'd think that it really wouldn't take much more of a carrot than trying to go off and support your future self, because that's ultimately you're doing. It's just trying to go off and make sure whatever you, whatever abilities you wake up with tomorrow, which may not be the same as they are today, are going to be, but you're going to be able to use the technology around and interact with the world as best as you possibly can.

That's not much to add, but it's not how people think. People think, people are very much more geared on what's the sexy new thing. Like. At FOSDEM, I gave a talk on accessibility, or sure, on sustainability, and I was, I've given talks on both of them, but it was, and I think it may have actually been both of them, an integration of the two and that, that one.

But the previous two sessions were on AI and they were packed. Like they were, it was like standing room only. And then my talk came up around sustainability. I mean, the room wasn't empty, but it certainly emptied out. And yet this is one of the things that like, this is life. This is us. You know, AI is neat and all, but ultimately it's not the thing.

But in terms of answering your question, what caused accessibility to go off and actually become a, a thing that people are paying attention to is the efforts of people with disabilities who have lobbied long and hard and protested often with civil disobedience in order to go off and to make sure that their rights are respected, seeing accessibility as a civil right.

And so that they are able to go off and have the rights to access by law. And in most Western countries, people with disabilities have rights to work and to employment, to housing, and to be able to communicate with their government. But that process is not being well respected. The digital world is being very much a move fast and break things kind of world, and, and actually supporting the bulk of their users is something that generally most companies don't invest in, even most governments.

Most governments have, far less accessible websites than they say that they should or say that they do. and, and it's something that, that you know, it's only through protest and through awareness raising and decades of effort that, that we've been able to get to where we are right now. and, and that's something that is still not where we need to be.

We're still not at a point where there's a final site that we can say, "Yes, this is accessible." This is like, it is... sites are more accessible, there is more awareness, but it's, it's just like security. It's about perpetual vigilance. So how do we make sure that we're more accessible today than we were yesterday, and that people are pushing that, that type of a framework forward?

Unfortunately, many people have a checklist mind set. They want to go off and say, "has this been done? Yes, let's move on." And that just. Unfortunately, it doesn't help people with disabilities who are actually experts around their own experience of where they're facing barriers with digital interfaces and you know, they need to be able to find ways to engage with their governments or engage with their stores or their friends in ways that are just like everyone else.

But that has, that does require effort and maintenance.

Chris Adams: I see, okay, so it sounds like there's a degree of, like, hard won essentially campaigning on various levels, and there's degree of, like, organ and there's maybe a degree of actual, like, organization with the people who are able to say, "well, okay, we need to find a way to translate these into concrete things you can ask for," or that people who do want to do the right thing, so they know how to go about doing that.

Mike Gifford: Right. Which is, which is how the web content accessibility guidelines. And the Web Accessibility Initiative from the W3C got that started because they saw that this was a need to try and, and provide some guidance and direction that everyone could agree on. And that would be a universal point that would allow everyone to agree that this is an improvement that, that benefits people.

And creating standards is a huge challenge, but it's been really important for organizations and governments to being able to lean on those and build legislation around that. Some of which actually have teeth for a lot of agencies or organizations around the world. So the, having legislation that requires accessibility is a, it's a huge deal.

Chris Adams: I see, and okay, so there's one thing that you just spoke about just there was this idea that, okay, you're essentially able to kind of capture this idea that society values this thing, which is not maybe immediately helping in a quarterly earnings call, for example, but it's still something that people tend to value, and that has ended up being translated into various forms of law, and one of the things that helped with the creation of that law was essentially things like the existence of some of these standards so that people could then say, "well, we want to have, you know, because we value access for everyone rather than just a very specific set of people, we're going to refer to these standards here saying you need to at least hit this kind of bar."

That's how some of it came about. Okay. And if I understand correctly, you, some of the examples you used earlier on about the situational example, like when you've got a broken arm or in your, I think I've seen some of that mentioned from Microsoft, actually, in some of their inclusive design things, so presumably there's an argument about the regulatory certainty that gets provided and means that organizations themselves might want to invest in saying, "well, okay, this is how we are competing in this by, you know, we're going to meet this bar when other people aren't," for example.

That's another kind of argument people take. Okay.

Mike Gifford: And Microsoft has had a great deal of leadership around the thinking around inclusive design and trying to think about how do we make sure that we are thinking about the whole self and not just ours, everyone on their best days.

Chris Adams: Okay, cool. So it sounds like there is a bit of a push. And so one way that some of these were adopted was essentially going through the public sector route to kind of demonstrate that there's demand and there's something that people need to go for and that and by creating that demand that allowed people to then respond to this.

Okay, so I've, are there any like particular countries or examples you've heard of? Because I've heard of like stories like say, I don't know, the things that started in the public sector that then kind of, percolated down. So like maybe I think things with Kindles, Kindle books, for example, or something like that, are there any examples we might point to show how something that large organizations might have been prioritizing first, but this end up kind of shifting them to start prioritizing something in a way that they weren't previously prioritizing?

Mike Gifford: I think that one of the challenges has been that, that governments have initially tried to go off and create their own legislation and, and not to build on a common standard. And one of the issues with this is that the tech is often global. And so if you're trying to go off and create a legislation, like so many governments individually in government agencies just don't have the clout to go off and to go to Microsoft and say, "Yeah. We want you to go to this and implement these standards because these are, if you want to sell to us, you have to follow these standards." Like maybe Microsoft will say, "sure, we'd love to do that. We're happy to go from bend over backwards and make that happen." But most likely they won't. They'll say, "sorry, if you're, this is what the product is.

And if you try it, we'll charge you X amount to go up in additional costs to go up from to meet those standards". But it, but it's not something that, that we can cover because you're such a small entity, generally how, how government procurements work. But, but if you're able to go off and work on a global standard, then technology, and especially if technology companies are involved, like they are, they have been to the W3C, that there's, there's an opportunity to, to get them on board and make sure that a lot of the kinks are worked out of the process, when it comes to implementation.

So that there are things that we know that can be implemented and can be maintained by the technology that we work with, whether that's open source or proprietary, there needs to be an engagement around how we're working on those same set of standards.

Chris Adams: So, one thing that you've just spoke about there was this notion of, okay, you, there's some work that goes into creating standards, because that allows both organizations, at governmental level to say, "hi, we need you to meet at least this bar," but from the other point of view, for people who are trying to meet these standards, let's say like, okay, you're calling from Europe instead of Canada today, right?

Everyone having 28 standards per country in all of Europe, for example, having one makes it a bit easier for someone who's inside the company saying, "hey, can we even at least meet this bar? Because this one will mean that we can, we don't need to do it across all these other countries as well," for example.

That's one of the ideas around the kind of some of these global standards. Yeah.

Mike Gifford: Absolutely. Having a single standard is much easier to get compliance around, particularly if you get a few larger economies that are adopting it early. Like if the European Union is able to adopt a standard around digital sustainability, that's something that will have real impacts around the rest of the world, because a lot of people want to sell their products into Europe.

So. So that's just like the GDPR has had a large impact on privacy outside of Europe as well as inside of Europe because it was something that was, was agreed to amongst a larger group of countries that there was a, an opportunity to really make waves across the digital sector.

Chris Adams: Okay, so that's quite interesting. So we spoke, we were speaking a little bit about accessibility and essentially the same things, because there is maybe not a short term gain for this, you might talk about it in terms of, okay, there is, you're creating some certainty and saying, well, these are things you need to actually have, and I guess in the context of some recent laws we're seeing passed, you might see some of that start to play out in the form of, I think we have things like, say, the CSRD, the Corporate Social Responsibility Directive in Europe, as one example, but even in America, or even in, I think, California specifically, because California is a big enough economy in its own right.

You're seeing some regulations there which end up acting as a kind of driver to kind of move to rather than just have people inside companies saying, "we'll save you some money on your green service." It's like, "no, this is the thing that's actually going to be, this is the bar we need to meet if we want to kind of keep selling to these people here," for example, or "this is a risk that we can head off by making sure we're compliant with this."

That seems to be the argument you're essentially kind of making around that, right?

Mike Gifford: That's right. And I think that, that the, it'll be interesting to see as scope 3 emissions get, get, more closely monitored and evaluated the, the impact of digital, I think is going to be a lot bigger than it currently is because most people don't, don't own a data center, they use services. So if you're, if you're suddenly looking at those people who are managing all the different technology that we're engaging with our lives on a regular basis, that has a third party impact and in terms of virtual pre-emission changes, and I think that we, we're going to, it'll be more of a challenge to try and figure out how to regulate those and to address them.

The web doesn't really affect most people for their scope one or scope two emissions, but it will certainly affect for, for scope 3 emissions.

Chris Adams: Right, I'm just going to take a kind of step back out for this. So I just want to check. So when you've mentioned the words scope 3 and scope 2 and scope 1, I assume you're talking about, say, the Greenhouse Gas Protocol. This is a bit like an international standard for tracking carbon emissions for an organization, for example.

So where scope 1 might be me burning, say, gas and emissions caused from, I mean, I use the hot beverage example because that seems to be quite easy way to get your head around it. Scope 1 is me burning wood or coal to heat up a kettle so I can have coffee, right? Scope 2 is me consuming electricity, so someone else somewhere is burning coal or gas so I can have coffee.

And then scope 3 is the entire supply chain, so that's like me going into Starbucks so I can have coffee. And then, there's still a supply chain, but I'm not burning anything myself, but, you know, there is still someone accounting for the emissions that way, and that's what you're talking to when you talk about these kind of scope 1, 2, and 3 thing here, right?

Okay, brilliant. So, we spoke a little bit about how there are maybe some patterns, some ideas, and some lessons from the accessibility movement where, with the creation of some standards, we've then seen people tie this to, say, how projects get bought. Like, someone help people allocate time and money to do a piece of work and I think the one thing that when we spoke before you mentioned about how you saw this happen at a kind of public sector level first saying "we have to make this accessible because we have a mandate to serve all of our population not just the rich people basically or not just like the this one target market we've decided to segment and go after" essentially but that ends up creating a kind of norm which then is allows, say, other sectors or kind of more other private sectors to start looking for that as well.

And I think there have been examples of, was it the Domino's pizza, for example? The famous example where once people had set a norm that you actually had accessible websites, then was it, maybe you could expand on this one because I think I'm going to get it wrong, but I do remember like Domino's had to start rebuilding their websites and making sure their sites were also accessible to people because the norm had been set.

Mike Gifford: There's a, the US has an unusual piece around accessibility where if your website isn't accessible, you can be sued. There's people who've set up lawsuits and yeah, and some of them are not as legitimate as others, there's easy enough to find accessibility bugs in various different websites. But, but there are, there are people who are chasing the court system in order to try and get revenue from lawsuits.

So then that happens quite a lot in the states and that's a huge motivator around accessibility in the world actually, is the, the sphere of lawsuits from or within the United States. But domino's got sued and they were like, "we're not going to pay for this. We're just going to go off and channel just in the courts.

We don't think this is necessary." They did end up losing the case and they also spent quite a lot more in, in, legal fees than they did in actually hiring web developers who knew what they, they were doing and actually were able to fix the issue. But it was a, it was an interesting case in that, that they, just the egos of the senior management at Domino's.

Were like, "well, we don't want, we don't believe this is the right way to do it. So we're just going to do it our way." Even though they were explicitly excluding customers who could not, who were trying to go up and buy their services,

Chris Adams: Like, "hey, want a pizza." And they're

Mike Gifford: yeah, "I want a pizza." yeah, so I was like, why couldn't a blind person order a pizza through Domino's pizza only ordering system?

It seems ridiculous. They wouldn't want to go off and to address their customers, but that, that is the case for some companies. And having a hammer to go off at, to come down and for those organizations who are opposing best practices is sometimes needed. Unfortunately, you need to have the carrots and the sticks.

Chris Adams: Okay, so one thing you've said, there was basically an immediate harm that was being felt by some groups who were able to then look to some specific laws and say, "hey," this is, that was essentially the basis for that. And that's how you've been able to kind of maybe compel some activity from organizations who might not be prioritizing otherwise, right?

Now, I, for that to be possible, presumably, you needed to be able to demonstrate that a website was unusable, for example, or something wasn't working the way it would. I mean, are there any lessons that we might apply from how we think about maybe normalizing some things around digital sustainability for this?

Like, is there anything we can use from there? Because it sounds like there are I don't think anyone's been sued for not having a green website or a green digital service or anything like that. And I don't I'm not saying you should reach for that first, but It's clearly one of the techniques that people have been using in other fields to kind of compell action.

Mike Gifford: It's interesting though, that Europe is suggesting now that you can't claim to have a green, green product if you don't. You know, they're making greenwashing illegal, which is, it's similar in some ways, but so much around accessibility has been around the creation of these web content accessibility guidelines.

Which have been a set of best practices that have been created by people around the world. Very much like in the, the, sustainability field, the web sustainability guidelines are much newer and they're only being developed by a committee, so they're not something that is a full W3C working group recommendation, but it's something that is moving in that direction.

But I think that in terms of motivating people, you need to have some bar that you can say, how do you demonstrate what the expectations are, and that you're meeting those expectations. And then have a common bar to, from, and to, to compare yourself against 'cause, I mean, how many people have a website that is green enough?

There are not that many. And there, even when you're looking at trying to create the standards and look at, "well, how large should a webpage be?" There's a lot of different views. And the average weight of a webpage, we know right now is about, I think it's 4 megabytes is the average weight of a web

Chris Adams: Bigger than Doom, then, yeah.

Mike Gifford: so it's only the initial release of Doom after all, but it's still shocking that it's that as large as it is, and there's so much cruft that is not necessary. But do you think we can get a website down so that the average site is a hundred kilobytes? Well, maybe, but it's going to be a huge amount of effort to go off and remove all that cruft and advertising and layers of JavaScript and to weed out what is actually necessary to convey. And then there's also elements of how do we build society that, that is where there's, privacy exploiting, JavaScript apps that are adding to the bloat of the infrastructure that we're not addressing. Security issues 'cause, again, if you've got a lot of software, you probably haven't monitored it to make sure it is meeting best practices. And having secure, sustainable code is really a huge part of just creating quality code. If you just redefine it as part of quality, then that's something that, that is really an important way to reframe the advantages of building for accessibility and sustainability.

Chris Adams: So thanks for that. So there's a couple of things that we spoke about and I realize I'm showing my age I'm making a joke about Doom and I realize that most people for them to like get Doom they need to know what a floppy disk drive is and you realize oh my word we... I realized that ages me quite a lot.

So for anyone who... doom was a very popular 3D shooter, and it was, it's definitely smaller than most websites these days. Okay, all right, let's move away from that somewhat depressing note and try and get back to what we were talking about. So you're talking about, there was essentially a way of demonstrating that there are some expectations you have, and then you're able to demonstrate if a website was meeting a certain bar of accessibility and I think when I've heard people talk about accessibility they use this term POUR, like is it perceivable, operable,

Mike Gifford: understandable and robust. Robust is a tricky one, but basically trying to go up and make sure that you're doing this in a real world environment, that it's something that it's not just from set up in a lab, but you've actually got real people, real engagements, real user flows,

Chris Adams: I see, and these and so perceivable is like okay if you're not blind you can't see obviously so you should be able to perceive something. Operable, I get the idea that you can operate it, right? Understandable, self explanatory, and

Mike Gifford: Plain language, all that

Chris Adams: yeah, and then the robust thing is basically this idea that Okay, you've opened this discussion about saying in certain parts of remote Canada you can't access something.

For example, that's like robust, like something might degrade gracefully, so you can still achieve a particular thing even if you don't have like the latest greatest iPhone on the super fast Wi Fi or something like that, right?

Mike Gifford: Right, right. Exactly. It's, it's looking at what are people using, how are people engaging with your site and how do you train and address those? And there's, there's not as many success criteria for robustness because it is a trickier one. Perception is certainly one of the easier ones to address and that's mostly about trying to go up and have textual equivalents for non textual content and making sure that information is visible for people with however they're able to, with whatever senses that they have that they're able to perceive that.

Chris Adams: I see. Okay, so one thing that I guess One insight I'm taking away from that is that these don't necessarily have to be like a binary yes/no thing every single time. Obviously it helps, and it means you can automate things, but that's not an absolute deal breaker for creating any set of guidelines, for example, or something that you might write into a commercial agreement, or might say that, you know, every public website should at least be doing this kind of stuff.

It sounds like you don't necessarily need to have that kind of binary thing.

Mike Gifford: One thing that they've tried to, I mean, the courts definitely want it to be binary. and there's definitely things that, that are, there's different levels of WCAG. There's, there's level A, AA and AAA

Chris Adams: Sorry, sorry, you said WCAG. Is it WCAG? This is Web Content.

Mike Gifford: Web Content Accessibility Guidelines

Chris Adams: got Okay. That's okay.

Mike Gifford: people just pronounce it WCAG, but I've gone with, with WCAG because that's.

Chris Adams: Yeah.

Mike Gifford: Somehow rolls off the tongue faster than, or easier than WCAG.

Chris Adams: Okay. All right.

Mike Gifford: The W3C three in general is good at many things, but not naming things. But the point of the level A, AA, and AAA is it's try to say what are the bare minimum things that organiza that things people should do.

Easiest stuff that affects the most people. What is sort of AA is what is legally required. So in most countries, it's WCAG 2.0 AA is the legal requirement and AAA is sort of aspirational stuff. So knowing that your users are at this base, if you're dealing with senior citizens, for example, you might want to go above and beyond what WCAG 2.0 AA provides, because you know that your users are going to have less perception of color. So you want to have higher contrast available for them than you might for other users. So it's really a, about stretch goals, for, yeah, so it's an interesting approach to structure content so that. And you're hopefully trying to go off and encourage your, the organizations to push themselves to better understand their users and be able to build a website that meets the needs of their users.

Chris Adams: I see. Okay. So if maybe one thing I'm taking away from that is that first of all, okay, there's value in binary, but there's also value in certain kind of grades as it were. And that might be more useful for the specific, I don't know, set of audiences you're trying to reach or who you're trying to cater, for example.

And presumably this, by kind of breaking into something like this, kind of removes some of the need to be an absolute domain expert so that if you're commissioning some work, you can say, well, "you need to be meeting AA," for example, for this, and that's what you could write into, say, a procurement thing to make sure it definitely does happen inside a new project, for example.

Mike Gifford: I mean, and that's what what a lot of organizations have tried to do is they tried to go off and put in contracting language that their product or service must be WCAG 2.2 AA compliant. what the current best practice would be. But unfortunately, procurement officers often drop off the necessities that if, if the client says that the site is successful, they'll leave it at that, and that there'll be no, not necessarily any follow up.

If the project manager on the client side doesn't care or isn't motivated to address this or doesn't have a subject matter expert that they can evaluate, then it's something that that can be quite difficult to implement it. And especially since like so many things, if you don't think about it early enough in the process,

Chris Adams: easy to count, yeah.

Mike Gifford: yeah, so you know, so often accessibility has been left to the end.

And then, well, the budget's spent, the time has run out and now you've got these accessibility issues you've just now discovered. Well, guess what? They're not going to happen, right? They're going to be, or they're going to be done badly. And in a way that's more expensive because it's being done after the fact.

And, and that's often a more fragile way to go off and implement technologies. Think about it from the beginning.

Chris Adams: I see, okay, so there's two things that come out of that that I think are possibly worth exploring in a bit more detail. One of these was this idea that basically, the cost of checking if something is still working the way it's supposed to. Like, you know, people say, "oh, it's totally accessible." But being able to check if something really is accessible, this is one thing that I think you've pointed me to some work where there's some software like Purple Accessibility, Purple A11y.

Yeah, Purple A11y, yeah. This is, I'm not sure where the A11y thing came from, but for people who are not familiar with the lingo, A11y is A one one y, which is the number of letters for accessibility, and Mike, you introduced me, what was the name of

Mike Gifford: It's a, it's a numerium. Anytime you go, you take the, the letters of the beginning of the end and you squish them together. Internationalization is I18N, localization is I10N. It's just, you squish them together and create a number of

Chris Adams: Ah, I see. And that would explain the whole S12Y that I think people keep trying to make happen with sustainability in the kind of digital world, right? Yeah. It feels a little bit mean girl saying stop trying to make, like, fetch happen, right? But this is something I struggle with as well. And if people who are curious why this episode is called A11y to S12y.

Now you know why. It's, numeriums at work, basically. Okay,

Mike Gifford: And they make a lot of sense, especially in Twitter, they make a lot of sense. But now that we no longer are limited by 180 characters, it's, and actually in some ways just complicates things because since sustainable development has been a term popularized by the Brundtland commission, I guess it was the eighties, the sustainability has been overused and generalized in a way that makes it very difficult to understand things and just sort of shortening it into, you S12Y doesn't actually make it more understandable or any more specific.

It's just made it more, added another, fewer characters and more confusion. So it is something that does happen.

Chris Adams: Does make for a cool hashtag, though, so I suppose, you know, swings around about. All right, so we digressed and the reason we mentioned Purple Alley or A11y was because I think this was something you were me about, how there are ways of, there are increasingly tools which, rather than just doing a one off check at the end of a project, might continuously be tracking some of this.

And this is maybe a way to kind of almost provide like continuous audit or something as one of the mechanisms that you might use to kind of maintain a level of quality, for example. I believe Wagtail is one example. They've built some tools so when you're editing a web page, they will basically give you a kind of check ahead of time.

So it's a bit like, shift-left, but for content editors or people who do this. So before you publish something, you'll see some of this and I would assume that other content management systems have these kinds of checks as part of the kind of workflow process rather than it being a thing that gets checked once a year, for example.

Mike Gifford: Right. Drupal has it with the, for the code level. We don't have it for the content level. There are plugins you can add in for modules. You can add in for Drupal that give the authors that sort of context, but we don't have it built into core. And that's something, so it's an effort to install that. It's hopefully it's something that will come into core at some point, but that's a much bigger discussion, right?

How that, how and what tool they use and how that would be implemented. But Purple A11y is a great tool for scanning websites and providing a, an understanding of where accessibility issues are across hundreds, if not thousands of pages, because there's a lot of tools that are built into your browser that you can use to scan a single page. So, like there's WebsiteCarbon.com and the EcoGrader from, which are great tools. But they're, they will evaluate one page and you can pay for services, certainly from EcoGrader that will go off and allow you to scan your site. They keep a sense of your site, but most people are, are not necessarily going to do that.

And not everything is designed for that. And site scanning tools. Yeah. It's nice to be able to have an open source solution that is able to give you a, a sense of the challenges and the barriers that you're running into as you're developing the site. They can see that you're making progress or your site is more accessible today than it was yesterday.

And the incorporation of something like CO2.js into that. It would be a really great addition and there's other tools that could be useful to go off and add to aggregate quality as well and get a sense of like plain language as a, as something that it's very difficult for people to go off and write.

People under think they understand it, but it's, it actually takes a whole lot of work to go off and write in plain language. But this type of language complexity can be evaluated on a code level, at least on the English language. It's more difficult for other languages, it can be. The more we can sort of build in testing early in the process, however, when a site's being developed, if it's built on a weekly or two week sprint cycle, the team, when they're doing their script planning, should know what their accessibility issues are this week compared to the last sprint, so they can see if they're actually making it more accessible today than it was previously.

And the same thing should be applying for sustainability, because there's things that do change in code, like a new JavaScript library is added. And things slow down or there's an image that, that somebody added and that's something that is not noticed by the development team and you want to catch that as early as possible, partly because it's easier to fix them because you've just added it, so you should be able to the issue and, and, and you're like, "Oh, it's just the last sprint.

We know we approved that" and you can learn from that process. I think much like, like spell checkers, like if spell checking was something that you only did right before you submitted your paper,

Chris Adams: Yeah.

Mike Gifford: like you're not going to learn, you're going to be frustrated by the process. It'll be something you ship off to somebody else to deal with and you're not going to go off and build that into your own practice.

But if you see the errors and you see them as you're writing it, eventually you will write better and you'll have a sense of what are the mistakes you often get and how do you try and avoid them going ahead.

Chris Adams: I see, okay, so it's basically a piece of, like, incorporating into, like, your workflow, I suppose. You, you said one other thing that I wanted to unpack, if I may, which was a little bit about setting expectations. Like, talking a little bit about how, like, because you, you mentioned there's things you can do for, like, language, for example.

And we spoke a little bit about this idea that, okay, there's a certain level. I think you mentioned WCAG AA or something like that. And I know the one example I've seen from other places are saying "we aim to meet this level and this is how we check for this and here's how to actually contact us for any kind of breaches" for example like these seems to be mechanisms that i can see being adopted to talk about digital sustainability like one of the reasons i'm particularly interested about Purple A11y and that one there is a it's an open source so anyone can use it for free assuming they have access to like a computer to do that.

But also the Singapore government have been talking about using this to check their own digital estate as well. And that's how I know that in the public sector, some people are using these to do some of this kind of continuous monitoring, or at least some way to kind of flag things up to make some of this a bit more kind of observable and actionable, I suppose.

And like, as far as I'm aware, we don't really have any way of, do we have a way of checking that for accessibility when a... so, you know, you go to a website or a digital server, you say, well, do you, is there a way of seeing a kind of consistent URL, like wellknown/a11y or something to see where they say what they are, say what level they're trying to reach, and then how to kind of test about this.

Because this feels like something you could at least for certain things, be able to kind of verify, you know, verify the claims that you're making, basically.

Mike Gifford: Right. I mean, I think right now for accessibility, the best practices is around trying to create an accessibility statement that's in the front of a web page. So if you go to any European government website, now you'll see a page that is about accessibility, that gives you information. And if it's properly structured, it'll say, it'll have a feedback form that allows you to go up and give feedback if you run into a barrier.

It'll tell you what they've done and how they've done it and get a sense of what some of those challenges are. Also, it's useful to highlight the, yeah, what are the processes that they've done to go off and assess accessibility. All of that can be built into an accessibility statement. And the same type of thing could, could very much be included as a sustainability statement.

And there are going to be pages, there's out,

Chris Adams: we target this, yeah.

Mike Gifford: That's right. I mean, there's no reason that couldn't be done the same, in the same way. You just have a little link in the footer that explains what the process is. And you evaluate that page on an annual basis to make sure that there are things that are being, so that if you're changing your processes, which they do, you come back and reflect on that and say, well, how do we make this better this year than it was the last?

Right. That, that reflection, I think is an important part of it, as well as the feedback with the users.

Chris Adams: Okay, alright, so there is a way, so maybe this is actually quite a nice segue for us to move from, okay so that's one pattern that we've seen adopted in the accessibility field where we know there's a tradition of being quite specific about things you're looking to purchase but also some ways of checking that people are actually like meeting the things that they're saying they're doing.

I mean, there's some parallels to the whole kind of green claims discussion I suppose and that you might see online where people are saying, you know, you can't mark this product as carbon neutral by buying offsets anymore, but they say like if you're gonna make any kind of claim you need to kind of substantiate these claims now rather than actually just having some woolly language around this.

So, okay, and maybe this is a nice way to talk a little bit about what that might look like for digital sustainability because I believe the I know there's these Digital Sustainability Guidelines. I don't, I, WSGs, I'm not quite sure what we would call them. I mean, in Germany, you might call them like WSG, as if it looked like a joke on ESG.

I don't, okay, ignore it. Like, we don't have a good word for them. I mean, nothing is, snappy as WCAG, yeah. So, come up with something for those. I do know that there has been some work to make some of these checkable. I think, I remember the name, it was Star, wasn't it? Star, which is. Sustainability, Technology, Accessibility, something.

Do you know the thing I'm to here?

Mike Gifford: I do. But I don't the

Chris Adams: Sustainable Tooling And Reporting. This is what it was actually. Yeah. I just looked it up in a tab. Yeah.

Mike Gifford: There's whole, there's a whole lot of effort to go off and get a pronounceable acronym that is also somewhat meaningful and doesn't offend anyone. And that was, yeah, that, that's what we came up with STAR, because it's a generally universally appreciated type of thing, so.

Chris Adams: Okay. Well, this is good. And this is like machine readable things that you could check. So you could check it each year, but you could almost Yeah.

Mike Gifford: And that's it. The more that there, I mean, I think that there's always going to be things that are going to be, you know, there's some things you're going to check by machine, but you're not going to be able to check it all by machine. Like even with WCAG, that's being around for as long as it has, you can only catch about a third of the issues using automated checkers.

So with sustainability, there is going to be need to be manual but there's stuff that you can check, knowing where the site is hosted and where the third party libraries are that's stuff that's checkable because it's just an IP address you

Chris Adams: Look it up. Yeah.

Mike Gifford: against database and determine whether it's there.

And the same, the size of the pages and the, the, the use of JavaScript. These are all things that can be. evaluated in a very, black and white sort of manner, like compared year after year to see how, how are you doing or print after sprint to make sure your, your. We're actually making movements on this and just to make, for people to realize that these tools are things that have been hidden, like the weight of a page has been hidden for people for so long.

They're becoming much easier to evaluate and to track and to identify. So a lot of the stuff is, is going to be more noticeable for people and for them to say, well, "why is my page taking so long to load? Who's responsible for that?" Like that's, that is going to be an issue that, the people will start to pay more attention to cause they have to.

Because there will be legislation, hopefully, that requires an organization to demonstrate that they are improving their performance and have improved customer experience and making a more sustainable site.

Chris Adams: Okay, so I'm speaking to you with a, because you're in France, and I know that if we're talking about digital sustainability, France is currently one of the countries which is kind of at the forefront, and they, I'm going to butcher the, the, the RG... I can't remember what they're specific, and it's basically the French version of digital sustainability.

I think it's Référentiel général d'écoconception de services numériques, but basically it's the equivalent to digital sustainability stuff. It looks a little bit like the digital sustainability kind of guidelines, but the French version that they've been written as a, and there's, as far as I'm aware, this is something which has been adopted nationally as a thing that you would actually include.

Now there's another discussion about how you might kind of maintain that and enforce that and make sure and what the remedy mechanisms are if you are in breach of that. But like, that's one of the examples there. I guess, you know, that's one people, one thing that people might be looking at as an example of, okay, here's some ideas, but here's what some of this might look like, for example.

Mike Gifford: Yeah, no, I think it's great. And there's things that I think would have been quite useful to have done for accessibility as well. For example, engineers are required to take a course on sustainability when they're getting educated in universities. Well, why isn't that part of how things are done in accessibility as well?

That's a big one. There's certainly a lot of elements around evaluating municipalities and evaluating other government agencies in terms of what their digital sustainability plans are. A lot of work around recycling, the recycling electronics, again, that's a big part of this puzzle is trying to make sure that the, all of the carbon and resources and energy that goes into these devices is managed as effectively as possible.

So again, that's really forward thinking stuff. And also it's wonderful that France has gone first on this and highlight that, that this is something that, that like somebody has to go first. Somebody has to demonstrate that it's a possibility. And France has legislation in place long before people are even thinking about it.

So it's been a real good thing for the Web Sustainability Guidelines to be able to look at and learn from, what's being done in France and, and to say, well, "how do we take this, this model that is being successfully implemented in France" and try and say, well, "how do we scale this to the rest of the world?"

Chris Adams: see, so like you said before about, okay, are there lessons you can make more general? We should probably give a shout out to the fact that, I mean, and one of the reasons I know about this is work, some work from Gautier Rosilla and also I think a gent, Thibaut Collas was one of the people who basically draw attention to basically a translated version of the old RGSN as examples of this and when I first read them, I was like, "wow, this is actually really, this is actually quite well written" and it kind of stays in its lane in the sense that it's just focusing on digital sustainability, it might refer to some other things, but it's not repeating too much of the other guidance that exists in other forms with other groups and other working groups.

Okay, cool.

Mike Gifford: Also, a shout out to Laurent Devenry, who's also been, I'll include a link to an article I wrote for Apolitical, talking about the, the French, the French legislation and trying to help make it more accessible to the English speaking world. Cause I think that's something that is often a barrier for good ideas is language and culture.

And so trying to encourage more, a broader adoption and understanding of what countries are doing effectively to make these changes.

Chris Adams: Okay, all right. Well, I think we're just starting to come to time, but it sounds like there are a few things we could learn from the world of accessibility, like it's not perfect, but yeah, the idea of like setting some basic standards that you can then write into procurement is one thing that we've got.

Like, we don't really have the equivalent to poor, like perceivable, operable, robust yet. I wish we had that for digital sustainability, and maybe one day will, but if that seems to be a thing, that would be helpful. But the whole notion of being able to like continuously check to see if something is or at least have a way of checking if people are saying something and then are they able to kind of deliver against that.

There's a precedent in accessibility that may also be relevant for sustainability too and we have a growing number of standards from outside of technology that we might even be able to refer to for the things outside of how a website is built that might actually speak to some of the kind of other wider sustainability considerations as well, by the sounds of things.

Mike Gifford: Absolutely. And, and I think that the more we can learn from other people's work, the better it is. It's not like we're anywhere where we need to be around accessibility, but there's been some advances around accessibility that are, that are, are useful to learn from. With sustainability, we just don't have the two decades or three decades of struggle to get us where we need to be.

We need to have countries very quickly taking on regulations or that, that allow us to quickly start scaling down the technology. There's so many organizations buy technology that is there for at least a few years, if not for a decade. And larger organizations are buying technology that, that is going to be in place for a long time.

It takes a long time to shift from one stack to another. So we need to be prioritizing sustainability in the procurement process as early as possible so that it's something that can be built in as we go ahead and as we're building towards a more sustainable future.

Chris Adams: Brilliant. Well, Mike, thank you very much for sharing your perspective and your experience with us today. If people are curious about the, what you've been discussing or some of the work you're doing right now, where should people be looking? Where would you direct people's attention to if they wanted to learn more?

Mike Gifford: Right now that the best place to go from find me is on LinkedIn and it's Mike Gifford is the easiest way to access me. I'm also on a bunch of different Slack channels, like the Drupal Slack or the Climate Action Tech Slack is a facility to reach out. I do have other social media accounts like Mastodon and Twitter, but I generally don't use them as much as I do LinkedIn.

Chris Adams: Okay, right, and CivicActions, that's CivicActions.com, presumably, and if there's any of the other kind of user groups or working groups you're part of, are there any ones we should be thinking about? Like, is it OpenWeb? I forget the name of them actually, you said there's another CMS

Mike Gifford: Open Web Alliance that's, that's, that's coming together. That's, that's an effort to try and get, different content management systems to be able to support and work together to go off and build that more effective open web, so that's one group. Also the Susti Web Community Group is definitely something to consider joining if you're interested in sustainability.

And there's an effort to turn that into a full working group, which will require more effort and leadership for people to be involved. And so, yeah, those, those are definitely some of the main places to see. On accessibility, you can also find us on our accessibility sub site, which is accessibility.civicactions.com.

Chris Adams: Brilliant, okay, and I just realized, you mentioned something about FOSDEM in a talk, There's, I'll share the link to the FOSDEM, Sustainability and Accessibility Talk, which was what prompted some of these discussions in the first place. This has been fun, and I didn't realize that you were in Europe, so if I ever do pass through Paris, or pass through Carcassonne, or France, maybe we'll get a glass of wine or something.

Mike Gifford: That'd be lovely.

Chris Adams: Thanks a lot, Mike. Lovely to chat to you again. All right. Take care.

Mike Gifford: See you, Chris.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

Community Publishing and Greening Software

2 maj 2024 | 43 min

In this episode of Environment Variables, host Chris Adams explores the theme of Community Publishing and Greening Software, by talking to Branch Magazine’s Hannah Smith and Marketa Benisek. They discuss community publishing and the lessons learned from producing Branch Magazine, which supports sustainable digital practices and includes diverse voices from the Green Software Foundation. The discussion covers the complexities of adopting digital sustainability and how Branch Magazine, through its innovative, carbon-aware design, has been a platform for expressing these nuanced themes in sustainability and how finding beauty in the imperfect might just be the answer to all your problems!

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Hannah Smith: LinkedIn
Marketa Benisek: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Topics:

Branch Magazine | climatetech.action [1:50]
Designing a Grid Aware Website [13:09]
Content Trends [20:38]
Sneak Peek Preview of issue 8 [27:12]

Resources:

Grid Intensity Library | Green Web Foundation [02:29]
Green Web Foundation [3:30]
Wholegrain Digital [4:04]
WebsiteCarbon.com [4:07]
Solarpunk [24:12]
The Wabi Sabi Web - Branch | Tom Greenwood [28:49]
Embracing friction: A conversation with Luna Maurer and Roel Wouters - Branch | Marketa Benisek [30:08]
Talking it out: Restoring information ecosystems through authentic human connections - Branch | Bárbara Paes and Olivia Johnson [31:26]
Echoes of electronic waste - Branch | Joanna Murzyn [31:37]
Connectivity, infrastructure and the defence of the Amazon's socio-biodiverse ecosystems - Branch | Hemanuel Veras [31:41]
One Movement, Four Wings: Connecting climate strategies - Branch | Melissa Hsiung [32:47]
Ministry of Imagination - Branch | Rob Hopkins [34:58]
Issue 8 community-assembled playlist - Branch [36:30]
Curiously Green, the green web newsletter - Wholegrain Digital [39:59]
co2.js | The Green Web Foundation [40:44]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Marketa Benisek: Perfection is essentially the enemy of progress. So it doesn't really matter if something is imperfect, but we can build on it. That's when Hannah came up with the idea that actually we could collaborate on the next issue of Branch Magazine, and this could be the theme.

I'm your host, Chris Adams.

Hello, and welcome to another episode of Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. The world of green software and digital sustainability has come a long way since the Green Software Foundation was created three years ago.

The Software Carbon Intensity Specification is now an international standard. All of the large hyperscale cloud providers in North American Europe now publish guidance about incorporating sustainability principles into building on their platforms, and we're also seeing unprecedented policy interest in both understanding and managing the emissions associated with digital services in general.

There's more to sustainability than just making our software greener, though. If we want a truly sustainable industry, we also need new visions of how we relate to technology and the role it might play in a more positive, regenerative future. One such vision. Is that laid out by digital magazine Branch, a publication featuring writing from many of the active feature figures in the Green Software Foundation and one of the first ever websites to pioneer a carbon aware design that changed in response to the carbon intensity of the underlying electricity grid that its servers ran on.

Originally from climateaction.tech and published by the Green Web Foundation, Branch is a twice yearly publication, and at the time of this podcast being recorded. It's now into its eighth issue. Now, in the interest of disclosure, I should probably share that I've written for Branch a few times and that I worked on the grid intensity library, the JavaScript library, that the site relies on for its carbon aware design features.

I'll try my best to keep my bias out of this episode, but it's worth acknowledging that before the GSF, The Green Software Foundation was as prominent as it is now, is one of the better known non commercial places to read about the intersection of software and sustainability and see how folks are doing it.

I'm joined today by two of the contributing editors, Hannah Smith and Maketa Benesek, to talk about community publishing as a tool for change and the lessons learned publishing a magazine that tries to walk the walk as well as talk the talk on sustainability in the digital realm. So, Han, Marketa, thanks so much for joining me today.

Can I give you the floor to introduce yourself and what you do?

Hannah Smith: Great. Well, thanks, Chris. Thanks for the introduction. And yes, you are definitely a little biased in talking about Branch. I think that's fair to say, but anyway, I'll introduce myself. So hi, my name is Hannah Smith. I am director of operations at Green Web Foundation, where some of you might know, I work alongside Chris and I'm very excited to be here.

I was a WordPress developer for a long time. I also run a meetup community called Green Tech Southwest. And quite recently, sort of over last summer, I was very involved in a project looking at grid aware computing as a, as an alternative or an improvement on carbon aware computing. So, Marketa, I guess over to you.

Marketa Benisek: Thank you. Thanks for the intro, Chris. So, my name is Marketa Benisek. I work as Digital Sustainability Lead at Wholegrain Digital, the authors and kind of creators of WebsiteCarbon.com, the online kind of first and original online carbon calculator. And I'm really, really excited to be here. And I'm super excited about this project. I also did a climate reality training with Al Gore, uh, during the pandemic and I'm a carbon literate professional. So that means that I essentially, I'm just really passionate about explaining people kind of the essentials and the basics of the climate science so that they understand the urgency of why we need to take action.

Chris Adams: Thank you for that, Marketa. Now, before we dive into the world of Branch and community publishing and all that, we normally ask where our guests are calling from to get a bit of background color, really. Han, can you just tell me a little bit about where you're joining me from? And then if I hand over to Marketa after that.

Hannah Smith: Yeah, so thanks Chris. I am dialing in from the temperate rainforest in Exmoor National Park, which is over in the southwest of England in the county of Somerset. And today, as I look out the window, all the trees are coming into leaf and it is a truly, truly glorious view.

Chris Adams: Wow, I spent 30 UK and I'd never realized there were rainforests. So you said temperate rainforest as opposed to a tropical rainforest. Is that how it works?

Hannah Smith: Absolutely. Yeah. I didn't know that temperate rainforest was a thing until I moved here either, but there is this growing movement to reestablish these, these ecosystems or, or just make people aware of them. It's basically a forest, which is on the coast and pretty wet, and that's how you end up with a temperate rainforest.

It sounds, it sounds exciting, but you know, you need good wellies and a good Mac and not to be afraid to get a bit muddy if you live here. Yeah.

Chris Adams: Marketa, over to you as well.

Marketa Benisek: I am joining this call from Prague in Czech Republic. That's where I'm based currently. I relocated here a couple of years ago from very busy London after having a kid. So yeah, it's just, it was a beautiful spring weather just about a week ago. And now we're back to snowy weather. So we got all of those like winter jumpers. And yeah, we'll see what happens.

Chris Adams: Wow. Okay. I did not know that you were in Prague actually, Marketa. I thought you were still in London. Okay. I should probably go myself and also introduce myself because I realize I've just done this massive introduction without saying who I am, what I'm here for. So folks, if you are listening to this podcast for the first time, my name is Chris Adams.

I am the executive director of the Green Web Foundation. I'm also one of the organizers of Climate Action Tech, which is a online community full of essentially climate aware techies, I suppose. I also worked as one of the policy chairs in the Green Software Foundation, and that's partly how I know Han.

As Han mentioned, I did work on the Branch Magazine, a few years ago, basically. So I have some kind of background there. I'm calling from Berlin where today it's been sunny and hailing at the same time. So that's my kind of contribution to the meteorological diversity, I suppose. All right. Okay. So I think we're all sitting comfortably and we've got a good idea of where we're all kind of calling from today.

Maybe we should talk a little bit about Branch. So I understand there was a new theme for this that the two of you wanted to kind of talk about because it's, we've had different themes for publishing different issues, but the theme this time was I think finding beauty in the imperfect. And I believe this was on, this is a kind of reference to some of the struggles we see when people start trying to adopt digital sustainability principles or kind of working around that.

Would either of you maybe just explain a little bit about how this came about or maybe what drove you to kind of come to it from this point of view, actually?

Marketa Benisek: Yeah, absolutely. So I should probably start by saying that Hannah was one of the first to I met a lot of female voices that I came across shortly after I met Vinita Greenwood, the co founder of Wholegrain Digital. And she was one of the first female voices that I've heard talking about digital sustainability and digital carbon emissions and all that. And I introduced myself, this is it must be about six years ago, seven years ago, something like that. And then we kind of stayed in touch ever since, you know, we collaborated over several projects and we just stayed in touch and connected about once a month over a call just to catch up and talk about what's new in the digital sustainability world, so to speak. And about six months ago, we got together on a call. We were both pretty upset about some recent articles and mentions of the different tools and, and projects that people criticize for their imperfections. And it felt, it wasn't really the criticism itself that was so upsetting, it was mostly the fact that they made it sound as if there's just no point in trying

if it's not perfect in the first place. And so we got to talk over this and, and about this, and we just realized that this is something that we feel really, really strongly and passionately about, and we wanted to address it and kind of just voice out that, you know, perfection is essentially the enemy of progress. So it doesn't really matter if something is imperfect, but we can build on it. That's when Hannah came up with the idea that actually we could collaborate on the next issue of Branch Magazine. And this could be the theme that would kind of give the platform to lots of different voices in this field and to people from CAT community on like, what can actually be done and what is already happening in this space that is imperfect, but it's still very much meaningful.

Chris Adams: Okay. So thank you for that, Marketa. That does provide some of the context that makes it a bit easier to understand this theme. Han, maybe I could maybe hand this over to you to talk a little bit about why you chose to explore this theme using a magazine like Branch, for example, rather than trying to get a white paper written or something like that, perhaps.

Hannah Smith: Thanks, Chris. Yeah. So, so as Marketa has said, we kind of had these conversations and we kind of realized that we felt that this narrative, which can be quite dominant in the tech industry of, you know, of always having perfection in your data or, you know, always striving for everything to be a hundred percent accurate.

We kind of felt that we had a lot to say on it, and we also thought that there'd be a lot of members within the climateaction.tech community that would have a lot to say on it too. Perhaps they were in a place of work where they were having to overcome some of these, you know, some of this pushback from their own team, or perhaps they're in, you know.

Developer communities that were maybe not embracing these ideas again, for these reasons. So we kind of realized that Branch would make an absolutely amazing place to bring loads of different voices. From across the whole tech industry together into one place. And, you know, Branch is very much founded on this idea of bringing together different voices and perspectives from even different industries, not necessarily just the tech sector.

You know, we know that there's people working in research or people who work with tech. You have an awful lot to say here too. So Marketa and I were like, yeah, okay. Let's get the CAT community involved in this. Let's see if people want to tell stories from their places of work, or maybe there are founders of companies that want to talk about how they're embracing things in different ways.

Maybe there are some people who are, you know, uh, practitioners, maybe freelancers, maybe independents. Maybe people from research organizations. Let's get all of these voices together, let's get all of these perspectives together, and let's kind of create, we've been talking about it as a bit of a tapestry, a kind of tapestry of views that start to answer these questions or start to explore what it actually means in practice when you're working in digital sustainability to find beauty in the imperfect.

So what is beautiful, and what is imperfect? So yeah, we've kind of thought Branch just lent itself to be the perfect place for people to do that. And again, the awesome thing about Branch is it's a non profit magazine, as you well know, Chris. So that awesome, we thought made it easier for people that were perhaps within organizations that maybe don't have a platform to speak within their own organization.

They could come and use Branch. And be able to talk about things in an open way that perhaps their organization might not be so cool with them publishing on their own blog. So it feels like a great opportunity to bring a really diverse, in the biggest possible way, set of voices together, I think.

Chris Adams: Okay. I see. So you've got this idea of diversity. And one of the key things is that it's not just one person, you know, the internet isn't just for like developers, for example, there's other, there's users and there's other people who are affected by it that it's worth hearing some of their voices involved in it as well.

Okay. Thank you, Hannah. That does make it a little bit clearer, actually, since you spoke a little bit about Branch and actually the design part of it, maybe we could just switch gears for a second and talk a little bit about some of the realities of trying to make some of this, because one thing we mentioned before is that Branch was one of the early magazines that essentially changed this design based on how green the grid is, basically.

And I figured maybe we could just provide, have a bit of kind of catnip for the nerds talking about how that was actually done. So maybe you could talk a little bit about, okay, yeah, that's a cool idea, but. There's all these considerations from a production point of view. So maybe are there any kind of particular challenges you found or things we would have bear in mind if they're going to take their first steps into designing, say, kind of a grid aware or carbon aware design like this?

Hannah Smith: Yeah. Well, thanks, Chris. So Marketa, I might pick that question up if that's okay with you. Cause I

Marketa Benisek: Yeah, please do.

Hannah Smith: yeah, so, I mean, Chris, I know you were sort of involved in building the first issue of Branch. I came on board to help out on the second issue and beyond. And so I think it's worth saying that first of all, the way that Branch responds to the grid is that it provides a different experience based on what's going on, on the grid on a certain day.

So if you've got loads of fossil fuels on the grid, on a particular day, you don't see the images by default. So you have to actually opt in to see the images. If you've got a nice windy, sunny day, maybe then the intensity will be low. And so you will see the images. And then you've got this middle, middle scenario there, where the images are converted into a really fuzzy kind of web format so that you can see the images, but there's a clear and obvious degradation.

The thing I really, really like about Branch and the way it does that is that it's, it's giving you a different, um, experience based on the renewables that day and in that place. So it's, it's more responding to demand or, or supply really of what's on the grid. And I think that's quite clever. So rather than sort of doing stuff like shifting around location or time based. I find that that is actually a very, very visceral way that people can understand what's going on in terms of the supply coming on to the, onto the grid. Actually, I find it throughout Marketa and I putting all the articles up for issue eight, I found it brilliant. I've now know exactly what's going on on the given day as to what the renewables are, because the background color of Branch tells me, which is super, super handy.

But actually we did notice just last week as we were kind of finalizing some of the images that well I don't know if it's a new level of carbon intensity or if it's a just a level of carbon intensity we don't get very often in the UK. But I, to my absolute shock and horror, went on Branch and it was grey and I was like, oh, That's not a color Branch.

Chris Adams: What's going on here?

Hannah Smith: That's not supposed to happen. What's going on here? And we looked into it and realized that actually the grid was running at very low carbon intensity that day, which kind of made sense because when I looked out the window, it was super windy, yeah, super sunny day and we were like, all right, cool. So I guess like, that's quite a fun bug in a way, or a fun thing is that, you know, I guess our expectations over different carbon intensity levels can change a little bit, or, uh, you know, if you're connecting through to an API, something, something suddenly shifts and, and you get a bit of a surprise.

So that's definitely a bit of fun to be aware of on the production side of things.

Chris Adams: I remember you, I remember you talking about some of this actually, like, like, you thought, oh, the grid's never going to get that green, right? And then you see it changing, like, oh, Christ, you need to update the design because of this. Yeah. This idea of like having different thresholds was, you know, Yeah, that was, that was a new thing because I remember you asking me and like, it took scratching my head a little while because it doesn't seem to be anything wrong with the code.

And then you realize, oh, the API that tells us the carbon intensity must then have started to introduce some new levels that we need to design for. So yeah, that's, uh, that, that, that was, that was actually kind of a fun thing from this. Okay, cool,

Hannah Smith: Happy days though. I mean, I'm happy that we were running on very, very low intensity on those days of, you know, Hey, that's what we're here for, right?

Chris Adams: Yeah, that was, I guess, the technology being somewhat more enthusiastic than the design five years ago, or four years ago, perhaps. So that's like one of the examples that we need to take into account. And maybe one thing that I'm glad you mentioned the idea that when you first visit this, you don't see the images.

One of the things that I think is There was some fair amount of focus, which is often overlooked is the accessibility of this, like deliberately trying to make sure that if someone can't see an image, you at least have some way to make it perceivable to people, for example, maybe we could talk a little bit about some of that thing, because people tend not to, this may be a thing that is overlooked in publishing, that's actually, I guess, part of the web, maybe, and maybe you could expand a little bit on the accessibility aspects before we talk about some of the actual content of Branch, perhaps, Han.

Hannah Smith: Yeah, sure. So I mean, something, when you're uploading content to Branch, you realize just how important alt text is on your images. Because that's one of the key features of Branch is that if you have a very carbon intensive day, so e. g. there's lots of fossil fuels, bad, that's not what we want, you don't see the images.

And what you see instead is the alt text and a caption, which describes it to you. You do see kind of, you see a grayed out placeholder of the image. So you're very aware that there is an image there, but you can click show image and you can see the image and opt for that to get downloaded to your machine and opt for for that, that additional action to happen. So accessibility is a really interesting aspect here as well.

Chris Adams: Yeah. Cause I guess on one level you're, that's kind of like forcing how other people might perceive the web, if it might be, might be partially cited or something. It's kind of some of the decisions there to kind of foreground the fact that the way that you experience a digital service may not be the only way it can be experienced, for example.

Hannah Smith: Yeah, it's, it's interesting. I love the way that Branch does that. It plays with almost messing with your perception of what a website is and can be, and I, I mean, that was always one of the key ideas of it was to imagine different ways that a website could be, it doesn't always have to be the same, exactly the same.

It should be responding to different things. And also remind you that people interact with these things in different ways as well. As you rightly said there, Chris, you know, not everybody will see the images all the time.

Marketa Benisek: It's a great preview into someone else's world and how they see the world, how they see the digital world as well. I think this, this often doesn't really get enough attention. That not everybody sees the images in perfect high res, you know, some people might see it blurry, some people might see it, I don't know, pixelated, whatever, you know, so paying attention to these alt texts and, and just a different setting for people who might have different needs and different kind of health conditions is really important and the entire web. should be built that way. I mean, that's why we have this technology in the first place. So the fact that Branch has that is really important, I think. And I'm just, that's just one more reason why I love it so much.

Chris Adams: So, Marketa, you're right, actually, there is this notion about, I guess, the web being something that should be accessible for everyone, because, like, the internet is for everyone like this. Maybe we could talk a little bit about, actually, some of the high level themes in the content for this issue. Like, are there any trends that you felt really deserved attention or could do with more people writing more openly about, for example?

Marketa Benisek: Yeah, absolutely. I think that, first of all, you know, the, the theme, we were pretty set on, on the theme, Hannah and I, you know, since. The very early days, finding beauty in the imperfect. I guess as two women working in tech, you know, it may come across, it may have come across slightly more poetic or feminine to some. And I think that some people, there was a need to kind of explain what we really mean by this. So we spent quite a bit of time on this explanation, you know, through the open call page, just to really clarify what this means. And obviously it was an invitation for everyone, not just for women in tech. I just wanted to make that really clear, but yeah, I, I mean, after we set the open call page on Branch, we received a really great number of submissions that we carefully read through and selected the best ones that really spoke to the theme and then we narrowed it down to I think seven or even eight categories and we got super excited and obviously we wanted to involve everything and you know, we wanted to give a platform to everyone who even reached out and things like that. But then we got a bit more realistic and we realized that actually, you know, we really do need to narrow it down. So we narrowed it down to four key categories that kind of spoke really strongly to just new ways of building just and sustainable web that is also a humane web. And so now we have four categories that, that they are meaningful connection, kind of solar punk and imagining different future, new ways of looking at the design philosophy and how we might be able to build the web.

You know, that is not perfect or not in that kind of usual perfect way, seamlessly perfect. And then finally, obviously the perfection itself and just not letting it be the enemy of progress. So those are kind of the four key categories that were really, really, important for the theme, I think. And yeah, I just really can't wait to share that with the world.

Chris Adams: Okay, and that's presumably the logical grouping inside the content. So someone, when they go to the magazine, they'll see it grouped like that. So they might look at dive into one, for example, and then the other, right?

Marketa Benisek: Yeah, absolutely. I mean, obviously this wasn't a straightforward process. You know, we spent a long time kind of going through all of those images, grouping them together. But shortly after we received all of the submissions, we could see patterns in some of these, you know, submissions and we obviously group them together and kind of organically these four key themes.

Chris Adams: Ended up being the logical group for this.

Marketa Benisek: And yes, ended up being the four categories, yeah.

Hannah Smith: And I think I might just add a little bit onto that as well, because the, the whole idea of this theme was that we wanted to explore the subjective ways that people Think about finding beauty in the imperfect in digital sustainability, whatever lens they're looking through, be that developing carbon calculation methodologies, be that designing things, being that the way that people use the internet, being that even questioning, should the internet even be a thing in the first place?

So we did throw it very, very wide and then as Marketa explained, you know, we, we grouped things. And then what we kind of discovered, the thing that makes me very happy because I'm quite nerdy and I do like everything to be neat and tidy, is that those four categories then yes, you're laughing because you know this is true because we then noticed that those four categories almost become a journey in themselves.

So, as Marketa said, that the first category we saw was around meaningful connection. And I guess the first thing that you ask yourself when you're Looking for beauty in the perfection when you're designing something is, why? What are we designing this for? And we realized that actually, at the very, very heart of it, what the internet should be about is creating meaningful connection.

Not just connection for connection's sake, but meaningful connection that's humane, that is just, and that is sustainable. And we kind of really noticed that theme. And then we kind of thought, alright, so you know what you want, So the next thing that happens is, well, in an ideal world, what does that success look like?

And that's where that imagination piece comes in, these ideas of Solarpunk. And if anyone's listening and hasn't heard of Solarpunk before, you're in for such a treat because it's just such a cool movement. It's all this idea of kind of imagining regenerative, renewable futures. So you kind of, once you know what you mean, you know what you want, EG, a meaningful connection, you start to imagine what it looks like.

And we had so much fun with all the pieces that imagined the future, like there's some absolutely amazing, beautiful, yet imperfect stuff in there. And then once you're imagining things, what we realized is the next step on your journey is the logical question is, well, how are we going to make it? And we realized then that we had this grouping all around design philosophy of people talking about all these different ways that they want to make this stuff, that they want to make things on the internet, that they want to design things on the internet.

And then we came to this fourth category and I wouldn't say it's the last one because the other thing we realized is that everything is just so deeply interconnected. And I think you realize that when you're doing anything in sustainability is that connections that are everywhere. It's never finished and it's never linear.

But we noticed this last question was around, well, let's talk about some of the things that get in the way of us designing our things or building our things. And we realized, well, perfectionism is this key thing that seems to come up and up again and is so often used as a kind of defensive measure for why we shouldn't bother.

We kind of realised that that was almost the last stop on the journey,in a way, probably before you go around again. So the categories do stand in isolation from one another, but can also be read and taken on a journey as well for kind of completing a whole process here as well. So when, when that kind of made itself obvious, I was very happy cause that, that felt very neat, very neat and tidy, so-

Marketa Benisek: Yeah that was a beautiful day that one

Hannah Smith: It was a beautiful day, but obviously there's imperfections there too, even in that.

Chris Adams: I see. Okay. Thanks. Maybe I'll just bring, hand over to Marketa. So we've spoken a little bit about themes and like the kind of, the motivation for this. Maybe we could just move to like some of the concrete pieces, for example, if we would have like a preview of some of the things you'd like to draw attention to, or kind of give a sneak peek, sneak preview of this.

Are there any articles you particularly enjoyed or you found a lot of, that you, would like to draw people's attention to, for example, if I hand over to you, Marketa, first, and then Han, I'll give you a chance to kind of name some of your favorite children, as it were, or favorite babies, or whatever, whatever phrase you want to use for this, then we can see what we've got there.

So, Marketa, you first, if I can.

Marketa Benisek: Yeah sure so I guess throughout this process, what I enjoyed the most was seeing these kind of similar patterns occurring in different places around the world. That was really just fascinating to, to witness and to be a part of, because, you know, all of these authors are from different parts of the world. They are talking about different problems, different solutions. And yet they were using very similar language. So for example, we mentioned perfectionism as kind of one of the obstacles to building a more just and sustainable and humane web. And interestingly, several people, at least my authors, you know, the authors from my group, they kept mentioning friction and the need of friction.

Somehow it seems like friction and imperfection is what makes any experience more human. And the web and trying to make it kind of perfect and frictionless and seamless and everything like that, that goes against that human experience, that humanness in the first place, which was really interesting, kind of realization out of this whole process, this project. Some of the articles that I really love, obviously, there is Tom Greenwood's article, the co founder of Wholegrain, who talks about the Wabi Sabi Web. And I just really love how Tom thinks and how he articulates his, his thoughts. It's just, it's kind of like a beautiful journey through his mind. And so Tom picked a Japanese philosophy called wabi sabi, and he used several examples of how we might use it to build a more humane web.

So wabi sabi is a concept where you kind of accept the imperfection and kind of the fact that all things All of these are beautiful because they reflect the time and the experiences that they have been through, but we don't necessarily see that on the web. And maybe that's the problem. Maybe, you know, all of the data are kind of like set in stone and we don't see how the users behind the data changed and evolved over time.

So this is just a really interesting and very fascinating exploration of how we might. Imagine a different kind of web called the Wabi Sabi Web, so to

Chris Adams: S W S W, instead of dub dub dub, right? So, W W W, W S W, perhaps.

Marketa Benisek: Yeah. Yeah,

Good point. Then also just going back to friction, I actually, Hannah pointed me to a really fascinating project called Designing Friction by two artists based in the Netherlands. And I did an interview with them and they really very much go against this whole concept of what the web is becoming. You know, it's becoming this unhuman frictionless space that doesn't really allow for playfulness. And they are against it, not only as artists, but as parents, as human beings. And so it was really interesting to read through their principles of how we should kind of make friction part of the digital culture.

It should be something that is celebrated and even sought for. You know, it shouldn't be something that we should avoid. And they give a whole bunch of examples of how we might be able to do that. And just to clarify, this is not to say that friction should make the internet slow or, you know, it's not about making the user experience awful. It's just about making the whole user experience more human. It shouldn't be about users, but it should be about people. So to speak, and then we've got a whole bunch of real examples. Uh, from authors who are actively trying to understand what the internet is like for people in different parts of the world. And yeah, for example, there's a really lovely article from Barbara and Olivia from the Engine Room who talk about their research on the information ecosystem in Latin America. And another one about e waste in India and another one from Hemanuel on the importance of the internet connection for the peoples of Amazon. In order to help them kind of defend not only their own rights, but also the nature around them that they are living in harmony with. So these real examples of what the internet is like in different parts of the world, not just in our kind of privileged, rich, you know, Europe or our part of the world, has been just really so eye opening and I really, really strongly recommend people to, to go and have a look.

Chris Adams: So you spoke a little bit about say, you know, friction, making things a bit more kind of legible or possibly understanding where it's not just kind of totally seamless, fast thing where there's no agency for people at the end use, but you've also spoken, and there's maybe some content about essentially this intersection of climate and technology, how it manifests in different parts of the world, like you just mentioned, those, those writers from various places.

Thank you for that. Okay, Hannah, can I hand over to you perhaps to talk about some of the things that have been catching your eyes, that you direct people's attention to now that you've been working on this magazine for the last, like, couple of months, for example,

Hannah Smith: Yeah. Thanks, Chris. I think there is a piece written by Melissa, who's one of the organizers at climateaction.tech. And she has talked about a concept using a four winged butterfly. And the idea of the four winged butterfly is it represents a social transformation. And I think if we're going to transform the internet to be a more healthy place, to be a more humane place, to be a more just place, to be a more sustainable place, I think this is something everybody should read.

She talks about the, it's not her concept. It's, it's a concept from somebody else. But she's put it into a digital sustainability context. And the basic idea is that you need these four different elements of change. So each element of change is a wing of the butterfly and a butterfly needs all four wings or all four parts in order to be able to fly.

So it's this idea of, yeah, it's a lovely idea. I think it's a really, really good one. And the way she talks about it is, you know, she says, well, look, if you're looking at something that somebody's trying to do, if you're looking at a project and you think, ah, this is rubbish, this is no good, this isn't, you know, perfect, what about X, Y, Z, she said, well, you know, you can look at these other wings of the butterfly or these other aspects and actually maybe understand that, you know, this particular change or this particular initiative you're looking at is only one part of the butterfly.

But if it's connected with other parts of the butterfly together, it actually makes a movement, you know, a social change that can move, that can fly. And I loved that practical toolkit that she was giving there to say, hey, you know, if you think something's imperfect, perhaps You're only narrowing in on a really tiny part of it and, and if you zoom out, there's yeah, look at the bigger picture, you might realize that this is an absolutely essential part of making other things move or, or of giving other areas flight.

So I loved Melissa's piece and I, I really recommend that for anybody who's, who's interested in, in creating change and, and making progress. And I'll say two other quick things. One that just, it was a bit of a late addition to Branch, but it's just absolutely captured me, floored me, is there, it's a piece by Rob Hopkins.

And he's just published something called the Ministry for Imagination, which is the work coming from a hundred podcast episodes where he invites people onto the podcast. To talk about the world. Yeah, it's pretty cool. So you've got a long way to go, Chris, a hundred episodes. So he has sort of through working with these different people, they're interviewing on the podcast, come up with this manifesto for what it means to imagine a better world.

And it's not really just tech related. It's far broader than that. It's across all of society, but it is an absolutely incredible example of why imagination is important. And how sometimes it can lack a little bit in the digital space. I think we, we can get a little bit narrow minded about things perhaps, or see things in a bit too black and white and maybe not, not be playful enough.

So here's Ministry of Imagination. I can't recommend strongly enough that everybody takes some time to have a look at that for like an absolute definition of the power of imagination and solarpunk. And then the last thing I'll, I'll mention super quickly. Is when Marketa and I were looking at all of the articles, we had a hope at the beginning that we would not just have loads of writing, but that we would have art and we would have poetry and music.

And we noticed that poetry and music was a little bit lacking in the edition. So we've set off and created a Branch issue eight playlist. Where we have asked a whole bunch of people from across the community to suggest songs that motivate them to take action. But then also pick out lyrics that really speak to them.

And actually I had a chance to listen to the playlist before launch. It is so eclectic. We've got rock, we've got classical, we've got trance, we've got acoustic, we've got reggae, we've got funk. We've got like, hardcore dance music. We've got all this stuff. And I think what that speaks to me is that there is no perfect music choice to get motivated by.

Everything is different. It's, it takes a diversity of perspectives and a diversity of, of vibes, energies to make change. And I think the playlist really, really embodies that. And I think it also speaks to. The wider movement as well around needing a diversity of viewpoints and a diversity of cultures.

Songs are all in different languages as well, but actually you bring all of that together and you get a real tapestry. And I loved listening to the songs and I loved reading the lyrics that people highlighted and their reasons for including these songs. So it's, I know it's not that much to do with tech, but we are all humans at the end of the day.

And I think music is, can be a really, really big motivator and something that really brings people together as well. So I'm a big fan for the playlist as well that came about and you can find that on Spotify too.

Chris Adams: Okay. All right. So it sounds like there's, okay. Let's say that we have prose, poetry and music and references and links to other forms of music as well. All right. This is, this is sounding quite rich actually. Okay. We're, we're just coming up to the end of time. So this is basically a kind of project. If there's something of a passion project, but there is still a focus very much on like digital culture and things like that.

I understand that you do more than just make magazines. So Marketa, we should probably tell people where to go to look for this. I mean, maybe is there a domain name or a search search term they should be looking for if they want to learn about some of the things we've been talking about?

Marketa Benisek: Yeah, absolutely. So if anyone is interested in reading any of these articles and much more, they can go to branch.climateaction.tech. They can find all the information there and all the articles, the playlist and everything else that we've been working on over the past four or five months, something like that. Yeah. So that's,

Chris Adams: so Branch. climateaction.tech is the way to look up, and that'll give you the most recent issue. So we're recording this on the 22nd of April, but I believe it's going to coming out in the next week or so, maybe the 25th, I think it was the date that I've heard. And presumably that'll be the big day.

Okay. So when this comes out. If you're listening to this, it's probably out already, so you can check it out yourself along with the other previous seven issues worth of content that you have there. All right, and if I understand it, this isn't all that you do, so maybe I could actually just, while you're here, give you a bit of space to kind of talk about some of the other things you're doing.

So, Marketa, for example, where else should people be looking if they want to learn a little bit about digital sustainability or some of the ideas or some of the kinds of things that we've discussed, for example.

Marketa Benisek: Sure. So we frequently publish new articles on our blog. So people can go to wholegraindigital.com and then go to our blog and they can find lots of stuff related to digital sustainability, humane web thinking, that sort of thing over there. And I'm also very happy and proud to be on the team that creates a monthly newsletter called Curiously Green. And that's all about. All the things, digital sustainability. So yeah,

they can just go to,

Chris Adams: I really like that as well, actually. I'm glad

Marketa Benisek: thank you

Chris Adams: And Han, if people are curious about where you've been coming from, I suppose, is, are there any projects or things you'd like to draw people's attention to in the last few minutes that we have for this?

Hannah Smith: Yeah, thanks Chris. So I guess my day job is at the Green Web Foundation. So you can head over to thegreenwebfoundation.org and you'll see a lot of my work represented there. You'll see a lot of blog posts coming out from me, where we talk about the Green Web Dataset. We talk about our open source co2.js package, and also there'll be some really nice work coming out soon.

With some really nerdy deep dive work into the practicalities of carbon emission estimates for digital. So Chris, obviously you and I are working on that together. Yeah, head on over there and I think you'll get a good insight as to what my day job involves on that website.

Chris Adams: Okay. And just before I go, I believe outside of life, you also have run, you mentioned Green Tech Southwest, that's another community that you're involved with.

Hannah Smith: Yeah, that's right. So Green Tech Southwest, it's, it's kind of location based. The community at its heart is in the southwest of the UK. Anybody is welcome to come. All of our events are online. We've got another event coming up in May the 2nd, where we'll be looking at a methodology for measuring CO2 emissions.

And we'll also be looking at some really cool visualizations of renewable energy projects in the UK. So if you can't make it in person, if you're not based in Bristol or near to Bristol, you're very welcome to join online. And that is applicable to anybody, wherever you are in the world. You're super welcome to come and be a part of that community.

Chris Adams: Cool. Thank you for that. All right. I think that takes us to time actually. So folks, thank you so much for giving me your time and talking to, and giving this sneak peek of a project that's kind of close to all of our hearts, I suppose. And hopefully some of the listeners who get exposed to this, I guess all we have left is to say, thank you very much.

Have a lovely week and yeah, best of luck with the launch.

Marketa Benisek: Thank you. Thanks for having us.

Hannah Smith: Thank you, Chris. Thanks for having us along.

Chris Adams: All right. Take care, folks.

Hey, everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show. And of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode!

Building Green Software

11 april 2024 | 36 min

In this episode of Environment Variables, host Chris Adams introduces the co-authors of Building Green Software - Anne, Sara, and Sarah. Through candid discussions, they explore the process of writing about green software development and highlight key insights gained along the way, touching on the interconnectedness between sustainability and existing best practices in software engineering, and emphasizing that embracing sustainability isn't about adding extra tasks but rather integrating it seamlessly into existing protocols such as security, resilience, and monitoring. Join for a thorough conversation on the lessons learnt writing the newest book on green software.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Anne Currie: LinkedIn | Website
Sarah Hsu: LinkedIn
Sara Bergman: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

QCon London [30:36]
NDC { Oslo } [32:00]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Anne Currie: A lot of the best practice in sustainability is often also the best practice in other areas. So it's just one more good reason to adopt best practice in security, in resilience, in monitoring, you know, it's not something you're going "oh my goodness, mate, we have a whole extra thing to do." It's just another reason to do all the things that you should be doing anyway.

Anne Currie: Absolutely. Yes. So my name is Anne Currie and I am the CEO of a green training company called Strategically Green. And I'm also one of the co chairs of the GSF community working group, and I am the lead on the GSF's new Maturity Matrix project. So yeah, those things, and I'm one of the co-authors of building green software along with, and I will hand over to Sara.

Sara Bergman: Hi, my name is Sara Bergman, or Sara Bergman, Bergman. I never know how to say it in English because it's so different from the Swedish. But yes, I am a senior software engineer at Microsoft, even though I'm currently on maternity leave, hence why brain is not fully switched on, and one of the co-authors of the book Building Green Software.

Sarah, over to you.

Sarah Hsu: I've been waiting for that sentence for so long. Sara and myself have been working together, but we never really shared a stage. So Sara over to you, or Sarah over to you, has been our dream for a long time, and now it's finally happening. But anyway, hello everyone. My name is Sarah Hsu. Similar to Sara, my surname is actually pronounced very differently with an English accent.

So yeah, Sarah Hsu, it's fine. So currently I'm a site reliability engineer working for Goldman Sachs, but I'm also one of the co-authors of the Building Green Software book. And this is why we are here today. I'm also the project chair for the Green Software course that we recently launched with Linux Foundation a few years ago, and we currently have over 70 000 completion, which is amazing.

But yeah, anyway, that's me.

Chris Adams: Cool, thank you, Sarah, sara, and Anne. I'm saying hello from Berlin and Sarah Hsu, you're calling from London.

Sara, whereabouts are you calling from today?

Sara Bergman: I am in Oslo, Norway. So yeah, I am Swedish, but I live in Norway. This does not matter to anyone who lives south of Denmark, but for us Scandinavians, it matters a great deal.

Chris Adams: Cool, thank you. And Anne, where are you calling from today as well?

Anne Currie: I'm calling from London as well. So we're not quite as international as all that.

Chris Adams: Oh, well, it's not bad. Four different cities in three nations isn't, isn't too bad, I suppose. All right. Okay, so if you're new to the Green Software Foundation podcast, my name is Chris Adams. I'm one of the policy working group chairs in the Green Software Foundation myself. I'm also the executive director of the Green Web Foundation, a Dutch non-profit focused on reaching a fossil free internet by 2030. And before we dive into what it's like writing books about green software, here's a quick reminder that everything we talk about will be linked in the show notes on this episode. And because we're a software engineering podcast, the transcript will be available on GitHub in markdown format. So if there's any typos or there's corrections, then we do accept pull requests.

And if there are pull requests that are accepted, we'll then credit you and shout you out in the next episode. All right, then. So, itooks like you're sitting with us comfortably, Sara, Sarah, and Anne. Should we get into it and talk a little bit about writing green software and or writing about writing green software?

Yeah?

Anne Currie: Absolutely.

Chris Adams: All right. Okay. I'm going to hand to you, Sarah, first. If I understand it, this is one of, your first book writing about green software. So maybe you could just talk a little bit about how you found that process, writing about, say, a developing field like this.

Sarah Hsu: Yeah. It's my first time writing a book and it's also my first time appearing on a podcast. So maybe we can make another episode on this experience. I'm joking. So what was it like writing a book on green software? Honestly, I don't want to sound too cheesy, but it has been a really fulfilling experience.

And why is that? It's because finally I can use what I'm good at, which is software engineering, and use what I've been training for the past seven to 10 years to have a positive impact on the environment. And it just, it's great to see that ripple effect from something I'm good at and instead of just individual actions.

So I feel that really, really close to my heart. And of course that Anne and Sara made, they're not, they're more than just the icing on cake. They're absolutely integral to this entire experience. I've always admired them since I first met them three years ago. And working with them has taught me so much.

Yeah, couldn't have asked for better co-authors. And, and thankfully we all share similar humor. So when one of us put a joke or put a pun in a book, it doesn't get like, "this is not funny," cross it out. So I think that's also another icing on the cake. Yeah. But yeah, hopefully that's not too cheesy of an answer...

answer but

Chris Adams: No, it's okay. It's okay to be genuine if you've enjoyed working with other people, especially if you're doing something you believe in. Absolutely. All right. So that's the skinny from someone who's basically been writing books like this for a while. Anne, this isn't your first rodeo, you've written a few other books.

And I know that you, and we've spoken a little bit about all kinds of sci-fi books before, so maybe you could, I could just touch on you about maybe some of the differences between writing about wacky sci-fi or writing about things when you're, if you forgive the term, at the coalface of software development trying to figure out what to do next here.

Anne Currie: Yeah, that's a good point. So this is my tenth book, eight of which though were sci-fi, were futuristic sci-fi, but kind of like more speculative fiction than hard sci-fi. So it was all kind of surprisingly similar to writing an actual book about stuff that's happening now. But oddly enough, writing the fiction is much easier because you get to make it all up.

Whereas, although I try not to make it up, I try to keep it quite, quite, quite,

realistic. It's hard sci-fi, Yeah. but the book was much more difficult because the book, I mean, I think this is true for all of us. It was really important that we got everything right and we, and we phrased it right and we made it so that people could read it. Which I also want to do with my books, but they aren't quite so important, if you know what I mean. This feels like an important book, or we need there to be an important book, so hopefully this will be the important book, because if it isn't, then it'll be quite a long time before there is another one.

It's a lot of work, writing a book.

Chris Adams: I can definitely concur with that. Alright, so, one thing I didn't mention was that this was actually published by O'Reilly, who have some form writing about sustainability in this field. There was a book by Tim Frick, in the mid 2000s, called Designing for Sustainability, which was focused on web. But as I understand it, this has more of a focus on parts of cloud, perhaps, or some of the other practices. Sara, you've been working with cloud for a good few years now. Maybe you could talk a little bit about what actually the content covers inside green software, because it's quite a wide term and there are ideas like web for sustainability. Maybe you could just, maybe give a high level on some of the areas you focus on, then we can talk a little bit about how you folks work together remotely.

Sara Bergman: Yeah. Sure, sure. Yeah. So definitely if you run on cloud, we have a lot of content for you, but even for people who have their own data centers or on prem and you're allowed to call like a few servers in a closet, a data center, as far as we're concerned, there are definitely material for those people as well. We all come from a more backend point of view, but there is.

Web and front end content as well. Some of the points are on a much higher level than this applies only to you if you deploy to this specific type of machine. We tried to go broader, so it's really for anyone who writes or works in the software industry. Not just for, for the engineer, but for someone who is in a role that's not so directly linked to, to code.

We focus a lot on ops as well. So if you are on that side, there is a lot of good content for people like that. And that was really important to us. We come from slightly different backgrounds and we don't want to exclude anyone. That was one of the things we talked about really early on in the process.

You should pick this book up and feel like there is something for me here. And that's been really fun, I think, to sort of go broader than maybe you do in your regular, like, nine to five job.

Chris Adams: Okay. All right. Thank you for that, Sara. So, I now. If I may, I'm just going to touch on a couple of things that you spoke about there. You spoke about the idea that you're writing a book for a number of different audiences, so it's not just for people who are like optimizing code. It's not just about green code, there's maybe aspects of ops as well and things like that. And, given that you're not all in the same space, and given that O'Reilly has quite a, like, established kind of setup for writing books, maybe you could talk a little bit about how you wrote this. Like, was it like, you're using a bunch of GitHub together? Was it Google Docs? Or is there some kind of magic platform behind the O'Reilly firewall that you might be using to keep track of this stuff?

Because I've never, I've never understood, and we're a bunch of engineers here, or at least coders here, because I'm curious, I'm interested in it, and like, it seems a bit weird to maybe use like email versions of Word around to write a book about code, for example.

Anne Currie: Yeah, it was all Google Docs. Well, well, one of the interesting things is O'Reilly have a lot of different options for how you're going to do it. So they have a kind of, there are kind of GitHub integrations where you can do all through GitHub. But we decided to do it all on Google Docs because we were so remote. And in fact, we've only met all the three of us in person once before. So it was all, all had to be remote. So we would use Google Docs. And I think it went pretty smoothly. That was quite easy. I don't know what you two thought.

Sara Bergman: Yeah, I agree. And especially because all of us have been traveling at some point, or some of us traveled quite a lot, I've had, been very pregnant and then had a baby. And there are circumstances where you don't want to boot up a laptop and like sign into a Git repo. Like that's second nature to some, but sometimes it's very convenient to be able to do it from your phone at 3am, if that's when you have the time.

Chris Adams: All right. Okay then. So that gives us a bit of idea of like how the sausage was made, so to speak. All right, then. Anne, I'll put this to you first and then see who else wants to bite. Last time we were speaking, Anne, you were, we were talking about, I think, lasers from space as a way to power data centers, about whether that's a good idea or a bad idea. And I learned a bunch of things about there. So are there any particular things you learned about green software along the way when you're writing this book? Because you can't, you know, you can't be the expert automatically then, and it's a very, very fast moving field. So were there any things that kind of leap out at you or that you were surprised by as you were writing this, that you had to kind of change your opinion on or rethink perhaps?

Anne Currie: Well, probably my favorite chapter to write, which was, was the one that I had to do the most work on, the most research on, we only decided to add later on in the course of the book, which was the networking chapter. Because I was quite interested in what, how does networking fit into green software? You know, is it, where is all the electricity being burned?

You know, and I've, the, the most interesting thing I think in that was I interviewed a whole load of people and I did a lot of reading of papers and things like that to find out what was going on. And the interesting thing for me, I think was the last mile, comparing how good or bad, or green or ungreen all the ways means of connectivity are. So, finding out that that fiber is by far the best by a long chalk actually. So fiber is fantastic. So the backbone is pretty good. The Metro, which is what takes your data across cities, that's pretty much all fiber these days, it's pretty good. And The Last Mile, if it's fiber, it's pretty good to your house. That's the best, followed by Copper. Copper is actually pretty okay, if you're still, I'm still on Copper here, and don't, you know, don't feel too badly about Copper. But once you move into wire free stuff, then Wi-Fi isn't too bad, but 3G is absolutely terrible. 2G is terrible. 3G is better than 2G was.

4G is much better than 3G was. And 5G is much better than 4G. But one of the things that interests me about it is that the mobile generations are a classic. Something we talk about a lot in the book is Jevons paradox, which is the idea that as things get more efficient, you use more of them or to a certain extent, you don't use more of them because the, well,

they, you,

Chris Adams: The savings can be eroded by the increased amount of use in absolute terms. That's what you're getting at, right?

Anne Currie: Indeed. Yes. Yeah. And 5G, I would say is the classic example of that, which is it's much better than 4G, but you know, people didn't invent it for that reason. They developed the efficient 5G because they wanted us to be doing a lot more stuff on mobile devices, and we will use it. The whole point is for us to do more stuff on mobile devices, more mobile gaming. And so, yeah, 5G is vastly better than 4G, but it will drive massive increase in the use of mobile. And mobile is very energy intensive compared to a landline or Wi-Fi. It's different use case.

Chris Adams: So there's one thing I want to just check on before I kind of hand over to one of the... Can I say one of the Saras, or one of the... Sarahs, Saras, all right, yeah. I'll hand over to you if there's anything that you found surprising. So when you're saying better, you're talking about in terms of energy use per gigabyte transferred here. That's your use, like it's efficiency that you're referring to there, right?

Anne Currie: Bits per watt or watts per bit is the industry term for it. And it's something they really do focus on a lot. And the interesting thing is, yeah, I mean, obviously networking is there because everybody wants to do more networking. There's, there's so much untapped demand, which is a classic Jevons paradox issue, really.

Chris Adams: Well hopefully the fact that we've got a bunch of dark, do we still have significant amounts of dark fiber from previous dot com booms and busts where we've got all this stuff available for us to use? Surely that's one thing that we have been able to pick up on, right? Right,

Anne Currie: Well, yeah, there certainly would have been, but to be honest, the past is of no use to us really. It's all the stuff that's been, we're just still laying cable at an unbelievable rate to handle people's, the desire for streamed content.

Chris Adams: So there's an aspect of embodied carbon that we need to be thinking about as well then. So that's one of the things we've learned. Alright, I'll open the floor to, yeah, Sarah Hsu, in London. Are there any particular things that surprised you when you were writing this or caused you to kind of change your mind or think, "Oh. This is probably something we should be spending a bit more time thinking about," for example.

Sarah Hsu: Yeah, so we have a chapter called Core Benefits. And our initial idea was to basically examine green software from different perspectives in software engineering. We have like security, we have reliability, resilience we've performance, data, all that. But while we were writing that chapter, we realized, "hey, actually, this is all very important, but one thing that we're seeing is, because of what's going on around the world, there's so many other things going on. People really are not taking green software as seriously as maybe say 2021." So we actually wrote a chapter on like how we can help our readers to convince others that green software is not as difficult as it's thought out to be, and we came up with like a three bullet points on like how you can convince others. But most importantly, that we want to say in this chapter is that there are so many other knock on benefits from doing other best practices in software engineering already. Green software is not its own ivory tower, like people shouldn't be worried like, "Hey, like everyone is already so overstretched," you don't need to worry about. "Oh, now I need to have a whole new team just to work on sustainability." People who are already very skilled in security, very skilled in reliability, they all can help our software to be more sustainable. I think that's a really strong information we really want to send out there, especially to like the grassroots people and not fail them to be really helpless in the current climate.

So yeah, that's probably quite interesting realization we had later into the book.

Chris Adams: Okay, that's quite an empowering message, and that sounds like, there's a really lovely quote from Adrian Cockcroft, who's leading the Realtime Carbon Project, so he says like "carbon is another metric," and yes, there are other, obviously more metrics that you need to track, but it's not like people in software engineering have never had to measure something before.

So that's one of the things, like, obviously there's things about transparency that can help, but yeah, to an extent, there are lots of transferable skills that can be applied into this domain as well. That seems to be what I'm getting from you on that one.

Anne Currie: Yeah, I mean, I really liked your chapter that you wrote on that. You put something in, Sarah, about "it's not a new work stream." It's, you know, a lot of the best practices in sustainability is often also the best practice in other areas. So it's just one more good reason to adopt best practice in security, in resilience in monitoring, you know, it's not something you're going "oh my goodness, mate, we have a whole extra thing to do." It's just another reason to do all the things that you should be doing anyway.

Chris Adams: Okay, that's a useful framing actually. Sara, I, are there any things that you've learned, that you found, that you want to share, or that you would think would be salient for the conversation here?

Sara Bergman: Yeah, sure. I think, and that maybe goes back to the process is we interviewed people, a lot of people for this book and got their perspective. We also have amazing tech reviewers, and all those people have been very instrumental to the book as well. And, and that's something where I learned a lot from to reach out to people and how willing people are to help and contribute and share their expertise inside the community.

So the community is great. So that's not maybe a hard skill or a hard thing, but that's something that I really enjoyed about the process. And that's something I learned a lot from specifically around the hardware chapter, because I don't work with hardware every day. So. That was a lot of learnings for me.

And actually, I listened to previous episodes of this podcast, for example, the one about the Junkyard Data Center, which is very interesting to include as well as a way to combat e-waste and how big of a problem that is. And then you also get to sort of step outside your bubble and like, okay, but e-waste is surely something that concerns software people, but also other people.

And then you get to realize, okay, but there is lots of legislation around the world working to combat this. And like things are already moving in other sectors as well, which is really fun to see. And then you get additional knock on effects because industries and countries come together to work towards the same goals.

Chris Adams: Okay, all right. Thank you for that, Sara. So, we've, you've written about this, and the thing, you know, when you write a blog post, and as soon as you've done that blog post, you realize, "oh, there's a things, a few things I wish I'd couldn't have put onto it." I imagine it might be somewhat similar when you're getting a book out the door, because once it's out, it's not like you can push updates to a book the same way you can push updates, you know, hotfixes and things. And now the Building Green Software, the book has gone to print, are there any things that you wish you had a bit more time to kind of write about, or that now that you've seen it, go, you know, "oh yeah, we totally missed this," or "here's a thing we should, if I could do that again, I'd spend a bit more time writing about." Sarah, if I can, if I start with you, then I'll kind of go around and then we can round this up perhaps.

Sarah Hsu: I definitely had that same reaction, like two weeks before we were about to submit everything... And this is not a software product. We don't have a process on how to mitigate an incident. What if we wrote something wrong? How do we go about it? I remember just like panicking as well. But it is very interesting, like looking at publishing book.

And like managing a production system. But anyway, I think the things that really, I wish I probably spent a little bit more time on is like in the observability space, like, especially how not just in green software, but like the traditional monitoring metrics and logging really isn't doing what it's supposed to do in the modern software system management, right?

As our system gets more intricate, there's a lot more microservices. Sometimes like if you want to go from one hop, one request all the way for the request to come back to give you a response that can go through a hundred different systems, how do you know exactly where things have gone wrong? And I think I really wish I've spent a lot more time thinking how carbon and how environmental aspects will marry into this space and how do we actually make sure that we are not reinventing the wheel and once whatever metrics, whatever data we are ready, we can slot right into what the world of DevOps and what of SRE is already very good at doing, which is production system management, especially on how to make our system observable, if that makes sense.

Chris Adams: Yes, it does. So what I think I'm taking away from you is that we've got things like spans, and we've got ways to kind of understand the impact of a particular API request or something like that. But when you've got a distributed system, we don't have something like the equivalent to that right now for

carbon, for example. You can't track distributed carbon across all your systems to kind of sum it up.

Sarah Hsu: To find out exactly which system is the trouble one, like, I have a really good analogy. I was writing this the other day, like, you can think of, like, three different elementary data as like a murder mystery, right? Metrics will tell you when someone has died, right? And logging probably will tell you how someone has died, and traces will tell you where this person has died.

So we need to figure out how all these three things will match with, like, the environmental aspect. And hopefully we can have, like, traces that will tell you that's exactly where the bottleneck is happening within our system that, okay, system C.5 is the one that's emitting the most carbon.

Chris Adams: Okay, we're going down a rabbit hole here, but like, we already have all tools like Grafana and stuff like that that give that. Surely we can work out these numbers. This is like a problem that can be solved by,

Sarah Hsu: Yeah. Yeah. But we just need to make sure we are going in the right step. I mean, and not come up with things that ourselves, if that makes sense.

Chris Adams: yeah, that does, okay, all right,

Anne Currie: It is,

well, I'm going to say it is a bit, we'll just spend it, we'll blow all our time talking about this. We need another, we need another podcast to talk about the similarities and differences between carbon metrics and other metrics, because there is a key difference, which we can't talk about now because we're going to run out of time, but we should, we should have that in another podcast. The key difference between carbon metrics and other metrics.

Chris Adams: Okay, all right, that's definitely a thing to dive into, because it's easy to get focused on carbon, and it turns out, yes, we're in a climate emergency, but there's other things we need to be thinking about as well. All right, okay, so we spoke a little bit about what's on your lap next, I suppose, and Sara, you mentioned that there's a baby on your lap, or that's one of the things that you have and you're coming back to. Are there any, like, what happens next now that this is out the door? Because, as I understand it, the book, Building Green Software, is available digitally for people, so you could get it through Safari. But I understand there was quite an interesting approach taken about making this available to a wider set of people.

Does anyone want to take that at all, to talk about the kind of like open aspect of this book? Because this was something which felt quite exciting when I, when I first heard about it.

Anne Currie: Yeah, I'm happy to talk about that. So at the moment, the book is, it's available on the O'Reilly Safari site. So anybody who's an O'Reilly subscription, or if you're a fast reader, you can, you can trial an O'Reilly subscription and read the whole thing. But I wouldn't necessarily recommend that because the book's very intense.

I think if you just sit there and read it end to end, your head will explode. But so it's available on O'Reilly. It's available to buy from all good, well digitally, it's available on the Kindle at the moment, so you can buy it from the Amazon site. And in two weeks, I think it comes out in physical form, so available from all good bookshops. So you can order it anywhere you like and read it in real life that way. Or if you wait a couple of weeks after that, until we actually have time to go around and set something up, it'll also be available under a creative commons license, so under a, it's a, it's a fairly restrictive one.

It's the O'Reilly creative commons license, which is, so you can't take it off and then write your own book, you know, you can't, it's, but you have to attribute this and it's non commercial and it's, but you can read it for free.

Chris Adams: So, source available. You can read it.

Okay, that's, wow, that's really, really exciting then.

Anne Currie: Yeah. That will be available. I'll have it a copy up on my Strategically Green website first, probably.

And then we will, we'll see where else we put it as well. But yes, we insisted on that because it's just so important, you know, it's saying, "we hope this is an important book and it's really important that people read it."

Chris Adams: Great, wow, that's exciting. It's really useful to have something in the, so it's not quite public domain, but it's definitely available in the same way that you can have source available licenses that you can still benefit from and apply there. All right, then. Okay, so we spoke a little bit about some of the things we're covering inside the book, about how you can find this, and when it may be available, and some of the lessons learnt along the way. I've got a question, I have to ask now, like, what comes next, now that we've done that? Anne, I'll put it to you, and then, if anyone wants to take the question after that. So, Anne, you've written a book, or you, plural, have written this book now. Do you just sit in your laurels? Do you just like wait for the royalties to come in or does something else happen after this now?

Anne Currie: If only, if only we could sit in our laurels and wait for the royalties... I don't think writing a book is something you, you ever, unless you're J. K. Rowling, you never really make a, maybe, maybe a film wouldn't be made from this book, but I suspect that's not the case. So, yeah. So the next thing I'm doing is a whole load of training around this, this stuff.

So I'm, I'm doing that through Strategically Green. So started to do some public training and, and one of the things that I'm doing in the public training is, it's focusing on the thing that I would have liked to have put in the book that I didn't. Which is that it's not all bad, you know, there's just enormous benefits.

And one of the reasons why we don't have a whole chapter of like, "Oh my goodness, mate, this is amazing. Why aren't you just stopping and dropping everything and converting your systems to run on renewables? Because it's, you're crazy not to" it's because I don't think a year ago when we started, 18 months ago, when we kicked this whole thing off, that was the case. It wasn't so obvious that renewable power was, was the win that it is becoming. I mean, if you look now somewhere like Spain and Portugal, I'll pick them rather than Scandinavia because Spain and Portugal were doing it the total new tech way. They're doing it with, it's all wind and solar. They're now even in the winter this year, they've had huge, they've had large numbers of weeks when power has been free for the bulk of the business day. So, you know, it's kind of like, this is the dream of humanity that, you know, through, through the ages that we would have effectively free power. And it's looking like that is a possibility, but we will need to change our systems to run on top of it because it's not all the time it's variable. And that's kind of the key message of the book.

But I wish we'd hammered it home a bit more, the wins.

Chris Adams: Right, I'm going to put the question to Sarah Hsu next, but just before Anne, I just want to check on that. When you're doing some research for this, we know that the power, the cost of power changes depending on how available it is. Are there any cloud providers who are making that visible to people yet? Because have you found this? It feels like it's the obvious thing to have to incentivize the use of green power, but it's almost nowhere so far. And maybe you, you might've come across this as well, or maybe the big providers might do something like this because they clearly are making significant decisions about this as well.

Anne Currie: No, I don't, nobody is, but I have spoken to people at most, apart from Google, who speaks to anybody, nobody speaks to anybody at Google, but I've spoken to, spoken to folk who don't necessarily know and have no ability to affect it, but they are aware that eventually we know that the clouds will provide dynamic tariffs. There's no doubt about that because everybody's going to have to provide dynamic tariffs at some point. So, you know, I don't believe that anybody does it at the moment, but I think there's no doubt that it is coming.

Chris Adams: Okay, so not from the big guys, but it's something available coming from somewhere. The one company I've found that is doing this so far, there's a company called TriBuild AI. They're doing this in, I think they're based in Texas, and they're strategically placing things, but the idea is that you can use the power, but not between 6pm and 9pm, when everyone's using power, and that's how they make it like a fifth of the cost of everyone else.

So we are seeing early signs of this, and like, there's examples everywhere outside of technology that we could learn from as well. Okay, all right then, so Anne, that's what you've got on. You've got the maturity matrix to kind of maintain, and the training that you have going on. Sarah Hsu, if I hand over to you. And now that the book's out the door, what do you think happens next? Or what's, what's on your plate after this?

Sarah Hsu: Well, I'm just speaking for myself, but I'm pretty sure all of us are going to be rock stars and go on a tour, go on a speaking tour and book signing tour. So yeah, I think Sara and Anne are doing QCon in April and myself, I'm not speaking this year, but we will be doing like a small book signing there. So it'd be great to see a few folks there.

Myself, we'll be going to Berlin for a conference and then hopefully we'll do some book signing there as well. And one thing I really want to do is I just want to find some observability for who is also really passionate about sustainability and just see what It's just really, really chat it out about like how we can do all this is within observability.

How do we actually make something observable in terms of carbon or like any environmental aspects if that makes sense.

Chris Adams: Yes, that does. If you are not already speaking to some of the CNCF TAG ENV folks, I'm really excited about some of the work they're doing, and they've recently presented some stuff at KubeCon specifically about the observability thing that it's definitely worth chatting about, actually. Okay, and so,

Sarah Hsu: Yeah, definitely.

Chris Adams: And Sara, I'll hand over to you now. What, now that this is out the door, what's on the plate for you, now that you're coming back into the world of, like, software engineering, as opposed to looking after the little one?

Sara Bergman: Yeah, out of the diaper. Well, I guess there will be still be diapers for plenty of time, but yeah, no, going back to work and then being a rockstar going on a speaker tour, like Sara said, so QCon London, which will be very fun. This is the second time only all of us see each other like in person. So very much looking forward to that.

Speaking at NDC Oslo later this year. Also, the Green Software Foundation has a meetup in Oslo, which is very fun. And next, next one of those, I'll be speaking as well. So that would be really fun to like see folk on my own home turf. And yeah, after that, who knows, you know, this was so much fun. Do we do a version two?

I, you know, we'll see. Right?

Sarah Hsu: I think we should, I forgot to mention, we've actually have the book translated to both complex and simplified Chinese. So that's something very, very exciting. I was telling one of my best friend, my best friend was like, "Oh, now your parents don't have an excuse to not read your book," because now, my poor mother will now have to read this book.

Chris Adams: Wow, I did not, that's, that's, well, it's more than a billion people. There's a, there's a readership to, to reach, and people who speak Chinese as a first language probably need to be thinking about the environmental impact of software as well, given it's the second largest economy in the world now, basically, so we've mentioned there's a growing body of resources out there for people to be looking at this. So we mentioned things like, say, some of the free training that's available from the Linux Foundation that people can pick up. And there's a book which we've spoke about, which is Building Green Software by O'Reilly that is going to be available from a number of different places. So Anne, if people do want to, obviously they know about the book, but beyond the book, where should people go if they're interested in what you were talking about or some of the things you're working on?

Anne Currie: Well, at the moment, LinkedIn is, I find, is the new place where all the green chat is happening. But yeah, yeah, LinkedIn. It was lovely whilst Twitter still existed as a thing, and I'm still there,

Chris Adams: drove it into a mountain, yeah.

Anne Currie: Indeed. It was, yeah, I'm still there. I still find useful things there, but it's not as active as it once was.

So LinkedIn is now my social media hangout.

Chris Adams: Okay, alright, and then Sarah Hsu, how about you?

Sarah Hsu: Similarly, I was never really on Twitter and I'm really glad I have persuaded Anne to move over from Twitter to LinkedIn. I also just can't I do want to quickly mention, we do have an email address that you can reach out to all of us. It's [email protected]. So if you have any feedback, any suggestions, or just want to say... please, please, yeah, just reach out to one of us.

Or if you want more direct response from one of us, reach out to us on LinkedIn.

Chris Adams: So if you're a retired engineer and you're going to tell people "you should be solving this all with nuclear," obviously, that's the way to, that's the place to send emails to. Okay, alright, please don't do that, actually.

And, Sara, for you, where should people look if they want to keep up with some of the things you're working on?

You mentioned QCon and a few things like that, but if there's a particular URL or website people should look at, then, now is your chance.

Sara Bergman: Yeah, I'm also, also on LinkedIn, missing Twitter. I still hang around, or X, I guess it's called now, and feel sad about what it once was. But no, LinkedIn is the best place.

Chris Adams: All right, then. Well, folks, thank you very much for sharing your time and sharing some of the insights you've learned along the way of building basically the first book about sustainability and stuff which wasn't written by men, because that's actually like about freaking time. That's very, very good. And yeah, lovely seeing you all again. Hopefully, maybe I'll see some of you in London if I get over there or possibly some of you see some of you in Berlin if you're coming over and maybe wonder if you might sign my copy if I can bring it. All right. Thanks, folks. This was fun. Really, really nice seeing you all again.

Sara Bergman: Thank you, Chris.

Anne Currie: Thank you.

Bye bye.

Sarah Hsu: Bye guys.

Sara Bergman: Bye.

Chris Adams: Okay, take care of yourselves. Hey, everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show. And of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundationon. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

Greening Low Code

4 april 2024 | 36 min

In this episode of Environment Variables, host Chris Adams engages in a fascinating discussion with Marjolein Pordon, a quality consultant at Praegus, about the exciting synergy between sustainability and low code platforms. They explore how low code tools not only streamline software development but also play a crucial role in reducing carbon emissions. Marjolein shares insights into the early integration of sustainability considerations in the development cycle, and together, they demystify the misconception that companies' unique needs necessitate custom software solutions, highlighting the efficiency and environmental benefits of leveraging shared components in low code environments.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Marjolein Pordon: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

Introduction to ESG [11:51]
ESG and Climate Change | Coursera
Mendix [14:32]
WordPress [14:39]
Zapier [14:49]
Creatio [14:50]
N8N [15:39]
Windmill.dev [22:42]
OutSystems [34:15]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Marjolein Pordon: With all the emails we send, inclusive, reply all, the, "yes, I'll see you in a minute." All those kinds of emails, there are 12 000 times from earth to the moon with a car on carbon emission. That's huge.

I'm your host, Chris Adams.

Hello, and welcome to another episode of Environment Variables, the podcast where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. When we talk about green software, we often talk about optimizing code we already have, or finding out ways to make the energy we use less carbon intensive.

So for the computation we do end up using, we end up with less pollution in the form of greenhouse gases and so on being emitted as a byproduct of our work. However, there's another way to look at this. If we accept that the most efficient database query possible is one that you don't have to make because you've designed a system not to need them,

then you can argue that the most efficient system can be one that you haven't had to spend loads of time, energy and money building, building an entirely custom version of, because you found an existing set of components that work well together. This is essentially the argument made when people make the case that a technology stack containing open source software can be more sustainable than a closed source one.

In that the cost gone into building the various components is shared across all the millions of people who would otherwise be duplicating all this effort, building their own versions of these open source, the components. But does this apply elsewhere as well, though? Low code and no code environments have grown in popularity over the last few years, and one argument in favor of using them is that by building the system, or building any system from a set of existing components offered in a kind of visual, low coding environment, you avoid the need to spend so much time, money, and yes, energy, building your own custom software in the first place.

This brings up all kinds of interesting questions. Does this just mean that we end up with more software in total, because more people are building their own software, rather than a relatively small number of professional developers? And are the needs of organizations so sufficiently generic that you can use low code environments for this kind of stuff, to avoid needing all that custom code?

And if you're stuck using providers who are moving slowly to transition from fossil fuel in their infrastructure, what are your options? With me to explore these ideas today is Marjolein Pordon, one of the few people I've seen talking about both sustainability and low code at the same time. Thanks for joining me today, Marjolein.

Marjolein Pordon: Thanks for having me, Chris.

Chris Adams: Okay, you're welcome, Marjolein. Can I give you a few minutes to introduce yourself and what you do? And because you're the first Marjolein I've ever met, if I'm mispronouncing your name, could you please just share the pronunciation so I can use it properly when I'm later on the call? Because I'm trying my best, but you're the first Marjolein I've ever met.

Marjolein Pordon: Yeah, Marjolein is a Dutch name and you're pronouncing it correct. It's like Mar and then Yo from the rapper, Yo Yo, and then a line, like a thin line. So, Marjolein, that's, that's completely correct. Well, I'm Marjolein Pordon. I'm, 38 years old. I live in the Netherlands and I'm a quality consultant for the company Praegus. Praegus is a test consultancy company, and we also have a branch that looks into low code. And one of our CEOs is very much into sustainability. So I'm really in my place there. She also thinks that we should move to an office with sun, sunlight energy, and wind energy. So it's not just me, but also in the company I work for that we have these standards and I'm now, I think eight years in the business and I really love working in IT, testing and things like that.

So yeah, I'm really happy to be here. Oh, and within low code, I've had a few assignments in low code and it got me the name, nickname Lady Low Code, because I'm so passionate about low code and talking on conferences about it.

Chris Adams: I'm glad you mentioned that actually. First, for the breakdown of Mar-yo-line, that's very easy to understand, but also having an alter ego like Batman is also very interesting and useful to have. And that makes it really easy to find as well for the website. That was one way that we came across you actually.

So, and you mentioned a couple of things that were quite interesting there. So, oh, let me just quickly introduce myself for people who've never listened to this before. So my name is Chris Adams. I am the executive director of the Green Web Foundation. It's a Dutch-based non profit focusing on a fossil-free internet by 2030.

I also work as one of the policy chairs for the Green Software Foundation's policy working group where we basically do work at the intersection of policy and green software. If you are new to this podcast, folks, we will share a transcript of everything and we also, we cover, we mentioned quite a few projects and we'll be listing those in show notes.

So we'll share links to that at the end of that. So if you're listening and there's something caught your eye, please do take a look out for that, because that's one thing that we have available. Finally, okay, back to the podcast. Marjolein, I think you're sitting comfortably?

Marjolein Pordon: Yeah, I am.

Chris Adams: Excellent. Okay, so you mentioned that you didn't always work at this intersection and you worked in a couple of fields where you might specialize, before specialize in this niche, and you said that they've kind of influenced your thinking today, like, could you maybe elaborate on some of that?

Because you spoke about, say, testing and stuff like that, and we know that some of the tools people use for testing end up being the same tools you might use for driving certain activities, driving a browser and stuff like that. So yeah, maybe if I just give you a bit of space to expand on that, then that might help listeners understand some of the connections here or how you got here.

Marjolein Pordon: Yeah, I started as a software tester in 2017. My boyfriend is also a software tester and he said, "yo, you, you would be perfect for it." I started on high code, traditional projects, worked there for, I think, a year or two. And then in 2019, I was available for new assignments and I get really obnoxious then because I don't like doing nothing.

So then a coworker said, "well, you can fill in for a week and a half, it's low code, it's easy. Just do your thing." And then it seemed to be not so easy because low code platforms have standards. You need to use them as required. And people didn't do that. So instead of a week and a half, I was there for a year and a half.

Yeah. So, and then I noticed that if you use low code well, the reusable components make the time to market quicker, but also your energy use lower. And the platform that I was working on then, also now doesn't refer to itself as a local platform, but a sustainability platform, because they say even more important than a quick time to market is that we even can go to market.

Because in the Netherlands, we see that the energy use is that high that companies are not allowed to be connected to the energy network. So that local platform said, "we need to change our way of working, because if we do not do that, we cannot make new apps. Because they're simply not the energy to do that, we'll be cut down.

So we need to do this something. And if other companies don't, then that's up to them. But we make the change because we see that it's necessary for the environment." And I really like that. So I got passionate about low code on the one side and sustainability became a thing. And then in 2022, I was asked for QA&TEST to speak there on green IT. And then I got in contact with professors in Spain and in Portugal, not only on sustainable green IT, but also on the connection with low code. And, well, it got me convinced that we really need to do something because it's, it's scary how much energy resources we use, but also hardware resources we use, because if we make quicker apps, heavier apps, we need new hardware, but we are losing also those resources.

So we need to change the way of thinking and low code, I think, is a good way to help us change the way we feel and think about software development.

Chris Adams: Okay, that's actually... so you said a couple of quite interesting things that I haven't heard people mention before. One of the first things you spoke about was this idea that, okay, given the fact that there's, there are limits that we need to stay inside, the thing we need to do is actually, if we're going to increase the kind of utility of the services we need to do, then we need to make them more efficient.

So we did, a couple episodes ago, we did one all about multi-tenancy, about the idea of like multiple people sharing the same platform and using it in a more kind of efficient way, rather than having the equivalent number of like multiple separate platforms which aren't used very much. And it seems like you're touching on some of those ideas there, because this is actually the first time I've heard groups actually take the kind of limit thing more quite seriously, because you often see folks talking about, "okay, well, we're going to make things more efficient."

But if you look at the large providers, say, Amazon, Google, and Microsoft, all of these providers seem to have plans to massively increase the energy use they're using. So this is actually quite a different approach that you've been laying out here. It's like, "well, actually, we're going to use efficiency to kind of book those savings rather than just use it to sell more stuff" basically. Okay. And that was actually quite cool. And the thing that you also mentioned as well was this idea about the testing aspects. Now, I think that there's a term that we spoke about before in the kind of preparation session for this. You mentioned this idea of like shifting left. So, there are things that you would do when you're a tester to make sure you can meet a set of standards.

And it seems like, there's maybe some ideas about applying those things as part of a kind of process, so that you kind of maintain a certain level of efficiency or a certain level of accessibility or sustainability as well. Maybe we could just expand on a little bit of that before we move to the next point.

Marjolein Pordon: Yeah, because if we look to, to testing, usually we come at the end of the development cycle. The things are made and then we test it and then we say, "well, we trust this to work or we not, and" then it needs some rebook. But what we believe as test consultants is the sooner we, the earlier we are connected, then we can think ahead.

We can look to the requirements, but, and not only about the product, but also about the sustainability. Because if we create awareness with the developer. How to make his application more energy efficient, it's better to be done before he starts building. Same with the product owners and the designers and the architects.

They are the ones that should design the system to be energy efficient. And if we all work together and if I as a tester know what they have done, then I can also check that. And it's going to be important because at least in Europe, we have the ESG legislation. And what I hear a lot is that small companies say, "Oh yes, but it's only for the big companies."

Is it? Because if you supply to a big company, you're part of the chain.

Chris Adams: Yeah.

Marjolein Pordon: and you need to also document and let know that you are efficient. And it could be that you lose some big clients because you do not have the things that they need for the legislation.

Chris Adams: Yeah, I see.

Marjolein Pordon: So, in the whole chain, we need to shift left.

We need to use sustainability as a nonfunctional. And it works broader than just environmental sustainability. Because on the marketing part, we all know that, especially the youth want a better world. If you want those as your clients, and they're gonna grow up, they're gonna work in companies, you need to be green, then you need to show that you care for the world. So, it's not just that you save the planet, but you also save your brand, you save the way people look at you, and you make sure that you are there in a few years.

Chris Adams: I see, okay, so you're referring to like, essentially retaining talent and actually attracting people who are basically looking around and thinking, "hey, we should probably be doing something on this climate thing." And if a company is not kind of living those values, that's going to make them think twice about even, "should I join this organization?

Should I choose to work in this field?" And so on. Okay, thanks, Marjolein. Okay, so we've spoken about low code and no code, and you mentioned that there's like one, you're working for one provider. Maybe it might help to just get a bit concrete, because it's quite a wide ranging term. So when we talk about things like, say, low code, maybe we could talk about some of the kind of better known services, so people get an idea of what they might look for.

So what is, like, a low code system in this case, or who are the companies that do this kind of stuff, perhaps? If you could talk about that, and then maybe we might see if there's any open source equivalents that might be useful for developers who are listening, or people who might want to take some other steps into this field.

Marjolein Pordon: I'm not connected to a particular platform. I have worked, I think, with five platforms, but as an independent tester, what we see is you have different kinds of low code. So some are in the workflow management, you have CRM systems. Webpages, ERP systems, all based on reusable codes and those can be modular.

So like Mendix likes to call drag and drop.

I have a component, I drag it and put it in. WordPress works the same way, but then you build a website. Those like Workflow and Aden, I think is a one, and Zapier,

Chris Adams: Mm.

Marjolein Pordon: and for CRM you have Creatio, and all that I mentioned are open source tooling. And I think that's important because like you said, if you share that knowledge and share those, that code, people don't have to make it their own and they don't have to invent the wheel again.

That's I think important because you don't have to know everything. You need to know where to get it. But also, the energy efficiency and the environmental impact is way lower if we reuse code.

Chris Adams: Okay, thank you for that. So to reiterate, so when we're talking about low code and no code, we might be talking about, so you mentioned a few platforms that people may have heard of. So one is Zapier, which is very, very common that will basically plug into an existing software as a service and then let you make a transform.

And then, and there was another one you mentioned, which was N8N, which is a little bit like a kind of open source, well, open source like equivalent. I'm not sure if they're technically using an open source, it's more like source available, but the general idea is you can run it on your own servers, so you can have a similar idea of pulling information from one place and having little bits of javascript or something like that you run there, with the idea being that you have multi-tenancy, lots of people using the same things, because you'll need a little bit of code rather than having a whole running one.

And you also mentioned that Mendix, which is a more visual option, and, okay, and some of them which are more specifically tied to an existing, say, enterprise system or something like that. Okay, thanks for that. Okay, so we spoke a little bit about, in the intro, some of the arguments people use about where low code might fit into this kind of, into this world of green software and reducing the total resource requirements of, of anything like this, and we've spoken about the idea of lots and lots of people using, say, shared low code.

Shared components in a centralized service. Now, this implies that lots of companies have problems which are similar enough to each other that you can actually solve something like this. And when I've spoken to people in various organizations, they often, one, a common kind of theme is, "oh, our stuff is so unique that the only way we could possibly do this is by hiring a really, really expensive specialized developer because only they, only they could possibly capture the uniqueness of our organization." And this seems to suggest that's not the case. Maybe we could talk a little bit about that, like, is it that companies are over indexing on how unique they are, or where does this fit in, or are there some parts that you can do, but there's still places where you might need a specialist, for example? Maybe you could kind of share a little bit about that, actually.

Marjolein Pordon: Yeah, well, it might not be a popular opinion, but I feel that companies need to convince themselves they're unique because how else are they going to stand out? And what I see with companies that embrace low code is that they say, "well, we choose to look for the components we have in common with other companies. And if we build those components, those modules, those parts that we need in low code, we have time left, we have money left, we have expertise left to build those few things that are customized." So we all know software projects takes ages, deadlines are never met, but what if we use those components that are alike, because every company has parts that align. If I sell something, I need a cash register.

I need a website, all those things align. And if I need a small part that's not aligned that I want different from my competitors, then I can build, let that build by that said expensive developer. But that developer only has to do that little part for maybe a week or two weeks. And the rest is built in low code.

It's energy efficient. I save money. I save time. And I invest for just those parts that I really want to stand out in. And it can also mean that you think that you're just like everything and you build everything in low code and what you save in money, you can add to invest in your company, add new products, do your marketing. So there, the, the low code embracers, I feel say, "yes, we are unique, but we use what we have in common to do in low code and what we do not have in common and stand out, we invest in that."

Chris Adams: Okay, I can follow that argument now, actually. And there's maybe one question I have for you, is basically how these platforms are built, because usually, like, rather than paying to have an entire server or something like that, you might pay for kind of per use, and this is like an argument people make with serverless computing is like if you're paying for the requests or you're just paying for the amount of computation on a kind of per second basis, then you have an incentive to reduce the amount of computation you're using because you have it linked much there, there's a much clearer incentive for you to do this.

Are, is this common? Do you, are there, is it fairly common for, say, low coding tools to bill for, like, provide granular billing in terms of, like, a request made or a kind of run or a workflow? Like, what are the units that people tend to use when they're looking at something like that?

Marjolein Pordon: The bigger platforms I noticed have trouble or are not quite there yet to do that. But there are smaller platforms who say, "well, this is the way we do it." And I think that that's creating of awareness. What does my system do? Because I pay for it. And we all know money is a big, big incentive to say, "Oh, wow.

Why am I doing this?" And when I did my, my talk in 2022, I did some research and what I noticed is what we do in social media every day. And I had a thing that if the miles from earth to the moon with all the emails we send, inclusive, reply all, the "yes, I'll see you in a minute," all those kinds of emails, there are 12 000 times from earth to the moon with a car on carbon emission.

That's huge. Every single day by every mails we send and that are sent. 12 000 times, so 6 000 times to the moon and back again every single day. That's huge. When I knew that I was going through my newsletters, which one am I reading? Actually reading. Not reading? Sorry, I unsubscribe. And that's only maybe a hundred emails in the year, but still it's a hundred less. So that was for me an eye-opener. But if I had a bill every month of a hundred euros because of emails that were polluting and that I wasn't even reading or even opening,

Chris Adams: Then you're going to think about that,

Marjolein Pordon: then the awareness would even bigger. So I think that's it. Yeah, so I think that the platforms that are doing that billing are really good. Because that's the best way for awareness.

Chris Adams: Okay, all right, thank you for that. That makes sense, and there are a few cases that, a few of them, I do believe, expose some of these metrics to you like that. At climateaction.tech, one of the communities I run, we use N8N that provides some of this for some of these things, and I've seen similar things with windmill.dev, which also gives some of these, exposes some of these numbers. So we spoke a little bit about the efficiency part. There's also one thing that came up, is, okay, if you're using one platform, for example, there's going to be there'll be things that you cannot change because they are kind of with the provider's decision.

For example, if you're, like one example I can think of is, there's a service called Webflow. This is a kind of low code tool for designing websites. And they run on AWS, who for the last few years have been less ambitious on climate in terms of greening their own supply than say Microsoft or Google, based on like the things that we see in the public domain, for example, and I know there was one example, there was a project called Flowty by a chap called Fershad Irani.

One thing, one thing he basically did was he looked at where you could do this, and because the thing that gets built is maybe a, say, a static website that's created by Webflow, he was like, well, if you've got this, maybe you can move this to a greener provider, for example, and, or maybe you can apply some kind of digital sustainability techniques to reduce the size of the page and do things like that, right?

Now, this was one example of like a low code service, which was like, if you're going through maybe Webflow, for example, you're tied to kind of infrastructure which where, the people providing that are not as ambitious as some other groups, for example, so you're kind of stuck there. There are cases where there are options.

So you can say, "well, I like what you're using, but can you move to another provider?" Or, "If you don't do this, then I will move to another provider" themselves. Maybe you could talk a little bit about like what your options are if you are using some of these tools. Because one of the things I've heard people say about no code or low coding tools is like, "yeah, they're great, but now I'm stuck inside this specific environment where I have no other way of moving away from that."

Maybe you could touch on what are your options if you do want to ask your provider to improve the sustainability further down in the stack, for example?

Marjolein Pordon: Well, what I noticed is one of the big platforms, Mendix, has made it a big issue. And that's because a lot of their clients pushed them and then the majority counts. Then they need to do it because they don't want big clients to walk away. Again, they are part of the chain of big companies, so the ESG legislation is probably also haunting them, which is good. But yeah, vendor locking is a big issue in low code because for example, I have an application built in Java, C#, whatever. The developer that I have or the service that I use or the company that I use, I don't like, we have issues, we part. And I just hire another Java developer, but if I do not like the platform, same with if I wouldn't like Excel or if I wouldn't like Word, everything that I did in there, I need to migrate. And it's not easy. I mean, I could open things in Google Docs, if I went from Word to Google Docs. But still, I could have issues with the outline, with the layout, things like that. So migrating from one platform to another is not easy. And the integration with platforms, like if I had used one platform and then I think, no, I don't want it, but the application can stay there, but I built something new in another platform. The connection is, is really hard because I always say, Lego wants you to use Lego and not K'nex.

So, and maybe you can connect K'nex to Lego, but it won't be easy. It won't be good on usability. So yeah, the vendor locking is definitely a thing. So before you start with a local platform, wherever you need it for, do your research search on, "is this the platform I want now and in the future?"

And not only on, "does it work for my company," but also "what legislations are coming, are they sustainable, what is my client group wanting?" Things like that.

Chris Adams: Okay, so there's not just, so there's, you're talking, there's almost like some, a case of about alignment you need to be checking for, but also kind of influence you might have as well as actually just saying, is this really convenient for me right now, for example. Okay, all right. That's actually quite helpful.

The one thing I can share with you when I was looking through this that other listeners might find interesting, especially if they are developers, there's, we've come across some tools which are open source. They provide kind of low code like environments, but they also allow you to kind of drop down into your preferred programming language.

One thing I've really been quite impressed with is one organization, one project called windmill.dev. So it's an open source, kind of low code platform, but it's more like a... it's a platform where you have visual ways of working, but there's also a way that you can drop into pretty much your own language, or use any kind of docker container.

And that, because the actual platform itself is open, that's one thing, but also the fact that you can drop down into languages that maybe developers might be familiar with, means that you, that reduces some of the lock in to an extent. And I think that's actually quite a promising path to go down, because yeah, you, have that separation that we know in other, in other kind of sectors.

Having separation of different layers does allow for you to have more options on the table. Okay, so can I, if I can, can I just come back to one thing that you spoke about before? Because I was really surprised, and actually quite impressed by the whole thing about saying, "well, okay, we're going to use this as a, as a absolute limit in energy we can use.

So the only way we can do this is by increasing, if we want to grow, then we need to make more efficient use of this rather than just say, we're just going to keep growing as fast as we can, and we're not really going to think about the resource requirements." So we spoke about this idea of, and this is one thing, an ongoing discussion in green software, which is about basically growth and the idea of the rebound effect.

Are you familiar? There's a term called the rebound effect, which basically refers to this idea that if you make something more efficient, you can increase the total usage of this just because it's become more accessible to more people. And one of the key things around tools like low code is that yes, you're democratizing access to computation, but it also increases the number of people who might be making systems as well, which can have some similar effects.

So maybe you could talk a little bit about the conversation and how you've seen that evolve in the Netherlands specifically, because that's somewhere where you have seen, like moratorium on new data centers being built, which has meant that you've had to, it's forced some of the conversations that are probably not happening quite so quickly in other parts of the world.

And this was, yeah, you brought this up at the beginning of the podcast, so it feels like it might be worth just spending a bit more time looking into that.

Marjolein Pordon: Yeah. Because if you look at data centers, we have, I think, three big ones in the Netherlands. And the one I think in Lelystad uses as much energy as the whole of Amsterdam. That's huge. One data center takes up the same amount of energy as the main city of our country. And that's huge. That's crazy if you think about it, because we cannot build at the houses we need at the moment because of the nitro legislation, because there's not energy, there's not enough water, and those data centers use water and energy of a complete city. So if we do not have that data center in the Netherlands, then we could build a complete city of Amsterdam. Think of all the houses we could build. So this is quite a discussion in the Netherlands. Arjen Lubach is a guy who makes, is a television presenter, and he makes those kind of reports on this.

So also to make us aware, because they are built for a part with our money from the taxpayers. So we are aware that it gives jobs, a data center. We need it and that it's good for Google and AWS to be here, but they're not energy efficient. And well, one of the politicians said, we need them to give back because they're here in the Netherlands,

they use, okay they pay for the energy and for the water, but they need to give something back also for sustainability. And what we now are looking into is that they heat water to cool down their data centers. But we could use that water to heat cities with city heating. So then we, then they should make sure that there are pipelines to the city next to the data center. And then the warm water from the data center can flow through the houses and heat them. And then when the water is cooled down, it can go back to the data center and then you have the rebound effect and you can reuse the hot water to heat the houses and the cool water can go back to the data centers. And data centers said, "no, no, that costs too much money." And we are in the Netherlands now working on a legislation that will make sure they have to do that. If you want a data center in the Netherlands, then you need to build it, and you need to build the system to heat houses. And I think that's good because then you're cooperating, then you're working together and making sure that we need the data centers, but they can give back and work together so that we can still have a better environment because if we go on like this, they had, I think, requirements for 2030 in the Netherlands. Well, we're not going to meet them. And then they said, "well, no, but we'll make them in 2040." Yeah. And in 2030, you'll say 2050. We need to do something now. And I think that by making this happen, saying to a big company like Google, "fine, you have your data center, but these are the requirements or else." It's just necessary to do this.

Chris Adams: Okay, that's a very different take that we've seen in other parts of the world when discussing this actually. And what we can do is we can share some links to some of those points because I haven't heard those stats and those numbers presented in that way before actually. Marjolein, I just want to check up.

So we've spoke, we've covered quite a lot of ground. We spoke about like low code and things like that. If people do want to, if they found some of this interesting and they are, they're looking to take some of their first steps into low code, into the set field. Could you maybe just like suggest a few places where people should be looking to either get either learn in their own time with some like training or if there are any particular projects that you would want to draw people's attention to for this?

Marjolein Pordon: Well, for the low code part, most applications have their own playground and learning environments, which are quite good. So I would start there on the low code part. There are not many general courses on low code, but I'll check them after we're done with the podcast and then I'll, we'll add them to the show links.

Chris Adams: And if there was maybe one platform people might work with or start with or is there one that you would suggest people take some of their first steps with for example or something like that if they're coming in to begin with they just want to start kicking the tires and trying it out, for example?

Marjolein Pordon: Well, I would start with like a Mendix or an OutSystems or things like that, where you have a lot of information and a community behind so that you get the hang of what is low code, what are the standards that you need to work with. And if you have done like one or two of those, then you see the common grounds and then the other platforms will be a lot easier.

Chris Adams: I see. Okay. All right. Well, Marjolein, thank you very much for that. If we're just coming to the end of our time now. So if people have found this interesting and they want to follow what you're doing, where would you suggest people look? Is there a website or is there maybe a pro, is there like, where, where are you online for people to follow your updates and see what you're, see what you're doing going forward?

Marjolein Pordon: I'll follow LadyLowCode at my tag at Instagram and Twitter is @LadyLowCode. LinkedIn is Marjolein Pordon and my website is www.ladylowcode.com.

Chris Adams: Brilliant. Okay, well, Marjolein, thank you so much for giving the, making the time to chat with us today. As we mentioned before, we'll run through this to make sure we've got show note links for all the things that we discussed here. And yeah, have a lovely week, all right? Take care, Marjolein.

Marjolein Pordon: Thank you, Chris.

To find out more about the Green Software Foundation, please visit greensoftware.foundationon. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

Greening Digital and the Rebound Effect

28 mars 2024 | 44 min

In this episode of Environment Variables, host Chris Adams delves into the fascinating topic of the rebound effect with Vlad Coroamă, founder of the Roegen Center for Sustainability. They discuss how improvements in efficiency can sometimes paradoxically lead to increased consumption, using examples like teleworking and online shopping to illustrate the point. Through their conversation, they explore why this happens and what conditions make it more likely. Their insights shed light on the complexities of balancing technological advancement with environmental sustainability, offering valuable perspectives for anyone interested in building greener digital services.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Vlad Coroamă: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

ICT4S 2024 [41:36]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Vlad Coroama: When there is rebound, but if your digital service makes the activity sort of more affordable or simply more desirable, and it will be consumed more, but it will have changed in such a way that the footprint of the new activity, the modified one, is much smaller than the original one. And then although you might have rebound, the overall balance will be net positive.

I'm your host, Chris Adams.

Hello, and welcome to another episode of Environment Variables. Where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. When we talk about green software, the notion of efficiency comes up quite a lot. Take two forms of efficiency explicitly called out by the Green Software Foundation, software efficiency and hardware efficiency.

In the first case, you're talking about how much energy is needed to perform a given amount of computation. And in the second case, you're talking about how much hardware might need to be created by extracting material from the environment, refining it, then turning it into electronics that we run computation on.

In isolation, it's quite hard to argue against efficiency, and we can point to literally years of data showing how increases in efficiency in computing have blunted what might otherwise be eye-watering increases in the amount of energy consumption and other resources we have gone through in absolute terms.

Thing is, efficiency has second order effects too, because making things more efficient can make them more accessible too. Increasing the number of people who can use them. And we can point back to published work in 1865, observing this happen with coal-powered steam engines. So the common term for this is the Rebound Effect. And joining me today to explore what it means for greening digital services is Vlad Coroama of the Roegen Center for Sustainability.

I first came across Vlad's work when at a green cloud procurement workshop in 2019 held by the European Commission in Brussels. And over the years, I've had his writing and presentations about the rebound effect and digital sustainability, some of the most incisive and accessible work on the topic, on the subject.

Before I embarrass him further though, I think it might make sense to give him a bit of space to introduce himself. So Vlad, thank you so much for joining. I've been looking forward to this. Can I give you a few minutes to introduce yourself before we get into the meat of it?

Vlad Coroama: Hi Chris, many thanks for your kind intro, of course, for having me on your program. So I'm Vlad, the founder of the Roegen Center for Sustainability, which is a small company based in Zürich, Switzerland which tries to do a research, actually, more research than consultancy in the field of computing and sustainability, or if you want more from a, more from a deployment perspective, digitalization and sustainability.

Actually before that, this quite recent and before that for my entire life, I've been an academic and I've worked in the fields of computing and sustainability and also the more technical one of smart energy for about two decades now. And with both these hats on, so both with the more sort of hands on engineering system developing hat on, and with the more theoretical hat, what my work, which of course in the beginning, it was not so clear where the path leads to, but it became more and more clear that I want to understand how we can both make computing more sustainable, which is if you want perhaps green IT to use a general term, but also to see, and perhaps in my view, more importantly, to see how we can deploy computing or digitalization again from a deployment perspective, to induce environmental benefits across our societies and economies. So in other sectors, sometimes that it's often called Green by IT and these indirect effects, as you said, it's not only about direct effects, this is very much about indirect effects, are, or can be, so much more powerful than the direct footprint. And unfortunately, it's not only the positive effect. So it's not only how can we do, you know, society and economy more sustainable, by the way, society, societally sustainable as well. We'll talk mainly about the environment today, but much of what we discuss applies to societal implications as well, but to come back, so it's not only positive effects.

We also have, unfortunately, this indirect detrimental effects to sustainability, both societal, I mean, we've seen elections and so on, right? But also environmentally. That is, computing or ICT can induce more energy and material consumption, increased emissions, increased pollution, and so on in other sectors.

And with this, I think we sort of arrive at the topic of our discussion today, which is rebound.

Chris Adams: Indeed. Yes. Thank you very much for that, Vlad. So, just if you're new to this podcast, my name is Chris Adams. I am the executive director of the Green Web Foundation. That's a Dutch-based nonprofit focused on reaching an entirely fossil-free internet by 2030. I'm also the, one of the policy chairs of the Green Software Foundation's policy working group.

And here's a quick reminder, we're going to be covering a few papers and a few projects and links and websites. We have show notes at the end of every single episode, and there'll also be a transcript. So if there's something you miss, we will have that available so you can kind of catch up with this or basically submit pull requests if you see things you need to correct. All right, Vlad, are you good to go? I think, should we start with this? Okay, then. So we spoke about this. The topic of this is Greening the Rebound Effect. And I've touched on what the rebound effect is, but for the uninitiated who want to learn more than what I just said, what is the rebound effect?

And maybe you could tell us a little bit about where it comes from and whether it's a new thing.

Vlad Coroama: Okay. Thanks. So as you said in the introduction, your introduction, the very first time that we know this has been mentioned, it was in 1865 by the British economist, William Stanley Jevons, who wrote a book, The Coal Question, it's called, and it was about what we today call rebound effect, we called it such back then.

And by the way, this is a very cool thing, as a computer scientist, you seldomly get to cite, you know, a paper older than 20, 30 years, so it's really, it's really nice when writing a paper, you know, to cite something from the 19th century. So what Jevons noticed was that the more efficient steam engines and other, you know, coal using machinery was becoming in the 19th century.

At first glance, paradoxically, the overall coal consumption was not decreasing, but increasing and increasing at a very larger rate. So this is not necessarily counterintuitively, but yet it requires an explanation why. And of course the explanation is that the amount of engines was increasing because

the more coal, the more efficient machines were becoming, and by the way, some of these machines were steam engines that were helping in the very coal extraction. So coal extraction itself was becoming more efficient and thus cheaper. So both the running the machines was becoming cheaper and accessing coal was cheaper.

And this means you could deploy the coal for many more machines doing the same stuff that had been done before and for entirely new applications. And this is basically what we now call the direct free buffer, which means a good or a service becomes more efficient because the energy is more efficient or some other material that flows into it.

So it is more efficient to produce that good or service, thus it becomes more affordable. And thus, as we know from neoclassic, since Adam Smith, basically the demand for it tends to increase. So that's the first phenomenon. And this, by the way, lay dormant for over a century and late seventies, early eighties of the 20th century, of course, due to new researchers, Khazzoom and Brookes,

Chris Adams: Ah, yeah.

Vlad Coroama: we rediscovered the phenomenon in wave of the two oil crises, was the context.

And basically they talked about the same, about the direct rebound effect. And then this developed also to what we now would call the indirect rebound effect, which is actually an umbrella term for a variety of mechanisms and phenomena. But all have in common that something becomes more accessible in a way.

Some resource is being saved or is being used more efficiently. And that resource can be energy, but it can be any material. But it can also be immaterial stuff, such as time, and this we call time rebound. So if a technology saves us time, we'll do something with that time that likely will also require energy and produce emission.

Or if it saves us money, as before, we might not spend those money on more of the same good. But, and this is called the income effect, and this is another type of indirect rebound effect, we might use the disposable income to, you know, do something else that in itself might be energy intensive and, you know, responsible for lots of emissions.

Chris Adams: Okay. So, so we've got the Jevons. So there's Jevons paradox where it initially started with, and I believe you spoke about Khazzoom and Brookes. I think it's, is it called the posh term, the Khazzoom–Brookes postulate? Something like that. It's like the kind of term you'd use at a cocktail party to impress people.

And then you've also described, there's a few different flavors of rebound that we might talk about, and they might have different degrees of magnitude. And this is something we're going to, we can talk about a little bit later. So in time, you know, if I'm, if I save a bunch of money by buying a bunch of things at, say, a supermarket, a cheap supermarket, I might end up spending a bunch of money either eating out or buying, getting coffee, you know, posh coffee and stuff like that.

So that's, those are the, some of the rebound effects you're talking about. Okay. All right. Thank you for that potted history there, actually. And I didn't realize that it was so much based around the seventies, because presumably that's like relation to the oil crisis and when people suddenly started caring a lot more about efficiency.

Right?

Vlad Coroama: Yes. And actually they saw that and then also a more fuel efficient cars came around also in the U S and, but still the overall fuel consumption, not maybe short term during the, you know, the crazy months of the oil crisis, but over several years, the oil consumption, the petrol and so on was still increasing.

And that's how, so for a similar sort of trigger as Jevons, a century earlier, they started looking at it, how does this happen? And what they saw is people drive their cars more.

Chris Adams: I see. Okay. Thank you for that. And, so when we're talking about things becoming cheaper and more accessible and more widespread. It's not a huge leap of the imagination to think about things like electronics getting cheaper or more widespread, even cloud computing becoming cheap or more widespread.

So I guess maybe it's worth me kind of moving to this. So if we're gonna talk about things like the rebound effect or things like this, or why efficiency gains matter in the context of building digital services, maybe I could actually ask like, why does this matter when we talk about digital computing?

And why does it matter in a world where we are seeing more laws being passed now and a kind of influx of new kind of legislation or people setting new norms about this?

Vlad Coroama: So it, it matters, as you say, of course, this happens within computing as well, or within ICT, let's say, I think it's Koomey's law that, that says,

Chris Adams: Is the network halves every 18 months or something, isn't it?

Vlad Coroama: Yeah, it's that basically we are, or a consequence of Koomey's law, a consequence, not directly Koomey's law, is that we use all our gains for more computing and not for, you know, less energy intensive computing. And this of course makes sense. But the problem I think with computing or what does it mean, the, the big challenge of computing, but also environmentally, what can become an issue is that it's general purpose technology and that it induces efficiency, not only within computing, but also, and crucially outside it.

So we, when we talk about the rebound effect of, again, computing/ICT/digitalization, choose the term that you prefer... we have to distinguish between the rebound within ICT itself, more computing, perhaps, you know, more cloud, whatever, it's more affordable. But crucially, I think also the rebound outside this in all the other sectors that digitalization makes more efficient.

Chris Adams: I see. So there's, so in the context of digital, one thing you're saying is that, yes, there is, obviously we should be mindful of an efficiency argument, but because if you just only talk about efficiency rather than consumption, you can lose sight of the full picture and. If you take a second to step outside, the efficiency that you might see at a kind of digital level could also have like

absolute increases or decreases accordingly. So you should be, so we need to take, we need to be looking at the two of these basically.

Vlad Coroama: Yes, exactly.

Chris Adams: Okay, cool. Thank you for that. That's actually, okay, that's quite helpful. And I suppose when we talk about efficiency, it's worth looking at some of the numbers, for example.

So we have seen, say computing get quite a bit more efficient, but we've also seen, basically, we've also seen, for example, some of the hyperscalers, we've seen hyperscale companies like say Microsoft, Google and Amazon, it's not like they've stayed the same size and they haven't grown.

We ;have seen them growing, even as things get more efficient. So these are one of the things where we need to be somewhat aware of, yes, the absolute figures in this as well as the efficiency part here. And we've spoken about how, there's, digital can have an impact on the outside world, and you might be talking to things like, so like transportation examples or like ride hailing, things like that is what you're referring?

Maybe you could expand on some of those. Cause I think these are the things that I've seen you talk about quite eloquently in other places, actually.

Vlad Coroama: Yes, because there is the hope, right? And very often we have the claims that ICT or digitalization, let's say now, makes so much of the world more efficient by coordinating it better, by, you know, finding patterns, by we all know how Google did its cloud more efficient and so on. But then there are many other fields outside.

And in all of them, I see a pattern of how in the beginning everyone says, or a lot of the voices say, "Hey, great, you know, now we have your efficiency. Now it will be so much better." It's a sort of a techno-utopianism, if you want. I will give two-three examples. The one I will start with this one, I'm writing now actually about the rebound effects of teleworking.

And I've, so I've been reviewing many studies and it's very funny because the very early studies, teleworking has been around long before sort of the World Wide Web made it into the homes. Since the 70s, they started talking about this. And the first papers have titles such as, you know, 'Traffic Reduction by Telecommuting' and then similar things.

And then through the work of Jack Niles, it was, and especially Patricia Mokhtarian in the nineties already, they started to understand, "oops, wait, wait, wait, there is also lots of rebound effect." And today's papers have titles such as, you know, 'Does home-based telework reduce household total travel?' So lots of questions marked there.

You know, does telecommuting promote sustainable travel and physical activity? Does telecommuting reduce commuting emissions? And so on. These all, and I have many more, but I will not go into them. But so the phenomenon that happens there is that, yes, teleworking in first instance, of course, if you don't travel to work and travel is energy intensive, much more so than, you know, the little bit of energy that we consume now to have a call, it saves energy in the sum.

But then, because you have more time, because you are more flexible, you start, and because before you used to do other things while going to work or coming, you used to have multi purpose trips. So you, I don't know, dropping kids at school, you know, going to the gym, doing grocery, whatever. those other reasons still exist.

So you will still undertake other trips and much more so than you have subtle effects. If you only need to commute, say, twice per week to work, you might be very tempted to move much farther away from work, you know, in a nice countryside where the kids can play, you know, in nature and safely and so on.

And then you only commute twice instead of five. Well, four or five times per week, but for much longer distance and perhaps no, you can no longer do it by public transportation because you're not urban anymore, but you have to do it by car. And that's a classic. And this became more and more clear. So teleworking is not clear actually now whether, you know, the net effect is a positive or negative.

And I will not go into this detail for others, but we have this for e-commerce or slash online shopping as well. Again, lots of enthusiasm environmentally in the beginning, and then you see that many other things happen. You know, you, all of a sudden you order much more. It's so easy to order from the couch at 11 pm, you know, you don't need to go to the store. So all the consumption increases, or now more recently with AI, with autonomous vehicles. And this is perhaps the last example I want to give, and I think you've heard this before because it's a favorite example of mine. Also in the beginning, we had lots of enthusiasm, you know, how the cars will like coordinate with each other.

And then, you know, they, at some point we won't need traffic lights anymore. So then don't need to brake and waste energy and then reaccelerate. But this is all peanuts. What actually will happen is that, you know, they will substitute, autonomous vehicles will substitute a lot of public transportation because it will be so much more convenient to be driven by car and be able to work in the car or, you know, read a book, discuss whatever. So use the time efficiently.

Chris Adams: Okay. So you spoke about these, there's quite a few examples then of the rebound effect resulting in basically first in like direct efficiency leading to increases in usage in other ways, with some kind of actually quite vivid examples there. And that's, that feels like a nice segue to talk about, okay, we have this idea around rebound.

And there's different kinds of rebound that can take place. But as I understand it, there are certain conditions that make rebound more likely or increase the effects of rebound versus, making them somewhat smaller, for example. Maybe we could talk a little bit about that because I remember hearing about the kind of like vivid example of autonomous driving.

Like you mentioned, there was, yes, it increases the, it lowers the threshold of you doing things to the point that, you know, there's there's a famous study about someone sending autonomous cars to just pick up a sandwich they left at home because it was so easy to do now. And that's obviously not going to be a sustainability win.

So maybe you could talk a little bit about when you do see rebound and what conditions make it more likely to happen versus maybe when it's not so likely to happen, perhaps.

Vlad Coroama: Yeah, there are, I do not have general rules to provide an answer. So I cannot tell you this precise, you know, class of applications or yeah, digital services are more likely to rebound than the other not. But I can give you a couple of hints or perhaps examples, and the easiest is to start with that, with the example of the vacuum cleaner that you mentioned earlier, and which is of course outside digitalization, but I think it's very nice to understand the phenomenon. When the bagless cleaners emerged to, Dyson was the first on roads to invent them.

They also became so much more efficient. So they used to consume 1.5 to 2 kilowatt of power, and now they are 4-500 watts. So a factor of it's like 20 to 25%, a factor of four to five, reduced power. And the question is, do we vacuum much more? So it is of course, cheaper to run them. Do we vacuum much more?

Probably we do it a bit because, you know, they are also more convenient. They are cordless very often and so it's easier to grab them. But certainly this rebound is relatively small and not, you know, 400% it doesn't overcompensate with certainty, because, well, you only need so much to vacuum your house and it's probably also not the most people... the favorite activity for most.

So one of this thing is when there is something like, when the demand is satisfied,

Chris Adams: Ah, like an upper limit.

Vlad Coroama: not have rebound, you need to not have the rebound mechanism. And this mechanism being often, not always, again, it can be with time or transaction costs or other things, but often it is monetary.

You save money and then you, you know, consume more of that good. And in this case, if the demand is satisfied, then you don't need more. So, for example, smart heating in a home, to come again now back to computing, if you have smart heating, I mean, we used to have our homes up to the seventies, even at 13, 14 degrees centigrade in winter.

We don't do it anymore. We all have whatever 19, 20, 21, wherever we feel comfortable and we don't need more than that. So, a smart heating system will make our, our heating more efficient, then we'll save 10%. And that can be financially quite interesting. We'll not use those money to hit more because there is no need for it.

I mean, we might perhaps, you know, let a bit more fresh air in and thus waste a bit more energy, but it will certainly not compensate the savings. So that would be one such example where at least a directory bound. It is, is unlikely,

Chris Adams: Ah, I see. I'm really glad you mentioned the vacuum one because I remember watching your talk just after I bought a cordless vacuum cleaner myself, and I remember saying like, "okay, there's an upper limit to how clean my flat can actually be." And like, yeah, it's a lot more fun to use, but yeah, I, it doesn't make me... making me slightly more efficient at vacuum cleaning does, it didn't double how much I enjoy vacuum cleaning, right?

Well, I might enjoy it more, but there's an absolute upper amount of vacuum cleaning hours I'm prepared to invest into my flat, for example. Okay. So that's, so there's this upper limit of satisfaction that if you have something like that, that's maybe one kind of hint that you might be looking for, for example, and we might be able to kind of take some ideas into another domain domain for that.

Vlad Coroama: And if you want, I can give another such hint. So when there is rebound, but if you sort of, if your digital service makes the activity sort of more affordable or simply more desirable and it will be consumed more. But it will have changed in such a way that the footprint of sort of the new activity, the modified one, is much smaller than the original. And then although you might have rebound, the overall balance will be net positive.

And again, a short example outside of digitalization is LED lamps, right? There is certainly, once you have LED lamps, there is certainly a rebound in the, in a sort of light rebound, in the amount of light that you're using. You will, because you know, they take six watts and not 60 anymore. You are not so concerned with like turning it off anymore.

So there is some light rebound, but in terms of energy, the rebound is really small because even if you leave it like twice as much, you will still save 80 percent and not 90 percent, but still the net save will be... and the same in digitalization. We did, for example, a conference for it in 2009 between two continents.

And that conference happened at two sites simultaneously in Nagoya, Japan, and in Davos in Switzerland. Of course, it was a seven hour difference, so there was just a four hour common slot in the Swiss morning and the Japanese afternoon. And why we did this? Because for a conference, the main environmental impact are flights of participants to the conference, and in particular, intercontinental flights.

So the hope was to save intercontinental flights. And according to our survey afterwards, we have indeed succeeded to save some, around 80 intercontinental flights of people who would have flown to the other side of the world. And we induced much more, around 200 intracontinental flights. But you know, a short haul flight has such a smaller footprint than a long haul flight that although we had many more participants

and many more flights. Because those flights were much shorter, the overall impact was still positive. And again, we only talk direct rebound here. I like to stress this point. The system boundaries, as they say in environmental sciences, the system boundaries of indirect rebound are basically the words. So it is very, very tough and we do not have yet the right tools to profoundly assess the, you know, the overall impact of digital technologies, unfortunately. And this is one of the areas I'm most interested in.

Chris Adams: I see. Okay. Can we just dive into that a little bit more before we move into the next topic? So you spoke, we spoke a little bit about say, forms of rebound where there's an upper limit where there is, it's like me having a more efficient vacuum cleaner is one thing. And then you mentioned this other thing where there is like,

where you're somehow, where the savings end up being kind of almost somewhat circular. So if I'm, maybe I can reduce, say the cost of cloud computing, you mentioned that in many cases, because I've made it cheaper, I would then recirculate those into doing more rather than actually reducing the total energy use.

Is there anything, have you, maybe you could expand a little bit more on that part, because when I think about things like, say, AI, or I think about some kind of, some things related to cloud computing, we can totally see this, and there are very much arguments basically being made right now, that say, "Oh, well, all you need to do is focus on your cloud bill going down by half," for example, "and then that will be, and that will achieve your savings."

Well, that's what you need to care about." And it sounds like, if I was to focus a bunch of time into halving my cloud bill, I'd then have a product manager or my CEO say, "well, okay, look at all this, look at all this money we've saved. Let's reinvest it into doing more so we can do, so we can have more of a competitive advantage in our particular field," for example.

Vlad Coroama: Yeah, of course, if you want... reducing your impact means always reducing your overall impact and not becoming more efficient. So in a sort of narrow view of what you asked, if your manager came to you and said, "hey, let's be twice as efficient," the answer should be "no, let's overall consume less." This being said of course, it's again, it's very difficult because AI in particular, and AI is the most uncertain domain, as you very well know, or certainly one of the most uncertain how it will develop in terms of energy consumption.

And so it might go through the roof or it might not, you know, and it will depend on so many factors, but it also has, it brings about, and I keep coming back to the indirect effect. Sorry for that. But, you know, you cannot forget them. So I hear very often the argument of sobriety, of digital sobriety.

And of course, it's good because to achieve various, you know, goals, whether they're achievable or not, or any goal, any limit we want to achieve, of course, all sectors have to go down. But then AI can use substantial societal or environmental benefits when it's environmental benefits, and if you can really put your finger on them, then it's easy.

Then you, then it's a no-brainer. Of course, it's worth spending, you know, the additional data center to train our models better. Also, by the way, the energy consumption in our devices for a model inference. Because, for example, I have now a paper under review where we measured a bakery chain in Germany and they deployed AI to predict the demand for bread and thus to reduce food waste and the results show, so we made a sort of a benefit-cost analysis.

And the benefits, energetically speaking, are so much larger than the costs of deploying that AI system, training it, inferencing and so on. So when there is, then it's a bit of a no brainer. Unfortunately, you sometimes have, you know, societal benefit at an environmental cost. And then it again becomes harder because then it's a scientifically non-answerable question, then it's an ideological question, right?

"How important?" Or a value-based question. But to come back to actually your question, because I think I went perhaps a little too far away from that, I keep, I keep bringing the indirect effects because I think they are so underrepresented in the discourse, both academic, but even more so societal.

So your question was about the efficiency of, or could you say it again? Sorry.

Chris Adams: Yes, so the question I'm putting is, say, if I want to reduce emissions, it's very easy for me to just talk about, "look at how much more efficient I can be" if I'm a developer, I often think that, you know, I've, I'm incentivized and I am rewarded by making things more efficient. It feels like, if I just focus on halving that cloud bill, for example, there's a risk that they'll just bring that cloud bill back up again, for example, or bring the environmental impact back up again by using some of the savings to do new things.

So the thing I would need to, things we'd kind of need to be able to do is basically have this notion that, okay, we do need to be on a kind of glide path downwards in absolute terms, for example. We can't just talk about emissions intensity, because this is a common thing that you see being, that's coming up in quite a few places.

And this is something that organizations tend to report now a bit more as a way to avoid talking about absolute figures. But it feels like if we're going to do this, we need to look at absolute consumption, just as much as efficiency. And efficiency is one of the strategies you would use to reduce consumption in total, in absolute terms.

Right?

Vlad Coroama: Yes. Yes. Absolutely.

Chris Adams: Brilliant. Okay. Thank you for clearing that part up. I...

Vlad Coroama: Well, it was more you clearing it... But yes.

Chris Adams: This is part of what we're now doing is we, running through some of these to make sure that I understand it and i, when I'm doing this, I'm basically standing, this is helps me explain it to other people as well. So this is a, yeah, this is totally okay.

All right. So we spoke a little bit about rebound. There's a few different flavors of this that we had. And you touched on this idea that if you just look at one aspect, then you can miss some of the kind of wider systemic issues and systemic impacts. And this feels like a nice kind of segue to talk about some of the other work that you've been doing, because it's actually where I came across some of your other work about trying to quantify the environmental impact of a service across multiple areas, basically.

And I found this really helpful where, when I first read it in 2020, because it found, it provided a somewhat kind of rigorous way to help address the fact that a lot of the time people will overstate either the savings, overstate the damage being done in this, in these areas. And I think the name of the paper was, sorry, it's a bit dry.

It's Methodology for Assessing the Environmental Impacts Induced by ICT Services. But one thing that was really nice about this is you would say, "well, you need to think about how much more efficient something might be, but you also need to think about what kind of take up that might be for something."

So, and all of these things here. So. Maybe I could actually talk to you a little bit about this because it's very, very common to see very, very kind of extraordinary claims about efficiency or extraordinary claims about savings under perfect circumstances. So maybe we could talk about like, when you look at this stuff, are there common mistakes or common kind of omissions to look out for when you see people talking about the savings delivered by maybe a new service, for example, so you can help develop some kind of intuition? Because this is one thing I think we don't really have the language to talk about this right now. And I think one thing that your papers did was actually introduce some helpful terms or some helpful language to talk about some of this.

Vlad Coroama: Yes, in all honesty, I... first disclaimer, we didn't provide a cookbook recipe how to do it and how to arrive at a net impact. Again, system boundaries are the word and we don't yet have the tools for that, but this being said, you can try avoiding the most common and sort of low hanging pitfalls if you want.

And perhaps the most common is that, that you read is, you know, a juxtaposition or direct footprint of a service and the benefits in fitting uses. So direct footprint, which is by definition, it's inherently negative from an environmental perspective, as almost any human activity, and the other side, the indirect benefits.

But then conveniently, and I'm not saying that this is necessarily purposeful, it can be out of naivety or, you know, but it is convenient for getting the indirect negative impacts. So you always have, so I think that's a helpful way of thinking about it. You have the direct footprint, negative by definition, and then you have indirect effects, which are both positive and negative. Ideally, you would try to cover them both. The very sort of high level indirect ones, which are systemic, you cannot cover. But you can at least take care of the direct rebound, for example.

Chris Adams: All right. Thanks for that. So you spoke about, so there's leaving omissions from here. There's one thing I've seen that in a few places, so I've seen like, say, caching services basically say, "well, look how much, look at the savings you've received," for example, without telling you the full amount. And this is something that I think Uber have released.

They've shown, there's now a calculator to see how much cleaner your ride would be if you use an EV car versus another one, but you don't see the absolute numbers, for example. I mean, what's wrong with this? Like, is this a thing that... is this a good idea or should, or if you were to do this properly, like, how would you make this more representative, for example, when you see examples like this?

Vlad Coroama: Well, I don't know those particular calculators. I haven't used them or seen them. But from how, what I understand about them, it's very often a question of baseline or of the counterfactual. What is your counterfactual? If I hadn't used this, you know, for example, green taxi service in your example, what would I have done?

If the assumption, if the baseline is, I would have used, you know, a very inefficient internal combustion engine-powered car instead, then, of course, depending also a bit on the electricity mix of the grid, usually it will be positive, right? The overall impact. But the thing is, it might have replaced public transportation or no trip at all.

And then it's a rebound effect. So if I am keen, you know, I am taking a taxi and you know, it just tells me, "Hey, if you now take the green one instead of this," then I would say it's probably a reasonable assumption. We have some production issues, so from a life cycle assessment, of course, it's a bit complex, than it's probably pictured, but in essence, it's probably not incorrect.

But overall, what happens is that if a taxi ride is cheaper, or if I have what is called a moral hazard, so I have a clear conscious, "oh, I'm going green so that I can take it," and then I'm taking one that would not have existed in the counterfactual, then of course the net effect is there.

Chris Adams: Okay. So that last part is like, I get an Uber or I get an electric taxi and, so I sort of, to a restaurant and I then decided to eat a big fat steak, as an example, like as a way to kind of balance these out. Like there's maybe an indirect, there's a direct saving, but systemically, I still have created more of an emissions. Like, I'm not going to try and do the calculations between steak and a drive, but you get the general idea.

Okay. So that's where some of that comes in. You mentioned this thing called a counterfactual. And I think this is actually one thing that might be quite helpful because we've seen a number of papers and reports being used to talk about how, you know, you can achieve sustainability through AI and we've seen them written in, say, in the early 2010s or even the early 2020s.

And then there is often a lot of interest in talking about how good something could be, but there seems to be less, historically, we've been less good at tracking whether those savings have been delivered. Is this something that maybe you could talk a little bit about that? Cause I remember you write, I saw recently you wrote a little bit about the, this kind of reporting, the fact that there's a kind of gap in how we talk about this and the following through part, maybe you could just follow, just expand on, on this and why you need this, this extra information to kind of see if things are working basically.

Vlad Coroama: Yes. So for AI, I think it's a bit too early to tell, we have not yet seen like a series or I'm not aware of any, a series of studies or even like one old study that made some predictions and the authors didn't come back to it to say whether anything was delivered on. And I think it's not also a typical, like, computing thing that we do lots of predictions. And as the old adage goes, "predictions are difficult, particularly about the future."

Chris Adams: Yeah.

Vlad Coroama: Yeah So first, you know, when our predictions were right, we might like to go back and highlight this that we were right already back then. And otherwise we might conveniently forget that we made those predictions.

For the computing domain generally and not AI, because again, AI, I think it's a bit too young for that phenomenon to be seen one way or the other, but for computing generally, there is a track record of various, both companies and sort of lobby groups of the IT industry or of the telecom industry doing sort of predictions and then continuing. One very known example is GESI, Global E Sustainability Initiative that published every couple of, every four to five years, starting 2008.

They publish a series of studies. The first one was called SMART 2020, then SMARTER 2020, then SMARTER 2030, and so on. And there are predictions, the first two ones are called SMART 2020 and SMARTER 2020 because they're where to the year 2020.

Chris Adams: Ah, I see.

Vlad Coroama: And and they predicted many gigatons that would be saved through digital technologies.

I think the first one was 7 point something and the second study around 9. 1 or 2. And that's, that's quite a chunk of the sort of fifth of CO2 equivalent that the humankind puts into the atmosphere every year. So that's a very substantial chunk. And if that would have been true, it would have been amazing.

But now, well, 2020 is past, heh, and we published newer studies, but they didn't look, you know, how did this stand up to the test of time?

Chris Adams: Wow. Okay. That's, that feels like quite a gap that we probably should be trying to close. If we're going to be talking about, we're doing research in this and seeing what is going to be effective as time is ticking down. Right. Vlad, that's actually, I really want to dive down in that rabbit hole, but we're coming up to time.

So I'm going to have to be a good boy and try my best to stay inside the time we do have. Vlad, we've covered quite a lot of interesting areas and dived into quite a lot, and I've shared a couple of links. The show notes have series of links to the papers and things like that. If people do want to continue this work or continue following what you're up to, where should people be looking beyond just the show notes of this podcast for example? Is there a website that we should point people to, or do you have a online presence you would direct people's attention to?

Vlad Coroama: Well, anyone, if, I mean, you can post my, a link to my LinkedIn account, if anyone wants to contact, I'd be happy to, you know, to engage in conversation and continuing discussions, this is what I do. So other than this, there is no value that is specifically say on rebound effects of digital technologies.

I wouldn't know of any, but there is, for example, the ICT4S conference. So the ICT4Sustainability, that, that conference that started back in 2013 and where indirect effects of ICT are quite a powerful presence. This year's edition will be end of June in Stockholm, Sweden, and I'm co-organizing with a couple of other researcher.

So, with Mattias Höjer at KTH, with Tristan Brehmer in Lancaster, Charlie Wilson in Oxford, and Dan Schien in Bristol, we are organizing a workshop on this very topic, indirect effects of... called wait, I'm no longer sure what acronym stands for, but something with indirect something. So a workshop on assessing indirect effects.

So there are a couple of venues that are scientifically dedicated to this, but I, there is no unfortunately, no like, you know, portal where everyone has the topic.

Chris Adams: Like Institute of Rebound. Yeah. Okay. And yourself, I understand that there's the organization that you work for. The Roegen, is it Roegen Center of Sustainability? That's the one. Yep. So that's roegen.ch is the place people would look to if they want to see any future publications and research in this field from you.

Brilliant. Well, Vlad, thank you so much. I've been looking forward to this and I have to admit, I'm a bit of a fan boy. I've really enjoyed a bunch of the papers and things you've been publishing over the years, and I really hope you continue to do them because they come up with really nice examples that I can help explain to other people.

So thank you once again. And yeah, hope you have a wonderful week. Take care, Vlad.

Cheers. You're too kind, Chris. Thank you as well. And by the way, thanks for your great work that you and your foundation are doing. So thanks for that as well. And thanks for having me. Cheers.

Cool! Thank you. Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

Nowcasting and Using Computers to Reduce Emissions

21 mars 2024 | 45 min

In this episode of Environment Variables, host Chris Adams is joined by Rachel Tipton, a full stack developer at Open Climate Fix, to delve into the intersection of AI, green software, and electricity infrastructure. Rachel emphasizes the increasing demand for electricity due to factors like server centers, EVs, and electrification efforts. They explore the concept of carbon awareness and its implications for software development, shedding light on the complexities of optimizing energy consumption and reducing carbon emissions. Through Rachel's journey into coding and her work at Open Climate Fix, listeners gain insights into the critical role of technology in decarbonizing the electricity grid and mitigating climate change.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Rachel Tipton: Linktree | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

The Tech Generalist | Rachel TipTon [04:06]
Women Who Code [04:08]
Subak [14:09]
GitHub - openclimatefix/Open-Source-Quartz-Solar-Forecast [15:09]
Real Decarbonization: How Oil and Gas Companies Are Seizing the Low-Carbon Future | Tisha Schuller [41:29]
html garden [42:36]
https://sceneries.site/#home [42:49]
https://github.com/search?l=&p=1&q=user%3Aopenclimatefix+label%3A%22good+first+issue%22&ref=advsearch&type=Issues&utf8=%E2%9C%93&state=open
openclimatefix (Open Climate Fix)

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Rachel Tipton: I think it's really important that people are aware that our electricity infrastructure is going to have to handle a lot more electricity, whether that's going to be like server centers, like computers in different places, but also like EVs. If we're going to electrify everything, we're actually going to be using much more electrical energy in the future because it's going to be replacing those carbon-based fuels.

I'm your host, Chris Adams.

Hello, and welcome to another episode of Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. When we talk about green software, it's common to start talking about energy efficiency, i.e. how much energy you need to perform some computation, or even hardware efficiency, i.e. how much physical hardware is needed to perform that computation. One idea that is less easy to understand is this idea of carbon awareness, an idea based around the fact that depending on where you are in the world or what time of day it is, the amount of fossil fuel being burnt to power the grid we rely on can change, which means the energy could be greener or less green.

This is because beneath the stack of technology that makes up the internet, there's an entire energy system that is at least as complex as the internet. And if you have visibility into that system, there are ways you can operate your systems at the level above to take advantage of these changing conditions, often to save on carbon emissions, and in some cases to save on cash too, as the price of energy can often change based on the amount of cheap renewable energy being fed into the grid that we draw electricity from.

As I mentioned before, this is an idea that takes a bit of getting used to, and when you hear it from publicly traded trillion dollar companies whose own reported emissions have only grown in the last few years, you may be right to be skeptical about carbon awareness. Now at FOSDEM, the largest open source conference in the world, in February, my guest, Rachel L'Abri Tipton at Open Climate Fix presented a fascinating talk about solar nowcasting, something we'll cover later.

But crucially, as someone who recently switched careers, both into software engineering and into working with grid operators, she seemed a good person to talk to, to learn more. Hi, Rachel. I've given you a bit of context about how we met. But I figured, can I just give you the floor to let you introduce yourself before we start?

Rachel Tipton: Hi Chris, thanks for having me on the show. I enjoyed listening to the other environment variable episodes

in preparation for the show and I think that I'm a convert. So I'm Rachel Tipton, I'm a full stack developer at Open Climate Fix, OCF, that's the shortening of Open Climate Fix, is a non-profit product lab, and we're developing renewable energy forecasts to decarbonize the electricity grid. We'll get more into that later, but I'm really happy and excited to be here speaking about my own coding journey today and my time at Open Climate Fix. In addition to speaking at FOSDEM, I was also part of organizing the energy dev room program that I spoke in, and that was organized by LF Energy.

So I was volunteering for them and I ended up submitting the application for Open Climate Fix to speak. So I then asked if I could present some of the slides

that they I had a plan for us to speak with, because I thought that we should be talking at the conference. I also write a blog called the Tech Generalist.

I only have about three articles in it, but that's something that I've been doing since I started my coding journey. And then I also participate in Women Who Code events with the London chapter of Women Who Code.

Chris Adams: Cool. Thank you, Rachel. And when you refer to LF Energy, you're referring to Linux Foundation Energy, right? That's one of the organizations that was, okay, got it.

Rachel Tipton: It's a small group within the Linux Foundation.

Chris Adams: Okay. Thank you for clearing that part up. Okay. So if you're new to this podcast, my name is Chris Adams. I am the executive director of the Green Web Foundation. Which is a Dutch-based nonprofit, basically focused around the idea of a fossil-free internet by 2030. But I'm also one of the policy working group chairs in the Green Software Foundation Policy Working Group.

I'm also one of the regular hosts of this podcast. Okay. And just a quick reminder, everything we talk about on this show, we're going to link to in the show notes, and there will also be a transcript provided. So if there's something that you missed the first time, there should be something available.

And because this is a software engineering podcast, we also have this transcript and the links, the show notes in markdown form on a GitHub repo that we'll be linking to. So if there's something you missed, or if there's something you feel that you need to be corrected, we do welcome pull requests. All right.

Okay, Rachel, I was going to ask if you're sitting comfortably, but I realized that you're standing where you are in the world. So I guess, should we start? Does that sound okay to you?

Rachel Tipton: Sounds great.

Let's go.

Chris Adams: All right. Okay. So we've teased the audience with a few kind of geek catnip terms, like solar nowcasting and so on. But before I dive in, I wanted to ask. I know that you grew up and studied in America, but I think you, I think you mentioned that you're calling me from France and you haven't always been a software developer.

So could you maybe expand a little bit about that before we dive into the nerdery?

Rachel Tipton: Yeah. So I'm actually speaking to you, Chris, from Olsene. It's a small village in Belgium near Ghent, for people know where Ghent is, that's the point of reference. I grew up outside of Chicago in a state called Indiana in America, and I've lived and worked in France since 2011. So I moved to France in 2011, and my time is split between Lille, which is a city in the north of France, and then this village in the rural area of Belgium. So we have a big garden and it's quite springish outside today. So we've got a magnolia tree that's in full bloom and it looks really nice. And we've had a lot of rain this year in Belgium, so our basement is flooding. Everybody's basement is flooding here. We've had twice as much rain as we usually get.

Chris Adams: Wow. So firsthand experience of changing climates. I guess. Yeah. You, you can't deny some of that then. All right.

Rachel Tipton: No, no, not at all. Yes. I think it's important to say that I'm a writer and a maker and for most of my 20s. I made my life decisions based on being able to have time to write and have time to make art. Some of that writing is out in the world. A lot of that writing is probably in some place in manuscripts, but that's like who I am at the core. And so I didn't study computer science or physics or engineering like some of my current colleagues, but I studied English literature and geography. I did have a stats course in my geography study. So that was kind of like the beginning of my exposure to data.

And I got out of university with critical thinking skills and really good writing skills, but no clear career path.

So I'm an interesting person. I have lots of interests, can talk about many things, I read voraciously, but really up until I decided to shift into coding, I hadn't felt like I had a career of some sort. So I had been, when I decided to try coding, I'd been working as a freelance technical editor and writer alongside teaching English full time at a university in the north of France.

I was also teaching at business schools as well. And I had ended up in France because I wanted to learn French. So that was part of, you know, "I want to be a writer. I think learning a second language would be great." Now that I have, you know, I spend part of my time in Belgium. I'm in the Dutch speaking area of Belgium.

So of course, I've also...

I speak Dutch now to like a intermediate level. So I think it's that language learning part of myself that actually made it quite easy to start learning coding languages because I understand syntax and I understand patterns within language and that

like, I really love grammar.

Like I love doing grammar exercises and I feel as though solving code problems are to some degree beginning to understand the grammar of a coding.

Chris Adams: Oh yeah. I see

Rachel Tipton: Remember we have a coding language and then beginning to understand like larger data structures is like beginning to understand like paragraphs that lead to arguments that, you know, that you can build something with your code structure in the same way that you do linguistically with language.

I know not all coders approach coding in that way, but that's the aspect of myself that perhaps made me attracted to learning how to code.

Chris Adams: No, I could totally see that. Like I live in Germany and when I came from England to Germany, one thing I picked up was, well, the way that you can just like bash words together to make longer words. When I first came here, I was terrified of how long all the words were. Then you realize they're just like one word linked to another word linked to another word.

It feels like, "Oh, this is just like jQuery. It's just like chaining," you know? And then when I realized that I was like, "Oh, okay, this makes a lot more sense." So I totally see where you're coming from actually on this, Rachel. So yeah, I'm also glad that it's not just me thinking about this. So yeah, that's reassuring.

Rachel Tipton: Yeah, syntax is really, really exciting. Okay. Yeah. Let's see, like in my story, in the, it was the middle of the pandemic when I decided I need to make a change. I think a lot of people had some sort of... I've talked to people who've had a moment of, "Oh, I need to shift something fundamentally in my life," and for me, it was my job and my career. I loved the process of teaching English at the time. I loved my students. I'd even been able to develop a, like a creative writing class for, at the university where I was teaching. So it was really great to get to teach poetry and short story writing to my students, but everything else outside of the teaching was really draining.

And even though I had time in the summer to do my writing, I was completely burnt out, which I think a lot of people who teach have experienced burnout before. So I was also very, very tired of a myth that I had for myself of being like, needing to suffer to be an artist. I think that was like a holdover from my twenties.

I've talked to other artists who have this.

Stepping into the

Chris Adams: artist stereotype thing.

Rachel Tipton: It's the starving artist stereotype. And I think I bring it up because I think it's false. And unless you are a well funded suffering artist, then it's hard to live that artist's life. And so

for me, I was also looking to make a decent living so that like the environment in which I was making art was a little bit more just like financially stable. And then I didn't feel as though I was like giving my heart to my job and not getting anything back from it. So

yeah, that was the place where I was. So, a friend mentioned coding and I started to just started to research it. Like I contacted two people that I knew who worked in tech and one of

them was really encouraging.

He was like, "you know, the tech space needs more women. You can do it. Like, you'll figure it out. Like JavaScript is not rocket science." I trusted him and then somebody else had mentioned boot camps and this particular friend is a business analyst. So I just asked her to put me in touch with coders that she knew. And so she connected me with

two female coders and then I've talked to them and like through those discussions, I started to be exposed to words like front end, back end, full stack developer, and I had no idea what those words meant. Yeah.

That was like three years ago. It was just a lot of jargon, but I had also, as a technical editor, I had already worked in spaces where jargon was used and I know that jargon means stuff.

You just have to listen to it long enough and eventually it's meaning will clarify itself. It's like being a

Chris Adams: I'm glad you mentioned about jargon thing.

Rachel Tipton: Mhm.

Chris Adams: Yeah. Because I, I think it's very easy for people to basically say, well, I'm not going to, you know, you can rag on jargon, but in many cases, it's actually a very efficient way to get something across once you've established a context. So I totally understand.

Like when you're using the term "MEAN stack" or 'Jamstack' or something like that, the fact that it means something is way, way, way, way, way faster than us having like another five minute discussion explaining what each letter means, for example, and things like that. So, yeah, I know where you're coming from there.

Totally.

Rachel Tipton: Yeah, it's really, jargon is useful for the insiders, for the outsiders it makes it seem like it's much more complex

than it is. As a technical editor, I specifically had worked for a society of doctors who basically wrote papers on blood diseases, so it's like all of the papers were on bone marrow transplantation and I was editing it.

And there's so much jargon in that space around like treatments and diseases, but with time I eventually understood what they were talking about, just in basic terms I was able to actually write about it and talk about it. And so I had the same approach to the coding space. Like from that experience, I was able to apply that to being in a tech space and be like, "this is just like being with the doctors who, you know, thought that they were really

Chris Adams: Yeah. It's not special. It's just new. Yeah.

Rachel Tipton: Yeah. Yeah. It's not special. It's just new. And I think if you show up with like a beginner's mindset, I like the idea of like the beginner's mind that's completely clean. You're just there and you're curious and you're learning. It's a really great experience to just move forward with that like confidence and humility at the same time.

Chris Adams: Cool.

All right. Thank you. So that gives us an idea of a potted history of where you were and where you came from and how Python or JavaScript might be a fourth or fifth language in the context, rather than just like a first or a second language in that sense.

Rachel Tipton: That's how I'd see it. I don't speak Javascript or Python, but I can read it. Yeah.

Chris Adams: Okay.

Rachel Tipton: Yeah.

Chris Adams: Like Latin, I suppose. All right. Okay. So that's some of the background that you had and you've been working with Open Climate Fix for the last few months or so. So that means like messing around with like Postgres or Python and like working up and down a technology stack, for example, and trying to figure out how to use computers for climate wins.

I found Open Climate Fix really interesting because Dan Travers, who was, he was actually one of the advisors and mentors for our organization a few years back when we were going through an Accelerator Progress program with the Green Web Foundation with Climate Subak, an organization. But I really like what they do and I really like the open approach.

So maybe you could just share a little bit about who they are, what they do, because I think that'd be quite helpful context in some of the discussions about computers, forecasting and how some of this fits and why you'd even care about any of this.

Rachel Tipton: Yeah. So the. Company Open Climate Fix was founded as a non-profit product lab five years ago. Like we just turned five, I think, what did it say, late February. So it was founded in 2019 by Dan Travers, who you've mentioned, and Jack Kelly, who worked for the Google DeepMind project. And yeah, the idea of Open Climate Fix is that we connect people who are doing ML research with the energy sector to help decarbonize it.

And the initial product that we've developed is called Quartz Solar, which is like a short term, like we've mentioned solar nowcasting, but it's a short term forecasting service. And I've tried to make this as simple as possible to explain to someone on the outside, but basically what we've done as like. The way, the place we've implemented the court solar forecast is with the National Grid ESO in their control room. So in the UK, there's funding that's called, I think it's National Innovation Allotment that an organization like National Grid can use to collaborate with an organization like Open Climate Fix to develop technologies that will

Chris Adams: I see.

Rachel Tipton: reduce carbon emissions.

Chris Adams: Can I just check with you? Sorry to interrupt. You mentioned the national grid. So, and you also mentioned this term ESO. And maybe it might just very briefly, just to kind of give people who may not be grid specialists, what those terms mean. So national grid is the grid operator for all of the UK.

They maintain the UK grid, although they're actually, the company is actually active in more than one place. But this ESO term, maybe you could just unpack some of that cause it might be a bit more helpful for people.

Rachel Tipton: Yeah. So when we say National Grid ESO, we mean the electricity system operator. And what that means is they're responsible for operating the electricity system. So the grid that is getting electricity from where it's generated to people's homes. And they need to make sure that the electricity, that the amount of like electricity that's demanded by homes, businesses, everywhere, and the amount that's being generated is balanced on the grid.

Chris Adams: Gotcha. Okay. All right. Back to where we were. Yeah.

Rachel Tipton: Yeah. From my own learning journey within Open Climate Fix, there's so much that's happening around the grid, but the basics is that, for example, when I turn on my electric tea kettle, the grid is aware that it's giving me power and then it'll be shutting off that power at some point. And like power grids are balancing, are doing these balancing actions and balancing decisions are being made very quickly. And that's how a power grid works in very, very general terms, which is beyond, of course, my expertise, but it's what we work with at Open Climate Fix, which I find pretty cool. So the product that I work on is the Quartz Solar web application, and it consumes data from several APIs, and those APIs are serving a solar forecast. And so National Grid uses the Quartz Solar app, and I'm using this just as an example. There's other, we have other customers that are also using the Quartz Solar application, but National Grid uses it to make like real time balancing decisions. And so what that means is that they'll see how much solar energy is being generated on the entire UK grid.

And they're able to determine whether or not they're going to be using like electricity that's coming from fossil fuels being burned or electricity that's coming from solar.

Chris Adams: Oh, I see. So the idea being here is that because they, if they've got better information to this, then, you... so things like fossil fuels may be polluting and kind of not great in climate terms, but because they're dispatchable, that you can turn them on, they have a degree of controllability. So this helps you kind of balance some of this.

And maybe this is actually one thing that you, maybe we could expand on some of that in a little bit more detail, because you spoke about that some of this is providing some forecasting to reduce the amount of uncertainty so that you don't need to have so much kind of backup fossil fuels ready to kind of ramp up, for example.

And I think when I saw your presentation at FOSDEM, A, the control rooms just look, look exactly like you'd imagine a control room in a James Bond movie to look, but also the, there was this massive honking gray picture of a, some kind of gas turbine. Maybe you could like talk a little bit about that and how that relates to some of this, because I think how, like you mentioned, we just see power going in and on, but there's a whole set of complexity under the stack beneath us that we're not really aware of.

Rachel Tipton: Yeah. So the, the purpose of spinning reserves for a power grid is that you always have turbines that are running. Then that might be a gas powered turbine or a coal powered turbine, but they're at the ready in case energy is needed. So one of the reasons that a good solar forecast is important is that a lot of these turbines are running on the grid and they're just sitting there at 50 percent capacity, and they're also, when they run at 50 percent capacity, it's also 50 percent efficiency.

And they're not generating electricity for the grid. They're just kind of at the ready in case they need to be spun up or spun back down.

Chris Adams: Oh, I see. Okay. All right. So that's not a million miles away from how, like, if you want to have a computer running, for example, even inside a cloud system. Right. You still need to have a bunch of computers, which are still there and they're still idling and they're still generating, they're still generating heat and using electricity and so on, but they're not necessarily doing useful computation.

So the same idea also works at the kind of one level down with like a honking gray coal-fired power station or gas. So even if it's not generating power, it's still emitting just because you need to have some of that capacity ready for it to respond. Ah, okay.

Rachel Tipton: Yeah. And the the benefit of like having an accurate solar forecast is that you'll have fewer of these turbines that are running at half capacity. And you'll be able to make the decision, okay, we don't need, let's say you've

got 20 of them running, this is not at scale, but there's 20 of them in random places running ready to support the grid. You can make the decision if you have a very accurate solar forecast that, "oh, we'll have, let's say 50 megawatts of solar energy that's coming onto the grid. So really we can turn those off." And if they're, if you're confident with your solar forecast, then you can confidently make that decision and you'll be

Chris Adams: I see.

Rachel Tipton: yeah, you'll be saving carbon emissions and then also the balancing costs of the grid, cause you're not burning fossil fuels at that point.

Chris Adams: Okay. So it's a bit like say if I've got a cluster of machines and I just know that I'm not going to have this much capacity, I can just like auto scale down a bunch of computers and presumably like with a cloud provider, they would switch that off. We would assume they would switch it off or something like that.

But that's kind of what it did. It almost like removing it from this pool of reserve capacity. That's what some of this is doing. And you said one other thing as well, is that just like how computers have like a area where they are, have optimal efficiency. It's the same thing with say some energy plants, right?

So they might not be a very efficient, say 60%, but if they're 80 percent capacity, then they're more efficient. There's a little bit like that, that you, you referred to as well, right?

Rachel Tipton: Yeah, exactly. I won't add any more to that. I also think of it as like, when we work like, for example, a part time job sometimes takes as much energy as a full

time job, but a full time job might be working more, like, it might be more efficient in terms of, like, bringing in income. Like, if you have to go to your part time job every day, you

still have, like, that get up and go in the morning, go to the job.

Chris Adams: Even though there's only that much. Yeah. I mean, even though you only do that many billable hours, you've still got the commute and all the other things, huh? That's actually a really useful comparison. I didn't really viewed it in that, in that way, but that makes it feel quite a bit more intuitive. Okay.

And so this sounds cool. Open Climate Fix has been doing this kind of work for a while. And when I look at the kind of GitHub repo full of OCF's work and speaking to various people, I know there are, there's a very, very kind of high nerd count over there. So presumably there'd been some calculations around like the annual savings or what you could achieve with this.

Because I understood that when I first, when I first saw kind of Jack doing some of this, Jack Kelly, I remember him writing these blog posts saying, "oh, I'm, I'm leaving Google. I want to work on climate change. I'm trying to figure out where I should spend my time." And I think one of the, I remember reading a post where he basically said, "well, I think this is the most effective thing I can do using my specific set of skills."

So presumably there's some number about the savings that could be delivered or that are being delivered by something like this, right? Cause it's been going for a while. Yeah.

Rachel Tipton: Yeah. So, our back of the envelope calculation, which is what we call it, is the back of the envelope calculation, is that the operation of our solar forecasting live service, Quartz Solar, in the National Grid Control Room, saves about or decreases like UK carbon emissions by 000 tons per year.

Chris Adams: Wow. That's not bad for a team of... You're not really a big team, are you? There's like

Rachel Tipton: No, we're around, well, around 12 people. So less than 10

people.

Chris Adams: Wow. That's, that's pretty impressive.

Rachel Tipton: For me, getting to work on a project like this that can like say how it is impacting carbon emissions or decreasing carbon emissions in a large scale way, not just by like using electricity from like solar in my home, but actually seeing that like the energy transition requires that our infrastructure changes and how we operate our infrastructures.

Yeah.

Chris Adams: Cool.

All right.

Rachel Tipton: Yeah, it's impressive. I don't know what percentage of UK CO2 emissions that is per year, but that's the amount that we're calculating.

Chris Adams: All right then. Okay. Thank you. So, so just to record some, some kind of, okay, forgive the term grid choreography, you can actually achieve some of those savings like that. All right All right. So, okay. We've spoken a little bit about OCF and we spoke a little bit about some of the grids and how some of this works.

Now, I know that where one thing that, one thing that's kind of key in the Green Software Foundation is that there's a lot of open source code, because this is one of the ways to increase reach. And also in my organization, the Green Web Foundation, we have a similar kind of take on open. And like you folks, you literally have open in the name.

And if you're coming into, switching into a new field, it can be a bit scary working in the open to begin with. So maybe you could just touch on some of that first, before we dive into some of the specifics about this Quartz thing, because so far we've been speaking about open source and things like that, but there was a really, really cool demo that you shared that I...

I think we have to stop teasing people soon and dive into talking about.

Rachel Tipton: First I'll, I'll talk about our, our approach to open source and then the Open Quartz Solar Forecast itself. So the intention is to have as much of our code and data available to anyone who wants to work on decarbonizing the power grid or, I don't know, use the code for other as well.

Last year, we had some mentoring from someone who works in the open source space and they explained the concept of coding in the open versus being an open, like a thriving open source community. And right now we're in that space of transition of trying to offer a pro like a product or a tool, which would be the open source, Open Quartz Solar Forecast that people in the coding community can contribute to.

So I feel as though I started my coding career coding in the open. So someone can go to the, I think it's the Quartz front end repo on Open Climate Fix and see the code for our next JS app,

which is the Quartz solar app. But to actually run that app, you can see the API that we're using, but you'll have to like set up your own database to be able to use that.

So it's not readily runnable, I would say, but it's out there. So if somebody wants to take it template for their own application, they can use it. So I, what I've said is like the, the code for my career is out in the open, it's the good, the bad, the ugly. There's a lot of

commits where I'm just like fixing linting errors and that kind of thing. But it's also nice because then I can point to those repositories when I'm applying for other jobs, or people can see that I've contributed to

an open source project and I personally love the community aspect of people collaborating on code. There's something really beautiful about that collaboration to me and that the community ownership of the code itself.

And so, the idea of Open Climate Fix is that we're providing code that other communities can use or creative people, smart coders who want to certainly implement like an ML model, we're showing them how to do that. Like if you want to implement a large deep learning model. Yes, an individual might not necessarily have the, the data infrastructure to do that, but they can see how we're doing it. And we're giving them those tools. Could be something interesting to add, the energy sector was like one of the first sectors to digitize, I think like in the 70s and 80s. And so some of it's like digital infrastructure is a little bit dated, let's say. Like the

images that you saw of like the control room.

It looks like older digital infrastructure. And so there's this other element to the, like, intention behind Open Climate Fix's work, which is to provide like an upgrade to that digitization for the energy sector. I don't know if that's clear, if that makes sense, but that's one...

Chris Adams: That does make sense. Cause we did an interview with some folks from Electricity Maps and they, they provide like universal APIs across bunches of countries. And some of the things they had to do, I think when I was speaking to Tony Van Sweet, who were telling me about it, I think it's in Singapore.

They need to, because they can't get actual data. They need to basically use optical character recognition on a image that gets updated every five minutes to pull a data point out. Because you just have a system that was designed a certain way and that's how it's been like that for however long. So yeah, I totally understand where you're coming from with that.

Okay. So we spoke a little bit about this and you spoke a little bit about working in the open and about how there's a difference between coding in the open and having like an actual open source project that kind of gets people involved and things like that. I figured, can we talk about the demo now and some of the things you shared at FOSDEM because that was super cool.

Rachel Tipton: Yeah. I ran in this morning to see like what the, if I had like a solar panel at my house, like what the solar generation would be. So what we presented is it's called Open Quartz Solar Forecast, and we have a repo for it on the Open Climate Fix GitHub account. And basically what it is, it's a, it's a solar forecast that is site specific. So if you have a latitude and longitude coordinate, you basically plug that in to the, the model itself, it's a Python package. So you install the package and then you run the function. I think it's like run forecast and you put in your, the lat long coordinates for your site, as well as a timestamps, which will be like

the initializing null of the forecast, Like right now. And then. It generates a forecast, we call it the 15 minute forecast resolution. So you'll have data every 15 minutes out for 48 hours for your solar forecast in a specific location.

Chris Adams: So when you were presenting this, so you mean you're, you're speaking to me from Belgium and you presented this in Brussels, but the national grid is based in the UK. So the, the coverage for this is, it's not just one country. It's, could you maybe talk a little bit about that? Cause that was the thing that really surprised me.

This is why I didn't realize it was possible, actually.

Rachel Tipton: Yeah. So to explain a little bit about the site specific forecast, it was initially developed by using, we had like a thousand different sites around the UK that the model has been trained on. And the idea is that it takes in numeric weather predictions, like those are the inputs, is the actual local PV data, and then also numeric weather predictions. And it'll create a solar generation prediction for that specific site. For like the machine learning nerds, the actual model itself is a gradient boosted tree, which is like a bunch of decision trees making decisions together. That's the extent of my talking about the model itself. It's been trained on UK data, but it can be applied then to any other like, lat long coordinate in the world.

Chris Adams: Yeah This was the thing that was really cause I think you were presenting with, I think it was Zach, Zach Watts, who,

wow, his hair is way longer than his profile photo when I saw him in person I was like, oh wow, so I can use this in Germany, or you can use this in India, you can use it somewhere else.

That was like super exciting for me because the idea that you could just run that on your own laptop to do that was, yeah, pretty freaking cool when I first came across that actually.

Rachel Tipton: And I like to see it as like a distillation of our, the technology and the research that's gone into developing a... like a solar forecast nationally in the UK down to specific sites that might be like a solar farm or like maybe an energy trader wants to have information on a specific site. And then that gets like crystallized down to four lines of code for someone that wants to generate a solar forecast for their home or any other like location. In the world, and it can be interesting because you might actually like run the forecast in a location to see if it would be useful to have solar panels there. For example, that could be a use case that you use.

Chris Adams: And there was one thing that I found quite interesting, is that we've got this whole shift to basically renewable energies. That's the kind of generational as it were. And we've also seen that I believe in various areas, like say subcontinental Africa, or even India, you're seeing an uptick in people adopting new forms of power, basically, or solar, for example.

And this feels like this is something that you could see these tools being applied in new other areas, which where there's, where you're seeing new sources of demand, new sources of load that needs to be served, for example. And like, essentially that you've got something a bit like a kind of, leapfrogging effect, like you mentioned before, like, okay, there's antiquated technology, but theoretically say in, I know India is actually one country where that you folks have been doing some work already, and there's a bunch of really interesting work using even just like kind of grid modeling tooling to actually see how you can get, get off fossil fuels.

This feels like some, it's almost like a kind of leapfrogging at that level over there. You've got the same access to, to these tools in other parts of the world as well now.

Rachel Tipton: And I think it's important to offer a tool that's easily usable and easily accessible, and then it could also be like connected to other sources of like, what are called, it's called numeric weather prediction. So it's not as accurate as our, like national solar forecast, which uses satellite imagery, but it is fairly accurate.

And I think in most cases, in many cases, it's like something is better than nothing. Balancing a power grid. It's good to have like a super, super accurate solar forecast, but having something that would give you an idea of like how much solar generation you would have in any location in the world is really useful.

Chris Adams: Okay, cool. Thank you for that. So you've explained this, so a Solar Quartz open source. We've got a link to that and we'll add it into the show now. So people who are curious or want to try running their four lines of code wherever they're in the world, just to kind of get their, their own forecast, they can do that.

If anyone has really enjoyed the episode we did about Wasm, there's an issue to see if there's a way to turn some of this into Wasm as well, because there's a bunch of Python and there's a bit of like using C++ inside it. But if you do know Wasm, there is an issue open that I opened because I figured, "well, browsers can do this.

So why not just make browsers able to do kind of solar forecasts?" So maybe that might be of interest to other people. So we spoke about this part here, and this is a chunk of this is talking about reducing the amount of spinning reserves you need to have. Like we're just going back to just some of the grid stuff and like back to this whole idea of like carbon aware.

The thing that you told me was that can most of these tools reduce the amount of spinning reserves that you might need for this. And that means that that's, you're kind of reducing the amount of generation you might need to meet demand. All right. Now that's, as I understand it, one of the key ideas around carbon-aware programming is, as you said, it's attacking the problem from the other side.

So rather than actually balancing out supply and demand by just creating new generation from fossil sources, there's this idea that you could, you know, balance this out by reducing the amount of demand you have as well. And I understand that this is not a new concept inside grids. Like people have been doing stuff like this for a while already.

And as someone who's has basically switched into technology and then also working with the grid, there's a bunch of things which are quite counterintuitive. So the question I have for you is that when you're thinking about like grid responsiveness and flexibility, are there any kind of rules of thumb or things you found counterintuitive that you might want to share with other people to kind of help them think about this, honestly, quite new and quite a bit of a complicated concept to kind of internalize?

Rachel Tipton: So something that was complicated for me to understand in all of these discussions is that we're talking about decarbonizing the grid, but there's not a lot of discussion around like minimizing consumption that's something to be aware of is that when we're talking about decarbonizing the grid, we are not necessarily talking about using less electricity or using less energy in general.

So that was something that was not necessarily hard to wrap my mind around, but I think it's an important point to just put out there because a lot of the conversations are around more developing grid technologies where we can connect more renewables, but that we would still be able to consume as much energy as we do, for example.

But in terms of like, what is it, carbon aware programming, I think that it's the way that I thought about this that was useful for me is that I have the use case of balancing a power grid and using a solar forecast to make those grid balancing decisions. And then for an individual or on like a smaller scale, if somebody has what are called flexible assets at their house, so they might have like a battery or a solar panel, they might have like a tiny wind turbine or something that's generating electricity, having a solar forecast in that setting is a little bit different than having it making decisions on the power grid, but it's still sort of a decision making and energy balancing tool that you might be using in your home. Not sure if that answers your question.

Chris Adams: That does help, I mean, because basically we're still trying to figure out a lot of the wording for this stuff. And we do, we know there's a shift from essentially having a grid, which is basically responding to load. You can have a grid, which basically always meets any demand by just creating more generation.

And there's this idea that now you could shift the use of power from certain places to where to a certain time. So like you mentioned for yourself, if you've got something like say, local generation of say renewable energy, for example, if you're able to store some of that, then being able to run off some, some of that, instead of having to draw from the grid later on at times of high carbon intensity, that counts as a climate win, for example.

And I think these are the things that we don't really have ways to really talk about yet. Or there are some papers exploring some of these ideas, but they're not particularly in the mainstream yet, I suppose. So I think this is something that we are trying to struggle with. And also when you see it coming from a number of large organizations, it does give this impression that, It can kind of almost allow people to not talk about absolute amounts of load.

And when we have seen massive new amounts of load coming onto the grid, especially with data centers, and we're seeing extremely aggressive projections and total energy use. You can see why people would be skeptical about some of these ideas, but I kind of feel that we're aware that this is happening outside of technology, that the idea of responsiveness to the grid is something that happens outside of technology as well, and has been going for a while.

And it's not like a totally new concept, basically.

Rachel Tipton: I think I have two things that I'd like to add to that. One is the idea that like, so there's certain technologies that will inform a user when renewable energy is on the grid. And so some people are using the grid in those moments when there is renewable energy, and I think that there is the potential for individuals to create change, not by just having flexible assets at home, but by saying, I'm going to demand energy and use energy when it's renewable on the grid, and then the grid operators and the energy ecosystem itself, I would hope would then respond by providing more renewables on the grid. I'm talking in vagaries, I feel like in generalities, but that would be one way to shift the, the type of electricity mix, I guess, as you call it on the grid. But then also from the other side that people, I think it's really important that people are aware that our electricity infrastructure is going to have to handle a lot more electricity, whether that's going to be like server centers, like computers in different places, but also like EVs, everything, if we're going to electrify everything, we're actually going to be using much more electrical energy in the future because it's going to be replacing those

Chris Adams: Yeah. Fossil fuel energy. Yeah. heat pumps instead of burning in a furnace, for example. Yeah.

This is one thing that I kind of like, just, I realized recently, actually, is now that lots of places which are taking on, for example, in Germany, right. We see a load of heat pumps being rolled out, especially, I believe you're seeing this in France and Belgium.

And these are places where we don't normally have aircon, right? And if you're listening in a place which uses aircon, the idea that like we've never really used aircon is kind of mind blowing. And now there's going to be a bunch of us with heaters, which can also cool things down. And like, like you mentioned, there will be a whole load of new energy being used because this is the only way that you can provide heat or cooling without actually having to burn fossil fuels.

So we are going to seeing a bunch of this. All right. We've totally gone down a bit of

Rachel Tipton: Yeah. Gone down a deep rabbit hole that we don't have time to elaborate,

Chris Adams: So I'm going to park that, but I'll share a couple of links. So we're just coming up to time and I just wanted to provide a bit of space to let you talk about like what, what's catching your eye that you would like to draw people's attention to.

So normally we talk about open source projects or maybe a magazine or a video or anything like that. Yeah. If there's anything that you would point people to who might have found this conversation interesting, what would you like direct people's attention to? Because yeah, now's your time, basically.

Rachel Tipton: So off the back of this particular conversation, there's a book called Real Decarbonization by Tisha Schuller, and it talks about how the energy transition is going to involve oil companies, Shell, BP, these large entities. And I feel like if people are involved in the energy transition, it's a really useful book to read, even though you might not agree with everything that she's saying and everything that she's talking about. One of the, the projects that has caught my attention lately, that's more, it's my artistic side. It came from an article in the MIT technology review called recapturing the whimsy of the early internet. And it's a movement called the HTML energy movement. I don't know if you've talked about it before on...

Chris Adams: Never heard of this. Wow.

Rachel Tipton: Yeah. So it's people who are making these HTML energy sites and they say, basically, they're celebrating the rudimentary character of HTML as a language, which I know a lot of people don't necessarily consider HTML to be a programming language, but they are pitching themselves as hidden gardens on the internet.

So I've listed a couple of links to those websites that are very, very relevant. Very minimalist in some sense, but sort of artistic places to poke around and see what is being offered. So there's one called HTML garden and then terrarium of many sceneries from the MIT technology review, which I find to be a good source of information about tech that I wouldn't necessarily be exposed to at my workplace.

Chris Adams: Okay, cool. So that was htmls.garden and sceneries.site. These are two examples of this kind of HTML garden or HTML energy sites. I'd never heard of that at all. Wow. Cool.

All right then.

Rachel Tipton: Might be a bit off topic, but they're kind of fun to explore.

Chris Adams: All right. Thank you for that. Okay. So if people have enjoyed this conversation and they want to follow on some of the work that you have, where should people look? I mean, Rachel L'Abri Tipton is a relatively strong name for Google, right? It's not like Chris Adams or, you know, John Smith, for example. So should we just be looking for Rachel L'Abri Tipton on LinkedIn or GitHub or where, where would you direct people to?

Rachel Tipton: Yeah, so I have my LinkedIn is Rachel L'Abri Tipton, GitHub is also Rachel L'Abri Tipton, and then I have a sub stack that I write, which is the Tech Generalist. So you can also find me there and I've provided just my Linktree for the podcast notes. So all of those links are in the Linktree.

Link tree Yeah. And then also wanted to just pitch a small pitch for myself. My contract with Open Climate Fix is ending at the end of April. So I'm open to new opportunities currently, whether that's in the energy space or other places, or just collaborating on projects. So if people want to reach out, I'm more than happy to talk full stack developer opportunities.

Chris Adams: All right. Brilliant. Well, Rachel, thank you very much for your time and explaining some of the inner workings of how. Basically grids work at a national level and yeah, I hope your basement gets sorted out with the flooding and...

Rachel Tipton: I hope so as well.

Chris Adams: Yeah. And have a lovely week. All right. This has been fun, Rachel.

Thank you.

Rachel Tipton: Yeah. Really fun. Thanks, Chris.

Chris Adams: Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

AI Legislation

2 mars 2024 | 49 min

This week, Chris Adams is joined by Asim Hussain and Anne Currie for an engaging chat on the present and future of legislation around AI and Green Software. Our guests share their hot takes on various topics such as the usefulness of proxies in measurement. With backgrounds in climate and tech, and a future full of green energy, they share their insights into what we might expect, and hope for, from the future.

Learn more about our people:
Chris Adams: LinkedIn / GitHub / Website

Anne Currie: LinkedIn / Website
Asim Hussain: LinkedIn / Twitter

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Sustainability, a surprisingly successful KPI: GreenOps survey results [05:36]
Cost as a Proxy for Carbon – the inconvenient truth | Kainos [13:41]
Technology Carbon Standard [25:11]
Executive summary – Electricity 2024 – Analysis - IEA [32:15]

Events:

Carbon Hack 24: Where measurement meets innovation, and impact knows no bounds [44:51]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Anne Currie: AI is eating the world, and it's incredibly requiring of energy. But I'm also going to come back to something that all three of us have said previously, which is, "does it matter if AI uses huge quantities of energy but it's all green?" Then it doesn't matter. And it's not, and that energy isn't necessarily being used for other things, because nobody wants it to heat their home in wherever it is, the middle of nowhere, where we've put our data centers.

I'm your host, Chris Adams.

Hello, and welcome to another episode of the Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. This podcast, Environment Variables, was initially set up as an example of what you might call pun-driven development between myself and Asim Hussain, back when we were both organizers at an online community called ClimateAction.tech. And, to be honest I kind of missed our chats, so I got in touch with him and asked if he'd like coming back on the podcast and talking about what he's been up to at the Green Software Foundation and I was very happy when he said yes. And likewise, Anne, our other guest today, she's been a fixture in green software for a while and has actually been working in the community working group as well as working on a book about green software called, unsurprisingly, Building Green Software.

It's nearly out, so I figured it was nice to kind of chat to her as well. So, that's what we're looking at today. Anne, Asim, lovely to have you on the podcast again. For people who've never listened to this podcast, can I give you the floor to talk about it? I'll go with you first, Anne, because it's alphabetical, and then I'll hand over to Asim.

So, yeah, Anne, the floor is yours.

Anne Currie: Hi, Chris. Thanks for that. My name is Anne Currie. I am one of the community co-chairs of the Green Software Foundation. I'm also the CEO of a training and consultancy company in green software called Strategically Green. And as Chris said, things that I'm doing at the moment, I've been writing the last year, I've been writing with, with two co-authors, also from the Green Software Foundation, Sara Bergman and Sarah Hsu on the O'Reilly book, Building Green Software, which will be out next month. So that's all very exciting.

Exciting. It's very exciting.

Chris Adams: Okay. Thank you, Anne, lovely to have you back. And Asim, I'll hand over to you as well, actually.

Asim Hussain: Thank you Chris. So my name is Asim Hussain. I am now a full time, the executive director of the Green Software Foundation. I'm really excited to be here again. What are the things I've been working on recently? Just keeping the Green Software Foundation running, has been basically my main activity. But the other thing I've been really heavily involved with is one of our projects which is called Impact Framework, which I'll probably mention a couple of times because it is completely on my mind 24/7.

Sometimes I'm asleep and I'm dreaming of it. If I'm getting my kids to sleep, I usually have the GitHub pull requests open and I'm reviewing them. So it's on my mind constantly. So yeah, that's me.

Chris Adams: Thank you, Asim. Last time you were on this podcast, we were talking about the wonders of fungi and the, the carbon fixing powers of mycelium. How are the mushrooms doing?

Asim Hussain: My mushrooms don't, I've not been doing too well. I seem to have lost my mushroom green thumb. My last attempt kind of really failed in a very moldy way, but I have some new ones growing them from scratch again. So I'm going to give that another go.

Chris Adams: Well? Okay, I guess. Do you say green thumb? I wish you green thumbs or brown thumbs, or what's the color correct color? For? For, for fungi? Yeah.

Asim Hussain: Well, you probably, you just like to have that, that mushroomy smell. That's the way you test it.

Chris Adams: I wish you lovely, mushroomy smells in your future then, Asim. Alright, okay. Hi folks, and if you're wondering who I am, as I mentioned before, my name is Chris Adams, I am the executive director of the Green Web Foundation, which is a non-profit based in the Netherlands, focused around meeting a fossil free internet.

By 2030, I'm also one of the co-chairs of the policy working group inside the Green Software Foundation and a regular host on this podcast. So if you're new to this podcast, what we do in this kind of format is The Week in Green Software, we look at some of the stories that we thought were interesting that we might want to share with other people to basically give an idea of what's happening in the field and in many cases, get lukewarm or hot takes from our guests on the day.

And I think that's pretty much it. We will share a link to every single story. And like we mentioned before, the transcript and the show notes are both available at podcast.greensoftware.foundation. But we also will link to a GitHub repo. So if there's a typo, or if there's a mistake, or there's something you'd like to link to, then you can also open a PR.

And we do accept PRs, so that we've got a useful record from other people who are trying to learn more about green software as well. All right, Anne, Asim, I'm assuming you're sitting comfortably, right?

Asim Hussain: Or somewhat, I bought a meditation chair recently. I'm sitting cross legged at the moment.

Chris Adams: Asim, I'm going to hand over to you for the first story because this is actually one thing that takes me back to some of our Climate Action Tech Days, the very, very early 2020s. So this is a story, Sustainability, a surprisingly successful KPI, this is the Green Ops survey results, we'll share the link to it, but the short version is, and the headline is basically, "if you want to save money in enterprise, turns out that sustainability as a KPI is more important than cost." This is one of the key findings of the recent GreenOps survey that was actually part funded by the Climate Action Tech Mini, Mini-grant fund that was running through this.

I find this quite interesting because it's, well, this is the first time I've seen a survey, which is actually scaled out like this. And I suspect there's gonna be some takes coming from this. So, Asim, maybe if I hand over to you first and then Anne, I'll let you come in and then we'll see what goes on from there.

Yeah,

Asim Hussain: Well, yeah, and this is really interesting. And I think it reminded me as soon as I read it, it reminded me of, do you remember when we ran the State of Green Software? We released it last year. We did a big survey with it as well.

One of the questions there, I, I, I still come back to it because I think it's very interesting. We ask people to rank in order of importance, cost, performance, sustainability, reliability, and security. So you actually, you couldn't just say, you have to actually literally drag and drop these things in the order of importance that you thought it would be. And I thought it was really interesting because the number one item that came out was security, which is interesting, but also very valid and it is, I suppose, you know, what a lot of people and organizations do you focus on. Second was reliability, which is, you know, understandable as well. Third was sustainability. And then came performance and the last, the very last item was cost, which was fascinating to me because to a lot of people, they would assume that cost is the most important thing. And there's something, I remember the researcher at the time, she talked about this concept. I think it was called onstage and offstage when you give information. Like if you're like, if you're on stage and someone asks you a question, you're going to say, "Oh yeah, cost! Yeah, cost is the most important thing to, yeah, we'd love, I love saving costs.

I love also, you know, getting promoted at the end of this quarter," but like off stage, and I think this is interesting, I think off stage, when you really ask people, when they know they're anonymous, when they know they can be truthful regarding what it is, I think this speaks to the fact that to a lot of people, sustainability is important and they do think it's more important than cost. And I think, I don't know how we can kind of get this circulated in such a way, and I suppose this article is doing a good step in that, like, how do we get this circulated in such a way that people understand, like, you know, this is important.

Anne Currie: One of the interesting things here is that all of these things are very aligned, aren't they? That being sustainable does cut your costs and it does improve your security because an awful lot at first stage of being sustainable is to turn off machines that you don't really need anymore. And those machines are your security holes. So, you know, there's a massive alignment between security and sustainability and resilience and sustainability as well. You know, the resilience of auto scaling rather than the resilience of hot backups or cold backups, it's more resilient and it's more efficient. So it's kind of like all of these things are highly aligned, but I guess it all comes down to that whole thing about extrinsic versus intrinsic motivations, isn't it? That people like to feel that they're doing good at the same time as they're doing other things.

Chris Adams: Well, there's actually, there's also a bunch of research that, I think the correct kind of term in psychology kind of jargon is this value perception gap, where everyone basically thinks that they're, they might care about sustainability, or they might care about fellow, they don't think everyone else does, so therefore they're going to self censor and say they don't actually care about this so much, or they'll talk about the thing that, as Asim mentions, makes them more likely to get promoted, for example, or something when they're on the stage like this.

The thing that we've done, Asim, I'm really glad you mentioned the State of Green Software report. I've shared a link to that and also the survey data that we had there. This story in particular, this GreenOps one in particular, the actual underlying data is available for anyone who wants to do any kind of research themselves.

And I think there's a couple of really interesting other things that came out of this, is that there was this idea about, if you look into the story, it talks a little bit about where people need to feel support for example like from the c-suite what are the kind of pre-determining things in order to you to provide some of the kind of cover for this there's a bunch of other stuff inside this that's worth a look actually so yeah I thank you both for like your actually quite warm takes today about this. This is quite cool.

Asim Hussain: But I think it's all about motivation and that's really what this is all about. It's not like what is actually the right thing for you to focus on in software. It's more a question of what am I motivated to work on? Like one of the things in the early days of this whole space is like, well, cost is important, but people aren't waking up in the morning going, "Oh, Today, I want to reduce, you know, annual turnover expectation by 0.02%. Like, and that's, what's motivating me today." It's important to whoever needs to do it. But what motivates people is actually, "I want to do something." This is important to me that, that actually reduces some of the emissions of my software. And I'm going to work a lot harder on that than I am on reducing costs by 0.

2%.

Chris Adams: Yeah I guess maybe some people are, I don't know, I mean, maybe we'll come to the day where someone says "do you know what? I just really like returning value to shareholders." Like, that is what, maybe that's a thing. I mean, they, presumably they... but, yeah.

Anne Currie: Well, I'm going to come in with hot take number two, actually, because I had a very interesting story this week, which is very related to this and that security being the number one motivator. If people actually acted on good security practice, it would massively cut carbon emissions. The going through and closing off all the machines that you really know you need to, if you did a proper security review and you close off all your old machines that you weren't using anymore, that would be a really good security move, would also massively be more sustainable.

But not only that, I heard a very interesting tale this week from a company that wishes to remain anonymous. They did a security review a couple of months back, which, which was about identifying and turning off effectively, well, not zombie services, but services they knew they didn't need anymore. And as part of doing that, the security team identified a machine that was running at 100 percent CPU. So they thought, "well, hang on a minute, that's been hacked. You know, someone's binding Bitcoin on this." They handed it back to the ops team and said, "what's going on here? Sort it out." The ops team took a look at it. It turned out it was running a load test that had been, that, that had been turned on three years previously and never, never turned off.

So the security review identified the problem as fundamentally a cost problem, but, but more than anything else a sustainability problem. So if you do security reviews, it's actually one of the best things you can do for being greener, because you, you just stop doing stuff you shouldn't be doing.

Chris Adams: And I think I'm going to need to share a link in the show notes to the meme saying, You cannot have a cluster if, if you don't have a cluster, because that seems to be a kind of reference to some of this here. Okay, thank you very much for that, Anne. Just for people who are curious, as a bit of context, this survey was actually carried out by a gentleman called Mark Butcher, I believe, who works at a firm called Posetiv.

The actual process of why this was funded was basically by donations from members inside a community who just put some funding in. And there's a couple of other mindy grants that are ongoing now. There's one to create a sustainable AI manifesto, and there's another one which is a playbook for becoming a climate-conscious product manager.

So it's kind of nice to see some of this stuff. I'm looking forward to more things like this being created because you don't need that much money to get some really interesting results and things coming out there into the kind of, I guess, the public discourse. All right, cool. Anne, Asim, are you comfortable with us moving to the next story?

Asim Hussain: Yeah.

Anne Currie: I am.

Chris Adams: Alright, okay, so the next story here is a blog post from Kainos.com. I believe they might be members of the Green Software Foundation, I forget, but this one is basically, the headline is Cost as a Proxy for Carbon, the Inconvenient Truth. And I'll just share a kind of quote from the paragraph, from the piece.

It says, "Many technology related articles and major cloud service providers promote the idea that cost is a proxy for carbon emissions. This suggests that optimizing and reducing your cloud costs can lead to increased sustainability. But is this entirely true? While there is some truth to this statement, it is critical that limitations to this approach are well understood."

So I guess now I should, while I've got access to two experts, I'm going to kind of put it to you two. Anne, when is cost a good proxy for carbon? And when is cost a bad proxy for carbon? Because I've understood that you might have some opinions on this as well.

Anne Currie: I have very strong opinions on this, Chris. I think that at some stages, it's an excellent proxy for carbon. And at some stages, it's gone beyond. It's not really a great proxy anymore. So it depends is the answer. So I'm going to, I'm going to pull in one of our new projects at the Green Software Foundation that I'm the lead on, which is the Green Software Maturity Matrix, which is about the fact that there are kind of, we're all at a different stage.

Most of us are at stage one. We haven't really started down the road of becoming green in our, in our systems. And at that stage, cost is actually a pretty good proxy. If you can halve the number of machines you have, or you can turn off stuff that you're not using, that's a really good way of cutting carbon. But as you get towards the end, you become quite sophisticated and you're doing really clever stuff with tuning and demand shifting and demand shaping. It's not quite such a good, a good measure anymore, but you know, just use your, we all went into tech because it's clever and it's interesting and it's complicated and it's... just use your brain.

Just everything that you save money on is not, is clearly not going to save you carbon and everywhere that saves carbon is not always going to save you money. But as a rule of thumb, if you're turning machines off, that's good.

Chris Adams: Okay. All right. So basically, if you have fewer machines, you have fewer emissions. That's the kind of broad strokes approach that you seem to be out talking about. Now, Asim, I see that you've added a note here that you've seen examples of resizing, reducing emissions, but increasing costs. So maybe you could talk a little bit about that.

And then I'll maybe come in with some takes at my end as well.

Asim Hussain: I've got, yeah, I've got some other hot takes on that as well, actually, but yeah, I have seen, I have not seen many. I have not seen many. So I think I just wanted to point out that, but I don't think that's necessarily, and the absence of information, or what is that statement?

Anne Currie: The absence of information, evidence of absence is not,

Chris Adams: Absence of evidence is not evidence of absence, right? Yeah, yeah.

Asim Hussain: There's a certain absence in our podcast here right now. The, but that effectively, like I have seen some examples of people like when, and where they've tried to do some sort of downsizing and they've happened to have just enough measurement infrastructure in place, which is usually the problem, to actually like determine, Oh, actually, Oh, wow.

That actually increased the cost. Cause you know, sometimes you can right size, but the machine you move to actually costs more and overall just ends on a bit more money, but the actual emissions is less. So it's rare, but I think that's more a statement of we're just not measuring this. But I think more importantly, I think cost has two problems as a proxy for carbon emissions. Number one, and the same, this is the same problem you have with carbon emissions, it's the same issue. A, it's a lagging indicator. Like if you're developing a piece of software, you only know the cost of it after you've rolled it out and you've done it and you've moved on for a month onto the next project, and then somebody goes, "by the way, that's like slightly more expensive than we were expecting."

Like, "Oh, I suppose, but do you want me to stop working on this other project now?" And I'm like, "no, no, no, you carry on that project." It's a lagging indicator. It's like after everything's happened. And I suppose that also means it also has all the other problems that you have with using carbon as a proxy for driving action to reduce carbon emissions, like If you like the, the issue with looking at cost is like someone saying, "well, I'm spending like 3 million on this database and 4 million on this other thing." And you're like, "well, I don't know. I'm, I developed an application. What is my costs?" And so like you have this problem of really it's the problem is more how do you actually get information to the people at the coalface so that they can make decisions to actually reduce the emissions directly? And I think that's the challenge with cost is that you can't, it has all the same problems as carbon in trying to get information to people to actually, so they can know what to do. Because you can tell a department what their cost is, you can tell a department what their carbon is, but can you tell developer X on team Y the impact of a choice they're going to make? Is, I think that's really where we want to get to.

Anne Currie: And I think we're saying exactly the same thing there in a slightly different form. You're saying that it's not a good measure when it comes to tuning what developers do, and I totally agree with that.

I'm saying that generally for where everybody is at the moment, a lot of the good tuning comes at, comes from ops teams rather than development teams. And for ops teams, cost is not so much of a leading indicator. It's actually quite close to what they're doing day to day. So it's a really good metric if your ops team are the one who are doing the tuning to reduce carbon. Not perfect. As you say, it's a terrible, you're better off with performance being a metric if you're doing the development. Yeah.

Asim Hussain: No, you're right. We're talking about different audiences. Yeah. So yeah, to the, to an ops team who are, whose job is to deal with the software after it's deployed, yeah. I suppose it's not a lagging indicator. It is their main indicator.

Anne Currie: But I think that we're in agreement that it's a great proxy for ops teams, but not a great proxy for dev teams.

But it's, but as you say, also, it's not even always a great proxy for ops teams, because often the best thing you can do in terms of operations is to move the whole, your whole shebang to somewhere with a greener grid, but that might well not save you any money.

Asim Hussain: True.

Chris Adams: Google have some interesting stuff inside this and that. They've got some region pickers which do expose some of this. You can prioritize late cost and carbon. And like, if you cared about carbon, you might move things to Switzerland, for example, compared to running it in, say, other parts of Europe or, say, even America, for example.

So there's stuff like that, that I think you can actually do. But I think when I read through this piece, there was a couple of things that kind of outline on this, like, yes, this can be a bit more complicated, because this kind of assumes, like, if you have this cost as a proxy thing, in many cases, you might have like a commitment to maybe a 50 or an 80 percent discount on your cloud bill, if you've agreed to spend a bunch of cash now.

So now you've got totally different incentives to be doing what you're doing. And like, when we start talking about different instruments that people use to essentially affect the cost of cloud builds, it can be a bit harder to talk about. And there's also this other thing that, let's say you're looking at something like, we don't know which services have particularly high margins versus low margins, which ones lose money versus making money.

Broadly speaking, in 2024, you might assume that for the same amount of work, doing something with, say, a serverless platform like AWS Lambda might be more expensive for the same amount of requests as a dedicated box, for example. So, if you looked at that, that might assume that it's that much more carbon associated.

There's questions there, and like, as Anne mentioned, when you start looking at the edges, you can definitely find all these different edge cases which make it harder to really kind of use. But as a rubric, as a kind of starting point, it may be better than having nothing, basically. And it's not like technology is immune to us looking at easy to measure numbers and assuming that's automatically the truth.

Asim Hussain: Now that is definitely one of the things I would say. I kind of came hot on cost then for a second. But like, what I would say is that in the absence of anything else, which is by large the problem that we have right now, it is one of the best measures that we have. But like, I think we need to move the world forward to better measures eventually. But it is the best that we have right now. If you don't have anything else.

Chris Adams: If we had actual energy

Asim Hussain: If you had actual, yeah,

Chris Adams: right? So quickly, it's a commodity.

Anne Currie: It wouldn't. Because even if you've got energy, you know, I would say this, it's still very hard to do a like for like comparison. It's not like performance. Performance is a really good measurement for performance. You know, it's like seconds. How many seconds did it take to do that thing?

I mean, you can put it on different sized machines, different speeds. There are even issues there. But as we all know, you know, sometimes the grid is green and sometimes the grid is not green. So even carbon measurements, you could be running exactly the same thing. Everything can be under your control, but unless you're doing demand shifting and shaping or work shifting and shaping, as now seems to be the more common and correctly, then the name seems to be shifting from demand shifting and shaping to work shifting and shaping, which

is a much better word for it.

Chris Adams: New to me as well, actually. Okay.

Anne Currie: It was, cause I used to complain all the time because

demand is

Asim Hussain: Did you start changing? Did you start changing the words, Anne?

Anne Currie: I, it wasn't. I, in the book, it's all demand shifting and shaping. And I do say it's a really terrible description because the demand hasn't changed. As far as we're concerned, the demand comes from users and the users still have demand and we're attempting to meet the demand that they have. We're not intending to shift that. But the name, I think the words demand shifting and demand shaping came from the grid.

And as far as the grid is concerned, we are the users, not the end users. So what they're talking about is our, as the tech industry, our demand is shifting and shaping, which it is. So there's some confusion about who's the end user in this scenario.

So I have seen some people start to use work shifting and I wish I'd known that they were going to because I've just written a book in which I didn't use those terms.

Chris Adams: Ah, well you heard it here first. Alright, what I'm going to do is I'm going to add a couple of show notes both to basically your book, Anne, and a couple of blog posts. I've written one post recently, a while ago, called Options to make software greener without changing the code and how to remember them, which essentially I'm taking some of the papers I saw at Hot Carbon and trying to translate them into kind of layperson's terms.

So that might be a bit easier, because I think that I agree with you. Demand shaping and demand shifting. These are ideas that have come from the electricity sector that aren't as intuitive to developers, actually. And maybe, I think we've probably got enough there because we've got a few more stories to run through.

Are you comfortable if we move to the next one, folks?

Asim Hussain: I am yeah.

Chris Adams: Okay. All right, so this one is introducing the tech carbon standard. So if you go to www.techcarbonstandard.org, this is a website for, not a standard per se, but a proposed standard from an organization called Scott Logic who are based in the north of England.

This was actually revealed at the state of open conference in February, and the TLDR is the Tech Carbon standard is developed as an approach to classifying an organization as technology estate to enable consistent analysis and benchmarking of its technology carbon footprint. The standard takes inspiration from the globally recognized GHG protocol and its emissions classification in scopes 1, 2, and 3.

And although you can't see this in the audio format, we've shared a link to it. It's basically a diagram, more or less highlighting where you might think emissions take place. Both in the view of the GHG protocol, which is like pretty much the de facto standard for reporting carbon, but also mapped to the kind of bits of software that you might have.

So there's a notion of upstream emissions, where you might be kind of purchasing cloud software, for example, there's operational emissions, which might be talking about you running servers on premise, or something like that. And there's downstream emission, things like end user, and like network transfer, and stuff like that.

So, this is, this is something that is kind of new, and I, I found quite interesting, because Asim, I know that there are very deliberate decisions behind the SCI, the Software Carbon Intensity spec, which has been put forward by the GSF to not try to tack too closely to the, to the GHG protocol.

And it might be worth just to see what you folks think of this, because I think this is quite interesting. And the idea of it being entirely open for other people to use and riff off, I think is quite helpful actually. So maybe I'll hand to, I'll hand to Asim first and then to Anne actually, because Asim has some very, he's got some things he definitely wants to say about scoped emissions.

Asim Hussain: I'm coming in hot.

Chris Adams: Yeah.

Asim Hussain: No, I'm not coming in that hot. No, I think it's, I think it's good. I think, look, the GHG protocol was built from the perspective of an, an organization that does manufacturing that, that is what it has been based on. That's why if you're a manufacturing organization, it makes complete sense. Your scope one, your scope two, and your scope one is, I don't know, should we summarize it for people? That the scope one is kind of, if you've got. If you've got an oil drum in the back garden and you're just burning oil. And I don't know many manufacturers that do that, but that would be your scope one emissions. Scope two is the emissions from the electricity that you purchased. And scope three is from all the things that you bought to build the thing, the physical thing that you sell. So it makes a lot of sense from that perspective, but for our world, it's really problematic and especially scope two. Like if you say the word scope two, it's almost meaningless in our world.

Because if you, if I say scope two as a proxy for electricity, well, it's not because if you've got an on premise data center and if you're running your software partly on your own servers and partly on the cloud servers, and you said scope 2, your energy usage, your scope, this is, I can't even explain it, it's so complicated, like your scope 2 is the energy, but only your application can choose, but only on the hardware that you own. The energy that your application uses on the cloud providers falls under Scope 3, Category 11. And so, like, it's really hard to kind of figure out, well, what is the energy? It's like, you have to kind of have this thing. And really, when you say those words, it's so contextual, when you say scope two, it's not what it is to you. It's what it is to the audience that's sitting there and it means different things to different people. So what I do like about this is the attempt that they've got to kind of really look at them and you know, they've got upstream emissions, operational emissions, downstream emissions. When you say operational emissions, it will mean the same thing to everybody you're speaking to, and it's not context specific. So I think it's good from that perspective. It clarifies some of the stuff. So I think there's some good stuff here. I do. Yeah. I think it's some good thinking. I think my advice is to get into detail, when you click on servers and storage, the only description is the energy consumed by on premise servers and data centers. And I suppose we all know there's a lot of, there's a lot of meat behind that one that you've got to, you've got to flesh out. But yeah, I think my rant on scopes is over. So I'll pass it on to other people.

Chris Adams: Thank you for sharing that particular point of view, Asim. Anne, I suspect you might have something to share as well here. So yeah, what did you think when you saw this?

Anne Currie: Well, I always think all work in this area is great. All discussion we're doing around this is great. Fantastically great. But I always think about taking it back to my Maturity Matrix projects on the GSF Maturity Matrix project. You need to do different things at different times. So this information is useful to you in different things at different times of your journey.

So for most of us at the moment, we're not doing anything at all. Just, you know, you really don't need that much data. You know, your scope one, scope two, scope three are not immediately critical to most people who just need to turn off the machines they're not using anymore, right. Do a bit of right sizing.

Have a think about what they're up to. Start to think about, are they in green regions? What are the future? What platforms are they choosing? Are they likely to be aligned with the green future? To start with, where most of us are at the moment, you don't need a great deal of data to really make a huge difference.

So it's absolutely fantastic that they're doing this work. And that we're doing this work. Everybody's doing this work. We don't want to lose sight of the fact that for almost everybody at the moment, we don't even need data. We just need to start to, to use best operational practices.

Chris Adams: All right, so you don't need, you don't need to wait until you have data. You're, or don't, sorry, don't need to have all the data, is what you're saying, because I don't think you're saying you don't need to have data to work on this. It's more like you're not, okay, all right, I think I know more or less where you're going with this.

Okay, one thing that I'll share with this that I thought was quite interesting was, this is the first time I've seen someone really outline scope one, actually, like burning fuel for generators and things like that, because this can add up, and there's also steps that you can take to basically, reduce the environmental impact of running generators themselves, actually.

So, you know how most of this stuff runs on diesel, basically, right? So, diesel is what you'll be burning in a data center, most of the time, if you're not running on electricity. And, this will be tested quite a few times a year, for example. There's a thing called, I think, Hydrogen Treated... oh God, what is it again?

There's a specific treatment for diesel, which basically is a way to remove all the kind of bits of dead, long dead animals and plants from it, which essentially make it less harmful for people nearby to be breathing it. And that can actually come from biological sources like, or biogenic, as in things which are non-fossil, for example, there are non fossil sources of HVO diesel that you can have, which is one way that you can reduce some of the footprints on this. And although it's only a few hours in a given year, maybe 100 hours in the 9 000 hours that you do have, that's very dirty stuff and it has a very kind of localized impact on people.

So it's kind of nice to actually see that called out for the first time. So I found this quite helpful and it was when I first came across it and I'm glad to see it open for a bit of work on. Alright, shall we move to the next story, or Asim, you've got something you want to share?

Asim Hussain: Just a hot take.

Isn't it funny that like these buildings, which contain some of the most advanced technology that human beings have ever created in the history of our existence, maybe, that are completely chock full, that are huge, just in case they go down, we burn dead animals to, to power them.

Like we, if, if we can do that with the technology inside the buildings, we can do something with the technology that powers them.

Chris Adams: I think, okay, alright, inspiring.

Asim Hussain: heard any of that. I don't

Chris Adams: we've got one more story, and then we'll wrap up a bit. So, we're going to link to the IEA's recent report. So the IEA is the International Energy Agency, and they made some headlines earlier on this year with their link to the electricity report.

Basically, they talk about what's happening over the next 10 years with electricity. And one of the headlines was they were saying that electricity consumption from data centers in terms of like AI and the cryptocurrency sector could double by 2026. So I'll just use the quote that I pulled from it, which was "electricity consumption from data centers, artificial intelligence, and the cryptocurrency sector could double by 2026.

Data centers are significant drivers of growth in electricity demand in many regions. After globally consuming an estimated 460 terawatt hours, in 2022, datacenters total electricity consumption could reach more than a thousand terawatt hours in 2026." So that's a doubling in essentially, what, two years?

This is particularly interesting because this is way higher than their previous figures that they used to have which were around 200. So Something's changed with these numbers for them to be that much higher. And I figured I'd open the floor to either of you about this because this has got a lot of pushback from the data center industry.

But there's also people who work in the energy sector saying maybe there's something to it. So Asim, maybe I'll ask you what's your view on this one?

Asim Hussain: Well $7 trillion worth of chips cost a lot of energy to power. So, you know, like I think AI is a lot bigger than we think. And I'm not saying that in a way that is in any way negative. I think it's just, we have to accept this is the future that we're living in. This is the current present that we're living in. I strongly suspect that some of these numbers are taking that into account. The growth in AI has been significant. We all know on this call that the previous dirty secret of data centers where most of those servers were idle. In a future of AI, those chips are not going to be idle. They're going to be running at a hundred percent.

So like, I think we've spoken on this call previously about, you know, various previous reports that talked about, you know, given the current trajectory by 2040, the tech sector will be like 14 percent of global emissions. I wonder if anybody's doing any analysis to revisit, well, now, given what we now know about the complete, AI will take over everything. It is taking over everything right now. How does that look now? Where will we be in 2040 with the current growth in AI? Will tech be half of all emissions? And will we just be sitting there, you know, being carried around by robots and being fed by little tubes like that robot show? But I think, I strongly suspect that they factor that in. And, and I wonder if it's an underestimate.

Chris Adams: Okay. All right. So that's what you got from executive director of the GSF, but that's not his official opinion. Yeah, yeah. Anne, what about you? Where are you on this one here?

Anne Currie: Well, I think Asim's right that, and those numbers are probably right, that AI is eating the world, and it's incredibly requiring of energy, but I'm also going to come back to something that Asim and I have both said previously, all three of us have all said previously, which is "Does it matter?" Because AI is a really good example of a workload that works well with the fact that some places are much, have much more potential to generate green power than others. It's a latency insensitive workload that could be running in Greenland, not that Greenland has a massive data center industry at the moment, but it could do. It could be running further away. But, but also I think it feeds into something that I was reading that almost comes full circle with the discussion we were having earlier about data centers and how they're used and whether or not they're used well.

I mean, the ideal at the moment is we'd start building data centers as places where there was water for cooling and green energy for powering the data centers. And then you have to worry about the embodied carbon, but you don't have to worry about the, you know, if AI uses huge quantities of energy, but it's all green, then it doesn't matter.

And it's not, and that energy isn't necessarily being used for other things because nobody wants it to heat their home in wherever it is, the middle of nowhere, where we've put our data centers. But we need to have a think about data centers more generally. Something that comes up quite a lot is water use in data centers.

So at the moment we can't necessarily put data centers that could run on solar in places where it could run on solar because there's a lack of water, but there was an interesting, I read a very interesting article about that and I shall, I'll send you the link to that about PUE driving companies to overwhelmingly cool data centers with water, because that doesn't count against the PUE score. But in fact, if you're in a place where there's tons and tons of electricity, green electricity, you could be cooling your data centers by using aircon and that would be not a hit on, on water, but it would make your score, your incredibly important PUE score,

because we're focused entirely on a metric, which is not a great metric, is forcing up water use. So, yeah, it takes us full back to the full circle of metrics are useful, but do not overly lean on them, or it can be worse than if you had never had them to start with.

Chris Adams: Okay. All right. Thank you for that, Anne. This actually makes me think of a recent interview between Jigar Shah, who is the head of the US loan program office, and David Roberts, who runs the podcast Volts. It's a really, really interesting one. It's particularly talking about nuclear in the specific, but I'm going to share a link to it in the show notes because it gives, it basically has the person who's the head of the loan program office saying "we have data centers speaking to us. They're expecting this kind of load. And the figures there are extremely high." And they're actually not that far away from the numbers I see here, which was surprising for me. Because the IEA, A had much lower figures before. But there's also another document that we can link to, which has numbers which might not be exactly the same as these figures here, basically.

We'll add a few more, because this discussion about how much power is actually being consumed and how much is going to this is very much a relevant one, especially when you think about, okay, if we've got this much energy, are we going to use it to meet demands of AI or are we going to use it to decarbonize the rest of the world?

Because there is a finite amount of green energy that you can actually deploy in a given amount of time and space. And this is something that we tend not to talk about right now. And maybe this is something that we need to touch on a little bit later. And this might be a nice segue to the next story, which we'll just kind of wrap up with.

One of our last ones. This is a piece of datacenter research from the European Commission, basically. Actually, we mentioned a gentleman called George Kamiya. George Kamiya was leading this research with another gentleman called Paulo Bertoldi. This is essentially energy consumption in data centers and broadband comms networks in the EU.

And the thing that's interesting here is this is country by country information for both data centers and for networks. And it's worth a read if you want to have some numbers to talk about this. And there's a really interesting chart that we'll also add to the show notes. Anne, I'll hand over to you first because you have some things you want to share about this and then Asim, I'll hand over to you if that's okay.

Anne Currie: It's a really fascinating report with a lot of interesting data in it that we haven't seen previously. So I thought that the main takeaway from the report for me was that generally speaking, most countries in the world, that the percentage of energy use by data centers and by telecoms is fairly consistent.

It's kind of like data centers are around one and a half, two percent of electricity use within the country. The only time where that really isn't true is in places where there are lots of data. They are the data center for the world. There are a dis disproportionate number of data centers in that country like Ireland, for example, Denmark, the Netherlands, and it, it'll be interesting to see whether or not those are the places where all the data centers are in future, because I would like to see the data centers in places where there is an awful lot of green Power that is not used, Chris, as you were just saying, in any particular location, you don't want all the green power being sucked up by AI, you want it to go into other things as well, but in places where we could build data centers and there isn't a great deal of other use for the green power, and it isn't competing on a local grid, I hate the whole thing of use your data center heat to heat municipal pools, because that's immediately saying your data center is somewhere where it is competing with folk, with people, for their energy requirements. Data centers should be in the middle of nowhere, which is terrible for people who work in data centers, so I'm sorry. But, that's what the robots should be doing, running data centers.

Chris Adams: Okay. All right. Asim, I'll hand over to you before we wrap up and then come to the last story. Go for it, mate.

Asim Hussain: Well, I don't, as we were speaking, I just realized that I, what, why? What is it about? I know, I've known for a long time that Ireland has, I've known this statistic for a long time, but I don't know why. What is it about Ireland? Is it a tax haven type?

Anne Currie: It's a tax dodge, I think.

Chris Adams: So you're talking about context. So for context, you cannot see the chart, but we're looking at a chart with country by country figures, which shows the energy used in terawatt hours by data centers, the energy used by the telecom sector, and then the share of national energy use. And for most countries, it's between 3 and 5 percent or maybe 1. 5 percent and 4 percent in most cases, except in Ireland where it's hovering around 19 percent or 20%. So you got close to 20 percent of all the power used in Ireland to run data centers, which are predominantly a set of hyperscale data centers. And if you read through the report, you'll see that figure was around 10 percent in 2019.

So in the last few years, you've seen a doubling in usage here. And the other kind of outliers you might compare to are Denmark, which is around 5 percent of total energy usage and the Netherlands, which is a little bit over 5%, but in particular it is very exceptional in that it's multiples larger than everyone else right now.

That's from what we can see so far, one of the largest sources of Load Growth inside the sector, inside there. And there's also, what we'll do is share a link to a actual consultation from the CIO, the Irish Grid Operators, where there's some really eye opening responses from both academics and from data center operators, basically saying "we should be doing this, and we shouldn't be doing this," and some of the charts are pretty eye opening. Yeah, that's, you're right, as Anne mentioned, Anne, you spoke about was Ireland has been considered one of the places with very low tax, and as a result, large corporates love it, basically.

Asim Hussain: Is that what it is? If it's a low tax? I mean, cause then. The solution is to reduce the tax in areas where it has greener energy or something like that? I think there's

Chris Adams: Do you think that's the case?

Asim Hussain: Well I don't know and if it is, I mean we have to look at this whole situation

Chris Adams: They're just the prescription of reducing tax in the one place that has very, very low tax compared to maybe increasing the tax there so that you have a bunch of companies that need to choose based on carbon. I think that there's different prescriptions you might use to solve this particular problem if you see that as one of the problems that we have here.

Asim Hussain: Well no, I think it's just acknowledging the fact that if that's the situation and it was being driven by a low tax. You're right, maybe tax is the solution if tax is the problem.

Chris Adams: If you look at Luxembourg, Luxembourg is the other place with extremely high percentage of data center usage, which is actively courting data centers, which doesn't have particularly clean energy. And that's also one of the high ones. And it's also known for not being the place with the highest tax in the world.

So there's this idea that essentially low tax increases the amount of data centers you're going to have in a particular place, especially if they're large hyperscalers. But that's not necessarily means that you're going to have green energy there all the time. All right, Anne, over to you.

Anne Currie: I was just going to say, in the long run, hopefully, the solution to high prices is high prices. The solution will be that green is much cheaper. So, hopefully, the actual cost of electricity going down will overtake the tax breaks and things that are currently distorting the market. We really want data centers to be in places where the electricity is cheap. That is the, green will be cheap. So...

Chris Adams: Okay, thanks, Anne. All right, then. Okay, last story, because we're just running a little bit over on time. The last thing we're going to link to was actually this carbon hack thing. Asim, given this is like taking over your life, I figured it might make sense to give a bit of space for you to talk about what on earth you're doing and why people might go to it.

So, why do people care about carbon hack and why are you working on this thing? Because that might, yeah, justify your existence, Asim. That's maybe the way to think about it.

Asim Hussain: Just to save me. So this year we're running CarbonHack. It's running in about three weeks actually now. And the thing that I'm really excited about it is that the main topic of running is a framework that we've been building here called Impact Framework. Which is many things, but it's a tool to help you measure and really be very transparent regarding your software emissions.

A lot of the stuff we've spoken about today is this question of, is something a good proxy, is it a good measure? What is it? Oh, and what impact framework is, it allows you to, through a set of plugins, build a bespoke calculator, which can take observations that you can make about your software system and turn them into impacts like water, energy, carbon, things like that. But the most important thing about it, and it's something I don't really talk about enough, but it is that everything is stored in this thing called a manifest file. Everything. So the future I want to see is a future where somebody's not telling me what the carbon emissions are, because what is carbon? It's somebody giving me their manifest file, and it has all the coefficients, all the models, all the inputs, every assumption they've made in that calculation is that manifest file. And if I don't like it I'll change it. I'll adjust it. I don't like your model? I'll change your model. I don't like your coefficient?

I'll change your coefficient. And I can rerun it and I can judge what your emissions are. That's the real power of it and that's what we're doing with CarbonHack. We've got a bunch of prizes. One of the things we really need for Impact Framework to grow is a lot of plugins. Everything I just described is aspirational.

We need plugins to help compute web, to compute AI, to compute all these other, all these other things, to compute water. And that's a large part of the hackathon this year, is to encourage people to build plugins for Impact Framework. So I encourage you to go to hack.greensoftware.foundation and you can find out more information on it.

And I'm also giving, if you go to that website shortly, I'm also giving a lot of training and workshops on Impact Framework over the next couple of weeks. So if you want to learn more about it, just learn how it works, there will be lots of training sessions coming your way shortly.

Chris Adams: Okay. All right. Thanks for that. So, hack.greensoftware.Foundation is what people look at if they want to take part of this distributed, online, weeks long hackathon that doesn't involve you having to be awake for three weeks. You can do it in your own time, right?

Asim Hussain: Yeah. Yeah.

Chris Adams: Okay, cool. Alright then. I think that'll take us up to the end that we have, actually.

So, Asim, Anne, it's been lovely chatting to you again. I missed you, actually, so thank you for coming on for this. Like I mentioned before, all the show notes and all the links will be on software.greensoftware.Foundation. So if you go there, that's what you can find. And there'll also be a link to the transcript of this podcast, plus the links in markdown format.

So if there's a thing you see that's incorrect, or if there's a thing that you think we missed out, we do accept pull requests and you'd be contributing to the greater knowledge or the greater good for people to explore green software themselves. Thanks again, and yeah, let's do this again sometime soon.

Asim Hussain: Yeah. Thanks, Chris. Great to see you again, Anne.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

Greening Serverless

16 februari 2024 | 47 min

Chris Adams is joined by special guest Kate Goldenring, Senior Software Engineer at Fermyon. Together, they ask the real questions “is serverless computing the greener choice?” and “if so, under what circumstances is this true?” Tune in for an illuminating conversation on the current state, news, and future of green computing, digging into the topics of cloud computing, soft allocation, WebAssembly, and more.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Kate Goldenring: LinkedIn | X

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Carbon Hack 24: Where measurement meets innovation, and impact knows no bounds

Resources:

tag-runtime/wg/iot-edge.md at main [04:24]
Bytecode Alliance [05:47]
Basic Alpine Climbing Course - Seattle - 2024 — The Mountaineers [08:09]
A greener, cost effective cloud with serverless WebAssembly [13:00]
Surprising Scalability of Multitenancy - Marc's Blog [22:34]
Introducing Spin | Fermyon [31:01]
Software Carbon Intensity (SCI) Specification Project | GSF [39:17]
Take it to the limit: peak prediction-driven resource overcommitment in datacenters [41:06]
GitHub - fermyon/spin: Spin is the open source developer tool for building and running serverless applications powered by WebAssembly. [42:36]
Fermyon Discord server [43:09]
https://hotcarbon.org/assets/2022/pdf/hotcarbon22-sharma.pdf [43:30]
Fermyon Technologies [45:56]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Kate Goldenring: It is very clear that you reduce operational emissions by not running your application when you're not using it, so by leaving the gym. But actually, by oversubscribing, we're also reducing embodied emissions by having higher density and not needing as many servers. And so, I think this whole concept can help us decrease our software carbon intensity.

I'm your host, Chris Adams.

Hello, and welcome to another episode of Environment Variables, where we bring you the latest news, And updates from the world of sustainable software development. I'm your host, Chris Adams. When we talk about reducing the environmental footprint of software, it's common to talk about making an application more efficient by using better algorithms or relying on languages that make more efficient use of resources than others.

However, the fact remains. That for most software in the world, most of the energy use comes from having millions of computers waiting for something to happen, rather than actively doing work themselves. So if you have software designed to let someone book a ticket, or buy a book, any time of day the transaction might only take a few minutes to complete, but the software still needs to be running 24/7 in a powered up state, constantly consuming resources to make that possible.

Now, common wisdom was that you need to design for peak amount of traffic you'd ever expect to see, and then just accept that paying to run all those servers all the time, waiting for something to respond was just the way it had to work. Serverless computing broke this assumption. Instead of paying per hour to have your software constantly running and available, waiting for responses, you'd design your software a specific way and pay to use it on a per request basis instead. When your software wasn't actively serving requests, it would scale all the way down to zero and you wouldn't be billed for these dormant periods. If an incoming request did come in, your software would wake up, serve them again, before going back to sleep.

And in aggregate, this would result in fewer machines needing to be run. Because computing resources were being allocated in a more efficiently across a large number of programs, running them only when they needed to do work. So that was the idea, at least. While we touch on the details later in this episode, there are definitely trade offs you need to accept when designing serverless programs.

And while the term serverless implies a lack of servers, there are still racks of kit required to run this infrastructure. So, is serverless computing the greener choice? If so, under what circumstances this is true, and what's the difference between a virtual machine, container, or even a WebAssembly runtime?

Joining me today is Kate Goldenring, whose work I came across when she answered a really specific, really detailed question I'd asked about serverless computing. In the Green Software Foundation discussion forums, when I was trying to make sense of it all, she responded with a really helpful answer, citing a number of peer reviewed papers, as well as linking to a really informative talk that she delivered at the Cloud Native Computing Foundation Sustainability Week a few days earlier.

So when I asked her if she'd be up for coming on the show to talk about some of the details of serverless computing, she agreed and I'm really happy to welcome her on to Environment Variables. Kate, welcome.

Kate Goldenring: Hi Chris, thank you for having me. So I am very happy that that forum existed and brought us together. I am a senior software engineer at Fermyon, and we're very excited about serverless WebAssembly. And so that's something that we'll be talking about throughout this and in general, to touch on my career, I started at Microsoft, where I was actually focused on edge computing solutions.

And I developed a project called Akri, which is a Cloud Native Computing Foundation, CNCF, sandbox project for managing IoT devices around Kubernetes clusters. So from the start, I was very into open source software and got to really kick off my career there, which I'm very grateful for. And I'm still in the IoT edge space.

I serve as co chair of the CNCF iOT edge working group and have authored a couple of white papers on edge computing best practices within that working group.

Chris Adams: Oh, Kate, can I just stop you just for one second? Did you say, so you were with Microsoft back then and you mentioned Akri. Now I think that was like a few years back. Was there a project called Deus that was floating around in Microsoft land there? Like they want a little bit like say, wow, that takes us 10, back 10 years or so actually.

Okay. So Deus was one of these platform as a service things, a bit like Cloud Foundry and stuff like that. Right?

Kate Goldenring: Yeah, and Deus was acquired by Microsoft and Deus Labs was how it was renamed. And it became this hub within Microsoft for its open source cloud native projects. And so Akri became a project under Deus, and then it was donated to the CNCF after that.

Chris Adams: Okay, gotcha. Thanks.

Kate Goldenring: Yeah, so a bit of an evolution in how open source worked within Microsoft as well, and I actually, while working on Akri and in that open source world within Deus, I learned about WebAssembly, and I got very excited, in part because I saw the potential of it for green computing and reducing the carbon intensity of software, and so I actually decided to leave Microsoft to join Fermyon, which is a startup in the WebAssembly space.

And I'm still in the open source world contributing to our open source tooling there. And we actually work with the Bytecode Alliance, which is an alliance, as it sounds, of companies such as Microsoft, Amazon, Adobe, Fermyon were involved. And it's all about building out WebAssembly specifications so it can do more in more places.

So that's kind of also where I am now within the weeds of the WebAssembly world.

Chris Adams: Okay, cool. All right. I'm, I'm glad you mentioned WebAssembly because we'll touch on that a little bit later about what it actually is and why some of that is relevant, but I should probably introduce myself too, for people who are new to this podcast. So my name is Chris Adams. I am the executive director of the Green Web Foundation, which is a Dutch nonprofit focused on reaching a fossil-free internet by 2030.

I'm also one of the policy chairs of the Green Software Foundation policy working groups. And I'm also one of the maintainers of a library called co2.js, which as you imagine is a library to help you work out the carbon footprint of various digital services. Alright. So, Kate, you're sitting comfortably, right?

Kate Goldenring: Of course.

Chris Adams: Okay, cool.

Before we dive into the finer points of serverless computing, I understand it's quite early. It's earlier for you than it is for me. So it's about 4 p. m. in Berlin, 7 a. m. or 8, 7 a. m. for you, right? In Seattle, yeah?

Okay. And as I understand it, when you're not in front of a computer, there are some upsides to living in Seattle.

So maybe you could tell me a little bit about that before we dive into the nerdery, because when you shared some stories here, it pleased me actually, and I quite enjoyed some of the kind of names of some of the groups you're, you're part of.

Okay,

Kate Goldenring: Yeah, I'm happy to be running advertisement for the city of Seattle. I absolutely love it out here. I am an outdoor enthusiast, and I was even before I moved here. But moving here kind of took that to new heights, if we're going to be a little bit punny about it. But when I, right, soon after moving here, I actually learned that I had some.

Some history with one of the mountains here, Tahoma is also known as Rainier. And I found out that my mom actually attempted to summit it when she was pregnant with me. And we did not make it to the top. I think you could probably blame me for that. But after hearing that, I was curious. And so I tried as well with a guided group and also did not make it to the top, but got to weirdly the same spot, which kind of continues this weird chain of events. But what did happen was that I really fell in love with the sport. And so I joined my local climbing club, which is called the Mountaineers. And now a few years later, I spend a fair bit of my week, three to six hours, probably helping instruct that course that I kind of kicked off this love with and teaching a cohort of students, leading trips and climbs.

So I definitely, when I'm not in front of my computer, am thinking about planning trips, going on trips, and definitely grateful to be in Seattle.

Chris Adams: Okay, cool. Thank you for that, Kate. All right. It seems we're going to continue this run of peak related puns with the next section as well, actually. So before we dive into this show and some of the things we'll be talking about, here's a reminder that everything we talk about, every project that we speak about will link in our show notes.

And the show notes are also published in markdown form on a GitHub repo now. And this is my point to thank one of the, one of the people from who actually submitted a pull request to our last set of show notes, Ross Fairbanks of FlatPeak. He made the first PR to the transcript where he helped us name, basically last week, we were talking about some of the specifics about,

how do I say, about a certain kind of attack, security attack that is actually named after an animal. We couldn't remember it, but it's, you need to know about this if you're going to do decent energy and carbon reporting. So we were asking, "can someone please name this?" And Ross was dutifully the person to come in and name, "yes, this is called the platypus attack," and not just that, it's one of these cool IT security vulnerabilities with a really, really nice looking website.

So we've now updated that. I will link to both the PR and this attack. So you now know when someone is telling you why they can't share information, you can say, "well, no, yes, we've heard of the Platypus attack and this is the mitigation for it." So yeah, that's like the depths of our nerdery that we go into.

All right then. So we've covered that. And if there are any typos or things that we mentioned that you don't see linked, you will be able to make a pull request and we will basically thank you in just the same way we thanked Ross. So you can get your 35 seconds of fame as well as you do this. Okay. And hopefully that should be a good incentive for listeners to basically help create a useful artifact for other people moving forward.

All right then. Okay. Kate, I tried to give a good intro into serverless computing. And when I saw the talk that you presented, you had a really interesting framing that I hadn't come across where you spoke about the value of a service versus the costs of making it available to people. And I figured this might be a nice way to kind of open up, because this is actually, I find it really, really enabling for me to think about kind of having websites available and how you even do that, or even why you might even have a website running in the first place. So maybe if I just like gave you the floor, it would help us set things up for some of the later conversation, actually.

Kate Goldenring: Sure. And I want to call out that I similarly had an epiphany when I discovered this concept and I actually found it within Mark Brooker's blog. He's a distinguished engineer at AWS and one of the original authors of AWS Lambda. And in this blog, he talks about how the cost of a system does not always equal the value you get out of the system. And in a cloud context, the cost of running an application on a piece of hardware is the amount of compute resources you need available for it. And so as you mentioned at the start, oftentimes we provision for the peak. So the traffic to an application is rarely constant. And when you're choosing how many resources to provision, you have to make sure there's enough of those resources of CPU and memory to handle a burst of requests. And so this means that we pay for the short term peak traffic that an application may get. However, when we think about the value we get out of a system, that's the long term average traffic. And you can think about this and even see this in payment models for serverless platforms, where you may be paying for average requests per month. And so the problem with this, when you're paying for the peak but you're receiving the average is that we have this gap and that can cause low resource utilization. And so we ideally want to close this gap and a way to do that is with multitenancy. And so multitenancy is running multiple independent applications in a shared environment. And if you increase the multitenancy of a system, so you throw a bunch of uncorrelated applications on the same hardware, where each has its own traffic ebbs and flows, the idea is that their peaks balance each other and traffic flattens. And this line gets as close to average traffic as possible. And so now we've had our costs equal our value and it's, we're all happy. But this is in some ways very idealistic. We're assuming applications are uncorrelated. We're also assuming that you have some underlying technology wherein when you aren't running your application, it is scaled absolutely to zero. It's using no resources. So this is where I get a little excited. Cause I do think we are coming upon a world where that can happen. And that is with WebAssembly, which can have these sub millisecond cold starts. And the whole goal of all of this from the context of green software is that we want higher hardware utilization, as you mentioned earlier.

Chris Adams: Okay. So what it sounds like you're saying is that this, like this idea of like multitenancy or maybe a little bit like a gym, multiple people using the same thing. This is actually a kind of a common theme. And this basically increases the average amount of use of the kit. So it's not sitting there just waiting to do stuff, essentially.

Right?

All

Kate Goldenring: Yeah. And just to point out a stat to kind of nail that home. Sysdig did a report in 2023, the Cloud Native Security and Container Usage Report, and found that 69 percent of CPU is unused in containerized cloud deployments. So we're really only using 30 percent of the resources there. And ideally we would flip that number and we would be constantly using 70 percent of the resources. And one of the ways to do this is with higher multitenancy and different technologies, and there's a really interesting paper from Meta about XFaaS, which is a serverless solution they now have created for private clouds that came out in 2023, and they found this phenomena that we talked about, the peak to average, and they describe it as the peak to trough, so not the average, but the lowest amount of requests at any given point, and found that it can be 4.3 times. So these peaks are quite high is the point of that. And the way they respond to it is by determining which workloads are delay tolerant. And so then they'll move the delay tolerant workloads to execute as a different time. And time shifting is a common thing and common recommended practice in green software.

So I also want to mention that this isn't always the way we expect it to be. Not, you can't perfectly fill a piece of hardware with uncorrelated applications and you might have to find some other techniques to balance out your request frequency.

Chris Adams: Okay. There's a couple of things that you mentioned that I just want to drill down on first of all. So you first mentioned this idea that most of the time, most computers are doing nothing, right, or they're waiting, like even with something like containers, which are considered a relatively efficient way to have lots and lots of programs running, more than two thirds of the time, they're just like not actually in use.

And this is at the kind of higher levels, like say when we talk about, say hyperscale providers, you might see people talking about, "Oh, we're really good. We've achieved say 30 percent level of utilization. And that's like 10 times higher than what you might have when you're using like a series of virtual machines or even like physical machines."

So there's a, basically a bunch of waste that this is designed to kind of get rid of, and I think the paper you referred to with XFaaS, they were, they basically said, "well, yeah, 30 is good, but we can go higher than that. If by, by doing some of these tricks, we've hit like 50 or 60%, which again means that many fewer bits of hardware to run this."

Okay. But you also touched on something there. You said that this was a private cloud, so there's certain assumptions. What about, about the softwares? Because it's the same company, they can trust each other's workloads and they don't need to keep these things as safe from each other as they, as other ones might, or there might not be the same, there might be the same priority compared to like, say, if you're running a cloud provider with lots and lots of different customers who don't necessarily want to share each other's stuff.

Maybe you could talk a little bit about this idea of virtualization and actually serverless and this idea of like isolation and why it's important. Because I think for a lot of people, they don't really, this is not something that you need to think too much about as a customer so much. You just need to know it's there, but when you're designing a system, there's actually quite a few things that you need to take into account for this.

And there's been different approaches that have been developing over the last, say, 10 to 15 years, for example.

Kate Goldenring: Yeah, I think we talked about how great it is to throw a bunch of different people's applications all together on the same piece of hardware. But that is a very scary statement to a lot of people to have their application running next to someone else's application. So there's this concept of isolation.

The idea that I will be okay with you running my application next to someone else's in a public cloud so long as these criteria are met. And isolation really just means that you know that your application won't be prevented from using resources because you have an application that's hogging them. And you also know that there's not another application that's going to access your application's data.

And we can actually look at the cloud as having gone through waves of this concept of isolation and keeping things separate from each other. And with each wave being motivated by a more finite mechanism of isolation, and also leading to a new type of cloud application that's come out of that wave and you hinted that kind of the start, which is virtual machines and that the advent of virtual machines, the ability to virtualize hardware with hypervisor technology has made it so that we can have multiple independent applications running on the same piece of hardware. And it didn't change the way people necessarily built their applications. It was still monolithic applications, but we could do more with the same amount of hardware now, and for context of time to start up and scale up, virtual machines take a few minutes to start.

So, if you wanted to scale up the number of virtual machines, it was a slower process. Next wave, you can think of as the wave of containers, and with containers, now we're virtualizing a smaller set of technology. We're virtualizing the operating system, and now we can run dozens of tenants on the same physical hardware. And this has brought about the advent of microservices, so I can scale up individual parts of my application instead of just scaling up the whole entire application.

Chris Adams: So could I just check, so you, so we spoke about VMs, virtual machines, so it might take a few minutes to spin up and down and then containers will be a little bit faster. Maybe Sub second or around a second. And both of these, like, these are, these are still an improvement on a physical machine, which can be weeks to get a new box set up, for example.

But the advantage of a physical machine is like you have like literal physical isolation, it's totally separate from computing. So that's the idea. And there are very different approaches you can have to separate this out. And you were talking about containers and VMs, and I understand there's a few other, there's a few other steps on this kind of continuum from a total separate machine to your share of a physical machine inside it in a VM or a container or something like that. Right?

Kate Goldenring: Exactly. And so we've gotten smaller. And then you can think of as the next wave being one of serverless. So being able to have even more applications on the same physical hardware. And this was really pioneered by the advent of the micro VM. And if you're familiar with AWS Lambda, the underlying technology for that is the micro VM, and with micro VMs now we, it only takes 125 milliseconds to start up a micro VM, and so it creates this opportunity to be able to run something only when you need it, which is a great, great future to realize.

The only issue with that number is that we, we know that a hundred millisecond is kind of the boundary of where you start to notice latency, and when you add execution time and network on top of this, this 125 milliseconds becomes a bit larger. And so the result was they had to pre warm these instances, so they had to get them running before you even needed them, so they have to use resources, even when they're not being used.

And that, that eats into our cost and value equation we had earlier. And so Serverless has these great opportunities, but hadn't been realized quite yet. And so the next way that I'm particularly excited about is powering and realizing serverless with WebAssembly, and achieving even higher density, and having those instant cold starts, and we're still isolated, just like a micro VM.

But instead we use a WebAssembly runtime to isolate your application using linear memory and WebAssembly also has a capability-based security model so that you only have access to resources you've been explicitly granted access to in the runtime. And so you still, through all these waves can have varying, but pretty solid levels of confidence that your workload is isolated, but we're getting into different levels and ways of computing and building our applications

Chris Adams: Uh, Ah, okay, thanks. Alright, so you spoke about this, this thing called micro VMs, which I understand was it, it sounds like basically an virtual machine, but smaller. And as I understood that, that might be, it's smaller because rather than trying to create totally virtual machine where you have all the kind of things that a regular virtual machine might have.

Like you can plug a printer in or plug stuff like that. It's just a much, much smaller service area. And as I understand it, there's a service called fly. io. They use Firecracker as one of their main options. Okay. I think I understand where you're going with that. And that, so that's an improvement on VMs and possibly has offered some of the kind of isolation benefits that some people have basically criticized containers for not having.

But then with like WebAssembly, this smaller thing, there's a, there are basically different approaches taken to make sure that other people's code isn't going to touch your code and vice versa. That's what you're talking about with things like linear memory and this capabilities based thing where you almost, you have to grant access to something first, right?

Kate Goldenring: Exactly. There's different methods to achieving isolation. And I also want to point out that we've been talking about one type of isolation, which is per-tenant isolation. So the idea that one tenant is isolated from the other, but there's also an even finer grained type of isolation that we can get in this latest wave of cloud computing.

And that is per-request isolation. So the idea is that my one request to my application should be isolated from the other. And in an ideal serverless world, you'd be able to guarantee statelessness across requests so that if my first request triggered a bug, It should at most affect that request. And the way you do that is by creating a new instance or a new isolation for each request.

And that's been really hard to do in the past when it takes 125 milliseconds to start up your virtual environment. But with WebAssembly, because it can start in a millisecond, you can actually create a new instance of your application for every single request on request. So you have an even finer interpretation of this isolation world.

Chris Adams: Ah, okay. All right. That helps me. Okay. So you've gone from like 125 to one, which is much, much, much faster. So that's the general idea that you have there. And I should probably ask. When you're doing something like this, I, I kind of understand the idea that if you are able to ramp up and down really, really, really fast, then you you can probably fit more peaks inside it because someone else's peak will end, will end faster than making space for someone else's peak to come in, basically.

So that's the general idea. I'm not that familiar with WebAssembly myself, and maybe it might be worth just talking about what that part actually is, because I'm a developer. I might code in Python or possibly JavaScript sometimes. Do I need to learn something new? Maybe you could just touch on some of that before, and then we could see how some of that fits in, because this sounds cool and something starting up 125 times faster also sounds good, as well as being much, much more efficient. But I quite like coding in Python and I'm getting a bit old, so I'm not sure if I want to dive into learning yet another language, if I can help it, basically.

Kate Goldenring: I definitely don't want to take Python from you. So WebAssembly, as its name suggests, was actually made for the browser. So the browser, most applications were built in JavaScript, and people were tied to their language of choice, say Python. And so WebAssembly was created so that you can build web applications in languages other than JavaScript.

It is just a target for a language runtime and a universal bytecode. And so you can write your application in Python and compile it to WebAssembly, and then you could run it in the browser. That was the whole idea. And because it was run in the browser, it got all these different characteristics, such as being isolated because the browser has a bunch of. public code running next to each other. And so you need to isolate it. And that's also why it's fast. But the result is that we have this portable bytecode that can be compiled to from a lot of different languages. And if you're curious which languages can compile to WebAssembly, there are different categories of the level of support a language can have for WebAssembly.

It can compile to Wasm. And also just to clarify, I've been using this word Wasm. Maybe as well, Wasm is an acronym, so a shortening of WebAssembly. So I might

Chris Adams: Ah, thank you. I was going to ask about that. Cheers.

Kate Goldenring: So their WebAssembly, it can compile WebAssembly. It can also compile to a WASI compliant version of WebAssembly. And if you've heard WASI before, that stands for the WebAssembly systems interface.

So it's a set of interfaces that describe how a WebAssembly module can have access to host resources such as IO, networking, et cetera.

Chris Adams: Okay.

Kate Goldenring: So different languages have better support for that as well.

Chris Adams: All right. So can I come in just, so can I come in just one second there? Because you've got me quite excited now, the idea that, okay, I don't need to learn a new language and me basically spending 10 years or 15 years, in my case, trying to get kind of basically some more competent Python. I don't need to throw that away.

And other people can, in whatever languages they use, they can also use that. And I think you said something quite interesting about, like, I know that under the hood, when I'm coding in Python, really something's being compiled down to some kind of like assembly. As for example, but I know that when I was using my, I'm using a MacBook right now, which has a Mac chip, and previously I had like an Intel Mac.

And then I had all these problems because there's different bits of, there's different architecture that these have. And I know that when I'm pushing code into say cloud sometimes. I'll be told that, "Oh, this doesn't work because you're not using an Intel machine." How does this relate there? Because it sounded like there was one kind of binary to like rule them all.

But I'm, I'm not, is it really like that? That sounds kind of quite helpful. And I'm, I would appreciate that, but that sounds a bit too good to be true, basically.

Kate Goldenring: I think it is good and it is true,

Chris Adams: Oh, wow.

Kate Goldenring: So, the thing about a browser is that it has to be able to run on all these different operating systems. So the code that runs in it needs to be able to do the same. And so WebAssembly, that .wasm file that you've compiled your application to, can run on pretty much the majority of operating systems and platforms.

So, Windows, Linux.

Chris Adams: I see.

Kate Goldenring: Et cetera. also like 64 architectures, ARM architectures, et cetera. You can execute that same .wasm file. And we're going back to like the people listening to this, who are passionate about green technology. This means there's no more cross builds. So if you think about how much your GitHub runners or whatever CI you're using, how they have to build everything for all these different

Chris Adams: all those

matrixes, like all for a version of Python that go, okay. All right. So you get rid of, wow. Okay. Yeah. Okay. I didn't know that hadn't actually picked up on that. And when you say this WASI thing, so if you are used to Linux or even Unix, there's this thing, idea of like POSIX compliant, right? Where

they're all going to talk to like a foster in more or less the same way. Is that kind of comparable? Was that generally the idea for that as well?

Kate Goldenring: People will definitely compare WASI to POSIX and I think there is some comparison there for sure. WASI is definitely different in that it is a set of interfaces. So, it's basically defining a contract or an interface between a WebAssembly module and the host, or even another WebAssembly module. And this is where we get maybe into something that is called the component model, which we don't need to dig into, but it's providing a new way of building applications so that these different languages can actually talk to each other within an application.

So you can imagine a universal URL library that's maybe built in Python, but can be called from Rust. And that's by adding a wrapper of the component model on top of a WebAssembly module, and they're all talking through these interfaces. And yeah, WASI describes a set of interfaces and actually. 0. 2 of WASI was just released, and that was a huge milestone.

It happened only a month ago, and that has promoted the standardization and the stabilization of these interfaces. So we now have a stable release that you can target with all of these interfaces and you can find them within the WASI repo, what exactly those interfaces are

and what capabilities it provides.

Chris Adams: Okay. So if I understand it, that's basically, so this, why is he thinking? Yes, it's a little bit like POSIX, but it's, it's, it's a bit more involved in that. Like it's, the idea would be that you could have that connect to maybe another system with a kind of predefined way. So rather than me having to kind of send a bunch of json in a serialized fashion, then you have to un serialize it, there's like predefined ways of these talking to each other.

Okay. I think I understand where you're going with that. And that does sound quite attractive. And that sounds useful to someone who's a developer or making some stuff here. Now, I understand that you are working on some of this yourself, and I understand that there is a, that you're, you're using one platform that, so I think when you showed me this talk, when you responded in this forum, you pointed to a talk with a platform called Fermyon Cloud, which does cover a sub, a couple of these languages that lets you run things in this new framework, Wasm thing, for example, could you maybe talk a little bit about,

about that part of what the experience looks like, cause I just, it has me quite curious and I assure you, we'll get back to other things, but that was, did sound kind of cool. And I just want to go down that rabbit hole for a second, if I may, actually.

Kate Goldenring: Yeah, and I think I've been painting this beautiful picture of WebAssembly, but it is still a fairly nascent technology. And so Fermyon is the company I work for, and we were really excited about simplifying the developer experience. So having someone come in and say, "Hey, I want to use this WebAssembly technology.

How can I use it?" And so if you're familiar with Docker, which kind of created that experience for containers. They created this Docker build, Docker run experience. We wanted to do that for WebAssembly with also the step of helping you even create that application. And so the open source tool for this that we've created is called spin and it is becoming the way to build serverless WebAssembly applications.

And so just with a simple spin new, you'll have templates that pop out for all these different languages. So we have 10 different languages that you can scaffold an application for. And there's SDKs for Rust, Go, Python, JavaScript.

Chris Adams: Ah, I see.

Kate Goldenring: Yeah, so now you've created this application, you've scaffold it, and then you can do a spin build and a spin up and run it locally. And remember, this is very universal. So whether Windows or Linux, et cetera, you can run it and then you can do a spin cloud deploy and deploy it to Fermyon cloud where we can host your applications and it's free and you can even share it with someone or if you use Kubernetes you can deploy it there.

So we're really trying to simplify the developer experience of using WebAssembly and also running it in your place of choice.

Chris Adams: Okay. You said a couple of things that caught my interest there. So first of all, yes, there is basically a tool, a little like how I might use Fly, Kotal, like to make an app or Heroku or basically some CLI to create a kind of harness for me to be developing in. And then I push to cloud, and presumably like Fermyon cloud, that's maybe a paid service or something like that.

But you said I can run it on Kubernetes, which isn't necessarily run by yourselves. And you, I think in the demo, you showed me, I think the demo I saw was running on Nomad, which is a service that we use as well, where I work basically. So. You're not tied to any particular platform and once something is in this kind of Wasm format, you could have it in a number of different places that do serverless.

You're not tied to like one ginormous provider or maybe two or three massive hyperscalers. You've got a bit of freedom in who you choose to work with in that scenario. Yeah.

Kate Goldenring: Yeah, and I think that is what is really exciting also about this wave of serverless is the cross platform agnosticity of your, your bytecode that you've created. So this same dot Wasm file ideally isn't locked into one cloud, so you can run it on Fermyon cloud, or you can run it on AKS's WASI node pools, or you can run it on other serverless platforms or locally and that same

application shouldn't be locked into anywhere because we're using the standard interfaces to run it. And so I think that's very exciting because what made Kubernetes so powerful was that people could switch from one cloud provider to the other. They didn't feel locked into their decision, and that's what was able to spread and kind of evangelize that kind of cloud computing.

And so I think the lack of vendor lock in is very powerful here.

Chris Adams: Okay. All right. So that's quite helpful. And this actually opens the door for some of the kind of, well, something we spoke about last week with on the last episode, there were some people talking about a company called Civo, I think Civo, who we talk about pools of resources, and we were joking about pools of resources in literal swimming pools, because they have a whole thing where they basically run computers

which generated lots and lots of waste heat, somewhere where the actual heat is useful. Because, the swimming pools would typically need to be, basically get the heat from somewhere. And they often do it by burning lots and lots of gas. So, the idea was by having something like a kind of, something which was kind of compliant in this way.

I think they run like a bunch of Kubernetes boxes somewhere, for example, and they don't literally put it in the bottom of this room for like some kind of piratical servers treasure chest. It's actually like, it's somewhere near, but this is one way of actually systemically thinking about some of the outputs from computing and basically doing it in a more kind of environmentally sustainable fashion.

It sounds like basically tools like this, or even serverless like that, you can run in these places. And that's actually one of the options.

Kate Goldenring: Yeah, I think, I think you're talking about Deep Green and that,

that work with, and we actually use Deep Green with spin applications. You have access to all these external resources cause they themselves are stateless, but you can still attach them to key value stores, SQL stores. And another resource we've provided is LLM inferencing on some GPUs that is powered by Deep Green.

And like you say, they're submerged in pools. And I believe. in vats of oil. And that captures first in the oil, then in the pool, like in a confined space. And that captures that heat externality and powers a pool. So I get very excited about the idea of capturing that externality of that compute. And yeah.

Chris Adams: Okay, cool. That sounds, all right. I'm glad you mentioned some of that because this is something we're trying to figure out because when, when we, if you listen to the episode last week, it sounds like we think that the servers are literally at the bottom of the pool. And obviously that that's not how it really works, but basically there is some way of using some of the heat from that.

Okay. All right. So I think I've got a good understanding of this, of where WebAssembly might fit into this. And I can see why when you make something, which is easy to run in lots and lots of places, then you can reduce the amount of hardware needed, but you can also reduce some of the impact that it might have by putting it in a, in a place that's more, I guess, sympathetically designed to like it's surrounding, for example, rather than venting heat into the sky, you're putting it, you're making use, you know, a sensible use of that heat.

Okay. Can I just talk about something else as well? Because. We've spoken about efficiency so far, mainly in terms of matching different peaks, going up and down and everything like that and averaging this out. So that's like one option. But I think that one thing I saw you speak about, which was a new concept to me, that I just want to like check if I understand it correctly.

You spoke about this idea of over subscription or soft allocation. So you've got servers and scaling them up and down. That's one way that you might make better use of the resources. But the other one, and I'm going to try and use a gym analogy, because I think that was what I thought of when you spoke about this.

You said like, if you go to a gym there's like more memberships that get sold than the amount of kit that's inside it. So if everyone all tried to like use the same machine at the same time, it would work and there'd be massive fights. But because people aren't using them all at the same time, you, people kind of, you can almost oversubscribe and oversell some of this.

And in the context of computing, you see something a little bit like this. So you might have like a physical server with, I don't know, let's call it say 32 gigabytes of RAM and 16 cores. Let's pretend it like a few years ago, right? And rather than only allocating 32 gigabytes worth of, of servers or programs and 16 cores, you do more than that.

You'd over allocate that. This idea of over subscription is a, is a common way to make better use of the existing hardware resources. Maybe you could talk a little bit about that because. I don't think this is something that people are that aware of. And in many cases, if you think about how you might allocate the environmental impact to a server to do some calculations, you might not be aware of how many other machines are really running on that computer.

It's not a one to one mapping in all these cases. I think that's my understanding. Maybe you could shed some light on that because this is something that I think a lot of people don't have too much exposure to.

Kate Goldenring: Yeah, I, I really like that gym analogy. And I think that idea is that say I haven't... that 32 gigabyte and 16 core machine and I deploy a thousand applications to it. If all 1, 000 got a request at the same time, some would be slowed, some would not succeed because there's not enough resources on that machine.

But I'm making a statistical bet here and it's informed. Usually you do some sort of understanding of traffic flows. That no, they're not all going to receive requests at the same time. In fact, I'm expecting only 30 percent to be running at once. I don't know. That's a random number. But because of that, you're able to over, you're over oversubscribing that machine. And what the result of this is, is that I'm using one machine instead of two or three, which is very exciting. And the reason, the only way you can do this is if you know that when it's not being used, it's not at the gym. So the key to this is that you leave the gym when you're not at the gym. And so you need a technology that does that.

Something that actually can scale to zero. And in the past, we've been oversubscribing already with technologies that at least scale down. And even microservices, when they're not in use, are scaled down. They're just kind of idly sitting there using less resources. And if we put this in a formulaic terms there, you were saying measuring is something that we like to do, and in measuring our software carbon intensity, the Green Software Foundation has that Software Carbon Intensity formula,

and that is a sum of operational emissions and embodied emissions, and it is very clear that you reduce operational emissions by not running your application. When you're not using it, so by leaving the gym,

but actually by oversubscribing, we're also reducing embodied emissions by having higher density and not needing as many servers. And so I think this whole concept can help us decrease our software carbon intensity.

Chris Adams: I see. Okay. And that degree to how much you're oversubscribing is like, that's flexible. For example, you might choose to massively oversubscribe and just accept that the performance might be a little bit ropey or in some cases, or you might be a bit more conservative in which case you're probably going to be using a bit more hardware, but you're going to have more of a guarantee that these things will work at their speed,

at the desired speed. And that was probably the approach that we've taken over the last 10 years, but there are new approaches available, like you're describing with Wasm and stuff like that. All right.

Kate Goldenring: Yeah, exactly. And I think when I talk about it being something that is hopefully it informs decision on how much to oversubscribe. We personally did load testing. So I had a fairly small machine, like eight cores, 32 gigabytes, and just a hundred gigabytes of disk and just threw as many applications as we could on there. Figured out what that density limit was and then hit it with a bunch of requests and a loader and a prober to try and figure out what are we comfortable with and setting our own limit there. And so I think everyone has to do that to kind of figure out what that, that important level of density is for them.

Chris Adams: Okay, cool. Thank you for explaining some of that. For those who are listening, we've linked to a couple of papers specifically about this. There's one by a gentleman called Noman Bashir, who has written a bit about how this over subscription thing works. And we'll also link to this paper. There's lots of reasons you might be ambivalent about Facebook, but the paper, this XFaaS paper is actually quite interesting.

And the thing that, one of the things that's interesting in the context of CarbonAware software is that, well, they, this idea that you might over subscribe and rather than that, you might also kind of offload to entirely separate data centers as another strategy to kind of keep serving things when maybe other parts of the kind of fleet of machines that are able to kind of pick some of that up, especially if they are maybe low carbon intensity because they're running a particularly green part of the grid, for example.

Okay. All right then. So we've covered quite a lot of ground and quite a lot of, you know, quite detailed concepts here. And we're definitely going to link to a bunch of papers for this. I wanted to just actually ask you about, about this, cause you mentioned a little bit about Fermyon and you mentioned a little bit about WASI and some of these tools.

If people are curious about some of this, where would you suggest people look? Because until I actually had this call with you, I knew there was a thing called Wasm. I knew like, it's this thing that I've seen in Pyodide as a way of like running Python in the browser. And I kind of understand that, but I didn't know where to start.

And it sounds like you don't need to be writing everything in Rust, so you can do things that you could write in, like, can you write in JavaScript, for example, to have something run that runs on a platform like this, for example?

Kate Goldenring: Exactly. You can use JavaScript. And if you want to try this out with Spin, Spin is on GitHub, but also Fermyon.com will lead it to you or developer.fermyon.Com as well. And. JavaScript works there, and we've simplified that experience by we have actually plugins to help you compile it to WebAssembly, and so you just install the plugin for that language, and then the spin build works correctly for you, and you can then immediately run it.

And so it really is hopefully making that experience easier. Simpler and, and getting you going and it is an open source community. So if you are an open source developer and want to contribute, we have a discord that you can join to join the discussion, or if you have an issue, put up the issue or even maybe grab one too.

So we're really trying to grow the ecosystem. And I think we have around 75 contributors currently, but we're always looking for more hands.

Chris Adams: Okay, cool. All right. We're just going to wrap this up. And I know there will be at least one person asking, how do I do some of this work and make sure it runs in a swimming pool? Not in a swimming pool, but in a kind of very, very green fashion. If you would be able to share a link for that for the show notes, I'd be very, very grateful because that sounds kind of cool.

And the idea of putting computers in lots of new places where the heat is useful rather than a waste sounds kind of exciting to me.

Kate Goldenring: Happy to do that.

Chris Adams: Okay, cool. All right then. So we spoke a little bit about... okay, this is like a kind of serverless that's somewhat flexible and quite resource efficient. And one thing that I think you demonstrated, which caught my eye was you were spinning up how many, it was like a thousand applications on your laptop to demonstrate that, yeah, you can run serverless in lots of places.

Can I just actually ask you a little bit, how about that came about because it was a pretty cool demo and it was lots of fun and I haven't seen nerds clap over for a demo for a while, but it was really, really nice actually.

Kate Goldenring: Yeah, so I hinted a little bit about the motivation for that, and that was to figure out how much we could handle in our cloud. So I was tasked with load testing our cloud and basically coming up with a number of, this is the size of our instances we use, and, and how many applications can we put on each instance.

And so, as you pointed out, I was using Nomad there, and we basically had a multi tenant version of Spin that had a listener for every single application that was deployed. And then just deployed as many applications as we could on my laptop, my poor laptop, until it couldn't handle it anymore and just cleaned it with a bunch of requests. And that was all in the vein of trying to figure out how dense we could go.

Chris Adams: That was really useful for me to know because I, until I'd seen that, I generally assumed that serverless almost always means either some massive, chunky Kubernetes cluster, or it's going to be some proprietary tool that, okay, I understand why they're, they're, they're necessary, but that makes you feel a little bit uncomfortable about, okay, this is my future is only one provider, for example, that provided a bit of freedom on that.

Okay. Brandy, if people wanted to follow your work, for example, where would you direct people's attention to? Cause we're just coming up to time and this has been lots and lots of fun and I've really enjoyed seeing some of your talks and some of the things you've been writing about. And I suspect there may be other, other ones as well, because yeah, this has been really helpful for me.

Kate Goldenring: Yeah. I tend to try to promote the interesting things I find or I'm putting out there on my LinkedIn predominantly and sometimes on X as well. And both of those are just my full name, Kate Goldenring, as the handle. But if you want to talk more specifically and personally about a question, I'm on the CNCF Slack, the Kubernetes Slack, Bytecode Alliance's Zulip, and then if you're just interested in following blogs about WebAssembly and Serverless, Fermyon, we have a pretty active blog on our, our website, so just.

permian. com, you can find it. And so, and that's where we can find more out more about those swimming pools that we have some GPUs running in.

Chris Adams: Okay. Kate, well, this was lots and lots of fun. I really enjoyed it. And I, if nothing else, I enjoyed actually reading all the papers and learning so much about this. So, Kate, it should be coming up to eight o'clock for you. So I should probably say, have a lovely day and thank you very much for that. Kate, I really enjoyed this and I learned a bunch of new things about serverless, scaling up, scaling down, gyms, swimming pools, and Wasm. Thanks, Kate.

Kate Goldenring: Thank you so much, Chris.

Chris Adams: Hey, everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

Carbon Hack 2024 Announcement

14 februari 2024 | 6 min

We're gearing up for the return of Carbon Hack 24 happening from March 18 to April 8, 2024. This year we're focusing on the Impact Framework, aiming to innovate how we measure the environmental impact of software. It's a chance for engineers, designers, and creators to dive into challenges, ranging from plugin development to non-technical content creation. We're inviting students too, with special categories just for them. Join us to push the boundaries of sustainable software. For details on how to register or to become a sponsor, check out the link below:

Register at https://grnsft.org/hack/podcast 👈

Find out more about Impact Framework:

Environment Variables Episode 58: BETA Impact Framework

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Asim Hussain: Hey everyone, Asim here. Carbon Hack is back this year. The hackathon will take place from Monday, March the 18th to Monday, April the 8th, 2024. Carbon Hack 24 is all about redefining the way we measure software to reduce its environmental impact. footprint. At the heart of this hackathon is Impact Framework, an open source tool that lets you compute and report the environmental impacts of software applications accurately.

It takes observations you can easily gather about running systems such as CPU utilization, page views, or number of installs, or anything else, and converts them into environmental impacts like carbon, water, energy, air quality, in an auditable, replicable, verifiable, and transparent way. With Impact Framework, if we can observe something, we can measure its impacts.

If you want to find out more about what the Impact Framework does, you can listen to episode 58 of Environment Variables.

Why the focus on measurement, you might ask? Well, we've learned that at the intersection of software and sustainability, measurement is the compass guiding our way. How we measure software's environmental impact is more than just a technical detail. It's the key to unlocking the magnitude of change we can achieve.

Here's the challenge. Alone or in small teams, participants will have the freedom to choose from a variety of prize categories, including Best Plugin. Impact Framework itself is just very, very simple. The power of Impact Framework is its plugin ecosystem. We've got plugins which take CPU utilization and turn it into energy. We've got plugins that take energy and turn it into carbon. We need more and more and more of these plugins to be built for my statement I said previously to be true, to take any observation and convert it into environmental impacts. We need plugins for that. So one of our prize categories is best plugin. We want you to build the best plugin you can build. If you submit your plugin idea on the Hack website, We actually nurture you, we help you, we guide you towards designing a plugin which has the best possible chance of winning.

There's another prize category called Beyond Carbon, which I'm very, very excited about. If you can build a plugin which outputs an environmental impact that's not carbon, for instance, water. You can submit your solution to the Beyond Carbon prize category. We really want to move the conversation beyond carbon this year. That's my personal goal.

This year, we've also got prize categories for non technical contributions. We've got a best content category. This is a category for people who would create the best piece of content, be it a how to video or a tutorial or a written booklet on how to use Impact Framework in any context, be it for a particular domain or in a generic sense, that's what best content is all about. How we're also expanding best content is if you're going to write a case study that uses Impact Framework to measure the impact of something and you write up that case study in such a way that it becomes a learning resource that also counts for best content as well.

We also want to encourage people to contribute to the framework itself. The framework is a piece of code which sits on the Green Software Foundation's GitHub repository. And if you can contribute to that code by going to our GitHub repository and picking up an issue and saying, I would like to work on this issue for the hackathon. Please go right ahead and we'll nurture you, we'll guide you, we'll support you because we want to make sure your contribution is accepted. But that is how you can submit the best contribution category.

And this year we're thrilled to open our doors to students. We have two prize categories for students. We have an undergraduates category. If your entire team is made up of undergraduates, you can then win a set separate accessory prize for undergraduates.

And again, if your entire team is made up of under 18s, we also have another prize just for you, an under 18s prize.

So how can you become part of CarbonHack24? It's as simple as signing up on our website at grnsft.org/hack/podcast that's grnsft.org GRN SFT. org forward slash hack forward slash podcast.

There you'll find all the details about the event registration and upcoming live sessions where I'll be talking every single Monday, giving you an overview of the hackathon, what's going on the latest information and just taking you some of the latest projects. This is your opportunity to be part of a global movement that's shaping the future of sustainable software.

We also extend a warm invitation to organizations that align with our vision to join us as sponsors. This is your chance to witness early stage innovations, connect with talented individuals, and showcase your commitment. If you're interested in sponsorship, visit the same link, which is grnsft.org/hack/podcast.

I'm thrilled about CarbonHack24. We're going to explore innovative ways to measure the environmental impact of software. Join us for three weeks of exciting challenges where engineers, designers, and content creators will use Impact Framework to measure software's environmental footprint.

We can't wait to see what innovations and solutions emerge from this incredible event. See you there.

The Week in Green Software: Embodied Carbon

1 februari 2024 | 44 min

This Week in Green Software, host Chris Adams is joined by Gabi Beyer and Brendan Kamp from re:cinq to talk about the recent initiatives to and consequences of the prolonging of the lifecycle of technology. The challenges of measuring carbon emissions in personal laptops and cloud services are discussed, highlighting the complexities of quantifying environmental impact, as well as companies reporting on these metrics. Tune in for a lively discussion on sustainable software development.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Gabi Beyer: LinkedIn | GitHub
Brendan Kamp: LinkedIn | GitHub | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

Carbon Hack 24: Where measurement meets innovation, and impact knows no bounds [39:52]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRASNCRIPT BELOW:

Brendan Kamp: Transparency, right? I mean, this is the key for the cloud providers, I feel, at the moment, is just become more transparent about some of the, this data so that we can do the work to calculate, but just give us the data to do it, right?

Brendan Kamp: Hi, nice to be here.

Gabi Beyer: Hi. Hello.

Chris Adams: Cool. Alright, so, I'm not normally sure who to give the floor to first in introductions like this. So I'm just gonna go for like, alphabetical sort here, which I think places Brendan slightly ahead of Gabi in this one, right? So Brendan, are you okay just introducing yourself and what you do at re:cinq?

Brendan Kamp: Yeah. First time I'm ahead of Gabi in anything, so I'm happy. So I'm Brendan. I'm a founding engineer at re:cinq. So I've mostly in my career worked with platforms and system level stuff. I did a small sense of developments on apps and I've recently just started learning about green software. And so I've started a blog forum called thegreencoder.io to capture all the learnings and things that I've been doing.

Chris Adams: Cool. Alright, thank you for that Brendan. And Gabi, I think you're next in the sorting order, so yeah, over to you.

Gabi Beyer: Yeah, so I'm Gabi. I'm also an engineer at re:cinq. I've done a bunch of different types of engineering over my years from Linux to platform to now doing green coding and optimization engineering. And recently we actually, Brendan and I both got accepted to do a talk at KubeCon, so we're excited for that.

Chris Adams: Oh wow, congrats! I need to check. Where is KubeCon taking place this year? Because I, I'm a little out of date when it comes to knowing where that massive conference moves around the world.

Gabi Beyer: Yeah, it's in Paris.

Chris Adams: In Paris? Oh, wow. That's not that far from Berlin. Maybe we'll see you in person in that case.

Gabi Beyer: That'd be great.

Chris Adams: Yeah, congratulations.

All right. Okay, so thank you for that, Gabi. Brendan, during our intro call, just to get to know each other and figure out what we're going to talk about, you mentioned that you're joining this call from the Netherlands. But you haven't always lived in the Netherlands. And you even mentioned living on a farm at one point before. So before we dive into the world of like your life with servers, maybe we could talk a little bit about Brendan pre-cloud or some of that first, before we dive into the nerdery.

Brendan Kamp: Yeah, perfect. Well, so I've lived on a couple of farms. The first one in South Africa, I was born on a farm. And then a couple of years ago, I think it's about six years ago now, we moved to Germany. Didn't enjoy Germany, so we decided to, mid pandemic, do the only option of going into the middle of nowhere, Norway, and living just on the edge of a fjord in Norway for a couple of years.

So yeah, and now jumping back to Amsterdam to get a bit more into the fray of things again.

Chris Adams: So let me just get this. So South Africa, very, very warm. Germany can be freaking cold, but also kind of warm. And you thought, "nah, this is too warm and temperate for me. I'm going to go all the way to Norway," where you're like, "ah, this is maybe a bit too cold." And now you're back in

Brendan Kamp: Too cold. Yes. A hundred percent. Definitely a weather thing.

Chris Adams: Well, I could sympathize with that as someone who's been living in Germany for the last 10 years and seeing a kind of recent cold snap, but also it getting chuffing hot over the summer, or comparatively speaking for Germany, at least. All right. And Gabi, when we were chatting about this inside the CNCF Slack, which is the Cloud Native Computing Foundation, Slack, I think, I saw your profile photo and I saw that you had a bike, but you also had a adorable dog on there. I guess there's a story there too, before we dive into the world of servers.

Gabi Beyer: Yes, that is Bruce. He is famous among all of my friends, and he is 13 now. He, I'm from Seattle, and he came with me to the Netherlands. And yeah, he is enjoying cycling around the city just as much as I am.

Chris Adams: Oh wow, so he's got a little dog passport and doggy visa and stuff like that, right?

Gabi Beyer: Whole European passport.

Chris Adams: Oh wow, that's cool. All right, thank you. And hello, Bruce, if you're listening, I suppose. All right. Okay, so thank you very much for giving us the time and introducing yourselves. Folks, if you haven't listened to this podcast before, my name is Chris.

I am the executive director of the Green Web Foundation, which is a Dutch nonprofit focused on an entirely fossil-free internet. I also work as one of the chairs of the Green Software Foundation's Policy Working Group, which basically does stuff related to new legislation, figuring out how to interact with new laws being made, and also creating some kind of fora for people who are basically trying to figure out how to deal with green software at a kind of legislation level. We also do a bunch of other things at the Green Web Foundation, but I'll maybe mention those as we discuss. All right, then. So, Brendan, Gabi, I assume you're sitting comfortably, right?

Gabi Beyer: Oh Yeah.

Brendan Kamp: Ready.

Chris Adams: Okay, cool. All right, just before we start, sorry, I almost forgot. While we talk about this, we will mention a few stories.

We will link to all of these in a transcript and some show notes at the end of this podcast. But we also will have these as a markdown file on GitHub. So if there's a mistake for anyone who's listening to these, or there's something that you didn't quite understand, you can totally submit a pull request to improve it.

And we will thank you in the next episode for your diligence in helping capture knowledge for future listeners. Alright, okay then, I guess we can start with the news stories, Gabi, the first one, I think we spoke about this before, this is one that I might let you kind of talk about first I suppose. This is Deep Green, which is the company based in the UK who put servers I think at the bottom of swimming pools, or in swimming pools where they use the waste heat to heat the swimming pool so you don't need to burn gas to heat the swimming pool, but that also has a nice effect of cooling the servers themselves. They've been doing a bit of work with a cloud provider called Civio, and I found out about this through, there's a serverless provider called Fermyon who are doing some really interesting stuff with scale to zero servers. And I think a bunch of this stuff seems to be running on Kubernetes. I imagine this, between the two of us, Gabi, you might have some exposure to.

I've got to ask, first of all, have you used any kind of serverless tools in the first place like this? And do you think this is a trend that we're going to see more of, this idea of the cloud not just being in gigantic data centers, but in some cases at the bottom of swimming pools where the heat is actually being used in a productive way rather than being vented into the sky?

Gabi Beyer: Yeah, I think that it's a great initiative. We actually, Brendan and I attended a talk a couple of months ago about a Dutch company doing a very similar thing called Leafcloud, and they would put them also in pools. They've created, I think, a heater that yeah, would take the energy from servers and transform them into heat for pools and also apartment buildings. So they're putting them into apartment complexes to heat the housing.

Brendan Kamp: I was always wondering, cause Gab's kind of explained this a little bit when we were there, but I was always wondering how they do the security side of things. Right? I mean, data centers are generally these walled off environments where you're storing companies' data. So understanding the technology underneath that to kind of like not allow those kinds of intrusions.

I mean, that must be phenomenal. Right?

Yeah,

Chris Adams: This is actually one thing where I'm moving out of my expertise, but I do know that this is actually something that is an ongoing challenge when you think about edge and stuff like that, because most of the time you, like you say, people talk about data centers as big buildings with big scary dogs and men, often with like some kind of weaponry patrolling these, and this shifts to a wider, more distributed set of places does mean that you might think about physical security differently to some of the other ones. Now, my understanding so far is that there is a bunch of different forms of encryption used so that the actual content is not particularly readable in many cases. But this is also something that, because we're on a podcast and there's more than just the three of us involved in this conversation, albeit many of them more passively at the moment, something we can find a bit more about or even raise in future, because you're right, Brendan, this is going to come up again and again and again. I'm also glad, Gabi, that you mentioned Leafcloud, because I'm actually quite a fan of them, because as I understand it, Leafcloud offer things like, compute as a service, storage as a service, all the things that you might see from some of the big cloud providers. But, so it's the same kind of API, like I said, like the idea being that you don't need, you don't need to care where the cloud is. You just need to know that your tests pass or you're getting the same things responding. Maybe you could just touch on that a little bit more because I haven't actually used them myself but I did read over their website and they've got a cool domain.

I think it's leaf.cloud or something like that which definitely got some nerd points.

Gabi Beyer: Yeah, I actually haven't used it as well either, but talking with some of their engineers, I found them very interesting, along with the domain. They have some funny lunchtime antics that they seem to get along in. But as for the technology itself, I Yeah, I can't really speak to it because I've never used it as well. But to kind of go off of the security question or thought Brendan had, I'm under the assumption that they might, for both use cases, since the Fermyon one is serverless, I'm assuming that they don't store any data at these pool sites or at these hotel complexes.

And it's a lot of, they do still maybe have that big scary data center where they send all the data to, but yeah, the actual operations are occurring elsewhere.

Chris Adams: I see, okay, so maybe there's a thing where it's just being loaded into memory for a short period of time and then it's being processed and then the results are being put back somewhere else, back into somewhere else in the cloud where it's somewhere more durable. Well, I guess there must be some people who seem to be okay with this because Deep Green, in the last literally couple of weeks, there was a really interesting announcement. They raised 200 million pounds for a massive raise to start rolling out these kinds of like integrated servers all around the UK. This was an investment by Octopus Energy, who are probably one of the kind of energy unicorns, I suppose, globally, basically. So that's something there, and we'll share a link to that as well.

So maybe we should ask one of them about, say," In your due diligence, how did you figure out how to answer this question? Because when we were looking at this, we thought this was cool, but we didn't quite figure out this part here." And that's probably the thing that we need to understand before we start diving into dropping servers at the bottom of swimming pools and getting them to execute cloud functions.

Brendan Kamp: Just like off the top of my head, this sounds, I mean, this sounds right in the realm of Kubernetes, right, where you have this idea of, you just have all these nodes everywhere, which would be the pools, right? Like these little servers are just nodes and you're just using them as a pool of resources that maybe you

Chris Adams: Ba dum tss. Very good there. Yeah.

Brendan Kamp: Yeah.

Chris Adams: A literal swimming pool of resources. Yeah. Okay.

Brendan Kamp: Exactly right. Which makes sense why they would partner with Xevo, I think, cause I think Xevo is Kubernetes first provider,

Chris Adams: I think that is the case as well. Yeah.

Brendan Kamp: Yeah, so, I mean, it sounds plausible, and I can kind of do the mental gymnastics to how it could work, but it would be so fascinating to see some white papers from these companies coming out and saying, this is how your data is secure, this is why you don't need to worry, right?

Chris Adams: Well, I'm glad you mentioned that because we actually have a future episode with Kate Goldenring of Fermyon, who basically was one of the first people to respond when I asked some really nerdy questions on the Green Software Foundation discussion board about "how do people do isolation for this?"

Because one of the things about some of these serverless tools is figuring out, "okay, how do you make sure that when you're running some code, you're not being affected by other people's noisy neighbors" and stuff like that, basically. So yeah, we will definitely be diving into that into almost excruciating detail in a future episode.

So yeah, I'll be looking forward to that. Okay! Thank you for that, Gabi. This was fun. Okay, next story that we had on this was this one in the register, actually. So this was about users now keep cell phones for 40 plus months, and this is hurting the second-hand market, apparently. So this is one thing that, Brendan, I was thinking of you when this story came up, because we have seen a slowdown in people buying new electronics, and actually this has been one of the real problems we've seen cited with embodied energy and embodied carbon and hardware. And now we've actually seen a new trend where people are not choosing to update their phones every year by the looks of things.

Brendan Kamp: I absolutely love the headline, because it's like, they're making it sound like such a negative thing. "Oh, you're crashing this whole market of secondhand users," whereas you're just utilizing your resources for a longer time, right? And so, I mean, when I was going down the rabbit hole of embodied carbon, cell phones and laptops were one of the things that majority of the carbon emitted for these devices is in the manufacturing side of things and not actually in the electricity usage. So extending the lifespan of these devices just by, I think it

Chris Adams: one of the best things you can do. Yeah.

Brendan Kamp: Yeah, exactly, exactly. But it's also going to hurt the companies that make the devices, right? And I think that's where there's the struggle, is "how do we still keep this growth going for these companies so that they can innovate, they can like do these things, but not destroy the world by littering it full of.

Chris Adams: Almost. Yeah.

Brendan Kamp: Exactly. Yeah.

Chris Adams: So I just realize that we're talking about this and you've used this term 'embodied carbon' that may be new to people. So when you say the word embodied carbon, I assume you're talking about the fact, I mean, you did a quite good job of explaining this, the idea that if you're going to have a phone, because loads of energy goes into making that phone, like you need to heat sand up to like 1200 degrees Celsius to make a silicone crystal before you can slice it and then turn it into a chip. That's going to be a bunch of energy used there. And that's usually come from fossil fuel sources, so that's the embodied carbon that you're referring to in that context, right?

Brendan Kamp: Yeah. A hundred percent. And it's also, they try and include the other things in terms of like transportation of the device from the manufacturing plant to the sales warehouses, the end of life, carbon footprint, and all of these they try and calculate all together and they have this global term of embodied carbon. And for these devices, and you might need to check my data, but I think phones and laptops are like, 60 to 70 percent of their carbon emissions are wrapped up in their, in the embodied carbons, whereas servers, funnily enough, because of the immense power or energy that they use, it's only like 15 to 30 percent of their emissions.

Chris Adams: So it's the other way around then. Alright.

Brendan Kamp: Yeah. So servers, the embodied carbon isn't as big, but it's still a nice slice. And funnily enough, this is something that Gabi's and I have been struggling is how to calculate embodied carbon for like using of devices, right? So if you only use a device for a small amount of time, like how much of that embodied carbon pie is

Chris Adams: is attributable to you, it's your share versus someone else's share, right?

Yeah, you do need to kind of, I mean, maybe this is actually one thing that we'll just touch on a little bit before, because, Brendan, what you mentioned just like, oh, it, you know, this, the, the framing of the story, like, "Dammit, these people should be consuming faster," right?

I think that's an interesting one because there's also one thing, one interesting development we've seen on the policy side is that Google, which the register always refers to as the chocolate factory, which always makes me giggle; they're one of the second firms now that has actually come around on right to repair.

So I think last year or the year before you had a whole bunch of actual shareholder activism. You had some people putting forward a kind of shareholder resolution to basically start looking into the ways to reduce the embodied carbon of owning a Microsoft Surface. And eventually it passed, despite there being a bit of pushback against management saying, "no, we don't want to do this." And now you saw Apple come around last year saying, "oh, we're kind of okay around this right to repair," but they had this whole notion of like parts pairing, which basically meant that it's right to repair as long as you only ever buy the parts from one provider at Apple's prices, which is probably good for Apple shareholders, but that does kind of mean that that is not possibly the intention of the, of this. And then Google have recently come through and said, "yeah we're actually coming around on this as well because we realize we've got loads of Chromebooks around and if we can fix them rather than replace them that might actually do something for the embedded carbon."

And also this whole parts pairing thing, which has been one of the last things that large tech firms want to hold on to, they've said, "yeah, we're okay with not having the parts pairing thing" as well, as I understand it. So this is actually an interesting development. I don't know if there's another loophole or something like that, but this really caught my eye and I figured it was, I was thinking of you when I read this actually, Brendan.

Brendan Kamp: But now also coming out with the new EU greenwashing laws that have recently been passed, for Google, I would assume this would start looking good for their data because very soon, these big corporate companies, if they want to do work in the EU, are going to have to start providing this kind of data of just how much emissions are their devices emitting, etc.

And if they are reusing, and I mean, I've always felt like this idea of "you can only use our things" is ridiculous, because, for instance, if your sole goal is to make money, it works, but now we're shifting to a world where there's other incentives, and if those incentives are, "hey, we want to reduce our carbon," then if a vendor is providing the hardware, those emissions are for the vendor, they're not for me, they don't go towards my numbers.

And this means that you have a more collaborative society. So yeah, I'm really interested to see where this goes, especially in the next couple of years, you know, Fit for 55, the new greenwashing laws, because I think these tech companies are going to really have to do some changes.

Chris Adams: All right, you've added a couple of things that, Gabi, I'm going to ask you a little bit about. So there's a couple of interesting terms that you mentioned there, so Fit for 55, this is one of the big announcements inside Europe to basically say "we're gonna reduce emissions by 55% by 2030." And there's a bunch of new laws and things tied to that.

And you also spoke about this anti-greenwashing resolution. We'll share a link to that 'cause I believe that was passed last week or the week before, inside European Parliament. So there's a few follow-on steps and essentially saying things are carbon neutral isn't going to fly unless you have a really, really clear way of backing it up.

So when Apple have mentioned things like they're having an entirely carbon neutral watch in Europe, that won't be a thing that you can do. And I imagine we might see this also percolating into other territories as well. All right then. So Gabi, there's a really interesting story here, which is the, this is actually UBS, the bank, who are basically saying, "Hi, we're trying to develop a baseline for some of our early steps." They used to talk a little bit about how this is quite hard. And as I understand it, you spent a bit of time, and you even wrote about your experiences, taking some steps to try and measure things just on a single laptop. Maybe we could talk about that before we speak more widely about this one here, because there's some really nice things to touch on for the UBS story, and they explain a few steps they've taken, but maybe we just touch on your experiences with your first steps and then go work a bit more widely because I think you did write about this after, was it the SDIA green coding workshop in Berlin? But I think you mentioned something that maybe you talk a little bit about that then we can go into the wider case here because it's very rare to see large organizations talking about this. And to get there, you need to figure out what to do on your own laptop first, right?

Gabi Beyer: Yeah. Yeah. Inspired by the SDIA. Yeah. I wanted to measure the carbon emissions and the energy consumption of different processes running on my laptop. And that was challenging enough in itself. And to do that, I used this Intel RAPL hardware component that can measure the energy, and I used Perf, which is also a Linux command line tool to, which utilizes RAPL, sorry, which then can get the amount of joules of energy a process is running. And then, once you can get that energy consumption, you must calculate it into kilowatt hours, and again, multiply it by the grid database, and you can get the carbon emissions. So there's a lot of steps just to get the energy consumption and carbon emission data for a single laptop. And to build off of that, we've been working a bit on figuring out how to measure cloud, different cloud services, their emissions, specifically like VMs, which is similar to what UBS seemed to do because they were running VMs for their application data. And that is another level of challenge because the hardware components are being shared within the services. So you have to kind of then determine how resources are being shared.

Chris Adams: Yeah, so it's not just a single laptop. It's like you're sharing your laptop. So if I understand it, so doing it, it's hard enough for you with a laptop when it's just you owning the laptop, when you don't have to share that laptop with other people. When you're looking at a server, the number of people who you're sharing it with could be changing over time. And that means that There's almost like a kind of integration, mathematical integration problem that you also need to wrestle with as well. And that's another thing that makes it somewhat more complicated.

Brendan Kamp: Or maybe just to touch on, also you did the CPU calculations for VMs, but what you're currently struggling with is memory, right? Because there's no data on memory.

Gabi Beyer: Yeah, that's specific to the cloud though. We don't have their, like, hardware memory information, like what kind of DDRs and RAM stuff they're utilizing. And each one kind of emits different energy usage depending on that. So it's kind of difficult to say, like, "this application is using this part of memory whereas this application is only using this much."

So that's been hard to calculate.

Chris Adams: It's a real challenge. Yeah. This is something that we've come across quite a few times. There is actually two things that may be worth touching on here, actually. So inside the Policy Working Group where we work, we've seen various sectors really struggling with trying to come up with some reliable numbers for this.

And there's actually a roundtable that we're organizing in the coming months specifically for the financial sector, because amongst all of the sectors so far, they're weirdly one of the more transparent so far. I think it's partly because they've bought so many servers already, because they, That they actually have access to these numbers, whereas some cloud providers don't have this, and we'll show a link to a post about, there's a bank called ABN AMRO.

They're the third largest bank in the Netherlands. They actually were on the record basically saying, look, we've looked at our figures and 40% of our emissions are attributable to our own infrastructure, and partly this because they own some of their own servers, and they're moving to the cloud, but this is one place where it actually has been easier for people to come up with these numbers.

But when you get to the cloud, it becomes, as you mentioned, it's way more complicated, because Gabi, you said, who's RAM? What's CPUs? What machine? The fact that you don't know where it is means it's really, really hard to model some of this stuff as well, actually. And, yeah, I think this is going to be an ongoing challenge.

But I wanted to ask you, actually, when you're on this, there's a project that I think I saw in the kind of recent GitHub repo called, I think, is it called Carbon Cloud? This is a Golang thing or something running on the server. Maybe one of you could talk a little bit about that because that caught my eye, basically.

Brendan Kamp: That's all Gabi.

Chris Adams: Yeah? Okay. So, Gaby, I'm curious, maybe you could talk a little bit about some of that, because it seems like you've tried to wrestle with some of these problems yourself, first hand.

Gabi Beyer: Yeah, we're trying to, written in Go, we're trying to measure the carbon intensity of cloud software. But our goal is mostly to do it in real time and to be able to provide like a metric that users and developers can utilize and get that quick feedback cycle loop to see what is changing and going on in their code base that could be producing these higher emissions or change in emissions and stuff like that.

Chris Adams: All right, cool. Well, in that case, one thing you may be interested in is, so in the GSF, there's a project called the Real-Time Cloud Project, which I've been involved with, and we'll talk on, talk about a little bit later, because this is one thing that, as someone who's working in an organization that cares about a fossil free internet, a significant chunk of that is going to be, well, decarbonizing the power. And this is one thing that we've struggled with as well. It's really, really hard to find the numbers for this kind of stuff. And that probably brings us to the story that I kind of put forward here. So there is a really interesting story from Data Center Dynamics about Amazon, basically saying they're claiming they're the world's largest buyer of renewable energy in 2023. They said this story and we'll share a link to this and it sounds, this sounds really, really good when you first think about it, right? Like they're a large provider, they, they basically mentioned that they've invested in something like more than 100 new solar and wind energy projects and stuff like that. This adds to a total of nearly more than 500 wind and solar projects globally, right? So these are the figures that we have here. And this sounds on the face of it kind of large if you were to assume how big Amazon is compared to other providers, right? And I've been really struggling to figure these figures out because if you look into sustainability reports from any of the large providers, Amazon, as one organization that hasn't really published its absolute energy use figures ever, or it's very, very hard to find this information.

If you sign an NDA, you might get the figures yourself, but most of us don't sign NDAs with Amazon because, well, I'm not spending 15,000 Euros at least each month with them myself. But, what we found out when I was doing some research into this was the Carbon Disclosure Project is one place where these figures are submitted.

So I did a bit of research myself and we've shared a link to this specifically. Google and Microsoft both published their CDP responses listing a bunch of really interesting answers to the questions and this is helpful because it allows us to have comparisons across each of these. And the short version of this is that when I looked into this, the figures for absolute energy use were about just under 40 terawatt hours, 39 terawatt hours, is what was listed inside this, which is, for most of us, we don't know what that figure would even kind of compare to, right?

Like, is it like, you know 'giga watts', 'gigawatts,' 'kagoolies,' like, what does that even mean? For context, Ireland, the entire country, used 34 terawatt hours of power in the same year as this reporting. So Amazon is basically, is using more power than Ireland, but they also in the Carbon Disclosure Project, they also list how much renewable energy they've used and therefore have purchased. And they've purchased 35 terawatt hours of clean energy. So that's, they've bought more clean energy on an annual basis than Ireland or the country has used in the same year, which is mind-blowing as a figure. And it kind of gives you an idea of the scale we're talking about here, for example. Brendan, you got something to come in there.

Brendan Kamp: Yeah, yeah. So I just have a question and I'd love to hear your opinion on this around carbon contracts, right? So my understanding of carbon contracts is you're buying energy that would be used by another company or green energy. And you're then saying, "all right, that is now my green energy." So it's not necessarily a reduction technique.

It's more of a auditing technique in my mindset.

Chris Adams: I see why you would say that. And to be honest, the market structure is such that it's actually quite complicated and not very accessible for most of us. All right. I think what you may be referring to are two separate things.

So if you want to report electricity usage and the carbon from electricity usage, there is a approach which is based on your usage tied to your location,

right? And there's another approach which is designed to recognize the fact that you've Bought a bunch of renewable energy. So one is called the market-based approach and one is called the location-based approach for this. And one of the challenges that we've had over the last 10 years is that you can basically buy power and separately from your use of that power, you can just buy some certificates that have been sold by a bunch of green energy producers.

Because it may be that in certain parts of the world, people don't value those certificates, and then you can then apply that to your usage to then say, "well, this is quote unquote Green Power." This has been the approach that has been used. And Ireland is a particularly interesting case because in Ireland, you've had the Advertising Standards Authority basically say, "yes, this may be how your market works, but it's still misleading to consumers. So therefore, we will not accept, you know, we're going to stop Irish energy providers selling green energy that uses this certificates based approach." Now, the thing that's quite interesting with Amazon is that there's a project inside the Green Software Foundation called the Real-Time Cloud Project, which Gabi, I mentioned before. We've actually got someone who's been literally counting up all of the output from every single project to compare this to what the figures might actually be, so that you can get some idea of what the physical production might be versus the amount of credits that might need to be purchased. And this is, we've actually, we can also link to another podcast where we talk about this in terms of additional, deliverable, and timely as ways of talking about green energy.

Where if you're going to actually have, if you're going to make a claim around green energy, you need to be able to basically address these things. We'll link to the podcast episode which dives into this in more detail, but the short version is if something's additional, like you said, it's not reallocating green energy, it's introducing new energy into the grid.

And this is one thing that some of the large providers tend to do now, they make a big story about, "Hey, look at this new wind farm that we've helped create." And that is usually financed through a power purchase agreement for where people do that. So that's kind of the additional part. Now, deliverable is another thing.

So deliverable in this context might be, "I financed a renewable energy project in Norway, but my use is in Spain," right? Which is the other end of Europe. Now, it's not very plausible that the power would go all the way across Europe without any losses to then power your data center in Spain, for example.

So that's deliverable. And then timely is basically saying, "well, I funded a bunch of solar panels, so therefore I'm allowed to say my servers running at night are green." Now again, that is not necessarily the most plausible claim that you would make. So this is why you need a combination of all these three.

And this is one thing that I believe Nina from Energy Tag in the episode we'll link to explains in a lot more detail. Anyway, going back to the original story. Yes, it's really, really good that you see a company like this saying "yes, we're the largest provider," but they're just one of the largest users of energy, full stop.

So just by dint of the size, that they're going to be one of the largest. And when you look at the amount of energy used versus their revenue, I mean, I've shared this, actually, this research, we can see if you assume a figure of 100 Euros per megawatt hour, 100 Dollars per megawatt hour used, which is a pretty generous figure for data centers. It's, what, 0.7 percent of Amazon's revenue, and it's less than a percent for Google's revenue, less than a percent for Microsoft's revenue, so this is actually a good thing, but this is often a more statement of how large these companies are, rather than how much of their money and their revenue they're dedicating to basically green energy.

And that also tells us that we could probably be moving way faster. So that's my, as a executive director of a nonprofit, talking about this, saying we should probably be doing this more, especially when you look at how much money has gone to basically buying your own shares to make your share price go up rather than make sure that we have a functioning society and habitable earth.

Brendan Kamp: So do you foresee any of the new regulations coming in increasing the percentage of revenue that they'll be spending on renewables?

Chris Adams: This is an interesting question because typically when you look at it historically, you've seen that a significant part of the funding for renewable energy has come from various kinds of subsidies to make it easier for companies to get on board. Like, companies will basically say, I'm not going to do this until it makes sense to me on a financial basis. And if you're a hyperscale provider, or if you're just a large user of energy in total, the reason you would have a PPA is, or you would use renewable energy, is yes, it makes you look awesome and looks like you care, and it's good, but it's also the cheapest way to buy power, right? So it's almost willful to choose to buy fossil-based power in lots of places, if you can get a contract like this.

However, there are complications about making sure that the power is both timely, deliverable, and additional, like we've mentioned, but these numbers, the figure of a hundred dollars per megawatt hour, this was something that I ran by some of the energy modelers who we interviewed before, and they said, "this figure is, yeah, that's not an unreasonable figure to use. You could do that."

So if each of these companies doubled their... the amount they're dedicating, they could absolutely all go to 24/7 fossil-free very, very quickly. And there are smaller companies that are already doing this, who are already at 99%. So we'll add some links to that, but I better move on to the next story because this is something that I could talk about literally for hours. All right then, so the next story, this was one I think that you shared actually. Tripling the lifespan of servers. Why we retrofitted 14,000 servers. This is a story from Scaleway, the European cloud provider.

And Brendan, is it okay if I just hand over to you for this, to explain this one? Because you shared this, and I think it's actually quite interesting to see a company talk about this.

Brendan Kamp: Well, this was just something that, you know, pops up on your feed and caught my attention, obviously with talking about embodied carbon, quite a lot recently, the lifespan of servers is very much at the forefront and Gabs, correct me if I'm wrong, what do we estimate in the lifespan of servers? Is it six years?

Gabi Beyer: Think it's four. Maybe by dell.

Brendan Kamp: Four years, yeah, so like a four year server. And what Scaleway has done is they actually went in to their data center and did some, you know, a bit of analysis on what parts of servers were failing. And it turned out in this case, that majority of them were hardware components, the RAID controllers, and in the RAID controllers, the part that was failing was the battery.

Now, RAID controllers are slightly more old-school components of servers. Nowadays, you have software based RAID. And so what they did was they thought, okay, we can actually remove this components and potentially that could prolong the life of these servers and just switch them over to a software based RAID.

Now, don't ask me what RAID is. I'll probably get it completely wrong, but.

Chris Adams: a Redundant Array of something Drives,

is what it is.

Brendan Kamp: It's, it's, it's, a hardware thing.

Chris Adams: Yeah.

Brendan Kamp: But just the understanding that they invested all the time and money to go and do, I think they did 24 or 25 POCs, just proving that this could work. Then they went in and took out 14,000 of their servers, moved them to a new data center, updated them with these new controllers, and put them in, and carried on running them. And they had no reduction of quality and service, right? So the servers performed as well, if not better, once they'd done this. The hyperscale is also starting to invest in these ideologies a bit.

I mean, the amount of servers we can save, the amount of embodied emissions we could like reduce by, it would impact server companies quite considerably, but I mean, something has to give at the end of the day. And another thing that they also did mention, which I found fascinating was because we currently have this big chip shortage due to, you know,

Chris Adams: Supply chain.

Yeah.

Brendan Kamp: Supply chain, etc. The servers that were redundant, that they couldn't migrate to this new setup, they stripped for parts, and they're using those parts to maintain the servers that they updated. So they've kind of really tried to look into every single facet of these servers and kind of see how much longer that they can make them last.

And I mean, yeah, from four, as Gabs said, to 10 years now, I mean, that greatly reduces the percentage of the embodied carbon of those servers.

Chris Adams: Yeah, that is, on that kind of scale, that's a meaningful thing to look for.

I guess we should probably be asking for these figures when we purchase cloud services. Ask, "can you tell us about your embodied carbon for your service as well in that case?" Huh.

Brendan Kamp: Yeah. Transparency, right? I mean, this is the key for the cloud providers. I feel at the moment, it's just become more transparent about some of the, this data so that we can do the work to calculate, but just give us the data to do it. Right?

Chris Adams: Gabi, I see you nodding along like this is a thing that you've, you might have some scar tissue developed over the months trying to wrestle with.

Gabi Beyer: Just a little bit of blood, sweat, and tears.

Nothing, nothing I can't handle.

Chris Adams: Okay. So you heard it here first, please share these numbers if you operate some kind of cloud service or you're providing the infrastructure that people run on. Maybe it's worth asking the Leafcloud folks from up there, because they might have some of these numbers.

Alright then.

Gabi Beyer: That's a good idea.

Chris Adams: Folks, I've really enjoyed chatting with you and I've realized that we've almost hit the hour that we have allocated and I haven't got around to rounding this up and basically plugging this big event that's coming up.

So on the subject specifically of transparency, the Green Software Foundation runs a kind of annual hack day or hack event basically. Rather than trying to compress building anything into 24 hours of Red Bull, chocolate, and delusion, it's actually spread over a longer period of time. So this is actually a hackathon kind of thing that starts on the 18th of March and runs until the 8th of April.

So that means that as people who might be involved, we do get to sleep and eat and see our families and like, live full lives as people rather than just machines that crank out code. And this year, the focus is around a new initiative called the Impact Framework, which is essentially a bit like a kind of manifest file, or like, you can think of like a Terraform state thing for your entire system that enumerates through all of the figures for like carbon emissions or something like that.

It's designed to be kind of open and flexible. And there's a hackathon all about that at hack.greensoftware.foundation. And there's some specific awards around talking about measuring things beyond carbon, so that might be useful for some of the things we touched on today. Creating the best model so that if someone wants to figure out the emissions of using a service, like say the carbon cloud thing, that might be an example.

The best content for explaining why this matters. There's the best contribution to the framework itself, because it's an open source framework written mostly in TypeScript. And finally, because there's loads of undergraduates who will be alive longer than we will be and have to wrestle with this, the best undergraduate project as well.

So people can register interest there. And I think that's me done with my contractual obligations for this podcast, I suppose. Folks, I really enjoyed chatting with you. This has been loads and loads of fun. Thank you so much for coming on for the day.

Gabi Beyer: It's been really nice.

Brendan Kamp: It was great. Thank you.

Gabi Beyer: Yeah,

Chris Adams: Before we go, I just want to check if people were interested in the work that you do or wanted to follow up, where would I direct people's attention? Is it like, is there a place, a LinkedIn or a blog that you'd like to talk about? If I ask Brendan once again first, because it's alphabetical, and then I'll ask you, Gabi, if there's anything you would point people to, then I think we'll just wrap up after that, actually.

Brendan Kamp: Yeah, I mean, my LinkedIn's quite active. I'm terrible at Twitter, so it's just brendan-kamp-757 I think is the end. And also we do write quite a lot of blogs, so I've got blog.thegreencoder.io and also blog.re-cinq.com, where you'll find a lot of our blogs and things. And Gabs, code-wise?

Gabi Beyer: Probably LinkedIn as well. Maybe, yeah you can look at our GitHub also on re:cinq and Cloud Carbon and, yeah.

Chris Adams: Go from there.

All right then. Well, thank you so much for that, you two. As we mentioned, I'll put a markdown version of the transcript and the show links that we have. And if anyone wants to be mentioned with a PR for anything we fix or a typo, then here's your chance for a tiny two seconds of fame in the next episode. All right. Thanks, folks, and have a lovely day. All right. Ta ra!

Brendan Kamp: Ciao.

Gabi Beyer: Bye.

Chris Adams: Hey everyone, thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

The Week in Green Software: CNCF TAG Environmental Sustainability

25 januari 2024 | 45 min

Join TWiGS host Chris Adams in talking to Kristina Devochko, tech lead of the Environmental Sustainability TAG at the Cloud Native Computing Foundation. Kristina shares her journey from economics to tech sustainability, and eventually joining this Technical Advisory Group. Further, they discuss the mission and projects of the group, as well as how anyone interested and willing is able to contribute. She elaborates on her experience of diving into this new field with no prior knowledge and acts as a reminder that no matter how scary it seems, you can do it too.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Kristina Devochko: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

TAG Environmental Sustainability [09:56]
The Week in Green Software: Carbon Aware Spatial Shifting [15:13]
Kepler | [16:07]
GSF Oslo Meetup group [34:08]

Events:

CNCF Cloud Native Sustainability Week (2023) [14:43]
Green Software Foundation Oslo - February Meetup [34:31]
Carbon Hack 24 | Online | 26th Feb to 15th March [39:02]

Resources:

Cloud Native Computing Foundation [01:53]
Kristina’s cat, Penelope [02:51]
Kristina’s cat, Sofie [02:51]
TAG ENV | Linktree [09:29]
Kepler GitHub repository | CNCF
Exploring Kepler’s potentials: unveiling cloud application power consumption | CNCF [23:23]
Environmental Sustainability Glossary [24:33]
Green Reviews WG task board[24:45]
Green Reviews WG project repository [26:45]
[WG Green Reviews] Design Document
TAG ENV Slack Channel [31:31]
GitHub - cncf/tag-env-sustainability: 🌳🌍♻️ TAG Environmental Sustainability [31:44]
Green Software Champions | Speakers [35:38]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Kristina Devochko: You know, there weren't many there at that meeting, but they were so passionately discussing the different topics around the topic of sustainability in tech, specifically in the cloud native technologies. And I really loved the passion and interest from those who were there.

I'm your host, Chris Adams.

Hello, and welcome to another episode of Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. Well, we talk a lot about green software on this podcast. If you've listened to some of these episodes, you'll realize that cloud is a really, really, really big part of green software now.

And last year, the Cloud Native Computing Foundation set up a sustainability focused working group. And given that it's the new year, 2024, it seemed worth checking in on it to see what's new. So today, I'm joined by Kristina Devochko, who's part of the CNCF and some of these groups. Kristina, thank you very much for joining us on the course.

Should I give you a bit of space to introduce yourself and talk about where you're coming from today?

Kristina Devochko: Yes, of course. Hello Chris and hello everyone listening into this episode. Thanks for inviting me. My name is Kristina Devochko. I am a platform engineer at Tietoevry and I'm based in Oslo, Norway, and I am very excited to be here. I am also a Microsoft MVP and a CNCF ambassador. And what comes out of that is that I do a bunch of different activities in the tech community that are mainly related to the topics of cloud computing, green tech, Kubernetes. And one of such activities is my involvement in the CNCF Technical Advisory Group for environmental sustainability, which kind of also correlates with, with a bit of my personal passion for this domain.

And a fun fact about me is that I am what you can call a cat mom at heart. And actually yesterday, right before we joined today with you, Chris, for this podcast, I, me and my husband welcomed two new family members, two adopted kittens that I will run to after we are done with this episode and kind of keep, continue training them to feel safe in their new home.

Chris Adams: Wow, I was not expecting cat pics so early in 2024, but it sounds like we might need to have some reference to this, because you can't talk about cats without sending pictures of cats. So maybe that's something we add to the show notes.

Kristina Devochko: Yeah, that's, by the way, one thing. I love adding the photos, fun photos of my cats into all the technical content I make. So if you would like, we could totally do that.

Chris Adams: All right, if anyone who's not watching for this, I have to say, while I'm using like a basic headphone, Kristina has an awesome set of headphones with glowing cat ears on this. So this may be a running theme for the rest of 2024. All right, so if you're new to this podcast, my name is Chris Adams. I work at the Green Web Foundation, which is a Dutch nonprofit focused on reaching a fossil-free internet by 2030.

I also work in the policy working group inside the Green Software Foundation, where we do work on policies and respond to future coming legislation and things like that. But for the purposes of this show, what I can share with you is that while we talk about this, and we're going to mention various projects, we do our best to share a transcript and set of show notes for all of these things.

And because it's New Year, we're going to be trying something new. We're going to be posting a transcript of this plus all the links onto GitHub. So if there's something that you're curious or you want to learn more about or we've got something wrong, we'll be accepting pull requests for the transcript if there's anything that you really were curious about.

And yeah, we'll be sharing that on podcast.greensoftware.Foundation as per usual. And I think with that, should we begin then, actually, Kristina?

Kristina Devochko: Yeah, let's do it.

Chris Adams: All right. Okay. So, Kristina, you mentioned before that you're calling from Oslo, and I no longer live in England, but I grew up in England, so old habits die hard, and we always talk about the weather.

And I'm calling you from Berlin today, which is coated with snow. So I should actually ask, how is the weather in Oslo today?

Wow.

Kristina Devochko: Oh, that's a very relevant question because it's quite freezing, actually. And like, personally, I love winter and I love snow. And we were lucky enough this year to have a truly white winter and Christmas and the New Year's Eve. But I need to say that this year we had like a few extreme cold waves happening in Norway.

And I was recently reading an article, I live just outside of Oslo and like, right after the New Year's, in the beginning of January, there was a new record that was hit in the Oslo municipality, where it was registered to be minus 31.1 degrees Celsius, and that's like, the previous record was from 2011, and it was 28.8 degrees, I think. And today I had an appointment at 9am before this podcast recording. And when I went outside, I saw that the temperature was minus 20 degrees. So that's kind of when you feel that your nose is full of needles at that point. So I love winter, but not when it's that freezing. But I'm here now in a warm room and happy to be joining you for this podcast. So...

Chris Adams: Well, I'm glad. I'm glad we're both in somewhat warmer temperatures than minus 31 degrees Celsius. All right. Okay. Fun fact for this, if people follow along, there is, if you have any kind of carbon aware websites today, the coldness is actually changing some of the carbon intensity for some websites. So there's a website called Branch that me and some friends maintain.

It actually broke the code because it was so cold and so not windy in the UK for a certain time that the carbon intensity went higher than we've ever seen it. So this is an example of how some of this can manifest in software itself. All right, so we'll get to that a little bit later on, but I should basically, before we dive into this, it's worth asking, you now work in green cloud computing and that's like your new specialization.

But that hasn't always been the case. So if I understood it, you did, you worked as, you trained as an economist first before you came to working with cloud and development, and then you changed directions into this field. Is that the case?

Kristina Devochko: Yeah, yeah, that's true. I mean, I never, I never thought of working in tech as being my kind of career. I wasn't born and raised with like all this programming knowledge from the start. So I came from a culture where it was natural for women to do something else than computer science. So for me, in my head, it was natural to do something like economics.

And that's what I started with. I got accepted to the bachelor's program at the university. And then I figured out that it was so boring. The first year I started, it just didn't work out. I couldn't find the motivation. And I had a friend who was studying computer science, but more into the NLP domain, into natural language processing.

And she kind of challenged me because kind of the atmosphere there was good and people were nice. And it was interesting. So I just took a chance and applied and got in. And from that point on, there was kind of no way back. So I started working with as a full stack developer and got really that spark after studies.

And for two years ago, I think there was someone from the community who challenged me, like, why don't you start contributing to the tech community as well? There is so much you can do there if you really love working with tech also outside your full time job. So I actually started doing that, and from that point on, it's like, I've tried a lot of different things, and I find it being a lot of fun, so it's for me like more, it has become more than a full time job, a source of income, it has become a hobby, and a sort of passion, and like if, since we are talking more about the sustainability in this podcast, personally, I've been really focused on that area, like in my life, in my everyday life, living a more eco-friendly, planet-friendly lifestyle. And from that, it kind of naturally transitioned into also putting focus on that at work, in my company. And from there, I ended up joining the Environmental Sustainability Technical Advisory Group. So it was like a very interesting journey starting, now that I look back at it, starting with economics and thinking that this is my only way to kind of take my career and then like ending up in a totally different space.

That's, yeah, that is somewhat, somehow that was a bit of magic for me.

Chris Adams: Okay, cool. Thank you for that. So there's from an academic transition to I guess an energy transition really. And as I understand it, so in the last year you got involved because you mentioned the Cloud Native Computing Foundation Technical Advisory Group for Environmental Sustainability. That's quite a mouthful.

I mean, I have to ask, is there a short name for it or is there something I should use instead when I'm using that? Because I don't think it rolls quite off the tongue and I'm sure there's a faster way I can refer to that, right?

Kristina Devochko: Oh, yeah, totally. I know. That's like, that's a bit of a wording there because you need to kind of also showcase where this technical advisory group come from. So, like, we call it the TAG ENV, like TAG ENV. So TAG for technical advisory group and ENV is That's kind of our abbreviation for environmental sustainability.

So we just call it TAG ENV or CNCF TAG ENV. So that's what we also use in our writing, writing content.

Chris Adams: TAG ENV is easy. TAG ENV I'm going to use from now on if I'm going to refer to this, because that's much, much faster, and it's kind of fun to say. All right, okay. So I should have to ask then. There is now a TAG that's focused on ENV, right? And maybe you could tell me a little bit about actually what it does, and what kind of work or what kind of projects might be associated with that, for example.

Or maybe even how you got involved in it, actually.

Kristina Devochko: Yeah, of course. So, CNCF has, and CNCF, the Cloud Native Computing Foundation, so I'll call it CNCF. It has a specific structure as an organization, so it has different groups in order to ensure that the different projects and users and organizations that are supporting the foundation can kind of develop and improve in a sustainable, expandable manner. So the, therefore there is a concept called Technical Advisory Groups. And this is basically a group that is led by people that are community members that are specialized in that specific domain that the group covers. And then anyone from the community can join those groups depending on their area of interest and participate in the meetings, participate by contributing to the different projects and activities that are going on in those, in those TAGs, which is the short version of the name. And there are many different areas. There is security, observability, there may be networking and like for, I think now it's two years ago, like one and a half years ago, the Environmental Sustainability TAG was founded by some of the community members. And at first we didn't hear about it. Like I personally haven't heard about it until spring last year, because I was starting to prepare more content on the topics of green tech. And I got in as a speaker at the KubeCon and CloudNativeCon Europe conference, which was in Amsterdam in April last year. And I start, and I was doing my research, preparing for the talk, and I, by chance, found this TAG, this group, and I was like, wow, okay, CNCF has a dedicated group now, but it, it didn't seem very active because it was just starting out. And then I, of course, highlighted it also during my talk and also saw on the schedule during the conference that they had the pro, a group meeting. So I just joined that, actually knowing nothing about the community in that group. And that was really for me as a beginner, a bit scary, but at the same time, when I joined, I was very warmly welcomed and I loved the engagement, you know? There weren't many there at that meeting, but they were so passionately discussing the different topics around the topic of sustainability in tech, specifically in the cloud native technologies. And I really loved the passion and interest from those who were there that I thought, "wow, cool. This seems like people really care about it, not just do it for, you know, to check, have a checkmark somewhere." Yeah, And I started just after that, I came back home and I started joining the regular meetings and just contributing with small steps, just doing some minor changes, asking questions.

And now it's less than a year actually. And now right before Christmas, I got appointed to be a tech lead for the technical advisory group. And I'm really happy about that because I've been putting a lot of love into supporting the TAG and we have grown and it's really great to see more and more people joining us and also joining on the activities that we are doing. And kind of the main mission of this group is to raise focus and raise awareness about the importance of sustainability in the cloud native landscape.

Like what is the current state of that?

What kind of challenges we have and how can we help both the CNCF landscape, the projects there, but also the companies that are utilizing those projects that are running in cloud, be it public cloud or private cloud, are running in a cloud native manner, how can we help them get the sustainability of their systems and software better, basically?

Chris Adams: Okay. I'm somewhat aware of the TAG ENV'S work. I remember that, I think, last week, there was this, last, towards the end of last year, there was a sustainability week with a bunch of online talks by various people presenting, yeah, and we'll share a link to that, and I know there's a couple of other projects that we're going to talk about in a bit more detail, and you've mentioned this idea of, well, what is a state of the art, how do you make some of this visible to people, and you name checked some things like observability.

Are there any kind of particular projects that you're particularly excited about, or you might be seeing kind of repurposed, because you, I believe you mentioned one project, and I think in July we had Niki and, Niki and Ross, who were also part of the TAG at the time, talking about some of the work in making carbon emissions more, more, more observable and visible to people.

Maybe you could talk a little bit about that, because I think that's one thing that you've mentioned to me that you're quite excited about.

Kristina Devochko: Yeah, absolutely. And Niki and Ross, it's great that they have been on the podcast because I know that they have been talking also about Kepler, which I also wanted to highlight in this episode as well. And they are also driving now the kind of subgroup in our Environmental Sustainability TAG that also has a separate project, which I also would like to talk a bit more about afterwards.

But like the first CNCF project that I would like to highlight that kind of touches upon the observability piece in context of measuring, like, the sustainability of your, of your system. And that is Kepler. And this, this project has now, for a few months ago, I think, entered the CNCF landscape as a sandbox project, so that's kind of a big deal for the whole project and everyone who was supporting it. And for those of you who are listening in and haven't heard about Kepler, it's the short name for Kubernetes-based Efficient Power Level Exporter.

So that. sounds like a very complex name, but that actually makes meaning. So that's a word with the meaning behind it. So this project, its main goal is to measure energy consumption so that this data can be gathered, collected, and afterwards you could feed it to the machine learning models that can perform a specific analysis on that power usage utilization data. And then you could visualize that. And by visualizing that over time, you could also understand how efficiently your resources, your power is being used and what kind of optimizations you can make based on that. So you could kind of...

This could help you make more informed choices on, in terms of optimizing your resource usage.

Chris Adams: So there's a couple of things I just want, so I just want to check if I, check my understanding and also for other listeners on this. So there's a tool called Kepler, and because measuring power usage inside the cloud can be quite challenging at times, this is essentially an open source tool that goes some of the way to looking to making that more accessible and observable to people who are responsible for, like, operating clusters of computers and things like that.

And, if I understand it correctly, it's using what was the kind of hip and trendy software, like, Extended Berkeley's Packet Filters. It's using something quite close to the metal, so it doesn't actually slow down the system to report information as much as other things. Or, at least, that's one of the considerations that people have been taking some steps for.

Is that correct?

Kristina Devochko: Yeah, you're correct. And I think an important piece to mention here, like coming on to your point that it's not that easy, necessarily, to gather this type of metrics for power consumption. Like if you think about bare metal, that type of data can be more, in a more easier manner gathered because it is available directly from the hardware when you're running on bare metal, but when it comes to cloud providers, or like when you're running on virtual machines, this, there is no support in virtual machines to be able to make these types of metrics available.

So there, that's why it may be challenging for example to get 100 percent accuracy of running Kepler on, on managed Kubernetes service. And to just make a notion of that, Kepler is mainly targeting the Kubernetes-specific workload. So when you're running Kubernetes clusters, be it on bare metal or be it in, as a managed Kubernetes service, for example, in public cloud. And like you mentioned, it also uses BPF to gather some of that specific metrics specifically that would be specifically useful in, in public cloud. But we would link a blog post as well that I would recommend to check out because it goes in more very technical details around the different deployment models and architecture for Kepler because they have, they are currently working on a specific deployment model that I really hope we will get to see public cloud providers collaborating with Kepler project on, which would potentially allow to not use a pre-trained model with some kind of pre, pre tested, yeah.

Pre, there is like a data set that was based on some of the other data that the Kepler community has collected, for example, from running on bare metal.

And it does make some assumptions based from that. So it provides some pre-trained power models that you could use in public cloud to get somewhat close to reality type of data, but still, for example, things like idle power, you will not get that information or that metrics from Kepler because there is no way for the tool knowing about how many VMs are running on the host when it comes to public cloud.

Chris Adams: Oh, I see.

Kristina Devochko: There, that's kind of the, there are some limitations there and this blog post goes very nicely into those details. But the deployment mode that they're working on, that would allow you to install Kepler on the host first, that public cloud providers could do. They would install Kepler on the host first, that would gather that hardware specific metrics for power consumption.

And then you would kind of have a layer.

Chris Adams: The visibility at that point, rather than like, you're inside the virtual machine guessing how many other neighbors are also using your physical machine, you're actually getting the physical machine saying, "yeah, you actually have 10 neighbors and six of them are mining Bitcoin."

Or they wouldn't be doing that. But like, you get the general idea. I'm not sure if you're allowed to do that in some, in a lot of clouds. Okay. I'm glad you mentioned that.

Kristina Devochko: Hmm. Yeah. So that would require some collaboration with public cloud providers since they have control of the hardware. But then in that case, you would kind of install Kepler twice. You would install it on the host and you would install it in the Kubernetes cluster. But then by doing that, cloud providers, maybe by making Kepler data from their hosts available as an add-on on, yeah.

Chris Adams: Service you could then subscribe to.

Kristina Devochko: So you could get that data. Yeah.

So you could get that data and you could get an even more, more accuracy on that power consumption metric. So I really hope that the adoption of the project takes off and we get to see it available in the managed Kubernetes services in public cloud, for example. So i, it's, it looks really promising, to say least.

Chris Adams: I'm glad you mention that, because there's one project inside the GSF, this is the Green Software Foundation for other people, called the Realtime Cloud Project, which is currently led by, amongst others, Adrian Cockcroft, formerly VP of AWS, and there's people from Microsoft and also Amazon who are working to essentially do the carbon part if, if things like Kepler are actually exposing some of this.

And for the nerds who are on this, listening on this podcast, I'm going to ask for your help. Because one of the reasons that's cited for not sharing this information is there's a very particular kind of attack that you can use if you have direct energy emissions being exposed at a very high resolution.

You can find out what programs are running or even some information like that. So there are reasons cited for not sharing this information. There's a whole line between zero knowledge and letting everyone get hacked and like the, the research that we've seen or that's been shared inside there is that around the minute resolution, that's still enough for operators like yourself to kind of like tune and manage carbon, but also is something that still mitigates against most of these attacks.

And I do forget the name of the animal that the attack is named after, but if you're listening to this and you submit a PR, I'll be very grateful for this because it's got a cool name. It really does. But I'm totally blanking on this. If you're curious about this as well, we did actually do a podcast in July last year with both Ross Fairbanks of Flat Peak and Niki...

I can never pronounce her surname. Niki, I'm so sorry. Niki, she works at Grafana and, Manoladeki, I believe. So that's the people. And we'll share a link to that podcast, which dives into a bit more detail about some of that. Okay, so we spoke about that. And that's one of the wider projects that has some application clearly inside this.

But there's maybe some projects, there's a, well, there's one or two projects inside the TAG ENV that you're also involved in, that you said that you'd like to kind of talk a little about. Maybe if we give you a bit of space to mention one of those, perhaps.

Kristina Devochko: Yeah, of course. So like Kepler and other projects that are coming out that are related to sustainability, we support them as a technical advisory group. But we also are looking for ways we can also contribute by starting some of the projects inside the TAG. And we have a few like content focus projects, like the landscape document that Kind of covers the state of sustainability in the cloud native space that we are working right now on creating a version two for. And we have recently published the glossary for cloud native sustainability glossary that could help clarify some of those specific terms. But when it comes to more technical contributions, I wanted to highlight the project that we are working on in the subgroup of the TAG that is called Green Reviews, so it's a working group that is run by Nicky, Antonio, Antonio DeRosso, which is one of our TAG contributors and now is his co chair in the group, and Ross Fairbanks as well, that is the technical lead for that group. And what we are working on in that group is that we are building a workflow that can be used to target the projects in the CNCF landscape, like Kepler for like, not like Kepler, but the other projects that you would be using, for example, Kubernetes, Kubernetes, I think would be a better suggestion in this case, because the workflow actually uses Kepler together, some of the, of those metrics.

So this workflow would then be connected towards those CNCF projects that will be running under load and so that we could simulate a load on kind of the application itself so that we could gather the sustainability specific power and resource utilization specific metrics over time so that we and that workflow could analyze the data and publish information about the sustainability posture of that project.

And this statistics can then be publicly available for all the projects in the CNCF landscape so that over time, the project maintainers and project contributors could look into the states of their projects from the sustainability side and improve the resource utilization of those projects over time, make them more resource efficient, which also is one of the goals for Kepler projects, for example.

Find those bottlenecks that, find those ways to improve and make those projects more lightweight and more resource efficient. So this is also something that is open for anyone to contribute. So if you like to code, you could just join and pick an issue and, you know, start helping out.

So if you would like Chris, you're also welcome to join.

Chris Adams: Maybe some, there may be some things we look into a little bit later on. So, if I understand that correctly, there's a couple of things you mentioned, and you use this term sustainability posture of a project. I'm taking that to mean like, what the kind of profile might be like, where, under what, so how it might perform under certain circumstances and what in certain places it's, is it more efficient here or less efficient there, is that what you mean when you say that?

So, the idea, you get an idea of where there might be common areas for improvement across the whole portfolio of projects, for example, based on some of the real world data. Is that the thinking behind it?

Kristina Devochko: Yeah, we are like still in very early phases. So this working group has been like a few months old. So the piece, for example, of what kind of methodologies we would use to calculate that, that kind of sustainability posture, how efficient a project is, we have still not 100 percent defined that. So I think we want to start like simple by just gathering some of the Metrics like the SRE specific metrics, you know, like CPU, memory usage, maybe use, use the Kepler specific power consumption metrics.

Start there. And then the next step would be to see what other metrics we would need, or those metrics that we have collected by now, should we use as, the Software Carbon Intensity Specification, should we use that in combination with something else? Should we use something like Impact Framework Engine, right?

From GSF. So we have been looking into that. There are some discussions around if we should use, for example, SCI as a Software Carbon Intensity methodology.

Chris Adams: That's one way of measuring the basic, it's a bit like, a little bit like PUE for data centers. It's like the amount of emissions associated, a kind of rate, as a software. Okay.

Kristina Devochko: So that we could provide some of that information, like you say, that we could provide, like how much power has been used, if we could correlate that to the amount of carbon emissions, maybe water consumption at some point, because this is also becoming very important these days with enhancements in artificial intelligence and machine learning.

So we're still figuring that out. So I don't have an exact answer for how exactly it will look like, but this is an opportunity for anyone also from... Yeah.

Chris Adams: Refer to some of the existing state of the art. Okay, that's cool. For folks who were interested in listening along to this, so, Kristina, what you've actually described might make you think of an episode that we did with Arne from Green Coding Berlin. We've interviewed him as well.

He spoke about some of this in a bit more detail. We should share a link to that because he dives into very deep amounts of data for this and also his organization has been sharing some really interesting stuff about the difficulties of tracking some of this stuff on the cloud. Because, you know, one of the, while the cloud can be extremely convenient, there are knock-on effects on basically having an amorphous blob of compute that you kind of tap into, rather than having a direct machine that you can direct measure directly.

All right, great, and I'm really glad you mentioned the glossary and some of the other things you had, because one thing that I found really helpful when helping other people actually to talk about something for the first time is having this stuff saying, "well, this means this," or even just having the landscape with you seeing that, well, "wow, there's actually a really wide range of projects where the sustainability aspects do touch on a bunch of these things."

So, thank you for mentioning those, actually. I should probably ask you now on relating to that. We spoke about, like, TAG ENV. And we spoke about some of the projects that are ongoing, but it might be useful to talk a little bit about how you folks organize and how people can get involved, because I was able to join the Slack and chat to a few people to kind of more get a kind of feel for it, but I figure it might be useful for other people to realize that it is actually quite easy, and this is something that you've invested a bit of time in to make accessible to people, and well, you're here, so I should, I guess I should ask you really.

Kristina Devochko: Yeah, I appreciate you asking that because we are always looking for ways to communicate to the other community members that it is not that scary and difficult as it may seem to start contributing to some of these groups and organizations. So in the TAG, it's pretty, I think it's quite simply structured.

So of course we have some of the chairs of the TAG, and we have also tech leads that kind of overview the different types of activities in the TAG and in subgroups are done in accordance with our mission and our goals and contribute in that area. But it is open for anyone to join. So we don't have necessarily like a membership or anything. So everyone can just join our slack our regular meetings we have both meetings for the TAG in itself, which are bi-weekly, and the working groups-specific meetings as well. And anyone can just join and start asking questions, participating in discussions, you can go to our GitHub repository and in the issues filter on 'help wanted' or 'good first issue' labels.

And then you could see if there is something that resonates with you, or you could just start by contributing to the existing discussions in those issues. You could contribute content. So if there is a specific topic that you would like to write a blog post on when it comes to cloud native sustainability, you could just submit that, and then we will help you to get it there, to get it on our website, to get it on the CNCF blog. So there are quite a few ways you could just start. And even if, yeah, and even if you feel like, it feels a bit scary, you are very much welcome to reach out to me. We could probably share some contact info, or just go on the Slack and send me a message if you need some guidance on how to join us.

And we have a separate blog post that we recently wrote.

Step by step guide.

Chris Adams: Like how to take your... okay, well you heard it here first, you have a direct offer from one of the members to help you in if you're curious about doing this. This is one thing I actually quite admire about the CNCF. With the Green Software Foundation we, while it's a membership organization, it can feel a little bit difficult to get in, and this is one thing that I've been really impressed by.

And I'm aware that yes, there are all these weekly calls and there's things where it's very easy to see what's going on if you're remote, for example. But as I understand it, you do a fair few things in Oslo as well for like people who like to meet up and physically share the same physical space as other people sometimes, like doing meetups as well.

Maybe you might want to just briefly touch on that before, because I know this is one of the first things we have coming up in 2024 that you mentioned actually.

Kristina Devochko: Yeah, that's true. And like, like you mentioned with the Green Software Foundation, I, when I first started learning more about it, I was like, "Oh yeah, I would like to become a member." And then I realized that my company was not part of that yet. So it was not, not that easy, but still, you, you of course have some opportunities to just follow and participate in discussions.

But what I also liked when I saw that GSF started with local meetups, local meetup groups and supporting that initiative. And then when I checked, there was no one, no such meetup group in Norway. So it was open for anyone to just start one under the GSF umbrella. So in September last year, I just asked if I can do that.

And then we started a group, Green Software Foundation Oslo Meetup Group, basically, which will, I hope can become an arena for the, to grow and strengthen the local community of technologists who are passionate about the topics of green software and green coding in general. And we had one meetup last year to launch the group in October and hopefully can have it on quarterly basis.

So now the next meetup coming up is in, in February, and we'll have some nice topics about the impact of AI and machine learning on sustainability and also the adoption of Cloud Carbon Footprint, open source tool in one of the Norwegian companies called ODA, and I believe one of the recent episodes was precisely on that topic, wasn't it?

Chris Adams: No one admits to using it, but it's basically underpinning so much of all this stuff. Yeah. That's really, I'm glad, I didn't actually know that Oda was actually doing that kind of work. And it's quite nice to hear that, yeah, it shows up in various places, because I, I really enjoyed chatting to, chatting to the guys last, for the last episode, and nerding out about like Legend of Zelda, if nothing else.

That was of

the

call

Kristina Devochko: yeah.

Chris Adams: was some of it. Okay, so we've shared a link to the Oslo meetup group, so people can find out about that and when that's taking place. I believe that's open to anyone who's physically in Oslo. You don't need to be a member of anything like that, you just need to be interested along those lines.

I should probably, at this point here, just give people a heads up that if you're not in Oslo, and you're not on remote event, and you're maybe organizing events, it's worth knowing that there is actually a directory of potential speakers at speakers.greensoftware.foundation. This is one place where, if you're running an event, and you're looking for someone who can contribute some expertise, or is able to kind of help with it then, that's one resource.

And also, I believe that you've shared a link here, which is run from the CNCF, which basically has a listing of other events coming up as well. Is that the case?

Kristina Devochko: Yeah. On the website for the TAG ENV, we have a separate events section and currently we are working on adding, implementing some automation because we have the tool that was donated by some of the community members so that we could use to scrape the data about different sustainability related conferences, sessions, meetups, and then we could continuously update that events page.

So hopefully there will be coming out there automatically in not too long. So you could check that one out as well.

Chris Adams: That sounds cool. I can't believe in 2024 we still don't have a kind of, like, solved problem. Like, we haven't solved the problem of events yet, like finding out when they're coming up. Because this is something I really struggle with, and I'm glad that you have that mentioned. And I think you just said something kind of cryptic but interesting there. So there was a Cloud Native Sustainability Week in October, and I will happily admit on air that I basically plundered that place for all these videos and people to speak to for future shows.

And we've got one or two of the speakers who've been presenting there talking about some really deep dives. We've got, we've got some of those episodes lined up. You're planning one for 2024 as well. Is that likely to be around the same time, like October-ish?

Kristina Devochko: Yeah, I hope so. I, we will need to see if this is the date we should go for. We haven't decided on the date yet, but we have seen that it has been a success. It was the first time in 2023 that we decided to do it. So we were not sure how it would work out, but actually, despite the virtual mini conference that we had like a two hour event with some speakers presenting virtually, we also had like, I think we were around 20 plus local meetups that were, we calculated that it was happening in 17 countries across four continents.

So we think that it has been quite a success. So we hope we could do it on an annual basis. And now we have just started planning for the Cloud Native Sustainability Week in 2024. So if you or any of the listeners would like to contribute, it's totally open. And we have a tracking issue in the repository for that as well.

Yeah.

Chris Adams: you for that, Kristina. We'll share a link to the last one because there was really, I actually filed an issue in the TAG because I was like, "hi, I missed all the talks because I was in the wrong time zone." Or moreover, I might have been asleep at the time or something like that. But and then I think within two or three days, I had someone basically show like here's a list on YouTube of every single talk with links back to the slides and everything like that. It was so, so nice to see that. And there was some fantastic content. I was really impressed by that. So yeah.

Kristina Devochko: Yeah, it was really good, I agree.

Chris Adams: Oh, we're talking about events and I almost forgot. I should mention that there is, there's a quick announcement I kind of have to share as part of like me working in the GSF.

There is an event coming up later in this year called Carbon Hack 24. This is the Green Software Foundation's kind of global event, basically. Essentially, the idea is that it's a little bit like what Carbon Hack was like last year. So it's a week's long kind of hackathon kind of thing. This time, the focus is around a piece of software I believe you mentioned called the Impact Framework.

This is an idea which is a bit like something like a kind of manifest file or an executable manifest file for us, for people to kind of understand and measure the impacts of software. The idea would be that there's this piece of work that the GSF has been working on called Impact Framework, and the idea is a bit of a hackathon around something like that.

And at the, right now, there, we often talk about sustainability in terms of carbon only, but Kristina, you mentioned water as one, one possible dimension. And there's a number of other dimensions that we've actually spoken about as well, which are also fair game for exposing or looking into and things like that.

So that's one thing coming up. And I should say, it's online. It starts on the 18th of March and ends on the 8th of April. You don't need to stay awake for that entire time, surviving on Red Bull and chocolate. That's not a good idea. But the idea is that there's space to make it kind of accessible to people regardless of where you are in the world.

And if you go to 0hack.greensoftware.foundation, then people can see and learn all about that. And the only thing to bear in mind is that the registration is the 22nd of January. So there's a, it will need to move relatively quickly on that. Okay, there's a, there are a series of awards for that, that we'll be referring to.

And yep, I think that's the plug that I've done while we were talking about events. So thank you for bearing with me for that, Kristina.

Kristina...

Kristina Devochko: Of course, I'm very excited for that. I'm very excited for the hackathon and it's cool to see that it's not just a one day, one day, 24 hours type of activity.

Chris Adams: Here's my free idea that I'm unable, that I would like to work on, but I don't want to promise myself that I'll do it, because that's, I'm getting older, I'm not sure how much time I have, but I know there's a thing called Node-RED, I want there to be Node Green, to use some of the ideas behind Node-RED to create a visual representation of all of the inputs of data that you might have in, so you can work out some of this, because when we talk about this kind of stuff, it can be a bit hard to understand where information about carbon intensity might be coming from or where, like you've described utilization or how hard a machine is working is coming from.

And I think if there's a visual way to present that, that would be useful. And if there's absolute precedent of things like Node-RED being used in all these different places. So that's like kind of my lukewarm take, I suppose. Alright, Kristina. Thank you so much for spending the time to talk with us.

I really enjoy, appreciate you coming on and talking all about the TAG ENV and how people can get involved. And yeah, this has been really fun. Thank you for giving us your time and talking about this. This has been good.

Kristina Devochko: Thank you, Chris. It's been a great discussion. It's, it's been fun. So thank you for inviting me. It's been a pleasure. So I hope some of you would be listening in to this episode, would be interested in joining us and don't hesitate to reach out if you're not sure about anything.

Chris Adams: That's a really good point. Just before we close off, we should, if people are interested in the work that you're doing, either personally or the actual TAG ENV ,what's the best, what should people be typing into their search bars, for example, just to learn about your work or learn about the CNCF's environmental work?

Kristina Devochko: Yeah, I think that if you would like to get in touch with me, you could probably just type my name, Kristina Devochko, and on LinkedIn, normally I use most, mostly LinkedIn, and then I have a separate technical blog where I write some technical content on those different topics, kristhecodingunicorn.com.

And if you would like to join the Environmental Sustainability TAG, you could just, I hope we will add the links to the website and to the repo, but we are also at the CNCF Slack. So if you join that, you could search for a channel called TAG-environmental-sustainability, and we'll warmly welcome you there.

Chris Adams: Brilliant. Thank you so much. I'm also really glad you mentioned Kris, the Coding Unicorn, because although this is an audio podcast, you need to Imagine that behind Kris there is a neon sign saying, Kris, the coding unicorn. And that's, it's... That's all I've been able to see or look at when on this conversation, so I'm really glad you mentioned that.

Kristina Devochko: Almost like product placement.

Chris Adams: Yeah. Okay. Well Kristina, thank you very much for this and hopefully you stay warm and cozy for the rest of the day. Take care of yourself.

Kristina Devochko: Thank you, Chris. Thanks everyone.

Chris Adams: Okay, bye.

Kristina Devochko: Bye.

Chris Skipper: Hey everybody, this is Chris, the producer of Environment Variables, just butting in to tell you about an additional resource that we have for the upcoming Hackathon 2024. Join us every Monday at 2:30pm GMT for our regular live stream, which explains how the IMPACT framework works and shares hack project ideas.

You can visit hack.greensoftware.foundation to find out more about that and sign up for both the live stream and the Hackathon. That's all for now, see you next time.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

BETA Impact Framework

18 januari 2024 | 43 min

Asim Hussain is joined by guests Srini Rakhunathan and Navveen Balani, the technical leads on the Impact Framework. They delve into how this innovative tool effectively models, measures, simulates, and monitors the environmental impacts of software across various platforms. The conversation explores the framework's unique ability to handle diverse environments, from cloud systems to mobile devices, with an emphasis on the practicality and necessity of measuring software emissions accurately. Highlights include intriguing insights into integrating the Impact Framework with SCI Open Data and the future of green software development. This episode is packed with valuable information and thought-provoking discussions that offer a glimpse into the future of sustainable software.

Learn more about our people:

Asim Hussain: LinkedIn / Twitter
Navveen Balani: LinkedIn
Srini Rakhunathan: LinkedIn

Topics:

The Impact Framework | GSF [3:00]

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Events:

Carbon Hack 24 | Online | 26th Feb to 15th March [40:08]

Resources:

The Software Carbon Intensity Specification | GSF [3:44]
CarbonQL | CarbonQL [09:19]
Joseph Cook | Project Lead on IF [13:43]
Kubernetes [16:45]
Node.js [17:13]
Cloud Carbon Footprint | Thoughtworks [21:18]
Boavizta [21:32]
Azure Functions [23:29]
Cloud Instance Metadata Model [25:22]
Teads Curve | Benjamin Davy of Teads [26:12]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
That simplification that you can just download the code. You need to know a little bit of Node.js or Yarn or how to use it to just be able to run it. As long as you are able to do that, you're pretty much up and ready. Maybe we can even set up a timer, Asim, how long it takes for someone to run it. I bet it will not be more than five minutes.

Asim Hussain: Hello, and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host, Asim Hussain.

Welcome back to Environment Variables, the podcast that covers just about everything to do with sustainable software development. I'm your host, Asim Hussain. I'm the executive director of the Green Software Foundation here to navigate you through the evolving landscape of green software. In today's episode, we've got a special segment lined up to you, joining us are Srini Rakhunathan and Navveen Balani, the brilliant minds leading the project on a much anticipated Impact Framework from the Green Software Foundation. They're here to give us an insider's view of this revolutionary tool designed to model, measure, and simulate and monitor the environmental impacts of software. From cloud environments to your mobile devices, the Impact Framework is set to redefine how we understand and reduce the carbon footprint of our digital world. Before we dive in, let me introduce my esteemed guests and colleagues for this episode of Environment Variables. Navveen, how about, let's start off with you, please.

Navveen Balani: Yeah, thank you, Asim. Hello, everyone. I'm Navveen Balani. I'm Managing Director and Chief Technologist with Technology Sustainability Innovation Group at Accenture working on the intersection of technology and sustainability, and an active member of Green Software Foundation from its inception. Happy to be here and be part of this podcast.

Asim Hussain: Thank you, Navveen. And we also have Srini. Srini, do you want to give an introduction to yourself?

Srinivasan Rakhunathan: Definitely. Thanks, Asim. And wonderful to see you again, Navveen. I'm Srini. I'm part of Microsoft's cloud sustainability team, and I am a senior program manager. I work on sustainable hardware, and as part of Green Software Foundation, I got an opportunity to work on sustainable software with this amazing team.

So glad to be part of this call and happy to help.

Asim Hussain: Before we dive in, just a reminder that everything we talk about will be linked in the show notes below the episode. So I've got a couple of questions that maybe we can go through, but I actually thought, we were having a chat before that, I actually thought it might be interesting for everybody to understand the journey of Impact Framework, because that might actually help people understand what it even is.

I struggle to even explain. I think that's one of the, one of the things I want to chat to you all about is like, how do you even talk about Impact Framework to others? Because I, it can be so many different things to so many different people. It's a little bit challenging. But, yeah. I thought it might be interesting to talk about, you know, the early days, like years and years ago. And I remember, this is how I remember it. So I remember it as: we were all working on the Software Carbon Intensity Specification together, getting really deep and diving into a lot of that stuff, and then we knew we needed to start actually using the SCI and writing case studies. Srini, I believe you were the first person in the entire world to have written a case study on the SCI.

You did it with the MST, FTE shop on sample one. And that was a wonderful moment. I remember that. And then a little bit later, Navveen, you wrote like a very detailed one with kind of a larger case study with, I think, an Accenture use case. Yeah. I don't know. Let's just start off there. Like, what did you guys like, do you remember much about that time?

Like, what were you thinking? Like, what was, what were the challenges I'd say that you had when you were implementing and trying to write some of those case studies, do you remember? You were the first people in the world, almost, who would like actually bother to measure software.

Srinivasan Rakhunathan: You know, it almost seems like, you know, you were on a journey for which you didn't know where to start or where to end. And that's how I used to think when I did that case study, because there was no reference material. There was no guidance. There was no, you know, right or wrong, right? We had multiple discussions with this forum and the extended forum pouring over the calculations again, again, and again, to even figure out whether we are doing the right thing or the wrong thing. So yeah, very, um, anecdotal, Asim, that you're talking about how we started off. I remember there was this SCI data project, SCI guide project. We knew that we were looking for something, but in parallel, we were building these case studies, it's almost seems magical today that we are at this place today.

Asim Hussain: That's true. I remember one moment, you just, we were getting really deep into trying to figure out the carbon emissions of networking. That was one of the things I remember that we're getting really deep into trying to figure that out. And we were, you were looking at energy expenditure of bits across a network and trying to estimate the carbon intensity and of the embodied carbon that was being used when you're transmitting data.

And I think eventually we just landed on using a coefficient of like gigabytes per, you know, carbon per gigabyte that we got from some source. I remember like that, there was all of trying to figure all of that out. Yeah, we'd had, you're right, there was all those, like, hard won lessons that we've kind of also forgotten about these days as well. Navveen, what do you remember about when you were writing your case study? What was the hardest challenge that you had when you wrote your case study?

Navveen Balani: So I think going back, I think the fundamental challenge that we faced was around data and particularly around whether the data is authentic enough, right, for measurement. Now that's a, that is a problem to solve at that point of time. So, what we did was, why not go with approximations, coefficients, so at least unblock us from a data perspective. And once we have the data sorted out, I think next was around energy calculations, measurements. And since we are all deploying on cloud, right, most of the applications are running on cloud. At that point of time, uh, a lot of these data was transparent to us, right? So whether it's watching machine, choice of processors. So how do you even get to the power emissions and then calculate energy? So I think it's a journey. We went through the data challenges. We went through the vendor challenges. We then we a lot of reference material, which we use. And I would say finally we then had projects like the SCI guide that we wrote, which provided instructions on how can you go about doing a calculation.

We started off with the ontology project where we thought, okay, we'll give them a model representation of your software boundary, then maybe do a one click and then calculate the SCI score. So a lot of these thinking projects finally landed up in the Impact Framework. I think it's been a journey over the last couple of years, right?

And now, yes, at least we have, we are working towards a mature product or an API, right? Which can help enable any developer, right, to calculate SCI based on their boundary.

Asim Hussain: you're right. I think it basically came around, this is, it basically came around, you actually were, with pen and paper, in its rawest form, trying to calculate the missions of some software, you realize that data was the biggest problem. So then we launched like the data project. And I remember the first thinking we had was like, we, maybe we need to go and create some datasets, but I remember when you guys were starting to investigate this stuff, you were saying like, actually, you know, these datasets already exist, it's not a case of like creating a dataset, it's like we're not even too sure which data sets to use for which contexts. So then you wrote the SCI guide, which then became the "if you want to measure this type of information, use this data set and use it in this way, or use this equation and use it in this way." And so then that evolved to, well, now that's just some text on a page, how do we programmatically help people to actually calculate this stuff in, in a, you know, more scalable way, rather than just giving them content. Navveen, you started to look at this ontology project, which I still, SciOntology, I still love that name. We should have run something with it, but that was all about, like, how do you figure out what it is you're going to measure? That is, that was trying to answer the question, "what is an application?" Like, what are you including?

What are you not including in an application? And then like Srini, you and I were talking about this old project called CarbonQL. And what that was trying to do was trying to create like an API where you could ask this API, like anything. Like I have like. Remember we even came up with like a key format.

If you remember, I'm looking at a server, I'm looking at a server of this type. I'm looking at a server of this type on this cloud. I'm looking at it. So you came up with this, like you could request anything from this API and it was like a facade or something, and it had lots of code we had to just figure out all this logic and come back in with an API. And I can't remember exactly how we got from there to like a YAML file and an Impact Framework that may be lost in time. But we evolved from kind of an API to realizing, actually, this is just not something that we could ever do. It's too big as an API. It's like, it's enormous. It's like, how do you measure everything in the world with one API? So we said, actually, this is software that you're going to write. And then we started talking about that. And I don't know, then we just lost in a whole load of a year's worth of conversations, and then we evolved into where we are right now, which is, how would you describe, let me ask you this question.

This is an interesting question for me. I'm going to ask Srini first. When someone asks you, what is the Impact Framework? How'd you respond? Like, what'd you say to them?

Srinivasan Rakhunathan: So, and this question has been asked many times by my engineering team when I have presented, how do we, you know, they try to push through my agenda of having Impact Framework. And I'm trying to do that more and more with the projects that I handle. The key difference, the key USP of Impact Framework is this is by developers and this is for measurement of your development efficiency, development processes, code, as opposed to what we had prior, which was focused on, you know, what are the standards governing emissions?

What are the reporting? What is GSG saying? What is X? You know, this is a star contrast where you're trying to do something that is going to help developers measure emissions from their day to day code, day to day, you know, CICD pipelines. So that's, that resonates, uh, with most of the developers when we talk to them.

And they're more curious, you know, uh, I think, like, one of the things that we also talked about, all of us, is to, how do we make it granular such that the measurement is easy. So developers don't just need the final emission number, they need to know what you did to come to that number. They will not accept, you know, just something that you give to them on a piece of paper.

So I think that is the USP of Impact Framework for me.

Asim Hussain: Like debugging the carbon emissions of your application is giving you that, you need that granularity. So, you know, like someone just doesn't give you like, "Oh, my carbon score is eight." And you're like, "okay, good. What do I do with that?" It's all of the workings out underneath. And I'm, and I thought it was really fascinating because we had an organization come on our last call, didn't we?

And they showed us their manifest file, which is like 13 megabytes large. So they really went granular and it still only ran in like 20 seconds. I was really impressed with that actually. So Navveen, like, how do you describe Impact Framework to, to whoever you need to describe it to.

Navveen Balani: So, so I described as, as Impact Framework. It's probably a framework that helps you realize the SCI Specification. Uh, and what it means is it helps you basically come up with an SCI score based on, uh, your software boundary. As Srini mentioned, right, it's basically a developer framework that can be used for measuring any software application. And I would even say it's a vendor-neutral framework where you can plug and play your own models or use open source models to come up with a measurement and finally arrive at an SCI score.

Asim Hussain: Yeah, I like that. I forgot. Like, yeah, we, you, you, you're the one that kind of, I remember and the calls like would, was gently nudging us back into like focusing on the SCI, which we, I suppose at that point we kind of got lost in some conversations, but that really focused the, the whole team again quite a few months ago. Do you know, I won't give my definition. I'll actually give a definition that Joseph gave me yesterday in a call that we had. Joseph's the PM for the project. And he described in a very interesting way. He said he's now thinking of it as an executable audit for a software application. And I thought that was a really interesting term, isn't it? An audit, like that manifest file that we're creating is like an audit. And the term executable is like, again, as you're saying, Srini, it's not the final number.

It's like this whole thing that you can re-execute and readjust and value. I think it's been coming up a lot. 'Cause everybody's asking me what is Impact Framework and we're constantly trying to come up with a language for it.

And I just want to maybe just acknowledge the fact that we've come up with something very different. We're kind of like setting like a paradigm here and that's why we're struggling so much to explain to everybody what Impact Framework is. And I'm very excited about that because I think that's when you really change things. We can't even really explain what it is right now. We can only tell you that anybody who gets involved gets really involved and they can really see themselves in it. It's our most popular project in the foundation right now. And when I talk to people externally, there's just a lot of interest in it.

And I'm, and I'm just so excited about it. Should I move on to the questions? I've got some questions. Yeah, should we go ahead? Okay. So here's the first question. So can you explain how the Impact Framework simplifies the process of estimating energy and carbon impact for various applications? So pretty broad question.

So can you explain how the Impact Framework simplifies the process of estimating energy and carbon impact for various applications? Hang on, let me ask this question. Does it simplify? Maybe 'simplify' isn't the right term for what it does. It's quite complicated, isn't it? Maybe it kind of surfaces information in a different way. This is one way of describing a framework. It takes observations about a running system and turns them into impacts. So that's one way. It takes observations about a running system and takes them into impacts. So I suppose one way it simplifies it is if you've got something you are already observing about your running application, like CPUtilization or something else. It can help you then turn that into energy. So that's one way it simplifies it. One other thing that, this is one of the misconceptions people have about Impact Framework when I talk to them about it, is they think it's something that you have to install on every one of your servers, right? Cause that's how most of the things that measure energy work.

They're like, okay, Impact Framework. So how do I install it on my GCP servers or my Azure servers? And I'm like, well, you don't, you just. You know, you just have to tell us what the utilization was of each of your machines and each of your processes and Impact Framework will try and convert that into energy.

So I think that's one way maybe you could say it simplifies it. You don't have to install anything at all?

Srinivasan Rakhunathan: I think so.

And, uh, when you use the word 'simplify,' Asim, to me, the very fact it's a command line utility, it's a huge bonus, right? You don't have to spin up Kubernetes clusters. You don't have to spin up Spring Boot services. You know, you don't have to host it on an API. You don't need an infrastructure for all that, you know, you can

Asim Hussain: You don't need to instrument any of your applications or code or anything. It's just, yeah. It's just like observing. It's

just looking at it. Yeah.

Srinivasan Rakhunathan: Exactly. so, so to me that simplification that you can just download the code, you need to know a little bit of Node.js or Yarn or how to use it to just be able to run it. As long as you're able to do that, you're pretty much up and ready. Maybe you, we can even set up a timer Asim how long it takes for someone to run it.

I bet it'll not be more than five minutes. You know,

Asim Hussain: I don't think so. I don't think so. To get the simple, like hello world version of it out with like, yes, sampled, maybe it's use it with your own data. It will take a little bit more time, but to like use it with a sample data, I think five minutes is absolutely yeah. Any thoughts, Navveen, or how, like anything you want to add to that?

Navveen Balani: So I would also say that it simplifies, I mean, if you're a beginner who wants to do software Carbon Emission Measurement, then, I mean, this API, I mean, the turnaround time for basically measuring, let's say, a virtual machine is, uh, maybe a few hours, right? Just going through the API, uh, setting things up. So the barrier to measurement, uh, I think is quite simplified. Given our history, right, where at least we took maybe a month or two months, right, just to go through the entire data capture process, which model to use, right? So all those knowledge, I would say, is accumulated in this tool, right? So the entry point to SCI now is drastically reduced.

Asim Hussain: I remember that being, I remember that was one of the, one of our original kind of design, I don't know if that's a term design decisions or requirements of the tool was to codify the expertise that was in like both of your heads and in the heads of other people, like the very small set of people who've been looking at this to make it easier, because now all you have to do is you just have to like plug in a utilization value, pick a model or a set of models, and it will like compute all that for you instead of like, "which coefficient should I use, which this, which that?" Is kind of all baked into these models. Talking of models, we call them plugins now, don't we? Sorry. We call them plugins now, which I'm having some name changes. So I remember I was chatting the other day.

We originally had this way of thinking, which is we know we need to capture some observations about our running system, be it utilization of virtual machines or billing data or whatever it is, some observations we have. And we always knew we were going to pass it through some sort of model to create some sort of impact, like energy or, or, or whatever. Then at some point we realized it's not, one model isn't going to work. We're going to have to break it out into lots of, and that was a very important decision that we made. I feel like that was like an inflection point. And there was a moment when we were like thinking about this thing in terms of one model rules them all, to hundreds of thousands of models that you can like combine in this kind of Linux piped command line process. I think that was a real breakthrough moment in the project. I think it was something about, were we trying to compare Boavizta, Cloud Carbon Footprint and Teads and these other models together?

Was that what it was? I can't remember what even triggered

Srinivasan Rakhunathan: I'm, I'm also trying to think.

Asim Hussain: Yeah, I think it was something to do with that because CCF did a lot. Oh, I can't remember. Anyway, what are some of, what are some of your favorite models? I was going to ask, like, what are some of your favorite plugins? Or, yeah, like pipeline plugins. I do remember the moment, actually, I think this is quite important, actually, as well, like, when what once we finally had this ability to grab some observations, put it in a YAML file and then pass it through a set. We know that's what we did. God, this is a memory lane. We standardize the interface to these models.

That's the first thing we did. And as soon as we standardize interface to models, we could then. Use the same data and pass it to different models. So we took the same utilization values, suddenly, suddenly, and I can't believe no one had ever really done this, ever. We took the same utilization values and we passed it through the Cloud Carbon Footprint model plugin that we created. We then took it to the Boavizta model plugin, and we then took it to the Teads, there was a Teads model plugin. And the numbers are so different.

I remember that was a real shock for all of us. And I won't go into details, but there was a, almost a 400 percent difference in the energy values from one to another.

Srinivasan Rakhunathan: I think the methodology differences between these datasets were so pronounced, like you talked about Boavizta, you talked about Cloud Carbon Footprint, that with the original concept that we had where we wanted to cover all, we would have had to build multiple flavors of the Impact Framework and there were cost issues in maintenance that would have caused issues in adoption.

I think the standardization of a model plugin was more a decision that we took once we realized that one model is not going to cut the cake for all of us. And You had different models depending on whether you are hosting it on AWS or Azure or GCP or your laptop or even your mobile devices. But if someone wants to just look at the raw emissions from the software, agnostic of the hardware, you could do that only if you have a very thin measurement tool.

Asim Hussain: Yeah. And for instance, also remember at the time I was still at Intel and we were building a model, a plugin for Impact Framework, which obviously would only measure Intel chips. So you had to have multiple models in a pipeline because if you had other vendors' chips, you would need that model in the pipeline as well to like calculate that model, you know, and the way I imagine the future is that vendors themselves would then be releasing their own models. Like my dream, Srini, is like Microsoft releases, I know it's doing the importer model, but then, you know, the future dream in the future is Microsoft releases models for every single one of its services. And if you're using that service, you plug in the Microsoft model for that service. You plug in whatever observation, like let's say it's gigabyte seconds for Azure Functions. You grab your Azure Function's gigabyte seconds, you've got the observations for there. How do you turn that into energy or carbon?

So ideally you just use the Microsoft Azure Functions plugin, plug

it in, and it knows how to convert gigabyte seconds for Azure Functions to whether, whatever it would be like energy probably, or maybe carbon.

I don't know how we'll end up there. So that'll be like, yeah, that'll be the future. My dream would be like, you come to Impact Framework and like, imagine you see like a page which is like hundreds of models and each of these models has like a logo of like some cloud service on it, you know, and you just drag and drop the cloud service that you're using. And then maybe you've even got some like automatic importer which imports the data from your cloud service into it. And boom, it's all working. It's all automatically calculating

Navveen Balani: It'll be probably like a GPT store.

Asim Hussain: A GPT. Yes. Oh yeah. The Impact Framework store where you go to, I could even imagine like, honestly, everything we've done is open source. Everything we're doing is publishing is open source. I'm fine if a commercial entity wants to come along and create a paid-for model, you know, "if you want to use this model, you need to sign a contract and pay us some money and, and use it" because honestly, maintaining a lot of this stuff is very challenging. Like get making these models is a lot of hard work. I'm okay with organizations, hint, hint out there if there's anybody listening, he wants to create a business on top of Impact Framework. It's, I think it's absolutely fine to, to create a plugin that is commercial creates, you know, let's say. I don't know what that would look like, but, you know, it pays money to, to do this properly.

Cause one of the things we're doing is like with one of our models is the Cloud Instance Metadata model. That model, given a cloud instance ID, tells you metadata about it, like the number of virtual CPUs, this, that, and the other. Because there is no good data set. There is no API, which you can go to for that.

So we need to maintain that as a GSF. And that takes time, effort. So if somebody out there wants to, you know, go do that, I think that's perfectly reasonable. Yeah. Why don't you each say what your favorite plugin is so far? I don't know if you have one. Navveen, what's your favorite plugin so far that we've built? I think I know what Srini's going to be. But, uh,

Navveen Balani: Let Srini go

Asim Hussain: Yeah, Srini go to first.

Navveen Balani: Let me

Asim Hussain: I'll Srini go first because it's a hard choice, isn't it. Srini, what's yours? If it's not the one I think it's going to be, I'm going to be surprised. Go on.

Srinivasan Rakhunathan: No, it's not the one you're thinking because it's not yet there.

Asim Hussain: Oh, okay.

Srinivasan Rakhunathan: Yeah, so I really like the TDP model. Yeah.

Asim Hussain: Like the Teads Curve one,

what we call Teads Curve? Yeah.

Srinivasan Rakhunathan: Yes. Because that's something that was very manual earlier. I don't think any of the other assets have that capability to give that Teads curve. And for me, while the solution, the technical solution was

simpler, I think, but the concept of having a model in the first place was awesome. I think that's how, you know, keep things simple, that same, the logic that works, right? Because it solves, you can extend the TDP model for any hosting infrastructure. It doesn't have to be a laptop or anything. I can just open my power configuration on my laptop, find out what is the TDP score, and then do a multiplication.

Asim Hussain: Oh, just plug it. Yeah.

Do you remember when, cause it's Benjamin Davies. We even brought him in to the standards working group one day, like years ago, just to talk through, cause he wrote, he did that. That's, it all started when he wrote that Medium post where they'd done that analysis and he'd like scanned through all of Amazon's bare metal servers, averaged every single server out to like, this is the generic power curve.

That's what we call the Teads curve. I don't know if that's the right term. We just call it the Teads curve, you know?

But like, that's the generic power curve. And then all you need is that data, plus what's called the TDP of a chip, and then without getting too wonky for our audience, then if you, then you can compute the, an estimate of energy for any single chip in the world. So open the door. I remember when that first came out, I was like, I felt like such a relief when Benjamin did that work. Cause prior to that, I was thinking, "Oh my God, we have to sit and figure out this data for every single chip in the entire world on every single device. How are we going to do this?"

And then he came along and went, "ah, I averaged it all out and it works pretty good." So we were like, "okay," and I see that used everywhere. So many places I see that used every, cause it's so easy. It kind of taught me a lesson actually, which is it was so stunningly easy. It was like one sentence could summarize that model. It's just everywhere. It's just absolutely everywhere. It's so simple to use. So yeah, that's a good one. That's the tease one. Yeah. Srini, Srini, you have one? Is it the SCI one? Maybe. No, Navveen.

Sorry.

Navveen Balani: Probably I'd like to have some AI model. Currently, we don't have one. So probably I'll wait for the hackathon to provide, right? Hopefully, as part of the hackathon, we'll get a lot of good models to evaluate.

Asim Hussain: What would be a good like AI model? Because like. Would it be something that kind of like, looks at it, like, what would the observations, is it prompts? It'd be like, what would be the observation prompts?

Navveen Balani: it'll be, yeah, it'll be an interesting model, I would say, based on the deployment. Uh, for instances, just to give an example, let's say if somebody is using an OpenAPI directly, right? They're just interacting with prompts. So in that case, how do you evaluate the carbon emission given just the prompts, right?

And a lot of these observability data may not be available from OpenAPI. But let's say if you're doing fine tuning or if you're doing your own model, then you will have access to infrastructure, and then you probably can use our existing models. So I would say the first one would be quite interesting where you don't have much information available from vendors.

So how do you come up with emission, right? Just maybe, just based on traction and prompts. Maybe it is prompts, length, latency. Let's see how the community as part of the Hackathon picks up this challenge.

Asim Hussain: I think, like, when I think of the future, like, it's interesting how our world's changed because we used to talk about AI emissions in terms of the training and, like, how long it

takes to train and we used to talk about inference in this very abstract way. And I'll be honest with you in the last six months, everybody just talks about prompts.

Like the word prompt is, and I use, I'll be honest, I use AI co pilots and all this other stuff all the time. And it's like, it's all about prompts, isn't it? I do think, Navveen, I think the future is going to be prompts. Like how many prompts have you made that generate images? Like an image prompt is probably going to be different. You probably have like a coefficient for an image prompt, code prompt, an LLM prompt, something like that. At least in terms of the way humans are working with AI these days. I think prompts is like the key. Yeah. And when you see like articles and stuff out there, it's always like, "Oh, on average prompts cause this impact" that I think I see the language being used that way. So, yeah.

Navveen Balani: I think that probably brings up to, I think, one of the challenges that we still need to solve around managed services, right? A lot of these, if you treat also OpenAPI, right, servicing offer as a managed service, now how do you calculate the carbon emissions? For instance, serverless is a very good example. So you have to use some kind of approximation, right? Because based on time or resources So expect similar models once we get more data and clarity, right? Yeah, we'll reach there.

Asim Hussain: Yeah. I think that's a really interesting, yeah. Maybe we'll get you, let's propose something in the, so there's a Carbon Hack coming up, a Hackathon coming up in a, in a couple of months around Impact Framework. I was going to mention at the end, let's talk right now, actually. 'Cause that actually, 'cause obviously there's going to be prizes for the best kind of models and plugins that we're describing right now.

And, and then Navveen, just as we're talking right now, I think I would love to see, as you say, like some submissions, and we can post in the ideas forum on the hackathon website, like. I'd love to see, you know, turn prompts into emissions.

That might be a right, that actually is turning into like a bit of a working session on Impact Framework that actually might be a good way for us to influence direction of what people are building in the hackathon is what we just described is like, "what are the observations that we want people to try and figure out the impacts for?" So I want to help figure out the impacts of prompts.

We talked about serverless, serverless, the observation for serverless is something called gigabyte seconds. Help me convert gigabyte seconds to energy. Like I can, I've got some ideas for that. Like, yeah, that's how we should think about this. We should be thinking about, we should be saying like, these are the observations that would be the most useful for us to create models, to turn those observations into impacts. So I don't know, I don't know what other observations that we have. Like billing data is obviously oftentimes a fallback.

Srinivasan Rakhunathan: GPU utilization is also there, I think.

Asim Hussain: Yeah. GPU utilization. Yeah. Or like, what about a non-cloud? Well, here's an observation. Yeah. Gigabytes transferred. Yeah. Write a model to turn gigabytes transferred into.

Srinivasan Rakhunathan: Emission.

Asim Hussain: What other observations are there, like from the web space, your observation might be page views. There's already models that do that, like page views.

Maybe this, oh, I think we've hit on something just now. I've been struggling to try and like, how do we encourage, how do we direct people? But I think this is how we direct people, which is: These are observations that we want to be able to model, submit, think it through. And we can, we can submit, but yeah, there's lots of others.

It's just late in the day. So just final question. So let's talk about this. Like, you know, we've been just building the alpha and beta version of this. And you know, there are some people using it right now, but let's just talk it through amongst ourselves, like in practical terms, how would you see, you know, a software developer, a practitioner or a company just start using the Impact Framework?

Like how would you. Imagine if you were talking to a company, what would the very first steps? What would you, what would the conversation be? Like, I'll tell you what I've been thinking about this stage. 'Cause I've really loved the idea of this term that Joseph used, which is like executable audit, maybe it's an audit. Maybe it's a, "bring me in, like what software do you want me to audit? Oh, you want me to audit your XYZ software. Okay. Give me your data. Give me your observations. I'm going to write a YAML file. I'm going to write a manifest file. And at the end, here's a file which gives you. All the emissions of this software over the last month," something like that.

Is that how you would start off or, or are you seeing the conversations go different ways? Srini, I'm seeing you like, I think you think differently. I'm just starting to suspect your mind looks looking, your face is looking.

Srinivasan Rakhunathan: No, I think, see. One scenario we've already seen, Asim, right, is many people are talking about region shifting and time shifting. So if you're running a workload based on the carbon intensity of the underlying region of data center or whatever, electricity grid, people are talking about, and I think many cloud providers have implemented it even to shift their workloads behind the scenes without.

With or without your consent to a different region which is more greener. Now, if people have accepted that as a new normal, then why are we waiting till the operational aspects of the software? Why can't we look at the development cycle, left shift the entire process, give the tool to, there are projects which have 80 developers, 100 developers, 120 developers.

Today, nobody has any idea of how many dev environments, prod environments, pre-prod environments. I think, unless you show people numbers and show people how to measure them, the real impact of what is translating into carbon emissions doesn't even come through. So I think a lot of value is there when it's a practical, when you talk to companies who are in this cloud migration initiatives or huge initiatives, multi year, multi semester initiatives, there is a lot of value in even calculating it at a development level.

Why do we even wait for the operational aspects of, you know, calculating emissions? So I think that is a USP for sure.

Asim Hussain: So I think you mentioned two things that were interesting. One was like. I think you were alluding to this whole idea of once you've got this kind of like Impact Framework manifest file, you can do this kind of what if scenarios that you're alluding to that, right?

So once you've got that manifest file, when you were saying shifting left, were you talking about once you got that audit effectively, once you walk in and you audit your application, and like, "here's my inventory," passing that to the developers and the team going, "by the way, this is everything that you're running. You can run some scenario, don't build CarbonAware computing, just add the CarbonAware scenario model to the end of the pipeline to simulate what it would look like if the CarbonAware was," it kind of gets that kind of, yeah, helps, it almost helps like model, like what if scenarios in, in that environment. And the other thing you were talking about, it was almost like, yeah, you mentioned cloud migration. In my experience, cloud migration is this kind of like, no matter what you do, it's just this messy thing. It's very confusing. You start off with like a very clear idea, a very clear plan. It's very messy and you're not really sure what's going on.

And by the end, you're very clear again, cause you're in the cloud and you've got some good metrics. I remember there was one organization that was reaching us out to the start. And they were like, we're moving to the cloud. And we wanted to measure using Impact Framework on premise. And then consistently measure as we slowly migrate our replication to the cloud. And that way, the audit from Impact Framework should then change slowly over time. Because theoretically, like the cloud is more environmentally better. Then that would have given them that clarity in that migration process. Which we were too late. They did the migration before we, we completed in where we are right now.

But anyway, that was an interesting thought. Navveen, I think because you work in the consulting scene, you actually speak to a lot more customers and clients than us. Like, I don't know, like, what do you have, do you have conversations? Like, how do you have conversations to them about the Impact Framework?

Navveen Balani: So to extend to the simulation example, right, I think a lot of cloud vendors provide calculators, right, to calculate, let's say, the cost based on virtual machines, the entire infrastructure, right? So if you could put the element of carbon and emission also, right, because they know how the infrastructure is set up. And now they can do a simulation based on both cost as well as the carbon emission and also do simulations that reducing carbon also reduces your cost, right? So a lot of, I think enterprise wants to do a lot of FinOps activity, maintain the cost also, right? Sustainability of this tool, right, can be good uh, outcome to reduce their cost also. That can be also a value proportion to lower, I mean, lower the carbon footprint as well as the lower, the total cost of ownership.

Asim Hussain: There is nothing, I would say, stopping anybody using impact for any, like cost as an impact. There's nothing stopping anybody using it just to look at cost and then scenario cost reductions as well. So, okay, I think we're running out of time now today. So let me just maybe just close out statement. I also want to give it a bit of a call out to the Carbon Hack. So thank you both again for like, I know we chat every week, but thank you both for having another chat here. I think it's been such an exciting journey that we've been on. So that's just about it for our deep dive into the Impact Framework. Before we head off, I just have a quick announcement about CarbonHack24. CarbonHack is a global contest for developers who want to make a difference in the fight against climate change. It challenges participants to use the Impact Framework, which we've just been speaking about in depth. The theme of CarbonHack this year is measurement. We want to see how you can use Impact Framework to measure the carbon emissions. You know, other, we're also talking beyond carbon this year. So we have a prize of, you know, water, carbon, other things like that. If you could help us to write plugins for Impact Framework, which help us measure beyond carbon. We would love it. That's really where we want to move the conversation. We have kind of basically three prize categories.

One is the best plugin, which is like what we've just been describing, which is like something that can take an observation and turn it into something else is a plugin beyond carbon is a price for us for a type of plugin, which can take an observation and turn it into like a water or a. Or another beyond carbon type of type of measurement.

And we also have a prize for best content. So if you can help us to explain Impact Framework better than we're explaining it ourselves, there's actually a prize for that. That's like a non technical prize, whether it's a video or a tutorial or anything like that. We also have an undergraduates prize. So if your entire team is made up of undergraduates who are university, there's a special prize for you in addition to the prizes. And we've also now got to confirm, I don't know if I told you, but we've now got to confirm that we can have an under 18s prize. So if your entire team is made up of under 18s, there is again, an additional prize just for you on top of everything else. So I'm very excited about that. So CarbonHack is online.

It's starting on the 18th of March and ends on the 8th of April. You're going to compete to showcase your best application in the prize categories I've just spoken about. The hackathon is open to all, but you must register to be part of it. Registration begins on January 22nd.

Please head to hack.greensoftware.foundation. It's a pre-register. And once you're there, then we'll send you further emails about how to get, how to get more involved. So please go hack.greensoftware.foundation, register now. It's all in the show notes. So we've come to the end of our podcasts. All is left for me to say, thank you so much to Navveen and Srini.

That was really great. Thank you for your contribution. And we really appreciate you coming to Environment Variables.

Navveen Balani: Yeah. Thank you everyone. Looking forward to the Hackathon.

Srinivasan Rakhunathan: Thanks. Thanks for the opportunity and looking forward to meeting you guys in future as well.

Asim Hussain: Awesome. That's all for this episode of Environment Variables. All the resources for this episode are in the show description below. And you can visit podcast. greensoftware. foundation to listen to more episodes of Environment Variables. See you all in the next episode. Bye for now.

Cloud Footprints with CCF

11 januari 2024 | 48 min

Host Chris Adams is joined by Arik Smith and Cameron Cash from Thoughtworks in this episode of Environment Variables. Together they discuss the Cloud Carbon Footprint (CCF) tool and the role it plays in sustainable software development. Join this conversation as they nerd out together about the future of measuring carbon in the cloud, open source collaboration, and some cool posters in the background.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Arik Smith: LinkedIn
Cameron Casher: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

Arik’s awesomely distracting looking poster [02:36]
Notes from the SDIA Green Coding summit [07:05]
(GreenOps) A forecast of green clouds | Thoughtworks [08:18]
Green Software Patterns [08:24]
Electricity Maps API Documentation [28:36]
Software Carbon Intensity (SCI) Specification Project | GSF [32:38]
GitHub - cloud-carbon-footprint/cloud-carbon-footprint [38:04]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Arik Smith: [00:00:00] At the end of the day, CCF is a community supported and collaborative tool, and I think that's the biggest and strongest part of all is that this isn't a methodology that one organization has owned or came up with, but that the, I want to say the entire industry who really cares about this subject has contributed and promoted growth within.

Cam, if I give you the floor, then Arik, I guess you can introduce yourself after that.

Cameron Casher: Thanks, Chris. Really happy to be here. I'm Cameron Kasher. I work at ThoughtWorks on software, data, sustainability, general responsible tech, and really happy to talk to you today about Cloud Carbon Footprint.

Chris Adams: Cool, thank you. And Arik?

Arik Smith: Hey, I'm Arik Smith. I also work as a developer consultant at ThoughtWorks, mainly as a cloud SME and a green software specialist. But I am also excited to be here, so thanks for having me.

Chris Adams: You're very welcome. I'm excited here, but I'm also quite excited by that awesome looking poster that I can see inside the video. I'm afraid that you can't see it in the audio, but there looks like there's a Japanese poster behind you. Is that Zelda or something? Arik, can you please tell me what that is, because it's totally distracting, but it looks really awesome.

Arik Smith: Yeah, for sure. So behind me I have my home office, which is basically in my basement that I call my culture cave. So what [00:03:00] you're looking at is what I call my wall of culture, where I just hang up artwork that I usually find at local vendors around town, mainly around my favorite video games and movies.

So yeah, the poster is a poster with Link on it and in Japanese katakana, it says Zelda, but there's also some other stuff on there. I have an Avatar poster right next to it for The Last Airbender, and a Final Fantasy VII, and an Arcane League of Legends poster. Lots of fun stuff going on.

Chris Adams: Wow that's an impressive nerd cave, and I think anyone who's been following Zelda this year will probably be excited about a reference to Tears of the Kingdom, amongst other things. All right, and Cam, we met in person, actually, about a month ago, when you came over to Berlin for the Green Coding Summit, and I A, I realized you were so much taller than me, oh my god, but also I didn't realize whereabouts that you were coming from.

So you were coming from Colorado before, and at the time of recording, this is the 21st of December, so there's a bunch [00:04:00] of really explosive political news that might impact what will happen with elections next year. But, as much as I do want to talk about that, and what the climate implications of having different leadership in America might be, I think I should ask you about one thing you told me about, skiing in Colorado. That's a thing that you do on a regular, right?

Cameron Casher: Yeah, it was great to see you in the flesh in Berlin. It was really awesome to be there. And yes, Colorado has found itself at the forefront of U. S. politics at the moment. So, that's exciting. And, yeah, we're in uh, ski season here. I'm excited to get out and snowboard a bit. But still waiting on some of the good snow to hit some of the Colorado ski resorts, but I'll be trying to get some days in this winter for sure.

Chris Adams: Nice. Okay, cool. I'm calling from Berlin, and if you haven't heard this podcast before, I should probably introduce myself as well. My name is Chris Adams. I work as the executive director of the [00:05:00] Green Web Foundation, a small Dutch non profit focusing on reaching an entirely fossil free internet by 2030. But I'm also one of the chairs of the Green Software Foundation's policy working group, where we do work with standards bodies and other organizations working in this field of sustainable software and general sustainability. All right, I think with that in place, I should ask you folks, are you sitting comfortably?

Looks like you are. Okay, then I guess I'll begin. Okay, Cameron, I'll ask you about this because you were talking about this notion of responsible tech when you came over in Berlin, and then maybe that will use that as a way to talk about some of the projects that you've been working on. Maybe we could just open with what this notion of the Responsible Tech Playbook that you're presenting, because that looks like it provides a bit of context and that helps understand where tools and carbon calculation might actually fit into the context of using technology in a more responsible basis.

If you could start there, maybe.

Cameron Casher: Yeah, [00:06:00] definitely. I know you'll probably hint to this, but we can definitely link to these resources in the podcast description. But

Chris Adams: Yeah, absolutely. Thank you for reminding me. If you hear any projects mentioned or anything like that, and if we don't put it in the show notes, please do let us know, because we realize that people who listen to this podcast often use every single project as a jump off board for their own research. So we will do everything we can to make sure this is useful for other people who are taking their first steps or trying to explore this new field. All right, Cameron, yes, you were saying we were, you were

presenting about this, yeah.

Cameron Casher: So ThoughtWorks has done some really cool stuff in the area of responsible tech. We had a playbook that came out a couple years ago. And really cool thing recently is we got in touch with some folks at the United Nations and they were very interested in our Responsible Tech Playbook. So we were able to do some really cool work partnering with them to create a more customized [00:07:00] version of it. They're very large around the world and they have a lot of people doing a lot of great stuff. So incorporating responsible tech principles and practices was very important to them and doing it in a more customizable way. So that's what I talked about in Berlin a bit at the Green Code Summit about what the playbook was.

So maybe it's worth just defining what we consider responsible tech.

Chris Adams: Yeah, go for it.

Cameron Casher: We consider responsible tech as the active consideration of values, unintended consequences, and negative impacts of technology. A lot of times these tend to go overlooked. It's not always necessarily a negative intention when developing technology, but really it's, the idea is to manage and mitigate potential risk and to be inclusive to all communities.

There are a lot of different topics we cover in the playbook, including data and AI. Accessibility is a big one, but what I really focused on in [00:08:00] Honestly, because a bit of my background was sustainability, so I was able to shift the focus at the conference talking about it, but also help get a really cool chapter in the playbook around sustainability and how we could incorporate the idea of GreenOps and incorporating Cloud Carbon Footprint.

Chris Adams: Okay, and GreenOps in this scenario is essentially what DevOps is to developing. GreenOps is to this same idea, like it's essentially operationalizing some of the kind of aspects of green software. Is that the idea behind it?

Cameron Casher: Yeah, and another ops really, there's a ton of ops these days, but the idea is tying it really close to DevOps or FinOps too, from a more financial standpoint, really considering green software, carbon emissions, carbon footprint within your development process.

Chris Adams: Ah, okay, so that makes sense. And now I understand the link between there and Cloud Carbon Footprint in its own right, which basically ends up essentially using some of the [00:09:00] usage data that gets produced as a byproduct of operating various digital services and then providing that in a way that you can act upon, or do something, or, or, you shouldn't.

Arik, I see you nodding, so I suspect maybe I should actually give some space for you here. Maybe you can introduce Cloud Carbon Footprint to people who have never heard of this, and why someone who is a software engineer who is curious about environmental sustainability in this context might care about it or want to know about it.

Arik Smith: Yeah, absolutely. Cloud Carbon Footprint is an open source tool for tracking your energy usage and carbon emissions across All the major cloud providers, AWS, Google Cloud, and Azure. And, which we can talk about later, we've also added, or begun to add support for Alibaba Cloud as well. It's a great tool, it serves as an unopinionated monorepository of multiple toolkits for you to go about grabbing your estimates.

So we have a CLI, uh, or a command line interface you can interact with, uh, we have an [00:10:00] API that you can integrate into existing workloads or directly carry, as well as an out of the box dashboard for visualizing and tracking your emissions over time. Um, and we present this at a daily granularity, sometimes even hourly.

Because what we do is we query your usage from your billing data. So think about AWS, Athena, costing usage reports, exports, or your Google Cloud, BigQuery, billing data exports, or your Azure cost details API or exports. We take all of that in the granular usage data that they use and convert that into energy where we can then convert that into carbon emissions to display.

Chris Adams: Ah, cool, okay, so anything that can basically expose billing data or usage data can then basically provide a single view across maybe in a state or a different set of services. So rather than having 3 or 4 competing and possibly incompatible calculators, you have one [00:11:00] consistent way of looking at this stuff, right?

Arik Smith: Absolutely, yep, you'll have one unified methodology to be able to view your usage across multiple cloud providers all within a single dashboard or a single place, which is great just to get a holistic view of where you stand and comparing apples to apples in that way.

Chris Adams: Gotcha. All right, okay. Now, I know that I've been somewhat familiar with Cloud Carbon Footprint and there's a bit of a backstory behind how it came about, because I think there was actually a few years ago there was some really like pioneering work from some folks at Etsy who started on some of this. I'll leave that open for either of you to actually talk a bit about that, because I think there's a really interesting story about how some of this work done in the open there ends up turning into a tool which ends up feeling a little bit like, you know, that XKCD cartoon where there's that one piece of project that everything gets built on? That's how I see a bunch of this. And maybe I, maybe I'll let one of you folks explain some of the origins of this and how that end up, how some of that led to [00:12:00] CCF, or Cloud Carbon Footprint, being like an open source project with a number of users.

Arik Smith: Yeah, absolutely. So within ThoughtWorks itself, there was a lot of growing interest within climate action and getting a hold of our own infrastructure and where our footprint stands with that, especially within our cloud resources. And what we found was that there was not really a tool to do so at the time.

So this was about, and Cam, feel free to correct me on timing, but I believe about 2020, where the bulk of this work was done. And being ThoughtWorkers, we typically decided to look into ways we could perhaps create a way for ourselves to measure our own cloud usage. And we came across Etsy's awesome Cloud Jewels methodology and saw the way that they were estimating compute resources within the cloud, and it served as a great basis and foundation for us to build and find out how can we do that for Google Cloud and AWS and Azure, all the [00:13:00] different cloud providers that we have some level of usage within.

So we started building out a proof of concept to estimate our own internal usage and eventually we made the decision that due to the absence of any tooling out there that this is something everyone should have because we want to empower all organizations and all users to be able to have this level of transparency with their cloud usage because the compute resources and overall energy usage of the IT industry in general is a big major part of our global footprint, especially in major large tech companies and large scale product companies.

We went open source back in March 2021, and since then we've been iterating on it ever since until the cloud providers came out with their own native tools, which is great, and we saw more and more tools and interests popping up in this space.

Chris Adams: okay. That's cool. So basically there's this OG climate calculation stuff, uh, from Etsy, and then that big, that served as some of the basis for a more polished [00:14:00] tool that could be used in lots of different contexts, rather than just the ones that Etsy used, because I think back then, Etsy had a bunch of on-prem stuff and a bunch of Google Cloud, so that was where some of it looked like, but when CCF became, I think the first version that I saw, actually had support for both AWS and a bit of Google Cloud, and now it's actually got support for, like you said, all the big ones, plus Alibaba, I didn't know about Alibaba, actually.

it.

Arik Smith: Yeah, Alibaba is limited to compute resources for now, but we're still looking into ways to expand that. Um, similar to how we did Um, at the beginning with Etsy, they were estimating cloud, um, compute resources. And then we started thinking about how can we expand that to memory and to networking. And all of that has been a collaborative effort, which is why we call out Etsy within our methodology as the basis from where we started from.

Like even just the networking estimations. We work with Benjamin Davey from Teads based on a write up he did for estimating network as well as embodied admissions and iterated within the other [00:15:00] usage levels within the same way.

Chris Adams: Ah, cool. I think I remember, actually, there's a bit of work by David Mitton there, when he did some work around the CPU coefficient stuff as well. I realize I'm diving into, like, loads of really nerdy stuff, uh Cam, I see you nodding, so maybe I'll let you just touch on some of that stuff, because I think it might be worth talking a little bit about where else this turns up, because we've seen a number of calculators kind of launch now.

A lot of them either reference CCF, or have started using that as a basis. But before, though, I'm getting ahead of myself, yeah. Maybe I'll give the space for you, Cameron, because I saw you getting a bit excited there when we were talking about some of that.

Cameron Casher: Yeah, what's funny is the opportunity to speak here today with you gave us, gave Arik and I a chance to be a little retrospective about our time with CCF and think about lessons learned and what's gone on, what's been challenging. And yeah, even just hearing Arik talk about our partnerships and collaboration, I think it's just worth saying that's been a really cool aspect of being an open source maintainer. [00:16:00] Just being able to openly work with the community, being able to have really awesome discussions with industry experts like David Mitton or Benjamin Davey or everyone at Etsy to just continually evolve the methodology for CCF and make it better, make it fit users, and getting into some of the challenges, I don't know if it's necessarily a challenge, but it's just a tough point is that. We don't have the best transparency into who's using the tool, and I think that's a pretty common trend for open source software, but a lot of it is just through word of mouth, or even when I was in Berlin and talked about it, I got the chance to meet a lot of people that approached me and let me know that they were using CCF, which is, it's really cool to hear about that and the impact that it's had.

Chris Adams: Oh, thanks, Cam. So you mentioned David, uh, David Mitton, and, uh, and I realize there's a couple of other people who are also involved in this as well. I mean, well, the original person on the podcast was, uh, [00:17:00] uh, Danielle, uh, oh, Cameron, I need you to help me with her surname. What was Danielle's surname again?

Because she was

the original product manager on this.

Cameron Casher: Danielle Erickson, and she was, yeah, one of the first guests, and she was a very integral role with the open source launch and continued feature development for CCF as our project manager. So, um, yeah, Danielle had a serious role. And also Lisa McNally, who is the, also on the policy working group with you, um, she helped spearhead a lot of our development too.

And I'd be remiss if I didn't mention Marco Valtas, who had a significant role on the dev team as well.

Chris Adams: Oh, cool. I'm really glad you mentioned that, actually. And, uh, actually going back, Jesus, in a few years back now, oh, uh, was, uh, I believe Chelsea, Chelsea, I think it's Chelsea Mozen. She was one of the people who was organizing Climate Action Tech back in 2019 2020, and, uh, I learned so much from her, and she was one of the people really pushing the stuff at Etsy as well.

So, it's, I'm really glad [00:18:00] you spoke about some of the other people, because there's a whole lineage of people who've been contributing to this now over the years.

Cameron Casher: And so many people are coming to mind and, you know, even Dan Lewis-Toakley, who's been a part of the Green Software Foundation, really spearheaded the open source launch too, in the early days. So important to include him as well.

Chris Adams: Yeah, happy days.

Cameron Casher: So I think we're just trying to look for ways to have a better understanding of who's using it.

We have an adopters markdown file in our GitHub repo. We're still trying to get the word out that we want to know what your story is and what your experience has been with CCF. And we have a blog that we just put on the microsite and we're very open to just sharing stories and experiences to providing help for anyone new who wants to try and get it set up at their organization or just for their own personal use.

Chris Adams: I can definitely speak to this because, uh, we maintain, where I work, we maintain a library called CO2.js, which we know is used in a few places, like, uh, we [00:19:00] spoke to Mozilla, and it's in their browser, which is cool, and we work with some other people where we collaborated, but then sometimes it turns up in totally random places.

So, like, I did a talk, um, in Hungary, uh, and I was invited to speak at this event, and then when it was launched, I found out that it was actually used inside one of the key projects that people were actually launching on the day. I was like, "oh! Wow, that's cool and unexpected, but nice to know," right? And I feel that seems to be some of the cases when I see with CCF, because I know that, um, there's a tool called GreenPixie, which is a quite well developed, quite sophisticated commercial product that initially had some origins with CCF, where they use some of that.

I understand, I believe, I think VMware uses some parts of it, as well as IBM. So it turns up in all these interesting places that I didn't, that I hadn't thought about. And when I actually was at the same event as you. I didn't realize the footprint it had in the financial services industry. And that was the thing that really blew my mind, actually, and this, it's, I guess it kind of does [00:20:00] make sense, really, because if you don't have access to information here, having an open methodology is quite useful, and we, at that event, we saw at least one person from a well known bank, ABN AMRO, this is like one of the third largest banks in the Netherlands, they basically said, we're a regulated organization, we, when we calculated our emissions, it's like 40 percent of our carbon footprint comes from our use of digital services.

So we want to have an open tool that doesn't have us rely on a single thing and lets us be transparent with both our regulators, but also our sustainability teams. And this was actually mind blowing that you'd actually have companies talking about this because for the longest time, people tend to either don't say too much about what the environmental impact of digital services might be in their organization, or they say it's very low.

So to have organizations say, yeah, we're using it, this is what we use as our basis, and this is how we calculate it, was mind blowing. It was really useful to see that actually because Like you said, it's really hard to see, and it would be so nice to know that it's [00:21:00] reassuring to find people using your projects, but sometimes it's so hard, you do ask yourself, like, could we have found out another, did I have to really fly across the Atlantic to find out that people are using this tool?

Cameron Casher: Yeah. Yeah. And I'll just add, I think we might touch on this a bit too, but our roadmap, how we decide what we want to continue to build and look into is really community driven. So it's really helpful to understand how people are using CCF to help us understand what to fix or implement. Yeah, it's just very important, really.

Chris Adams: So maybe we can talk a little bit about that then, because we spoke about roadmaps, and like I saw you nodding along when you were thinking about some of the things that are on this list, and I realize that you've been implementing some of this. Are there any kind of, like, exciting things that you're looking forward to seeing the light of day, or being able to work on, that you're really thinking, "Ah, I cannot wait for this to be visible," for example?

Arik Smith: Yeah, absolutely. So we have a few things in the works that I think are [00:22:00] exciting. One of the big things is a lot of improvements we've been trying to do this year is to our overall implementation of Azure, as well as just performance and scalability, which has been one of the biggest lessons learned during the time that we've been developing CCF.

Because ThoughtWorks itself isn't a product company, so we don't have a lot of internal cloud usage. So really learning and leveraging partners and collaborators and sometimes even clients to figure out what type of issues or what type of considerations that enterprise level product companies will have when using CCF has been really helpful and what we've been really leaning into.

And even just Azure as well, because Azure has a Different implementation where it's more API based instead of SQL based like your Google Cloud BigQuery or your AWS Athena, which we interact with for usage. So we currently have a migration going on to migrate to some of the new ways that Azure has made available to [00:23:00] grab billing data.

And one of that is through their new Cost Details API, which we were previously using their Consumption Management API. And then as well as the ability to export cost into an Azure blob and be able to parse a large amount of data from that. So that's currently in the works and we can't, we think this will help a lot of large scale Azure users, which unfortunately due to the differences in implementation.

tends to lag a little bit behind in some of the ways we deal with the other cloud providers. But also, even beyond that, we've seen a lot of interest in Trying to get a grasp of usage and cost for Kubernetes workloads and clusters. And we have a really cool partnership coming up in ways we are looking into collaborating with OpenCost, another open source tool for diving into Kubernetes spend and finding ways to use CCF as a basis to showcase energy [00:24:00] usage and carbon emissions alongside the data that they show within the tool.

We've done a similar collaboration in the past with Backstage, making CCF a plugin for their platform. And that is also another cool opportunity coming up because people use open costs and we're trying to really leverage the ways that CCF is flexible and extensible to where you can still use CCF's methodology as we've seen

in some of the other tools we mentioned, but also in your existing toolkit, and that's why we keep referring to it as a tool and not just a product, because it's really just a means to an end to be able to get transparency into your cloud usage and carbon footprint. I think those are two of the biggest things.

Oh, and we also have Boavizta, but I may actually toss that to Cam because that's something he's been a little bit closer

to to talk about.

Chris Adams: Alright, Cam, before we go into that, I just want to, I realise that we didn't actually speak about this now, um, CCF [00:25:00] is a open source product, um, it's presumably written, is it TypeScript or JavaScript? Maybe we could just touch on that briefly because I realize that we didn't actually talk about any of that, uh, and it might be useful for people who are curious about taking their first steps, right?

Arik Smith: Yeah, for sure. Yeah, CCF is an open source monorepository, all written in TypeScript. And that was a deliberate decision because we wanted to have a single language used across the entire codebase to make contributions and onboarding to the project easier. So if you're looking for an open source project to contribute to, we welcome it.

Even if you just want to read the methodology and provide feedback, we welcome that as well. But yeah, so the dashboard itself mainly leverages React. It has a simple React dashboard for visualizing your data with the API being in Express and the rest being in native TypeScript. Tried to keep it simple, but,

Chris Adams: so it's stuff that people are likely to be, to use, it's not something like written in Sorry [00:26:00] Haskell Friends, like Haskell, or something which is not that popular. Okay, cool.

All right, Cam, yeah. Cam, you were about to talk about some of the other things on the roadmap that you, uh, touched on. I think, maybe it might be worth just briefly explaining what Boavizta is, because if people haven't heard of Boavizta, they might not know about the cool French guys and, uh, women working in that particular project.

Cameron Casher: Yeah, Boavizta actually had a really cool workshop at the summit that we were at in Berlin. And they were explaining some of the ways that you can go about measuring the energy from some of your server usage, which is some really cool work and props to those guys who did an excellent job there. Because I think there was a really good collaboration in that workshop and it was pretty well received and spawned a lot of great conversation.

Chris Adams: I agree with you on that. I remember, so the, so the thing with, the interesting about Boavizta is like, we'll share a link to the workshop, because one thing that was nice is they did this, Boavizta led one of these workshops about understanding [00:27:00] the both embodied energy and the usage energy in cloud, but they did another one about AI as well, but they were gracious enough to share the entire decks and everything online as well, so we'll add some links to that.

So we've established Boavizta, that was one thing, but you mentioned there's some other products or other kind of integrations that are on the roadmap that you might be talking about, possibly also with French speaking people involved, I suppose.

Cameron Casher: Yeah. So I guess for context, we, with our methodology, we've had to rely on a lot of other public and open datasets since a lot of the information around carbon or energy estimations, we haven't been able to grab directly from the major cloud providers, at least just yet, we are really reliant on some of the open data sets we find.

And with that comes a lot of manual updating and monitoring to make sure we're aware of when some of these data sets get published with new information, this really led us to the idea of wanting to [00:28:00] more automate the way that people could get some of this data, whether it's energy estimations for servers or even carbon intensity values. What I really wanted to try and do lately is almost integrate feature toggles. You can always default to some of this average data from the publicly available sources, but if you have a Boavizta subscription or however you'd connect, then maybe you could select that as your way to get some server information from some of your, let's say AWS vCPUs. Or another thing that we wanted to express was our partnership with Electricity Maps. This is another company over in Europe that's doing a really great job around understanding real time carbon intensity values from regions globally around the world. This is really nice for CCF because you can really get more accurate carbon [00:29:00] intensity data, which is a integral part of our estimation for carbon emissions in our methodology. And some of the work we were able to do lately was map the Electricity Map zones to cloud provider regions. So you're able to directly connect. If you have a subscription to Electricity Maps, you essentially just pick a box and say you want to use this and get to see some of their data and map it to some of the daily values that you'll see from your cost and usage report.

Chris Adams: Ah, I see. Okay, so if I, to make that a little bit more concrete, when you often have, say, some values which might be, so people often talk about the idea that you can maybe change a region to reduce the environmental impact. Let's say you're using AWS, and on one part of the United States, on the Eastern Seaboard, New Virginia, not Eastern Seaboard, but North Virginia, yeah, that's going to be using a bunch of coal, and you, but on the other side, you might have, say, in Oregon, it's going to be a very [00:30:00] kind of like cleaner grid by comparison, but there may be other places which might fluctuate at various times and that will capture some of those changes specifically that an Electricity Map would give you, right?

If there's maybe times of day where it's particularly sunny and windy, you'd have green energy there and it would actually reflect that in the billing, for example, or show it in your stats in a given day or hour, right?

Cameron Casher: Right. And the really cool thing about their API is that we are able to grab historic data. And a lot of people using CCF are wanting to see what their usage looks like for the past year or two. So we're able to actually map on those given days of their usage, backtrack and backfill. And yeah, they also offer hourly estimations too.

That's something we're looking forward to potentially implementing for CCF is estimations on an hourly basis.

Chris Adams: Ah, I did not know that. That's very cool to know. All right, because this is one of the scenarios that I'm curious about. Because once you do actually have like enough usage data, you often, the question is, it's almost hypothetical. [00:31:00] "Now I know I've been doing this year. What could I have done so that I can inform like future discussions?"

This is why, because a lot of the time having maybe a bunch of historical data and then being able to see would it have made a difference if I switched to another region, or would it have made a difference if I moved a particular piece of job that happened, a computing job that happened every single time at this time, could I have measured that?

And it sounds like this, these are some of the tooling that actually makes some of that possible, so you can start comparing what you would do then, say this is what we should be doing next year, now that we've actually got some of our own usage data and some carbon information about what, where we might go from there.

Cameron Casher: Yeah, and I know a lot of organizations that are trying to set their sustainability goals for the future and even in the case of ThoughtWorks, having that baseline year is very important where you can set a year and be able to say, okay, this is our baseline. This is where we want to measure up against as we try and hit some of our goals that we're setting, like our 5 or 10 year plans.

Being able to see that historic data is very important today. If you want to [00:32:00] set 2023 as your baseline, you can see what your cloud carbon footprint looks like from that year and just move forward and try and improve it.

Chris Adams: Ah, okay. That's helpful because basically the entire sector needs to halve its emissions by 2030 in order to stay on the kind of pipe, on the trajectory for 1.5 degrees. And I guess that gives you some context that works out to be about a saving, you need to reduce emissions by between seven to 10% per year depending on what you have.

And this might give you a way of saying, "okay. Here's what they are now, this year, this is what we need to beat or stay inside if we're going to be on track." Huh, I didn't know that.

Okay,

Cameron Casher: Yeah. And actually that's a good plug for the, the SCI standard at the Green Software Foundation because, uh, CCF does a really good job at giving you your holistic view of all your usage and emissions, which a lot of companies is only going to grow as they scale. Also using this alongside the SCI could be helpful to see more of the rate of carbon emissions.

Chris Adams: [00:33:00] Okay, so this is one thing that I think is interesting that really caught my eye, because when I was initially looking at Cloud Carbon Footprint, there was this idea that you can tag certain services, so you can say, here's my team, or there's a particular there's maybe a particular digital service I make available to people. Presumably you could track that, and you could say, "over this week, the SCI has been this much, which is the Software Carbon Intensity. And then I want to improve that by X percent," and you could set that as a target for the next month, and so on. That's essentially what you would be able to do. Is that correct?

Cameron Casher: Yeah. So we did implement tagging, which was a really crucial feature for us in a lot of ways to help with this whole idea of GreenOps and just understanding how you can do different breakdowns for teams or resources. And actually Arik was a really pivotal developer for the tagging. So I might let him speak a little bit about,

Chris Adams: Yeah, please do. Yeah.

Arik Smith: Yeah, for sure. So yeah, so tagging [00:34:00] basically allowed for us to display the same tags that you would use within your AWS or Google Cloud or Azure Resources, but also expose them on the resources that we were estimating within CCF. And it's great because not only does does it allow you to filter at a more custom level when you're using the CCF API, if you only want to see

resources tagged with a specific value, but it's also great for when you're trying to see aggregated estimates for a custom grouping of resources. So Cam mentioned whether this may be at a team level, or a sub organization level, or a project level, or however you wish to organize it. So, it has been great in getting a grasp in a much more custom level of where you stand within your organization and what your big contributors are that go beyond just the account and service level.

Chris Adams: No, that's actually really interesting. So, what I think it sounds like you're saying is, with tagging like this, you could basically say in a given [00:35:00] month, I know that all the billing for this particular service has been this much. And I know I've done maybe 100, 000 or 10 million requests. Then basically, because the software carbon intensity is essentially, it's the number of things you've done, divided by the emissions, or the other way around basically. That sounds like you could plausibly get SCI ratings for almost any service very quickly that way, across an entire project, basically. I didn't really, I didn't really think of it that way, but that, that, that would be right. If you just have a bunch of things tagged, you know how much usage you have, like, how many people have used this, or how many requests have you served? That will be enough to put together some calculations for a rate like that that you could track over time, right?

Arik Smith: Yeah, absolutely. I think it also opens a lot of possibilities for the ways that you can use CCF. So it's something that I forgot to mention on our roadmap. We're trying to continue to expand upon right now, a lot of those features are available at the API level, which just so happens to be the most [00:36:00] popular way people interact with CCF.

But we're also trying to leverage the dashboard to be able to view a breakdown of your tagged resources in a much more custom way and be able to filter through data through there as well. But absolutely, yeah, it opened up a door to where the world is your oyster in terms of how you choose to bring your estimates or certain groupings into your overall footprint.

Chris Adams: Okay, so I'm glad we had this conversation, because the next thing I want to ask is about community involvement, because when you're running an open source project, there's, there's a, there's a challenge about working out, do you use a mailing list? How do you actually get feedback from people? Let's say we just had this conversation about, okay, I want to have automatic SCI readings for every single thing that comes through, like, where would I ask?

Is there like a Slack channel? Is there a mailing list? Or is there a GitHub discussions board or something? Maybe I could ask one of you about where I would ask something like this to have this conversation, because I'll be honest, I might look at some of this over winter, if nothing else, over the winter break, [00:37:00] because I'd, that didn't occur to me until we just had this conversation now.

Arik Smith: Yeah, absolutely. It's been a learning experience for us. Um, ThoughtWorks has had a few open source projects previously, but at least for us personally, we've never been maintainers, so we've definitely still have been trying to figure out that cadence of what is our preferred communication method? What do people prefer?

And with this project being targeted not only to the software engineer that's curious about their own team's carbon footprint, but also IT execs that are can actively make decisions and utilize CCF to provide a lot of information behind those decisions. We have to be flexible. So we started and currently still use a simple Google group, which is on our website that you can join and become a community member of.

And we get lots of emails that way, either asking direct questions or troubleshooting information from our team or just the community at large, which has been [00:38:00] great because sometimes people may chime in and share their experiences or give their feedback. Of course, since our project is on GitHub, we leverage a lot of GitHub issues and the typical cadence that you would find within the open source repository there.

So people usually raise new issues for bug reports or features, and we've also recently enabled the GitHub Discussions board for our repository to allow for further discussions, especially around feedback or general questions to hopefully try to lessen the load on our email group a bit and keep it closer to our code since we've been seeing a lot of technical questions and bug reports and troubleshootings coming through there and more recently we've been trying to experiment with a Slack channel to get some of those conversations going at least a little bit more seamlessly because one of the main things we've been troubleshooting and learning from is the way that we interact with collaborators.

A lot of times when [00:39:00] other organizations or companies are standing up CCF within their own organization, they often want to get back to the tool or they may find a bug or something to improve and may open up pull requests and a lot of that conversations have been through the email group or GitHub at large, but for instance, that Azure API migration that we're working on is actually being collaborated with some engineers from Resync that Cam met during his time in Berlin.

So we've opened up a Slack channel to hopefully make some of that collaboration and communication and questions easier. Before it's been ad-hoc, "okay, we'll join your Slack or we'll join your, your Teams account or maybe we can do a simple Google chat," but hopefully to centralize and encourage a lot of those communications and collaborations going forward.

Chris Adams: Ah, cool. So you said one thing that was quite interesting there, about that it's not centri It's, this gives me the impression that it's, that Cloud Carbon Footprint is almost like yes, there's obviously some work from ThoughtWorks, and some [00:40:00] contribution there, but it exists as its own project, and I think when I'm looking up now If I go to GitHub, it's its own organization, so there's like a, there's the implication there that this would be like a community project that, in the long run, would have other organizations contributing and maintaining it, because I didn't know about Resync actually contributing in that way.

That's really neat.

Arik Smith: Yeah, absolutely. That's the stance we've always taken with CCF. I know we mention ThoughtWorks a lot, and that's why we're trying to pivot the language to ThoughtWorks just being the primary sponsor of the tool.

Because at the end of the day, they have invested a lot. Within Cam and I's team and the sustainability solutions team at large to put in a lot of the day to day work within the tool, but everything we spoke about from that initial methodology and building on top of Etsy Cloud Jewels, like there were conversations with Etsy actually being had to get Feedback on the methodology, we're coming up in the okay to move forward that way.

Uh, we've collaborated with cloud providers [00:41:00] themselves to get feedback on the way we're doing things and to see if our estimates make sense. And yeah, even with other organizations on our website, we have a list of innovation partners, which really don't even cover everyone we've worked with that have contributed to the project in some way.

And that's why even on GitHub, it's organized within, at least all the, all of the CCF related repositories are organized within the Cloud Carbon Footprint Organization. Because at the end of the day, it is a community supported and collaborative tool. And I think that's the biggest and strongest part of all is that this isn't a methodology that one organization has owned or came up with, but that the, I want to say the entire industry who really cares about this subject has contributed and promoted growth within.

Chris Adams: okay, so like an open place where there's some consensus to actually figure out, "okay, this is how you measure this stuff." All right, wow.

Alright, we're coming up to time now, so I just want to make sure, are there any things that you folks would like me [00:42:00] to draw attention to, or are there any questions I haven't asked yet that you'd like me to ask or you want to talk about before we come to the end of this?

Cameron Casher: I'll just say that I just want to thank you for letting us speak to CCF. Um, If there's anything that's become more apparent to me recently, it's been the impact that CCF has had in the community and in this domain. And I know we talked a bit about this, but just stressing that we want to hear more from the community.

We want to understand the people using it. So really feel free to reach out to us in any of the ways that Arik just explained, Google or GitHub. And. Yeah, we'd love to talk more and understand about how you're using CCF and one of the things we didn't really get into today was just how CCF can really be like the starting point for a lot of companies because just understanding where you're at in measuring is a first phase of even this Green Ops idea where you get into [00:43:00] understanding, measuring, and then the final act of reducing your carbon footprint. So, we'd like to think that CCF really helps enable that act of reducing, and I know the Green Software Foundation has a lot of great material around where to start, especially around the cloud with the green software pattern.

So, using CCF to have sort of a perspective and baseline of where you think you can start to tweak and make some changes is really a cool and important way to understand how different people are mitigating their carbon footprint in the cloud.

Chris Adams: Brilliant, thank you for that Cam. And to Arik, I realize that I should have asked this at the very beginning, but if someone wants to find out about Cloud Carbon for the project, is there a domain name, is there a website, what would you, what would they type into their favorite search engine or browser?

Arik Smith: Yeah, absolutely. Usually, if you Google Cloud Carbon Footprint, we're appreciatively [00:44:00] one of the first, if not the first, to pop up. But, yeah, you can definitely find us at cloudcarbonfootprint.org. Um, it contains our website, our main documentation, and a breakdown of our methodology. There's also a live demo there, if you don't want to download the tool but want to get familiar with what the dashboard is and the capabilities of it, um, and also links to our GitHubs and some of the community groups like the Google groups that we mentioned is available there as well. Otherwise, like Cam mentioned, we're pretty open, so if you still have a question or want to get involved in some way or even have feedback, which we welcome a lot, feel free to hit us up on LinkedIn or shoot us an email.

Maybe we'll provide a link or something and, or any preferred channel that you can find us on.

Chris Adams: Brilliant, thank you for that. Okay then, and maybe we'll just wrap up there, but I feel like there's maybe one thing that it's really worth emphasizing. If you're using tools like Cloud Carbon Footprint, please do tell the maintainers that they're using them. [00:45:00] Because usually you only ever hear from people when they're unhappy with your product or project rather than when they are using it. And it does make it easier to feel better about your life choices. So that was the last thing I might share with you, uh, because it's something that we actually experience ourselves with CO2.js as well. But folks, I really enjoyed learning about cloud carbon footprint with you here. And, uh, I'm looking forward to some of the cool new things that come up from there.

Once again, thanks again for your time. And Cam, enjoy your skiing. And Arik, enjoy adventures with Link, alright?

Arik Smith: Will do. Thanks a lot, Chris.

Cameron Casher: Thank you, Chris.

Chris Adams: Thanks guys, see you around. Hey everyone, thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the [00:46:00] Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

Driving Climate Change Solutions with AI

4 januari 2024 | 16 min

In this episode we take a look back at another talk from our Decarb 2023 Event while we’re all on the New Year break at the GSF. Titled Driving Climate Change Solutions with AI, Charlotte Degot of CO2 AI and Diana Dimitrova from Boston Consulting Group talk about the Third Annual Carbon Emissions Survey titled Why Some Companies are Ahead in the Race to Net Zero.

Learn more about our people:

Charlotte Deogot: LinkedIn
Diana Dimitrova: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Topics:

CO2 AI and BCG’s Third Annual Carbon Emissions Survey | Boston Consulting Group [2:55]

Resources:

CO2.AI | CO2AI [1:19]
Boston Consulting Group | BCG [1:19]
Klöckner & Co. [6:00]
Nexigen® Data Services (PCF) | Klöckner & Co SE [6:45]
The Economist Group [8:55]
CodeCarbon.io | Code Carbon [14:53]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Asim Hussain: Hello and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

Chris Skipper: Welcome to another episode of Environment Variables. Happy New Year, and I hope you are all enjoying this holiday period and looking forward to 2024. Seeing as we're all on a bit of a break here at the Green Software Foundation, we've got another highlight from the recent Decarbonize Software 2023 event.

We'll be showcasing a talk on driving climate change solutions with AI. From Charlotte Degot, CEO and co founder of CO2 AI, and Diana Dimitrova, managing director and partner at BCG. They are introduced by the wonderful Namrata Narayan, director of communications and member relations at the GSF. So it will be her voice that you hear first.

So, without further ado, here's driving climate change solutions with AI.

Namrata Narayan: Now we're going to hear from another steering member, BCG, and their partner, CO2 AI. I'd like to welcome Charlotte Degot from CO2 AI and Diana Dimitrova from BCG. The GSF is focused on addressing the environmental impact of software. With the rise of AI, we wanted to take a moment at Decarb to discuss how this technology can assist our efforts to decarbonize software.

Charlotte and Diana are here to cover this topic and highlight the results from the Carbon Emission Survey Report that was just released. Charlotte and Diana, over to you.

Diana Dimitrova: Great. Thanks so much. And thanks so much for hosting us today. We're super pleased to be here. And as you mentioned, today was the launch date of our third annual survey that actually found some fascinating insights around green software and how that propels climate action globally. So Charlotte and I are thrilled to take you through it.

So maybe as a first step, we can just introduce ourselves. Charlotte.

Charlotte Degot: Yes, very nice to have us, I'm Charlotte Degot, I'm the founder and CEO of CO2 AI. Just in a few words, CO2 AI is a sustainability platform that helps large organizations on their end to end net zero journeys from measuring and reporting their emissions down to reducing and proving their impact.

Diana Dimitrova: Excellent. And, um, I'm Diana Dimitrova. I'm a managing director and partner, uh, in BCG, and I focus on building digital solutions for clients, um, that advance their climate goals, um, that's specifically in our unit called BCGx. I'm going to kick us off with the main findings, um, of the report, and, and let me just contextualize that a little bit.

So, we publish a third annual report around carbon emissions, uh, and measurement. It includes. 1,850 organizations that reported back to us. It does represent over 20 countries in 18 major industries. And based on the reported emissions, it's about 40 percent of global emissions. And the report is called Why Some Companies Are Ahead in the Race to Net Zero.

Now what we found, the unfortunate bit, which is the first column, is that comprehensive measurement of scope 1, 2, and 3 actually hasn't improved. It was 10 percent this year, it was 10 percent the year before, and 9 percent the year before that, so real stagnation there. Now, we did see some elements that gave us quite a bit of hope, and that was really in scope 3.

So in scope 3 measurement has landed at 53 percent or partial scope 3 measurement has landed at 53 percent which is up 19 percentage points in two years. So we are seeing organizations actually be selective of which categories they measure but really starting to get at scope 3. And more interestingly is they've actually set targets and that's the 12 percentage points that you see on the bottom there.

Organizations at this stage are setting targets for scope three categories at 35 percent rate, which is a notable improvement. And regionally, we are seeing changes and we are seeing certain regions pull ahead in the race. And again, this is a scope one, two, and the comprehensive measurement that they have.

South America, APAC are really pulling ahead when it comes to their measurement. And then finally, the most important bit is why are companies doing this? And 40 percent of the folks who responded are telling us, well, they do this because they're seeing a hundred million or more dollars in annual financial benefit when they get on the reduction journey.

So there's a real financial incentive for them to be tackling these massive challenges. And then let me tell you a little of how they did it on the next slide. So these companies are doing four things twice as well as the average organization. The first thing that they're doing well is they're collaborating with their suppliers, meaning they have workshops together, they have joint targets, they have joint programs.

So they're engineering products together for a lower carbon solution. The second bit is Digital solutions that allow you to measure at a product level, so it's not good enough to just know what Apple's total emissions are, it's actually much more interesting to know what the iPhone's emissions are, and getting at that product level is fundamental to enable consumers to make a greener choice.

And the highest number on this slide, as you see, is the use of digital technologies, and Charlotte will give us a few examples of that, but those that adopt digital solutions or green software really make a dent versus those that don't. And then lastly, I heard our host kick it off, regulation, and those that view regulation in a positive way, catalyze around it, really are seeing those benefits when it comes to meeting their reduction targets.

Now, to give you an example, I'm going to take you through a client where we've done this. The client is called Klöckner & Co. What does Klöckner do? Klöckner actually distributes steel. So steel is a heavy emitting sector. It represents 7 percent of the world's emissions. We all have a stake in the ground to really decarbonize steel.

Unfortunately, there is no net zero steel, but there is a lot greener steel than what we use on average. What Klöckner did is recognize that they needed to get to that product carbon footprint element, which is the second pillar I just talked about. So they actually built a software that was able to calculate their product carbon footprint for over 200,000 products, and you see that's called Nexigen PCF algorithm.

It's cradle to gate emissions. And it covers, as I said, 200,000 products, and it is certified, which makes it really important that whoever gets a number actually gets a certification. And then they decided that wasn't good enough, they needed to push a little bit further, and then they created Nexigen Data Services, which is an online purchasing platform where anybody who's buying steel can log on and look at what the carbon footprint of their basket was in the last purchase, and how, what greener alternative is available to them.

And so this platform allows procurement individuals to actually really arbitrage where they're going to spend their carbon because they're given the transparency and the choice to make the lower carbon solutions. So it's a great example of leveraging technology and getting at product carbon footprints, which is still a bit of a panacea that organizations are gunning towards.

Having said that, I'm going to hand it over to Charlotte and she'll take us through some of the other details on digital.

Charlotte Degot: Yes, thank you. So, as Diana was saying, one of the four key things that companies will succeed better on their reduction journey do is adopting digital solutions. As you can see on the left part of the slide, what we see is that companies who use automated digital solutions are actually two times 2.5 times more likely to comprehensively measure their emissions.

And this is very important because what gets measured gets done. This may look like a basic, but it's a basic that just sets the foundations right for you to then decarbonize. And what's also encouraging is that when we ask the 1,800 companies who answered our survey, what is the number one enabler that they see and that they think they need to adopt to accelerate on emissions reduction, they quote, technology, uh, as the first enabler.

Before leadership buy in, before sustainability focused, uh, culture. So it really means that they have, uh, perceived the importance of digital and that, uh, we can hope for a wave of adoption in the years to come.

I just want to give you one example from real life of what it means to be using tech and how it can bring value to a very large and complex organization. We are talking here about The Economist Group, so the press media group as they are very large and they, um, they issue a lot of newspapers. They are very committed to sustainability and what they do with technology is using CO2 AI to help them steer their end to end net zero journey. So what they use it for is first to measure and automate their carbon footprint calculations. They have a very complex carbon footprint. They need to capture a lot of data points, and CO2 AI helps them collect those data points, structure those data points, and get to a level of granularity in their footprint, which is really good and really helpful to make decisions.

The second thing that they do with CO2 AI is to really set up a road map. And define the hotspots and the reduction levers that go with the roadmap so that they really have a plan and they can cascade the plan across the organization. This is a big pain point that sustainability face team, um, sustainability teams face today.

They tend to have targets, but high level plans and no resources to actually make that happen. Technology can really help broadcasting the information, make sure that operational teams get involved into the decarbonization journey. And last but not least, as we were saying, scope three. So the emissions that come from outside the direct operations of the company, typically the emissions from the supply chain, the suppliers, et cetera, are, um, extremely important.

They are on average 90 percent of total emissions. And those are the hardest to tackle. What The Economist is doing with CO2 AI is mapping suppliers and being able to really prioritize with who, with what suppliers to engage and what to discuss with them to go at a level of granularity. We were talking about product level at a level of granularity, which is good enough to have proper discussions about reductions and, and impact measurements.

Finally, I want to say a few words about AI before taking questions. We talked about tech and digital in a broad sense, which is extremely helpful and will drastically help us accelerate on our sustainability journey. In this field, artificial intelligence is Extremely important. You see here on the page, 30% of respondents who plan to adopt AI, uh, to, uh, steer their net zero sustainability and journey moving forward.

Uh, they see artificial intelligence and, uh, as an enabler and a helpful way on many that different dimension. Uh, a couple of ones are making intelligent decisions. Typically on energy usage or decarbonization initiative. Artificial intelligence is a key lever to automate and increase the quality of carbon footprints.

This is what we do at CO2 AI, and I could talk for hours about that, but there is a huge accuracy issue and granularity issue with carbon footprints. On carbon footprint when made manually, and artificial intelligence can really help solve this topic. And finally, making predictions and making sure that roadmaps ahead, both from a business and from a sustainability standpoint, are optimized.

So we're reaching the end of our presentation. You have two QR codes here to go deeper into either the report or what we do more generally. And I think it's time to take questions.

Namrata Narayan: Thank you so much for sharing. I know we've got a ton of questions to follow up with, but first, I do see we've got a few audience questions, so I'm going to ask you both. The first one is from Jean Luc Bonnet, and he's asking about the baseline. He said two times of 10 companies is not so great, but two in 100,000 companies is a great achievement, so could you comment on that?

Charlotte Degot: I think the question is related to the four levers that accelerate and that multiply by two the likelihood of managing your reduction journey. So what I can say is that what we ask for in the survey is what percentage of companies are measuring the emissions comprehensively. And as Diana was saying, this has not moved and this is stagnating at 10%.

And the other key metric that we asked for in the survey is how, what percentage of companies have actually managed to reduce their emissions in line with their ambition. And here this number is 14%. So I will I will see the glass half empty on this question, and I will say that the baseline is not high enough, and we need to really accelerate the adoption of those four levers, and that will increase the baseline.

Namrata Narayan: Thank you so much, and then we have a question with regards to the carbon emissions survey, a couple which I'm going to you know, group together. So, one is, how many companies were part of the survey, and did the survey find any advice regarding how to decarbonize AI itself, since it's also a cause of emissions?

Diana Dimitrova: Great, so let me take the first part and Charlotte, I think you can comment on AI itself. So it was 1,850 organizations up from 1,600 last year and up from 1,450 the year before, so we have been seeing a steady interest in folks giving us this information back and the sample is hoovering for some of the kind of large organizations that are between 10 and 25,000 employees, and it said based on their self declaration it was 40,000 percent of global emissions, so we're quite comfortable with the sample that came through this year.

Charlotte Degot: And on, uh, uh, how to reduce the footprint of AI, this was not part of the questions we've been asking, uh, on the survey, but what I can say is that, um, there are, uh, many ways to optimize the carbon emissions of any model. Also, tools that exist like CodeCarbon, et cetera, I'm not going to teach anything to this group about it, but clearly, the impact of AI needs to be monitored as any other type of impact, and it needs to be used wise.

Namrata Narayan: Well, Charlotte and Diana, thank you so much. There are several more questions, so we might have to come back to you at a later time to get them answered. Thank you so much for joining us at DeCarb.

Chris Skipper: So, that's all for this episode of Environment Variables. If you liked what you heard, you can actually check out the video version of this on our YouTube channel. Links to that, as well as everything that we mentioned, can be found in the show notes below. While you're down there, feel free to click follow so you don't miss out on the very latest in the world of sustainable software here on Environment Variables.

Bye for now!

Asim Hussain: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show and of course, we want more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. Thanks again and see you in the next episode.

The Week in Green Software: Google, Grids & Green Software

28 december 2023 | 40 min

TWiGS host Chris Adams is joined by guest Savannah Goodman from Google, to bring you the latest news and updates from the world of sustainable software development. They discuss insights from Google’s Sustainability report, the need for a fossil fuel free internet by 2030 and the importance of microgrids and nanogrids. They also highlight Google's sustainability tools including Cloud Region Picker, ActiveAssist and Carbon Footprint.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Savannah Goodman: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Topics:

We need a fossil fuel free internet by 2030 | Branch Magazine [6:38]
Tesla Megapack | Microgrids & Tesla [12:27]
Google Sustainability Goals | Google [21:19]
Google & Fervo Geothermal Partnership | Google [22:58]
Google Cloud Region Picker | Google [26:44]
Google Active Assist | Google [29:20]
Google Carbon Footprint | Google [30:13]
The Realtime Cloud Project | Green Software Foundation [32:15]

Resources:

Octopus Energy [15:04]
The Week in Green Software: Modelling Carbon Aware Software | Environment Variables episode with Iegor Riepen at TU Berlin [20:07]
UN Global 24/7 Compact | UN [24:46]
Vincent Poncet LinkedIn | Linux Foundation [32:08]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Savannah Goodman: [00:00:00] While it's important for us to reduce our own carbon footprint, we think the opportunity to provide these flexibility services to the grid will actually help drive broader system decarbonization and allow for broader carbon emission reduction.

Chris Adams: Hello, and welcome to another episode of This Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. your host, Chris Adams. when we talk about green software, whether you're diving into papers about carbon aware computing. or hearing someone talk about GreenOps, a new [00:01:00] variation of the term DevOps, or even as a consumer, looking at the work that the company's been doing to make it easier to understand how you can reduce the carbon footprint of things like flying. It's hard to escape the work of Google when you look at climate and tech. be honest, it's not that surprising. They operate at a scale that's hard for many of us to really comprehend. Alphabet, the holding company for Google, has a market capitalization of 1. 75 trillion dollars these days. And they use enough digital infrastructure to really care. Their own sustainability report in 2023 shows that Google used something in the region of 22 terawatt hours of electricity last year. This is not a small company. Since 2007, they've generally been one of the leaders when it comes to decarbonizing infrastructure. And recently, they set a public goal to match every hour of power they use with an hour of generation from carbon free sources by 2030. This is something we've seen a [00:02:00] number of organizations get behind, and even the UN with their recent 24 7 compact. So we know there's a lot of smart people working at Google, working on the greening of digital. And today, I'm joined by Savannah Goodman from Google to talk about the finer points of doing green software at scale. Hi Savannah, the floor is yours to introduce yourself.

Savannah Goodman: Thanks, Chris. Hi, everyone. I'm Savannah Goodman. I lead the data and software climate solutions team at Google, and I'm super excited to be here. I'm an energy nerd at heart, and I've been studying and working in the climate space for the last decade.

Chris Adams: Thank you, Savannah. Okay, so if you are new to this podcast, here's a quick reminder of how it tends to work. Uh, When we mention projects or papers or any links like that, we'll add them into the show notes. And if there's a thing that you heard us talk about that you'd really like to learn more about, please do leave a comment.

It really helps other people who are trying to actually learn things about this as well. The other thing I should probably do is introduce myself. My name is Chris [00:03:00] Adams. I work at the Green Web Foundation, a nonprofit focused on reaching an entirely fossil free internet by 2030. And I'm also one of the chairs of the policy work group inside the Green Software Foundation, which Google is also a member of. All right. I think that's pretty much it. Savannah, if you're happy for us to go and sitting comfortably, shall we start?

Savannah Goodman: Yeah, let's go for it.

Chris Adams: All right. Okay. So we're going to talk about some of the cool carbon intelligent computing stuff that Google does shortly, but before we dive into there, I figured it might help to talk a little bit about how you got here first, because as I understand it, you've worked with technology that interacts with grids at various scales in your career.

And if we start at the small end, I realize that if you're going to be talking about data centers, it helps developers to understand some of the principles that are behind the kind of power grids in the first place. And this is something you worked on before. And I think what I learned about was there was a project that you worked [00:04:00] on previously with the idea of an off grid internet cafe in a box.

This seemed too cool to not ask you about basically. So I should ask, how did this come about before you worked at Google?

Savannah Goodman: Yeah. Yeah. Thank you for asking. It was a super interesting project and happy to get the chance to start small and then go all the way to data centers. But yeah, I was in grad school and I was interning at a company that was focused on improving energy access in East Africa. And they were working on a couple of different products for pretty specific applications and they were really tailored to the needs of the local communities that we were working with.

So just a few examples, they had very small microgrids, you'd maybe almost call it a nanogrid to provide lighting at night or to charge a few cell phones. So really small kind of mobile solar and storage. And during their research they identified one of the other kind of key pain points in these [00:05:00] communities was the need for better and more reliable internet access.

And while some internet cafes existed in the cities, they could be cost prohibitive, they may be very far from some of the local communities, and oftentimes they were actually unreliable because of grid reliability issues in the region. And so my job was essentially to do the research and propose what an off grid internet cafe in a box might look like as a product, and how it could actually solve the pain points in these

from these local users. And so I did a lot of really interesting landscape research, developed some product specs, and ultimately also came up with the business case. And I think what I really liked most about this project was the fact that it had a ton of potential impact. I was really excited about the idea of enabling information access for everyone, not just folks in developed countries and big cities, and especially through more [00:06:00] reliable clean energy access.

Chris Adams: It's actually timely that we're having this conversation now, just after COP28, because we've seen a few specific announcements that tie into this, actually, things like this goal of tripling renewables globally, for example, by 2030. And there's actually, it's interesting that you talk about some of those things there, because T that was maybe a few years back, and I want to ask you, if you were to come back to this, were there any trends that you saw back then that you might have seen playing out that you'd probably look for immediately, or you'd probably want to dive into if you were coming back to this?

Because you mentioned some things like nanogrids, and the idea of moving past basically fossil based generation, because we know in lots of places, like for example Nigeria, You still have two thirds of the power is coming from generation rather than the grid. So it's extremely dirty power compared to what you have now. Maybe I'll just give it a bit of space for you to say what you would be looking for, what you're excited about seeing over the next few years in this particular field.

Savannah Goodman: Yeah, thank you. It's definitely some really interesting trends and a [00:07:00] lot has changed even since I was in grad school. And I think one of the first things I remember thinking about is how big, right? What is the optimal size of a microgrid? I mentioned this term nanogrid, which was, like I said, a few solar panels, a small battery that was very mobile.

I think that that sort of solution can target a few very specific and acute pain points, but didn't necessarily, I think, scale to the level that these companies were looking to have, especially given the impact that they wanted to drive. So I think one of the interesting trends we've seen is maybe moving from these nano grids towards more true micro grids that can enable multiple applications and are not necessarily mobile, but actually require more robust grid infrastructure.

And so that's something very interesting and especially in the context of the developed region, especially in the U. S. where I'm from, grid infrastructure is a very big focus right now. And I think it'll be [00:08:00] really interesting to see how we can take some of the learnings of how we've grown our own grid infrastructure and hopefully leapfrog in some of these areas that don't quite have the existing distribution or transmission build out.

Certainly a challenge as I don't think we've fully figured it out for ourselves, but there's definitely a lot of lessons learned along the way. So I think that's . Something really interesting that will play out. Another area that always really stood out to me was the energy and water nexus. And this is something that is really important in a lot of developed countries where there's also a lack of clean water access.

And there's been a lot of kind of technology development over the last decade, and I'm sure there will continue to be. And I think the challenge with this is more, again, the kind of go to market of those products. How do we actually ensure that these products fulfill really specific needs and have an appropriate business model that actually meshes well with the type of [00:09:00] communities that need it most.

So that's an interesting one that I'm really curious to see how this will continue, especially as there's more and more focus on the global south and adaptation for climate change, which is closely tied both to water and energy as well. And then the last thing.

Chris Adams: Oh if I can just stop you there, you said you introduced this term, which some other listeners might not be so familiar with. You mentioned the energy water nexus. Could you maybe expand on that a little bit? This is the idea that basically generating water uses energy, but there's also a flip side to that.

Could you maybe expand on that? Because I think that's quite an interesting topic

Savannah Goodman: Yeah.

Chris Adams: up some people's ears actually.

Savannah Goodman: Yeah, definitely. And, um, I think there's, yeah, this sort of exciting concept of the fact that energy and water are two absolutely necessary resources for thriving communities. Like you said, you can, you need usually energy to make water accessible, especially to make it clean and to make it drinkable and portable, right?

There's usually some sort, usually using [00:10:00] electricity to do that can be really effective. There's other ways to do it without electricity, but it's a lot more challenging to have infrastructure that scales and is robust enough. The flip side too is water can be a really great energy source. We think of pump storage and hydro, and that's really important for providing clean, dispatchable resources.

Of course, hydro is very dependent on the weather and the climate, and as climate changes, it can become actually a less reliable resource. And so there's just a lot of interesting intersections there. We can also think about the climate impacts and droughts and how that might impact both water access for drinking, but also for energy.

So, it's a generic term, but it's intended to capture the fact that there's actually a lot of intersection and dependencies between these two fields both ways, and I think it's a really important focus area, right, as we think about you know, impacts on the climate and how communities are going to need to adapt.

Chris Adams: Thank you. That's really useful. I [00:11:00] didn't, I've been looking for a term to describe some of this and this immediately makes me think of, say in France or in Germany, where we saw in the heat wave last year, for example, we saw some of this, we saw the reliance on water from things like say nuclear power stations in France, for example, which were essentially. They end up having to come offline in some cases simply because they didn't have enough access to water to actually keep them cool, for example. And this is something that a lot of us don't really think about necessarily in the first place. Okay. It's power, but where does the water fit into this? But flipping, that's actually a really good example of some of this. Okay, cool. So we spoke a little bit about grids. You mentioned that there's like nano grids and say slightly larger grids, like a micro grid. It's worth moving to there, actually, because I understand that when you were working at that company, then you've also worked at Tesla working with microgrids. And when most of us think about, say Tesla, they might think about electric cars more than microgrids. And in fact, I suspect that most developers might not have heard what a microgrid actually is. So for people who are not familiar [00:12:00] with the term, you maybe just explain what a microgrid might actually be? And just to give people some context, because we'll talk about how this relates to data centers a little bit later, but in the short term, maybe you could just briefly provide a kind of microgrid 101 and why you might care about it, why an electric car company might care about some of this, because that will probably be useful context when we talk about data centers and how they integrate with grids and stuff like that.

Savannah Goodman: Yeah, absolutely. So a microgrid is a localized group of energy sources and loads that can typically operate independently of a larger grid system. And so this sort of independent network of resources can be useful for a few reasons. First, microgrids can improve energy security and reliability. They tend to be less vulnerable to large scale outages because they can usually kind of disconnect or what we might call island and continue to [00:13:00] operate even if the main grid system goes down.

So this is especially beneficial for critical loads. So oftentimes hospitals or emergency shelters may have their own kind of generators and may be able to create their own sort of micro grids. Further, microgrids can actually help reduce, you know, the cost in carbon, uh, depending on the situation. So, for example, a lot of island nations typically have some form of microgrids.

Because they're not physically connected to large land areas that have the privilege of having these large scale grids and a lot of infrastructure and robustness. And so while a lot of these island microgrids have historically relied on diesel generators, they can instead integrate wind, solar, and batteries, which can then actually reduce the cost compared to diesel, importing diesel, and will actually reduce the carbon emissions.

So that's another big advantage of, or opportunity for microgrids. And to answer your other question on [00:14:00] why companies like Tesla might have incentives to build batteries for the grid, it's really a way for them to scale their battery production for EVs. So Tesla Energy was created actually a while back and it really helped drive down the costs

of manufacturing batteries at scale, and it also has the, you know, added benefit of actually creating new business value, new business lines for the company. And the other thing I would say is there's also some really interesting synergies between electric vehicle charging and using grid connected batteries to manage consumption spikes and, and some of the charging costs.

So when Tesla deployed the supercharger network, they also looked at deploying grid scale batteries at those charging networks to help manage the costs that they were paying to the electric utilities. So it not only can be a business line, but batteries for the grid can be used to help manage their overall kind of infrastructure costs for the [00:15:00] EVs.

Chris Adams: Ah okay, so that makes it a bit clearer for me to understand, and also this makes me think of some of the stuff we've recently seen in the UK, for example, there's a company called Octopus Energy, they're an energy company, and they basically They have this new deal working with people who are building homes and they have this kind of this deal basically if you have batteries or renewable energy fitted into the house when it's being built they're offering zero energy bills for the next for the first five years as long as you're able to make the house integrate into the grid and it seems to be like you said because sometimes energy is expensive when there's, it might be cheap when there's lots of renewables on the grid.

For example, it's taking advantage of some of that and storing energy when it's cheap and then using it locally rather than having to buy expensive energy from the grid and vice versa. Ah, okay. That helps me understand that now. Thank you. you said something else, actually, about the fact that you've got grids which are interconnected.

So they're independent, but they're connected in some ways because there's a larger thing. This made me think a lot about the internet, actually, how the internet is [00:16:00] basically made of a series of smaller networks which are interacting with larger networks. So there's some parallels there. And I can imagine how some of these ideas might scale all the way up to something like a large hyperscale data center. I suspect there's probably a bit more to it than that. And I guess this is probably where the work that you've been doing with Google might come in, actually. So for folks who are following along for the ride, when you're talking about, say, hyperscale data centers versus micro grids or small grids here, what are the differences when you're working at that different kind of scale, for example?

And is this idea that, okay, you've got a series of small grids connected. That's a little bit like the internet. Is that a comparison that you could really make? Well, maybe you could just expand on that a bit more. Cause that's what I immediately think about when you talk to me about series of grids connected to each other, for example.

Savannah Goodman: Yeah, I think that's a great analogy. Like you said, the internet is made up all of these different kind of nodes and is really a network of nodes and connection points. And that's [00:17:00] exactly what the grid is. And some areas are better connected, right? They have more interconnectivity and nodes, and that can make them more reliable than others.

Also similar to the internet, because there's this network and different demand and supply pockets. There can be congestion of just like there may be network congestion for internet or data transfer. There can be grid congestion for the flow of electrons. And so keeping this in mind, as I alluded to earlier, there are a lot of opportunities to really better optimize the grid when you have

flexible resources, whether it's EV charging or storage for the grid. This flexibility makes it a lot easier to manage the grid during peak times. The example you, you mentioned with Octopus, I think is a great one, where they're fully recognizing the value of having flexible grid to the point where they won't even charge you for the energy because they know that's really important for the way they manage their portfolio.

And so for data centers, the way that we [00:18:00] contribute to this load flexibility is we can actually shift our compute load. And we can do that in response to signals from the grid in specific locations, where there may be congestion or during kind of the peak times of the grid. And this is what we would typically call demand response.

And just,

Chris Adams: Ah, okay.

Savannah Goodman: yeah, and it's- Demand response has actually been around for a long time, but what's really, I think, new and innovative about what we're doing at Google is leveraging compute demand to be able to participate in these demand response programs. Historically, demand response has typically been from industrial factories who are turning down their manufacturing or from thermostats, right, who are turning down the heating.

But what's really great about compute load is that it's a virtual load. And so the fact that we can shift it not only in time. But also in space is really, um, the unique part of, of, um, compute. And we've [00:19:00] actually been shifting compute at, at Google for a few years now. Historically, we were shifting in response to a carbon intensity signal from the grid.

In order to minimize our own carbon emissions. And for demand response, we do it a little bit differently. We will typically receive a notification from the grid operator, or we'll agree on the local peak time. And then we dispatch our kind of global compute planning system to overwrite that existing schedule that was carbon optimized.

And that basically limits non urgent compute tasks for the duration of that event at the data center. And I think we're really excited about these capabilities because while it's important for us to reduce our own carbon footprint, we think the opportunity to provide these flexibility services to the grid.

Will actually help drive broader system decarbonization and allow for broader carbon emission reduction through enabling the increase in clean [00:20:00] energy and just being able to turn down load when the grid is being supported by gas peakers, for example.

Chris Adams: I see. Okay. So for people who may have listened to this before, we, we did a podcast with, I think, I believe his name is Igor Repin at the Technical University of Berlin. He was talking and going into quite a lot of really nerdy detail about how some of this stuff was modeled on the European grid to explain this and saying, if you're able to smooth out these kinds of spikes, then you don't need to actually have quite so much infrastructure in the first place.

Or it may be that you don't need to have things like, say, as many peaker plants, often tend to be very carbon intensive and tend to burn a lot of fossil fuels. I see now. All right then. So maybe we can actually just talk a little bit about this being part of a of broader strategy. So we spoke a little bit about there being a kind of target to be entirely fossil free by 2030, for example. Is it a chance we could maybe just dive into some of that a bit more? Because for most of us U understanding [00:21:00] why you'd have a target to be running entirely a fossil free energy by 2030 might not be obvious to everyone. And sorry, I think the term that Google uses is carbon free, but basically this idea of you want to have things running 24 7 rather than just saying, having an annual kind of claim, for example. Could you talk a little bit about some of that and what some of the thinking behind that might have been?

Savannah Goodman: Yeah, definitely. So As you mentioned, Google has two main climate goals. One of them is to be net zero by 2030. The other is to be running on 24 7 carbon free energy by 2030. And just to clarify too, 24 7 carbon free energy is much more complex. Than the annual matching schemes that have been most common to date, because we're essentially moving from global annual matching to local hourly matching.

And so you can imagine how, especially over a global, uh, system, how complex that gets and there's no playbook. But we see these goals as a way to actually help scale new [00:22:00] global solutions that drive broader system wide decarbonization because we're actually aligning our own goals with what the grid needs.

Through these hourly local matching, that's how the grid operates, right? It operates, you have to have local constraints. You have to match supply and demand every hour. And so we've seen research from folks like TU Berlin and Princeton and the IEA that show 24 seven procurement is one of the best ways.

For corporates to help accelerate the energy transition towards grids operating on clean energy every hour, every day. And for 24 7, load flexibility is really complementary because it provides this nhe gew sort of flexible resource that can help us better match the clean energy that's on the grid or the clean energy that we procure on a local hourly basis.

So we're also looking, besides load flexibility, at other new next gen technologies like geothermal. We've [00:23:00] announced the starting operations of our geothermal plant with Fervo in Nevada. We're also really excited about battery storage. And there's a lot of other technologies that we're looking at, and research shows that having this diverse portfolio, both of Load flexibility and next gen technologies is what can make 24 7 more cost effective and more accessible and actually meet the grid needs, especially when we consider rising demand in electricity from things like electrification and data center growth.

So, so yeah, all of this demand response effort is really a key part of our kind of sustainability strategy, and we've recently announced a couple of pilots to prove out that this is real, this can work, it has benefits for us, it has benefits for the grid, and we're going to continue to evolve our capabilities and work with our utility and grid operator partners to make sure that we're maximizing the shared impact [00:24:00] of this system.

Chris Adams: ah, okay. So it sounds like rather than just we're saying I'm going to buy a green energy tariff and that, it's all more I want to shift the entire paradigm that the grid actually operates on. And I know that Microsoft have also come on board, but they're not going to call it 24 7, of course, because that's what's called by the competitor.

So they're using the term 100, 100, 0. I think it's 100 percent of the time. Coming from a hundred percent renewable energy or zero carbon. It's something like that. It's not as easy to remember as 24 seven, but that's one thing we've seen. And we've also seen the federal government. I believe they've made us, they made it last year, actually saying they're aiming to have by 2030, 50 percent of all of their power. And that's the entire federal government, not just a single company, for example. So that was like another example of this. Okay, cool. And I think we might have alluded to this. This is actually a kind of wider scheme. The UN has this global 24 7 compact that any organization can sign on to and get on board with as well, I think, right?

Savannah Goodman: Yeah, exactly. And we're really excited to see during COP that the US government officially also [00:25:00] signed on to the compact. There's over 100 signatories at this point. And I think what's most exciting to me is that it is really a community of different kinds of companies and organizations from all across The energy sector that need to come together.

So there's some energy buyers. There's energy suppliers. There's governments. There's cities. There's software and data providers. There's hardware providers. And I think we're not going to do this alone. But what we're really excited about is creating this ecosystem And developing technologies to advance that ecosystem so that we can all collectively work together to meet the ambitious decarbonization goals for the grid that are really needed to enable the broader climate targets for 1.5 degrees.

Chris Adams: I see. Okay. All right. Thank you for that. So we've spoken a little bit about data centers and how flexibility there can actually have some kind of impact inside this. And we spoke also a little bit about, [00:26:00] uh, the kind of wider context of why you might think about this in terms of decarbonizing a grid. This gives a bit of visibility there, and you spoke about some internal things. Maybe we could talk a little bit about some of the things that end users of services that, say, Google might offer, might use, because we've got a lot of developers who use, say, Google Cloud Platform or even other such platforms as well, and I know that I've used some of Google's tools previously, and I know that there's some projects inside the GSF, the Green Software Foundation, that talk about this and are contributing to that, but maybe you could just tell touch on some of the tooling that you've seen in use or made available for end users, because there's a couple of cool things, which I think are worth develop, worth some of our listeners not knowing about.

Savannah Goodman: Yeah, absolutely. And just to set the context right now, as part of our Carbon Intelligent Compute or our demand response programs, we're not actually shifting customer workloads. We're focused first on shifting some of the internal workloads that operate Google [00:27:00] products like YouTube videos and things like that.

However, we're very keen on exploring with customers who are interested in reducing their footprint to shift their customer workloads. In the meantime, we've developed some tools that can help these cloud customers reduce their carbon footprint themselves. And so, just to talk through a couple of the different tools, one of the first ones we launched was called the Region Picker.

The Region Picker allows customers to look at all of the different kind of characteristics that they may want to optimize for, in particular, latency, cost, and then carbon footprint. And based on,

Chris Adams: Hmm.

Savannah Goodman: you can essentially adjust the weighting of those different Aspects, depending on what is most important for you and your business.

And based on that, the tool will actually provide you a optimal region or a set of optimal regions to site your new workloads in. And this tool, we've also embedded [00:28:00] the essentially green leafs into the cloud platform. So when customers are choosing regions, they can actually see which ones are cleaner.

by based on meeting a certain threshold of carbon free energy. The reason we started with the region picker tool is through our own analysis and data we've seen that one of the most, pretty much the most impactful factor when siting workloads is location. So, eventually we'd love to be able to shift workloads in time as well for customers, but this spatial sort of shifting or site selection is really impactful because customers can move their workloads from a dirty to a clean grid, and that makes a really big difference.

Chris Adams: So the region picker, for people who haven't seen this before, as I understand it, when you're using this, it basically gives you an idea of saying if your audience is in Germany, rather than running it in say, one part of there, you might want to consider looking at Switzerland. Who are still in the same place, but have much, much cleaner power, for example, [00:29:00] and still would be staying inside your latency requirements is tools like that.

And that's the kind of stuff that I, uh, that I saw. And I think that's the first time I've ever seen any large organization sharing some of that stuff. So that is included in active assist now, or some of the tools inside Google. Is that what you were saying?

Savannah Goodman: Yeah, that's exactly right. That's part of our region picker tool. The active assist tool can help customers reduce their footprint in a slightly different way. So the focus of the active assist tool is, let's say customers have some projects that have just been running in the background. The ActiveAssist tool can recommend automatically certain optimizations, could turn down a project, or minimize the runtime of a project, which will not only help reduce cost, but also reduce carbon.

So the ActiveAssist tool is using machine learning and AI to serve predictions and recommendations for cloud customers.

Chris Adams: So the active assisting, so it's a bit like, so yes, there's AI and things, but it also just tells you, by [00:30:00] the way, mate, you've left your computer on, or you've left this project running. Maybe you want to turn it off if it's not serving any traffic, because this is one thing that comes up. It'll basically do some stuff like that, as well as providing some specific, much more tailored recommendations as well.

Yeah?

Savannah Goodman: Yeah, exactly. That's right. And then the last tool that we've developed is called Carbon Footprint. And this is really a reporting tool. There's also some kind of insights that customers can glean to optimize how they're setting up their infrastructure. But this is the rounding out the suite of tools we developed.

It helps customers understand what their actual carbon footprint is from the use of cloud services. So as they make changes, They can see how that impacts the trends of their carbon footprint over time, and they can also use the data for their own corporate reporting. Usually the use of cloud falls into a customer's scope three, indirect emissions, and so this is an important, an important [00:31:00] reporting tool for our customers to be able to meet their reporting needs.

Chris Adams: I see. Okay. And this is something that is just built into the system from day one, or is it part of something you need to purchase separately, for example?

Savannah Goodman: So this tool is available and free for everyone in the cloud console, can access it directly and it's organized per billing account. You can see pretty granular data too, per month, per project, per region. So there's a lot of different ways you can slice and dice the data. There's also a data export, right?

If you want to integrate that data directly into some of your own dashboarding and marry it with other kind of cloud operations data. So yep, it's free and available for everyone on Google Cloud.

Chris Adams: Cool. Thank you for that, Savannah. All right, then. So I know that you work for a specific vendor, and I am mindful of us spending too much time talking about this, because I know that, I can see why it's useful for one company to share this visibility. And from the perspective of a non profit, I'm glad that there's one company pushing this stuff and has been. forthcoming sharing this information compared to some of the other providers in this [00:32:00] kind of space. And I know that from speaking to one of your colleagues, I think Vincent, I never pronounce, I'm going to pronounce, mispronounce his name, Ponset or Ponset. He's also, he's been involved in some of this as well.

And I met him at the Linux Foundation at LF Energy Summit in, in, in Paris, basically. I understand that there's a project called the Green Software Foundation Real Time Cloud Project, which is, being led by both Adrian Cockcroft and Pindy. I, oh, Pindy, I totally forgot your surname, but there's, we have a couple of people inside the GSF who are members who are leading on this, maybe we could talk a little bit about why that's important and why as a member of GSF, you've been involved in some of these projects to create some kind of consistency and conventions around this data that gets exported or exposed to customers.

Savannah Goodman: Yeah, definitely. So we've heard from our customers that the data we're providing is super helpful, but they also need more, more transparent data that's consistent and comparable across their cloud providers. [00:33:00] Often customers will have multi tenant use cases and will have multiple cloud providers. And so it's really important that we can provide more accurate data, more transparent data and comparable data to help them, like I mentioned, with their reporting needs.

And to also make better informed decision for taking action within the cloud environment on how to reduce their carbon emissions. And so we're really excited to be working on the cloud project with Green Software Foundation. Um, that's looking at two main use cases for the project. The first one is this emissions reporting, and the second one is really carbon optimization.

Ideally, these two use cases would be tied closely together, but in reality, because of the way the current accounting and auditing systems work, the reporting Use case typically has a significant lag in the data. And so we're hoping that this project will enable better standardization across metrics for better comparability for reporting.

But at the same [00:34:00] time, we think there's a lot of opportunity to provide more real time, more accurate data for the optimization aspects. And so we're hoping to develop methodologies and tools that can enable cloud providers to offer this more granular and realtime Data. That will support the carbon optimization use case in the near term and then ultimately with the goal of reducing the lag of carbon reporting in the long term . So we're super eager actually to explore working with our cloud customers and seeing what tools are most useful for them, what metrics, how can we help them optimize their carbon footprint, whether it's in an automated way.

Through our own carbon intelligent compute system or through the tools that we're providing through the console.

Chris Adams: gotcha. Okay. Thanks for clearing some of that up. And I know that because I'm also in that working group right now, I can actually speak to you and say, quite honestly, that some two years ago, Google started publishing some of the information at a kind of region by region basis, showing [00:35:00] like the carbon intensity or like the amount of power that is considered carbon free for every single region, um, sharing it as like a CSV on GitHub. This has actually been really helpful to provide a kind of starting point for sorting out some kind of consistent metric, consistent way of reporting this or requesting this from different providers. Cause we know that Amazon has a cloud calculator and Google has a cloud one and Microsoft have one, but having a consistent thing to refer to has been really helpful for this and it's made it quite a bit easier to then say. For example, I'm glad that Google has been one of the earlier companies to talk about information, emissions across all three scopes, scope one, two, and three. And this is something that we don't have from all the providers right now. So being able to point to an example and a data set has been really helpful in this scenario here. I wanted to ask you, coming to the end of the time for this chat, and we've spoken a little bit about like changing some of the kind of paradigms around, rather than basically just buying green energy, like essentially changing how [00:36:00] people think about using power to actually incentivize more carbon aware or approaches for reducing the emissions associated technology, for example. I want to ask you, is there anything that you'd like me to draw attention to? Or you, you reckon we should talk about as well, because we've covered quite a ssions ssions lot of ground from like nanogrids all the way up to ginormous data centers that use as much power as small cities, basically. So what else are you looking at or what else would you like to draw people's attention to right now in this field?

Savannah Goodman: Yeah, I think overall we're super excited about the potential for green software to reduce emissions. I think there's a lot of different ways that we can go about making software more green, whether it's through our hyper scale carbon intelligent compute type platforms or even individual developers choosing cleaner regions for their workloads.

We're really optimistic about how these solutions can come together to enable broader capabilities across the GSF community and beyond. [00:37:00] We think that load flexibility is such an important part for the future of the grid and we think software has some really unique Flexible capabilities that other kind of loads don't necessarily have.

And so, that's why we're really thrilled to be part of this community to see how we can really maximize the, you know, level of flexibility that green software can offer to really help drive the energy transition at the broader grid scale and reduce carbon emissions.

Chris Adams: Brilliant. All right, then I think we're coming up to the end of our time. So I just want to ask if people have, if their interest has been peaked by any of this, where would you suggest people look, or if people were interested about what you're doing specifically, where would you suggest people look to follow on for updates about what's taking place in this field, beyond the projects we just mentioned, like the real time cloud thing in GitHub, or like Google's account on GitHub, for example, or some of the other things. So maybe, let's say I'm a developer. I'm [00:38:00] curious about some of this. I don't know where to look. Where should I be looking to learn more about this or continue my research?

Savannah Goodman: Yeah, definitely. So we publish quite a few updates through the sustainability topics in the Google Cloud blog where you can see the updates on the latest work and not just for green software, but a lot of the different sustainability areas that we're developing new technologies for. If you're interested in learning more about kind of Google's overall strategy, not just cloud specific, Sustainability work, then google.com/sustainability

. It's a good place to start. That's where we publish all of our papers and annual reports. But yeah, otherwise, we definitely recommend the Google Cloud blog and following the sustainability topic. Lots of exciting updates to come.

Chris Adams: Brilliant. Okay. And what we'll do now is we'll just try. And if, if you're a listener and you've listened to some of this and you caught your eye, we're going to take a moment now to just get as many links as possible to all this stuff that we've spoke about, because we've covered a lot of ground. All right. Savannah, I've enjoyed this. I've learned [00:39:00] a lot and I think our listeners probably have as well. I want to wish you have a lovely week and yeah. Have a lovely winter break. If you're, if you celebrate your time away over winter.

Savannah Goodman: Thank you so much. Have a good one.

Chris Adams: Okay. Cheers.

[00:40:00]

Responsible AI

21 december 2023 | 20 min

From the recent Decarbonize Software 2023 event, this episode showcases a fireside chat on Responsible AI with Tammy McClellan from Microsoft and Jesse McCrosky from ThoughtWorks. Jesse shares his thoughts and experiences from years of working in the field of Sustainable Tech on the topics of risks, sustainability, and more regarding AI, before answering some questions from the audience.

Learn more about our people:

Sophie Trinder: LinkedIn
Jesse McCrosky: LinkedIn | Website
Tammy McClellan: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Events:

Decarbonize Software | GSF

Resources:

AI Transparency in Practice | Mozilla Foundation [04:19]
Reducing bias and improving safety in DALL·E 2 | OpenAI [09:31]
Responsible AI: Fireside Chat | Decarb 2023 [18:41]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Asim Hussain: Hello, and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

Chris Skipper: Welcome to Environment Variables. Today, we've got another highlight from the recent Decarbonize Software 2023 event. We'll be showcasing the fireside chat on Responsible AI from Jesse McCrosky, Head of Sustainability and Social Change and Principal Data Scientist at ThoughtWorks, and Tammy McClellan, Senior Cloud Solution Architect at Microsoft and Co-Chair of the Community Working Group and Oversight Committee at the Green Software Foundation. They are introduced by our very own Senior Technical Project Manager for open source projects, Sophie Trinder, so it will be her voice that you hear first. So, without further ado, here's the Fireside Chat on Responsible AI.

Sophie Trinder: Hi everyone, I'm Sophie, the technical project manager for our open source projects at the Green Software Foundation. Today I'm going to introduce a special fireside chat to our Decarbonize Software event. We're continuing the conversation that began on the 5th of October at our panel on responsible AI. The conversation surrounding responsible AI is dynamic, oscillating between optimism and skepticism. one side, practitioners believe that AI has the potential to drive sustainable development goals, from responsible consumption to waste management and energy conservation. The promise lies in our improvements in measuring software's environmental impacts, and innovation across energy-efficient algorithms, hardware optimizations, and the growing use of renewable energy sources. On the other side, the rapid expansion of AI, particularly large learning language models, and the insatiable demand for this technology, are raising concerns. If left unchecked, the energy consumption and resource utilization associated with AI make many feel like we're endangering a future where software causes zero harmful environmental impacts. To help us explore the path forward, I'm thrilled to introduce Tammy McClellan, Senior Cloud Solution Architect at Microsoft, and Jesse McCrosky, Head of Responsible Tech and Principal Data Scientist at ThoughtWorks. Thanks, both. Take it away.

Tammy McClellan: Thanks, Sophie. And a hello to all you sustainability addicts. Jesse, hello. Let's start the question with, how do you see the relationship between responsible AI and sustainability?

Jesse McCrosky: Hey Tammy, great question and nice to see you all. So at ThoughtWorks we use a framework that I like which we refer to as the greening of tech and greening by tech and I think this is the best lens through which to view that question. Greening of tech refers to the fact that these systems and especially generative AI as we're talking about now have serious energy consumption, they have serious sustainability issues that need to be tackled.

The other side is greening by tech and recognizes the potential that this technology has to actually improve sustainability of other processes, either within or outside of the tech world. And I think what ties these, these two questions together is issues of transparency and information and ensuring that people have the information they need to make the right decisions for our environment.

Tammy McClellan: I like that, greening of tech and greening for tech. It's my new mantra now. So how can, uh, we use this to make more sustainable solutions?

Jesse McCrosky: So it's a big question. To begin with, I think that I refer to transparency, and when we talk about transparency, a lot of people think that means you share your source code, or you share your model weights, and then you're transparent. Or it means you have to explain the decisions the AI is making, and that's transparency. Transparency is more than that. There's a report I did with the Mozilla Foundation on AI transparency, and we talk about meaningful AI transparency that needs to be legible, auditable, and actionable. And this means that we have to consider the specific stakeholders that the information is being provided to, what are their needs, what are they going to do with this information. So it comes down to the old adage that you can't manage what you can't measure. So for example, in order to support meaningful policy, meaningful regulation, we need to have information about the sustainability characteristics of these systems.

Tammy McClellan: So talk to us a little bit about some possible solutions in this area.

Jesse McCrosky: Yeah, absolutely. So when we're looking at solutions, especially using the kind of transparency lens, we can think about who is the transparency being provided to. So, for example, we can talk about consumers. And right now, consumers are very excited about ChatGPT or whatever else, Stable Diffusion, DALL-E, and everything like that. It's a lot of fun to play with. And they do not have meaningful information about the carbon implications of that play. So someone was suggesting to me that ChatGPT should have a real-time counter across the top somewhere that's telling you how much carbon have you emitted so far in your session, how many, you know, gallons of water have been consumed, whatever else. And it is not a matter of just shaming people, but it's helping people make the right choices, because there might be applications for which ChatGPT is really worthwhile to use, but there's other times that somebody's just idly playing or something like that, and if they realize the implications of doing, they might make other choices. This becomes more interesting when we talk about communication between, for example, model developers and model deployers. So, for example, if somebody is using the OpenAI APIs in their product, they need to be able to have information about what the implications are of those API calls so they can make good choices in how they build their software.

Tammy McClellan: So awareness is key, absolutely. So Jesse, what is the potential for Gen AI to support greater sustainability?

Jesse McCrosky: Yeah, it's an exciting question, and I think there is some potential here. There's a case that ThoughtWorks took, it's a couple years back now, I think, in which we worked with a international manufacturing and services company. They were interested in finding solutions to meet their sustainability goals, and they just weren't sure which way to go.

They weren't sure, "should we start sourcing our energy from a different place, or using different sorts of transportation, or using different industrial processes or offering different products?" And so what we did for them was built a mathematical model of their operations and their supply chains. once we had that mathematical model, we were able to build a sort of scenario modeling dashboard where we could show them like, "hey, if you switch to delivery trucks that are using electricity instead of gas, this is what happens to your emissions, this is what happens to your bottom line, this is what happens to your customers."

And likewise, depending on / considering different product mixes, considering different sourcing, whatever else. So the mathematical model here was not rocket science, to be honest, it was fairly simple stuff. The hard part of this engagement was really understanding the business at the level that we needed to in order to build that model. There were many hours of interviews and poring over notes and internal documents and everything else, as well as actually some basic desk research to determine the necessary carbon emissions factors, that sort of thing. I'm excited at the potential of generative AI to make this sort of process more accessible and more scalable. And I think that we've seen evidence so far that these models do a very good job of looking at these sorts of documents, looking at recordings and interviews, and it may be possible that you could create this model semi automatically with far, far less of the kind of very heavyweight and expensive all sorts of interventions. As well, it was challenging to understand the exchangeability. And so, for example, if the company is buying cotton in one particular country, it might be obvious to us that they can instead buy the same cotton from some other country, and that's the only possible change that could be made. But it's not so simple for the model to figure that sort of thing out automatically.

Whereas GenAI, I think when we connect to these sorts of emissions factors databases, has the potential to make this process much easier.

Tammy McClellan: Yeah, awesome. Let's move a little bit and talk about risks. How do you think businesses can manage the risks of AI?

Jesse McCrosky: Yeah, it's it's a big question. I think everybody's talking about this. And I think what I would say is it's critical to understand that risks must be mitigated, not removed. I think a lot of people are talking, for example, about bias and discrimination, and they say, okay, we're going to produce a model that's perfectly fair and perfectly unbiased, or we're going to eliminate this bias from our model or whatever else. And this is just not the way things work. We live in the real world, and these systems are based on data from the real world. And the real world is unjust, and so we need to be able to be ready to tackle that. So, one example that I like is OpenAI with their DALL-E interface generation system. For a while, maybe some months ago, I think, if you asked it for pictures of lawyers, it was going to give you eight pictures of white men, basically.

And OpenAI recognized that there was a problem there, as did the community, of course. So eventually, OpenAI had a short blog post where they talked about how they were going to fix this. And it was apparently fixed, so when people tried to get pictures, they would see pictures of lawyers, and some of them would be women, and some of them would be of different ethnicities, and everything else. So People were curious how this had been fixed and it turned out that all that OpenAI was doing was just randomly appending words like women or black or Asian or whatever else to these prompts and people were not super impressed with this solution but I think it's an important illustrated example, because it's a mitigation, there was a problem with a model, there was a problem with the data, this is not a problem that can be solved fundamentally, it needed to be mitigated, and they found a way, they said, "here's the harm that's going to come from the system. It's going to not be producing an adequate representation, and we found a way that we can show more representation." So this is the sort of mitigation that companies need to take. So when there's issues, and this is where transparency comes in around the carbon impacts as well, so that they can be mitigated, so that if I'm an engineer sitting in front of my laptop writing some software, I need to have awareness that if I call this Gen AI call or whatever else, I have to understand this is going to spike the carbon emissions of my product, and I need to find another solution.

Tammy McClellan: Gotcha. Yeah, that makes sense. Tell me. So are you optimistic or pessimistic about Gen AI at this point?

Jesse McCrosky: I think I'm mixed. I think that ultimately solving the climate crisis means simultaneously solving a social crisis. And I think it's very hard to solve climate change without also solving issues of social justice globally. And I think that Gen AI is a tool that might enable some of these conversations to be tackled in a more interesting way.

So I think as long as we're mindful and honest and clear eyed about how we apply this technology, there can be some optimism there. We need to ensure that we have adequate transparency so that people understand the carbon implications of the choices they're making when they're using these systems, but given that, there is potential to do better.

Tammy McClellan: Gotcha. So I know when you and I chatted before, you said that you had a fun story of AI. Did you want to tell us what that is?

Jesse McCrosky: Ah, so actually, I think there's a misunderstanding. The fun story was an expanded version of what I was talking about before, but

Tammy McClellan: Gotcha.

Jesse McCrosky: if we have a moment, I think one thing I want to add when we come back to the idea of how, how transparency can help Gen AI be used more responsibly. So, a lot of people are familiar with the concepts of DevOps or MLOps or CD4ML, these sorts of processes. And I think this is a really critical place for transparency around carbon emissions to be integrated. I think the point I would make is that right now, a software developer that's working in kind of a modern setup has the ability, as they're writing code, to see immediately if the code that they're changing is causing some test to fail, or is causing some performance degradation, or is introducing some bug or whatever else. And I think we need to have the same process for carbon so that it if an engineer is making a choice and for any devs out there, maybe you have a case where you need to use a regular expression, but it seems like too much work to figure it out. "Hey, I can just call a Gen AI model and it'll do it for me as well."

It'll work just fine. And you might make that choice because it saves you a couple minutes or whatever. But if you then see that all of a sudden your dashboard turns red and says, okay, your carbon has just increased like 100 percent or whatever, you're going to come back and you're going to revisit that decision. And also your team is going to see that, the trail of what's happening because of what you've done. And so it creates this sort of accountability in the development process.

Tammy McClellan: All right. So I'm curious. What are the top three recommendations you would give to people who are interested in reducing carbon emissions of AI?

Jesse McCrosky: Good questions. And yeah, I think that's something I didn't really touch on so far, but there are a lot of choices that can be made when applying AI. So we don't need to use the biggest general purpose models for everything. I think that there are cases where a general purpose model is really needed. But um, I think that in most cases, no. And so we can talk about using much simpler application-specific models. We can talk about using a smaller model and fine tuning it for the particular task. There's processes like quantization and distillation that can make models much more carbon efficient and nearly as effective. So investigating these options, and again, I think this kind of hinges on the MLOps setup where you need to be ready to evaluate performance. You need to be able to say "how small can I make this model and still actually meet the requirements in my product." Beyond that, I think it's a matter of providing transparency to the end user. So if you're producing something, if users understand the choices that they're making when they're using that product, there's a lot of different ways this can play out, and this can mean some Gen AI chatbot or something like that, but this also can be, maybe you have an e-commerce product. platform and you're using AI to make recommendations to your users and the recommendations that you make can influence their behavior and it can encourage them to buy more products that are disposable or made in very carbon-intensive ways, and so considering these sorts of externalities as well is really critical.

Tammy McClellan: Gotcha. I'm curious, do we have any questions from the audience at this point?

Sophie Trinder: Yes, we do. And thank you so much, Tammy and Jesse. It's been a really great session on AI here at Decarb, and it really shows the passion in the industry for these technologies, plus the responsibility that we all must take when it comes to AI. I know we'll be hearing a lot more in the coming months. But yes, we've got a few questions from the audience. I just want to shout out first, Jesse, thanks for the fun story on OpenAI, how they were mitigating the problem with data to show more representation through mitigation. It was a really interesting insight, thanks. So one question from the audience, how important is prompt engineering for improvement of AI efficiency?

Jesse McCrosky: Great question, yes, and it's it's really extremely important, because the energy being consumed by the model is going to depend in some complex ways, depending on how many tokens are coming into it, and in quite a direct sense, how many tokens are coming out of it. So if we can reduce the number of tokens going through the system, we reduce the carbon emissions. And this again, I know I'm sounding like a stuck record, but it really depends on the MLOp setup, where we should be able to test and see how short can we make our prompts and still accomplish what we need to do. And this is both the length of the prompt itself and the length of the output. So for example, go back to that example I was talking about where maybe ChatGPT has a little indicator at the top telling you how much carbon has been emitted in your session so far. Maybe if you see that number growing as you're chatting with it, you're going to say, "hey, ChatGPT, please be a little bit more brief with your answers. I don't need the whole kind of colorful language and going on and on about everything." So yes, it's very important.

Sophie Trinder: Super interesting. Thank you. We've got another one on training the AI ML model, which obviously takes a huge amount of data and processing, which in turn causes a lot of emissions. How do you think that we could best counterpart the same?

Jesse McCrosky: Yeah, good question. And I think that I have an article out where I actually talk about how the comparisons are a little bit overwrought, talking about how training a model is equivalent to driving a car some distance or whatever. I think that, um, the comparison, at least so far, thankfully, is not quite accurate because we have many cars on the vehicle and a relatively small number of models being trained. I think the important thing is to keep it that way. I think the important thing is that we need to encourage use of open models and shared models rather than every single organization in the world trying to train their own LLM. And this is why I would be a strong supporter of open-source models. I think it's nice to see that movement.

I think it's potential. It means that organizations, first of all, save their money, but also save their carbon when they want to be able to explore elements in their business. And there's always the potential for fine tuning, for whatever other tools need to be applied to open models to make them suit people's applications.

Sophie Trinder: Amazing. Thank you. And jumping back to sort of problems on data and representation, we've got another question centered around that. So do you think we should promote digital humanism and ethical AI to raise awareness about the need for sustainable AI?

Jesse McCrosky: Yeah, absolutely. I think we're existing at a moment where responsible AI and such is being discussed everywhere. There's very active regulatory work in many different regions of the world. There's many people in academia, in civil society, and in industry doing this sort of work. And I think that green AI should come along for the ride, so to speak, and it should be an important part of how we think about the risks and the potentials of these models.

So, yes.

Sophie Trinder: Amazing. Thanks very much.

Chris Skipper: So that's all for this episode of Environment Variables. If you liked what you heard, you can actually check out the video version of this on our YouTube channel. Links to that as well as everything that we mentioned can be found in the show notes below. While you're down there, feel free to click follow so you don't miss out on the very latest in the world of sustainable software here on Environment Variables. Bye for now!

Decarbonize Software 2023: Recap

14 december 2023 | 41 min

Chris Skipper hosts this episode of Environment Variables with guests Sophie Trinder and Adam Jackson to discuss the unveilings, questions, and highlights from the recent Decarbonize Software 2023 event. The GSF has announced its Impact Framework during the event that had taken place in November, and together with Chris, our guests discuss the realities of (and dreams for) the state of green software in this recap of Decarb 2023.

Learn more about our people:

Chris Skipper: LinkedIn
Adam Jackson: LinkedIn
Sophie Trinder: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Events:

UN World Environment Day: The Green Software Revolution (Virtual Event) - Monday June 5th [31:27]

Resources:

Software Carbon Intensity (SCI) Specification | GSF [11:02]
Impact Framework | GSF [12:48]
Introducing the Impact Framework | Decarbonize Software 2023 [13:04]
Green Software Practitioner [17:37]
State of Green Software | GSF [21:02]
Green Software Patterns | GSF [29:38]
Opensource / Carbon Aware SDK | GSF[30:43]
Dan Benitah's Projects and Posts [35:52]
Decarbonize Software [38:19]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Sophie Trinder: We've also had some big tech companies adopting the Carbon Aware SDK, which has been great, including Microsoft and NTT Data to shift some of those high workloads, like training machine learning models to a different time in the day, potentially even a different location. So yeah, it was really great to see the Carbon Aware SDK being talked about at Decarb.

Asim Hussain: Hello and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode we discuss the latest news and events surrounding green software. On our show you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host Asim Hussain.

Chris Skipper: Welcome to this episode of Environment Variables. I'm your host, Chris Skipper. And in this episode, we will be doing a recap of the recent Decarbonize Software 2023 event with the Green Software Foundation's Community Project Manager, Adam Jackson, and Sophie Trinder, Senior Technical Project Manager and Relative New Face at the GESF. Hello, Adam and Sophie, and welcome to Environment Variables.

Adam Jackson: Hi Chris, it's great to be back.

Sophie Trinder: Hi, Chris.

Chris Skipper: Great. So before we dive into the meat of this podcast, for our listeners who don't know you, could you please introduce yourself? And let's start with Adam, seeing as you are a well-seasoned veteran now of environment variables.

Adam Jackson: So I'm Adam Jackson, and I'm the Community and Tech Culture Lead at the Green Software Foundation. And once again, it's, I'm very excited to be here. I look after our community working group, and also many of our projects and programs, including our meetup program, Green Software Champions, and our recent Decarbonize Software event. Plus in 2024, I'll be leading the efforts on our Carbon Hack and our GSF Summit.

Chris Skipper: Great. Thank you. And Sophie?

Sophie Trinder: Yes. Hi, I'm Sophie Trinder. I'm the Senior Technical Project Manager at the Green Software Foundation. Excited to be here. Recently, I've been focusing on the tooling pillar of the Green Software Foundation's three pillar strategy, known as the Theory of Change. Alongside tooling, we also have knowledge and tech culture.

Like I say, I've been focusing on the tooling and driving forward the open source projects that sit under the umbrella of the open source working group. This includes tools for measurement, like our Impact Framework, and also tools for reduction, like our Carbon Aware SDK.

Chris Skipper: Awesome. Thank you, Sophie. And how long have you been in the GSF now? When did you join?

Sophie Trinder: Just over a month ago,

Chris Skipper: Just over a month ago. Great. Great. Cool. So exciting things to come. And for those of you who don't know me, my name is Chris Skipper. You've probably heard my voice before on this podcast. I am the producer of Environment Variables. And recently I found out that Environment Variables is the top podcast for 55 amazing listeners on Spotify, thanks to Spotify Wrapped. Just a note that spotify is not the only platform that environment variables can be consumed on. And if you want to listen to it directly from the source, please go to podcast.greensoftware.foundation to listen there. Um, anyway, so, uh, I feel like I've, I've spoken enough about that before we dive into the actual topic of today's podcast, which is the debrief, uh, which will be the title of this episode. Just to remind everyone, everything that we talk about in the show will be linked in the show notes. So, to kick us off, I have a few questions about Decarb 2023, seeing as I wasn't there. We'll be going in sort of a, uh, tit for tat. That's a bit, that's a wrong phrase. I'll be asking one question to each of you at a time, which we, we've kind, you've kindly delegated out to each other. So my first question is, Decarbonize Software 2023 was an overwhelming success in the eyes of the Green Software Foundation. Let's start with some statistics, seeing as we've been talking about statistics, just in my little intro there. How many people attended and how many talks were there? Did it meet the goals of the GSF's intentions with this event? Adam?

Adam Jackson: Yeah. So, uh, well first of all, Chris, no, no pressure that you weren't there live because it was the middle of the night for you. And I wanna start by telling everyone that it is available on demand if you haven't watched it already, at decarb.greensoftware.foundation. But yeah, let's jump into the numbers.

So we actually had five community talks and we had three talks from the GSF itself, announcing new programs. And yeah, we had, I think we had a really brilliant event. We had over 2, 500 viewers and almost 400 live questions and comments. And it wasn't all just "hi from London" and stuff like that. It was actually, there were some really great and deep questions.

Also some challenging questions as well, pushing the boundaries. So it's not just the quantity, it's also the quality of those questions that really stood through. But what's really important to me is that we didn't just connect with folks who are already members of the GSF, but actually brought in lots of new folks from the community. And they're the folks who'll tell their friends and colleagues about green software, as well as getting involved in our open source projects in the future. So, yeah, excited about how we did number-wise, but also excited about the type of people that we met along the way.

Chris Skipper: Wow, amazing. So yeah, I didn't expect to hear that statistic of that many questions. That's really great. And hopefully I know that a few of those questions were answered live on stream. Obviously there wasn't time to answer all of them, but hopefully there will be a source for people to go to. I know there's talk of there being a document or an article to do with the questions.

Adam Jackson: Yeah.

I think it's, I think it's actually published.

And we'll put that in, we'll put that in the notes. But there's an article where we, we pick up on some of the main question themes. Yeah, it's tricky, isn't it? You want to get in all of these different community stories, and we were really careful to get a good balance of different stories from across Green Software. And then we had five minutes for Q& A, and really you can only ever get into two or three questions. And Sophie and Namrata and I were there in the background trying to answer as many questions as we could in the, in the live chat and Asim was there as well. But yeah, we thought it was best to actually look at what, what the main themes were and try and answer as many of those on demand. And, and it's actually also given us some ideas as to the sort of video content we want to create in the future and probably might even spawn a couple of Environment Variables episodes as well.

Chris Skipper: Yeah, well, I was just going to say, if there are any unanswered or burning questions that haven't been answered, perhaps we can rope Asim into answering some on this podcast. So that would be cool. Anyway, let's move on. Sophie, so can you share a standout moment or a particularly surprising insight from decarbonized software 2023 that really stuck with you?

Sophie Trinder: It's so hard to pick just one. But yes, if I have to, one interesting insight that stuck with me and actually that gives me optimism is from a talk given by Diana Ditrimova, managing director and partner at BCGX and Charlotte Degot, CEO, and founder at CO2 AI. So to give a little context for those who may not have joined us at our Decarb event.

Diana and Charlotte shared some of the key findings from their joint carbon emissions report, which is called 'Why Some Companies Are Ahead in the Race to Net Zero.' So, this report is a survey incorporating feedback from 1, 850 organizations across 23 countries. And they estimate that all of the sort of organizations represents about 40 percent of global emissions.

So, we're looking at a good representation of global emissions here. And one interesting insight for me that stood out was that 40 percent of those surveyed estimated an annual financial benefit of at least $100 million for meeting emissions reduction targets. And so often at the GSF we hear the question, "what's the business advantage associated with making greener choices when building software?"

And I think that this stat, it gives a really interesting insight that helps form that answer.

Chris Skipper: That's, yeah, it's interesting that you mentioned that talk in particular. I'm hoping to get them together for an episode of Environment Variables to talk about that in particular, that study, that that's incredible, that, that reach that they have of 40%, yeah, of, yeah, I don't think, I think that's the one thing that that's been covered a lot on, at least on Environment Variables and my engagement with the Green Software Foundation is just having the sheer amount of data is sometimes it was probably the biggest issue out there with any sort of measurement to do with carbon emissions. So, and software in particular, if you haven't seen the Decarb 2023 video, you should go back and check that talk. Cause I think it's only about 10, 15 minutes long. So it won't take a lot of time out of your day if you want to have a look at it.

So.

Adam Jackson: Yeah, we're really hoping to actually make all of that content available in bite sized chunks as well. So, so that people can find that on demand. And one thing I like as well, just to add on what Sophie said, is that Um, that talk was beyond just green software itself, but it was also "how can you use software to further sustainable outcomes?" So that's, that's another thing that we don't directly cover every day at the Green Software Foundation, but a lot of our members and our community are really interested in.

Chris Skipper: Yeah. And it goes to show also the diversity of just how big the community has become in the last sort of two years, just because of the existence of the Green Software Foundation, it's attracted people outside of the software industry, which is fantastic, and I'm sure this probably echoes what Adam is fighting for is we want people from all walks of life to join the cause and come and talk and just increase the dialogue around it.

So that's fantastic. Anyway, so let's move on. I'll pitch this question to you, Adam, given the emphasis on measurement at Decarb 2023, what's one common misconception about carbon emissions measurement in software that was debunked or clarified during the event?

Adam Jackson: It's a really good question, Chris. It's actually quite a challenging question as well. So one of the biggest misconceptions that I see is that it's from people that are coming new to the GSF, is that it's been easy to do measurement at all. It really isn't. Measurement isn't easy today. It's about to get easier.

We're definitely on a journey, but it's important to look at the very short history of measurement. What we have today is some amazing standards, such as the Software Carbon Intensity or SCI standard, which, which actually has been submitted to ISO. Hopefully we'll get an approval and see that in the catalog very soon. But we're still super early days. And we're also going to be talking about, um, some of the, some of the key elements in producing tools that can actually do the measurement for us. It's been very manual working out. We'll see spreadsheets or a few people we've seen have implemented some dashboards. I saw a really great talk from Amadeus, which is one of our GSF members, over the summer at a conference in Berlin, and they'd done some really, really great dashboards, but it had all been very much custom work for them to get to that stage. And yeah, what is great to see is, particularly this year, we're actually seeing a lot of effort put in to measurement and visualization. And if I bring it back to the Decarb event, we did see a wonderful talk from Yvette and Johann. They're from CODE University in Berlin, and they've been building a measurement project. It's really focused on measuring consumption of energy. So if we think about the SCI equation, for anyone that's familiar with that, is the letter E in the equation. If you don't know what the equation is, take the Green Software Practitioner training and it shows you all of that. Anyway, these folks from Berlin, they've actually implemented some visualization. To that energy measurement and it was really great to see because visualization is hard and not everyone agrees on it. And there was a lot of, there's a lot of chats during the event on visualization as well. So, but the important thing is we'll need to see folks like Yvette and Johann. And they'll have to show us what we can do with that visualization. And it's also going to help us build out the capabilities of things like the new Impact Framework, extending it, show us what's going on as well.

Chris Skipper: Yeah, yeah. I just mentioned that data was hard to come by, but yeah, measurement, itself, obviously, is a big challenge. And if you are interested more in the measurement and the struggles of measurement, we actually have a previous episode of Environment Variables with Srini Rakhunathan and Navveen Balani with Asim talking about all the other attempts that have been made at measuring carbon emissions of software prior to the SCI.

So that's a really fascinating episode if you are interested in that. And yeah, it's great that you mentioned visualization as well. I know you love a dashboard, Adam. So that's something you mentioned last time you were on the pod. So I'm just going to bring that up again.

Adam Jackson: Yeah. Actually, one interesting thing is you measure, you mentioned data. Obviously, you need data in order to be able to measure. But also the act of measuring puts pressure on others to produce data. And we do hear, and I'm not going to single out any GSF member, but we do hear, "Ooh, why doesn't company X provide this data?"

And actually a lot of the time it's because people haven't asked for it yet, or people haven't shown that they have a need for that data. So, so I really do truly believe that the more we measure, the more data sources will become available.

Chris Skipper: Yeah, absolutely. Good point. Cool. Okay. Switching back over to Sophie in this game of question tennis. The next question I have is, we just mentioned the Impact Framework and that was obviously the big surprise that was the big mention that was going to happen at this year's Decarb 2023. So for our listeners who don't know what it is, Sophie, do you mind elaborating, please?

Sophie Trinder: Yeah, of course. So, yes, you're right, it was the big unveiling of the Impact Framework in our introduction at Decarb this year. So, the Impact Framework is one of the Green Software Foundation's open source projects. We have been working on it for almost a year now, but it's still in incubation, which is one of the project life-cycle stages.

And that basically means it's still evolving, it's still in its explorative stage. But we did manage to do an alpha release in November ahead of Decarb, which is where we introduced it, and showcased it through a live demo of the tool. So the framework itself is an extension of that software carbon intensity standard, the SCI, that Adam just mentioned.

So just to dig a little bit more into what that covers, the SCI is this equation, which is the energy consumption times the emissions factors plus embodied carbon over a functional rate. So this is the what we should be measuring when considering negative environmental impacts of software. And then the Impact Framework takes that equation and provides a tool into how we can measure that impact.

So if we think about some modern applications, they're composed of many smaller pieces of software, which we call components, running on many different environments. For example, private cloud, public cloud, bare metal, we've got mobile, laptops, desktops. Every environment requires a different model for measurement.

So the Impact Framework provides a framework for running all these different models using model plugins to calculate the output of impact through that SEI equation. So right now at the, the GSF, we've got a standard library of models and a repository of community models, and we envision a future where the number of models just keeps expanding out into the thousands to cover all those different environments.

So it is an open-source project, like I just mentioned we're welcoming contributions in fact, our hackathon, which is happening next February, is Impact Framework focused. So that's where we're hoping to see lots more of these models get made. The opportunities is endless that we see in the ways that we'll be able to start improving the measurement of the negative environmental impacts of software.

Chris Skipper: Great. Yeah, that's amazing. And yeah, I love the community focus of it. I think that we, we talked about the importance of open source in general, and I think everything that the GSF has done has been open source. So the possibilities are endless when we all join in and, and I think the hackathon, which we'll talk about a bit later in this podcast as well, next year is going to be phenomenal, especially when there's a bit of incentive behind it. So yeah, it'll be really interesting to see what comes from that. Moving on, let's talk a bit about the Practitioners course, the Green Software Foundation's Practitioners course, which you mentioned, Adam. This, obviously that's where you can learn more about the SCI, but it's trained over 50, 000 participants in less than a year, which was something huge that was announced also at Decarb 2023. What do you think is driving the surge in interest and how can more professionals be encouraged to participate?

Adam Jackson: Yeah, I've just literally opened the stats. 55, 000 now, so yeah, doing really well. We need another 5, 000 people to do that in the month of December, so we end on a nice round number. And actually, we know that some more people have enrolled as well, so another good thing. One of the cool things we've certainly got some people that have started the training and hopefully they'll, they'll come back to it as well.

But, uh, yeah, it's interesting. If we think about, uh, Chris's talk, that's, uh, another Chris, Chris Howard from EPAM. EPAM is a GSF member. So his talk at Decarb really showed us how important the knowledge, this knowledge is. And why it's so important for both individuals and also organizations. So I've no doubt that a lot of people do take the practitioner course because they're concerned about what has happened, happening to the planet. And they're worried that we're experiencing our hottest year ever, which was just announced at COP, and they're concerned that software is between 4 and 5 percent of global carbon emissions, and that's twice that of air travel. But I think there's something more than just an individual concern. I think that concern is becoming more than something that people worry about, or campaign about and actually, it's starting to turn into something that's also becoming a business driver in a way that, a way of individuals distinguishing themselves in the job market, so to speak. So for companies like EPAM, which is a consultancy, they don't just want to develop their staff. They also see the difference it makes to the organization, um, when winning a customer.

Because their customers are demanding green solutions. Um, and it's also, of course, it's a massive motivator for their employees. You know, they've a fantastic internal community who want to learn these things. But yeah, I'm not surprised we've seen this growth. My, my colleague, Russ, who leads our knowledge efforts at the GSF, he says it's just the beginning, and I think we will really continue to see us moving from just being worried about the environment to actually having people that really want to do something, they really want to take action, learn more about it, and then really measure and reduce.

So, yeah, I'm looking forward to seeing six figures on number of people taking that course, hopefully at some point in the early months of 2024, and I think we'll continue to see both that individual and that organizational pressure to learn and then be able to take action.

Chris Skipper: Yeah, amazing. Yeah, it's gone up to four to five percent already, the emissions of software.

Adam Jackson: Yeah, yeah, That's right. That's actually a stat that we, that we announced during our State of Green Software report back in May 2023. Yeah, it's, I think it's quite startling to a lot of folks because we see, I think most people think it's in the sort of 2%, 2 - 3 percent band, but we actually did a lot of research and found that it was actually quite a bit higher. And I would certainly recommend people go to stateof.greensoftware.foundation to read the report for themselves. But just a reason that it's higher than people think is because a lot of people just think about the data center or the device, but actually you've got data center, you've got all of that network infrastructure in between, and you've got the device. And I can't remember the exact numbers, but actually I think you can even sometimes think of them, depends on the application, but sometimes it can even be a third each. So yeah, reducing, for example, reducing network bandwidth is a good way of decarbonizing software as well. And people never think of that. So check out that report if you want to hear a bit more. Yeah,

Chris Skipper: Yeah. Cause I think it was just probably this time last year that it was 2%, which is quite, yeah, I think we announced that on the podcast, so that's quite, yeah, really scary, but yeah, and also, yeah, sorry to focus on more positive things and the growth of the green software practitioner courses intake or the amount of people that have gone through it, I think. That number is really impressive, firstly, and then secondly, just that more people are going to be using it as a business decision, that's something you pointed out, I think is something that I think might be a great way of people to think about their approach to green software as a whole, because whether you like it or not, businesses are going to have to adapt.

And yeah, so really great that the GSF is leading the way in providing qualification for people, consultants in particular, that's a huge industry consultancy, so yeah, and obviously we have consultancies that are part of the GSF. I think Accenture, am I right in saying that?

Adam Jackson: We've got a few, and yes, yes, so, other consultancies do exist as well for folks, but yeah, if you want to see who, see the, the big list, our membership's grown massively this year, by the way, so just go onto our website if you want to find out who, who is contributing as a member, and that those members really help us set our standards and they're the ones that are very heavily involved in the day to day running of the foundation. And yeah, consultancy is a big part of that. We also have a lot of representation from the, the cloud and tech companies themselves, but also industry, a lot of, banking in there right now, a lot of regulated industries in particular.

Chris Skipper: Yeah. Yeah. Great. So obviously we just talked about the frightening statistics of the amount of emissions that are being created by the use of software. I think part of that is down to the, the rise of AI and the use of LLMs. ChatGPT, et cetera. And so with that in mind, Sophie, I'm going to ask you a question a bit about that.

The relationship between responsible AI and environmental sustainability was a hot topic at Decarb 2023. What were some enlightening points brought up by Tammy McClellan and Jesse Mccrosky in their fireside chat as part of the event?

Sophie Trinder: Yeah, thank you. It definitely was one of the hot topics of the Decarb event. And so many enlightening points were brought up by Tammy and Jesse, but one particular point that stood out, which actually we've touched on already, is that transparency of data and information is key. We touched on this when we were talking about measurement, weren't we?

We were saying the more we measure and we request measurement, the more likely people are going to be to giving us the data we need in order to measure. So Jesse was touching on that transparency of that data. So we need the full picture on emissions of AI systems. And that's not just looking at open source code, but full transparency of what the emissions of AI systems are.

And he was talking about how In order for software developers to make sustainable choices on whether to integrate AI into their products or how to develop the type of AI that they're developing, they need the full picture. Similarly, he touched on how policy makers need that in order to make good regulation and policy on AI.

But also for us as consumers, how we can make sustainable choices when using AI. I like the term, be a conscious consumer. But for us to be conscious consumers, we need the knowledge in order to be conscious. So we need that transparency of data. And Jesse painted this hypothetical picture. What if there was a carbon counter or there was a water gallons clock at the top of ChatGPT?

Would you use it differently? Would you only use it when it was essential? Maybe you would write a shorter prompt. Would you ask it to write its answers briefer? Similarly, he painted another picture where, what if a software developer's screen went red? If the developer made a decision that, yeah, might make them save some time, might make them slightly more efficient, but what if it came with a huge carbon cost?

So their screen went bright red when there was a carbon cost associated with the code that they were writing. I think it was just a really interesting way to start thinking about being more transparent with that data so that we can be more conscious with our decision-making.

Chris Skipper: Yeah, absolutely. And there's also a gap in the market there for perhaps a browser plugin or something that would really inform day-to-day users, because I know people have, well, I've spoken to people that have been using ChatGPT as if it were Google and that's quite worrying, you know, so, um, you know, we've, there's an episode of environment variables where we speak about the environmental impact of AI and, um, just the sheer amount of water that it uses is quite incredible.

So those ideas. That was spoken about during Decarb 2023 were really great. And I think maybe we'll see some of that as a springboard for what will happen in Carbon Hack 2024.

Sophie Trinder: Yeah, definitely. We've seen it in certain other industries when you're buying a flight and you see the little green label now on certain flights where they say if you buy this flight ticket it's going to be better for the planet in terms of emissions. Maybe it's It's going to be a fuller aeroplane, those sorts of decisions, it's, it would be great to see some of those, like you say, user interface platforms where we'll be able to start helping people make those more environmentally conscious decisions when it comes to software too.

Chris Skipper: Yeah. And I think that's one of the biggest challenges facing organizations right now, um, at Decarb 2023, there were a lot of challenges that were raised. Um, speaking of which, what's your take on the biggest challenge organizations face when embracing green software right now, Adam, and how do you think it can be overcome?

Adam Jackson: Yeah, so, well, first of all, we've talked a lot about measurement already, and I do think that measurement is probably the biggest challenge for us all right now, but I want to look at a different challenge that some, that organizations specifically have embracing green software. And I touched on this a little with the EPAM example, but in one of the other organizations we had on the Decarb event was Siemens, and Siemens, they're a brand new steering, steering member of the GSF, they joined in the last few months, but their journey with a sustainable software goes way further back than that. They've spent a long time building sustainability into their entire software lifecycle. And that's what their talk was about. And I think actually that, beyond the knowledge, is the hardest part. Often, I often tell folks when they ask me what green software actually is, that there's a lot of parallels with well-crafted software. So well-crafted software, like green software, focuses on built-in quality throughout the complete software lifecycle. So architecture, development, operations, getting all of the stakeholders together to take responsibility. And we need the same thing for green software. So at the GSF, we often talk about software practitioners, and this is a broad term that goes beyond developers alone.

It includes product managers, program managers, designers, UX, testers, IT operations. And what takes the time, and GSF does have some materials that can help here, is developing the best practice, the processes, and the learnings that bring all of these people together into a really holistic software lifecycle, and I think that's why we're really keen to get organizations to hear and listen to others, even if they're competitors, what's the best practice that we can all share that's gonna deliver green outcomes. And then we also have some materials inside the GSF, such as our Patterns Catalog, which is like a set of best practices and people can use that to, to go beyond just the learning and start thinking about how to integrate it into their organization. So yeah, a bit of a long answer. It's, it's a bit of an abstract challenge as well, and it's different for every organization, but really think, yeah, thinking about how, if we think of the Learn, Measure, Reduce, that reduce bit, it can be, sometimes it can be simple for a single product or a single workload, but when you're trying to think about how to get your whole organization on that journey, that takes a bit longer.

Chris Skipper: Yeah, amazing. And also, I think it highlights, there's been an immense amount of growth in the last year, just in terms of the amount of news and the amount of events that are happening around green software, which is fantastic. So Sophie, with that in mind, what are some of the exciting innovations or projects that are making a significant impact that were highlighted at Decarbonize Software 2023?

Sophie Trinder: Thank you. Yeah. So another project that was highlighted at Decarb that's been having a significant impact is another one of our open source projects I mentioned earlier called the Carbon Aware SDK. So we had two of our principal contributors come and talk at Decarb about the SDK. That was Dan Benitah and Vaughan Knight.

They gave a fascinating talk into some of the use cases of adopters of the SDK. So for a bit of context for anyone who is new to the Carbon Aware SDK, but also to Carbon Awareness, that basically means being aware of the carbon intensity of the electricity on the grid and doing more when and where the electricity is cleaner.

So less carbon intense, i. e. generated by renewable resources, and doing less when and where the electricity is dirtier. So the Carbon Aware SDK is a web API and command line interface tool that helps assist in building carbon-aware software by providing the intelligence of knowing when and where the electricity is cleanest.

So you don't need to do all of that work for yourself. There's a tool out there already that can help, help you work out when and where would be best or cleanest for your consumption. And what was really interesting at Decarb, Dan showed us how he's been adopting the SDK himself at home to successfully reduce his impact by 30%.

So, he used the Carbon Aware SDK to simulate different hypothetical scenarios. So, for example, he was looking at his big consumption uses, his dishwasher, tumble dryer, those sorts of things, washing machine. And he used the SDK to, to simulate Different scenarios of if he was to shift his consumption, so that's the kind of the when part of the when and where of carbon awareness, when you consume, he was, yeah, looking at different scenarios and he learned that if he time shifted by seven hours to time where his electricity was cleaner, then a 7 hour shift would be best.

So that's what he did in order to reduce his impact by 30%. So it just shows that with the right data, like we were talking before, with the transparency of that data, we've all got the potential to make some significant impacts in our reduction of consumption. And so whilst Dan shared this example of carbon awareness in the home, and the small things do add up, we've also had some big tech companies adopting the Carbon Aware SDK, which has been great, including Microsoft and NTT Data, for example, like you mentioned earlier, to shift some of those high workloads like training machine learning models to a different time in the day, potentially even a different location. That was the where part of Carbon Awareness, when and where, time and location. So yeah, it was really great to see the Carbon Aware SDK being talked about at Decarb.

It is still growing. We're just about to do a version 1. 2 release in a couple of weeks. But again, it's all open source, open to contribution. And definitely would recommend people taking a look and seeing if you could simulate different hypothetical scenarios for your own workloads. Be that at home or at work.

Chris Skipper: Awesome. Yeah. I haven't heard that story about Dan using it in his home. That's fantastic. I know we spoke a little bit about how people can make small changes in their own life to do with green, with green software and just being more environmentally friendly as a whole the last time Adam was on the podcast.

But with that in mind, Adam, can you tell us a little bit more about how people can use the SDK in their personal life to become more carbon aware?

Adam Jackson: Uh, yeah, actually, Sophie's already given a really good overview and, and Sophie's quite an expert on, on energy. So I've had some really interesting chats with her already, but, but yeah, the, the added dimension for me, if you, if you actually watch that final session from Decarb, you probably can't help but see how excited I was personally to hear from, from Dan and Vaughan talk about carbon awareness to hear in their own homes. So I have a home assistant set up at home. So it's a Raspberry Pi 4 with the Home Assistant software on and that's open source and that piece of kit talks to all of the, all the smart home, all the, all the devices in my house. And we're getting better and better. I feel like every time I open Home Assistant, I can connect to more data sources in my own home. And that makes me really excited as to how I can not just think about the energy, but also everything that's going on in my house. So, yeah, we touched on, upon how the SDK could be used, and Dan is actually writing a blog post right now about how to integrate Carbon Aware SDK into home assistant setup without doing lots of custom code. And I was going to put a bit of pressure on him because he hasn't finished his blog post yet. So hopefully by the time this podcast comes out, it'll be done or soon after and we'll try and get that linked. So, yeah, I'm excited about this because not only do I want to, I want to understand more about my own carbon impact, but I want to combine that with the data that all of the devices in my house provide. And I want to combine that with my, my battery and my solar setup and think about the dimensions, such as having an electric vehicle. So looking at how I can really balance and optimize my use of electricity. And so, sadly, I live in a, in quite a high-carbon area. So if you touched upon the, the location being a dimension, I can't change where my house is sadly, but the intensity does, the carbon intensity does vary dramatically at different times of the day. So I'm hoping I can take something from my work life and make that work for me at home as well. So you definitely watch that, that short video as well from the end of Decarb about, about Carbon Aware SDK. And yeah, you'll see some of the discussions I had were along these lines as well.

Chris Skipper: Yeah, that's incredible. I only learned about a Raspberry Pis through this podcast and yeah, just the use of it to do that is, is a great use of a Raspberry Pi firstly, but then secondly, also just the application from your, in your personal life like that. Yeah, it's great. It's amazing. I think more people should try and do that sort of thing.

I wish I had more smart home devices. I don't have any, I don't think so.

Adam Jackson: Don't get it. Don't fall down the rabbit hole, Chris. This is never ending. My, my toothbrush can talk to it. It's, it's crazy, but it's testament to what the open source community does when you let them get excited about something, you see something like Home Assistant. And this is why I'm excited about Impact Framework as well, because I'm hoping we'll see the same passion. Being able, being able to talk to maybe not my toothbrush, but certainly things in the cloud, network infrastructure, and the devices that are really creating that carbon impact in our life. And see, let's see what the, what that open source community will come up with over the next few months.

Chris Skipper: Yeah, exciting stuff. We're going to keep, at least on the podcast, we're going to keep track of the work that's being done with the Impact Framework. So that's it for our episode regarding the decarbonized software at 2023 event. Just before we go, I've got two quick questions for you guys. I'm sure Adam, you could probably answer these. If people want to find out more about the rundown of what really happened at decarbonized software, where should they go?

Adam Jackson: Yep. That's a, an easy one. It is decarb.greensoftware.foundation. And we'll make sure we put that link in the show notes.

Chris Skipper: Yeah, absolutely. Link will be down in the show notes. And the next thing is what's the next big event for the GSF?

Sophie Trinder: So up next for the GSF, we've got a hackathon in February 2024. So definitely watch this space and watch our newsletter. That hackathon is all going to be centred on Impact Framework. We're really hoping to see not just carbon emissions being recorded, but also all negative environmental impacts like water, waste.

We've said before, the opportunities are endless with measurement.

Chris Skipper: Amazing. And yeah, looking forward to that and there'll be a pretty big announcement about that hopefully sometime soon. We've come to the end of our podcast episode and all that's left for me to say is thank you so much, Adam and Sophie. That was really great. I personally have learned so much. And, um, if you do want to watch, this is another thing about, um, Decarb 2023.

If you do want to watch it, please head over to our YouTube channel, where you can check out the talks that we mentioned in this podcast. So yet again, thanks for your contribution guys, and we really appreciate you coming on to Environment Variables.

Adam Jackson: Thanks, Chris. Yeah, it was great to come on and talk about the Decarb event and hopefully you'll have me on again very soon.

Sophie Trinder: Yes, thanks Chris. Thanks for having me.

Chris Skipper: Thanks very much, Sophie. Yes. Great. So that's all for this episode of Environment Variables. All the resources for this episode are in the show description below, and you can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables. See you all in the next episode. Bye for now.

Adam Jackson: Bye.

Sophie Trinder: Bye!

The Week in Green Software: Greening the Front End

7 december 2023 | 49 min

Chris Adams is joined by Ines Akrap from Cognizant to talk all-things sustainable web design. Together, they delve into the nuances of designing energy-efficient websites and the challenges of green coding in frontend development. Ines shares valuable insights from her experiences at the Linux Energy Foundation Summit and the SDIA Green Coding Summit. The episode also explores common mistakes in optimizing sites for carbon efficiency and discusses exciting projects in the field of green software that are generating buzz. Resources like Website Carbon, Ecograder, and Lighthouse are highlighted, alongside discussions on the Software Carbon Intensity Specification and the CarbonAware SDK. This episode is a must-listen for anyone interested in the intersection of web development and sustainability, offering practical tips and exploring new research horizons in the quest to decarbonize the digital world.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Ines Akrap: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Topics:

Designing Energy Efficient Websites | Linux Energy Foundation Summit Paris ‘23 [4:08]
Green Coding in the Frontend: The Good, The Bad, The Ugly | SDIA Green Coding Summit ‘23 [16:15]
SDIA 2023 Green Coding Workshop Slides | Chris Adams Green Web Foundation [23:41]
Mistakes when optimising sites for carbon [29:53]
Which projects in this growing field have you excited right now? [41:53]

Resources:

Website Carbon [10:01]
Ecograder [10:01]
Lighthouse [10:11]
W3C Sustainable Web Design Group [12:19]
Decarbonize Commerce 2023 | Green Software Foundation [18:56]
Careful Trouble [19:57]
How Thoughtworks uses Cloud Carbon Footprint for sustainability | Cameron Casher ThoughtWorks [20:32]
Power profiling with the Firefox Profiler | Florian Quèze [21:34]
Sitespeed [26:20]
Software Carbon Intensity Specification | Green Software Foundation [26:46]
DIMPACT [27:06]
The Week in Green Software: New Research Horizons [27:10]
Netcentric/progux: Progressive User Experience Library [31:42]
COP28 [36:16]
Firefox Profiler [40:30]
Linux Perf | [40:56]
CO2.js [42:12]
The CarbonAware SDK | Green Software Foundation [43:31]
Showcase — lowwwcarbon [45:22]
Sustainablwebdeisgn.org [47:07]
Green Software Practitioner Course | Green Software Foundation [48:13]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Ines Akrap: You cannot fix something that you cannot measure because you just don't know that you have an issue. So being able to detect this issue and to show these numbers to the decision makers, I think this is something very important.

I'm your host, Chris Adams.

Hello, and welcome to another episode of This Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. In this episode, we're joined by Ines Akrap, a developer who I first met in Paris at the Linux Foundation Energy Summit earlier on this summer, and most recently with whom I shared a stage at the SDIA. The Sustainable Digital Infrastructure Alliance Green Coding Summit in Berlin earlier on this month in November. We ended up talking about digital sustainability for front end developers, and we had such a good time that we figured it might be worth continuing this conversation on the podcast for other people who are curious too.

Ines Akrap: Hi Chris. Thank you so much for having me here. I cannot stress enough how excited I am to be here since I'm listening to you so much that it just came out on my Spotify wrapped at the end of the year.

Chris Adams: Wow, I, thank you, I did not know about that. Okay, so, and for people who are new, or who weren't at either of these events and didn't see you speaking, I figured it might be worth just briefly giving you the floor to introduce yourself and talk about some of your background for this before we dive into some of the specifics about what we'll be talking about for the front end, so optimising for carbon at the front end.

So yeah, Ines, the floor is yours.

Ines Akrap: Sure. So hi everyone. My name is Ines. I'm a web performance engineer and sustainable web consultant and cognizant and centric. And yeah, I've been doing this for a few years now with many different clients and hope that I can share some of these things with you today.

Chris Adams: Cool, thank you for that, Ines. And Ines, now, I know that we met in Paris, which for me felt really quite exotic and exciting. And then we met recently in Berlin, but as I understand it, you're based in Munich, in South Germany. Maybe we could talk a little bit about that before we dive into the inner nerdery of what we're going to be covering.

Ines Akrap: Sure. So actually I'm originally from Croatia, from Zagreb. I lived there and finished my master's degree until I was 20, 23, so more or less. Then I went for actually exchange year to Portugal, so I lived a year in Portugal. I came back and thought, Oh yeah, no, I'm not staying here. I need something more, something different. And I actually ended up in Barcelona. So I lived in Barcelona for five years and then this move happened, which people still think I'm crazy about, but I moved from Barcelona to Munich where I am now. And. I'm still very fond of this move. I am very happy to be here because my mind is way more German than people think.

I really love how things function here, but I'm definitely super fond of my memories in Barcelona as well. So yeah, currently I'm living and working in Munich.

Chris Adams: Okay, cool. Thank you for that. All right, then. So if you're new to this podcast, folks, as I mentioned before, my name is Chris Adams, I work as the chair of the policy working group at the Green Software Foundation, and I'm the executive director at the Green Web Foundation. It's a Dutch nonprofit working towards an entirely fossil free internet by 2030.

And as a reminder for people who are curious about what we discuss. Whenever there is a project or a story or a paper or a project online, we try our best to link to it in the show notes below this episode. So if there's something you hear about and want to learn more about, please do get in touch or please check the show notes because we spend a bit of time to make sure they're useful for other people who come along to this.

Okay, Ines, are you sitting comfortably?

Ines Akrap: Absolutely.

Chris Adams: Well, then let's begin. Okay, so as I first mentioned, I first came across your work when you were presenting with, I believe, your colleague Claire Thornwell at the Linux Energy Foundation Summit in Paris, where you were talking about your experiences building calculators and websites.

Tools to help clients optimize for carbon. Uh, and that was mainly for an audience of energy nerds and techies who don't really work at the web, or the web, website, and were actually really captivated by this. I figured it might be worth just talking about how that even came about in the first place.

Because yours was the only presentation specifically about web. And I enjoyed it and appreciated it. But I'm sure there's a backstory, so yeah, take it away.

Ines Akrap: Sure. Okay. So how long of an origin story would you like? Yeah, actually. So I, I was working a lot years as a front end engineer and with my background, so my master degree is actually in telecommunication and information technology. So I actually do have a background in networks and all these like data transfer. So this was always a very interesting part of a front end for me. So bit by bit, I ended up in web performance also as a person who doesn't really like to wait and is a huge rage clicker. This was a logical way. You know, I want to optimize it for myself as much as for everyone else. And I think it was end of 2020 where I was looking at this, of course it was end of 2020, so it was an online conference that I was looking at from the conf comfort of my sofa about JavaScript and just about this sheer amount of unused JavaScript. And the presenter at that point showed calculated just this rough number of unused JavaScript and started comparing it, of course, to carbon emissions and to trees.

And this was actually-

Chris Adams: Just a quick one, this is JavaScript, when you say unused, like when someone visits a webpage, you're talking about a bunch of stuff that is sent as like dead code that isn't actually used by the browser, is that correct?

Ines Akrap: Exactly that. So there is, these days, there is so much JavaScript that is just not being, well, that's being processed, but it just doesn't serve for anything useful on the website. So he was talking about this because there is this tendency in web development where you just put everything on a website and with the big systems, and of course, people coming and going, a lot of people are afraid to remove these things, you know, so this is how this thing came about and it was a moment of the mind blow turnout for me and for my career, when I realized this is even a thing, no? And then I went a bit like to search into the topic to figure out what, what this sustainable web even is and realize that in my work as a web performance, I'm already covering quite some strategies and optimizations that are leading towards, towards cleaner and greener web. And at this point in time, we already had a small web performance group inside of the Cognizant and Centric and Claire was part of this group as well. So we always function really well together. We are quite a dangerous combination because I'm an engineer and she's a business design person. So especially when we go together, we have the same kind of goal. We can cover quite a broad spectrum of generally topics and like points of view, like storytelling and all that. So we'd started working towards this topic a lot together, of course, with our team as well, who was the web performance team, quite embracing this whole topic. And this is how this part slowly came about. I guess the extra just to add on top is that we were building a tool that was a real user monitoring tool. So we can go a bit inside of that, that later. But basically, this tool was originally supposed to measure, you know, core web vitals, web performance metrics, and all that. However, then we came across CO2.js and we realized that we are already collecting the data that we can just plug in and actually give users, well, give our clients in a way, almost real time assessment of the carbon of their website.

Chris Adams: Okay, if I may just check with you one thing, you use this term real user monitoring, and this is, I think it's referred to as RUM sometimes, and this is the idea that rather than using an entirely modelled approach when you might, say, have a virtual browser visit a web page, you're actually, it's much more tied to having a well, as it says, a real user tracking this, so you've got a much better idea of, okay, you're collecting information about, say, it's this particular browser that's doing this particular work, rather than just a generic browser in that sense.

And that ends up being slightly more accurate than an entirely modelled approach. Maybe we could talk a little bit about some of those things, because you mentioned being able to translate from energy to, say, carbon and stuff with CO2.js, and I'm very happy to refer to that a little bit later, but as I understand it, when you, these two approaches of using, say, a RUM based approach or a kind of synthetic based approach, this brings out this this ends up making people aware that there are assumptions that you might use in a typical kind of website carbon calculator tool that might not be, might not hold true when the real user is accessing a website, like whether they're signed in, or they've got cookies set, or stuff like this.

And this can also have an impact on visiting a site, for example. Maybe we could talk a little bit about some of that stuff there, because I realize that this is something that you end up having to deal with when you're building some kind of, uh, actual tools that provide the kind of more robust numbers for this.

Ines Akrap: Exactly. So this was, to be honest, a bit of a game changer, no? Because when you think about, you explain as well, RUM tool very well. So what happens there is just, of course, you add a very small script to the website, to every single page of a website. And this is the huge deal as well. So when you use any synthetic, uh, monitoring, which is no, things like Website Carbon, EcoGrader, or any of these tools, especially in web performance, we usually, if you're in front end development, it's usually going to be web page test, it's going to be Lighthouse, something from that you're emulating. And what's happening actually recently is that you're emulating such a bad scenario that web performance may be quite good. In the case of carbon, actually, what happens is that there is no interaction with the page. And actually, 90 percent of session is happening in these interactions, no? After this first load. So with carbon and at this point in time, kilobytes transfer being a proxy for carbon, what's happening is that there can be a huge difference in between just this first load and actual number that we get if the user is interacting with page, if they're scrolling. So if the page is implemented well, and please do that, implement lazy loading and strategies like this, if this is implemented. We don't know if users ever loaded these assets or not. And this is where the RUM tools comes very handy. So as you already mentioned, accepting cookies and then starting to scroll, starting to click, starting to open stuff. This is where the number, let's say, of the transferred, transfer size can lead up let's say from, I actually, once I examined the website, It was four megabytes on initial load, five megabytes after the cookies were transferred, and 24 megabytes if you would scroll all the way down because of the amount of images. So measuring carbon by just this first view, by just this four megabytes,

Chris Adams: be five times smaller, yeah.

Ines Akrap: inaccurate, inaccurate, number.

So even though this number is still not super accurate, we are all aware of this, it comes much closer to the real scenario than I would say the synthetic one does.

Chris Adams: Oh wow, thank you. Okay, so that's um, a useful thing to take into account for people who are starting out with this, because yes, there are tools which you can begin, but there's often, there's basically quite, there's quite a lot of depth in this rabbit hole if you were to start doing this on a kind of monthly basis or kind of really use this to manage, um, uh, the performance of a website over an extended period of time.

Okay, cool. I also mentioned, I also saw that you have had some involvement with the W3C Sustainable Web Design group as well. And I figured, and I think I've actually heard you talking about that on one other podcast or seen you speak about some of that before. Maybe we could talk a little bit about that as well. Maybe you could talk maybe about the motivations for that and just show, just show maybe how that's going or what some of this might be, where this might be going in the future. Because we did do an interview, I think a couple of weeks ago, with some of the people who have been very involved, but maybe It's interesting to hear your perspective as well, because I think you're speaking and you're working at a larger organization, and the two people we were speaking to were from a smaller organization there, so I was, it was, I'm, I'd be curious about your perspective and how that's been going.

Ines Akrap: Absolutely. So yeah, it actually, this whole thing came to be because one of my colleagues, he was at the web performance conference and he met Zoe from Google. So she told him about this group even existing. So the moment he came back to Munich, he was like, yes, you have to join this. So this was actually very cool.

This was how I found out about this group even. And I actually came right in time because we were still, it was actually about a year ago. It was right. It Now, I think I'm probably hitting one year mark now in W3C, Sustainable Web Community Group. And yeah, it's been very interesting because there is such a nice diversity of people talking about this topic from different perspectives. So we started first talking a lot of, there was, of course, a lot of. so much for joining me today and I look forward to having a lot of interesting discussions about this approach because of course, as you and everyone listening to this knows, it's a very complex topic, there is like so much to it, so many nuances, and I was so happy to actually be a part of this conversation, especially because I was already working with some of the organizations trying to push this topic, so thank you And yeah, a lot of times it happens. I've seen this in web performance as well. A lot of times happens that it's a, quite a different scenario when you have a control, complete control over the things you're trying to improve. And when you don't, and a lot of times in these big companies, for example, the client I'm currently working with, they don't really have control over whole, like all the code because they're inheriting some of the parts. So this is a very tricky part to be in that you have to navigate around. So yeah.

Chris Adams: So just two things there. So you mentioned, so Zoe, this is Zoe Maria Lopez, I think, at Google. Okay, yeah, she's been on the, I've seen her on the mailing list talking about this and asking some questions already. Okay, and if I understand it, there is one flip side, which is, yes, when you've got, say, a very small organization and freelancers, there is possibly a kind of, people might feel like, okay, yeah, this is very much in line with my values, but you might not have the scale.

And then at the large companies, there's this idea that, well, we have the scale, but we're also publicly traded companies, and they might have different incentives. So the idea of having different people in that group should hopefully mean that you don't end up going more towards something which is really lovely but doesn't achieve any kind of scale, or the other side which is totally scalable but not really, maybe slightly more in the interest of shareholders than the planet, or everyone else, or the entire group, basically. Okay.

Ines Akrap: I think it's a very, again, like same as web performance, like it's very tricky topic because it's such a fine balance. No, it's such a fine balance. And actually one thing that I was just discussing recently with the client is just this. Of course, installing this culture of realizing the cost of something because you see the benefits, you see the business value. However, I think a lot of times. Especially the business perspective does not see the cost when it comes down to performance of the website, UX of the website, and of course, the energy efficiency and carbon of the website.

Chris Adams: I guess it's where the costs are being shifted to, for example. You might have something which means you've got nice developer experience, but if everyone's having to download three or four megabytes of JavaScript before they can use a site, that ends up being pretty ropey experience, for example.

That's like an example of a cost in this scenario. Okay, cool. Thank you for that. And we will probably come back to the working group stuff there, because I think that's something that's of interest. But maybe we could talk a little bit about the kind of fireside chat we ended up doing in Berlin together at the SDIA Green Coding Summit because I had a lot of fun. And we ended up talking about quite a wide range of topics given the time available. And I figured before we dive into some of the specifics about common anti patterns or easy mistakes you might make that you could learn from, or good examples of, say, galleries and things like, or things to learn from other people doing this well, are there any kind of takeaways from the event that you had?

Because I really enjoyed it and it was the first time I've seen where you've got, this was largely funded by the German government actually. They were investing in this particular field in a way that I haven't seen so many other countries doing so far. So maybe I'll ask you, are there any kind of takeaways that you had that, that you'd like to share here? Because you were there for the two days with me and there was a lot to take in, wasn't there?

Ines Akrap: Yeah, I was just going this morning, I was going through my notes and I actually quite like that. A lot of talks just confirm some of the things that I also stand behind that are my opinions as well, which is I would probably emphasize the part of the three important things to to move forward that was know that the fireside chat about how to make people care the way forward is awareness responsibility and regulation, so those three I think are very the three pillars in general that we need know so awareness responsibility and regulations I think those three are definitely a way forward to have it in mind And then, of course, a bit about the fact that companies move very slowly. This is something I felt on my skin so, so many times. So it took actually many months to even put this measuring tool that we offer on the client website, just because of the privacy concerns and all that. And it's actually very, it's made in a way that collects only. So even collector script is open source.

So it's making a way that it's very transparent. It collects only, uh, performance data. However, it took quite some time to get all the approvals to even start measuring this data, of course, calculating carbon. So I guess that one, and what else? Oh, yeah. And a great point from Anita as well, that we need sufficiency.

We need to learn to need and to use less. I think that's like a general, even when it comes to development, even when it comes to general life, we just, I have to have a feeling that we're already sufficient. We have enough of, again, I'm a bit generalizing here. Sorry. A lot of times we kind of especially. In this kind of middle, upper middle class, we have a lot, so just to start living with less and start delivering less.

Chris Adams: There were two things you said that I think are interesting that really caught my attention as well because at various, when you see other conferences take place, and we had Decarb from the Green Software Foundation just run last week, or maybe at the time of recording, when this lands, it'll be a few weeks, so please check the website for more details to check about the dates on that.

The thing that really struck me was A, there was this idea that there were organizations basically saying, yes, we're expecting to be regulated, and we're seeing regulation coming here. And that's something that I've generally seen lots of other organizations been pushing back against. But there was an argument being made that, essentially, you need to have this, to provide a kind of level playing field, for everyone to take this first step because, in the short term agreeing to some of this stuff either will be expensive or will incur some costs that will have to happen. And that was one thing that you tend never to see inside technology companies. So seeing organizations say that for the first time was quite a thing. And also, this idea of the notion of, okay, there is a, I think there's a really lovely quote or there's actually, I've ordered, I've ordered a hoodie from an organization called Careful Trouble, and I think it's Rachel Coldicutt, she mentions this lovely quote, just enough internet.

The idea being that there's a certain amount that you do need, and you, and there, and that's not the same as having to just like, just mash the accelerator pedal and just build as much as you can, basically. This is a interesting, in my view, a contrast to how I've seen other people talk about sustainability where there's this idea that, oh, we're going to have exactly the same kind of lifestyle, and, but everything's going to be electric, for example, and we're just going to keep using more and more.

So this was something that really caught my eye, actually. I think maybe there's actually one thing we might talk about when we were there, actually, was there was a number of open projects that were, that we saw, I think we saw Cameron Casher from ThoughtWorks talking about Cloud Carbon Footprint. And, uh, one thing that he mentioned, and we'll share a link to that, which is an open project that people in the GSF may be familiar with.

That kind of caught my eye. I can't remember where, there were multiple tracks and we'll share a link to this. Were there any other kind of talks by particular people that really caught your eye that you'd want to mention? Because the cloud carbon footprint thing was interesting to me because when we had one of the financial institutions listed there, they basically came out and said about half of our emissions are associated with ICT, and we're using tools like this to track this because we are getting numbers from different providers, but we also have things on premise, and this is one of the only ways we have right now to end up with some kind of numbers that are comparable across all the different providers and organizations we do have.

Ines Akrap: Yeah, I would say that my favorite deep dive tech, tech nerd talk, uh, was the one from Florian from Florian Quèze,

Chris Adams: Ah, yeah.

Ines Akrap: pronouncing your surname wrong Florian, but yeah, it was meeting him and talking to him. So he is working on Mozilla Power Profiler. And that was for me, it was just so many insights.

I love that tool and

Chris Adams: That was really deep.

Ines Akrap: out. Yeah, that was super deep and, but it's, it was super amazing for me personally, it was like, again, I'm familiar with this tool, so learning what's behind and then also getting an opportunity to pick his brain, I think that was one of my highlights of the event.

Chris Adams: His story about reverse engineering every single operating system to come up with some numbers was like, pretty like, wow. I didn't realize there was, it was that involved to do this. We'll share a link to both his talk to the slides, and I'm not sure when the recording will be online, but we definitely can share a link to the slides straight away, because that was actually a real, that was a real faith for me as well.

And later on, we'll talk a bit about that, because well, we basically did a workshop as well, and maybe that's one thing we'll spend a bit of time touching on. So, I actually ended up leading a workshop that Fershad, one of my colleagues, Fershad Irani, had designed with me. Basically giving people a way into looking at tools around optimising for carbon on the front end.

So we spoke a little bit about coming to some of the forcing factors that you might have in terms of basically there's regulation driving some of this, like in Europe you have the CSRD, the Corporate Social Responsibility Directive, which ends up manifesting in the laws in all various countries. But even in California, there is the, there's a recent law passed with a similar alphabet soup acronym, which I'm afraid I can't remember off the top of my head as I speak, but I'll add it as soon as I can to the show notes.

But we also speak a little bit about introducing people to tools like CO2.js. And other ones, which basically track this. So we started with that. Then we showed website carbon. Then we actually did a bit of a deep dive into the profiler as well. Actually, maybe I'll ask you actually, like you, you were in that workshop with, although all the way through with me and, uh, I, I realized that you also, you helpfully shared some specifics about the differences between cookie websites and stuff like that for people.

Are there any things that you might share for other people because we've made this an open source workshop and we'll link to the all 100 slides that we did for this. But for people who were about this and coming into this front end developers, are there any things you might share for other people who are coming to this for the first time and trying to talk about, like, maybe the, their first steps of tracking this or what kind of the common things you might need to take into account, for example?

Ines Akrap: Like first and foremost, I love the workshop. So for me as well, like there was a, at the same time on the, so on the green coding summit, they were on the second day, there were different workshops at the same time. So that was the, the worst part because I wanted to see them all. I gave that as a feedback. I was like, please give me, give me time to see them all. Of course I chose as well, I chose to be at yours to see as well how you approach this. So hopefully I can as well serve as an advocate for something like this because this was something I was doing this whole year was trying to raise awareness of this topic and I think what you did as well, did an amazing job. So for me, I knew most of these things. However, I think it was very well done. It was very like comprehensive and it was very, I was going through these basics of how to actually understand it and calculate it. And I think this is the biggest issue that we generally have is that people just not being aware and just not knowing how to do it and then where to start.

Chris Adams: Yeah, I think, I, thank you, first of all. The thing I, one thing that we were trying to do was give people an idea that, yes, there are models and there are tools, but there's also a lot of uncertainty in a lot of this, and there's a really nice quote that I think was shared, that Fershad added in, but that Simon Wardley has used in multiple places and goes back to the 80s, which basically says, all models are wrong, but some models are useful, with the idea being that, yes, we spoke about data transfer as a proxy right now because we currently do not have that much access to the information that you might need to come up with some more to draw more rigorous conclusions when you're trying to track the power used on a website.

And this is partly why we started with high level models, then moving into some tools like Website Carbon that you might be familiar with, and then diving into the profiling tools, because, to my knowledge, they're one of the only ways that you can get actual resource consumption figures from end user devices.

And before that workshop, we hadn't actually realized that these are usable for mobile devices as well. So that was actually a really nice thing to take away. One thing we got was that we had a couple of people from Mozilla, people who work on the Firefox profiler themselves. They joined the workshop to basically help figure stuff out.

And I think actually one thing that happened is over the weekend, the manual profiling steps we were demonstrating during the workshop, by the following weekend, some pull requests had been opened to fix some of the tools to automate this, and now there's a piece of software called Sitespeed, which now can run these on an automated basis.

So, that was actually quite cool, but it means that me and Fershad will now have to redesign the workshop for this, actually. There was also a really nice, it was a good chance to actually talk about some of the other models that we do have that are in the community right now, because we spoke a little bit about the Software Carbon Intensity, which is used by the Green Software Foundation, but it was also useful to be able to compare that to some of the other models that you might see, like the Sustainable Web Design 1 and the 1 byte model, or even Dimpact, which is like the digital impact one, where we had Daniel Schien come on a few weeks back to talk about that in a bit more detail. We'll link to that in the show notes as well for other people. But while I'm here, I figured I could ask you, are there some, there are ways we can do it right, but there's also, there's also some kind of common mistakes that you might see people making when they, when they take their first steps in this field, and I figured it might be asking you while you're here, and when you've seen people do some of this, are there like mistakes you might want to be looking out for if you are first taking a step into this field that you would give people a kind of heads up on if they were maybe a special, they know how to build websites, they might know a little bit about web performance, and they're trying to figure out how to transfer some of this knowledge into this kind of slightly new or updated domain.

Ines Akrap: Yeah, sure. I think the most important thing to have about this topic is that this topic is very vertical. So it's not one role in the development process that is responsible, which usually comes towards the developer. It's, oh yeah, they didn't implement it well. Yes and no. No, like it depends. It could be implemented, potentially not in the most optimized way.

However, a lot of times what I've seen is that no matter how good the implementation is, the moment the content comes in, it's when the things start shifting. No? Because we're talking again, there is just this sheer amount of assets, a lot of times of images, of fonts. I have a feeling that this overconsumption that we were already talking is happening without thinking what is actually beneficial to the user because now working in, in, in, in field of, well, I'm just going to call it now sustainability because the web are sometimes used interchangeably, however, green. Of course, just goes towards just reducing the energy and carbon footprint and sustainability goes more towards people as well, designing for the people, for the planet. So includes stuff like accessibility and in the end of the day, good user experience. And I think this kind of benefits the business a lot as well, because if you design something in a way that's fast, that's responsive, that's efficient, it's probably going to be better user experience, which will potentially convert into better conversions rate into better, which it's something that business is probably trying to do. So I guess my first advice here would be involve everyone. Make everyone in the process aware and make it a joint goal and not just the fight between developers or designers trying to optimize something and then let's say business kind of stopping them because it doesn't make sense on the business. It does, however, everyone has to understand how it actually does.

Chris Adams: Okay, and there was also one thing we spoke about, um, during the fireside chat was that let's say you're going to be doing some things with, say, carbon aware, this idea that websites might change or respond or functionality might be different based on the amount of fossil fuels on the grid, because this might be one way to materially or measurably reduce the carbon emissions in aggregate across a number of users, for example.

Uh, we spoke a little bit about there's some pitfalls and pratfalls around that. Yes, it's a cool idea and you can start using this, but there's, it feels like it's quite important to provide maybe a way to override some of this in many ways. You might be setting a default, but you're not setting a concrete limit in some cases because you might not understand the full context of a user.

Maybe that's something that you, we could talk a little bit about because I've, this is something we've seen done right, and then done poorly in some places, or maybe it's not done poorly, done in ways which, where there's a lot of room for improvement perhaps,

Ines Akrap: Yeah, for sure, because so this is something that I definitely advocate for is this progressive enhancement, which also goes towards the part of respecting your user's decisions, like what your user wants, what their current status is, because of course there is such a differentiation between the user, like just put yourself in the place of a user, which is something I usually try to do, since I'm quite a user of internet myself. It's just when you see this point of just trying to get this information. No, I just need an address of the place I want to go to on my phone. That's let's say in a Metro and my battery is dying. And then it's definitely not the same if I'm planning this trip in advance and I can see this website, whatever it is on my M1 MacBook Pro with my wifi in Germany.

So these are definitely different conditions that users have, and I have a feeling that this is something that's not being accounted for at all in general and going towards this road where we decide to design in a way that progressively enhance, that kind of respect these usual decisions, because not only this, so I worked on a small open source project that's called ProgUX, so of course it's out there, it's on the NPM, we can also link it.

So. Its point is that it adds to your website, adds a few of these parameters. So it adds the user network conditions. It adds data saving mode, if it's on or off. It adds the prefer, reduce motion. It's very important because some of the animations can trigger medical conditions at people. And not respecting this

Chris Adams: Wow,

Ines Akrap: only be mindless, but can be dangerous for some people.

So people can access this so they can enable this on their devices that they don't want the motion to happen. And I'll, well, most of the website now may, there may be some, some edge cases, but most of the website don't respect this. So what we created is just, it's very just few lines of code that just add this to your, to your local storage as well to CSS, however, the useful part is the developer part, which offers you a bit of a simulation of changing between these modes to help you develop faster, switch into these modes. It's been a while that we changed anything because it's, again, it's quite simple. It's quite straightforward. So we didn't improve it a lot, but again, feel free as well.

It's open source. Feel free to build on top of that. So having something like this, where we take these choices into consideration, know the network, the phone, the memory, the current data saving. And of course, these reduced motion stuff could make a huge difference in user experience and of course, with that, no, if we respect data saving and reduce, no, amount of processing, we're also again, reducing amount of energy consumed and working towards something greater.

See, okay.

Chris Adams: I can see a scenario where you might use some tooling like this to essentially simulate, okay, here's what someone's going to experience in a kind of low carbon version or a high carbon version, because one thing that we've seen is that, uh, either people might look at, say, a low carbon version of a website as an afterthought, rather than being something you progressively start with and then progressively enhance, like you mentioned.

But there's also this idea of, like, the ability to override is actually extremely important, because we, as designers or developers, we don't or we don't always know the context something's being consumed in and uh leaving the leaving that at the end user is actually very important and there's a really nice example that i think from Apple was that's worth looking at so Apple rolled out carbon aware charging leeron this year for all their phones which sounds like a cool idea with the idea being that on average, if you were to use this, what happens is that Apple devices, if the energy, if the grid has a particularly high amount of fossil fuels being burned on it, Apple's devices won't try to charge up right now.

They'll wait for the grid to become a little bit cleaner. So, over time, the average carbon intensity for the power you're using in a phone becomes somewhat lower. The thing is, when this was first introduced, because people were by default opted into this, the idea was that if you maybe were to go home, you plug your phone in, or which like you said, is on a really low battery, you leave it overnight, and then there's a like high amount of carbon, high carbon hours when you're asleep.

When you wake up, your phone isn't charged in anymore, and then you've basically had a really, really terrible experience of anything related to grid aware or carbon aware usage of tools. This is why this override seems like quite important. And being able to, like, make this more accessible to users, so they can see what it might be, think through some of the edge cases, actually sounds very helpful, actually. I didn't know that existed. Thank you for sharing that.

Ines Akrap: Yeah, yeah, I completely agree with you. And I think again, it's super important to stress without naming names as well, is that this progressive enhancement is more important than graceful degradation, where we do want to serve the lightest version first and then give an option to enhance it. You can, let's say.

Not serve images and then give users options. Okay, you have data, you have whatever you want. Do you want these images? Yes or no? Yes. Okay. No worries. We'll deliver them to you and not other way around because once the images are downloaded, the fact that you hide them, you didn't do much. No, in a way of this kind of processing power and transferring and all that, because they're already there. So they might as well be there.

Chris Adams: Okay, I work in a non profit and so there's, so I can say this, the, we're just about, I think we've just started COP28 and COP28 is a really good example of this right now. COP28 is the, for the first time we've ever seen, is a very high profile website that has a kind of green mode for this. But unfortunately, the way it was initially implemented was basically to serve the full fat version of a website so that any device that has actually used it has basically done all the hard work and used all the energy for the first time.

And then once you've done it, you've got the option to switch off the images, which does feel like it defeats somewhat some of the point of some of this. And I feel like this feels like a kind of concrete example of the opposite of this progressive enhancement. The idea would be that rather than do that, you would have a nicer experience where you start, and then you would gradually layer on the kind of more consumptive and resource intensive elements.

Is that what you're referring to here?

Ines Akrap: Exactly. That's the exact one, because let's go again in the, for example, the off motion, there's so many, for example, animations on the page that can trigger some people conditions, like you already triggered a condition. So the option that you can turn it off, it doesn't really make sense. The point would be like, okay, you don't serve them at first, but then if someone wants this kind of enhanced performance, no, and again, there is definitely a cool ways to automate that.

No, it's one of my favorite example is that like for shots Carbon Aware Edge Delivery, which is also a very cool thing. Link in the notes. So yeah, which is basically on the edge, it decides how clean or dirty your energy is. And then depending on that, it delivers something different. So you are already taking into consideration user's journey, user's conditions.

And of course, you will get a notification that says you're seeing the downgraded version. Would you like to be, so it leaves this decision to a user, which I think it's a very crucial part in this designing for people part. It's give users a choice. Yeah.

Chris Adams: We'll link to that, that thing in particular, the one that Fershad Irani has worked on, because this was something he did in his spare time, which I think is quite cool. It's essentially some of this carbon aware stuff, but rather than happening at the, um, client side inside the browser, it happens at the server side, so it decides to make a decision to possibly change some of the functionality at that point, rather than having to do something at the client side, where All, in some cases, you might have already downloaded all the information, so it's a little bit late at that point.

Okay, I'm really glad you mentioned that, because I was playing with that last night, and I thought, wow, that's cool. You also said something interesting about motion, and this actually made, it reminds me of one of the things that Florian Quèze, Florian, I'm so sorry if I've mispronounced your surname, Florian's talk, when he was talking about the profiler, yeah, about, he was showing that there is a, he was talking about how basically, depending on how you see these CSS gradients sometimes, like you might have a page which gradually changes colour over time.

He was saying, how basically, if that's done incorrectly, you can essentially cook a laptop, because you're just changing the colour of so many pixels and, oh, so many times, uh, in a given, uh, uh, If you're changing something maybe 120 times a second, and it's a 4K display, and you don't have any way of knowing what you had changed before, something just as innocent seeming as like a CSS transition can end up generating like a lot of power usage.

And like, that was cool, but the thing that was really exciting I saw was the fact that he was able to share a profile of this. So you could visit the page, and we'll share the link to the page so you can cook your own laptop or trigger your own fans if you want to, but he also shared a profile showing exactly where the energy, where this was happening.

So if you don't want to cook your own laptop, you can basically see a jump in power usage as reported using his profile run. That was so, so cool. I hadn't seen, I didn't know that was actually possible.

Ines Akrap: I love that, especially also because it shows as well a bit of this limitations at the moment with the models now, because what we were saying that also you still don't know this processing energy of the laptop and like basically one kilobyte. I don't think the CSS code was big, so I think it was around, no, like the CSS transition could have been like a kilobyte or two and it cooks your laptop.

So this is again, I think this is a huge topic to move forward with now, like to actually figure out how do we see that? But it was quite a cool thing to see with all of also with all of these, his gadgets, he has such a cool gadgets.

Chris Adams: Yeah, we didn't, I, the thing that really blew my mind was him explaining how, now that he's built these profiling tools, he's using it for all kinds of unexpected uses, so the profilers, so we'll show a link to the Mozilla Profiler for this because it's an open tool that you can use on any website, but the thing that was actually interesting is that it It doesn't just work for websites.

So he was showing how he can use it for basically Linux. Linux has a performance tool called Perf. So you can use it to track any kind of piece of software. And he was talking about how, yeah, you can use that. But after this talk, he said, well, Chris, I've also used this to profile my own house. And, uh, it turns out that if, because he knows where the data is from, say, a fridge, or solar panels on his house, or stuff like that, he's been able to just have a track, so like you see a track for, maybe, a web renderer, or something like that, he's got fridge, solar panel, front room, or something, and he's been using this to figure out, okay, well, where are the power hogs in my house, for example?

And the idea that you've actually been able to use, essentially, industrial grade profiling tools to figure out how to optimize your house was I think that was the nerdiest point of the entire conference, but that was pretty mind blowing when I saw that. There's a blog post about it that has to come out, and I think he might be presenting about FOSDM about some of this, because I didn't know that was possible, but it was very fun to see that.

Ines Akrap: Yeah, absolutely.

Chris Adams: Okay, we're coming up to time, and we've been, we've mentioned in a few, we've been nerding out over a couple of projects that we've that's been, that we're both excited about, but I wanted to give a bit of space. Which projects, open source projects, are in this kind of growing field have you excited right now that you would like to give a bit of time to talk about before we wrap up for this?

Ines Akrap: Yeah, I don't know. It's a on the spot. Well, I guess like my still all time favorite and again, like huge kudos to you is co2.js. No, I'm, I'm actually super excited as well about the prospect of having some more time towards the end of the year and to help out possibly with some. And what I've seen now in the pipeline of doing this no actually making it even better and better in a way of providing better carbon intensity data, so I think that's a huge thing no, specially with someone who actually uses this in real life . Also having this huge conversation with the client yesterday, about this measurement reporting and like let's say correct air quotes, correct these number from the ROM tool that we built are actually are having this underlying technology, which in our case is CO2.js getting more and more precise and giving us more and more power and flexibility. This is something I'm super excited about, because again, you cannot fix. Something what you just said with Florian, like you cannot fix something that you cannot measure because you just don't know that you have an issue. So being able to detect this issue and to show these numbers to the decision makers, I think this is something very important. And then improving these tools, especially again, I'm working the front end. So this is for me, big deal. I'm also quite, quite sure that the Green Software Foundation is, is doing quite an amazing job.

Chris Adams: The CarbonAware SDK, which is, which is on the server side primarily. Uh, so that's, uh, there's a somewhat slightly different area that they tend to do. But, and there's a couple of other projects. There's one called the Impact Engine Framework, which was actually introduced at Decarb conference, which does actually, which is extendable in this way.

So it can take data from a number of different providers or models or things like that. But yeah, thank you. That's very nice of you to actually say. This ProgUX thing, I I didn't realize there was actually a tool to make it that easy to simulate these different scenarios in such a fast way. Uh, we definitely should share a link for that one, actually.

Ines Akrap: It's not a simulation of actually how it's, how this reflects, but it's just a simulation of an implementation. So basically the small simulator just changes these CSS classes so you can see, of course, how your implementation changes in different scenarios.

Chris Adams: I see. Okay. Thanks, Ines. So, we spoke a little bit about ProgUX, and you mentioned CO2.js, and we spoke a little bit about the CarbonAware SDK, which is looking to graduate, to be a kind of graduated project inside the GSF, and we also mentioned the Impact Engine. I wanted to ask, if you're coming to this field for the first time, one thing that, maybe I'm showing my age when I say this now, is that when I was first messing around with CSS and realizing I didn't need to use tables for everything, was a website called the CSS Zen Garden that let me show and learn from other people.

And there was tools like CodePen and stuff where I could view source and understand this. I haven't been able to find something for sustainable web design, or kind of low carbon design, or anything like that. And I figured I might ask you, have you seen anything that you might direct people's attention to so they could learn from copying, or at least, or try and figure out where they start, because like we all know, this is open source. We learn from imitation before we create something ourselves for the first time, for example.

Ines Akrap: Absolutely. I'm actually very glad you asked because I just found out recently about this site called A lowwwcarbon. Link in the show notes. And yeah, it showcases all of these low carbon web designs and developments. So it's very cool. I haven't had time to see how I get featured there because I tried to do that with my website, but yeah.

Chris Adams: Okay, so this is low carbon, so it's low with three w's, like the www, so low www carbon dot com. That seems what it is. Okay, wow, this is cool. This is actually really nice, because it's These are real world websites, and some of these are run by actual large companies, as well as just personal pages. So this might be a thing for people who are curious, and want to look at this and see for the first time, see how it's being done by other people, if you're looking for inspiration.

Okay, well, thank you for that, Ines! All right, Ines, we're coming up to time now, and I just want to check with you. For people who are interested in some of the things we've discussed, we'll be talking, we'll be sharing these things in the show notes, but if people want to follow your work, where would you direct people's attention to if they wanted to do this?

Ines Akrap: Sure. So if you're interesting, of course, I'm always happy to answer any questions about web performance, about sustainable web. So I guess, of course, LinkedIn, Twitter, because I just still refuse to call it X. And as well, so also you have my, my, uh, my personal website. So it's InesAkrap.com. However, there is a shortcut to it, which is Ines.codes. So it will also redirect you to there.

And.

Chris Adams: Ines.codes?

Ines Akrap: Exactly. It will give you the bigger URL, but yeah, I have that domain for a while now. So yeah, but also I would definitely, if you want to learn more about this topic in general, I would direct you to W3C Sustainability Guidelines. So again, I think this is a great place to start.

So we put quite some, like a lot of effort in this. Also sustainablewebdesign.org Implemented recently these guidelines. So in the last podcast I was giving, I was, I said, bear with us. It's going to be implemented. So a kudos to sustainablewebdesign.org team. They implemented these guidelines. So it's a bit more user friendly because as you may imagine, W3C documents are not really a beach read.

So it takes some time to go through everything. So it's actually 93 guidelines. So there is quite something, but it's also, it's segmented by roles. So every single role can find their own starting point and to see they're also quite, they don't have a way, like it's not one, two, three, four, five, it's just. Pick one that you like, I guess our personal goal on this would be like, choose the one that says high impact, low effort. This is a pro tip, but yeah, basically I think this is the first place that you can of course learn something. There is also a huge pool of resources added to the guidelines that we used in shaping them.

So I think this is a good place to start familiarizing yourself with the topic.

Chris Adams: Brilliant, thank you for that Ines. I will also mention that for people who are coming to this, the Green Software Foundation actually has a free course available which will get you a kind of nice certified badge with a series of videos to provide a bit of helpful context for this. That's something that's of use, but Ines.codes, I N E S dot C O D E S. Okay, excellent. Thank you so much for this. I really enjoyed this, Ines, and um, thank you again for taking the time to, well, share a stage with me and share a podcast with me. Thank you.

Ines Akrap: Thank you so much for having me.

Chris Adams: All right, have a lovely weekend, Ines. Bye!

Ines Akrap: You too. Ciao!

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware. foundation. That's greensoftware.foundation in any browser.

Thanks again, and see you in the next episode!

Introducing the Impact Framework

30 november 2023 | 38 min

In this episode of Environment Variables, we dive into a key highlight from Decarbonize Software 2023, featuring Asim Hussain's talk on the Green Software Foundation's newly introduced Impact Framework. This innovative framework aims to model, measure, simulate, and monitor the environmental impacts of software. Asim begins with an insightful overview of the framework's capabilities and objectives. Later, project leads Srinivasan Rakhunathan and Navveen Balani join the discussion to delve deeper into its applications and potential. This episode offers an essential understanding of how the Impact Framework is set to revolutionize the way we assess and mitigate the ecological footprint of software development and use.

Learn more about our people:

Asim Hussain: LinkedIn / Twitter
Navveen Balani: LinkedIn
Srini Rakhunathan: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

The Impact Framework:

https://if.greensoftware.foundation/

Events:

Keep an eye out for the announcement about CARBON HACK ‘23 - coming soon!

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Asim Hussain: Send us a manifest file so we can see not only how much is this thing going to cost us and how are you going to architect it, but what are the carbon emissions of your proposed approach? And let me compare versus the other people proposing approaches and not just compare on cost, but compare on carbon as well.

And I thought that was really, every time we talk to people about Impact Framework, people always come up with Another use case.

Hello and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode we discuss the latest news and events surrounding green software. On our show you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host Asim Hussain.

Chris Skipper: Hello and welcome to Environment Variables. Today we've got a highlight from the recent decarbonized software 2023 event. We'll be showcasing the talk from Asim Hussain about the recent Impact Framework, which is a framework to model, measure, simulate, and monitor the environmental impacts of software.

That has been introduced by the Green Software Foundation. To begin, Asim will do a brief overview of the framework and later on he will be joined by the project leads Srinivasan Rakhunathan and Navveen Balani. He is introduced by our fabulous director of communications and member relations, Namrata Narayan. So it'll be her voice that you hear first.

So without further ado, here's the introduction to the Impact Framework.

Namrata Narayan: It's the moment we've all been waiting for. Well, one of many. I'm delighted to introduce our chairperson and exec director, Asim Hassan, back to the stage. The GSF believes the most important step in greening software is measurement. An Impact Framework will enable us to measure software across environments and software types. But don't take my word for it. Asim, over to you for a full, brief, and exclusive look.

Asim Hussain: Thank you so much, Namrata. And I was really excited today to see several talks now focused on measurement from Siemens, we just heard from, and from Code University. We're all aligned, and we're all converging to the same point. 2024, I believe, will be the year of measurement. That doesn't surprise me, because when we first launched the foundation, We asked all of our members what their pain points were, and the overwhelming answer was measuring.

And we've been trying to explore all the different avenues of measuring over the last three years of the foundation's existence, and we've made lots of progress. And I mentioned earlier on, we, the stuff, the standards working group launched the software carbon intensity specification, which has reached ISO.

We're very excited about that. It should be in the catalog shortly. We've incubated, incubated various data projects because we knew data for measurement was a problem. And we've drove the creation of, of a number of case studies, software measurement case studies. But now we're in the next evolution to have software measurement be and mainstream activity.

For this to be an industry with thousands of professionals working to decarbonize software, for businesses to grow and thrive in a commercial software measurement ecosystem, we need to formalize software measurement into a discipline with standards and tooling. And we believe Impact Framework is the tooling and SCI is one of the standards.

Today I'm going to explain a little bit about Impact Framework now, then I'll give a demo. I spoke to some of our other co leads on Impact Framework about potential future use cases for the tool last week when I still had some hair. And we'll play that afterwards. So Impact Framework aims to make the environmental impacts of software easier to calculate and share.

Impact Framework allows you to calculate your application's environmental impacts, such as carbon, without writing any code. All you have to do is write a simple manifest file. Sometimes we call it an IMPL. And Impact Framework handles the rest. I want to be very clear. It's called Impact Framework, not Carbon Framework.

Some of the things I'm going to show you today is about measuring carbon, but our goal for the framework is to go beyond carbon, to start measuring all sorts of different environmental impacts. There are three concepts in Impact Framework. There's the manifest file, which contains either raw data or information about where to get the data from to do your measurements.

Then we have models. How do you get from that raw data to, let's say, carbon or water or some other impact? To do that, we need to use models, and we can chain models together to calculate your impacts. Just in a pipeline, just like a Unix pipeline. Then we have Impact Engine itself, which is the CLI tool, the command line interface.

You pass the manifest file to it, and it computes your impacts of your software application. I would say it's saying transparency builds trust. You don't just calculate and share the number, the final number that the Impact Framework gives you. You share the manifest file. You share your workings out. You share.

You share all of your data, all of your assumptions, all of your coefficients so other people can verify it, can challenge it, can compute it themselves, can fork it, can adjust it, can change some of your values and run it again. It's all about being transparent. Only when you share your working out can anyone trust your calculation.

The project is entirely open source and composability is a core design principle. We want you. To be able to create your own models and plug them into the framework or pick from a broad universe of open source models created by others. We've got some to show you right now, which I can do in the demo, but our goal is for there to be thousands of models all working together.

What I'm going to show you today is a result of almost a year's worth of effort. We first started a talk at this project was proposed in December last year as an incubation project, and as it evolved into something incredible. And today we're ready to release the alpha version. Alpha. Means use with extreme caution.

There are bugs. I was trying it for the demo today. I found some more bugs. This is a buggy product. We are looking for early adopters who are comfortable with buggy software, non backward compatible changes. It's moving very fast, but if you are, if this is something that you're interested in, please try it out and please give us some feedback.

But you understand you're on the bleeding edge here. The sense of excitement from everybody involved is something I've rarely seen in this space. It's a bit nerdy. I'm about to show you a bunch of YAML files and get really excited about it. But trust me, these YAML files, I think are going to change the world.

One of the things I say about... My long term goals, and I hope the Foundation's long term goals, to do with sustainability is that we want to do to sustainability what open source did to software. I'm not talking about small, iterative changes. I want to change the game, and I think the Impact Framework will do just that.

You can find more information about Impact Framework at if. greensoftware. foundation. That's if.greensoftware. foundation. IF will also be the main focus of our yearly hackathon, which will be coming in February 24. We'll have all sorts of prizes this time around, including an under 18s prize and prizes for non technical contributions, so watch this space.

So now let's start the demo. So I've got an empty folder here. Well, it's not empty, but first you would create an empty folder where we can do our work. Impact Framework is built in TypeScript, so you can either use NPM or Yarn to initialize a space to work in. So I'm not going to run these commands in the space of interest, but you would probably run Yarn in it to initialize the space.

And then we want to start adding the CLI and the framework itself. So we do Yarn, Add. Run soft slash Impact Framework would bring in the framework itself. And we're also going to be using some models, some of those models I spoke about. So we've built some of our own models and we would install them in using IF models.

All of this, by the way, this is a tutorial at if. github. com. greensoftware. foundation. So you don't have to be furiously writing this out. And we've also got some other models in an unofficial models package as well. So after you've run all of that, what you need to do is you need to create a manifest file.

I mentioned that manifest file earlier on. Now I've already got a manifest file written here, so I'm just going to open it up. Here we go. So, this manifest file is a very simple one. It's going to compute the carbon emissions of a server we have running on Azure. Manifest files can be much more complicated than this.

Compute different software components, compute network, front end servers, user stories, campaigns, whatever you want. But for this demo purpose, we're keeping it very simple. Just one server. Let's walk through the YAML file. I'm just going to show you, let me just make some space here for a second. I'm just going to go through a couple of the aspects, the areas in the YAML file so you can understand it, the manifest file.

I'm going to first go through the initialize. So this is where we load up and globally define all those models that you want to use in your application. I'm going to go through some examples of this. You're going to see what I'm talking about. You would define where you're loading this from because you can, you could create a model.

Which you install. This is not, we expect a universe where there are thousands of models and you're just cherry picking which models you need to use to calculate your emissions. Some are loaded from unofficial models. We're not Microsoft, so it's unofficial for us to maintain the Azure model. But some of these are from our official models.

If it's coming from IF models, we're going to maintain this. We're committing to maintaining this. So this is the path of the module where it's coming from. This is the name of the class you're loading. And this is just how we're going to refer. to this model in the rest of the YAML file. Let me scroll down to another area, which is called a graph.

So this is where we describe all the software components we want to measure. In our case, we just have one server. Uh, cause again, we're keeping it very simple, but you can have dozens, hundreds of software components in it. And it, but it's up to you how you want to group them. Do you want to group all your servers by region?

Do you want to have a front end, a back end networking? Database, however you want to split it up and group it is up to you. We leave that very much up to you. It's just different components. Um, let me dig into my server then. So, this little, we call it a component node. This component node contains all the information required to measure this.

Server. I'm going to start from bottom to top. So I'm first going to start at this node called inputs. These are the inputs. These are the things you've gathered and measured about the server. What we call a key concept in Impact Framework is something called an observation. This is one observation. An observation is just something you've measured about your running system at a given time and for a given duration.

And these are the parameters of your observation. You can actually have multiples of them. We would expect you to have, maybe you grab observations every five minutes throughout the day. So we'd have a large array of observations. I've just got one here. The next thing to really explain is this thing at the pipeline at the top.

I mentioned before that we were all about models and a pipeline of models. The data here is passed in to the first model at the top, and it's just a piece of code. It's just a piece of code. And you pass in that observation and whatever outputs it passes it to this model. And whatever that outputs, it passes it to this model, to this.

And the game we're playing, the game we're playing is we've got this data here, this observation, some data about a virtual machine, and we want to convert this in this use case into carbon. And in order to do that, we chain these models together. They all work and play nicely together to convert all of that data into carbon.

Now to explain this all, the best thing to do is for me just to comment out. Some of the other models, I'm going to go through it one at a time, and I'm just going to run. The command, which is impact engine, you pass in the input file and you pass in an output file and you just let it run. The final thing I'll just measure, talk about while, while that's running is this config section here.

There's a bunch of different places you can configure models. This is one of the places you can configure models. And here. We just have some configuration values, which we're passing into some of those models. And this configuration here is specific to this component. So different components can have different configurations.

There's also a global configuration as well. So you can see here that file has completed, that execution has completed. I'm going to open up the results on this slide here. And excitingly, what we've got at the bottom is it's the same file, but as well as inputs, we now have outputs. And what the Azure importer model does is we've passed it some data.

To, to query and it's going into the Azure Monitor system and it's extracting out information about CP utilization, the memory available, it's 'cause this has all been captured on Azure and we're extracting it as a series of time stamps. What I've said here is, for a one hour period of time, give me 12 five minute buckets of the average CP utilization and, and other data.

And it captures other data as well. And a very important piece of data is what is the instance type of this Azure, the Azure machine that I'm running on. What is the instance type? We know it's Azure because the Azure instance, the Azure importer gave us that data. We know the instance type, but in order to calculate emissions, we need to know what processor is this instance type.

What processor is it? And for this, what we've built is another model called Cloud Instance Metadata. I am going to just run this whilst I talk. So cloud instance metadata is a module we're actually maintaining. There's actually very challenging, there's no real globally maintained database of telling you this instance type is this processor.

Every cloud provider is different, every cloud provider has lots of different rules. So we tried lots of different solutions, but what we ended up with is we're actually maintaining this model ourselves. We're maintaining a database of, we've only got Azure for now, but for Azure. This instance type is this processor and other information about that, and we will, we commit to keep maintaining this for the long term, and hopefully it's all open source, so if you see something missing, please write a pull request and add it yourself.

So you can see now it's complete. I'm going to go into the results. And in the outputs, again, you can see a couple more bits of information have been added. For instance, the physical underlying processor and another very important value, which is the thermal design power. So you can see here, very interestingly, that one instance type, if you actually look at the Azure documentation, it doesn't tell you exactly which underlying processor you're going to use.

Because when you use the instance type, you may get one of these three, one of these underlying processors. So it's quite challenging then to figure out what to do. So we have certain rules and logic which we've Worked with the community to describe and we effectively picked what's called the most energy hungry processor of that list because it's always better to overestimate energy than underestimate energy.

But now we have physical processor, we have thermal design, we have a bit more information from that cloud instance metadata. We still need, we need to convert this into energy to get to carbon. We need to know energy. What are some models that we can use to convert this data into energy? Now a really, uh, popular model that people use all the time is tes.

It was generated by a gentleman called Benjamin Davies, who worked at tes. It was a couple of years ago. It was published, and he figured out a, a, a generalized power curve by analyzing hundreds of servers, different types of servers, and it's quite well used. It's using a lot of our case studies. It's used in a lot of places is, is this tes curve and what Tes Curve needs in order to estimate energy is it needs a C utilization and it needs to know the thermal.

Design power of the processor. Now we have that data, we can run the script and it will effectively do a lookup of both of those bits of data to generate some energy. And, uh, EagleEye, if you would probably see that I also, uh, uncommented sci e. Sci e is just a little helper function that we have to, to copy some data into other, into an energy field, which is useful for the following models.

And you can see now we've calculated energy. So we've basically now, we've got to the point of... We know how much energy for this five minute period we are consuming. And this is in kilowatt hours, 0. 00284 kilowatt hours. This is one of the bugs. I don't, I want to make sure there's one of the bugs we discovered today.

That's slightly off. We will be fixing it tonight, but this is the, now we have energy. Now we have energy. How do we get from energy to carbon? Now, what you need to get from energy to carbon, operational carbon, I should say, is you need to know how clean or dirty was the energy. That is something called grid carbon intensity, and you would need to know every grid has a different measure of grid carbon intensity at different times, and we have several models that we can use in order to calculate grid carbon intensity.

There's a supplier, one, uh, Two electricity maps on Watttime are two of the biggest suppliers of this data. They're both members of the Green Software Foundation. We could be using the Watttime model because we've created one, but just to demonstrate another piece of functionality, what we've done in this case is we've actually just hard coded it.

There are some cases where you want to actually know what the grid carbon intensity every five minutes. There are other cases where you just want to use an annual average. And here's something that's very important. We've put in here 951 grams per kilowatt hour. This isn't hidden away in Appendix 96C of PDF 43.

It's in the YAML file. It's transparent. You don't, when you're being shared this YAML file, if you don't like this, you can go, Well, you know what? I actually want to use the Watttime model to actually get the real time energy or real time carbon intensity. Or actually, I don't agree with your value. I think it's 951.

This is what's so important. Transparency builds trust. We want to share everything, all of your workings out, all of your coefficients, all of your data. That's what you share, not the final number. So now we've run CIO. If we go back to the results, you can see at the bottom, we now have operational carbon.

So we finally got some carbon value. So for this five minute period, it's 2. 7 grams of carbon. And let's scroll down, that's 2. 708. And I think there's, hopefully there's one that's a bit lower to prove that it's working. Hmm. Well, it is working. It just says 2. 7. That's because it's hard to find the peaks right now.

But it's there. 2. 7 grams of carbon per five minute period. But we also want to include embodied carbon. We want to include embodied carbon. We want to include other types of carbon emissions. So, other... Components of the SCI calculation. So obviously being a GSF, we focused on SCI. So now we've also got SCIM, which is, it gives you the embodied carbon.

It's very basic right now. You have to provide a lot of data manually, but we're looking in the future, some of the things we're looking to expand is to, is to make this a bit more automated. So again, it tries to figure out what the embodied carbon is from the instance type and just calculates it more manually.

And SCI then actually calculates a SCI, a SCI score. Thank you. So let me run that again, and then now I have a slight, uh, awkward pause, I'll wait for that to run. And again, it's just that this, the really important, I was trying to, I cannot reiterate it strongly enough, you do not share the final number, you share the manifest file.

Manifest files are executable, there's no need to share the final number, you just share the manifest file and the next person just runs it, and they, and it gets the same data. You can actually hard code the actual data, you don't need to put the Azure, So you can actually put the results of Azure as inputs, and then that would actually be a static file that you could share.

So now that it came through and we can see, here we go, we've got the size score, we've got the embodied carbon per 5 minute period, and we have a size score of 32 grams per hour for that 5 minute period, and a 32. 4 per hour for that 5 minute period, and a 34. 2 per hour for that 5 minute period. So now we've got it.

We converted some raw imports into carbon emissions. Okay. So that is the very final thing I will just mention really quickly is, let me talk about another model that's been built by Intel called IEE, Intel Energy Estimator. Intel Energy Estimator is a drop in replacement for other energy models like TEEDS.

It just has much more accurate energy estimates for Intel policy. It actually has a... Data for each and every Intel chip type. Um, but the, and this is how you would configure it. And the power of IEE is that I can just go into here. And just drop in IEE. Each of these models has standardized interfaces and standardized parameters.

Standards is what creates the ecosystem. By having standardized models, we can plug and play, we can cherry pick, we can tune it and get exactly your calculation pipeline that works for you. So let me run that. Really quickly, and as it's running, I'll just talk a little bit more about how it actually works, because you can see here, it's a little bit different.

I mentioned that Impact Framework is TypeScript, and that means that, um, every model that we've seen so far has been written in TypeScript. They might call out some API, but it's written in TypeScript. You can see here that the IEE was built in Python, and what you can do with Impact Framework is you can actually call out to a CLI tool and do a shell command instead.

So you don't have to write your models in TypeScript. You can actually write your models in any language that you want, as long as you can create a CLI tool, which adheres to a specific open standard that we've created. It will plug and play into the Impact Framework, which is really exciting. That's all for the demo today.

You can find more information again, as I said, at If. greensoftware. foundation. And as a reminder, also in Feb next year, we'll be hosting rCarbonHack again, and Impact Framework will be the focus, so please watch this space. Next up, I spoke to Navveen and Srini, two other co leads of the project, about potential future use cases.

That was last week when I still had hair. Roll the tape, please. So with me today are two of two of the co leads of Impact Framework, Srini and Navveen. Navveen, why don't you give a quick introduction to yourself?

Navveen Balani: Yeah, thank you Asim. Hi everyone, I'm Navveen Balani. I'm a Chief Technologist with Technology Sustainability Innovation Group, working at the intersection of technology and sustainability. I've been involved in the foundation right from its inception and leading various projects in the foundation. Over to you Shimi.

Srinivasan Rakhunathan: Hey, thankNavveenen. Hi everyone, this is Srini. I'm part of Microsoft. I've also been with the foundation since the beginning and really have got an opportunity to do a lot of learning, multiple projects, all open source, so many people, so just love the energy here.

Asim Hussain: I remember the journey of all the projects that both of you have been part of since the very early days, and we've all started working on kind of measurement with the SCI, and you both started looking at... To calculate carbon emissions of software, we needed, we knew we had a data problem and you both went away and started taking the lead on the SCI data project and really started becoming experts in just the data aspects of software.

I won't go into all of the iterations, but we had a couple of iterations there until we landed on Impact Framework. I think everybody's just seen, we've just given a demo of Impact Framework to a lot of people out there in the world. And they've probably seen like how it works, like in a very practical sense.

But I wanted to just have you two here so we can talk a little bit about future thinking. This is what the Impact Framework is right now. What do we think it can be used for in the future? It's a very generic, it's very low level. So I'm going to throw the floor to Navveen probably to start off with.

Maybe just give us some ideas. Like what, what are your, what, what, what inspires you about this? And what do you think about applications for the future?

Navveen Balani: Yeah, so I think the future roadmap, uh, I think would be to extend the back engine framework for measuring the SCI scores for an AI application, uh, both during training as well as inference. And, uh, from a carbon emission measurement, uh, like during the training phase, uh, so we need to capture the server types like, uh, GPU and TPUs, have data for embodied emissions and capture various utilizations like, uh, memory and CPU. Uh, I think next we need to look at how we could integrate with available models or build models that provide the carbon emission calculation based on available data and obtain the SCI score. And also include various other factors like data processing and storage as part of the SCI functional boundary for overall AI training. And for the inference part, I would say, uh, this might be a bit, uh, tricky, uh, based on your deployment model. Uh, for instance, uh, let's say if you're hosting and have a control of the servers, uh, you can follow a similar measurement approach as in training, uh, and using function units, uh, such as number of, uh, inference calls. But if you have a deployment model, And let's say, which involves maybe a generative AI system, like sending prompts to a charge GPT or any LLM models. You basically need to rely on some proxy data such as API latency, maybe a number of iterations to get the right prompt response. Or overall prompt, uh, context and maybe the length and so on. So this is an area of exploration, uh, where we need, uh, open source, uh, communities to collaborate and contribute models, uh, based on various, uh, research and findings. Probably in the end, our goal is to basically look at impact engine framework to realize the SCS specification and make it easier to report SCS code for any software workloads.

Asim Hussain: Would you imagine a future where, as people develop papers and as they develop more advanced algorithms for AI, alongside this research, you deliver a manifest file, an impact manifest file where you're, you're explaining, this is this new type of AI algorithm, and this is the impact that the algorithm might have.

Copy it, fork this file. Play around with the numbers, see how it works for you. Do you see something like that as well in the future?

Navveen Balani: Yeah, definitely. That's a good point. Having all the research. Papers, right? Have this manifest file, right? And which can make the research also more compatible. And this can be a framework, right? Which can be used by any practitioners, right? To, uh, deliver their CS code.

Asim Hussain: Yeah, great. Yeah, we definitely need to apply this to AI. It's a big burgeoning field out there and it's a lot of computes going into that space. Shrini, what about you? What thoughts have you been given to this? What have you given to this area?

Srinivasan Rakhunathan: I think AI is a fantastic concept because both the Impact Framework and AI can learn and build and evolve and emerge together. That's a great synergy opportunity, but also on the, the enterprise side, or even the consumer side where you have multiple devices, multiple environments, multiple cloud providers to hyperscale providers. There is a big opportunity. With Impact Framework, because it truly is an open architecture of bring your own model or bring your own calculation algorithm. And at the very, I think when we spoke about it at the inception, the very concept we wanted, we wanted to do is to make it as granular as possible. So you have building blocks on top and any technology, whether you have Node. js, you have React. js or on the server side, anything that you can. You just have to make sure that the granular component is straightforward so that you can apply it for any kind of architecture, whether it is mobile or desktop or cloud provider. And that's a beauty of the model. And that's the beauty of the architecture I feel so that it can scale to any use cases, whether it is a consumer use case or an enterprise use case.

Asim Hussain: Yeah, I love that term, bring your own model. I think let's put that in all the material. I think that's a really good way of explaining it, isn't it? Bring your own model. Yeah, I remember when we were, like, one of the things, like, both of you have been also to the people who've been really at the cutting edge of actually calculating software.

There's not many people in this world who've actually. Put the time and effort to really dig deep and figure out the emissions of, of, of a piece of software. And I think one of the things that we discovered throughout our entire process is, as you're saying, Srini, everything's a little bit different. One model does not work for everybody, you need to tweak it a little bit for every single use case and every single, yeah, there's a lot of tweaking and can bring your own model to the situation.

Navveen Balani: Yeah, absolutely. I think the other area is how do you make sure that you get more people to do case studies, which we had been doing through manual calculations. And that's where the open source community can bring in user models for all their use cases. And we can learn.

Asim Hussain: Yeah. Oh yeah. There's been so, so many conversations to, to get us to this point where this Impact Framework, it's, yeah. So many things we can talk about. One of the things I've, I, I found quite interesting. I, I was chatting to a large enterprise organ, an airline, a large enterprise organization that's an airline.

And they were describing, and I was explaining Impact Framework to them. And they were quite interested 'cause they were, because they obviously put a lot of RF qs out, a lot of requests for quotes, requests for business. And what, and they also have stringent rules regarding the carbon emissions. They won't have carbon limits on the software that they build.

And what they were proposing is when an organized, when they put an RFQ out and they get a bid back from an organization, they were suggesting as well as a bid, send us a manifest file so we can see not only how much is this thing going to cost us and how are you going to architect it? But what are the carbon emissions of your proposed approach?

And let me compare versus the other people proposing approaches and not just compare on cost. But compare on carbon as well. And I thought that was really, every time we talk to people about Impact Framework, people always come up with another use case, which is what gives me so much excitement about this project, because people are seeing solutions that we didn't think about when we've been talking about this for as long as we've been talking about it.

Navveen Balani: I think this is definitely a good use case, uh, think of this as a calculator, right, which you can first create, right, how much emissions, uh, your infrastructure will take, and maybe that can be integrated with your CI, CD, Terraform, just to know upfront, right, what would be the impact. And if you can make it simplify this whole integration via plugins or calls, right, and make it native to each, uh, programmers, right, the tools that they use, I think we'll achieve what we want to achieve.

Asim Hussain: Awesome. I think we're coming up, coming up to time now. So one of the things I wanted to make sure. That we cover is actually we're going to be running a hackathon. We ran a hackathon last year. It was very successful. It was the Carbon Hack with a Carboner SDK, but the CarbonHack will be coming again.

First quarter of next year. And the main theme of the carbon hat will be measurement and measurement with the Impact Framework. And we're going to be looking for people to build models, that'd be cash prizes. People build models, build impulse, and we're going to have. prizes for best documentation, non technical contributions, all sorts of different prizes.

And I'm going to have an under 18s prize as well, which I'm really excited for. So we've got the Carbon Hack also coming up next year, which is, and the main theme is going to be measurement and Impact Framework. And I'm just so excited to have. Have that on our agenda. I just want to thank again, Navveen and Srini, you both have been just key players in the Green Software Foundation since day one, this project evolved on all the work that you've done from the start.

And I want to thank you and I'm really excited to, really proud and excited to have you here working all of us together on this project.

Srinivasan Rakhunathan: Thanks, Asim.

Navveen Balani: Thank you, Asim, and thanks to the community.

Namrata Narayan: Thank you. Okay, so really excited to finally see Impact Framework. I hope everyone sees why we're so excited and hopeful about the impact this open source tool will have on the industry. We've got a lot of questions, so we're going to try and get through, um, two of them, um, as best as possible. So, as an Annie panel asked, Would Impact Framework be recommended for use on a simple website, or is it better applied to a more robust application? And Neil Clark added to that question and asked if a website is hosted via a CMS hosting partner, like Acquia, abstracted quite far away from the actual server, how can we capture observation data?

Asim Hussain: Yeah, and to the first question, absolutely. Impact Framework is very general purpose. Even right now, I've been talking to you about how to measure software emissions of a server. One of the samples we have in the repository, I believe we still have it, is measuring our website's emissions, which website is built using GitHub Pages, it's a static website.

And obviously it's like a normal website. We haven't actually built the models that are required to measure that. Unfortunately, that's a limiting factor is you need models. You need models that convert network bandwidth to, uh, CO2. We were going to use CO2 JS. We need models that perhaps take data from, uh, Google analytics to help then use other data to figure out, well, what are the, what is the energy consumption on the client side?

So it's all limited by models. That's what we're limited by. But yeah, you can absolutely, uh, measure, measure websites. In fact, we've had the spoiler alert. You can measure anything with it. You can measure physical products. And I think the last question was around, yes. How can you measure when you don't even have access to the server?

Absolutely. That comes up all the time. We're building this thing. You build what's called an adapter model. So for instance, there's a lot of things, even in the service base, you don't have access to, like when you're running like, uh, serverless functions, you don't know the server. You just have some other.

bit of information like gigabyte seconds. Again, it's limited by models. We need models that convert gigabyte seconds to carbon, but we've started to think about how to create those models. And again, as we come on to carbon hack next year, we're really hoping the community comes together and starts filling in those gaps.

This is what I'm talking about. We need thousands of these little models. We want to fill those gaps and build those models. And you can cherry pick them together in order to calculate all sorts of carbon impacts. And other impacts, my dream is to like add water, add other impacts. I want to, we, it's called Impact Framework, not Carbon Framework.

Namrata Narayan: Thanks, Asim. Another question from Cristiano Elias. I hope I did not butcher your name. Is certification support of the results output in the roadmap of Impact Framework?

Asim Hussain: It's not currently in the roadmap of Impact Framework, but it is something that we've discussed. For instance, one of the things that's been, we imagine is the models that you use, perhaps in the future, we will generate some official models. which may be used more in a certification, but essentially if there is any certification, it's a certification of the models.

So you might get a whole bunch of models which are useful for a variety of use cases, only this, this certification agency will only approve this small set of models or for this type of. Computation. We only accept this model pipeline. We imagine it's going to evolve something around that, but we're a little bit early.

We're just alpha today. So we're a little bit early on to have any kind of specific roadmap items on certification. But it's something that we've thought about.

Namrata Narayan: Thank you, Asim. I hope everyone was happy with those answers. We did get a lot of, many more questions, but we will answer them at a later time. So stay tuned, subscribe to our newsletter.

Chris Skipper: So that's all for this episode of Environment Variables. If you liked what you heard, you can actually check out the video version of this on our YouTube channel. Links to that, as well as the Impact Framework, can be found in the show notes below. While you're down there, feel free to click follow so you don't miss out on the very latest in the world of sustainable software here on Environment Variables.

Bye for now!

Asim Hussain: Hey, everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we want more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. Thanks again, and see you in the next episode.

The Week in Green Software: Modeling Carbon Aware Software

23 november 2023 | 58 min

This Week, host Chris Adams is joined by TU Berlin researcher Iegor Riepin to talk about the benefits - and trade-offs - associated with load shifting over both space and time. Together, they nerd out over the specifics, discuss numbers, and weigh alternative methods of computing with green energy all around the globe. Iegor and his team did a study alongside Google, where they modeled the entire European electricity grid in order to study the effects of different types of load shifting, and how it can be most efficiently applied to the world of Green Software.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Iegor Riepin: LinkedIn | GitHub | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

https://pypsa.org/ [03:21]
RE100 [09:47]
What? The sun isn't always shining?! - by David Roberts [10:16]
XFaaS: Hyperscale and Low Cost Serverless Functions at Meta [47:30]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Iegor Riepen: Capacities of wind, solar, and storage required to achieve hourly matching of demand with carbon or clean electricity are reduced when we increase the share of computing jobs, and that means the share of power loads that are flexible. So, in short, demand flexibility makes carbon-free computing more resource efficient.

If this was interesting enough, But the modeling also revealed some interesting findings about the cost of transitioning to digital services that run on fossil-free power every hour of every day. This sounded like absolute catnip for sustainable software engineers, but because this is also one of the first open studies published around carbon-aware software, it's possible to understand the assumptions behind these results and figure out how they might change in future. So if you want to learn a bit more about this, to help in this quest, joining me today, we have one researcher from the Department of Digital Transformation in Energy Systems at the Technical University of Berlin, Iegor Riepen. Hey, Iegor.

Iegor Riepen: Hi Chris, thanks for having me today.

Chris Adams: Iegor, it's lovely to hear from you again. Before we dive into the world of grids and carbon-aware software and the like, can I give you just a few moments to introduce yourself?

Iegor Riepen: Yes, of course. Hi, everyone. I'm Iegor Riepen. I'm a postdoc researcher at the Technical University of Berlin, where I'm part of the Energy Systems Department. In our department, we use methods from operations research and mathematical optimization, researching the cost-effective opportunities for climate neutrality.

This means I spend most of my time writing, solving, and debugging mathematical models of energy systems. Our research group also maintains an open-source Python environment for state-of-the-art energy system modeling, which is available at pypsa.org and GitHub/PyPSA.com.

Chris Adams: Okay, cool. Thank you for that. So, I know that I actually met some other folks in the same department as you from actually an event called Clean Coffee that I used to run at ClimateAction.tech. Tom Brown came along to one of those Zoom calls back in 2019 and from there we just kicked it off. So, And that's how I found out about anything going on in Berlin. And I understand that you've been in Berlin for a while, but you weren't always studying and working in Berlin. You've been in other parts of the world as well, correct?

Iegor Riepen: Yes, that's right. I'm originally Ukrainian. I lived up until I was 20 or so, and then I moved to Germany to make my masters' studies, then academic career. Before Berlin, I was doing my Ph.D. in the Brandenburg Technical University, where I worked with Professor Felix Müsgens, and I worked on my Ph.D. in various questions of modeling energy systems.

When finishing Ph.D., I joined, as a postdoc, Tom Brown's group in TU Berlin.

Chris Adams: Cool. All right then. And that's how we ended up here then. Okay, so we're just about to dive into the meat of the show. And here's a reminder that everything we talk about will be linked in the show notes below in this episode. So if there's a project mentioned, or a site, or a paper that we refer to, please do write in and tell us so we can update the show notes for other curious souls and help you in your quest for knowledge. All right, Iegor, are you sitting comfortably?

Iegor Riepen: Yes, I am standing comfortably.

Chris Adams: Okay, all right then, let's begin. So it's really tempting to jump right into the nerdy specifics about this research and so on. But before we do, can you share a little bit about the background from your current research for this to provide the context for the work that we're about to discuss?

Iegor Riepen: Yes, of course. So over the past year and a half, my colleague, Tom Brown, who is the head and the heart of our research group, and I have been working on the open-source, model-based research dedicated to various aspects of 24/7 carbon-free electricity matching. This concept usually goes under the name hourly CFE matching.

The 24/7 CFE is a new approach for voluntary clean energy procurement, where companies aim to match their electricity consumption with carbon-free or clean energy supply on an hourly basis and around the clock.

Chris Adams: Okay, so let me just check if I understand that. So normally when people say they run on green energy, they might be talking about things on an annual basis. And this is so hourly, so it's what's 8, 760 hours in a year. So it's 8, 760 times higher resolution. So you don't have this whole thing where if you say you're running on green power, you're not making a claim of running on clean energy at night, for example, something like that, right?

Iegor Riepen: Yes, exactly. That's pretty much about that. So, it's interesting because the 24/7 carbon-free energy hourly margin aims to eliminate all in-house gas emissions associated with electricity use of an energy buyer. So, the strategy also aims at addressing the main problems that exist in much in demand with clean energy supply.

When using so-called Established Certification Schemes, the audience of the Environmental Variables might know all this pretty well, but I'll just name a few to outline the context. So, companies who would like to demonstrate their sustainability credentials might opt for buying the guarantees of origin, this has a story related to Europe or the renewable energy certificates or renewable energy credits.

This is a story related to the United States. A common feature of these schemes is that renewable energy credits are, so to say, unbundled from megawatt-hours of energy. When using the schemes, there could be several problems that would arise. So one of the problems is that assets procured with these credits, or the credits you procured from certain assets, this might not be additional to the system.

This means if you buy renewable credits or guarantees of origin from an asset located in a particular system, Norway, and the asset was located in Norway, is located in Norway, and will be located in Norway, no matter what your procurement, it's hard to claim the change of the system associated with your matching strategy.

The second problem is typically arising with using guarantees of origin is so-called mismatching location. So if demand is located in Germany, and the energy asset generating the Renewable Credit is located in Spain, there will be hours when the grid between Spain and Germany is congested. So it creates an accounting problem.

It's likely not those electrons were consumed, that were produced. So, these two are the so called additionality problem and location matching problems that arise.

Chris Adams: Okay, so if I just check if I understand that. So one of these issues was about additionality you mentioned. So if I understand, what you're saying is, let's say that you got some form of generation. Yes, you get paid for the power you generate. But there's another thing that people are paid for, which is like the kind of greenness of it. And this, as you say, ends up being unbundled and traded separately and there, you, basically a bunch of problems tend to happen once they're unbundled in that scenario there.

So you spoke about something in Spain like, yes there might be a grid where you could theoretically have a solar power, solar farm in Spain generating electricity, but whether it's actually really deliverable to somewhere in Germany is another matter, because it might not be physically possible to deliver that power.

So there's, this is some of the kind of complexity that it's trying to address here, if I understand that right.

Iegor Riepen: That's correct. And some energy buyers do think about this and recognize these problems, and they opt for power purchase agreements. These are bilateral contracts between buyer and supplier. And when signing power purchase agreement, these companies would pledge to buy both the energy, so megawatt-hours and the environmental credits bundle to it. The problem here is that under power purchase agreements, renewable energy supply is typically matched over a long period of time with buyers energy demand. So, for example, there is plenty of companies who are joining the known Renewable 100 group who claim to procure right enough renewable electricity to match their consumption on an annual basis.

The problem here is, guess what, renewables are not generating at all hours throughout the day. And this concept is sometimes called in the literature as renewables are non-dispatchable. Sometimes it's called that they're intermittent. Sometimes people say that renewable energy is variable. There was a recent awesome podcast from Volts, where host, David Roberts, had Jesse Jenkins, who suggested the term weather-dependent fuel-saving technologies, which I think is an excellent way to put it forward,

Chris Adams: yeah,

that's

Iegor Riepen: the way we will likely use it today. So when energy buyers sign PPAs with weather-dependent fuel-saving technologies, there definitely will be times when generation is low, so energy buyers have to depend on procurement from the local markets that likely have some carbon content in it in a given point of time.

Chris Adams: Okay, so, just a quick translation. You said weather-dependent fuel-saving technology. That's opposed to, essentially, if I'm going to be burning gas or coal, I don't have to care so much about the weather, but I'm having to burn a bunch of stuff, and the whole combustion of fossil fuels is one of the problems we have here.

So, that's how you might frame it, right? So, because once you've built something, you don't, presumably, once you've got, say, wind or solar or something like that, it's renewable, so you're not having to purchase the fuel. That's the distinction that we have there, right?

Iegor Riepen: So the term which I referred, which was coined by Jesse Jenkins in that podcast, referred to as dependent fuel-saving technologies, refers to the content that when we use renewables, we do not have to buy fuel, which comes at cost. We do not burn fuel, which comes with carbon emissions associated with burning it, and we don't have to think about the fuel.

Chris Adams: Okay. All right. Thank you. Now we've got a kind of, we've got some of these terms sorted out now. Now we can talk about how some of this might work in the context of data centers and things like that. So maybe I should ask you a little bit about this study. So we've, we've done this work now. Now maybe you could just talk a little bit about what the study that you were doing with Google and with TU Berlin was about for this.

Iegor Riepen: All right, so we have an ongoing research projects where we collaborate with Google. As audience might know, the company has claimed a commitment to achieve 24/7 carbon-free electricity matching in all of their data centers worldwide by 2030. And from our side, we bring an academic look to a broad range of questions relevant for making data centers, and more broadly, any consumers from commercial and industry sectors carbon-free. These questions are like, how can one achieve hourly matching of demand with carbon-free electricity? At what cost this comes? How can advanced tech help? Or what would be the impact on the background grids? We released a series of open studies aiming to address some selected aspects of this complex question. So, where demand flexibility comes to the story? Demand flexibility is the degree of freedom that companies pursuing their goal of reducing carbon emissions can use and benefit from. This refers to a broad range of companies with various degrees of flexibility, which takes many forms and colors, mostly of temporal demand management, and perhaps even more interesting, this refers to the demand flexibility of computing infrastructure or data centers that can be geographically scattered and managed collectively by one entity or one company.

So, data centers as electricity consumers have some special perk. They have an ability to shift computing jobs and associated power loads in space and time. And in July, this year, we released a new study where we focused specifically on space-time load-shifting problem. So we look at the role the space-time load-shifting can play in reducing the costs and resources needed to achieve perfect 24/7 matching.

We thought about the signals that companies can use for shaping their load, and we also looked at the trade-offs and synergies that arise from co-optimization of spatial and temporal load-shifting.

Chris Adams: Okay. So I'll just try to make sure I follow that from a kind of layperson point of view. So there's basically maybe two things that came out of that. So one of the, one of these things is this idea that in addition to, if we know that the amount of power we might have be generated from clean sources might change over time, it sounds like there is a chance to, rather than have to generate more, just actually make supply and demand match just by scaling back some of your own energy usage.

And this is something that isn't just done inside the technology sector. For example, I know that buildings might do this to cool a building down where the energy is cheap, for example, at night, so that when you walk in, it's nice and cool, for example. Or if you're in Texas or somewhere, where it's a, it's a nice, comfortable environment.

So it's taking, that's the kind of moving things through time. And that's one temporal thing, but there's something special about computers and data centers in that. Rather than just moving the energy, you can move the work to somewhere else where they have an abundance of power. And that's the kind of special thing that you ended up doing a bit of study, studies into, right?

Iegor Riepen: Yes, exactly. We focused both on space and time flexibility. We write in the study that space-time nowadays mostly is a story of computing infrastructure, data centers. Temporal flexibility applies to a broad range of companies who have various forms of temporal demand management.

Chris Adams: Okay, cool. All right, then. So we've spoken about that. And as I understand it, maybe I should just ask a little bit about, so what's the benefit of doing it this way? Like, why would you even think about trying to match some of this stuff up rather than just buying a bunch more green energy for example or a bunch new solar farms.

Iegor Riepen: Well, thinking about this, imagine a company with an inflexible demand wants to match its own consumption with carbon-free electricity. It could source carbon-free electricity from various sources. This would be a combination of re-imports if the grid is clean or clean enough for your purpose. Uh, it could be generation of renewable generators procured with power purchase agreements and dispatch of storage assets. This is a pretty challenging task, since battery storage is helpful for bridging some hours of no wind, no solar, but it's not the right technology or better frame, not the right economic technology for times when you need to firm weather dependent wind and solar over an extended period of time.

So, pre research done by Princeton University, NetZeroLab, folks from Peninsula Clean Energy, and also from us last year, did show that 24/7 CFE hourly matching is possible with commercially available technologies like wind, solar, and lithium ion batteries. But it comes at large price or cost premium and with some curtailment of renewable electricity.

If consumer following our imagination have an access to advanced tech like hydrogen storage or clean firm technologies, their price premium could be reduced. Data centers could use this special perk, so the ability to shift loads over space and also time, to relax this problem of matching demand with carbon-free electricity supply.

Chris Adams: Okay, so what I think you might be referring to here is the fact that because there's this flexibility that reduces the amount that you need to buy to have ready, uh, to match the entire time. So it may be that you don't need to buy so many batteries or have so many wind turbines or something like that.

Yeah.

Iegor Riepen: Yes, so put it simply, you can move loads from the places where you have access to the loads where you don't have enough carbon-free electricity, and by moving the slots you could possibly save potential storage needs, and you can also save or reduce the amount of excess electricity you would have that you might have curtailed otherwise.

Chris Adams: Okay, and by when we just briefly touch on curtailing, that means that rather than just wasting, just not being able to use this energy, you're, you're able to put it to use to some kind of productive use, which basically improve the economics means like you might reduce the cost of running something, for example, so it might pay for itself doing something like this.

Iegor Riepen: Exactly. If you have signed PPAs with an asset or you built an asset on site, you likely have some access of energy in some uh, hours. You could sell this access to the background grid, if the background grid takes it at a price, you could potentially store it, but you would store it up until point. The storage is economical and some part which you would not store or you would not sell.

You would typically curtail.

Chris Adams: Okay, cool. And curtailing for the purposes here is basically having to throw it away because you can't use it productively. All right then. Okay, thank you for explaining that. So we spoke a little bit about some of the details on this, and it might be worth just briefly touching on some of the open approaches for this, because you've mentioned a tool called PyPSA, which I think is Python Power Systems Analysis?

It's something like that, right? And this was, there was a the open part of the study is quite a key thing for, for this project. Is that correct?

Iegor Riepen: Yes. So we use um, um, PyPSA, which is an open-source Python environment for state-of-the-art system modeling. This is a tool which our group develops for and maintains for our own research, but also it's been used by wide range of companies, institutions, NGOs, TSOs, by some others who might find a use in open-source tools.

So PyPSA itself stands for Python for Power System Analysis, as it was originally scoped for the power system analysis. However, nowadays the tool is used for many other applications, but for Power, which includes transport and heating and biomass, industry and industry feedstocks, some carbon management, sequestration, hydrogen networks, and what else.

So the open-source Python environment which we ship or which we maintain includes PyPSA itself, which is a modeling framework, but it also has several individual packages that make it possible to go all of the way through the data processing, such as calculating renewable energy potentials in different countries that we model or collecting energy assets data to creating and solving complex energy optimization problems.

So why would open-source modeling be interesting? There are a couple of things which usually comes in place answering the question. This could be transparency and credibility. So by doing open-source modeling, we show that we have no cherry-picked assumptions. In our, um, studies that would drive certain results. Open-source is also pretty useful for reducing wasteful multiplication of work. We can think about it so there is plenty of energy modeling groups in academia worldwide, but also energy modelers in consultancies or in industry. All of us are doing basically the same job, but if all of us have to create an

Chris Adams: your own model. Yeah.

Iegor Riepen: Before we fly the airplane, we are not progressing much, and doing open-source is quite helpful too.

Sometimes you can just copy from somebody who did a good job and put it in this open license and go ahead.

Chris Adams: Okay, cool. That's actually, so that sounds quite exciting. So essentially, if someone's going to say, "oh, I think we can decarbonize this industry by this date." You can essentially model it and say, "these are the assumptions I'm making. This is why I think it's possible. This is why I think we can afford it. And, uh, this is how much I think it would cost," for example. And, uh, the idea that because it's open, it becomes easier for, say, policy makers or civil society to basically say," Hey, you've made a really weird assumption here. I challenge that and vice versa." And with it being modern, open, there were, although we've mentioned Google a few times, this could be used by any company or any organization that might also want to see if this would be applicable to them for in running any kind of infrastructure themselves.

Is that about right?

Iegor Riepen: Yes. When we do, for example, our study, which we today talk about, which is released in GitHub is on a special kit release. And everybody who can run a Python script could reproduce our results, could see how our assumptions are formed, what we put into our optimization problem as parametrization, and if you wish, you could basically get the same result on your local machine.

If you have this way, you can be pretty sure that we did not do some cherry pick stuff to, to shape or drive results in one direction.

Chris Adams: cool. That's really interesting. I didn't know that I could model the entire European grid on my laptop to actually try and settle a bet in the pub. That's quite cool. All right then. So maybe we just, let's go back to this study then. So we've spoken a bit about this and we haven't actually discovered or discussed the findings.

So maybe I could actually ask you, um, are there any particular key findings you'd like to share so far? Or is there any nuance we should be, be aware of before we dive into some of the kind of juicy results here, for example?

Iegor Riepen: Um, yes, sure. I think we could address this nuances topic first. We could just briefly go through the study design and key assumptions driving our results, so the audience would understand where from are. So, for this study, we used the computer model of the entire European electricity system. With this model, we simulated the early operation of the electricity system and the so-called system development, so we looked at the cost optimal investments the system would have in generation and storage assets for the model tier.

We placed five data centers in and we chose the regions where we placed our data center source to capture grids with different sets of features, unique renewable resources, and national characteristics. We assume the data centers have a nominal load of 100 MW for simplicity. This assumption of "What is the exact nominal load of data centers doesn't play any big role in the results?"

If we have smaller or higher capacity assumption, we would observe the same trends.

Chris Adams: Ah, okay.

Iegor Riepen: we And we configure our mathematical problem that all datacenters follow 24/7 CFE goal. So, to achieve this goal, datacenters can co-optimize electricity procurement from the local grid and procurement of additional resources such as storage, wind, PV generators that are additional to the system and located in the same meeting zone.

And we assume that datacenters have some degree of load flexibility, which we vary with scenarios. Stepping from 0%, which would mean that there is no flexible workloads, so all loads are inflexible and must be served at places of data center location, up to 40%, meaning that 40 percent of data center loads are flexible and can be shifted to other places or delayed to other times.

This is what we do. What we do not. First, we do not quantify the actual costs and technical potentials of achieving certain share of flexible workloads, we just say, "hey, this data center fleet have some share of flexible workloads. How would you optimize the flexibility utilization and what benefits it might bring?"

We also treat data centers simply as large consumers that can shift a certain share of loads, which is not too far from reality. But what I mean here is that we abstract from the technical aspects and properties of flexible workloads and some physical constraints of quick ramping. There is, uh, tech folks who could focus on these topics with their, uh, uh, with their, uh, knowledge.

And we base our model inputs only on freely available raw input data. So for electricity systems, we parametrize the system mostly from Danish Energy Outlook. And for data centers, we assume pretty generic transparent assumptions. Where, so we by, by this, we try to keep our study design broadly applicable for other companies who might have their specific flexibility shares or forms or shapes.

And we try to keep our workflow available at GitHub for everybody to access and reproduce.

Chris Adams: Okay, so if I understand that correctly, essentially, you're doing this with as much open stuff as possible so that someone can reproduce this, and the assumptions you're making about these data centers, it doesn't, although the size doesn't matter that much, 100 megawatts is about, that's like a medium sized to large hyperscale data center. So this is, again, it's, it's somewhat reflective of the reality. And you also mentioned that they're in different parts of Europe. So I, from memory, I think this was like Ireland, which is like windy and in the West, Denmark, which has loads and loads of wind, there was like a few other places in Europe as well with different kind of generation and different geographics.

So they're in different places. So it was somewhat representative of the regions we might use in cloud, right?

Iegor Riepen: Yeah, so data centers, eventually we scatter it in Ireland, Denmark, uh, west one zone, uh, Finland, Germany, and Portugal. Uh, the idea was that we would take regions with different renewable resources. First, regions that would be pretty far from each other, and also we would take regions where there is data center consumption in national energy mix.

And by that we take different enough regions and we would capture all the system dynamics that we would want to.

Chris Adams: Okay, cool. That sounds somewhat, I can recognize that with my kind of cloud hat on, thinking about running something in Ireland versus running it in Germany in this scenario. Okay, well, that sounds like we've given enough background for this. Should we dive into some of the findings? Is there anything you'd like to, so yeah, maybe I should ask, what was maybe one of the first findings that really caught your attention that you'd like to share from this?

Iegor Riepen: So I think we could go through several steps. That's the first finding that usually has been caught by people looking at our study is this topic of resource efficiency and cost reduction. Just for audience to understand, from our model, as a result of optimization, we get procurement strategies for each data center, which optimize to match demand with carbon-free electricity around the clock with some desired quality score.

So the cost optimal technology mix that we get as an output depends on various factors, for example, renewable resource, this would be wind to solar average energy yield, or cost assumptions and many more. So the clear trend that we observe across all scenarios is that capacities of wind, solar, and storage required to achieve hourly matching of demand with carbon or clean electricity are reduced when we increase the share of computing jobs, and that means the share of power loads that are flexible.

So, in short, demand flexibility makes carbon-free computing more resource efficient. What we also could do, we could retrieve the cost of any procurement strategy from our model, and thus we can map the resource efficiency to the cost effectiveness, meaning you pay less to achieve exactly the same. So the degree of this cost effectiveness scales with the level of flexibility that we assume.

So for the corner scenario where we assume that 40 percent of flexible workloads perfect hourlCFE matchaging and co-optimized space for achieving the overall energy costs of the model data center fleet are reduced by up to 34%. So this refers to the cost saving of an individual data center, but it refers to the cost saving of a group of data centers scattered geographically and managed together by one company.

So these data centers consume basically the same amount of megawatt hours, but do shift their consumption in space and in time to optimize the resources.

Chris Adams: Okay, so if I just, maybe I'm just trying to run that by you so I understand. So essentially, you model different amounts of flexibility in a system if you're controlling multiple data centers here. And essentially, the more flexible you make it, the more you can actually reduce the amount, the cost of actually having to buy all these solar farms and wind turbines and batteries and all the way that, all the way up to the point where, if you're doing 40, if you're got 40% of your loads being flexible, then it reduces the costs by about a third, essentially. That's what I think you're saying there.

Iegor Riepen: Yes, that's right, but we should see it in perspective that this 30 percent cost saving is basically our corner scenarios. We scale the cost very up, so scaling very up means we look at the perfect matching between uh, demand and carbon-free electricity. 100 percent that you are not allowed to have any gram of CO2 peak allowed hour of consumption.

And also we assume in this scenario that 20 percent buyers only have access to wind, solar, and lithium ion storage. So this is a palette of commercially available technologies which are just hard to use out of these technologies to make 24/7 strategy. If these two are right, then your costs are reduced by up to a third, and the costs are reduced less than that for all palette of asset technologies and scenarios that we considered in this study.

Chris Adams: Okay, cool, well thank you, that's, that was bigger than I was expecting it to be. Uh, that's, that's, that's, that's, that's pretty impressive if you're going to be spending, um, literally billions on power, like some large, uh, data center providers or data center users will be using. Okay, then, um, Iegor, so, if I understand correctly, the amount of flexibility you might introduce, this kind of carbon-aware computing, that, if you say that you need to be running everything on 100 percent carbon-free or fossil-free or clean energy, based on this, then this will reduce the amount that you need to purchase, which essentially makes it more affordable or more accessible to a wider number of operators, I suppose. Are there any other findings that you would draw attention to in this study?

Iegor Riepen: Um, yes, of course, what we do in the study, we take a look on the signals that companies might use to shape their load following strategies. So to discuss the signals, we could firstly go for special shifting story and then for temporal shifting story. So for the special shifting story, the one signal which comes up front is the fact that hourly profiles of wind power generation have a low correlation over long distances due to different weather conditions. So as a rule of thumb, you can think of the following. If two generators are located as far as 200 kilometers from each other, the hourly feed in from these assets have very low correlation, and data centers could arbitrage on this effect, or put simply, they could move load to locations when and where there is a high wind generation.

That's saving the cost of energy storage and thus reducing the amount of solar curtailment. So, the hourly profiles of wind generation is not the only signal. Another signal that we discuss in the study is the difference in quality of renewable resources in regions where data centers are located. So, the quality of local resources, or in other words, the average capacity factors of Solar PV in a given region, they translate to the cost of electricity.

The higher the quality of renewable resource, basically the lower the average cost per megawatt hour. Special load-shifting is possible. A rational buyer could just adjust own procurement strategy to contract generators in better locations. So those locations where renewable assets have lower costs and co-optimize special shifts accordingly.

In the study we illustrate this mechanism with a data center located in Ireland, so it's not the most sunny region in Europe, that would tend to shift loads away during daytime through mid spring to mid autumn. So data centers located in Germany and Portugal, the regions with much better solar resources than in Ireland, they would tend to receive loads during this period.

This feature would just work about reciprocally for wind-related load shifts. Data center in Germany would benefit from having partners in Denmark or Ireland that have much better quality of wind resources. And, so there are two signals for spatial shifting. There is one more, which we did not put much of focus in our study because of our geographical focus scope, but it could play a role for spatial load-shifting.

So if you look above Earth from the North Pole, and the Earth would rotate counterclockwise from West to East. And we are pretty sure that it rotates with a constant predictable speed, roughly once per 24 hours. So if data centers are scattered across the globe in distant locations and operated by one company, one could imagine a load-shifting strategy where loads would follow the sun.

Chris Adams: Yeah.

Iegor Riepen: So these are basically three signals for spatial load-shifting for temporal load-shifting story, we illustrate cases when the variability of the regional grid emission intensity could drive the carbon-aware temporal load-shifting. So the grid signal can play a role in load-shifting strategy if data centers have electricity imports from the local grid in their energy mix.

And the temporal flexibility could also be helpful in aligning the demand in time with the generation of procured renewable resources.

Chris Adams: Okay, so it sounds like there's almost two kind of scales you're working at here. So the first thing you spoke about was like, say, Ireland and Germany and Portugal. Essentially, so basically Germany and Portugal are sunnier, and Ireland is windier, and during the summer, they're going to be, Germany and Portugal are way sunnier, so if you were running say computing in these three places, you might choose to run more of it in those two during the summer, and then as it gets a little bit darker, you basically choose to run everything in Ireland instead, and that's going to be a much more efficient way to actually, essentially run, maybe, if you're going to run computing jobs at 100 percent carbon-free energy, that's a way that you can do that at one of the lowest costs.

So that's, there's one thing happening at the annual level, but you also said there's a kind of another thing which is much more tied to the kind of, you know, day and night cycle that you're referring to as well there. So there's different speeds that you might be thinking about, different trends that you might take into

Iegor Riepen: consideration.

Yes, we talk about pretty complex optimization problems that spans across space, spans across time, and the signals that would drive optimal utilization of flexibility through this space-time graph have various shapes. Some signals have stochastic pattern, like wind and feet, um, which is uncorrelated over long distances. Some signals have predictable pattern, like solar profiles that follows the Earth's rotation. And some, um, some signals have something in between predictable and unpredictable.

Chris Adams: Oh, cool. Wow. That's, I wasn't expecting that. Yeah. So we spoke before about, okay, one thing that you could do is essentially during summer, you're running your computing jobs in Germany and Portugal where there's loads of sun and loads of clean energy and that's relatively cheap. And then in winter, you'll choose to run it maybe Ireland, or somewhere where it's a bit darker, a bit gloomier, but way windier. And, uh, where there's loads, oodles of green energy there. But you certainly have some trade-offs that you have to make here, if you were to choose this. Maybe you could just expand on some of that a bit more, so that people understand, so that it doesn't sound too good to be true, for example.

Or people understand some of the specific nuances here.

here

Iegor Riepen: Yes, very right. What we do in the study, we take a look on scenarios where we co-optimize um, and Isolate Utilization of Spatial and Temporal load-shifting. So these scenarios with isolated flexibility can be seen by just academic exercise, but it's pretty useful for us to take a look on the system mechanics and get a feeling of the numbers.

So as a result of utilization, we can retrieve the value, which would represent something like reduction of the overall annual energy cost of a carbon-free electricity supply. If a data center utilizes either spacial load shifts or temporal load shifts or both, if we compare the value of spacial and temporal load management, uh, when, um, spacial and temporal stories are isolated, we come to the numbers of something from 6 to 1, depending on the scenario, in favor of spacial load-shifting.

So this means, spacial shifting workloads across locations brings you 6 to 1 high amount, high value. And this takes place because datacenters can arbitrage on differences in weather conditions and take advantage of. So this is a mechanism which we have just discussed. Shifting workloads across time to bring a higher value, uh, requires a few things.

So for that to, to have a high value, datacenters would need to buy electricity from the background grid, which is high variability of the original grid carbon emission density. So if a local energy mix is flat, dirty, or flat, clean, there is no, basically, value to shifting workloads from one time to another.

And the tr-de offs would appear uh, between spatial and temporal load-shifting when both are implemented together. So one can think about this in this way, if you have a certain share of flexible loads and you would like to shift some in space and some in time, but whatever you shift from other places and other times cannot exceed the upper cap, which would be the computing capacity constraint.

And whatever you shift away, that means to other places or to later times, cannot seek the lower cap, which would be the flexible workloads cap. And, uh, one thing on synergies, what we do show is that co-optimization of space and time load-shifting can yield benefits that go beyond the value of each of the two individual mechanisms could bring alone.

It's sort of an expected outcome for any operation's research problem. If you have two degrees of freedom, and you co-optimize them, uh, you could co-optimize in a way to get the benefit from synergies of them both. So in the study, we come to this point from various angles, but here's just a good example.

Imagine if you have, say, three data centers scattered far from each other and operated by a single entity. Then imagine we have each data center, which has a mix of wind and solar capacity built on site. Let's assume that this data center can shift workloads with any fixed volume of flexible workloads.

Now, somebody comes and says that, "hey, we have one long duration energy storage asset that we could place in either of the three data center locations."

Chris Adams: A big-ass battery, basically. Yeah?

Iegor Riepen: Yeah, the question is here, where would you place it to reduce the energy cost of the entire system? If you would write an optimization problem for this, solve it uh, we would tend to see that the optimization problem suggests us to harvest renewable electricity in the best locations. So those locations where the lowest cost per megawatt hour is achievable. For example, Denmark, Ireland, with good wind conditions. And integrally opening access to this cheap, clean electricity for all locations through the spatial load-shifting.

Chris Adams: So that thing you mentioned there was, you've got this notion of moving things spatially or moving things temporally through time. So it's, essentially, you get more of a gain from moving things geographically, spatially, if you're going to do nothing, if you just only do one. But that can be a little bit harder for organizations. So there is some gain from doing things temporarily, but on the temporal scale, you do need the grid to be a little bit more volatile, moving back and forth between very suddenly and then not very suddenly, for example, so they start using back, go back to dirty energy essentially, but the thing that you can do is these do work together, so you can move through time and space, and you do end up with a, basically the benefits do compound in this case here.

Iegor Riepen: Yeah, and it's worth saying that the benefits actually do not compound simply together, because the spatial and temporal stories are subject to the shared set of computing capacity constraints. So when you co-optimize both, you inevitably have to trade off among them.

Chris Adams: Okay, compound is the wrong word then. All right. But basically, the, by doing the two things together, you can get, you can get a better saving than just doing one of them by themselves, for example.

Okay. All right then. So we've spoken about this, how it's been applied to one company. And we've said that this could be used for multiple organizations. Um, presumably, if you... If someone was to do this, you could do this for an entire sector to figure out what the power might be needed for an entire sector to see how much you might need to deploy to, to displace all the kind of fossil-based energy generation that data centers use, for example. Is that plausible that you could do something like this with this kind of modeling?

Iegor Riepen: Um, well, in our study, we try to keep our assumptions on carbon-aware computing, in the way how we treat data centers, how we treat flexible workloads, as general as possible. So the study results should be applicable to the broad range of companies operating data centers, with their specific features and their specific workloads.

And as well as the study should be applicable to a broad ranges of companies from commerce or service or industry sectors for which, say, only a temporal story is relevant. One cool thing here is that data centers can pave the way for space-time load-shifting applied for other industries. Other applications we are not yet even aware about.

So just to mention, I recently visited, um, the group in the University of Wisconsin, Madison. It's Victor Zavala, uh, Scalable Systems Lab. So this is a bunch of awesome people. I spent with them three weeks. I'm not really sure what they work about, because they work about computational chemistry, on energy systems, on graph theory, optimizations, programming.

I think they crush every problem that people are throwing at them, and when I was there, they were launching a new project which was focused on exploiting space-time interdependencies between electrochemical manufacturing and power grid. So the idea here is that the electrochemical industry would shift loads so as to co-optimize the economics and to reduce the carbon emission intensity of the electrochemical manufacturing, which is pretty dumb.

Uh, because before I thought that, that space-time shifting is more about data centers only. But now, well, there are applications beyond only this sector. And I think this is a future where we are going to.

Chris Adams: Okay, so basically, as more and more, as more and more clean and, what's it, variable fuel saving technologies come onto the electricity grid, it's going to get more and more kind of upy downy variable. And it's not just data centers this would be applicable to. So electrochemical stuff would be like synthesizing fuels or making plastics or things from carbon captured into, by the, from the air, for example, or things like that. I think Tom Brown mentioned a little bit about making methanol in this kind of way, or some of the green hydrogen stuff around splitting water into hydrogen and oxygen for creating chemicals that way. So is that what you're referring to in this scenario?

Iegor Riepen: well, I don't know what they will do in the project. It will be very interesting to take a look. I'm not sure that it's very simply mapping one to another. So for data center special shifting means that you move. Workloads and associated

Chris Adams: space, that

Iegor Riepen: one place to another. But this moving means that just, um, computer jobs are being executed in one data center and not in another data center.

While the end consumer is somebody waiting for the YouTube video to be rendered, consuming the goods at the place it would be consuming no matter the shift. How exactly it works for the electrochemical industry, not sure, but we will see once the project is developed.

Chris Adams: I see. Okay, cool. All right. You mentioned one thing about this being something which is more generalizable to a wider set of technologies, and one thing that actually this makes me think of is a new paper that was published by Facebook. They have a serverless platform which they call XFaaS, and one of the key things that they were doing is actually having this kind of geographically movable computing that they refer to, and one thing that, what really strikes me is that Facebook basically said by allowing the actual computation to be flexible in terms of where it's actually run, they were able to massively increase the use of the data centers they were using.

So if you think about it, for most data centers, there might be a single digit percentage utilization. So most of the time, not doing very much. Cloud might be estimates of around 10, 10 times more efficient. So maybe 20 to 30 percent efficiency for most very well-run hyperscale data centers. Facebook themselves say that we've, by introducing some of these ideas, we've been able to get up to 66 percent utilization, which basically means there's a bunch of extra hardware that they don't have to buy and build, which is good because they are spending lots and lots of money building with data centers in lots of places and anything that you can use to reduce the number of data centers you need, in my view, is a good thing because that's a lot of buildings that don't get built, for example. But the key things they mention in the paper, and we'll share it in the paper, we'll share the link to the paper, is that if you have maybe a computing job or you've got a function or anything like that, you basically have like, they make this stuff possible by adding kind of deadlines or saying how tolerant of being moved through time or moved through space a particular job might be. And this feels like, this kind of hints that this might, might be a way that might become a norm for working with computing, where if you, if you don't need to have something happen right away in the same place, then you can basically get all these extra benefits by being a bit more flexible about this and saying up front, saying this stuff up front when you submit a job to a computing cluster or something like that.

Iegor Riepen: And it's pretty relevant to ask the question, it's not only how, what are the benefits for the operator of the data center, but what are also the benefits more broadly to the background system in terms of costs and emissions.

Chris Adams: All right. Okay. So we've, I know that you've just, you released a study in the summer and there's some stuff which people can refer to here, but I also am aware that we are in a fast moving field and you just mentioned some work in the University of Wisconsin, I think. Are there any other things you would like to have included into this kind of research, or you think people should be looking at over the next 12 to 18 months inside this, that might influence how people might think about carbon aware computing or set of changes? So, flexible computing like this with a view to reduce the emissions associated with running infrastructure that we all rely on right now.

Iegor Riepen: Maybe one study or research paper which I would love to see is that if somebody would take the courage and to illustrate the system-level benefits of carbon-aware computing across different contexts and different states of the system. So by system-level benefits, I mean from a society perspective, so we look at the costs or total carbon emissions or total curtailment of renewable energy and so on.

And by different contexts and states of the system, I mean the following. Let's think, nowadays, there are mostly companies who buy electricity from the local grid, they have some flexibility, they would go to the data providers, such as electricity maps, providing the short term forecasts, or carbon emission intensity, and they would factor it in, into their load following strategies.

This can work for the temporal shifting, also soon will work, I believe, broadly for the spatial load-shifting. So in this case, space-time shifting can help if you just buy from the grid. So there is a follow-up to this, so some companies might go beyond that and buy additional resources to eliminate all of their carbon footprint completely.

So for that, space-time shifting could also help, and this is basically what our study is about. It would help you to be more resource efficient, it would help to be more cost effective, it would open 24/7 cFE for a broad palette of companies who would not maybe jump there otherwise because of high cost premium.

And in the future we will hopefully be reaching the net zero electricity systems or more broadly net zero energy systems. And space-time shifting can be of help there too. So we would need some set of solutions where we would firm the variable wind and solar. We could think about a palette of solutions on the supply side.

It could be grid connected battery storage. There could be hydrogen storage if in the region where there are salt caverns. Or there could be even energy storage in the liquid hydrocarbons like methanol storage. My colleagues Tom Brown and Johannes Hump recently published a paper on this. But these are all the solutions from the supply side.

There could be solutions on the demand side, where there are large data centers that can move large loads across space and time. They could help the system to firm the variable wind and solar and provide the service for the system and get some sort of remuneration for that. By the way, Victor Zavala's group has also published a research paper where they make a mathematical modeling sketching out what type of remuneration they can get for providing the service for the system.

Chris Adams: Okay, if I could just quickly stop you there for a second because I want to check I understood it correctly. You're essentially saying that rather than it just being about looking at the cost only to say, in this case, it was like one tech firm looking at how much it would cost them, you're essentially saying it's possible to model this to say how much this kind of flexibility can save everyone else. If you actually had these providers, like a data center as a kind of active participant inside the grid, because that might reduce the amount of generation that the grid might need or that society might need. So essentially, it's like flipping it around saying, well, actually, is there some kind of value that can, or are there benefits that could be shared just outside of just the corporate, just outside of that company? Can it benefit other people as well?

Iegor Riepen: Yes, exactly. And more, more than that, so whenever we look at this context, either your company buys electricity from the grid and tries to move load across space and time, either to reduce costs or to reduce emissions, depending on what signals the company takes. Or if the company goes 24/7 and wants to eliminate all of the emissions and have a high impact on the background grid.

Or even if we even reach net zero systems, in all of these contexts, space-time load-shifting might be of help and might bring benefits, both for the companies operating it and also for the background systems. So if there is a study who would take courage to highlight all of this transition phase and illustrate the benefits for systems, this possibly would be a really good read.

Chris Adams: Cool. All right. Well, that sounds like something for people who are curious about playing around with this on GitHub or want to mess around with some of this modeling themselves to see. And I know there are a number of organizations and people like in software development houses who are actually trying to extend various tools like Kubernetes to incorporate some of this stuff so that you can actually see, so you can essentially design from a very, from the get go, just like I mentioned with Facebook.

So like Facebook's XFaaS paper talks about how they encode a degree of tolerance into this. But I believe that Intel is doing something similar to this for their versions of Kubernetes. There's, I'll share links to that for people who are listening. This has been pro, possibly one of the nerdiest episodes we've ever done, but I've enjoyed myself for this, Iegor.

Thank you, Iegor. But before we wrap up, are there any things that if someone. So, if someone did, if someone has followed this and was able to keep up and was really curious and would like to learn more, where would you direct people to look if they wanted to dive into this some more themselves?

Iegor Riepen: Well, if people would like to know more about our research on spacetime shifting or more generally on 24/7, they could possibly visit our GitHub page. So it's github.com/pypsa/247cfe. There in readme we explain what other research we're doing, how to clone our work on how even to reproduce it.

If people are interested in PyPSA ecosystem for open-source energy modeling, they could visit PyPSA.org or more generally if people are interested in the open energy research in general, it is living an open mode initiative do to which collects various research groups and open models all about energy.

And finally, if com, if there is somebody interested in voluntary energy procurement, one could visit 27 Compact, which is gocarbonfree247.com, which is place collecting people and companies working on this.

Chris Adams: Great. And Iegor, if people want to find you or follow some of your work directly, is it TU Berlin the best place or is there an Iegor Repin on LinkedIn or something that you would direct people to for future questions?

Iegor Riepen: I have my LinkedIn, which I could attach and there is TU Berlin email.

Chris Adams: Brilliant. Okay. Well, Iegor, I've really enjoyed this. Thank you so much for diving down into the depths of CarbonAware and time-space shifting computing like we did today. Oh yeah, and happy birthday, by the way. I forgot. Yeah.

Iegor Riepen: Thank you, Chris. Thanks for having me today.

Chris Adams: All right. Take care of yourself and yeah, have a lovely week. Cheers, Iegor. Hey, everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware. foundation. That's greensoftware. foundation in any browser. Thanks again, and see you in the next episode!

The Week in Green Software: Greening Web Standards at the W3C

9 november 2023 | 45 min

TWiGS host Chris Adams is joined by special guests, Anne Faubry and Alexander Dawson of the W3C Community Group. This week, they discuss the Web Sustainability Guidelines as well as the Content Accessibility Guidelines. Web standards as specified by the World Wide Web consortium, play a huge role in shaping expectations about how the web is supposed to work, and for whom. This week’s guests talk about their respective roles in the W3C and in publishing the guidelines, as well as how they came to be part of the group. Listen in to learn more about the differences between standards and guidelines, and what the Web Sustainability Guidelines aim to achieve.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Anne Faubry: LinkedIn
Alexander Dawson: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Resources:

afnor spec 2201 [10:37]
Référentiel général d'écoconception de services numériques (RGESN) [10:53]
Web Sustainability Guidelines (WSG) 1.0 at a Glance [22:28]
Checkboxes | W3C [23:40]
JSON API | W3C [23:58]
GitHub - w3c/sustyweb: Sustainable Web Design Community Group [24:16]
Web Content Accessibility Guidelines (WCAG) 2.2 [26:31]
GRI - Standards [29:49]
Issues · w3c/sustyweb · GitHub [38:05]
Sustainable Web Design Community Group [38:30]
Sustyweb Community Home Page
[email protected] Mail Archives [38:51]
The Tarot Cards Of Tech | Artefact [39:37]
Sustainable Web Design Community Group Wiki [42:02]
Frequently Asked Questions - Sustainable Web Design Community Group [42:11]
The intro guide to digital eco-design | Designer Éthiques[42:57]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Anne Faubry: This is what is called externalities. It's a debated word. It's a bit controversial because it's, it comes from an economic background, but it's important to have this in mind when you design a product because sometimes you might have the best intentions, but when it comes to environment, you can sometimes make things worse than what you wanted to do.

Alexander Dawson: Yep, okay. I'm Alexander Dawson, and I'm a freelance web developer from the UK. I also am a part-time sustainability researcher and writer, and lately I have had the pleasure of being the co-chair of the Sustainable Web Design Community Group, and I'm a co-editor and contributor to the Web Sustainability Guidelines, which of course we're all here to talk about today.

Chris Adams: Thank you, Alex. And Anne, the floor is yours.

Anne Faubry: Hello everyone, I'm Anne Faubry, so your pronunciation was great, from France, I'm a UX/UI designer. I work as a freelancer for environmental projects like NGOs or public services, and I try to implement accessibility and eco-design in all I do. And I also work part time for an association called Designer Éthiques, Ethical Designers in English, which tries to research the field of responsible designing and also more specifically eco-design. So we published the guidelines on this topic and, and that's how I ended up joining the W3C group and becoming the community co-chair of the UX part.

Chris Adams: Okay, cool. Thank you, Anne. All right. For folks who are new to this podcast, I mentioned my name, Chris Adams. I am the executive director of the Green Web Foundation, a Dutch nonprofit focused around an entirely fossil free internet. And I am the policy chair for the Green Software Foundation, the large industry body that publishes this podcast. Before we dive in, there's also a quick reminder that everything we talk about to the best of our ability will be linked in the show notes. below in this episode. So if there's a project that's mentioned or a site we discuss that isn't in these show notes, please do write in and tell us so we can update these for other curious souls and to help you in your quest for knowledge around all things sustainable and digital. All right, Alex, Anne, are you folks sitting comfortably?

Anne Faubry: Yes, we are.

Chris Adams: Okay. All right. So the first question is that sustainability on the web. This has been a long time coming, and I know the W3C Sustainable Web Design Group, or Sustyweb, if you look at the URL and it's to its friends, it's been around for a while with varying levels of activity. And I've got to ask, why did you look to the W3C group in the first place? And maybe one of you might help explain what the W3C is in this kind of context, because we've done a bit of an introduction, but it might be useful to give your understanding, and for one of you to show your kind of story about why you got involved.

Alexander Dawson: I guess the best way to start things off is to, well, essentially mention that we are effectively a community group, so we are not governed by the W3C. That's an important point to underline from the start. Community groups are a great incubator for groups that are effectively volunteers to gather together their collective experience and to talk about how they want to improve parts of the web and to look at topics that interest them and to progress those subjects further. And obviously our group, the Sustainable Web Design Community Group, it was started in 2013, and it really started as a link sharing group where anything to do with sustainability, people were passing useful bits and pieces to do with research and interesting URLs and such.

But it was really around the time of the pandemic hit that people started to sit up, take notice and think, "actually, something could really be done in terms of sustainability on the web and real improvements could be made." And we had a real influx of new members. And the community really blossomed in terms of life.

And it was around that time that myself and Tim Frick, who's the other

been an

Chris Adams: institution over this,

Yeah,

Alexander Dawson: we got talking with all the other members in the group and it came to consensus that maybe looking to other W3C bodies like the Web Accessibility Initiative, they've done such great work with things like the Web Content Accessibility Guidelines in terms of improving accessibility on the web, that actually being able to produce some good solid guidelines that are evidence-backed and collaborate in terms of producing evidence and research, and hook into policy and all the best practices that are out there, all within one collective source, that having something out there that people can quickly and easily reference would be a good way for people to make their websites more sustainable.

So over the course of coming up for two years, we worked very hard and, using various different groups, so we had groups for user experience, which of course Anne was involved in and I was involved in web development and we had groups for business and we had groups for metrics and we had another group which was for hosting and infrastructure and people effectively took their choices as to where their expertise lay, and we built up a huge big collection of these Google documents, these drafts, filled with all of these different guidelines and they've effectively been formatted into the guidelines that we now see today, and that's pretty much a good jumping-off point where I can pass over to Anne to give her experience as to how she found it and what her experience has been throughout the creation process.

Chris Adams: Yeah. Go for it, Anne.

Anne Faubry: Thank you. Indeed, I jumped in the project much later than Alexander. So for me, it was a whole new experience. It was my first contribution to the W3C, but it was not my first contribution to this kind of guidelines and standards because I had done this work previously in France on governmental standards and also by writing the Eco-design guidelines for designers. So, my point was to take this experience and help the group move forward by contributing with all the work we had done in France already on this topic so we didn't have to start from a blank slate and that was really exciting because it was also taking things to a next level, to a worldwide level, and, and so it's not only France in its little part of the world working by itself, and it was contributing and also challenging sometimes the guidelines with people from all, all different countries. So it was very exciting. And how it worked was a lot of collaboration with online tools, frequent video conferences, sharing tools, and contributing on the guidelines all together, and commenting on each other's work, and it was a lot of live exchange, and overall very efficient. Also thanks a lot to Tim and Alexander's work, who overviewed all of it.

Chris Adams: Okay. And this is Tim Frick of Designing for Sustainability, uh, that he wrote in 2016. So it's, uh, nice to actually see this kind of path all the way through. And if I may, I just want to touch on a couple of things that you said there that sounded interesting. You said you did some work earlier in France already, and I understand, would that be a I know there's one organization there, and I think you mentioned Eco-design as in Eco-design FR.

Could you maybe touch on some of those two points, please?

Anne Faubry: Uh huh, so there's actually a lot of different guidelines existing in France, and they are now converging. So there are two governmental ones at the moment, one by the AFNOR, which is the French Standardization Organization, and it's now being taken to the European level. And there's another one, which is the RGESN, which is more, like, only guidelines at the moment, not necessarily a standard so far. But it's, I was actually this afternoon working on this project again. They are now consulting a lot of different professionals and experts on this question to take it to the second version and to make it eventually, if also the legal context and political context allows it, to a standard level. At the moment, it's just guidelines, but we can see in the digital sector industry, it's still, it's already much used by many organizations to promote what they're doing in terms of a sustainable responsibility. And then there's different guidelines, most specific for developers or designers who are more focused on the web, easier to implement. So that's what we've done also with the, within the association for designers. Green IT is a great association doing this for developers and other, also other profiles. And so a lot of associations have popped up within the last years to help understand those guidelines, because they're not necessarily easy to handle.

Chris Adams: Or in English, yeah, as well. Okay, so, it's, I didn't know there was such a strong francophone influence, and it's, this is one thing that we found with the Green Web Foundation. We end up working with Gautier, whose name I never pronounce properly, for the Fog of Enactment, specifically because there's so much good work being done in France, that in many ways is quite a bit further ahead than some of the things that might have been outside of, say, the francophone, like, in, in particularly in primarily English-speaking communities, for example. And maybe this actually speaks to one of the things we might talk about later because there is something in the region, I think it's 93 separate recommendations in the guidelines, or is it 93, Alex? It's more than 90 at least, isn't it?

Alexander Dawson: Yeah.

Chris Adams: it's a lot here. And the thing I was going to ask you is, these guidelines cover a wide range of topics, as you mentioned, so like UX, business, web development, just, and they also don't just talk about carbon, because at the Green Software Foundation, there's a focus primarily with the, on software carbon intensity, and that's the actual specs specifically speaks to this. And while there is an interest in addressing some of the other kind of dimensions of environmental impact that you might care about, that's been the focus for the time being. I wanted to ask, Alex, if I could put to this, this to you first and then come to Anne. Could you maybe talk a little bit about the reasoning behind this decision to cover a number of different areas and then some of the trade-offs you'd have to, you had to make while deciding this kind of scope?

Because you can go wide or you can go narrow and they both have pluses and minuses.

Alexander Dawson: When we came to the decision of producing the specification, we really decided to go with an ESG approach. And obviously with ESG, it's not just about the environmental factors, it's also about social and governance and also economical. And... One of the things that we came to realize when also incorporating and looking towards things like the web content accessibility guidelines, especially for myself, I come from a inclusive design background, and it's often in the cases of things like inclusive design and climate change, the people who are most risk of being affected by subject matters such as climate change, those in at-risk scenarios, so older generations, younger generations, people with accessibility issues, people from low-income backgrounds, they're the individuals who are most often affected by climate related issues, such as people in developing nations and people, as I said, who may have accessibility needs often can be the ones who are at the blunt end of the climate related instrument. So it's one of the areas that we wanted to implicitly tackle with the guidelines to be aware of the human factor. So incorporating people and the planet within the scope of the guidelines. So things like accessibility, privacy by design, security measures, and well, um, there has been debate as to whether these things should be covered within the scope. There's a very good argument that obviously taking into account things like accessibility, for example, if a website is accessible, then it's going to actually reduce carbon emissions by default because someone with accessibility needs is going to be able to access that information much easier and they're going to spend less time on their screens or scrolling through information or having difficulty encountering barriers to access.

And that's going to waste less, not only time actually on the screen, but their device actually having to work around the barriers trying to get through those particular issues so there's a multitude of complexities and variables involved but implicitly working around all these issues such as security issues and privacy issues and accessibility issues, actually, as a secondary factor, can help reduce the carbon factor.

So it's something that we wanted to include within the scope. And it's something that we've run through, not only with the benefits, as can be seen throughout the guidelines, but it's something that we've integrated within the, the ESG measurements.

Chris Adams: Okay, so you're speaking not just about carbon and there's almost, it sounds like in sustainability lingo people talk about co-benefits, as in something that might help the climate but it also has a, kind of a societal good as well. That's one of the angles that you have there. And, Anne, I want to give a bit of space for you to talk about some of this as well because I I do know that France has had a typical, has a history of speaking about more than just carbon. Possibly because of the fact that the energy is already very clean, but it also means that there's other things we need to take into account like water and the material depletion and I forget that, I can never remember the other one that we actually refer to for this, but I know this is some stuff that you might be talking about.

Maybe you could talk a little bit about the range of different communities that you're coming from because you're a UX and UI designer and that's not the same as development and these cover quite a broad range of functions I suppose is the way you might talk about them.

Anne Faubry: That's a very good point. As a designer, you have to think, all the jobs to think about maybe even more as a designer, think about the externalities of what you're designing. And of course, I'm not going to come back on what Alexander said. It was, uh,

Chris Adams: a second, sorry Sam, and externality in this case, that's like shifting the cost outside onto outside of the organisational brandery? Maybe you could just touch on that because I think it's a word that people might not be so familiar with.

Anne Faubry: So, when you design a product, sometimes you're going to have some rebound effects or other effects which can be good or bad, but which were not anticipated. And this is what is called externalities. It's a debated word, it's a bit controversial because it's, it comes from an economic background. But it's important to have this in mind when you design a product because sometimes you might have the best intentions, but when it comes to environment, you can sometimes make things worse than what you wanted to do and what was the previous case. And that's what happens sometimes, if you only look at carbon, you might have some pollution transfers, especially indeed if you have a clean energy. I don't want to get into the nuclear debate, it's not that simple, but indeed regarding the carbon, it's pretty clean. And what we can totally see in the digital industry is people tend to replace devices for new ones because they're more energy efficient. And they tend also as designers or developers or architects to make their choices regarding a lot of the energy consumed. And uh, if you look at the other planetary boundaries, there's nine of them, carbon is only one of them. But there's also the water use and the pollution of the water, the soil, uh, the air. So if you only look at, uh, energy consumption and carbon related to it, you might forget the mineral extraction and all the biodiversity, uh, destroyed with that. And, and this relates also to the human aspects Alex was mentioning. So it's very important not only to look at carbon, but also to look at different environmental indicators, like water or resources consumption, to make sure you don't have these pollution transfers or bad externalities, eventually even the best, which might be IT for good and everything, but if it's not useful, or if it's actually consuming more resources to run, like with the AI or 5G, then it might be worse than the previous solutions.

Chris Adams: Okay, thank you for explaining that. So it sounds like there's a number of different criteria or dimensions that you're looking at rather than just looking at carbon. And there may be different degrees to which they may be easy to act on or even measure the impact on for some of this stuff. But one thing that I was aware of and one thing that I think was high tied in the second part of this was actually trying to make these standards something a bit more easy, a bit more usable for people. And like we can talk about some of the specific things, the nuances around, say, carbon and water and stuff like that. But, Alex, I thought I might just ask you a little bit about this because this is the first time I've seen some of these standards where they're designed in such a way that they can be consumed by other digital tools so they can, some of this can be surfaced. Maybe you could talk about some of the process for trying to find ways to make this stuff easy for people to see because I know at the Green Web Foundation where we work, us having a really simple check has been a really good way to start a conversation over something that is incredibly nuanced when just talking about energy, for example, just like Anne mentioned. But, yeah, there are some really cool things about these draft guidelines that I haven't seen in a spec before, so maybe I'll just give you the floor and let you nerd out on some of this bit.

Alexander Dawson: Yeah, absolutely. One of the things obviously that we wanted to do after getting the draft specification out was to focus on trying to make this as easy for people to consume as possible because obviously we want people to start looking at the work and trying to start implementing these practices.

Obviously, as it stands, the specification is a really long document. So I think if you include all the stuff that can be expanded on, it's close to 300 pages in length. So quite a lengthy piece of work for people to scroll through. But...

Chris Adams: including the references, right?

Alexander Dawson: we've got supplementary documents which we are consistently working on to help people digest that material and to help other toolmakers and people who want to utilize it to actually take advantage of that work and to more easily consume it. For example, one of the first things that we produced alongside the very first draft was an at-a-glance document. And essentially it works as a functioning summary of the guidelines themselves.

So just giving a brief overview of exactly what is contained within the guidelines. And it's something that the Web Content Accessibility Guidelines also has. And within the second draft, we also released a quick reference guide. And the quick reference essentially boils down all of the guidelines into a easy-to-reference sheet which

Chris Adams: A TL:DR version, right?

Alexander Dawson: Yeah exactly, which people can just scroll through and get the gist of exactly what they need to do in order to complete and work through each of the individual guidelines.

Obviously there's links through to the full in-depth guidelines, if people want to get more detail onto them, but we're obviously trying to help people get through these guidelines as easily as possible. And there's a PDF checklist to go along with that, because obviously we know people like checklists and we know that it's helpful for taking to meetings, being able to highlight and work through it with clients and be able to mark off your progress and such. And we do have a JSON API as well, which is something that got launched with the second draft, which is really useful because it's a JSON file that contains the complete unabridged form of the guidelines in full, which can be queried against via our GitHub repo and people can remix it, utilize it, do what they want with it to be able to use or implement our guidelines, as they wish.

A few people have played about with it, showcasing some of their favorite guidelines within their website. We've got a copy of it, which is going to be used on the Sustainable Web Design website, which we're going to be relaunching soon. So there'll be the ability to filter the guidelines by tags, by specific disciplines, being able to search through individual guidelines, so that will actually make looking through the guidelines much easier, and for helping people to be able to work through the content as well.

Chris Adams: Okay, cool.

Alexander Dawson: And also we've got some additional stuff which is on earmarked in the GitHub repo, which we're going to be working on in the future as well. Things like we're going to be producing an introductory document, which will give a generic sort of overview to web sustainability and why it's important and guiding people into the guidelines and effectively working as a good solid standardized overview, and also a test suite which will provide implementability.

So it will provide some actually measurable impact reporting for how effective individual guidelines are and hopefully some guidance regarding tooling and user agents. Because hopefully we would like to see our work implemented in things like Google Lighthouse in the future as well. So a lot of

Chris Adams: linters things

Alexander Dawson: come.

Chris Adams: Okay, cool. Thank you for that, Alex. All right, we'll link to a few of those things for any people who are interested in this and might want to have a go at building some of that stuff. I'm just going to touch on the next question. So, we've heard the Web Content Accessibility Guidelines mentioned a few times, or, I don't know, WCAG, is that how you pronounce it? I can never remember. Yeah. So, these have been useful for helping people, basically, make space for accessibility, so making sure people allocate time and money to this stuff in a project, and they've been useful, uh, for courts, uh, to actually recognize the rights of people with disabilities to access services, for example.

Like, we've seen cases of government, uh, websites mandating this has to happen, and we've also seen pizza websites being sued for not having accessible websites, like the examples of Domino's and stuff like that. Maybe we could talk, Anne, if I start with you, then maybe we could hand over to Alex. Could you maybe talk a little bit about how much this might have influenced the creation of these guidelines?

Because this is, it's been cited quite a few times now, and this is one thing that I really, that really leapt at me when I was looking over the guidelines.

Anne Faubry: The WCAG is, is indeed a, a reference, um, it's crazy to see how in France it became a compulsory as a law. And currently there's also other companies being sued for not respecting it. So it's really moving forward finally, because for a long time it has been in place, but it was not acted upon. And there was just threats, but nothing was happening.

So nobody really cared, uh, regarding the companies. And, and so you can see how it can shape the online environment and if we can do the same with the sustainability, which is actually a broader, broader side of accessibility because as Alex said, everything is intertwined, if we can do the same, then it can be extremely powerful. It can go up as high as governments, but it can also give very practical guidelines for the designers, the developers, everybody working on digital services. So it was definitely an inspiration. The other thing, which is great with the WCAG is the fact that you can check box and that it's very powerful to see if you're making improvements and how you can compare to each other.

And the comparison is also a very strong drive for many organizations and people. And, and again, to see the W3C publishing these guidelines, it's really accelerated also what happened in France, in the government because they, they saw that they were being caught up worldwide.

Chris Adams: Okay, and I, Alex, I might hand over to you for some of this, because when you work with web accessibility, there's this phrase, POUR, perceivable, operable, understandable and robust. And we've done some work, we being Green Web Foundation, we did some work with Wagtail, a content management system, to work on something we were referring to as GOLD, which was Green, Open, Lean, and Distributed as a way of talking about some of these things you might want to design for in a kind of somewhat memorable fashion. Maybe you could talk a little bit about how, how you might make some of this memorable for people or easy to actually talk about because as Anne just mentioned, when you can make it really explicit when something is or is not, isn't being done, it becomes very easy to actually then specify as a product manager or someone along the way to say, "here's what I'm asking for. Can you please deliver this or deliver that?" And I think that might have influenced some of the reference to, actually we'll talk about a little bit later, about the global reporting, the GRI, because that's something that's referenced. So maybe we talk about POUR and some of those things there, and if there's anything like that you've seen in sustainable, in the world of sustainability, then we could talk a little bit about this other reference because this is something I haven't seen in the, in guidelines before, a reference to other global standards, for example.

Alexander Dawson: Yeah, certainly there are quite a few of guidance out there, as Anne mentioned, things like AFNOR and RGESN, which we've certainly tied ourselves and linked up to whenever possible. And we've hooked ourselves, as we're going to be mentioning later, to GRI. But as far as sustainability goes there aren't a lot of catchy acronyms or things which we can say to people that "you need to do this or do that."

So what we've tried to do is really focus primarily on providing people with things that can be implemented. Much of our work is obviously based on things which people are hopefully already doing, good accessibility practices, good performance practices in terms of web performance, things which are usually, in a good agency setting, things that web developers were already trying to be integrating with their practices, all we're trying to do is showcasing it from a sustainability angle and trying to reinforce it and showcasing how it can actually reduce emissions and where those additional performances can be measured and where the best practices could be additionally tweaked, as it were, so hopefully the work that we're doing can actually be utilized alongside the standards and best practices and actually be used to, um, the benefits with any additional legislation that's coming out, because obviously the EU is seeing legislation as we speak, things like, um, CSRD, which is going,

Chris Adams: what that means, because that's, so the Corporate Sustainability Reporting Directive is what the CSRD stands sorry, we're trying to avoid alphabet soup.

Alexander Dawson: yeah, sorry. Yeah. And obviously with this upcoming legislation, one of the things that they have mentioned is that they are actually going to be aligning this directive with the GRI standard as a measurement of implementation, effectively for people who are listening, if you want to conform with this piece of legislation, what the EU is saying is, "if you are abiding by the GRI standard, that is a great way of showing that you are complying with this legislation when it effectively comes into force."

And because our standard, as we are producing, is aligning

Chris Adams: referring an existing

Alexander Dawson: Yeah, exactly. Because what we have produced is aligning itself with GRI and we've got GRI measurement impacts with every single guideline, the great thing is that if people can make those measurable improvements, they will have effectively our weighted evidence that they can use in their reporting to show that they are complying with this upcoming legislation, which hopefully will be a sign to show that they are conforming with sustainability regulation as it is coming into force, which is another hopeful, beneficial move to show other governments that potentially they should be moving in this direction as well.

Chris Adams: Okay, and, Anne, I might just ask for... GRI, we haven't defined what GRI means, but we've said it lots and lots of times. GRI, it's the Global Reporting Index. Is that the,

is? Yeah, Initiative, Global Reporting Index, which is a, maybe you could just briefly touch on some of that before we move to another questions about, and we need to be quite short about this one on the W3C because there are recommendations, there's standards, and there's different things, and we might be useful to talk about that. But in the meantime, yeah, maybe you could just briefly, for the uninitiated, explain the role of the GRI and what some of this actually is, because it comes up a lot again and again and again, just as Alex says.

Anne Faubry: Well, actually, if you allow me, I'd rather have Alex answer this because don't use the GRI much.

I didn't know actually of the GRI before joining the W3C, so I don't think we use it that much in France.

Chris Adams: Alright, quickly then, we'll stay on the GRI, because there's another, I didn't mean this to be such a kind of TLA, sorry, three letter acronym soup of an episode, but we are going to talk a little bit about, okay, you've got these guidelines, what happens next? So Alex, if I ask you briefly what the GRI is

then we'll just touch on what happens from here to turning this into like standards or what that process looks like.

Alexander Dawson: Yeah, absolutely. The best way that I can describe it for listeners in a really simple, easy way, the really simple crash course is that effectively you have best practices, you have standards, and then you have regulation. Essentially what we are producing with the Web Sustainability Guidelines is what's known as a best practice.

So it's something which helps you conform with any standards or any regulation because it's effectively out there but it's effectively not endorsed by a regulatory body, but it's something which is bound by evidence and we try to link it in with what standards already exist. You also have standards.

Now these are things like the GRI and you also have the International Standards Organization, otherwise known as ISO. And the GRI have produced a standard and GRI, the Global Reporting Initiative, it's been around for about 20 odd years and it's used by businesses and corporations and later bodies everywhere.

And it's free for anybody to use and it essentially lays down a whole set of different guidelines for things like mineral use and for water usage and for all the things that you would measure under ESG. So that's essentially set down in stone because they're a standards body like the World Wide Web Consortium.

And then above that you obviously have regulation because they're regulated by governments, and they are the people who effectively lay down the laws. So, essentially, that's the crash course method. You've got the governments, the standard bodies, and then you've got people like us who are producing materials to try and comply with those different things.

Chris Adams: Okay. All right. Thanks for spelling that out, I suppose. All right. So we spoke a little bit about the fact that, okay, you have these different kind of tracks and there is a way that you might go from here. Maybe we could just briefly touch on the fact that you've, there's now a second draft of these guidelines that are, that are out for people. Um, what's next on the roadmap? Or maybe I should ask, like, is there a roadmap? What happens over the next six to 12 months now that these are out here, for example? Anne, if I put that question to you, maybe you could talk a little bit about that, and then we'll touch on some of the other questions.

Anne Faubry: So people are actually asking lots of questions, uh, suggesting improvements, debating some of the facts, um, and sometimes contributing with new points of view, uh, and submitting issues on GitHub. And Alexander and Tim are doing an amazing job trying to integrate all of this feedback into the second and then now the currently the third draft of these guidelines.

Chris Adams: Okay, so we got that. So basically, if you read through these, and we'll be linking to these, you're able to run through... Basically, the idea is, go to GitHub, ask some of the questions there, possibly make a pull request or something like that, but that's the main way for collecting people's feedback right now.

So that's the case. Yeah?

Alexander Dawson: A pull request, add an issue to the issues system, and obviously it will be reviewed and integrated into the next draft and yeah, everyone's welcome also to join the community group and come to our meetings because we hold them every month and we discuss current progress.

Chris Adams: Okay, so if you can't use GitHub and you're or maybe you don't want to use GitHub, there are other channels you can either use and I think that there's maybe even something as old as old as low tech as a mailing list. So there's like a mailing list and video and there's also calls people can join to go. So there's a more than one channel by which you might share some of your feedback into the formation of some of these guidelines. Okay, alright, that's really helpful for people to understand. We're just coming to the end of the time, so I just wanted to give a bit of time for you two to talk about, now that we've just been speaking all about these guidelines and them being out there, and how they can be consumed and used, and how they are made, are there any other open projects or communities that, that outside of this, you think are really exciting and interesting?

So Anne, what kind of projects are inspiring you right now that might have helped you shape your thinking for this work here, but also that you spend your time thinking about or wanting to contribute to outside of this as well?

Anne Faubry: There's a great tool I keep using, but it's not even new, actually. It was done by Artefact, which is, I think, a company from England. And they created the Tarot Cards of Tech, a list of questions you can use. I use it in my projects, I use it with my students, I use it with my colleagues to ponder your choices, think long term, think strategy, and think externalities you might have, good or bad. I love the questions. There's actually two of my favorites, which are, "what would you do differently if nature was your client?" So you can think really differently about your service, thinking "maybe I could reduce this footprint or add this feature to raise awareness on this issue with the users." And another one I love is "what happens if 100 million people are using your product?" And then you can also think what externalities you might have on jobs, on social habits, on housing, and on many different topics to rethink more systematically and, and to think about more/other sustainable issues besides environmental ones.

Chris Adams: Okay, cool. Thank you for that, Anne. And Alex, I'll give a bit of space for you before we wrap up on this one. So, what projects are catching your eye or that you're really interested in right now that might have helped inform some of the work that you've been doing recently?

Alexander Dawson: I think the main one that really informed me most lately is CO2.js.

Chris Adams: Fershad will be happy, yeah.

Alexander Dawson: Yeah, I'm working on measurement because that really inspired me on when I was doing some of my own research to investigate more on how carbon affects various different aspects of the web ecosystem.

Chris Adams: I assure you that wasn't planned ahead, alright? Um, for reference folks, CO2.js is one of the projects that, uh, the Green Web Foundation works on and maintains that's used in a number of tools, uh, for this. I'm a little bit embarrassed, so I'm gonna move on quite quickly, uh, to the final question. Basically, we, if people have found this interesting or finding out about the fact that there are, there is guidance from the W3C trying to do this kind of work, and they want to find out about what to do and what's happening with this, could you just tell people where they should look or where they go?

If there is a particular website you would, you could possibly pronounce and spell out for people so they, if they're listening, they know what to Google.

Alexander Dawson: Yeah, if they just Google the Sustainable Web Design Community group, they will find our homepage and there's a great link to the wiki there which has all the information about all of our past meetings, frequently asked questions and links to all the social channels and how they can get involved. So that's a great place where people can start getting involved in the project.

Chris Adams: Cool, thank you, Alex. I've just checked while you're describing that. I've also, if you type in the word Sustyweb, it's a pretty good Google hack. So, there is nothing related to Sustyweb so far, other than this stuff. Okay, Anne, I'm going to hand over to you and then we'll get the final word with Alex. If people are curious about some of your work and if you have any kind of professional presence and you're comfortable sharing that, is there a particular place you would send people to or if they've found what you were talking about as interesting, where should they look?

Anne Faubry: Our main English resource, uh, is the guidelines that we produced. It's called Eco-design. I will share the exact link so people will know where to look. And otherwise I'd be happy to talk with anyone on, on LinkedIn. That's the best place to find me.

Chris Adams: Okay, so Anne Faubry on LinkedIn or, or Eco-design. Okay, and, Alex, yourself? If people want to find you online, what's the best way to find you in a professional context?

Alexander Dawson: With me it's relatively easy, they just go to alexanderdawson.com

Chris Adams: All right, that was easy. Okay, this has been fun. I've learned a lot. And this has been a nice conversation. Thank you so much for taking the time to actually run through this with me and tell me all about how this works and hopefully this should be interesting for other people who are curious about how standards are formed and how the W3C might be having some kind of impact in how people think about sustainable web. All right, thank you folks! Hey everyone, thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: Mapping Green Software on the Grid

2 november 2023 | 33 min

TWiGS host Chris Adams is joined by special guest Tony van Swet from Electricity Maps, to talk about the mapping of the carbon intensity of electricity grid. Tony shares some of the work that Electricity Maps has been doing to make it easier to understand how clean or dirty electricity is around the world, as well as how they’re making this data more accessible and usable to consumers. Join in on this candid conversation discussing the uses of such data and how to access it, as well as Tony talking about carbon intensity, open data, and open source.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Tony van Swet: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

How to trace back the origin of electricity (Smoothie Blog Post) | Electricity Maps [06:16]
The value of space-time load-shifting flexibility for 24/7 carbon-free electricity procurement | Zenodo (TU Berlin’s Study with Google, using PYPSA) [12:11]
Electricity Maps | Client Story: Monta (EV Smart Charging use case) [15:41]
GitHub - electricitymaps/electricitymaps-contrib: A real-time visualisation of the CO2 emissions of electricity consumption [21:01]
Electricity Maps | Reports - Hourly Residual Mix Methodology [27:13]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Tony van Swet: Looking at Google's use case at their data centers, they have the huge potential to shift their computation based on time or location, so this enables them to manipulate their energy consumption through using our API to increase their consumption when the sun shines and the wind blows.

Chris Adams: Hello, and welcome to Environment Variables, brought to you by the Green Software Foundation. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host, Chris Adams. Hello, and welcome to another episode of This Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. When we talk about green software, it's often common to talk about energy efficiency, and one of the reasons we care about it at all, is that right now we burn a lot of fossil fuels to generate electricity used in data centers, networks, and end-user devices. But how much of that comes from fossil fuels? And is that changing? This data exists all around the world, and sometimes the data is open, but it's often very messy. In 2017, the Electricity Map project was launched to make it easier to understand how clean or dirty electricity was all around the world. And as the name suggests, it took the form of a map showing the carbon intensity of electricity in as many places around the world as possible. Over the subsequent years, an open source project has grown with hundreds of developers around the world, contributing open web scrapers for data in their parts of the world to make the data more accessible. And earlier this year, the company behind the project released a new open data portal for historical data about how clean electricity was for anyone to use how they wish. So, what does this have to do with green software? Having access to this kind of data makes it much easier to understand the carbon footprint of your software. And this week, we're joined by Tony van Swet from Electricity Maps to talk about carbon intensity, open data, and open source. Hey there, Tony.

Tony van Swet: Hi, great to be here.

Chris Adams: Okay, Tony, before I get ahead of myself, I think we should give you a bit of space to introduce yourself properly. So can you tell us a little bit about what you do at Electricity Maps? And for folks new to the field, what Electricity Maps does these days, please?

Tony van Swet: Yeah, of course. I'm a senior software engineer in the advocacy team at Electricity Maps, and I'll give you a bit of background on what we do at Electricity Maps. So our mission is to organize the world's electricity data to drive the transition to a truly decarbonized electricity system. And as part of the advocacy team, my focus is enabling climate action with transparent insights.

We do this with the help of the open source community, building products such as our map visualization and the data portal that we're here to talk about today.

Chris Adams: Cool. Thanks for that, Tony. Okay. So if you're new to this podcast, um, my name is Chris Adams, as I mentioned before. Um, I work as the executive director at the Green Web Foundation, a Dutch nonprofit focused on an entirely fossil-free internet. And I also work as the chair of the policy working group inside the Green Software Foundation. And before we dive in, here's just a quick reminder, everything we talk about, we'll link to in the show notes below. So if there's a paper that caught your interest, or there's a story you've heard about, we'll do everything we can to make sure there's a helpful set of links that you can follow up for your own research a little bit later. But back to Tony. Tony, I've got to have to ask you, I know you're working in Denmark, but... I suspect you might not be coming from Denmark in the first place. What does a Kiwi end up doing on the opposite side of the world in Denmark, working for a company like Electricity Maps? I'm sure there's a story behind that.

Tony van Swet: Yeah, absolutely. It's definitely a bit of a career shift for me. So I started out about 10 years ago as a truck driver in New Zealand. I was full of self doubt, a bit depressed, struggling to find my place in the world. And to lift myself up out of this, I made it my mission to create technology to combat climate change, and I identified that software was the most powerful way to effect change at scale.

And this led me to enroll in a computer science degree. From there, I worked at a few cool startups in New Zealand, eventually looking to integrate electricity maps data when I saw their job postings and applied, and within a few months, I had the job and was waving goodbye to my friends and family to fly across the world to Denmark.

It's definitely been full of challenges, but it's been amazing to find a company that really shares my values and aligns so perfectly with my mission.

Chris Adams: Wow. So you, when you say you're a truck driver, you're talking about the massive, like 18 wheelers crossing from city to city, right? Something like that.

Tony van Swet: Yeah, I actually worked with the HIAB trucks, which have a crane on the back. So I was delivering building supplies around Auckland. It definitely gave me a lot of time to think about the world and take in the kind of sights and sounds of the city.

Chris Adams: Wow. Okay. So I think you may be the first former truck driver we've ever had onto this podcast. So yeah. Wow. Thank you for, thank you for coming along. That's also a fun story. I, it's, it's quite nice to hear something like that because, uh, I myself, there's a lot of us who are self taught technologists and to hear a nice story about switching careers you're going, "that's cool, actually." All right, before we digress, let's go back to what we were supposed to be here talking about, which is open data and carbon Intensity. So one thing you mentioned is that we're here to talk about open data and there's some recent work at your end that's made that visible. But before we do that, could we briefly just cover what carbon intensity means at your end, because this is something that isn't obvious to most people.

And I remember there being a kind of nice introduction on your website using metaphors like blenders and so on to explain that there's more to electricity to it being just gray versus green, for example. So maybe you could just. provide a bit of a background or how you explain this to people, then we can dive into some of the details about open data.

Tony van Swet: Yeah, the blender analogy is really great. We even did a smoothie maps version of our app for April fools, renaming all of the power sources to different fruits and vegetables to illustrate that. So yeah, carbon intensity to us seems like it's relatively straightforward, but if you're not familiar with this idea, it's quite hard to understand.

And in this case, we refer to carbon intensity as the CO2 equivalent for a given zone where energy is being consumed. We calculate this by determining the carbon intensity for each generation type and then weigh it according to its proportion of the grid mix. We also then calculate the neighboring zones and account for all the imports and exports of the connected zones to figure out a final number for the carbon intensity where you plug into the wall and consume it.

Chris Adams: So basically, if I understand that correctly, you're, what you're saying is you look at all the various parts of the world, and when you say zone, you're referring to maybe a country or a part of a country, depending on how a grid is designed. And then when you're talking about the kinds of generation, you're talking about, say a coal fired power plant or a gas fired power plant or a solar farm or something like this. So these have different levels of CO2 that get emitted for each unit of electricity and you're mixing those together, something like that. Is that correct?

Tony van Swet: Yeah, definitely. When we take a look at a coal plant, it's going to emit a lot more carbon than the equivalent solar or wind farm.

Chris Adams: Okay, cool. So that talks about the consumption, the how, where the electricity comes from. So maybe we can talk a little bit about, okay, how we experienced that and how, like, when I plug something into the wall, for example, what happens next?

Tony van Swet: Yeah, so when we, um, plug into the wall, the energy we consume is, um, considered a mix of all the generation types of the grid you're connected to, um, and it's almost impossible to determine whether an electron comes from a wind farm or a coal plant, even though this will have a significant change in the carbon intensity of the energy you consume.

So this is where it's really useful to consider the grid as a giant blender, mixing together all those generation types. And then we can evaluate the true carbon intensity of the energy that you consume.

Chris Adams: Okay. If we're going to continue this blender analogy, if you put lots and lots of, say, strawberries in a blender, it's going to look one color. And if you had lots of Kiwi fruit in a blender, it's going to look another color. So that's a little bit like what you expose to people and how that might change over the

day.

Right. Okay, cool. I believe what we'll do is we'll share a link to the blog post, because

I found it one of the clearest ways to actually help people get their head around this kind of concept, because it is a bit of a leap when you're first starting to get into this field. So with that, we've got a kind of grounding there. Maybe it's worth talking about this from the point of view of a software engineer. So. Let's say you do know this and you have access to this information. Why is this helpful if you're a software engineer? Like where does this fit into what you might do, for example, or affect your job?

Tony van Swet: I think it's super useful as a software engineer to, to have this information and I see a few main categories where you can apply this data, particularly around raising awareness of when to consume energy. We want people to use power when the sun shines and the wind blows. So I think that there are ways to present this information so people can make decisions in their everyday lives.

But particularly for me, I find it interesting of automating solutions where we can get carbon-aware products that will shift their consumption or the load of the power consumed based on how sustainable the power is available to them.

Chris Adams: Okay. So in this case, this scenario here, you're basically saying, if you have an abundance of power, which is very green, you might kind of tune or change your usage to use more of that, and when the power is particularly dirty, for example, you would try to use less of it so that you're shifting your power through time or possibly through space so that the average carbon intensity might be lower than it otherwise would be.

That's what I think you're saying, right?

Tony van Swet: Yeah, exactly. So the two main ways to optimize your consumption here is over time or via location. Um, so we know that different grids are much cleaner and, um, some people have the luxury to be able to shift their consumption via location as well.

Chris Adams: Okay, cool. So. We've got the kind of general concept for this. Are there any kind of favorite examples that you might point people to of people using this to actually demonstrate their behavior, either at a personal level or an organizational level? Because yeah, having a concrete example would be really helpful for people who are listening to this for the first time.

Tony van Swet: Yeah, I think, um, my favorite example is, um, looking at Google's use case at their data centers. They have the huge potential to shift their computation based on time or location. So this enables them to manipulate their energy consumption through using our API to increase their consumption when the sun shines and the wind blows.

Chris Adams: Okay, so if I understand it correctly, they're like a client of yours or a customer of yours, they pay for this, and then they use it then to essentially either scale things up or down, depending on the amount of power they might be using, depending on where the data centers are. That's, that's what it sounds like, what you're suggesting there, correct?

Tony van Swet: Yeah, definitely. So there is the location aspect and we see a huge variation of the carbon intensity throughout the day. So they also do time-based or scheduled computation based on the carbon intensity available to them.

Chris Adams: Cool. Okay. I'm glad you mentioned this because this is something we've had people come on the show before to talk about some of this, but since we have spoken about this, there's actually, uh, some interesting data. Uh, there was a study published with TU Berlin where we're, I'm based in Berlin so we've, I found out about this study and, uh, there's. I found this actually quite a nice example of this to talk about, because a lot of the time, when you see companies talking about this, it's quite hard to actually find meaningful numbers to say, does this actually translate to a saving in carbon? Or does it translate to a saving in even money, for example? And this is the first time I've seen with really detailed information, which has been modeled through this. Um, we'll share a link to this paper, but there's a few kind of headlines that I saw from this. And as I understand it, one thing that Google is doing, for example, they've basically set a commitment to say, "we want to have the average carbon intensity of our power to be this much." So we want to have a certain percentage coming from what they call is like carbon-free or fossil-free sources of generation. And, uh, the study that I saw basically showed that by moving the load around, it reduces the amount of renewable energy, renewable kind of generation that needs to be deployed in the first place for this.

So there's an embodied carbon saving in the, in not needing to have a bunch of wind turbines or solar all around the globe. And this study that I see, it was modeling five data centers. So five out of say 14 data centers that are around there. And there were. The savings are pretty good, or actually like measurable.

I think with the combination of moving things through time and moving things through space, so moving a compute load to where it was going to be greener, the figures that I saw, some of the headlines were that they're able to reduce the cost of doing this by something in the region of a third of the amount of investment that would need to be possible. And, uh, they also, this is one of the first examples I've seen, which even explains like what the costs on a yearly basis might be for this. And, uh, I think the. There was a couple of scenarios inside this. So there's maybe with zero load shifting or moving, say, about 40 percent of the compute loads that to to different parts of the data centers, where maybe one part of the world might be particularly windy or sunny. When I look at the figures here, I see something in the region of, if you, the savings that are here and we need to, and I will share a link to this, to the actual study for this, so that people can look into this a bit more detail, but with the five data centers modeled in, I think, Germany, in Denmark, in Portugal, in Ireland, and in Finland they were basically able to model savings of around at least 200 million US dollars each year by, in terms of the amount of power that you would need to be, the amount of like generation you would need to match this, to actually hit those targets. Now this is, I think this is actually useful to understand because this actually speaks to the fact that there's economic drivers as well as actually just environmental drivers for this. And this kind of speaks to the wider kind of trend, but. I think it's useful to, for this to be, people to be aware that there's actually something in the public domain to interrogate and look at some of these numbers and see how some of these are modeled and what some of the assumptions are. So we spoke about that. Are there any other use cases that you might point to that may be a little bit more closer to home, for example, or something that you might, that people might experience on a more kind of daily basis or close to themselves, for example?

Tony van Swet: Yeah, absolutely. Yeah. We have a few customers in the EV smart charging space, and we have also done some research with the Frederiksburg commune here in Denmark about the benefits of smart charging. And we... We were quite impressed to see a 10 to 15 percent reduction in carbon emissions if we have grid-aware smart charging products.

So this is plugging your car in the evening and letting it decide when the best time is to charge the car overnight. And even with a small shift in that load, we see a significant reduction in the carbon emissions of the energy consumed. So we were really positive with the results of that. And particularly find it a very nice use case that you put the decision-making power in the hands of the consumers here.

So people can choose whether they want to use these products or not.

Chris Adams: Okay, cool. All right. If you're in the UK, I believe there's a number of companies that do things like this. Octopus is one of the better known examples of this. And I think under some of the tariffs, there are scenarios where you can actually be paid to charge up a car rather than pay to charge a car or to use a car.

So the cost can go negative. Because there's maybe an abundance of power in the grid or like we have here. So that's actually, okay. That's quite useful. So we've covered a couple of use cases now. Maybe it's worth talking a little bit about, little bit about what kind of software supports the use of this data. So I know that at the Green Software Foundation, there's a carbon aware SDK, which is designed to allow people to embed this in some of their software. And where I work at my nonprofit, the Green Web Foundation, we have a library, a Golang library, which is used in a project called Carmado, which is a kind of federated Kubernetes operator. Could you talk a little bit about some of your experiences of what you've seen people use for some of this stuff? For example, maybe you could talk a little bit about some of the pieces of software that you've seen in the wild using some of these tools or using some of this data, for example.

Tony van Swet: Yeah, definitely. Firstly, yeah, we're hugely appreciative of the Green Software Foundation and their work to make it easier for developers to use data like this. We do our best to enable developers and hobbyists with our free data through our API. Previously, it was known as CO2 Signal and we've now incorporated that into the Electricity Maps API.

And we see lots of amazing tools being built. We see people building dashboards so they can make decisions around which data centers they use. And we do see a big community from Home Assistant also integrating our data. So people can connect their smart homes to become carbon aware and give information on the carbon intensity of their homes.

Chris Adams: All right. So we've got some, some stuff like that. And I think we've done a decent job of now establishing what carbon intensity is and how some people might be using it so far. And, uh, we spoke about this idea. There's a, like a free tier, which basically implies that people pay for a data service. But one of the things that we're here to talk about today is open data and this open data portal.

And as I understand it, this is your baby, so to speak, right? So maybe you could talk a little bit about, okay. What is this that we, that that's actually gone live because I've got a history with open data, but I suspect it'd be useful for people who are coming to this to understand what this data portal is and why it's useful and what it lets people do, for example.

Tony van Swet: Yeah, I was super excited to take the lead on the Data Portal project and really happy to come on the show today to talk about it. Providing free and open data really motivates me. And the Data Portal is a product on our website where anyone can download free carbon intensity data for over 50 countries in hourly, daily, monthly, and yearly for both 2021 and 2022.

Chris Adams: Okay. So let me just check if I understand that. So, uh, if people want to start using or experimenting with this data, there's a free tier which you, which folks like yourselves provide. Uh, there's another provider called Watttime that does a, of a free, a free tier. And there's commercial kind of real time feeds from both yourself.

And, uh, this part here is this high resolution historical data that has typically been quite hard for people to give access to. And this is openly licensed in the sense that people are free to use this how, however they wish, is that the case or is there any, or maybe we could talk a little bit about the licensing part so people understand how they could use some of this.

Tony van Swet: Yes, so we have provided the data free for anyone to use. We particularly look at Carbon Accountants and researchers to use the data. People are welcome to use it under our license, as long as they, if they're building a new product with our data, then they'll be required to open source that new product, but if you're using this data for Carbon Accountant, then you're fine to use it and charge for that accordingly.

Chris Adams: Okay, cool. All right. Uh, what we'll do is we'll share a link to the message, to, to the licensing. So people have an understanding for this. So I think when I looked at it was the open database license. So you're able to use it for free in any, in any form, as long as, uh, you're prepared to share under similar terms yourself.

That's basically the kind of general approach that I understand for that. And you, you spoke a little bit about there's an intended audience of people who might be carbon accountants or researchers or energy geeks. Can you talk a little bit about how this data gets published in the first place, where it comes from? Because as I understand it, the data can be quite messy to actually put into a kind of API for someone to consume.

Tony van Swet: Yeah, yeah. It's, it's a huge challenge to collect all the data. So we have an open source repository full of parsers that collect this data from TSOs and data providers around the world. We have an incredible open source community that helps us to maintain those parsers. We then process this raw data with the kind of smoothie idea that we talked about earlier, run data quality checks on top of the data, and then present it in a way that's easy to navigate and consume.

Chris Adams: Okay. So you've used a bit of jargon that I'll need to unpack on there. So you said that you're getting data from a few places and you mentioned a TSO. I'm assuming a TSO is a transmission service operator, like someone who operates part of the grid and they publish information. So that's where some of the data might be coming from.

Is that correct?

Tony van Swet: Yeah. Yeah. Spot on.

Chris Adams: Okay. And one of the challenges is that not every, so maybe I, as I understand it, when I've looked at this data, the data comes out in like grams per kilowatt-hour, what I would typically be paying for, but different places might have different ways of reporting it or different units. Is that the kind of stuff that you, that ends up having to be munged so that there's a kind of clean interface for people to consume?

Tony van Swet: Yeah. So the data providers, the TSOs tend to give the data in the format of a energy breakdown. So the various production types, whether it's wind, solar, coal, gas, and we then process this data and apply emission factors. So we add a direct and life cycle emission factors to each of the generation types, and then compute that to give a final carbon intensity number for each zone.

Chris Adams: Okay. All right. So if I understand that correctly, you're basically saying we know what this kind of coal power station is likely to be doing for each unit of coal. And because we might have some information about it being an old machine, old one or a younger power station. So you'll have some figures like that, and you essentially run through every single form of generation so that you've got a kind of up-to-date, accurate number for that based on what, what people are doing rather than have to look that up because yeah, it's quite hard to find.

So. You've, you've created this data portal. People are able to download it for a set of countries or different parts of the world, and you said that there's data for 2021 and 2022, and this kind of begs the question, what happens next? Is this, is the idea, is the intention to keep having this available on an, on a, on a annual basis so that next year there'll be data for 2023, for example?

Tony van Swet: Yeah, absolutely. I listened to your podcast a few weeks ago and I heard you mention that we were looking to raise the bar of energy data available out there. And I really like that term. It's exactly what we want to do. We plan to release new data early 2023. We want to enable carbon accountants to do granular carbon accounting based on our data.

And we really hope that providing this data for free gives the industry a push to be more open and transparent around what energy data is available.

Chris Adams: Okay, cool. All right. So for the energy nerds here, I, it might be worth just briefly talking about the fact that this currently provides average carbon intensity data. Is that correct? So that's basically the kind of location-based figure. So there, this isn't trying to take into account water or anything to do with market-based figures at present.

That's something that might be on the horizon in future. Could you maybe talk a little bit about what things are on the wishlist or what people are asking about What would they like to use in future from here? Because you alluded to some things about, uh, the life cycle intensity of, of, of energy, for example, and there's a whole other set of footprint impacts that people often ask about when they talk about carbon intensity, or even just the environmental impact of the use of electricity in any kind of service.

Tony van Swet: Yeah, absolutely. So carbon accountants are most interested in the direct emissions that we provide in this data because they're doing their accounting based on the Scope 2 emissions of a company. Um, we do also provide the life cycle analysis emissions for each zone as well. And this is taking a cradle to grave approach of the emissions.

We use the numbers from the IPCC and the,

Chris Adams: So IPCC in this case is the Intergovernmental Panel on Climate Change.

So that's one thing. And then the UNEC, so I'm guessing it's United Nations.

Tony van Swet: United Nations Economic Commission for Europe.

Chris Adams: Okay, great. Okay. So, so that's basically the kind of bona fide place where you're taking some of these numbers from. And when you talk about the life cycle emissions there, that means that let's say you're talking about solar or wind, for example. That includes the fact that someone has to make the panels in the first place, and there's going to be some pollution that may be caused there, carbon pollution from making the kind of silicon panels or the turbines.

Is that correct? And then the dispose disposal.

Tony van Swet: Yeah, exactly. Yeah. And even in the case of nuclear, the lifecycle analysis takes into account the storage and disposal of nuclear waste over hundreds of years and applies the costs of that to a carbon equivalent.

Chris Adams: Okay, cool. All right. So we've spoken about carbon and we will talk about carbon a bit more, but. It's very, one thing that has come up when people talk about the environmental impact of digital services, there's this term called carbon tunnel vision, where people only look at the one figure, or the one kind of dimension. Is this actually something that, is this on your wishlist, for example, because we know there's a, say, there's a water impact. People talk a lot about machine learning and AI and tools like that, having a water impact, and there's also an impact from the actual generation, for example. Could you maybe talk about a little bit like that?

Is that something that you'd like to be heading towards, or is that on the roadmap, for example, in the, in the long run?

Tony van Swet: I think we'd love to take a step back and, and have a broader look at the impacts. We're relatively limited with our capacity, so, so we do focus on what we know and what we're experts in. But I would love to see us work with partners to be able to provide our data alongside other sources to take a bigger picture approach to this.

Chris Adams: Okay, cool, Tony. So back to carbon then. You spoke a little bit about working with other providers and I realized just as I was doing some research for this recording, this podcast, there was a new paper that came out from Electricity Maps specifically about, this is a, this is a really nerdy, I'm afraid, residual carbon emissions when you look at the environmental impact of electricity and If I understand it, I'm just going to try and run my understanding by you if I can, and then if you can give me an idea about if it's in the right direction, that'd be really helpful.

So, as it stands, electricity maps gives you figures for location-based data. So you look at the carbon intensity of the generation all around the world through, and like dams or wind or solar, uh, you'll look at that part, green energy, they often talk about, say, using green energy in some parts of the world where they've purchased essentially certificates to count electricity as green.

And this is a kind of like a market-based approach that people use. And this is the basis that various companies use to say, "we're using a hundred percent green energy," for example. Now, if I understand it, this paper that goes into this and basically says, if you're going to look at the carbon intensity of electricity, you need to account where these certificates that people use, where they're actually being used in various parts, because that's going to have an impact if, because if someone is claiming green energy in Ireland, for example, and they're claiming that on the basis of certificates that were traded from Norway, that's going to have an impact on how green the power might look in Norway compared to Ireland, for example. This is what I think some of the research is that was in this paper. Is that directionally correct? Is that moving in the general direction of correctness for this stuff?

Tony van Swet: Yeah, absolutely. Yes. I think a lot of companies are buying renewable energy certificates and it has to be a zero-sum game. So the residual mixed paper that our policy team has just released goes into a huge amount of detail into how you calculate the carbon intensity after you have sold those renewable energy certificates for each zone.

Chris Adams: And as I understand it, this is something that's on the roadmap that will be looked at is A, a thing that is unsolved right now, but people are looking to figure out how to incorporate into how they work. I know that in the Green Software Foundation, there's a group called Realtime Cloud, who are working to come up with hourly figures to make it possible to provide this kind of reporting. This seems to be one thing that comes up because when I was looking through this paper just last night, actually, there was a few things which are really eye opening for me. So Ireland and Germany are two large markets in Europe, for example. And as I understand it, something like eight times the certificates were consumed as were issued in Ireland, for example.

So this basically means as I understand it, that eight times as much green energy is being claimed as is generated in Ireland. So therefore you've got a bunch of generation in somewhere else in the world that needs to be accounted for when you look at the carbon intensity of say, a place like Iceland or Norway, for example, but the same things seem to be in Germany as well. Germany has something like seven times the certificates consumed as were issued in Germany. So that suggests to me that seven times as much green energy has been claimed as is being generated. So if Germany had to have an entirely green grid, you would need something like a sevenfold increase in order for them to be saying, "yes, we're running entirely on green energy." That seems, this is pretty eyeopening. I'm really glad this is actually out there because I haven't seen this data provided in this resolution, particularly in hourly level before.

Tony van Swet: Yeah, I think it's really fascinating and definitely highlights why we need the transparency around this market based approach. And it's very early days, so we are hoping to inform the methodology of how we approach this in the future.

Chris Adams: Great. Okay. So what we'll do, we'll share a link to that. The things I've just shared are in the first 10 pages of this paper. It's about 80 to 90 pages long, and it's a really impressive tour de force. So Cyril, I'm really impressed with this work. Really mad props for you to actually get this together. Cyril is the policy lead, Electricity Maps, and he's also on some of these working groups, which is why it really caught my eye. So Tony, we've just spoken a little bit about Open data, different ways of measuring the carbon intensity of electricity here for informing your decisions as a software developer. Is there anything that you would like to draw people's attention to? Any projects or things that you are particularly interested or that you're excited about right now?

Tony van Swet: Yeah, absolutely. Yeah. First off, I'd welcome people to jump on and take a look at the data portal and I would appreciate any feedback around that. And. If anyone would like to contribute to our open-source work, we're also always looking for contributors there. To find out more, jump on our website at electricitymaps.com.

Chris Adams: And if I understand it correctly, you folks are still, it's still mostly Python scrapers and a kind of React app that you had before. Is that still the case for people?

Tony van Swet: Yeah, definitely. Yeah. Python and JavaScript.

Chris Adams: Okay. So common languages that people know their way around. Okay. Brilliant. I think that's pretty much it. I've, I've really enjoyed this, actually. Thank you so much for giving us your time and diving into some of the finer points of carbon intensity of electricity and, uh, some of the nerdery there. And, uh, Tony, thank you so much.

I've really, I've really enjoyed this. Cheers, mate.

Tony van Swet: Yeah. Thank you, Chris. It was really great to be here. I also wanted to say I went digging through our open source repo and found your name on there. So I want to give you a personal thank you for contributing in the past.

Chris Adams: Thank you very much. Um, there, I think there'll be more PRs coming in future when I find the time. Okay. Cheers, Tony.

Tony van Swet: Amazing. Thank you.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

And please, do leave a rating and review if you like what we're doing, it helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser.

Thanks again, and see you in the next episode!

The Week in Green Software: New Research Horizons

26 oktober 2023 | 43 min

This Week in Green Software has Dr. Daniel Schien from the University of Bristol, UK, joining host Chris Adams to talk about old research, recent news, and future prospects all revolving around digital sustainability. This conversation touches on some of the work Daniel has done in the past (and plans to do in the future) as well as their thoughts and reckons on this and how it can be used to steer our efforts towards a sustainable future. Together, they cover topics such as streaming being the new flying, and some ways in which new research has changed their perspectives on some problems in green tech.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Daniel Schien: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Daniel Schien: If we are saying streaming is bad and instead consumers jump into their car and go to the cinema, that would be the worst thing that people could do. And same if we are saying streaming is bad and consumers swap to terrestrial broadcast, makes hardly a difference in the short term to the energy consumption that is associated to the delivery of the service.

Hey, Dan.

Daniel Schien: Hi Chris.

Chris Adams: Dan, if any of our listeners lurk on the Green Software Foundation forums, or the kind of IETF E-Impact mailing list, they might be familiar with you. But for folks who are new, would you mind taking a second to introduce yourself to the pod for our listeners, please?

Daniel Schien: Sure. Thanks for having me. My name is Daniel Schien, and I, as you said, teach and research at the University of Bristol in the United Kingdom. I'm a computer scientist and work here in the Faculty of Engineering, researching questions of sustainability of ICT. And here in Bristol, I'm part of a small but dedicated group of researchers looking at the environmental impact of ICT.

Chris Adams: Okay. And I did a bit of research before this call and I realized the first paper I saw from you was back in 2011, researching the dynamic carbon footprint of digital services back in 2011. So it sounds like you've been doing this a while and you've worked with a number of organizations in the industry as well.

Is that correct?

Daniel Schien: Yes, that's true. Indeed, I joined the University of Bristol at the end of 2010, working on a project for the Guardian newspaper, actually. And at that time, they were interested in understanding the transformation of their business from a paper-based delivery to a digital delivery, reading the news online. And at that point there were no methods, no existing similar analyses of environmental impact from the distribution starting at data centers over networks to user devices. And in this void, we helped to come up with an assessment that can compare the quite well understood print side of delivering news with the digital version.

Chris Adams: Wow. Okay. And you've worked with a few other organizations as well. And I think we're going to touch on that a little bit later on. So wow. Thank you. The little story about the Guardian. I had no idea that the Grawn was doing that stuff so long ago already, actually. Okay. Um, cheers for that, Dan. If you're new to this podcast folks, as mentioned before, my name is Chris Adams.

I work as the chair of the policy working group at the Green Software Foundation. And I'm also the executive director of the Green Web Foundation, a Dutch nonprofit working towards an entirely fossil-free internet by 2030. I'm also going to share that whenever we mention a story or a project or a paper in this podcast, we'll do everything we can to find a link so you can actually research some of this yourselves and learn somewhat more if anything caught your interest.

And if there's something that you heard us talk about that you didn't see in the show notes, please do send us an email or get in touch and we'll do everything we can to capture that information and make sure it's available for everyone else. Okay, Dan, are you sitting comfortably?

Daniel Schien: Yes, I am. Thank you very much.

Chris Adams: All right, I think we should begin then.

So a few years back, one of the tropes that just would not die when people started talking about sustainability in the digital realm was this idea of streaming being the new flying. And, um, I saw your name linked to some work that was published in 2021, uh, working with Netflix that actually, in my view, helped shed some light into this and, uh, hopefully might put away some of the kind of poorly written part articles that you often see in the newspapers.

Could you talk, talk a little bit about some of this work that you did? Cause this was one of the first places I really came across your more mainstream work and it, I found it really interesting, actually.

Daniel Schien: Sure. So I mentioned earlier that my, my work here in Bristol on the sustainability of digital media started working with the Guardian and they wanted to understand the transition from print to digital. A couple of years later in 2015, we then worked with the BBC who also wanted to understand a similar transition. For them, it was the transition from terrestrial broadcast of BBC television to IP-based delivery. There is a broader context within which this transition happens. Terrestrial television, terrestrial broadcast uses up spectrum, and that spectrum is a scarce resource. And so there's a, there's an interest in potentially freeing the spectrum up for other purposes. So that's one of the drivers and the other is of course the affordance that IP-based delivery provides. You're moving from a broadcast mode to a unicast mode and, and enabling a lot of new ways of consuming. First of all, you can, you have video on demand as we know, and the other is more interactive forms of media. However, there was a concern about the environmental implications of it. And this concern was, was born just out of a lack of knowledge. At that time, the colleagues at the BBC, they, of course, had very good understanding of the energy consumption of the servers they were operating themselves, but what they did not have a firm grasp on was the energy consumption in networks that transport the data from data centers to the users and then the energy consumption of the devices in the homes. And this product system is, in its structure, relatively similar to the one that delivers the Guardian News Media. So we could leverage some of the methods that we developed in the work before to, to then come up with an estimate, an assessment model that then helped develop an estimate of the energy consumption for delivering BBC television, uh, over IP, but at the same time we delivered this, uh, we used this model to assess the impact from cable and from satellite as well, including terrestrial. And then we were able to also complement this with scenarios of how this might change in the future as the technology becomes more efficient, as the energy transforms, becomes more green, and as user behaviors change. So, so this was, this was a really, I think, yeah, a very successful collaboration. And there's a couple of things that we can dig into later on that are interesting here when it comes to the interpretation of those assessments. But one of the, one of the great things that came out of this was that at that time, that was one of the first studies that worked with an organization that provided the content and so had primary data of the impact in the infrastructure to some degree. And that was an important advantage compared to other work that is based purely on secondary data, on publicly available data, that it provides maybe a more robust.

Chris Adams: Yeah, you get to speak the people, surely. You're asked, you're asked like, "Hey, have I got this data right? Am I understanding it correctly?" And presumably, assuming they're being transparent and honest with you, you've got much greater access about what assumptions are going into that, rather than what you see elsewhere.

Yeah, that makes total sense, actually, Dan.

Daniel Schien: Yeah. So, and then a couple of years later, we, we had, um, a conversation with the uh, Responsible Media Forum, a small stakeholder representation run by Carnstone and, and they said, "can we somehow create a self-service assessment tool that lifts the, the methods that were developed in the work with the BBC, but makes them available for other organizations to also assess the environmental impact from the delivery of their

Chris Adams: Oh, wow.

Daniel Schien: services?" That was then the start of the DIMPACT project.

Chris Adams: Oh, I see. I hadn't really realized that there was a link going all the way back to public service broadcasting, essentially, in that model, actually. Wow. Okay. Huh.

Daniel Schien: That was the idea. And now many global media organizations are using this tool to report on the environmental impact of the distribution side. So I think this is a great success story. I always say that our role as academics is to create knowledge that we want to make the actual impact more transparent. And in this way, we have enabled the organizations providing those services to do exactly that.

Chris Adams: Wow. Cool. Thank you for that, Dan. All right. So there's a couple of things that I'd like to jump into there. And it sounds like... I thought I was just going to be asking you about Netflix, but it sounds like there's a whole set of different media that we might be thinking about and where there's changes and surprises along the way, basically.

I've got to ask you, were there any kind of counterintuitive findings that you had there? Because I immediately, for example, when you spoke about the idea of switching from analog to digital, I assume there's an energy impact in being able to switch off all the analog TV broadcast, for example, right? Is that something that people talk about?

Cause I hadn't really put those two together until you spoke about that, but surely that's actually going to be an impact, right?

Daniel Schien: Yes, absolutely. So, so this is maybe one of the nuances that, that we can spend a bit of time on. In the publication, in the academic publication, that, and I'm sure you will link to it in the show notes, there's a graph that compares the energy intensity of those four different modes of delivery of television from the BBC, so IP, cable, satellite, and terrestrial. And even though they, they differ, so between, if I remember correctly, between 60 watt-hours to 180 watt-hours, that's in 2016. There's a potential step in the interpretation of those results that consumers might take that needs more support. If you see this graph, you might think as a consumer, "if I change from streaming to watching something via terrestrial broadcast, I am going to save 100 watt-hours per viewer hour." However, that would not be a correct interpretation because all of those delivery modes, they are provided by an infrastructure that is inherently inelastic in its energy

Chris Adams: So you're talking about always being on. So if I've got like an analog television, even though I'm turning off the analog television, it's still broadcasting. So the same, so you've got a lot of the same analogs, sort of discussions we have now about "should I, is me using Netflix less, does that have an impact at the network level?" We have similar comparisons. It sounds

Daniel Schien: Exactly, exactly. So it's not that... your television, if you turn that off, you're saving energy right now. However, the modem in your home, that's going to stay on even if you're not streaming. The routers in the network, they hardly change. So in your ISP network, they're hardly changing the energy consumption. The terrestrial antenna network outside of your home in the country is not going to change its energy consumption. So on a very short term, your behavior as a consumer between swapping between those different modes of, of consuming television has hardly any difference outside of the home. If you are moving from a television to an iPad or to a phone, that will make a difference, but it won't make a difference outside of your home.

Chris Adams: Okay. I'm glad you said this because I ended up being, uh, on a podcast with, of all places, the Daily Mail, who really wanted to basically, wanted me to say that you're an absolute terrible person for watching Netflix. And if you watch twice as much Netflix, you are twice as bad a person. And, uh, I, I ended up basically not saying what they wanted me to say, which was that in context, not going, it's not going to be that much of a difference.

And yes, you probably going to have the impact at the device level rather than the other parts. I'm really glad that there's an actual doctor saying that as well now, because I feel slightly more validated in actually making the argument because yeah, it does seem somewhat counterintuitive, but it does seem like a kind of somewhat actionable thing for someone to be able to do.

Daniel Schien: Yeah, it's, it is unfortunate that the debate is now being tainted by a lot of, a lot of bias that comes in. You've got, it's been elevated on a very prominent level. There's a discourse that everyone now has, has an opinion on, and everyone has a right to have an opinion on this, but what you have, for example, in the media, you've got a bias to tell sensationalist stories, right?

Headlines that are emotional, that evoke emotions, they sell better. And there's sadly such an enormous amount of attention-grabbing exaggerations out there. It is, it's very sad, actually. And I've been subject to this. There was a Guardian interview that I spent several days in preparation, trying to provide enough data to have a really nuanced perspective on what is known about the carbon impact of streaming, but it's been completely contorted in the final article. That was very upsetting, actually. And I've been in touch frequently with newspapers when the reporting of work of Environmental Impact of not just streaming, but other forms of digital media and ICT services was just represented in a very wrong way, a headline grabbing sensationalist. So there's no, no silver bullet to preventing misinformation.

It is really a Sisyphus kind of work. These tropes, they come back again and again, and there's some, there's some work where the scientists that carry out the work apply methods that aren't robust, that are still under development, and that that happens. That is the nature of, of research and of knowledge evolving over time, our understanding improving, but then there's, once the work has been written down in academic articles, then there is the interpretation and the presentation of that work in the public sphere, or in a political sphere or elsewhere, that, that is, that potentially provides another layer where there's risk for misrepresentation.

Chris Adams: This is particularly relevant given the fact that we're seeing an explosion of laws being passed specifically around this stuff right now, both for and against a bunch of this stuff. We've seen the, actually just this week, I think even yesterday in Germany, where I live, we've seen what translates into English as the Energy Efficiency Law, specifically referring to data centers.

And you see the same thing at the European level as well. So you do see a real role for people's interpretation of this. And I actually wanted to ask you, because we spoke about some things which are like counterintuitive. I wonder how much of this you think might be related to the kind of mental models that people have, because a lot of the time, the mental model that you might have when thinking about digital sustainability, or like build it using a digital service, I can see why you would bias it towards the use that I see being the thing that infor that has the most impact, for example.

And I wonder, are there any papers or any reports or any guides that you've seen that help form a different men- mental model? Does it make sense of this, for example?

Daniel Schien: Yeah. So I want to maybe prefix this by saying that I don't want to be interpreted by saying that there's no environmental impact from our consumption of digital, of digital media or of ICT goods. There is uncertainty in the existing assessments. Um, but I think those that position the, the impact of the sector, the present impact of the sector in the order of 2 to 4%, um, they're probably reasonably robust to work with. And, and those 2 to 4%, they are high enough that the professionals working within the sector need to do something about it.

Chris Adams: Now this is two before, 2 to 4% of global if, of carbon emissions?

Daniel Schien: Exactly, exactly. Yeah, so of the 50 or so gigatons of anthropogenic greenhouse gas emissions per year, about 2 to 4% are allocated to the use of ICT services and the making of devices, and that is, and that's too much. We need to do something about it, especially as the rest of the economy hopefully decarbonizes and as our use of ICT increases.

So, so the dynamics that are at play here going forward, they're really, we need to pay attention to so that we can keep this in check. And at the same time, it is relatively small so that when we are trying to do what we can to reduce carbon emissions, then we need to prioritize our actions such that we have a firm grasp on, on the decarbonization of, of the sector. And so one, coming back to video streaming, if we demonize video streaming and the consumer who has to rely on guidance for action, that is straightforward to put into action if we are saying "streaming is bad" and instead consumers jump into their car and go to the cinema, that would be the worst thing that people could do. And same if we're saying streaming is bad and consumers swap to terrestrial broadcast, it makes hardly a difference in the short term to the energy consumption that is associated to the delivery of the service. I

Chris Adams: I'm glad you mentioned this one about terrestrial because I realized until you mentioned the BBC thing and the fact that people switch off a bunch of spectrum, I hadn't realized that there's actually a kind of net impact on that because you do see the same thing with, uh, cellular right now. We have 4G and 5G, and we are now seeing people switch off like old 3G transmitters, and there's like an energy saving from that. But in many ways, that's something that because it's seen as such a kind of industrial level, it's a kind of implementation detail. We're not really aware of a bunch of that stuff. And I feel like that's actually an interesting kind of angle that you presented on that.

So if I can move on a little bit further from this, so you did a bit of work with DIMPACT, I am pronouncing it correct. Is DIMPACT the way I should be pronouncing it? Is that okay?

Daniel Schien: DIMPACT is the pronunciation I use, yeah.

Chris Adams: Okay. Now, and DIMPACT, as I understand it, you mentioned there was a firm called Carnstone, they provide like services around this.

But the methodology is entirely open for other people to actually implement and use as well. And as I understand it, you are the advisory council with some of that, along with a number of other academics. So there was some guidance you were sharing for them as they were developing that. Is that correct?

Daniel Schien: Yeah, so indeed I chair the academic advisory board, and so the advisory board is, there's the methodology document and, and the exact constitution of the panel is there. There's sometimes the personnel changes, so it's best to see there who's there, but I think that the people that we work with are, uh, the absolute top academics, um, in this space, for example, Eric, Eric Masanet from the University of California in Santa Barbara is, is on the advisory board, um, or has been on the advisory board and he, together with Jon Koomey, have released an article in Jewel magazine a few years back, where they identified some of the principles that can lead to a misinterpretation of the academic impact of digital media, actually, that's really worth looking at.

Chris Adams: I think you're referring to, the paper is called Does Not Compute or something along those lines. Yeah?

All right. We'll share a link to that. Cause as, I found that really helpful as well for me to essentially make sense of this as someone who's not an academic. So actually on this subject of changing your mind and, uh, in face of the data, I should probably ask you, you've been researching this for a while, and, uh, we've already come up with a few kind of counterintuitive things.

Are there any areas where you've really changed your own view on something, where you thought something was the case and now there's more data available or the research has improved? You're actually thinking, "actually, that wasn't the lever for change I thought it would be. And there's somewhere else I should be looking instead."

Maybe you could talk a little bit about that, because this is one thing that. I think we're probably going to have to do more and more over the coming few years as the picture becomes clearer, both at the research and the industry level, as more things become more open.

Daniel Schien: So, less of a surprise moment is the gradual transformation of where I believe the emphasis of the attention should go, I started working on, on trying to, to, to understand the environmental impact of, of digital media now, 12 years ago, and with a keen focus on the use-phase energy consumption, because from a methodological perspective, it was one that, that was set in a, in sort of a gap, it's a distributed system where you've got data center networks and user devices, but the availability of data is poor.

No single party in this system has access to primary data. No single party can know what is the energy consumption in the data center because you can't get a power meter there or in the network or in the user devices. And so there's, methodologically, there was really quite a rich set of questions to explore. Um, now, the understanding of the environmental impact from streaming video has solidified. There's some variability around it and some uncertainty. However, the great new frontier is the impact from the making of the devices and, and in particular now with the electricity production in many countries becoming greener, there's a great success in the UK with decarbonizing electricity and hopefully we can continue along that trajectory. But as the electricity becomes greener, it becomes more and more important to understand and mitigate the carbon impact from the making of devices and, and we said, we spoke before about the interpretation for consumers. So when someone opens the newspaper and then they see something about streaming or about digital media, it is the metabolism of devices.

It is the frequent purchase of new devices that now is one of the key areas of environmental impact that consumers can actually do something about.

Chris Adams: When you say metabolism of devices in that way, you're talking about essentially built in refresh cycles or obsolescence, like a new iPhone each year or something along those lines, yeah?

Daniel Schien: Exactly so. I think that's, that is something where all consumers can have a meaningful impact by protecting the screen of their phones so they can last longer, and, and by, by not being swayed by the latest marketing stories to buy a new phone that is marginally faster, I think that that's something where, where, where the consumer can do something about it.

But the sector, the professionals within the organizations building the services, designing the services, of course it's their responsibility to do this as efficiently as they can. And, when I'm sitting here on my laptop and I hear the fan kicking in, then I'm often looking what are the threads that are currently ongoing.

What tasks are running there in the

Chris Adams: What things can I switch off on my computer to make it stop blowing away, right?

Daniel Schien: Yes, exactly. And sometimes you've got, you've got processes that are consuming CPU cycles where I'm pretty convinced there's waste going on. So what are some good examples? So I've got a mouse from a well-known manufacturer and it comes with a device driver, but the device driver consumes sometimes up to six gigabytes of memory. Because there must be a memory, there must be some sort of memory leak.

Chris Adams: This is for the mouse, right?

Daniel Schien: It's...

It's this one here. It's a good mouse,

but it is just carelessness. There are too many examples that I come across where the software running on my machine is made in a careless way. And I think the work that the Green Software Foundation does there can really help in increasing and improving the efficiency of software that runs on a single device.

I think we need to think more holistically and consider the entire system, not just on, on one device, but the whole balance between servers, networks, and use devices, but there is, there's still a rich space to make things more efficient.

Chris Adams: Okay. So we've got one about, so there's steps you can take at the use phase there. There are things you, as you describe, there are steps you might take in the design of something to make it run more efficient in use. And then there's another set of interventions you describe, which are design interventions you might make to reduce the need to cycle hardware quite so often, for example.

And it sounds like the former, where you might just make your code run more efficient, that's really common and relatively well understood, and we have some incentives aligned there. But this next part, that's something of a kind of newer horizon because the approaches you might take to essentially design something so it doesn't need to be replaced so much, you've got different drivers from an economic point of view, and we haven't figured out all those yet.

Are there any examples that you would point to that give an idea of where some of this might go? Because we talked about things like the Fairphone before as one example, and I know there are some cases in other sectors now. The first thing I think of is maybe like the framework computer, for example, which is designed to be taken apart or designed to be modularly put together.

Do you reckon there's some legs and stuff like that? Or yeah, maybe, maybe you could like elaborate on some of that stuff there.

Daniel Schien: I have the feeling that a cloud-based distribution model or, or a model to run software has got an enormous potential to reduce the impact for, from the manufacturing of devices and also for also for the runtime energy consumption. Just the maintenance of applications that are distributed over a traditional model where you've got a sort of a fat client living somewhere on a customer device for an organization constructing software has enormous implications for the, for the, for the maintenance, yeah, for the, for the ongoing support.

Whereas if you have a, um, if the, if the software runs in the cloud, um, then just by the, by the interface um, that you have over the web, um, you, the, the, the, the cloud software provider has, has, has much higher control, um, over the anatomy, the architecture, and the implementation and the evolution of the software to get, to make it more efficient.

Chris Adams: So to make some of this more concrete, let's say that you wanted to do something at a hardware level. You're like, yeah, you could buy a really fast machine, for example, but there are tools where we work. We use Gitpod because one of the reasons is that basically shifts all the kind of development work into a machine in the cloud.

And there's things like GitHub Spaces, which do this idea of shifting the work to a more centralized place. That's what you're, it sounds like you're talking about, and that might be easier to green, for example, as well?

Daniel Schien: Yeah, uh, so. I think there's probably the devil in the detail. Not all applications, um, will, will suit this model of a, of a cloud-based, um, distributed system equally well, but if you can have a thin client to, to just access that cloud service for many applications, there, I believe this is the opportunity to then hold on to this thin client for longer.

Chris Adams: Okay. So there you have it then. All right. So that's one of the ideas that we might be looking at. And there's a bunch of other questions about, okay, which, whose cloud are you talking about here? And there's a sovereignty issue there. And if you're going to concentrate with it, how can you be sure that you're going to get the data out of it?

Or there's transparency issues and things related to that, which you've, I think you've alluded to elsewhere. Are there any areas in the research community right now where you see some, I don't know, opinions or ideas that are probably a few further along than where the rest of industry might actually be right now, that you think is probably something that people could spend more time looking at? Because I know that you read papers and you publish papers on a regular basis, but we also see that there is a gap between what being discussed at, say, HotCarbon or the E-Impact events, for example, and then what's more mainstream and like the industry at present. Thank you.

Daniel Schien: So, so my own concern, has moved from trying to understand what the current impact is and in the past, relying on accounting style methods to a concern of how can we robustly enable all the stakeholders to take action that can result in the reduction of that environmental impact. And that includes the consumers. What can I, what concrete decision support, what guidance can I give to consumers when they decide between different options to consume digital media? What can I say to the designers, to the engineers building those services to do this in the least environmentally impactful manner? And infrastructure operators, is there something that they can do so that we are moving towards decarbonizing the system? And what I said before, the concern that there is a high base power consumption of devices and the lack of energy elasticity. There is the overhead from the manufacturing of devices that, that lock you in, um, to amortize this impact over the lifetime, um, of the, of the device that has in, in my mind eroded the confidence in the set of existing accounting-style methods to really provide that decision support in a robust way. I think one of the key questions that we still have to solve is that of electricity consumption and carbon intensity of electricity. There have been some examples how, again, a lack of awareness about the nuances and complexities in, of the electricity system have resulted in guidance and advice that is potentially not correct.

So there's a, for example, there's a difference between the carbon intensity of electricity, uh, that is, the average carbon intensity over the course of the day as it changes with the change of the mix of electricity and the marginal electricity intensity in the sense of, if I bring additional load onto the network or if I take load off, then the effect of the generation of electricity

Chris Adams: Is it going to be particularly dirty? Okay, I think you're referring to the fact that you have different companies who have different ways of talking about this. Watttime is one example, they have a marginal intensity and there's maybe other ways which use average intensity. And as far as I understand it, these end up with different incentives.

If you're going to follow the marginal approach, and it's a lunchtime, where there's loads of solar coming in, for example, you might have a very high marginal intensity, thinking that there's lots of power on the grid, so therefore me doing that, the marginal unit of generation might be gas, which is very bad, but that gives you the idea that you shouldn't charge your car when the sun is in the sky, you should wait till it's in the evening instead, when there isn't so much demand on the grid.

It gives you different conflicting signals. This is what you're talking about here,

Daniel Schien: That's exactly what I'm, that's the knowledge of when is it better for me to consume? Or rather, how can I use a carbon intensity for electricity to actually make decisions that are really resulting in fewer carbon emissions being released into the atmosphere? That is the question. And, and the, and, and ultimately we use the term marginal to refer to the, the, the, the relative, uh, change of the intensity as a result of my action.

That's how we want to use the term marginal, but the metrics that are behind some of the released marginal factors, they don't actually represent exactly this goal. Yeah. And so, so this is, that needs to be better understood. We can't use the average variable and interpret this as the marginal, but I'm, this is, for example, one of the, one of the questions on my, in my mind that that needs to be better understood.

And there's a lot of, there's a lot of people from a variety of disciplines that are now looking into this. And it is related to, to this question of causally, how can we, what's going on in the wider system? How do my actions now translate into a change of the system in the future?

Chris Adams: I see. So you're talking, this is one about the time scale that you might have. So on a very short term, this might make sense, but in a longer term, it may be that, let's say we had this example of don't use being in center with a number telling you not to do something when there's loads of sun in the sky.

It may be that in the long run, it makes more sense to actually have a signal for demand so that people just build loads, loads more solar there. But that's not going to be something that's going to be captured inside your kind of short term marginal thing. That's a longer term signal that you'd need to be thinking about, that you'd need to somehow capture in, in a number or an index or something like that.

Right?

Daniel Schien: Exactly. And we've got similar effects for other parts of this networked system that we're working with that provides those services. So the electricity, the electric grid is a network, but our.

Chris Adams: network is a network! Yeah.

Daniel Schien: is a network. And, and so you've got, you've got similar, um, similar problems. Um, maybe the answers will, will look slightly different, um, in each case, but they all need to be looked at, um, collectively, uh, in order to, to, to make decisions.

Chris Adams: Okay, so this idea of the timescale where you might see an impact, as I understand it, this is some of your future research that I think you mentioned to me that you might be looking to do going forward. Is there a chance maybe you might touch on a little bit about that? Because as we just round off the hour, I feel like it might be worth asking you, like, where you feel some of the more kind of promising and really interesting areas of research are. Based on what you've said so far, it sounds like there's probably, I've got a good idea, but maybe you might just talk a little bit about where you expect some of this research to go, because, oh yeah, congratulations.

I saw a chunky grant landed for this kind of research, and I'm very excited to see where it goes now.

Daniel Schien: Yeah, thank you very much. Indeed, I've been, um, very pleased to have been awarded a research grant, um, to investigate the, um, a change-based perspective to the environmental impact assessment for digital services. So everything that we said before around the, the complexities in the system that prevent us from, from making easy conclusions, drawing easy conclusions with the existing methods. We want to now work together. We've got a, uh, a really fantastic set of partners that have that primary data to allow us to to understand robustly what is the current impact, and then also evaluate what is the future impact. And together with the engineers working in those organizations, we want to have conversations that, that help us understand what is, what's going on when digital services are being designed and their interaction with the network.

It's the, it's the relationship between the service and the infrastructure and the consumption that we want to untangle. It's this dynamic system and how it changes over time. And that's uncertain to some degree. What is in the future will always be subject to some uncertainty. But unless we, we dare to ask those questions and understand the basic principles, we are leaving this entirely to chance. And that really is a risk to our ambitions to decarbonize um, all parts of this society. Uh, yeah. So very happy, um, that, that, that I've been trusted with, with this piece of work. And here's a call out, um, to, to listeners. Um, if you're interested to work in this space, um, we are, we are looking for researchers to join us.

Um, there will be ads for a job for a senior research associate coming out into the future and Chris might be able to include

Chris Adams: Yeah, we'll be very happy to share links to this, especially if we get to use some of this research ourselves going forward. Because as you say, we've been struggling as a practitioner when I work in this field. It's very hard to figure out, okay, what am I going to do in the short term? And then what are the things I need to do that work on the longer or medium term?

Because I work in an organization which has a 2030 deadline for an entirely fossil free internet, but also we also have like science-based targets, like the science is informed, basically the industry is saying that we need to halve emissions globally by 50%, particularly in the tech sector by 2030. And there are some papers and things being published to give some indication of this.

But as you've described, Daniel, there's a lot more to it than that. And the more we can actually have to influence policy, the better in this way, in this way. Okay, Dan. I've enjoyed diving into the weeds with you like this. I just want to ask, as we round up, if people have found any of this interesting or would like to continue some of this research, where should they go online to find out more about either your work or some of the papers and things that you're describing?

So we'll add the show notes, but is there a particular place you would direct people to?

Daniel Schien: So, maybe a good starting point is our 2019 CHI paper on YouTube, where we carried out an evaluation of a design change to YouTube that can reduce carbon emissions. Now, this is, this has been done purely with publicly available data. YouTube was not part of this. And so this is, there are assumptions in there and some uncertainty, but methodologically it describes quite well the general approach.

It's also important to note that In this, in this model, we are using an allocation-based approach to estimating the energy consumption of the network and the changes. We are, we're transparent around this and the shortcomings of it. So, so it needs to be seen in perspective, but broadly, I think that's a good text to start with.

Chris Adams: Great, okay, thank you for that. All right, Daniel, I just realized that you are also the author of the Low Carbon Kubernetes Scheduler, a piece of work that has influenced a bunch of nerds inside the GSF. We'll have to talk about that another day, about going from Kubernetes to Low Carbonetes. But Daniel, thank you so much for spending the time on this.

I've really enjoyed nerding out with you on this, and I'm pretty sure we're going to have a bunch of really interesting links and papers for anyone who fancies doing some more research from here. So once again, thank you, Daniel. Alright.

Daniel Schien: Thank you so much, Chris.

Chris Adams: Lovely seeing you again, and take care of yourself, have a lovely weekend.

Daniel Schien: Likewise. Thanks a lot. Bye bye.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again, and see you in the next episode!

The Week in Green Software: Net Zero Cloud

19 oktober 2023 | 44 min

This week, join host Chris Adams in a conversation with Tereze Gaile, global Sustainability SME at MuleSoft, a Salesforce company. Together they cover topics in sustainability from her organic sustainability role at MuleSoft, to various tools that are available to aid in the decarbonization of software. They touch on resources to help you learn important concepts and skills, how to bring sustainability into an organization, and much more on this episode of TWiGS.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Tereze Gaile: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Salesforce Launches ‘Green Code’ Initiative to Reduce Carbon Footprint of Software [14:49]
Sustainability Guide for Salesforce Technology [15:14]
Cut Carbon Emissions with the Developer Carbon Dashboard | Salesforce Developers Blog [16:40]
Trailhead | Salesforce [19:28]
Developer Carbon Dashboard GitHub Repo [21:18]
Emissions Reduction - FY23 Stakeholder Impact Report [34:41]
How much of the internet is covered by credible Net Zero targets? An update - Green Web Foundation [37:02]

Resources:

Adopting Sustainable Engineering Practices | MuleSoft Blog [7:36]
http://principles.green/ and the GitHub repo with the original priciples.green repo website [7:49]
Consequence scanning, as introduced by Dotevertone, as described by non profit MySociety and how to plan a workshop using it on trailhead [20:45]
COMPASS EDUCATION The AMOEBA of Cultural Change [42:35]
https://outrageandoptimism.libsyn.com/ [43:01]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Tereze Gaile: Sometimes, the challenge is not that they don't care necessarily, but that they have multiple conflicting priorities. And I've seen this time and time again, especially when we look at today, there's a cost of living crisis, and if an organization's bottom line is taking a hit as well, sustainability can be a really hard sell.

Tereze Gaile: hi Chris, very happy to be here and excited for our conversation. Thank you for the opportunity.

Chris Adams: You're welcome, Tereze okay. So Tereze, when we first spoke, I have to confess, I'd heard of Salesforce and I think most people working in tech have heard of that company, but I didn't, didn't know that much about MuleSoft and I didn't quite understand the relationship between the two. So for the uninitiated, would you take a moment to introduce yourself, just briefly outline this relationship between the two and give a bit of background about what you do there, please?

Tereze Gaile: It would be my pleasure. First of all, my name is Tereze Gaile, and I'm a technical architect at MuleSoft, a Salesforce company. I'm also MuleSoft's global sustainability champion slash subject matter expert, but more on this later. First, let's start with a little story. to get a better understanding on Salesforce and MuleSoft and how they are connected.

So Salesforce was founded in 1999 and its core product back then was Salesforce CRM or Customer Relationship Management and its vision in a nutshell was to be a world class internet company for Salesforce Automation, hence why the name. Since then, Salesforce has acquired quite a large number of organizations and a vision has evolved to include AI, data, and CRM.

So some of Salesforce's acquisitions include companies that you might have heard of, so like Tableau, Slack, and also MuleSoft, with the latter being acquired in 2018. And MuleSoft is essentially a product or integration platform as a service. Which means you can use our platform to design, build, and manage your APIs and integrations.

But something that's quite interesting is it actually started as an open source project. Back then, it was just a Java-based Mule enterprise service bus. But of course, over time, we have evolved, keeping in tune with Salesforce's own evolution. And also, we do have capability around logic process automation. At this point, maybe you're wondering why we are called MuleSoft.

It seems like a slightly name. Well, back then, data integration used to be very manual, painful process with a lot of configuring of how to connect with different components. And the core idea behind the software itself was how do you take the donkey work out of that? So hence why we're called MuleSoft.

Chris Adams: Okay. Thank you. Thank you for finally making sense of that because I really didn't understand where mule came from. But now mule is in the same place as my head as say yak shaving and stuff like that. Stuff I want to get done but might not be that much fun to do myself. Okay, great. And so last week we spoke to a Kiwi who had moved from New Zealand to Denmark, Tony van Swet from Electricity Maps. Now, as I understand it, it's the other way around. So I'm speaking to you from Berlin to, I believe, was it Wellington in New Zealand? But you started out in Europe, right?

Tereze Gaile: That's right. I was born in Latvia, but when I was seven years old, we moved over to the UK. As you can probably tell by my accent as well, I spent majority of my time in the UK. So I went through the school system. I studied aeronautical engineering at Imperial College London, and I also started my career in London too working at companies like, uh, Deloitte Consulting. Relatively recently, so 15 months ago or so, I moved to New Zealand. So yes, now I am based in the windiest little capital in the world, wellington in New Zealand, a-Tara. And it probably explains why I do have a slightly. I have a little bit of Latvian, a little bit of British, and definitely starting to pick up some of the mannerisms from this of the world as well.

Chris Adams: well. Yeah. Okay. Thank you very much for giving a bit of context. That makes a lot more sense now. Okay. I suppose I should just introduce myself too. Cause we just dive straight in without actually knowing who I am. Okay. So for people who've never listened to this podcast before, my name is Chris Adams.

I am the executive director at a Dutch nonprofit called the Green Web Foundation. I also work as the policy chair in the Green Software Foundation, which is a larger industry-based body. And I also organize an online community called climateaction.tech, which as you might imagine is about climate action in the tech sector. Okay, so before we dive into the real meat of this podcast, it's useful to know that everything we discuss, we'll try as much as we can to find a link for any kind of story or project, and we do put a lot of effort into these show notes, so if there's something that you've heard about, you found interesting, and you didn't see in the show notes on our website, please do let us know, because we would really like to make sure that there's a useful resource for other curious souls who are listening and want to find out as well.

Thank you. Okay, with that out of the way, I guess we should start. That sound okay to you, Tereze?

Tereze Gaile: Oh, absolutely. I'm very impressed with all your different titles, by the way, I'm blown away.

Chris Adams: All right. Okay. So, Tereze, you, as I understand it, you weren't brought in as an outside sustainability expert, but you've, this role that you're in has emerged somewhat organically. And I understand that basically blogging and internal comms played a part of it. Could you maybe just elaborate on that a little bit?

Because I found this interesting and useful for people who are starting out in this journey as well.

Tereze Gaile: Absolutely. When joining MuleSoft, our architects essentially have a choice what they want to specialize in, and they can pick those areas. And I personally from the very beginning chose sustainability. Why? The first reason is I wanted to embed sustainability into my professional life and not just more on a personal level.

And also very quickly it became apparent that no one else had really embarked on this path yet. So my first milestone was writing a blog post. I know it doesn't sound like a very big deal, but it was the very first thing in this space, and it was called Sustainable Engineering Practices with MuleSoft.

And another fun fact, it was actually inspired by principles.green, that Asim wrote

Chris Adams: Ah, cool.

Tereze Gaile: how many years ago, but it's probably also explains why I've been following the foundation very closely as well. And the blog itself was a catalyst in the sense that it allowed me to get connected with others across my own organization who are also passionate about sustainability and keen to be change makers in this space.

So fast forward two and a half years, needless to say, we have come a really long way. We've created a cross-functional team that essentially spans our core sustainability team, it also has members from our technology and product organization, and also from field teams like myself from customer success.

And we're really proud to share some of our milestones like, and I'm sure we'll talk about this momentarily, and other things like the Carbon Developer Dashboard and the Sustainability Guide for Salesforce Technology, our Enablement Content or Training Module on Green Code, and many other things.

Chris Adams: Okay. So it sounds like there's actual physical running code like services. There's actually some guidance that you might use in a professional engagement when you're working with a client. And it sounds like there's a bit of self service stuff for people who maybe don't have the skills or don't, don't have a, an ongoing engagement right now, but might want to learn where to start and how to get some background on it. Okay, cool. Thank you. All right. So... And maybe this is a nice way to move into the next question. So I get the impression that like lots of other organizations, there are people who've been working in various other parts of the problem in their own way inside an organization. So it's not like you're starting from scratch and there's usually someone working somewhere in your organization, but they're usually doing it in maybe an ad hoc fashion, or it might be seen as isolated heroic actions rather than something where there's an organization really getting behind it. But you also mentioned a little bit about working on a kind of client-facing role. You said you're in the field and one thing we've seen before is people talking about client work as a way to actually tap into interest that might be there as a way to even support a discussion or kind of highlight that people are caring about this outside of your organization.

If you might not know who those people are inside your company, could you touch on some of that, because I think there's a few things worth touching on. Then you actually just said, "I came in as an architect and I want to look at this and there wasn't this data, there wasn't really this proof just yet, but maybe there was a, maybe there was interest outside the organization."

Tereze Gaile: Yeah, that, even just that outline echoes a lot of the story I'll share. It's hard going from zero to a hundred in one go, right? So our strategy from the very beginning was how can we weave sustainability into our existing roles and responsibilities, essentially prove the value, and then build on that momentum.

For me, as you're well aware, I work in customer success. So, if customers are interested in sustainability, then I 100 percent have the green light to spend time on this topic. But of course, it's a little bit of chicken egg scenario. We needed to generate demand. When we initially started and we put MuleSoft and sustainability in the same sentence, people, most people had a really puzzled look on their face, like, "what do you mean? What's there?" Um, so we have to essentially, first of all, create a budding sustainability point of view. And then I relied heavily on our account teams to actually share that messaging with our customers to essentially plant a seed. So things like "actually, data centers emit as much as the aviation sector," right? I think that really gets technologists sitting up and taking notes. So it's really like to plant that seed initially. And it didn't take long at all for us to get a lot of interest. And those initial customers were integral in helping us build a more robust approach.

And also understand where are some of the gaps, what's important for them? So things like, for some individuals, they're really wanting to understand what their platform emissions are and what they can do to reduce those. Others have questions like, "Hey, I'm deploying on Anypoint platform," which is our cloud platform, essentially.

"What is, what are the emissions? What can I do to mitigate some of the adverse environmental impact?" Whereas for other customers, there might be more, let's say, thinking, not necessarily from a green code or green software perspective, but more how they can use the technology for good. It could be things like, "great, MuleSoft is an integration platform, but how can it help us with carbon reporting?" Be it with something like Net Zero Cloud, which Salesforce has, or another carbon reporting platform, right?

So it's really helpful for us then to see those two areas and actually start investing a lot more in that.

Chris Adams: Okay, cool. So one thing I think you mentioned before was this idea that, so. I know that if I'm using, say, Slack, or I'm using, say, Tableau, there's a limited amount of control over how I might use that, or how I might deploy things, right? So, I don't have the same kind of spectrum of interventions I might have available. But one thing you touched on before was, MuleSoft was an open source piece of technology. And as I understand it, there's a bit more freedom into kind of changing how you might deploy it inside an organization to have some different impact there. And like, I guess that gives a bit more wiggle room to have some of these conversations and inform and inform.

So those conversations with other people, when they're saying my priority is X versus Y, for example, maybe we could touch a little on some of that and then dive into some of the other parts. Cause as I, as I understand it, there's actually some relatively high profile external work that's come out of some of the cross functional team stuff that you mentioned before.

Tereze Gaile: Yeah, so to address the first point, MuleSoft, as I mentioned, is a, um, platform as a service. So actually you do have a lot more wiggle room. We essentially deployed these applications on, um, workers or replicas in the cloud. So, same sort of approach that you would have to managing your workloads in the cloud, if you have Google or AWS, all of that is very applicable still, whereas for most of Salesforce products, they actually are more software as a service.

So then again, the practices or the advice that we might share is very different. You know, you don't have an option where you deploy it, but you might have options how you configure your data strategy or how often you're refreshing that platform or the kind of automation you're building on top of it, like you, those best practices will be tailored or slightly different for those.

In relation to your second point about some of the things. We've created to one, raise that collective awareness and start taking customers on that journey. And the first one I'll talk about is our green code initiative. It's more than a couple of months old now, we shared it externally in May. The whole idea really behind it is "how can we help technologists reduce the emissions associated with using Salesforce products?"

And this is across the lifecycle. As I already hinted, of course, we have a myriad of different products. So as part of this initiative, we released a sustainability guide, and that has a number of those practices that basically show you what you can do, it's not an exhaustive list of every single thing you possibly could do, it's just more to raise awareness, start the conversation, and give technologists something to do already. And we're very open to have feedback and evolve it over time. But that's essentially what was a core part of this Green Code Initiative is this Sustainability Guide for Salesforce technology. And as part of this initiative as well, there was an interesting piece of research that came out, and it was, they found that 75 percent of technologists do want to develop softwares sustainably, but they don't really know how.

So it's half them had no idea. So it's really like tapping into that gap and it's early days, but I think we're quite excited to see where it goes over time as well. And the other one that I'll mention very quickly as an example, cause I think, while it's awesome to have awareness of what you can do in terms of how you can build or design applications. It's really hard to know or improve if you don't know where you are already. So you need to have that visibility. You can't improve what you can't measure, right? So, to tackle this, we release a Developer Carbon Dashboard. And it's really exciting because for the first time it's something that a customer can download on their own Salesforce instance.

Again, this is open source, so they can customize it for their org and actually get information in terms of the carbon dioxide for different parts of that core platform. So across business logic, the UI layer, API layer as well. So it's really cool because actually it's, you have that granular data now, whereas before it was, "Oh, we can estimate it" or like finger in the air.

But now actually there's a tool that can help you get that visibility.

Chris Adams: Uh, okay. Those sound quite juicy. So let's address those separately. So the first thing as I understand it, and I've got to open in just a tab here. So the sustainability guide for Salesforce specifically, this is like a PDF, about 22 pages long, and as I understand it, this is people who might be using these tools, not sure where to start, and they have some awareness or they, they have an interest, like you said, they're part of that 75% of people. And as I understand it, when I've looked through this, it's broken down into different parts. Say, uh, there's maybe something, "if you're a designer, then these are some of the steps that you might take." And when I look through this, I was actually. I'll be honest, I was pleasantly surprised to see some of this stuff because I saw things like, say, consequence scanning, some of the kind of activities that you might use as a designer at the beginning of a project to say, "how do we actually design out some of the unintended impacts in the first place, rather than actually just trying to compensate for them for them later?"

And I see a number of like other open source projects mentioned inside this. So this is actually quite cool. And, um, it's so this was, this basically talks about design part and then the architecture thing, which you might mention too, but it goes into quite a lot of detail and actually saying, "these are some of the ways that you might measure this."

And maybe we'll touch on this a little bit later. It does actually seem to refer to some of the early work. So I see Etsy cited for their work before, and I can see there's a number of concrete examples of this stuff, rather than things being entirely academic, actually. All right. So, and I'm sure the thing that I, when I read through this, there was some stuff I didn't quite understand because I don't use Salesforce myself.

So there's some mentions from specific tooling that I haven't seen elsewhere, like DataWeave or the various kinds of languages that you might be using for this. So, so, this is really helpful. I didn't realize that this was actually visible and it does actually provide a decent, it's quite readable to run through as well. So you have this here and there's chances to dive into this a little bit deeper later on in the, some of the online training tools or some of the other kinds of content that gets published for Salesforce, right?

Tereze Gaile: Yeah. So. In addition to the guide. So the guide is just one piece of how we're tackling the whole problem of, that there's just not enough awareness around, especially the intersection of climate and tech. Another thing that we have is a, we have an online learning platform called Trailhead. and the whole idea behind this is, it has a lot of content for Salesforce specific products, but it does have really nice modules on topics that are completely unrelated to anything that we really do at its core. So we will have numerous sustainability topics like creating a sustainable future, which includes soft sustainable development goals, also strategies for positive environmental impact, or how we approach sustainability with our suppliers.

So there's loads of things like that as well, among things like wellbeing and a bunch of other things. So I think It's a really powerful tool to open up that information and share it and it's free, right? Anyone can sign up to it as well, which think is really important. And yeah, any, where would you like me to go next?

Chris Adams: All right. That was good. That's helpful because this was, I was surprised by the breadth of what we had here. And I don't want this to sound like an advert for one company, but when I saw that, I was actually quite impressed by the fact that, okay, yeah, you're talking about the supply chain stuff as well as all these other things, and a lot of these seem like initiatives or projects, which didn't start at your company, but were interesting ones from other parts of the world. So consequence scanning is an activity for talking about who you're prioritizing in a design project, for example, and that came from a a group called Dot Everyone back in the late 2010s, and likewise, the supply chain stuff. You can see some of these ideas linked to the Chancery Lane project that we've mentioned on previous podcasts about, "okay, if you're going to do this, you need to be thinking about supply chain as well." So yeah, that was actually, it was nice to see some of these references to things which are not invented here, but still worthy of people's attention. So if I may, I just want to dive briefly onto the developer carbon dashboard thing, because this feels like something which is a bit on brand for our podcast. So you said it's an open source piece of tool that people deploy themselves or can deploy themselves, and there's a breakdown of different stages because in many ways, you might only, when you're working with cloud, for example, you might only see one part of it. You work without seeing stuff that's happening at, say, the end user, for example. Could you maybe talk a little bit about that part there? Cause I feel like the nerds will probably really like that part, especially if they know it's open source and they can look at some of the source code as well.

Tereze Gaile: Yeah, so I think in comparison, again, just using MuleSoft as an example, we don't have a front end, right? So the apps you create, it's more optimizing those, the underlying infrastructure, things like that. But for core, which we call things like Service Cloud, Sales Cloud, there's a myriad of different clouds, all under the hood has, um, the same sort of architecture and stack. Idea is based on how you built on top of it.

So you might have had some processes or flows, and this is really trying to give you a little bit of an insight on where some of those emissions are coming from. So, because it is a software as a service, it has a user interface component, so you might have done something, um, like really rich customization and experience cloud, had maybe heavy images or like quite complicated logic, so all of a sudden, you will see on the user interface side, more emissions from a percentage perspective, um, arising from that area, perhaps you have a lot of logic that do a lot of back end calls to different parts of the platform or different, um, external systems.

So then again you might see a spike in, um, API, API calls. And Apex essentially is one of the Salesforce's propriety coding languages, I guess. And again, you might have had a lot of Apex customization, which the dashboard will flag as, okay, maybe there are some inefficiencies or even just the fact that it's like a higher percentage than maybe another instance that you're comparing could give you an insight, like, okay, let's take a look.

Let's try and understand.

Chris Adams: Okay, cool. So basically is, so why, what it sounds like is that there are tools like Cloud Carbon Footprint or various things which look at maybe usage data, but because there's a bit of awareness of maybe a particular language or some of the architecture, it's giving more specific guidance about what you might do to mitigate some of the increases in emissions from maybe deploying a new system or something like that. Okay. So you've spoken a bit about this and about the actual, the new dashboard. You mentioned that it was open source. So presumably it's on GitLab or GitHub or Iceberg or Gitea or whatever tool someone's going to have online.

Is that the case?

Tereze Gaile: Yeah, absolutely. It's on GitHub. So any interested techie can go ahead, take a look, download, give feedback. And, um,

Chris Adams: make an issue, ask a question. Okay, cool. And as I understand it, there's a chap in the Climate Action Tech Slack, Boris. Oh, Boris, I'm so sorry for mispronouncing your name. Gamasch Boris G. So he's the person who's been leading this project and has a background working on the AI part as well. So there's, so that's, that's how I understand it.

He, he's been what he, this is mainly his baby so far that he's been spending his time putting into this and he's been leading on this. Is that the case?

Tereze Gaile: First thing I'll say, very recently, I actually had a chat with Boris around the pain of our names being mispronounced. So, he'll find this funny for sure. And, uh, for the Carbon Developer Dashboard, it was a real joint effort. So, in our core sustainability team, so he's across all of these sort of initiatives from a technical perspective.

We also have, um some of the people who are like really hands-on with the product itself, so like Kenny, I will also have people like, um, Tyson Read, who's essentially the product manager behind it. And I'm missing a lot of other people. Uh, there was, you know, I don't know, 10 or something like that. There are a lot of people involved in getting this off the ground.

So again, like a massive kudos and well done to them.

Chris Adams: Cool. All right. And that's been, that's quite new. That was launched just last month, right? Something like that? The Developer Dashboard. I think I saw a reference to it saying, "Hey, this is the new thing that's out." So it's open and there's still a chance to get some of those early kudos points for making a, I don't know, fix a typo in the docs or something like that.

Credit, commit into the project. Okay, cool. So we've spoken so far about, okay, all the nice happy path stuff. Yeah, everyone's want to be on board, you've got a client who really loves this stuff. We shouldn't kid ourselves that everyone is really into planet friendly tech, right? And we've come across pushback before. So I figured maybe we could talk about some of the common ways people push back on incorporating sustainability into what they do. And also some of the strategies you've seen work when working around some of the resistance, or even to respond to some of that pushback when people are saying, "Hey, I don't want to do X because of Y," for example, because this sounds like something you may have come across. So, yeah, if I give you the floor, maybe you could just talk about some of that part as well, actually, Tereze.

Tereze Gaile: Absolutely. Sometimes the challenge is not that they don't care necessarily, but that they have multiple conflicting priorities, and I've seen this time and time again, especially when we look at today, there's a cost of living crisis, and if an organisation's bottom line is taken hit as well, sustainability can be a really hard sell.

Reality is we're in the middle of a climate crisis. But it's still business as usual, and it's really hard to change this thinking. I think we all understand at a high level what's happening, but to actually change how we live, it's hard. But this is where I try and get inventive. So something that I've already hinted at earlier, I've tried to generate that customer demand around sustainability, that in turn means that I can start influencing our own product roadmap. So customers are interested, the more I can prove that, the more I can bring it in internally. Product managers will start prioritizing that, they'll understand how important it is. That's something that I do in, in my own remit, essentially.

Also, if customers want more granular data, then we need to do our homework. We need to start working with our suppliers, our vendors. And that actually puts pressure on them. So, for example, there have been multiple workshops between our engineering and technical teams with Amazon Web Services on this very topic.

We can't really be super transparent if we don't know what we're hosting our on top as well. Exactly. But these are just some of the ways we're tackling it at the moment, and also I will say persistence goes a long way, like for many. I've heard quite a few episodes in the past where it takes time and a few tries to find the right people or the right leader who will really elevate what you're trying to do, and individuals who are happy to partner up with you on that journey as well.

And lastly, what I will say with my Salesforce hat is sustainability is actually one of our most recent values. It's only two years, a year old, something like that. And we always have that in our back pocket as well. So to truly operationalize that value, we do have to embed it in everything. So whenever there's some kind of pushback or someone might not be on board, we're like "actually it's a core value now and we have to be accountable."

But for that, we need both top down and bottom up approaches as well to make it successful.

Chris Adams: Okay, that last thing you said was quite interesting, because it sounds like there's a series of different strategies you might use. So some of this might be the common thing that people are often like that. So there is perceived wisdom that you might lead with, "Oh, this is going to save you money. So therefore you save money and save the planet."

That's one example. And one of these, you just, the thing you just mentioned was almost like, no, this is actually more actually just an appeal to authority. So like the organization has said, this is one thing that we care about. So you're using that. That's quite similar to what Asim has mentioned as well. So this is Asim, uh, the, who now works at Intel, but previously was working at Microsoft. He was saying, when he was working at Microsoft, he would get pushback there on this, and he was basically able to say, our boss, Satya Nadella, has said we need to, uh, halve emissions by 2030, or something like that. And that allowed him, that gave him the kind of cover to basically say, "the boss says this, so even if you don't agree with me, because the boss has made this a priority, we're just gonna, you know, we, this is a, this is why it's important."

So it feels like there's different ways that you can actually make sure this is actually on the table, at least being discussed on a regular basis. And I, I like what you said about the, the, the persistence part as well, because if you've ever worked in a kind of kind of customer facing thing, you know, when you write a proposal, you propose something and then it's crickets.

And then at the end of the year, someone will come through and say, "Oh, Could you do this thing for us?" For example. That's happened a few times in the organization I work in and it does feel like you can write these things and often you won't know when it's going to come back up again. A lead might seem to go quiet and then they'll say, "oh, it turns out we've got a bit of budget or we need to spend it before the end of this financial year.

Can we please try this experiment or can we try to do a thing together?" That's nice to hear it coming from you as well. Okay, cool. Thank you. Let's, so we spoke a little bit about some of the kind of moves that you have, or some of the kind of strategies you might use, and some of the things that you've either used yourself or had else in other places. I think it's easy to, and maybe we should just touch on this idea of, okay, not everyone's super into planet friendly tech, for example, but. It's also important. There's a flip side to that. We can get a little bit ahead of ourselves and thinking there is nothing else but green code. And it's also easy when we're excited about something to overstay or get a little bit carried away with what we talk about. But there's a, there's an issue about keeping people's trust. And like you mentioned this idea about granularity and transparency. I want to be as transparent as I can, but I'm somewhat limited inside this. And there's a piece about you communicating with customers or people say, "this is what I can share with you based on the fact of what we're using." And the fact that you're communicating some of those limitations makes it easier for people to trust you. Maybe we might just touch a little bit about on say how you can sound credible when talking about sustainability because, uh, various organizations do tell this wonderful story, particularly in tech sector, like how there's zero impact and it doesn't matter what we, that we have any kind of impact because we're just gonna net out in zero or anything like that.

And we now see a real kind of shift in policy away from making these kinds of claims. And in Europe, there's almost the direction is actually quite a bit more developed. Like the idea of saying things are carbon neutral is very, it looks like it's going to be controlled a lot more. Maybe you could talk a little bit about that part because we spoke a little bit about companies being like say net zero or stuff like that and there is fact that emissions kind of grow and we need to have, be, be credible when we're talking about this, specifically in this sector. So maybe if I could just give you a bit of space to kind of share some of your thoughts or how you see this, that might be helpful for us to kind of like round some of this up with.

Tereze Gaile: That's a great question, Chris. We've actually seen this quite a lot recently in this part of the world as well, where a large number of organizations essentially will rely on offsets instead of reducing absolute emissions for their net zero claims. So, actually, I think one thing that's quite timely to share, recently the Australian Competition and Commerce Commission published guidelines on anti-greenwashing, essentially, um, and it resulted in a number of organizations taking down their sustainability claims.

And something that might be quite shocking, part of the reason why these guidelines were published was because actually when they did an investigation, over half of organizations had maybe some dubious claims about credentials. Or maybe it was slightly misleading or not very clear for the consumer.

We're definitely seeing, I know Europe is well ahead in this space but we're starting to see that a lot across ANZ and APAC as well. And this is something that I'm personally incredible mindful of. And I try to look at the bigger picture as well in terms of what's going to have the most impact. So for example, taking just a handful of long haul flights can completely reverse the good progress you're making when it comes to green code and green software.

So I think it's really important not to lose sight of that. For us, because we are a technology company, it does make perfect sense to actually focus on reducing emissions of our technology stack because the reality is, I'm not kidding when I say over half of our emissions do come from cloud technology.

And if we're serious about meeting our decarbonization goals, we do have to put our product and tech front and center. And, and yeah, like one, one of the ways how we're tracking this is through a metric called Carbon to Serve. And. I think Salesforce actually coined it, so if you're not fully aware what we mean by that, it's essentially how we measure the quantity of greenhouse gas emissions that it takes to deliver a specific Salesforce service or product.

And the reason why we have this as well, because actually in the beginning we had a lot of products and services on first party data centers, but that's changing. Like we're moving more and more of that into the public cloud, but it's very important for in those early days to actually understand what can we do to make improvements here as well.

And since establishing that metric, uh, in 2020, we have reduced it by 26. So there is good work being done, but we hope with the whole green code movement that we can really not just look at optimization of your infrastructure, but also look at software itself. And going back to just, I know we touched on a few different things here, but just a question on trust.

I can't echo enough, just, that piece around, especially in the guidelines anywhere on greenwashing or things like that, they say time and time again, just make sure that you're crystal clear in terms of communication and you tell the whole story. So maybe you're working with a customer on their green code or green software, but if that is just a small, I think it's important not to lose sight of, okay, can we help them tackle that part as well? An example of how I try to embody those principles, so whenever I work with customers to help them understand um, MuleSoft platform emissions, for example, share every single piece of source, my methodo methodology, rationale.

I'm very

Chris Adams: Yeah.

Tereze Gaile: explicitly clear that this is also baby steps. Like, this is like the first time we're doing this and we want to evolve it. We want to automate it. Um, and it might not be perfect, but that's okay. We'd much rather have something that at least you're consistently tracking something, so then you can see "ok, if I adopt this practice, what's the actual impact?" And it's a start.

Chris Adams: Okay. Thanks for being so transparent and being so honest about this, because this is a thing that we've, we found as a nonprofit where we work. So we did some research on essentially tech companies and tech companies and net zero targets and things like that. And generally we found that in the top 20 companies that we saw that made up the internet usage, we basically found we had a real hard time finding information about companies talking about having absolute reduction targets between now and 2030. This stuff is not published by most large organizations. In fact, BP, the oil company, the, the company you might associate with, like feeling guilty about your carbon footprint, these guys are more transparent than almost every single tech company we had inside this. And I will share the links to this 'cause it basically means that, yeah, most tech companies are failing the BP test, where the BP test is basically reporting your Planned Reductions in Absolute Terms by 2025 and 2030. This is, feels like a fairly clear thing to be asking for that we don't see in lots of places. And I'm glad you actually mentioned that. Yeah. Okay. We have some of this and as I understand it, the Carbon to Serve metric, that's a bit like the Software Carbon Intensity metric in that it's a relative metric, right?

So it's not absolute, but it's saying "this is what we do and if we increase that we'll be able to track that" with the idea being that tracking that will result in absolute emissions over time. For example, if that's high enough that should balance out any kind of growth you might have in say activity or something like that.

That's the general idea behind that. Okay. Thanks for that.

we just come up with the last? Yeah. Okay. We'll come up to the last question now. All right. So we spoke about some of this tension inside this, and we've spoken a few times about the idea that for the most part, technologists haven't really had an incentive to like, a explicit incentive inside their organization to develop knowledge around sustainability. So this might take the form of not really being, having a solid foundation when talking about this. And like you said before, It can lead people to over index on the importance of code versus something else. So like you mentioned it being half of your organization, I know lots of consulting firms. About 40 percent I know that among the big four consulting firms, around a third to 40 percent of the emissions come from not so much their use of code, but them flying to speak to clients about code, for example.

So there is an idea that you need to be aware of and need to be thinking about some of these parts here. And I think part of this comes down to the fact that as a technologist, you've got to learn about design patterns, about how efficiency, because that's going to help in your career path, for example. And we do see a kind of growing interest in green software, so it does feel like there is now a kind of growing set of incentives to start skilling up as a technologist in green software, understanding that there's a sustainability aspect that we need to speak to. And I wanted to ask you, are there any resources that you'd recommend or you've seen people use?

Because before this call, I saw that you've got some various kind of qualifications you've listed, but you've also being very hands-on yourself, trying to create some of this stuff. So I figured I'd ask you about some of that, because that feels like a nice way for people who are listening to figure out, "okay, what can I learn for free?

Or what are the examples I can start with so I can start building some of that knowledge up so that I can either interrogate these claims or even understand what's going to be most effective in my particular context?"

hmm.

Tereze Gaile: so the first one I'll mention, and genuinely this is not because I'm on environmental variables,

but The Linux Foundation for Green Software for Practitioners is awesome. So I did this when it first came out, but actually it's had such a good response internally that we're trying to work out a way to essentially get all of our engineering and technology organization to also sit this course. And again, it's not a huge amount of hours,

but it really nicely articulates the core concepts around carbon awareness, how things like energy efficiency, carbon efficiency, hardware efficiency, like how they're uh, connected and so forth.

So genuinely is, it is something that we're actually in conversations at the moment to say, "okay, how can we incentivize or encourage our own teams, and to have a look at this?" I did mention earlier a little bit about Trailhead, which is our own learning platform, and it does have content on sustainability, but we also created a very short module on Green Code, which at a high level, essentially covers some of the areas that you saw in the guide.

But I think from a personal level, I've really benefited from some that pure sustainability courses where they also teach you some of the frameworks for creating change in an organization, so the amoeba model for change, for example, or even having just that base understanding of the state of the world, how human activity's impacting global warming, and all this kind of stuff, and just like some of those frameworks that come with it, because then at least you feel equipped going into a conversation, and you know what the three scopes are for greenhouse gas protocol, and you can articulate what some of the regulation is across, at least at a high level, you don't have to be an expert, but just having some of those, you know, in your toolkit, I think it's really valuable.

Chris Adams: Okay, cool. So you mentioned a couple of things I've never heard of before. The amoeba model for change is something that was totally new to me, and I think we're going to have to add it to the show notes. And you spoke a little bit about scoped emissions, for example, which is something related to the greenhouse gas protocol, which is like the most common way of measuring organizational emissions.

We'll share some links to that as well. All right. I think we've covered quite a lot of stuff in this conversation, actually, Tereze. Before we wrap up, is there anything you'd want to discuss or just share with people before we finish up on this one?

Tereze Gaile: Final words for me, uh, being in this space can be really tough at times. So this is just a friendly reminder to look after yourself and a healthy dose of Outrage and Optimism.

Chris Adams: All right. Okay. Thank you very much for that, Tereze. We'll also share something about Outrage and Optimism as well, because that's actually something I'm really glad you mentioned that. That's one of my favorite podcasts, and it's really useful for understanding this wider thing that you mentioned as well. Thanks, Tereze. This was a really fun conversation. I'm glad we're able to actually make it despite being on other sides of the earth. Cool.

Tereze Gaile: Thanks, Chris. I really enjoyed myself and I hope there are a few nuggets in there that resonate with your fantastic audience.

Chris Adams: All right. See you around, Tereze.

Tereze Gaile: Cheers, Chris. Bye.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again, and see you in the next episode!

The Week in Green Software: Automating the Software Carbon Index

12 oktober 2023 | 39 min

This Week in Green Software brings a conversation with Arne Tarara, CEO of Green Coding Berlin. Join host Chris Adams as they discuss the new programming language Mojo and its efficiency, as well as other news from the world of sustainable software development. Together they share their thoughts on progress in Grid Forecasting in order to reduce high-carbon electricity use, in particular a new announcement made by Apple in this field. Finally, Arne and Chris share what’s on their wishlist for green software in this latest episode of TWiGS.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Arne Tarara: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Modular: Mojo🔥 - It’s finally here! [4:22]
How engineers can make IT more sustainable | Scaleway [10:44]
SCI (Green Software Foundation) [16:50]
Apple brings Grid Forecast to Apple Home to tell you when you’re using clean energy - The Verge [21:18]
Wagtail 5.1 gets a bit greener and leaner [24:31]
Detecting Green Impact in Source Code - Meet Greg Rivera, VP of Product at CAST [29:03]

Events:

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Arne Tarara: You can also say, "why don't you use C?" Right? This would be the direct response to people. And then there's always people saying, "yeah, but Python developers are easier to get. The development speed is so much faster. So obviously you have a time to product, but you also have a time to fix." Like, software sustainability is not only about green stuff, it's also about maintainability.

Hello, and welcome to This Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. In this episode, we're going to talk about working out the carbon footprint of any open source project on GitHub, myths around decarbonizing cloud platforms, and how to make your code 68, 000 times more efficient with one weird trick. But before we dive in, let me introduce my guest for this episode of This Week in Green Software. With us today, around the corner, we have Arne Tarara. Hey, Arne.

Arne Tarara: Hey, Chris. Yeah. Hi, I'm Arne Tarara. Uh, I'm the CEO at Green Coding Berlin, and I'm super excited to be on the show. Some regular listeners might've heard the name before, very often in a slightly, I would say more humoristic sense, because the name is more expressive than it is creative and I heard Asim lately making a pun about it, but in all fairness, he was not the first.

But on the flip side, the name says exactly what we're doing on a daily basis. Like we are a very small team. We have four people of senior engineers that work on open source tools that allow companies and engineers to measure and quantify carbon and energy costs of software. I think our most relevant software, the Green Metrics Tool, we will talk about later, but apart from that, to give a bit more insight, what we're doing is we consult companies and also work in government initiatives like, for instance, the Germany's Blue Angel Certification for Software, which I think was also slightly mentioned in the last episode, and with software open source communities like, for instance, Wagtail, that I know Chris, you're also involved with, and advancing the software industry as a whole through the open source tools and research that we're doing.

Chris Adams: Cool. Thanks for joining us. Arne, just before we dive into the world of technology, how was your weekend anyway?

Arne Tarara: We have the luck currently in Berlin, as you probably also know, that it's very sunny. We just... We had this very rainy phase before, and so my partner and I ventured out into the Tempelhofer Feld, which is this super big area. I have to make a guess, but I would say it's something like one or two square kilometers at least.

It's really big, so you can sometimes not really see the end of the other side, because there might even be some mist, or I don't know what's going on. And there was this kite festival, so there was this advertisement in the newspaper about it, and I was directly hooked. You get these 50 meter long kites.

It's, well, something like a bear or something like a, um, I actually don't know what this animal is called in English, but we call it a rochen, I think a stingray is the English term. So it's like super cool looking, like you look up and 50 percent of your view field is full of this kite. It was really nice and obviously lots of beer, as usual in Germany, and lots of food.

So it was very nice venturing out, slight sunburn was, I had to take in, but apart from that, super nice weekend.

Chris Adams: Cool. Good stuff. Nice to hear. Okay. If you are new to this podcast, what we typically do is we just run through some stories that we've been collecting over the week and we share some of the kind of takes and reckons on this. What we also do is any links or any stories or things we discuss, we share in the show notes for the podcasts that are going to be available on podcast.greensoftware.foundation. So that's the plan. And I should, just before we go, I should also introduce myself. Sorry. My name is Chris Adams. I am the executive director of the Green Web Foundation, a nonprofit working towards an entirely fossil free internet by 2030. And I'm also one of the chairs of the Policy Working Group. Okay, Anne, are you sitting comfortably?

Arne Tarara: Perfect. I'm ready to go.

Chris Adams: In that case, let's begin. Okay, let's look at the first story. The first story we have here is actually the reference for this 68, 000 times faster, uh, trick. Uh, this is about the new programming language called Mojo, which has been addressed, has been published, and is largely, as I understand it, it's like a really, it's like a, a fork, uh, of a much faster fork of Python. So per- Python might been one of the slower languages. There are some clever things that this language does to make it much faster. And Arne, I I know that you're a bit of a Python coder. I figured if there's anything I, I was, when you read about this story, what kind of springs to mind, maybe you could share a little bit or share a few reckons on this one.

Arne Tarara: Mm-hmm.

Yeah, sure. You sent me the headlines before, so I can at least have a quick look into it. And I was also directly surprised 68, times, never seen this before. I think the general term is every... In German, we have this saying, "every statistic is infation." But I hope everybody gets what I mean by that. So a benchmark always only works with the boundary conditions you give it and obviously everybody gets a different result. Nevertheless, what you hear very often if you're in this field for a bit of time is this one study from the Portuguese people who made this Energy Efficiency across Programming Languages, and C was the baseline, which is very often said it's, it's the fastest language because it's also so simple and so near to the OS itself.

And then Python clocked in, I think at 75 times, um, worse than, the task that they gave to C. It's always a bit biased. It was only compute, like no network I/O or nothing. But again, coming back to the topic here in particular, 68, 000 times, was even a bit surprising to me, like, where do they leverage all this performance, and it's actually a three part blog series, and if you, I think it's in the second blog post where they highlight what they have actually been doing, it's like the lift up they get from using just Mojo is something like in the, in the 80 times or 90 times.

So it's not that super different, at least from this initial Portuguese paper. And then they leverage functionalities that you typically can't use in, in Python directly. So they use these vector instructions. Some people you know them as, um, uh, as SSE or I think MMX was also a vector instruction at the time and the.

Yeah, we're still in the 486 processors, so way, way back, uh, but it's basically the same, um, structured set. So you have, you have a broader register set, you can do multiple calculations at the same time and they're getting bigger and bigger over time. So then they feed it in there and then I think they get something like a couple of hundreds and then they parallelize it.

So we just, in the end, the trick, so they use an 88 core processor or 88 thread processor. And then they get 88 times the boost out of it and at some point you arrive, I think over 10 iterations or so, they then arrived at the 68, 000

Chris Adams: This incredibly large number.

So for people who are Python programmers, this, and you're, and you think about the efficiency of the code that you're using, it's both. You can think of this as like Python plus, and, uh, this isn't open source yet, although the organization and one of the people who's really pushing this does have a track record of open sourcing tools.

So the people who worked on the Swift language have also been working on this part here. And as I understand it, the general idea is to make it as. Essentially take some Python code and do a lot of clever stuff at the compiler level. So as a programmer, you don't necessarily need to do loads and loads of really, know too much about the internals of, of a program.

You let the compiler do a bunch of really clever stuff, which ends up speeding up things impressively. Whether this is, this is also something in my view, there is a bit gimmicky in that sense that this is the only programming, programming language or piece of software I've ever seen where instead of having like .py or .js, they have a fire emoji as the way to actually identify some code. And I'm not sure if there really is a kind of actual more human readable version of this as well, or easier to type kind of extension, but that kind of gives you. It, it, it seemed interesting enough to share with other people and whether some of these ideas will make it into Python mainstream, we, it'll be really interesting because Python for the longest time has been maligned as a very slow language. It does give this idea that, yeah, if you make changes at the kind of compiler level, maybe we don't all need to become green software specialists to actually realize some kind of meaningful reductions in the actual energy usage of the stuff we're using.

Arne Tarara: Yeah, I'm totally on board with this one and I think it's also one of the more interesting and fancy news to share first because it has this major full mouthed promise that you bring with it. Python, the, what I think I can also bring to the table here in terms of discussion because I had it so often is, you can also say, "why don't you use C?" Right?

This would the direct response to people. And then there's always people saying, yeah, but Python developers are easier to get. The development speed is so much faster. So obviously you have a time to product, but you also have a time to fix like software. Sustainability is not only about green stuff, it's also about maintainability.

And I think we'll come later into the podcast to this topic because this is one of our research field. It's like software lifecycle. Looking at the software lifecycle in particular, you can only really save Python or C is better for you as a company if you also know. How much does the development cost you in terms of touching code again, in terms of getting developers, et cetera,

also in cost of runtime.

And I think we'll bring up the notes in there. When we talk about it, but one, maybe what I have here, because you said it's not open source, what I've seen lately, it was on a German news blog, I will put the German link in here,

is something that is basically, I would say the underlying part of it, like the optimizations that are applied in Mojo are automatic.

But if you want to know where your code is slow and what you can do on a programmer level to also advance yourself as a programmer, there is this, I think it's from the MIT, it's like an open source thing that scans your code in Python and gives you the recommendations that you need to do, which means be offloading to the graphics card or to CUDA, or which would be touching code that is really badly written or is just worse in Python.

Chris Adams: Maybe this will be a thing that shows up in Copilot or various code rewriting tools that seem be making up into people's different environments for writing applications these days. All right, thanks for that, Arne. Should we look at the next story that we have here?

Arne Tarara: Sure.

Chris Adams: Okay. So this one is, uh, from Scaleway, which is one of the larger French, uh, cloud hosting firms. They have, they have a four-part series specifically about how engineers can make IT more sustainable. And there's a couple of really, in my view, quite interesting tidbits that come out of this based on how they've seen people using their own infrastructure. Arne, was there anything that jumped out at you when you were looking at this stuff in particular?

Arne Tarara: Yeah, actually, I have been following the Scaleway series for a while because, um, as a CEO of a green coding company, I have the most full fledged thing you can do. I have a Google alert that alerts me about new green coding topics. So this one comes up often, like Scaleway is doing actually some PR work in the sector and, um, from what I know from the outside, they seem to be, yeah, very authentic about what they are doing in terms of the data center.

And yeah, they have these series in order to highlight new tips to developers, what should be looked at, or what are typically best practices, et cetera. And what they have been talking about is a study from Google in this latest post where they say that Google with their cloud product assumes, I have to look it up in the article, but it was something like 60, 60 gigatons, 60 megatons of CO2, which just is happening from machines that are idling or storage that is not even read or written to for a while.

And Google is actively, sorry, is actively talking to these customers through There's also something like a notification in the interface that says, "Hey, you have to stale services. They also incur costs, but more problematic for, for us as a whole is they also occur a lot of carbon. Do you still need them?"

Which is obviously a nice story to tell from Google because they're slightly cutting to their own flesh because obviously for storage, they also get money, but their concern seems to be bigger and they're going for the green route. And these numbers are so large very often, right? Like if you, if you look at some tools and they, then you have this milliwatts or something that you're looking at and somebody talks about tonnes of carbon, um, yeah, it directly, I also find it directly interesting. But on the flip side, what we have been looking at into a research project, was leveraging the YouTube API and basically looking at all the videos that nobody's

Chris Adams: Oh,

Arne Tarara: because there is, I would say there is more in terms of dead storage, so to say, that nobody's accessing, but it might be the case that somebody accesses it at any time and then Google can, like putting the bad complaining hat on here, then Google can still play out an advertisement.

So maybe this is why they don't clear. And isn't that actually a proverb om English? Like, "sweeping in your own front yard."

Chris Adams: Sweeping your own front yard is, I think there is, we have that content with that, that, that concept. I'm not familiar with the term. Can I just check with you? So one of the things you've mentioned is this thing that Google is doing, which essentially sounds like the cloud version of basically telling people, "hey, you left your lights on when you left, when you left the office today."

So they're telling people that they're leaving things on and that you may be, you know, "do you really need that light to be on?" Essentially. That's one thing they're doing here. And there's this other part, which is that you folks have been looking at which videos are on YouTube, are getting the, getting the views. Is that what you're referring to just then, Arne?

Arne Tarara: The idea was to, um, we never actually executed the research project, so the idea was to query the YouTube API for videos that have, have been uploaded a year ago, but have two views or so, then kind of get an estimation of how many dead videos might there be on YouTube, and then through some papers that have these calculation formulas, like how much that storage costs, which is somewhere on a hard disk or on a tape or whatever, and then just make an estimation how bad it is and make an article about it.

Yes, something like this. There are different variants you can go, but I think this is the gist.

Chris Adams: Ah, do you know what? That's a very, you've raised a good point about what you do with very infrequently accessed information, and what, how that should be stored. We'll come back to that a little bit later, because there are some guidelines around moving, this is actually one thing that I saw in the Scaleway piece, was basically this idea that if you've got storage, there are ways that you can store it in a way which is going to match it to how often you think something's going to be used.

So if you store something in RAM, for example, really easy to get, but, or if you store something in say object storage, you can put something in a colder form of storage. That's where you might not access it quite so quickly, but it's going to be a lot cheaper and you can go all the way down to storing things on tape, which basically doesn't even need running power at all in many.

So yeah, there are some ideas like that. Okay, cool. Should we look at the next, the next story we had here actually, Arne?

Arne Tarara: Yeah, absolutely. So I think you still mean in the Scaleway article, right? Because it has very many bullet points.

Chris Adams: yeah.

Arne Tarara: The second one that I found most interesting is they talk about this software life cycle and how this is a topic of concern and soon coming because they are a EU company and we have this corporate sustainability reporting

directive.

Chris Adams: CSRD, the

Corporate Sustainability Reporting Directive that lands at the end of 2024.

Arne Tarara: Yeah. So, so they are talking about this because as a data center, they are, I think in two ways affected, like they are affected directly and also as a supplier because they host their services for so many companies in there. As I said, this is a, an upcoming thing. So methods will be standardized, how you even look at this stuff, but also software lifecycle will become a different view, I would say, as a developer.

So you not only can say, "I'm looking at what is my software using in the runtime phase." But also if you are a container provider like they for instance are, there is something like a boot phase or there's something like an installation phase if you want to make up phases that are specifically for software.

But if you go through the classic means of a lifecycle assessment, you have something like a removal phase

what does it cost you to remove stuff from caches? You have something like a development phase. So how much does it cost you to develop the software and also to test the software? So how much do the pipelines cost?

And I found it super interesting that this is now, and this is all pushed forward by one of the bigger data centers because there's so much entailed with the work that we do, for instance, like if you, we mentioned the green or I mentioned the Green Metrics Tool earlier, which is, say our flagship product. This is, in essence, I would say a benchmarking software that has this lifecycle approach integrated because you have to provide it with infrastructure and code, so something like a Docker Compose file for instance, and then it will spin up the containers, it will install all the packages that are in the Docker files or whatever you have, and then it will coin this as the, as the boot phase when the container started, and before that when the containers are created on the system with all the downloading, it will coin it as the installation phase. And even before that, it will also have an idling phase, so it will also tell you what your system is doing, if nothing is happening in particular. Then you go through the classic runtime. So the software is run against whatever it has to do, so if it's a number crunching then it will do that, but if you have a more elaborate scenario, let's say you have something that also GreenFrame uses, which I think was also mentioned on the podcast before, so something like an end-to-end test.

So you have a Selenium test where you instruct the browser, go to this page, click here, upload a form, query an API, behave like a user, so to say. So this is then the runtime, and then later on everything's removed, which is mostly clearing databases and removing caches. And this is the small window that you typically come across when you interface with the software, uh, and when you run the software.

And in order to also give a holistic view for developers and enterprises, then you need something like our pipeline tools. Like we have tools that you can

plug into GitHub and GitLab, and then you see what your pipeline costs on GitHub, for instance, in terms of carbon and CO2. And for development tool, I hope we have something to share soon because this is still internal R&D.

It will be something like a small tool that you have in your top taskbar and it will always scan all the processes you have running on the system and then match them to the software you're currently developing. So every time you change your GitHub repository on the command line, it will attribute it to a different project that you're developing on.

All the auxiliary products will be split on and I think there will be a lot of discussion coming up because the question for me that I have for you is: when you are coding, would you say your Spotify that is running should be attributed to the software development cost or is it something that you should cater for privately, right?

Although Spotify might help you concentrating, this is a discussion to be had, I think, but I think the interesting part is, I haven't seen any tool out there that does something like this, that works alongside a developer on its own machine, maps it to the project, and then integrates it with a bigger picture of the pipeline, runtime, installation, etc. to give a full picture of the software.

Chris Adams: So this thing here of expanding what people talk about when they measure it to look at the life cycle at the beginning and the end. This is actually... I think this actually may be one of the next kind of frontiers of what we have, because when a lot of the time people are talking about, there is this kind of risk, kind of received wisdom that it makes a lot of sense to use serverless tools because you switch things off all the time, you're only paying for the actual requests you're using. And this is kind of this idea that that's going to be clearly the greenest way to think about this. Obviously, there's going to be an impact associated with keeping caches warm so you can serve responses and everything like that. Just like you said. I think. This is interesting, I'm looking forward to seeing some of this come out the door, actually, because the example you mentioned of, does the music you're listening to, does that count when you're writing some software, the only organization I can think of that's been including anything like that so far, is Mozilla, when they started trying to work out the environmental impact of the Firefox browser itself, where 98 percent of the environmental impact of their, of their reported organizational footprint was associated with the end-users. And for that, they basically looked at the entire computer being used. So rather than actually splitting up to say, here's just Firefox's part, they looked at the entire machine thinking, well, you need an entire computer to be using this. So that's how we're going to report that. And whether they're going to do that going forward is another matter, because if you have a way to cut down the biggest chunk by a significant amount, I can totally see a lot of effort and a lot of temptation to do that, especially when other organizations aren't currently reporting that, for example. So yeah, thanks for the thought provoking point, actually, there, Arne. Okay, shall we have a look at one of the next stories that came

Arne Tarara: Sure, uh, do you wanna, do you wanna go through and then I chip in,

Chris Adams: Yeah. Yeah. Okay, Arne, next story. This one is from Apple. They made a big splashy release last week about A, unveiling their first carbon neutral Apple watch, but also they've done some work to start mainstreaming grid forecasting. Have you, did you get a chance to look at any of this stuff? Cause I think this is really interesting seeing such a large company really try to start talking about this, especially when an organization is based in the home, but also to have it run through the entire system. I want to see how you thought about some of this because Apple has generally been quite a leader on a bunch of this stuff, but I know there's been some pushback in a number of sustainability circles talking about the idea of branding a piece of electronics as carbon neutral these days.

Arne Tarara: Yeah, I mean, I got the article actually from a co-worker who is a big Apple fanboy, if you want to put it like so, but saying that, I also use myself a Mac because I just think it's the best machine for me. And I was actually pretty happy about the news statement, I looked into the official Apple statement directly, so what they had in their newsroom, and although I don't use an Apple Watch, I think all the moves they had made total sense, like they switched the armband, which was leather, for something which is way more sustainable but also carbon neutral to produce in the end. I think they cut down on the cost for making the casing and then on top of that the whole story spun around the the Apple Watch, as I read it, at least, on top of that they introduced this functionality that you can actually look when is the best time to charge all your other electronic devices.

So you have it basically in arm's reach to save carbon, so to say. And I found this pretty cool because also the GSF had this hackathon, I think in 2022, if I'm not mistaken, to also use the open source SDK in order to, in order to get the grid intensity and write an application around it. And I found it very lacking that in my operating system, I don't know, for instance, when to plug in the laptop or when compute intensive task.

So we handed in at the time a small tool where you, whenever you run an intensive task, it will give you an info that this resource is currently blocked, like a network, for instance, and you can unblock it if you're on a green carbon or not. And I think this gives a very strong intention to people and drives this whole topic forward.

The downside that I've read, because you mentioned some people give pushbacks on this, is that obviously you can make the argument that Apple, as a big company with so many money reserves, can do so much more, and they focus on this small Apple Watch. But if you, I think the podcast is also about opinions and if you ask my opinion, I think it's a really cool way to do it.

They probably can do more, but it's still the best that I've seen lately from a big company because I know how slow enterprises can be and how heavy it is to push such a project. So, yeah, for me, it's a positive news, I have to say.

Chris Adams: Okay, cool. Thank you for this. All right. So we've shared a couple of links to both, uh, the wide story about why this won't be carbon neutral, as well as why this is considered carbon neutral, as well as talking about some of the most recent things they've done about essentially exposing the grid intensity or, um, in America to pins.

So, so you can basically make somewhat more informed decisions about when and how you use various kinds of electronics. Okay. All right. I'm going to come for one story, which is actually. This is something that, this is the culmination of some work with the Google Summer of Code. This is from Wagtail. So there's a story about Wagtail 5.1 gets greener and leaner. This one here is, this is basically a follow on from a project that, um, we did where we work. We did a works with a open source project called Wagtail. It's a very, it's a quite well known piece of software that's used on the NASA website, on Google's blog, on Mozilla's website, for example. And, um, we basically had a early career technologist work to include AVIF encoding into the actual platform. Because when Wagtail looked at all of their, the emissions associated with the work that they're doing, from the software point of view, the images was the biggest thing for them to work on. And this got merged into the project. And I figure this might be one that is, might be worth talking about because I know that we were using, uh, the sustainable web design methodology to work this out, but there's some other work we're doing now for trying to understand the environmental impact of the project at various other places. And the other being is if you work on a high profile Python project here, then you've got a kind of reference, reference for other projects to follow on from here. So Arne, anything catches your eye or any reckons that you're in for this one here? Because. We did start looking at some of the code that you had and we were using, we were looking at a Green Metrics Tool to come up with some green metrics with a tool,

Arne Tarara: Yeah, true to the name, so to say. Yeah, I'm actually super excited about the project. Can you, can you actually tell me right away, because I think this is also the most interesting for the listeners, in which version will it be integrated?

Chris Adams: It's already out, so we started in summer, uh, uh, we got the first part finished and merged in. And, uh, then it was released, I think in, I think it was in the middle of August is when it came out actually. So the release date for this was 1st of August. So the project started in, I think, midway through June and, uh, then yeah, it got, it was merged in time to make it to the release, uh, and then came out in the 1st of August. There's a bunch of other part on there, and if you look at the Wagtail sustainability kind of section in the docs, it's really good and it has a bunch of really useful guidance. And there's also a roadmap of new things uh, following on from this, like different places you can scale down the project with the idea being that if you got something here, then it can be adopted by other similar projects who are trying to adopt this 'cause you do see, you, you do start to see some open source projects trying to optimize for this. Like WordPress is an example of doing things in this field as well now.

Arne Tarara: Yeah, I think I see it here now, because my question was mostly related to the version number of the Wagtail, and I think 5.1,

Chris Adams: That's correct.

Arne Tarara: it, yeah, sorry, I did overlook it first, but now I see it, because you pinged us, like, a year ago or so, and we had this mini project with Thibaud, which is one of the main maintainers of Wagtail, by creating something like a, Wagtail comes along with these super nice bakeries, so it has like a reference implementation. We were super interested to look at it with the Green Metrics Tool to just see, "okay, how much does a typical visit to a Wagtail site cost?" And then Thibaud implemented all these pages, a contact page, he implemented a caching system.

You could go to a list page and whatever page types Wagtail has. But I think our project is still on the 4. branch, so we have to update that in order to look at the whole AVIF thing, because the interesting point that I see here, and I think this is also what, what I think Wagtail wanted to do in particular, is to estimate also the net gains of it, because

Chris Adams: AVIF

Arne Tarara: is obviously better, this is why it was developed, and it's also way newer than JPEG, right?

So in all fairness, it's just a newer standard, so chances are high that it can be more efficient. But the question is, "at what cost does this come along?" Because AVIF is a stronger compression, it needs more CPU, it might need more memory to do it, but on the, on the gain side, you then save the network transfer, but is it really worth the cost implementing that?

Chris Adams: This is exactly the thing. Yeah.

Arne Tarara: Yeah, and we had this software lifecycle topic before. I think this touches all in the same domain. You have to broaden your picture as wide as you can in order to make the boundary there where the optimizations or losses that you are creating through the, um, code that you write are not outside of this boundary and you create a false closed picture of it.

Um, this is actually super interesting and we really have to update our code so let me write it right now and then I'll write Thibaud an email right after the call if we can get this pushed to the 5.1 branch and then make the measurements available as soon as we can, because we have this in so many other places.

Have you, for instance, read the article on the Green Software article blog recently about a super cool product from Cast? I hope I pronounced it correctly. They are a member of the Green Software Foundation and they have developed basically a tool that can look at a code repository, look at a running piece of code also, it's like an agent you can plug into.

At least that's how I understood it. So there might be small bits and pieces that I didn't get correct. But the end result is you get a dashboard with so many green coding optimizations that you can do.

Chris Adams: Hmm.

Arne Tarara: there's also another project, I think it's called EcoCode, which is Python, JavaScript, supports a couple of languages and taps into SonarQube, like this many enterprises I see use it, typical static code analyzer.

But the question that I have always, when I use these tools is, "how much costs do these tools need in order for the optimizations that they give you?" We also contacted Cast about it and we had a video call and we pitched them a joint research project about it and yeah, this is still, this contact is still in the works if that is going to happen or not, but I think this is a very interesting thing to get all these recommendations and then you can do 50 recommendations on it.

You have to sync in 20 developer hours, your product gets 1. 0.1 percent better. But SonarQube used, I don't know, five, five kilowatts of energy. Too much, obviously, now, but a tough question to ask because sometimes the optimizations might not even worth taking you don't save anything in the long run, right?

This is why I'm super interested in the AVIF case because I think this will be a net gain and it's even better if you can underlie it with some hard numbers.

Chris Adams: This is exactly the reason, this is actually where the next step for some of this is to, after doing this work, there were some trade offs associated with this, and we'll share a link to some of the issues where we explore some of this. One of the key things is that, okay, we might have made this part a bit more efficient. But have we just shifted all the work onto an end-user's device, and are we making them do the work for this now? This is, this really speaks to this idea of, okay, where in the boundary does this actually take place? Cause if you assume that you have some responsibility, then you do have it, but you can end up with a scenario where just by making your thing super efficient, you just push all the load onto someone else's kind of balance sheet, as it were. This is some of the problems when we think about carbon accounting that we need to kind of address, especially in the kind of digital realm, because yeah, how you design a system can very much change whose computers are doing all the hard work to actually make an experience possible. But yeah, I'll happily share the links for that because there's a bunch of stuff. I don't know of many other projects that are in the open that are doing this right now, and it wouldn't be that difficult to actually get the numbers out because there's actually decent data set available for that already. Okay. All right then. I think we're coming up to time. So maybe we'll just do a quick run through of some of the events that are coming up. And then Arne, I think, um, we might have just one, one, one little question and then we might wrap up actually. So the Green Software Foundation runs a series of meetups as well as having events in the future. And in the last month, we've seen a new number of new meetup groups open up. There's one in Karlsruhe, Germany. Arne, can you tell me what the deal with Karlsruhe is? Because I see it showing up in a few places and I've never been there. And I'm a little unclear cause I, I used to know of it as like, uh, old and previously industrial area. Is there like a science area behind it? Or is it like a tech, a real tech sector in the country?

Arne Tarara: Yeah, I actually can't tell you how it came to be. Karlsruhe is really... I know Aydin as a friend who is, who's running this meetup over there and he has a big software company where they develop software for other people. And he told me that the, how do you say, like the industry share of IT in Karlsruhe is above 50%.

I think I have 60 in my head, it might slightly more or less, which is crazy.

Chris Adams: So half companies are tech companies in this one city.

Arne Tarara: Yeah, and this also entails the money that flows into the city in particular through taxes and stuff. So IT is a big topic over there and they have something like the CyberForum, which is I think even Germany wide known, how do you say, group of people who are in something like a industry, yeah, I don't want to put a name on this one in particular, but the CyberForum is very big. So if you advertise something there, you reach a lot of people that also work for big enterprises and there's a lot of IT stuff going on there, which might be under the radar when people always hype Berlin or something as a hotspot for developers.

So it's a big area and he has very good topics on the meetup groups. I think there's even something out with the Cloud Native Sustainability Week. I think you also wanted to touch on that, right?

Chris Adams: Yeah, we'll touch on that. Okay. So there's, okay. Thanks for the kind of minor geography lesson there. Thank you, Arne. So we have, um, new meetup groups being opening up in Scotland, in Brighton, in Oslo, Norway, and Belfast, Ireland. So if you're in any of those countries or cities, that might be something for you. And the other thing you did mention, Arne, is this Cloud Native Sustainability Week. So the second week of October, we'll share a link to this, there's a series of events all through that, like really chock full of events for people who have an interest inside this. And then finally, on the 16th of November, there is the big event from the Green Software Foundation called Decarb. So if you go to decarb.greensoftware.foundation you can learn a little bit more about that event taking place there with a bunch of speakers. And if you have something that you want to share to an audience, the CFP is still open. That's what we have going on there. Arne, I've really enjoyed nerding out with you and a bunch of this stuff. We normally have a kind of little wrap up question towards the end. And I'll just share it with you to give us a nice kind of wrap up for this. So do you have a green tech wishlist? And if so, what's at the top of it for you these days?

Arne Tarara: Yeah, so my, we touched about this Apple topic before and I think I mentioned that I would really like to have the possibility in my operating system to directly see if any code that I can run now will be greener or less green because of the grid and everything that is entailed to it. So this would be something that I would really like to see Apple also integrate into macOS and also be super reliant on the data that I get.

But on the other side, we are working in an R&D company and we typically, as a team, develop the tools that we really want to have, so in a very beneficial position that we can usually

Chris Adams: scratch your own itches basically here.

Arne Tarara: so it's very often happening that at least we can do a prototype if we think a topic is interesting, but I think when I talk, look at enterprises and stuff like this, the easiest thing to do and which is very easy in reach is like combining the grid and the operating system to give developers the empowerment to choose at least on that.

Chris Adams: Alright, okay, so you said something interesting there. Right now, with cloud computing, I can do this really easily, right? Not really easily. It's doable, where you can get a, uh, hour by hour figure of the carbon intensity of the power you're using, so you can work out the average carbon intensity of an application or something, right? But I can't do that with my own laptop. I can't do that with my home. And I feel that, in the same way that you... If we're going to be doing, if we're going to talk, talk about how green the energy is, and we've, we've accepted that the energy will be greener at certain times of day, and less green at other times of day, then I don't know of any organizations that currently give me a figure of showing this was the average carbon intensity of the power you've used last month, or even every day. I would love that to exist, and I'm pretty sure that can be buildable, but I haven't seen it being done at a kind of personal or house level. That's my kind of wish list for someone to make, because it's totally buildable, but it doesn't exist yet, and I don't have the time to build it myself.

Alright! Right. We could, it's a doable thing.

Arne Tarara: totally know what you mean. I know you come initially from the UK and then just did you know that Octopus, which is I think UK based initially, and they already have that because the grid in the UK is so much more smart than the German one is, that they just came to Germany also. Have you had a notice of that?

Chris Adams: I did know about that and Octopus, they, so they are, they're one of the people who do expose this. And actually I didn't think about that, but they do, they may well have APIs that can expose that. So if you're listening to this podcast and you use Octopus, please do set, or if you work for Octopus, please do tell us if this is, it can actually be exposed because I think it'd be a nice bit of nerdy fun to share with people.

Arne Tarara: Yeah, super nice.

Chris Adams: All right, Arne, thank you very much for coming onto the show and thank you for working on all the various open source tools that we've used in the organization I work in. And yeah, have a lovely week and hopefully you get to enjoy some more kite surfing and gigantic kites in the Tempelhofer Feld of Berlin.

Arne Tarara: Chris, thank you so much for the invitation and also good week to you.

Chris Adams: Cheers, Arne. Okay, bye.

Arne Tarara: Ciao.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again, and see you in the next episode!

Decarbonize Software 2023 Preview with Namrata Narayan

5 oktober 2023 | 38 min

Join Chris Skipper and this week’s guest Namrata Narayan in discussing the upcoming Decarbonize Software 2023 event taking place this November, as well as the role Namrata plays in the GSF. In this episode, they cover the relationships between different member organizations and their role in green software and how they can work toward the same goals in a competitive environment. They touch on how and where this year’s Decarb event will take place and even how it has been set up to reduce its own carbon footprint. Hear about the planned sessions of the day and how to register in this episode of Environment Variables.

Learn more about our people:

Chris Skipper: LinkedIn
Namrata Narayan: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Decarbonize Software | [29:08]
Register for Decarbonize Software 2023 | [34:58]

Events:

Decarbonize Software 2023

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Namrata Narayan: If we see software as an agent for climate action, then we are going to be a lot more successful in articulating why software aligns with sustainability.

I'm your host, Chris Adams.

Chris Skipper: Welcome to this episode of Environment Variables. I'm your host, Chris Skipper. And in this episode, we will be discussing the upcoming Decarbonize Software 2023 event with the Green Software Foundation's Director of Communications and Member Relations, Namrata Narayan. Hello, Namrata, and welcome to Environment Variables.

Namrata Narayan: Hi Chris, excited to be here.

Chris Skipper: Cool. So before we dive into the meat of this podcast for our listeners, could you please introduce yourself?

Namrata Narayan: Sure. So, as you said, my name is Namrata Narayan and I really lead communications and member relations now at the Green Software Foundation. I've been with the foundation for about a year, actually more than a year.

Chris Skipper: That's amazing. And Just so people know, I'll say a little bit about myself. You might have heard my voice in this podcast before. I'm Chris. I'm the podcast producer. The other Chris is the host and I'm an absolute noob when it comes to green software. I don't have a software background at all. In fact, I have a musical background and a podcast background.

So, but yeah, so other than doing podcasts, I like drinking lots of coffee and at the moment avoiding all the cherry blossoms here in Australia because it is spring and I have a grass allergy. So there you And before we dive in, here's a reminder that everything we talk about will be linked in the show notes below this episode.

So before we talk about Decarbonize Software 2023. Let's chat a little bit about yourself. So what does being the Director of Communications and Member Relations in the GSF entail? And what are your responsibilities and goals in this role, Namrata?

Namrata Narayan: This is a role that was just recently created. I was doing this work for the last few months and it just made sense to formalize it and make sure that everybody in the community knew who to reach out to and in what capacity. So my role is really focused on building awareness about green software, the work we're doing at the foundation, and nurturing relationships with our member organizations and their people to support knowledge sharing, participation in our projects, and ultimately really support a culture change towards sustainability in tech. So in nutshell, I oversee content marketing, run our social media and weekly newsletter, advise our leadership and working group teams on branding and messaging and identify opportunities to support our members within the context of our projects and initiatives.

Chris Skipper: Awesome. And so you've been with the foundation for just over a year. Have you seen, in your role with marketing and promoting the growth of the community, have you seen a change, A, in the number of members that have come into the Green Software Foundation, and B, in the type of members that have come into the foundation?

Namrata Narayan: Yeah, that's a good question. I certainly have. I think since I started working with the foundation, we welcomed probably 10 new members at this point. They're all different shapes and sizes, which I love because it really speaks to the fact that this is an issue that everybody should be concerned about, and all types of teams should be working on, it's not just a problem for the big giants, but also for small engineering teams, for consultancies, for service providers.

So we've got a really nice variety of members that we're now working with.

Chris Skipper: Yeah. Awesome. Yeah. So I think I probably came on around about the same time as you then, because Environment Variables is just over a year old, or maybe I think it's been longer than that because there was a lot of planning for this podcast to actually go ahead and then eventually Asim got around to recording something and sent it to me.

But yeah. So, and I'm not as in touch with members, I would say, as yourself probably, but I've definitely seen global membership the Green Software Foundation. For context for listeners, I'm currently in Australia, but Namrata is sitting in Canada and that's just within the people who work for the Green Software Foundation.

But we have had people as far away as Japan and I think, I believe even South America attend events. So yeah, it's very much global. I don't need to, I'm preaching to the choir here. I've forgotten how the term is, yeah, but people, people who are listening to this podcast are probably members of the Green Software Foundation.

And if you aren't already, you should probably join. How easy is it to join the Green Software Foundation?

Namrata Narayan: We've got a few different tiers of membership. So based on what organization, yeah, what they can do, they can either come at the steering committee level or at the general level, there's, we have a standard agreement because we want to treat all of our members equally. And yeah, it's relatively straightforward and as soon as they become members, as soon as everything's signed and sealed, then that's really when they get passed on to me and I support the onboarding process, which we've now made a lot better, there's still definitely room for improvement and we're looking forward to just making it even more seamless, but are able now to provide so much information right off the bat in terms of what we do, how they, how different, how different people within member organizations can get involved, where they can find all the information.

I mean, it's all available and ready for the taking.

Chris Skipper: Great. I think one of the best inductions into the Green Software Foundation, if you're not familiar with it and you've just stumbled upon this podcast episode in the riches of the amount of podcasts that there are on the internet, is actually the Green Software Foundation newsletter, which is actually one of the sources for a lot of the material that goes on to Environment Variables.

The newsletter is fantastic. It's a fantastic resource. Let's talk about more about your green software journey. When did you first encounter green software and how did you come to find yourself at the Green Software Foundation?

Namrata Narayan: So it's actually quite a serendipitous story. I've been working in the sustainability and SDG space for over a decade. I've worked on, I've worked with mostly nonprofits, not-for-profits, think tanks, et cetera, that are really focused on meeting one or more of the sustainability development goals. And then a couple of years ago, I decided to start my own practice so that I could work with more organizations as opposed to just one at a time. And soon after I launched my own company I got approached by the Green Software Foundation leading up to their first ever hackathon called Carbon Hack that took place last year, I believe in October, and they were looking for additional support, and I had never even heard of green software. I had some idea of how we could be more sustainable with our sort of, in our digital practice, but the term, the concept, the theory was all quite new to me. So I just found it so interesting. It aligned so much with what I care about that I really didn't give it too much thought. I said, "yes, how can I help?" Um, and that was a really wonderful experience. It was a deep dive into the foundation, um, and all of its inner workings. Um, but it was a love, it was a lovely experience.

I fell in love with the team pretty quickly and once CarbonHack wrapped up, they were like, "we would like you to keep working with us."

Chris Skipper: Awesome.

Namrata Narayan: I was like, "yes, please." I want subversive movement. I want this to be as big and as successful as it can be. I took it on.

Chris Skipper: Yeah. Carbon Hack 2022 was a real success. It was great. Uh, I spoke to Adam a little bit about how it would be a source of inspiration for people that are doing talks at Decarbonize Software 2023, because there are some brilliant talks and some brilliant ideas that came from Carbon Hack 2022. And if you're interested, you can go to the Green Software Foundation's YouTube channel and you can view actually all the videos of submissions from that, and they've all been put together in really concise videos, but looking forward to Carbon Hack 2023, the flavor of things that will come, they're community focused, obviously, they're not, it's obviously not in the hackathon style of events.

But we'll talk about that a little bit later. So let's talk more about your involvement with the Green Software Foundation in your role as the director of communications and member relations. Has this lead led you to a deeper understanding behind the aims of creating sustainability focused goals within an organization?

Obviously, you're very experienced in that already. How would. People go about communicating sustainability focused goals within their organizations, particularly in relation to green software, and what are the first steps that someone would take to achieve this?

Namrata Narayan: There's probably a really intelligent way of answering this question, but I'm going to answer it in a slightly unconventional way,

which is something I, which I hope is also intelligent, but one of the things I realized a few months ago is if we see software as an agent for climate action, um, then we are going to be a lot more successful in articulating why software aligns with sustainability.

And we'll also, I think, be able to make sense of the metrics that we need to use to measure software and its environmental impact. So, I would say the first thing we need to do is really see software as not just a thing, not just tech, but really something that can move the needle in our broader sustainability pursuits. And then it gets really fun to then tell the story once you look at it from that point of view, then you don't get bogged down in the numbers and in things, frankly, people don't remember. You focus on the narrative, you focus on why we're talking about software. It's one of the easier things to fix and get right, right now, when it comes to reducing carbon emissions, when it comes to developing product and creating processes that are more climate conscious, that are more carbon aware, to use some of the language that we use at the foundation. So I would think that's really the first thing we should do, and then everything gets a lot easier after that.

Chris Skipper: Yeah, absolutely. Yeah. It's interesting that you mentioned it as an actor, right? Because I think most people that aren't in the software industry, that aren't developers, see software as just, or see using their computer as just this guilt free, it is a guilt free exercise, but they don't see the consequences on the environment, especially when it comes to data and using cloud services, for example, that type of thing, or engaging with services that have a big pull from data centers.

Namrata Narayan: Yeah.

Chris Skipper: I was shocked just through editing this podcast at the sheer environmental impact of AI, for example.

That type of thing is terrifying in my eyes. So I think for everyday people, it can be enlightening to hear that type of communication from organizations like the Green Software Foundation. Yeah, I think more, more people should adopt that within their organizations, taking the approach that, yeah, this is something that's easy to change right now, um, and that can have a dramatic effect on, yeah, on climate change and just generally more, have a knock-on effect towards more sustainable goals as well within organizations.

This is a question that kind of sprung to mind, but because of your involvement in the Green Software Foundation, do you find yourself adopting more sustainable changes in your day to day life at all? Like through, in other means, not necessarily software related ones?

Namrata Narayan: Yeah, absolutely. So I was aware of, I was aware that there were things we could do digitally to just be a little bit more responsible, but when I started working with the Green Software Foundation, I never thought about how software was built and what made it, what made certain applications and certain interactions with the software we use, so seamless, and so easy, and almost desirable. And ignorance is bliss, I'm no longer ignorant. And, for example, now, when I'm using, oh, this is a great everyday example, tabs.

Chris Skipper: Yeah.

Namrata Narayan: I am notorious at having a thousand tabs open, at all times. Okay, I have, yeah, I have, like, tabs open for work, I have tabs open for personal, I have tabs open for, like, every facet of my life, and I don't close any, any of them ever, or I used to not close any of them ever, which also tells you what, how my brain works, a bit of things going on here at all times, but now I've gotten a lot more disciplined about closing web pages that I'm not actively working on, closing documents that I'm not actively using. And those are really small sort of actions. I think it's a meaningful one because it tells me in that moment that I'm being really thoughtful about what I'm doing, how much energy I'm consuming, how much energy I'm taking, and what I'm able to give back. So, that's one thing. Also, ever since ChatGPT. Boy, do we love it. But ever since it came about, and I, one of the first articles I read was, I think this was for chapter two or three, I can't keep up now, but for a conversation with 24 prompts, that consumed, what was it? No, a conversation with 50 prompts consumed the equivalent of 24 bottles of water. And I was like, that's ridiculous. I'm not always asking really good questions. At the start, I was just playing around with it. I just wanted to see what it knew and what kind of information it was pulling and what it, and where it was pulling it from. And I quickly learned that ChatGPT just loves to make shit up.

Chris Skipper: Yeah, it does.

Namrata Narayan: So now I've gotten a lot better. If I do use ChatGPT or any sort of generative AI tool, I'm really careful about what I ask, which means I have to do a little bit of homework beforehand. So it's maybe not as fun. I don't go down this crazy rabbit hole of Q&A with the application, but I try to limit it so that I'm being a little bit more resource sensitive.

Chris Skipper: Yeah. I think it's made, made a lot of, yeah. People rethink the way they do things. Like you said, like just having fewer tabs open is, yeah, starting point. And I think that's the, go on, no, sorry.

Namrata Narayan: I'm sorry, I didn't mean to interject, I just want to say that's on the personal front, and then

professionally, I think I'm in a unique position to then take AHA that I received from the GSF and pay it forward, so now when I work with clients, when I do any sort of consultancy work, when it's around web development and web design, I do talk about green hosting, I do talk about you can actually design your websites in a way that are more, um, environmentally friendly, that require less energy, um, about the images you use, think about how many videos you have, where they're placed, do they load automatically, do you, do they have to be triggered?

All of these things make a difference and I'm now building in this knowledge into the conversations that I'm having with people that are actually looking to create websites.

Chris Skipper: Yeah.

Namrata Narayan: So I'm hoping, I'm hoping I'm helping.

Chris Skipper: Yeah, no, absolutely. I think that's the way everyone can help is by paying the information forward. It's interesting that you mentioned, yeah, website design is one of the things the way I think most people, at least people that are in the, in the sort of freelance or creative sphere where they have their own personal website can really make an impact.

And I think one of the great examples of that, I don't know whether you've had a look at it, is Branch Magazine.

Which Chris Adams helps to run, which is just incredible. I have no idea how the coding works behind it, but yeah, it adapts basically to the way that you, to the power in your area. So when it's clean, you get more kind of images more color on the screen.

And when it's a period of dirty energy, you get, how can I put this, binary black and white version of the magazine. And it's still just as informative. You can still get the same amount of information across, which is fantastic. So, and I want to do that for my website as well.

Namrata Narayan: Same here. I didn't even, this is the other thing when I started, where I did not even know we could, this was possible already. My mind blown when I started. It was, it's one of the first things Asim showed me when I started working with the foundation. He said, "hey, have you seen this?" Because he obviously thought I would think it's the coolest thing ever.

You can do this already, like I just think it just. That, to me, is a smart solution. That, to me, is smart software, is the ability we give it to be responsive to what is actually happening in our environment and in our climate. And the fact that we can do it already, it's not something that we have to work towards. We can do it today. Having to know about it is a really, to me, is a really powerful message and also a really inspiring one.

Because we can make significant change today.

Chris Skipper: Absolutely. Yeah. And yeah, that's obviously where you come in. And so, and that's obviously where the event that we're going to be talking about, Decarbonize 2023 comes in because it's very important. And so now when this podcast goes out, we're at the stage where the registration for talks unfortunately has already closed.

But, and you're probably in the midst of picking who you're going to choose for the final tracks and that kind of thing. So let's just talk about a little bit about the rundown of the event. So it's going to be happening on November the 16th this year. For those who don't know, it's entirely virtual, right?

Okay. I'm particularly interested in hearing about the event's objectives related to advancing green software practices and principles. Could you give us a little bit more information about that?

Namrata Narayan: Yes, so this year, I think with Decarb generally, it's, we want to really focus on action and solutions, and not so much dwell on, um, I think our community, our audience is more interested in what we can do, and how we can be better, and so that's how we frame a lot of the events, and a lot of the sort of spaces that we organize. So Decarb this year is really about our community and giving our members and individual contributors a platform to inspire and learn from one another. Whether it's green software patterns or the software carbon intensity specification. Our members have actually taken the knowledge and tools that we've shared with them and spent the last several months applying them and learning how to make them work for their systems and within their infrastructure. So we want to really create a space where these stories can be told because they provide a path for others when organizations are able to hear what their peers or their competitors in some cases are doing and how they're addressing a very similar issue. It gives them additional motive and also guidance on how they might be able to do something very similar. And one of the things I'm personally very passionate about doing at the foundation and for the foundation is helping our, encouraging our members to actually see each other as peers. We obviously have members that are competitors in the market, but when working at, when working on software sustainability and working within the Green Software Foundation, we don't want them to treat each other like competitors. We actually want them to operate like they're peers and they're collaborators and they're helping support one another towards a shared future and a shared goal. So that's really the directive and I'm excited about hearing what they have to say and hearing how they've taken the patterns and what they've done with them, how they've applied the SEI to develop base measurements that they didn't have before, and what those calculations have told them about their, there's a lot to be excited about.

Chris Skipper: That's awesome. Yeah. I like how you framed it in the fact that it's not a competition. I think one of the unique things about the makeup of the members of the Green Software Foundation is that you do have competitors in the market that have come together to Fight for a greater cause, for want of a better phrase, but yeah, there, there's, there are people from Avanade, and from Linux, and from Accenture, big companies like that, that are part of this organization, as well as people who are just starting out, and people that are from other industries.

We had Jo Lindsay Walton, who's a university lecturer who has nothing to, he doesn't, he does have a relationship to green software, but his relationship is a little bit more tenuous compared to other people. And I'm sure he won't mind me saying that, but yeah, so that that's one of the joys of it. And I think people who attend the Decarb 2023 event will see that and will be able to not only benefit from, like you said, learning from their peers, um, and getting some direction, but also just networking in, in general and making new connections through, uh, the event itself, because as with it being online as well, will there be opportunities for people to go into sort of breakaway rooms and chat to each other and that type of thing?

Namrata Narayan: We're not going to do breakaway rooms, but we, there is definitely an opportunity to ask questions during the event, and last year we got a plethora of questions, so we are anticipating a fair number of questions to come our way, to come towards our members as well. We also, through GSF Discussions, which is our sort of open forum on GitHub, are going to really be encouraging people to participate, ask questions, answer questions, share insights, connect with one another, and continue the conversation. It's not something that needs to end after the two and a half hours of decarbonized software. We want people to keep taking these questions forward. Go further because chances are the people that are attending the event have answers to questions others are attending, others who are in attendance are asking, so I would say that's part of what we're looking for and looking to create is just an opportunity for that knowledge to be shared and exchanged so that we can move forward and accelerate the pace of change.

Chris Skipper: Yeah. I like the idea of people educating one another. I think with that in mind, there's, I think you've, you've probably said this already, you've implied it in the way that you've said it, but it's for everyone. It's not just for the that are super experienced in the green software sphere, but also just if you're a student and you're at university and you're perhaps learning computer science or you're even at school, is there an age limit?

Not.

Namrata Narayan: I actually have gotten this question a couple times, especially last year, because last year Decarb was the end of the hackathon. This year, obviously, it's its own event. It gets its own time and place to shine, but every individual, regardless of their seniority or role or industry, if they believe they have the power to make a meaningful difference and drive sustainability forward, then they should attend this event. We want students. We want practitioners. We want, whether they're developers, designers, architects, data scientists, analysts, because everybody is part of the matrix. No one is spared and it's everyone's responsibility and I, and similar to a lot of others, I think environmental problems, you know, when it comes to solving carbon emissions, when it comes to reducing software's harm on the environment, we need a real mixed bag of people working on the problem, it can't just be engineers. They need the support of designers. They need the support of project managers. They need communication people. So we really want diverse audience and we believe that's only going to add value and, I would say, support to everyone who really is eager to do something.

Chris Skipper: And also, just a reminder to people, it is free as well, don't have to pay anything to come, and it's only two and a half hours long as well, so it's, and it's probably going to be, it's going to be what is going to be completely jam packed with really diverse things. So with that in mind, do you have an idea of the sort of tracks that people are going to be able to attend?

What sort of, can you give us an idea of any specific sessions that will be featured at Decarbonized Software 2023?

Namrata Narayan: So for this event, we don't have specific tracks. The way. In terms of the format, we will have a series of community driven sessions showcasing stories and demonstrations, which will really show how different practitioners across industry are using tools and resources available through the Green Software Foundation and others in reducing their emissions, improving the way they measure their emissions, how are they making their software more energy efficient or less resource intensive? Sprinkled between those community-led sessions, we are going to have a five-side chat on responsible AI and introduce new initiatives and projects. So, what I'd like to say, you won't be hearing from us, aka the Green Software Foundation very much, but you will hear a lot from your peers and organizations that you're really looking, that you, that inspire you, that you're really engaged with and interested to hear from.

Chris Skipper: Yeah. Awesome. Yeah, I mean, we, we already touched a little bit on the idea of, yeah, responsible AI. I know just from looking at the website for Decarb 2023, which is dcarb.greensoftware.foundation. You can go to that website, you can have, that's where you can register as well. But just from looking at the website, you can tell that there, if from this, there's some reference to last year's event, which was huge because that's where the SCI, so the Software Carbon Intensity, Software Carbon Intensity Specification was announced, um, as well as the Linux Foundation, um, uh, training program, the Linux Foundation, let me say that again.

Presume there will be a lot of talk about that. We've already mentioned that they're gonna be talk about the way people have used the SCI and there's also going to probably be a lot of talk about the SOGs, the State of Green Software Report as well, which we've featured on this podcast before. And you can also have a look at that website if you go to stateof.greensoftware.foundation and you can find, you can find a heap of topics there. So I presume there will be crossover between the, what we see on the website and what we're going to be hearing at Decarb 2023. So with that in mind, also on the website, there's this statement that I find really profound and it was a statement that I think is, probably sums up a lot about what the Green Software Foundation is about, which is having software at the forefront of climate action. Can you delve into how Decarbonize Software 2023 plans to highlight the role of software in achieving climate goals, particularly in relation to the upcoming COP28?

Which I think will be happening in November.

Namrata Narayan: Yeah, it starts at the, it starts at the end of November and goes into December. So at COP28, for anyone who might be unfamiliar, global leaders will discuss how to reduce global greenhouse gas emissions. Their aim is by 43 percent by 2030. This is a pretty ambitious and admirable goal. In my time working with the GSF, I've learned that nothing really happens without software anymore and it makes a lot of sense when you really think about it. I can't even cook a meal without an app, but it's a perspective in terms of what my life looks like. Advancements in technology and the way we live, I think, have flipped the script on engineers, they're no longer necessarily being told what to build and do, they're being asked how it can be done, and how it can be built better, and any organization that isn't looking to their engineering population as sustainability problem solvers, we really hope Decarbonize Software changes their mind, because it's this community of software practitioners that are going to be able to have a really meaningful and tangible impact on what tech companies are actually doing about their environmental footprint. If they're a tech company, the first thing they should care about is what is their tech doing. And I really do feel that software practitioners are now going to be seen as real critical players to solving sustainability problems. And I think that's new. I don't think that's always been the case.

Chris Skipper: No, it hasn't always been the case. You can tell just by the existence of the Green Software Foundation and how young it is, it hasn't always been the case. And just from the people that have come on the podcast to talk about green software, they will talk about it in, a lot of the, sorry, a lot of the terms and a lot of the, the phrases that are being used around green software are so new.

And that's just the nature of, I think the nature of the industry. But like you said, yeah, software developers are going to have to be at the forefront of this battle against climate change for most tech companies, if not all of them. So yeah.

Namrata Narayan: And it might be a challenge for some. I think, I don't want to speak out of place, but there's, there are probably a good number who haven't been perceived that way. They haven't been trained to see themselves in that sort of position. Hopefully the next generation will. And so it's a bit of a, it's a bit of a culture shock for them too, right?

"Oh, what do you mean? What do you mean I'm, reducing carbon emissions in the way that I code, what are you talking about?" But the, the fact of the matter is it's something that we can, like I said, do today.

It's possible we have the knowledge, we have, we have the SDK, we have these tools to make it possible and work for different types of applications across different types of domains.

So I, I really do think that software is where everybody's focus is, will be in the, in the next few years.

Chris Skipper: Yeah, absolutely. And Decarbonize Software 2023 is the event that you should come to if you want to learn more about it. And in particular, because like we said, this is such a new term, the Green Software Foundation is just over two years old, I believe, and so if you do attend this event, and if you do want to join the Green Software Foundation, or just be a part of the community and, and just involve yourself in green software in any way, you are at the forefront of this movement.

This is the start of it. And it's exciting. And it's something that we can all get behind, I think. So everyone should be able to attend this event to learn more. And my final question to you on this is how do they go about doing that?

Namrata Narayan: Yeah, so everyone can register online. We've created a short link so you have to type less, which is grnsft.org/decarb. We'll share the link in the notes as Chris mentioned. That's it. That's all you have to do. You just have to register and we'll push any and all information that's important to your experience to you as soon as we, as soon as you register and you have your information.

Chris Skipper: Yeah, so it seems like it's going to be a really exciting event. You're all set. Obviously, it's not too far away now. It's probably a month away from when this episode goes out. So before we head off, we've come to the end of our time now. And before we head off, we have a closing question that we normally ask our guests on Environment Variables.

And so with that statement that I talked about on the website in mind, I want you to know, as the Director of Communications and Member Relations, at the GSF, you're obviously very in touch with the message of the Green Software Foundation and promoting the goals of it, as you've spoken at length about, if there's one tagline or catchphrase you could use to convince people to join the GSF, what would it be and why?

Namrata Narayan: So I think I'd go with, let's say, Green Software Foundation: Where Software Meets Sustainability. And I think it's short and sweet. It reinforces our commitment to align these two domains, which are often approached separately, it's inclusive, it doesn't leave anybody out. I'd like to think that it's evergreen.

I don't think it's something that we're going to necessarily solve in our lifetime, but it forces us to keep working on it. It doesn't have an end date. It has to, we have to continuously ensure software is meeting sustainability. And I think it's easy to remember.

Chris Skipper: Yeah. Yeah. Yes. I like how you used evergreen. No pun intended there.

Namrata Narayan: No pun intended.

Chris Skipper: Cool. All right. So we've come to the end of this podcast episode. All that's left for me to say is to say, thank you so much, Namrata. This was really great. I really enjoyed this chat. Thanks for your contribution and we really appreciate you coming on in Environment Variables.

Namrata Narayan: Thanks Chris, it was great fun.

Chris Skipper: Awesome. So that's all for this episode of Environment Variables. All the resources for this episode are in the show description below, and you can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables. See you all in the next episode. Bye for now.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again, and see you in the next episode!

The Week in Green Software: Cleaner Energy with Molly Webb

28 september 2023 | 47 min

TWiGS host Chris Adams is joined by guest Molly Webb from Energy Unlocked for another week in Green Software. In this episode, follow their conversation covering the latest G20 summit in New Delhi and a special announcement they made relating to renewable energy, a surprising move by Apple to now support Right to Repair and their reckons on the subject, as well as news from W3C’s first draft of Wed Sustainability Guidelines, and more. In this episode you can expect to be well informed on the going-ons of the past week and also upcoming events in sustainability and tech. Tune in for an open and exciting chat on The Week in Green Software.

Learn more about our people:

Chris Adams: LinkedIn | GitHub | Website
Molly Webb: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Big weekend news in the G20 - a commitment to amongst other things tripling renewables by 2030 | [4:33]
In a historic about-face, Apple publicly supports right-to-repair bill | Grist | [13:16]
Introducing Web Sustainability Guidelines | 2023 | Blog | W3C | [19:24]
Data centers in Ireland use as much metered energy as all the cities in Ireland | Central Statistics Office of Ireland [24:52]

Events:

CNCF Cloud Native Sustainability Week | [42:14]
Decarbonize Software | GSF [42:23]
UN Climate Change Conference - United Arab Emirates Nov/Dec 2023 | UNFCCC | [42:55]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Molly Webb: [00:00:00] It's about, "is my device that's sitting in my home behind a meter somehow going to be able to make itself visible to the grid and then be coordinated as a virtual power plant?" And even though we've heard the jargon and we talk about virtual power plants, we're really not there yet. That's the key part for software.

Chris Adams: Hello, and welcome to another episode of This Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris [00:01:00] Adams. In this episode, we cover some ambitious new goals at the recent G20 meeting on renewable energy and how that might affect green software.

We'll cover Apple's recent about-face on right to repair, the first set of web sustainability guidelines from the W3C, and finally some eye opening stats on data center energy usage in Ireland. Finally we'll share some exciting events from the world of green software as well. But before we dive in, let me introduce my guest and colleague for this episode of This Week in Green Software.

With us today, we have Molly Webb of Energy Unlocked. Hi Molly.

Molly Webb: Hi, Chris.

Chris Adams: Thank you very much for joining us. Molly, I know that we've gone back for a while actually, but for people who are new to you, would you take a moment to just introduce yourself and see where you're, just explain where you're coming from and then we can begin the show.

Molly Webb: Great, thank you. I'm Molly Webb. I have been living in London since 2003 in [00:02:00] spite of my American accent. I set up an organization called Energy Unlocked in 2015 after working in climate change advocacy for many years and my focus on climate change advocacy was all in software to IT and telecommunications industry and how you would apply novel software or digitally enabled solutions to solve climate change. So Energy Unlocked in 2015 was focused on challenge-led processes to support new entrants and businesses and cities to accelerate their energy transitions. And then in just recently, I've set up a spin out called Pure Carbon, which I'm sure we'll talk about later.

Chris Adams: Cool. Thank you, Molly. Now, Molly, it's nice to be on a call with you again, because I believe we might've met back in the, at least, oh my God, in the early 2010s, [00:03:00] when I was working at a wacky startup called AMEE, which stood for Avoid Mass Extinction Engine, there was a meetup, I think it was called CleanWeb London, and I think I might've seen you presenting one time there.

So yeah.

Molly Webb: It I was back in 2007 that I started working on "what is the role of IT and telecommunications in climate change?" And I feel like I've, it's still a small world, but at the time it really was a small world.

So I'm not surprised we didn't, that we met then. Great. Yeah. So yeah. It's

Chris Adams: It's good, lovely to have you back and to be running through some of this show with me again. So for people who are new to this podcast, my name is Chris Adams, as I mentioned before. I am the executive director of the Green Web Foundation. That's a small Dutch non-profit focused on reaching an entirely fossil free internet by 2030, and I also work as the policy chair for the Green Software Foundation. [00:04:00] And as you might know, I am the host of this podcast. All right. So. If you are new to this show, what we do on this weekly show is we basically look at a series of stories that have showed up on our radar, that are of interest, where there's a kind of green software angle, and we basically just talk about them and share some of the, some of our own reckons on this.

So that's the plan. We'll share all the links to these stories along with anything else that comes up from this. And yeah, that's generally the plan. All right, Molly, should we start with the first story?

Molly Webb: Great.

Chris Adams: Okay, all right, so the first story we have, we know that there was a big G20 meeting in New Delhi, but one of the key things was there was an agreement to basically pursue tripling renewable energy capacity globally by 2030 and another thing about accepting basically a phase down to get off coal power. There's a few other things that took place here, but for our kind of niche podcast, this is the thing that was of interest and we thought might be worth [00:05:00] talking about. And we figured this would make green software easier, right?

Because there'll be a lot more green energy. But it turns out there's a bit more complicated to that, and you need to design for green software for this. And Molly, I figured, like, it might be worth asking you about some of this, or see if you've got any reckons on this particular story, because, yeah, you've got a background with some of this as well.

Molly Webb: So, yeah, it's really encouraging when governments commit to changing the energy system, typically, and this is electricity, right, so we need to power grids with cleaner electricity, and then the flip side of that is that we need to actually shift more of our consumption to electricity, so we can't just heat our homes with gas and then applaud our efforts to accelerate grid decarbonization, we have to actually electrify our demand. So I think while I'm excited when [00:06:00] I see this, I often think once these big announcements happen, it's because the problems have been solved. So it's not really controversial to accelerate the renewables. It's tough to get permitting for transmission networks. It's tough to get permitting for onshore wind or anything that is going to impact on the pristine countryside or how people view where are they where you should do the the manufacturing of these things but what's exciting what's more exciting to me is we have a massive challenge in shifting to all electric transport shifting to all electric heat and doing that at a, at least three times the pace that we're doing it now. So yeah, it needs, it's "yes, and."

Chris Adams: Absolutely,

Molly Webb: I have a bit of a rant about that, I guess. Yeah.

Chris Adams: No, this is absolutely correct. This is one of the, there is this kind of slogan, which is electrify everything, then clean the electricity. And then that's basically one of the big plans to get through to this. And you mentioned there's [00:07:00] a few ways, some of the kind of big challenges facing us, like decarbonizing transport largely revolves around if you're going to be replacing, say, ICE, internal combustion engines, then it's often going to be things with EVs, which means there's going to be a significant demand on the grid there.

And that's assuming you're going to replace it all with cars, which is not necessarily true where, and isn't always the one golden path. And also you spoke a little bit about the idea of dealing with how we stay warm and one of the key things is moving away from burning gas in furnaces to using things like heat pumps, which are still considered somewhat kind of magical, even though they're actually quite old now, I think, Molly, don't you

Molly Webb: Hmm. And there, the problem isn't the technology anymore with heat pumps, it's really the installation. It's having enough people who have the skills and the awareness of what they're doing when they install a heat pump, that all of the standards that go around that, how you make sure that there's going to be maintained and kept up. So there's a lot of, I would say the small stuff that [00:08:00] doesn't happen at a G20 summit that's got to happen if we're going to see heat pumps and other technologies that are tested and tried and completely ready for commercialization. They just start still at small enough scale that no one quite knows how we're going to pull all those pieces together.

Chris Adams: I agree with you. This is actually, now that you do have this kind of statements of intent, it really gets down to the kind of boring, but necessary stuff around legislation or making it easy to actually do some, do some of this stuff and make sure you know how to get the right supplies and everything like that, for example.

Molly Webb: But back to 'why software?', it's very tempting to think, "Oh, I just put a heat pump and then I'm done." That heat pump needs to have connectivity. It needs to be able to be coordinated. It needs to be carbon aware, not just aware of the grid needs, but aware, is it consuming low carbon or high carbon kilowatt hours?

And now there's a difference, right? You can have, my home consumes maximum couple of kilowatt hours [00:09:00] at any at any time. So your home isn't consuming very much. And so the challenge of having all of those small, tiny loads connected and then coordinated in a way that doesn't break the grid is an absolutely terrifying challenge for grid operators. And no one's quite figured, figured that one out either. So as much as it's hard to actually deploy it and get it into a site, then you need it to be connected and coordinated. And there are some new standards, like the MATTER standard, I mentioned, um, before this, um, recording, but there's some new standards, some new thinking about how you would make it easier to do that coordination. With Pure Carbon, we're trying to give another financial incentive for that coordination to be possible. So adding a layer of essentially paying for the Carbon Aware value of coordinating these devices. But it's such small stuff, right? It's not, "oh, I'm going to build a gigawatt scale power plant. And [00:10:00] that's really easy to ensure that over 40 years I'm going to have a market for that power."

It's about, "is my device that's sitting in my home behind a meter somehow going to be able to make itself visible to the grid and then be coordinated as a virtual power plant?" And even though we've heard the jargon and we talk about virtual power plants, we're really not there yet. That's the key part for software,

Chris Adams: yes

Molly Webb: software challenges there.

Chris Adams: Yeah, software is good for coordination. And this is actually one thing we'll just briefly touch on because you've mentioned this term, which for people who may have never heard of it, it might sound like, "what is a virtual power plant?" So the idea basically speaks to this notion that there are two ways of making sure that supply and demand are actually needed, like the amount of power that we need and the amount that gets generated.

So. One way is to basically make sure you've always got supply. Like the massive G20 announced to have lot announcement to have lots and lots of power. But the other option is [00:11:00] just to scale back some of your demand. And the notion of a virtual power plant is basically, if you get enough people together or organizations together or devices together to coordinate how they scale up and down their use of power, you basically don't need to build so much supply in the first place.

So it's a bit like a kind of, it's a coordinated equivalent to having a power plant. This is generally idea behind a virtual power plant and it's a pretty cool idea. And it's something that is, I feel like it's, Molly, it feels only now are we really seeing the kind of support from a kind of policy point of view to actually do this now.

It's been like, a cool idea for 10 years. And now we're finally seeing people start having some interest in actually being able to do this, particularly in America, but also possibly in how people are redesigning how the grid works in the UK and in Europe as well.

Molly Webb: Yeah, I mean there are energy communities coming in the new, in Europe, and that is spurring investment so that you [00:12:00] can be paid to coordinate between each other in a community, so that you're essentially balancing out your load, the shape of your load in a community before it hits the grid, and there's a legislation in the US as well for 2 2 2 2 that's going to allow much smaller and smaller loads to participate. But it is, your dishwasher is basically a kilowatt capacity. It's going to consume not very much at once, so you need a million dishwashers to coordinate, to have this sort of scale that the grid is used to dealing with.

And that will be in different places the grid is used to dealing with one power plant in one place, and it's a known quantity. So in spite of the fact that we're starting to see the frameworks, we don't necessarily have quite the same focus on technology to deliver it as we have traditionally had when you focus on technologies like building networks and transmission lines, building [00:13:00] power plants.

Chris Adams: Indeed. We'll talk about that a little bit later on, because if nothing else, I learned about the, some work with MySociety about trying to bring around this new law around community energy, and we'll talk about that later because it's just, it's too good a pun to miss. But in the meantime, let's have a look at the next story.

So our next story here is actually about Apple. This is from Grist Magazine, and uh, it's an,

it's, yes, yeah, it's a historic about-face. Apple have come out to support, uh, right to repair for their equipment. Now, this is interesting because for the longest time, Apple has basically been saying, "no, we think the best way to maintain the quality of any kind of electronics is to make sure everything comes back to us and then we'll basically sell it back to you later on," for example, or "we'll take these apart and then you'll purchase an entirely new machine."

Now, Apple have changed their tune to basically say, "yes, we think that rights to repair is a good thing and, uh, it's good for the, it's good for business and it's actually good for, [00:14:00] uh, end-users and consumers." This is really interesting because Microsoft, Apple is now the second of the very, very large companies to come, come out in favor of this after Microsoft did a somewhat similar event, had a similar about turn in the last six months ago, and we'll share a link to that.

The reason I figure it might be interesting for us to talk about this, Molly, is that you have this shift from holding on to something for a very long period of time now. So rather than having a kind of 18 month kind of cycle where you have a phone and then you throw it or, and then someone buys another one, which is really bad for the planet, this kind implies a much longer life cycle and a lifetime for equipment.

And I guess a kind of very different for want of a better word, do people use a duty cycle for something? If it's gone for a long period of time,

or it's something like that, I think. And I figured this changes how you design for things for, for a start. And it makes you think about designing for a longer period of time and making devices [00:15:00] work a bit better and make something work on lower spec machines, for example, but I guess there's a bunch of other considerations if you have equipment that might last five or 10 years, for example. And Molly, I think this is something that you might've had some experience with, or that you did a

Molly Webb: Yeah. I wish I knew more about the competitive reasons why they would have switched. I always think "what's their competitive advantage?" Are the supply chain issues making the return and refurbishment just a real nightmare, and so they're like "Sure. Let's get a bunch of mom and pop shots out there, refurbishing our equipment. That's fine." Are there bottlenecks with materials in the supply chain that make them want to slow down the sale or something? I'm, they obviously don't want to slow down sales, but there's got to be something in there is what it makes me think of,

but I do think if you have things for longer, I remember, was it Fairphone in the UK?

Didn't they? Yeah. But it does make me think you need more in the [00:16:00] cloud, more software. You need to be able to do things that once you have some, or aren't reliant on, say, the chip in the phone or whatever it is, or you'll need to be even more modular about the hardware design and be able to switch out some of those key components. So yeah, it's exciting because you really don't need to replace that phone all the time, though, if constantly you're breaking it too, so. For that reason, you might need to replace it. You need better cases that when they drop, when you drop the phone, it doesn't completely shatter.

Chris Adams: It can be a bit harder to have a waterproof case if you have to make it easy to, for people to open up, because if you make it easy to let people in, it can make it easy to let water in, which is one of the challenges that I think Fairphone had previously as well, and

recently, yeah, Fairphone, so they, I believe they announced this Fairphone 5 in the last few months, and this is the first one which is somewhat waterproof compared to the previous ones, which were definitely not waterproof, [00:17:00] and they too they've actually, I think they are, they've gone on the record saying they expect to support this phone for 10 years from this, which is

a far cry from what we had before.

It also, like you

Molly Webb: I, yeah, no, I was just over the weekend in London. It was still the heat wave, went to the pool and saw someone in the pool with their phone,

and I thought that is, it's just to say there's a human, we will have to change our behavior a lot around these devices if they're going to last longer.

It will be really interesting and I think a lot about the materials that supply these things. So I think shifting to a more life cycle approach with phones, like you need to do with other, when you sell a wind turbine, you know how long it takes that wind turbine to work off the embodied carbon associated with it. When you sell a phone, you don't really think that way. So it'll be interesting if we can

shift to more [00:18:00] lifecycle analysis, where some of the data you were looking at with AMEE back in the day would help you do the embodied carbon assessments. And we're just maybe getting to the point where we could do more of that.

Chris Adams: You're right, actually. There is, there was a recent legislation passed about how all, especially in Europe, about every device having to have replaceable batteries, but that, it was either coming in 25, 2025 or 2027, there was something like that. So maybe one of the things that's come up here is that actually there's a bunch of really low value repairs that you don't want to do yourself, or it's really hard for you to justify doing.

And then if you open a fig leaf here, then. You're able to at least look like a very good kind of corporate citizen whilst giving away the thing which didn't make you that much money in the first place, for I don't know.

Molly Webb: There is also Scope 3 reporting, mandatory reporting coming into Europe in 2025 which will affect a lot of companies with European operations.

So that will be interesting to see how [00:19:00] far companies take that.

Chris Adams: You're right. Okay. This is, right. This is, this is going to be an interesting one. Um, for people who've been listening last week, we did a bunch, uh, in about Scope 1, 2, and 3 reporting. So, uh, if you are really into, uh, the GHG protocol, that's the place to go. Um, in the meantime, though, uh, we were talking a little bit about how you might design for, uh, devices that last for a longer period of time.

And the good news is that W3C, as in, yes, the people who define what the web is, essentially, the next story is actually about them sharing their first ever web sustainability guidelines, the version one of this from the sustainable web design group. So this is interesting and I figured this might be a nice one for us to talk about, Molly, because this is essentially a full spec, full set of recommendations that you can follow. And, uh, typically when you have large organizations or large specs being put together, it's often considered that these are [00:20:00] only large organizations that are able to really influence how some of these guidelines are put together. It was a real community push.

So there were people representing say large companies associated, but there was also like Tim Frick, for example, one of the guys who's been, who wrote Sustainable Web Design way, way back in 2016, for And there's a bunch of freelance developers and people associated with this as well who've been working on this. And I'm going to make sure that I'm going to look through to make sure I haven't forgot some of the names, but Molly, I figured this idea of actually something which is being pushed from the kind of bottom up, as it were, rather than being where you tend to have one or two very large organizations influencing the surface, I figured it might be something that you might have some reckons on because, yeah, you've seen how some of the sausage gets made a few times and

this, yeah,

I is there, are there any kind of corresponding things

Molly Webb: I just, I.

Chris Adams: brings to mind for you?

Molly Webb: I wish there were more successful case studies of real bottom up standards or communities coming together that then have a really big [00:21:00] influence, but because generally right now, I think we're going to talk about 24/7 carbon free push for accounting for electricity at the hour, as opposed to annually, and that is certainly enabled by software.

So the big software companies like Microsoft and Google and even Amazon have different strategies for how that should happen. And so they've really been able to influence that agenda, not necessarily in a bad sense, but it has really been large companies with a lot of capacity, able to hire whole teams that can look into this and then push it out, I guess, I look at industry associations, and often those industry associations can be captured by the sort of lowest common denominator, so I do what's going on with LF Energy, with the work on the carbon data standards, Working Groups, and I like, so some of the open [00:22:00] source work is really, I think, great to look at. But yeah, if it would be Interesting if there were more, if there was a way to have this bottom up small startups, I mean, Energy Unlocked, we set out to work with startups and to give them a bigger platform and a bigger voice. And it's really tricky, not just because it's hard to get a common message from startups, but also because they just don't have the time to be working on this.

They're in survival mode.

So this is a great example. I'm going to look into that one.

Chris Adams: Yeah. I really, I think I'm really happy to see this because, so there's two things with the actual article open, there's a couple of things that it's worth referring to. So, initially, there was actually something called the Sustainable Web Manifesto that was put together a while back, like a good few years ago, that has its own website and we'll share a link to that.

That people, a bunch of people started signing and, uh, it got some people really excited and interested [00:23:00] in this. These guidelines also actually refer to the GRI, which is the global, it's the Global Resource

Institute. Thank you. The Global Reporting Initiative. So there is, yeah, there's a whole kind of like techie angle on it, but there's also actually that there's this real push to bring in some kind of rigor that you see in the sustainability world to really address some of these issues here.

And the thing that's also really nice is that. It looks like the, some of the strategies that we've seen successfully used for accessibility are being adopted here. So there is a, this set of guidelines called the Web Content Access Guidelines, which are specifically around making it easy for you to tell if a website is considered accessible.

And, uh, they each have a kind of, there's a rubric, which was, I think it's POUR, P O U R, which is perceivable. As in, can I perceive the information that I need to access? Operable, as in, can I operate the actual device? Can I do something? So this is a lot about not being tied to have to use a mouse, for [00:24:00] example, if you maybe don't have access to that, for example.

One of them, which is understandable, I believe, which is relatively self explanatory, and the final one, which is robust. So R being robust, being you want to have some kind of technology that's together. And it's worth looking at the guidelines at a glance, there's a short version of this with the editors named Alexander Dawson, Tim Frick, and Neil Clark of TX Impact and Tim Frick of MightyBytes I've mentioned before.

There's a bunch of things to look through. I think this is a real triumph and it's really nice to see this make it out the door because it's been in work, it's been, people have been working on it for a while. All right. Okay.

Molly Webb: It's amazing, thanks.

Chris Adams: All right, Molly, the last one is going to be a little bit about, um, you did allude to some of this before about the power being used by different organizations, by different, like, in a house or something.

This is the story about Ireland, actually. Over the last month, a pretty impressive or eye opening stat was [00:25:00] published by the Central Statistics Office of Ireland. Basically stating that in the last year, data centers' metered electricity consumption in 2022 was 18 percent of all of Ireland's electricity usage.

And that was basically the same as 18 percent of all the urban dwelling. So the data centers in Ireland use the same amount of electricity as all the cities in Ireland. And when you would have this coming from say a campaigning group, that might be one thing, but when it's coming from the statistics body, this is a really interesting and eye opening stat that I haven't, I haven't seen before. And this is, I think it speaks to some of Ireland's unique role in kind of infrastructure and particularly digital infrastructure, because Molly, you mentioned before that this felt like a bit of an outlier. There's a global figure for electricity usage by data centers and things, and then this one here, which seems to be massively higher than what you might typically expect.

Molly Webb: Yeah, it's [00:26:00] really interesting. I study urban carbon footprints, and of an urban carbon footprint, it's often buildings that are the biggest contributor, and that's commercial and domestic buildings. And then, of domestic buildings, of the city's footprint, I guess you'd say domestic is one third of energy, but that's also heating, which isn't electricity, right?

And then, what percentage of a household's consumption is electric as opposed to other forms of energy? And that's usually like a third again. So you could see how the pie shrinks, but to say that urban dwellings are 18 percent of final energy electricity consumption, and then same with data centers, is to me the first time I've seen them on par, because the International Energy Agency is still saying 1 to 2%. I think this is where my big sort of fear when I heard of the Pace of AI and ChatGPT kind of solutions, [00:27:00] I was like, "that is really power hungry stuff" and we're gonna see a real increase there and we know how, how much data centers just continue to increase in terms of consumption. So I'm surprised to still see the 1 to 2% of global emissions from data centers as the kind of, that's what we said back in 2008 when we did this work with McKinsey on what's the direct footprint of the industry, and it was like 2%, and that included devices, it included laptops, not just data centers. So I'm thinking data centers have grown as a proportion globally. But when you see a stat like that from Ireland, you can really feel the, the the scale of it and we're going to just see that grow. I would love to see better numbers. I felt like I had seen some better numbers on data centers and now I can't find them again. I'm like, did someone post something and then remove it cause it was too incriminating? So I, if anyone [00:28:00] has good stats out there, send them.

Chris Adams: This is a nice kind of elegant segue to some of this whole policy fight that's taking place right now. So just for context, there's a couple of links for this. So we spoke a little bit more about, okay, this is energy usage, and we've spoken before about how You need to think about demand usages. You think about matching the amount that you need, uh, for power.

You can basically, yes, you can have more supply, but you can also think about how much demand you need to have. This is one thing that you might want to be aware of, or that it's worth being cognizant of when you think about this kind of stuff. And there is actually a story I've linked to, which is actually from earlier on in the year.

Uh, this is basically about Microsoft trying to, uh, take steps to essentially integrate data centers into the grid, because as Molly just mentioned, Molly, you, if you've got a big load, if you can reduce that, yes, you can get like a million dishwashers to scale back. But if you have a really chunky facility, if you can scale that back, that's actually [00:29:00] one other way of dealing with this kind of varying amount of power that's going to come in and out of the grid based on how, how sunny it is, or how much the wind's blowing and so on and things like that. And I believe this is something that you've spoken about before. And I think this is some of the things that you're doing right now. The story we'll link to is basically a story about Microsoft building data centers such that they can do this and they're expecting to be paid and compensated for this.

And as I understand it, it's not just data centers who, who. Who are basically trying to find ways to build like a economic model to integrate or, I don't know, what's the word we use? Grid choreography? I don't know, there's got to be something you call it,

Molly Webb: I say carbon aware

Chris Adams: Carbon aware, yeah that's used in lots of places.

Molly Webb: Yeah, but it is choreography,

Chris Adams: Yeah.

can't.

Molly Webb: Algorithms.

Chris Adams: Yeah, I think choreography's cool, right?

It stops it being so much about, I don't know, there's all these other metaphors we could explore which aren't about military, but are evocative and [00:30:00] visual, basically. So yeah, I know,

that's

Molly Webb: A dance.

Chris Adams: Yeah, it's a dance, all right? Orchestration, we

Molly Webb: certainly. Yeah. Yeah. We do use orchestration and instrumentation. Google and Microsoft have been saying it's important to look at every hour because the kilowatt hour is no longer the same depending on the time of day when it's generated, so a kilowatt hour's carbon intensity will be different if it's sunny or windy. And so when we consume it, we can look at not just reducing demand, but changing the time, deferring or shifting demand. One approach is hourly matching, so just look at the match of what you procure and what you use. And you also, what you consume on the grid, what you procure, and what you actually use. The problem with that approach is... You can get around not shifting your demand if you can find a good power purchase

agreement to buy some cleaner [00:31:00] energy from somewhere nearby. That's great. And then you have what is called like emissions first, where you look at emissions only. And so you would say, "Hey, I'm Microsoft and I have a data center in somewhere with completely clean grid and somewhere with a completely dirty grid."

So like, Finland versus South Africa, and you would say, okay, I'm going to move my computing load to the cleaner country. And then what I work on is cities. So cities can't just move all of their buildings to another country. They're stuck in place, but they also can't control how the entire city procures. So, we're in this combination of an hourly matching strategy and an emissions first strategy, and what I do and have been working on for the last three years is analyzing the carbon value of this carbon aware demand, and sometimes you can get up to 30 percent annual savings by just [00:32:00] shifting your demand to be more aware and not use the high carbon intensive kilowatt hours. And then now it's the question, is that value worth a while enough to someone to pay you for it? And if they're going to pay you for it, are you definitely using that to invest in electrification

and low carbon alternatives? So that's the sort of path I've been on to create more of a location based, carbon aware strategy for cities and for companies that operate in cities and want to be able to maximize every option now to reduce carbon, even before they can afford the big new change, like a heat pump or, or electrifying their vehicle fleet.

Chris Adams: I think we're going to need a word for this third approach that people are taking, because it's actually,

Molly Webb: I call it carbon flexing.

Chris Adams: Carbon flexing. Okay, we'll go with carbon flexing for now, because

Molly Webb: But we do need,

Chris Adams: it's, uh, because I believe [00:33:00] so that, I forget the name, Magee, I forget his surname, but he

was also, Magee, he he was also talking about this idea that, yes, there are things you can do to arrange a big power purchase agreements, you buy all the power, like the 24/7 thing you mentioned, or there's an emissions first approach, which Amazon and Facebook quite pushing quite a lot now, right now, where they're saying,

"I care less about the hourly time, that's important, but what's more important is matching it to the dirtiest energy I can find to get, take that off the grid as it were. So I'm displacing some of it that way."

And then this third approach, which I understand that what you folks are doing and also what, I think what Carbon is doing in America is basically finding another way of saying "we have all this decarbonization that has to happen and if we don't have the actual policy support to do that right now, there needs to be another way to pay for that to happen."

And there are ways that you can incentivize people to change their behavior. As I understand it, when I heard you describing, that seems to be the same, not, a similar approach, there's all this other really dirty emitting use of energy we need to do something [00:34:00] about, and there are, there are steps that you can take to shift people's behavior.

And if they can do that, then if compensating them makes them more likely to do that, then that's an option. And if people are able to build a service around that, then that's something that corporates or people with the means may choose to do as one way meet some of their own kind of climate contributions, I suppose, for this goal, this shared goal that we all have.

Molly Webb: Yeah, the way I look at it is that a lot of the policy is focused on 2035 when our grid is fully decarbonized and what should the tariffs be and it actually does take that long to change fundamental things like pricing and tariff design because it needs to be tested and there are experiments going on. In the meantime, we can use voluntary initiatives like the ones that companies can instigate as almost like testing grounds now, but they can do it at scale to demonstrate that their demand is going in a certain [00:35:00] direction and then that informs policy as well. So we help increase the pace of the policy change, but we also do it before there's a regulation in place, and that's, I think it is very aligned with WattCarbon.

It's, I love that they're, what they're working on.

Chris Adams: It's good that you both have the word carbon in it, so you at least have some idea, and you agreed to carbon flexing, whatever that is. All right, we've just got a bit off track, because we were talking about data centers in Ireland a minute ago, and one thing that we were saying was that one of the things, these numbers seem hard to get your head around, and the link we've shared actually shows a link directly through to like the spreadsheet that you can look at yourself, if you fancy, but the thing that I, you said, which I found quite interesting was that you're saying that houses, we've got this figure of lots of power being used here right now. And I know that right now there is a law that is currently going through in Europe, the energy efficiency directive, which basically says organizations, if any kind of data center facility, that's [00:36:00] using more than a hundred kilowatt, kilowatts of demand, not kilowatt hours, kilowatts of demand. They all need to start sharing information about how clean the energy is, as in how much of it is from renewable sources. Uh, what is the kind of water footprint, how much power is used over the given year and stuff like that to inform the kind of policy discussion.

Now, when I was looking at these figures here, a hundred kilowatts, I felt like that's not a small amount. That's, it's not ginormous, but it's relatively sizable. I think, uh, you mentioned say, dishwasher running at full blast being a kilowatt hour for example and I know that I think in the UK I think the average usage for a house works out to be about over a year it's maybe, say, three, I think the

average is

Molly Webb: to 5, 000 kilowatt

Chris Adams: Yeah, exactly.

So that works to be around, what, 300 to 500 watts solid at the entire time. And a data center, that's a hundred times that. So if 100 kilowatts was quite small, what we've seen in the last month or so is that the [00:37:00] threshold for reporting has now gone up to 500 kilowatts, right? So you've basically got, gone from, let's look, what's it, maybe a hundred kilowatts, that's what, 200 houses, assuming roughly, right?

You've gone from 200 houses and now the most recent kind of threshold for reporting is now at 500. So basically you've gone from 200 houses to a thousand houses, households worth, worth of energy usage. That's the threshold for being able to actually report and share any information now. Now it feels, if we are struggling to understand these figures and we're struggling to actually have a data informed discussion, I really feel like this multiplying the minimum threshold by five times is not going to make it easy for us to have a kind of honest discussion about how we actually share the supply of electricity to various places, or how we actually integrate with the grid. Now, I just wanted to just speak to someone who might have some figures on this, because yeah, they, the intuition that you shared of three to 5, 000, that's what I thought.

And these numbers [00:38:00] don't seem all that small. Is a hundred kilowatts a big load these days?

Molly Webb: No, it's a small load from the grid's point of view.

That's the thing. So, anything under a megawatt, like 500 kilowatts, 100 kilowatts is small. I'll just give you the example of UK Power Networks

in London again, because I'm here. They had a minimum threshold 500 kilowatts for participation in their procurement of demand-side flexibility.

So, what are they going to call on when the grid is congested to turn down? They had a 500 kilowatt threshold minimum, and then they lowered it to 100 because they thought, "oh, we're not getting enough of this small stuff." So, from the grid point of view, they think a megawatt and under is teeny. But from a person who's a company that's trying to aggregate a thousand homes, the lower the minimum, the better, because then you don't need as much contracting and commercial

arrangements with all of these decentralized players.

Right? So my [00:39:00] first thought is 500 kilowatts, still small. So it's still under a megawatt. I'm happy about that, but it is the wrong direction, right? Everything else is going more and more small scale to, for reporting and other things, and this seems to be going the other way. So I guess I think of the, I think it was called the Medium Combustion Directive or something like this with diesel

generators. I'm trying to think of the exact acronym, but there's regulation on the size of a diesel generator that you need to report on. And so the way that the industry gets around it is just splitting up these diesel gen sets into smaller and smaller chunks so that they can get under this threshold. So the higher you make this threshold, the more a megawatt plant suddenly is actually two 500 kilowatt, 499 kilowatt plants and they get away with it. So that's what you've, I think, got to be thinking about is, [00:40:00] there's some gaming of this going on behind the scenes so less reporting has to happen. We definitely don't want that.

Chris Adams: Yeah. And what you just said just now about if you've got like people basically designing an entire grid or a market so that people do get compensated or paid to integrate more, integrate into the grid, then surely having this information is going to be more helpful. I just, when I, when I learned about this, it just seemed like such a step backwards and I really hope that we don't see this taking place because for the longest time, this was like some of

most impressive legislation specifically to help with things like green software, because these kind of data, this data that's necessary really helps inform some of the decisions we make as engineers, if we're going to design a system to, you know, work in a kind of choreographed manner with the grid, if we're going to use that language.

Molly Webb: I think we have to realize that the increase in consumption of all of our computing is going to make a megawatt seem smaller and smaller, [00:41:00] at the same time that we want to get more and more visibility on what's going on in order to choreograph it, and a lot of that electrification of heating and transport is going to also increase electricity consumption, so we're dealing with a growing power sector just, well, at a time when the power sector in Europe, at least, hasn't seen demand growth very much.

It's seen it stayed pretty flat or even declined because of efficiency measures. So we're in a new world where it's going to be a much more decentralized coordination challenge, a much bigger electrics, electricity demand growth than we're used to at the same time that we're changing, we're, we're seeing the impacts of things like weather events on the resiliency of all of this infrastructure.

So lots to keep the grid operators awake at

night. Yeah.

Chris Adams: really does feel like a thing to have more visibility on, not less visibility on, even if it

is a bit harder to actually collate some of this information, which [00:42:00] should be easier now because it's 2023 and we have computers to tabulate stuff.

Molly Webb: I know.

Chris Adams: All right. Okay. I'll get off my soapbox and just run through the last events and I think we'll wrap up.

Okay. So we just have a few events, which, uh, may be of interest to listeners. So we have a Cloud Native Sustainability Week taking place in October. We've got a link for that. That's a virtual event all around. There's lots and lots of talks. And then finally, there's this decarbonize software on the 16th of November.

This is open and this is a follow-on from the kind of carbon hack we did last year. And those are like the events that are coming up, but there's a bunch of other big, significant events taking place if you think about climate and not necessarily tech. And Molly, this is something we were talking about before, because you mentioned there's NY Climate Week and there's COP, and there's always a COP these days.

So yeah, what's, what's on your radar for the next couple of months that you might point people to if they have an interest in climate?

Molly Webb: Yeah, I tend to follow tech or climate and the two meeting [00:43:00] is rarer, but we're all definitely looking toward COP. This is the UN General Assembly coming up at the end of this month, and that's where we have Climate Week NYC, lots of announcements, people get together and talk and set targets and then I'm hopeful that there'll be more at this COP on the buildings and more decentralized.

So, I think that's the initiatives side of things, because often that's missing the organization that will probably do something on the 24/7 carbon free hourly matching, carbon emissions, emissions first, or carbon flexing strategies aside. SE4ALL is the organization that's been pushing a UN Global Compact on 24/7 carbon free energy. So, they'll likely be doing something at COP. So we're looking at all of that. I'm very curious if anyone sees anything to do with the power [00:44:00] hungry AI chips and climate, because I think it's a huge, it's a huge change this year when we were starting last year when we've seen breakthroughs in AI and we're seeing the weather events and other really scary impacts of climate change. And so you want AI to be applied to those challenges in a very rapid, urgent, systematic way or agile way, maybe not systematic, opportunistic. I don't care. Apply the, apply it because we are going to see that there's all of this power. The power needs to drive AI, and then you want to see it doing, having the right outcomes on the other side from a climate perspective. So that's what I think

Chris Adams: Um, very much so, I think, um, just as you were talking about that, and when you mentioned AI, it got me realizing, I, so this hundred kilowatt figure, which I thought was large, which, which, which is all right, I was trying to see how can I do that if I've just got a bunch of a, [00:45:00] say GPUs, like these chips used in AI, apparently I, I asked around and I think If you, you could plausibly get, I think with the new, the NVIDIA A800, they use enough power that you can put a few of them, I think, I reckon you can basically use two racks, just two racks by themselves.

It could be enough to hit a hundred kilowatt hours, sorry, a hundred kilowatts of demand. So just two racks is going to be sufficient. And when you compare that to two racks or 200 houses that really puts some of this in perspective about how much power is being used for this stuff and why it's important for it to coming from renewable places and actually that it's the way that it's deployed is actually in a kind of, in a more equitable fashion.

Oh my God. I'm going to stop there because we're just going to go off off on one because.

Molly Webb: Yeah. That's a whole other podcast.

Chris Adams: All right. Molly, I think that's taken us up to the time that we have. I'm really glad you came on. It was really lovely to see you again. Uh, and it made me feel like I'm back in my kind of clean web London days back, [00:46:00] back

there actually.

Molly Webb: Thank you so much. Thank you so much for having me.

Chris Adams: All right. Okay. That was it and thank you very much and have a lovely week. See you around, Molly.

Molly Webb: Brilliant. Thank you, Chris.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again, and see you in the next episode! [00:47:00]

Sci-Fi Fantasies with Anne Currie and Jo-Lindsay Walton

21 september 2023 | 38 min

Joining host Anne Currie, is Jo Lindsay Walton, a research fellow in Arts, Climate, and Technology at the University of Sussex. Together they will explore the dreams of a green future inspired by Science Fiction and the practicality of these as solutions to climate change. This adventure will cover interdisciplinary approaches to viewing and tackling climate change and green software from angles of technology, politics, and especially literature. The discussion will touch on the Digital Humanities Climate Coalition and it’s toolkit that can help researchers minimize their carbon footprint, and will revolve around the ASCEND programme as well as other opportunities and missions to attempt the clean and efficient use of data centers in environments like our moon, and the complexities of protecting and cooling the servers, and also the aspect of polluting the moon

Learn more about our people:

Anne Currie: LinkedIn | Website
Jo Lindsay Walton: LinkedIn | Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

UK-Ireland Digital Humanities Association | [4:42]
Data Centers in Space: The Promise of the Moon | [15:26]
Beyond the Stratosphere: Computing in Orbit | [31:29]

Resources:

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

[00:00:00]

Jo Lindsay Walton: Building data centers on the moon is very productive of fuzzies, but not utilons.

Anne Currie: Indeed. Yes, indeed.

Jo Lindsay Walton: And I also feel like any, any file I saved on the moon, I would also want to save somewhere else as well.

Anne Currie: I think that would be sensible. It's not exactly your ideal disaster recovery location.

Asim Hussain: Hello and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field, who have a passion for how to reduce the greenhouse gas emissions of software.

Anne Currie: Hello and welcome to another episode of Environment Variables, where we bring you the latest from the world of sustainable software development. I'm your host, Anne [00:01:00] Currie. For those of you who are regulars to the show, you've probably heard me on the other side of the microphone, but in this episode, I'll be host, so this is goint to be an interesting episode, because we'll be talking about kind of science fiction approaches to climate change. What's going on and what's actually useful to us to be thinking about and what probably isn't useful for us to be thinking about what we, we might be distracted by? But it should hopefully be a very interesting episode that we have a guest today who is also massively interested in science fiction.

So I would like to introduce to you our guest, Jo Lindsay Walton. So hi, Jo. Welcome to the, uh, the podcast and please introduce yourself.

Jo Lindsay Walton: Hello, Anne. Hi, everybody. Yeah, I'm Jo Lindsay Walton. I'm a research fellow in arts, climate and technology at the Sussex Digital Humanities Lab and I'm really excited to be here. I'm a relatively new member of the Green Software Foundation and I've really come here via the Digital [00:02:00] Humanities Climate Coalition, the DHCC, which is a kind of community-led initiative, which I guess we'll be speaking about, around digital decarbonisation, around climate justice, and I also do some work on climate communication, how do we talk about climate, bringing in interdisciplinary angles there, games, arts, literature, including science fiction. So this overlaps with my interest in science fiction, including the sort of post cyberpunk fiction of writers like Cory Doctorow, who directly explore contemporary issues around tech, law, and climate as we encounter them today, as well as more classic works by people like Ursula Le Guin, Samuel R. Delany, that like to imagine life under better or just radically different social institutions. So I'm really happy to be here.

Anne Currie: Excellent. Thank you very much. Um, so, uh, just a little bit about me for, for people who perhaps aren't regulars. My name is Anne Currie. I am one of the co-chairs of the Green Software [00:03:00] Foundation Community Group. I'm also currently writing the O'Reilly, the new O'Reilly book about green software, co, um, a co-author of that. It's called Building Green Software. It's being published as we go on the O'Reilly website, and I'm doing that with my fellow GSF members, uh, Sara Hsu and Sara Bergmann. Uh, and, and my sci fi, the reason why I'm, I suspect why I'm hosting this episode today is that I'm also the author of a science fiction book series, the Panopticon series.

It's similar in some ways in terms of kind of time in which it's set and, uh, and ideas to, to Cory Doctorow, so it's post cyberpunk. Yeah, and it covers a lot of the stuff that we'll be talking about here.

Jo Lindsay Walton: Including the moon!

Anne Currie: Including the moon, including the whole book on the moon. Oh yeah, the moon is great. I love the moon.

Before this podcast, I listened to, just to get myself into the mood, I listened to the theme music to Space 1999, which is a, it was a really good show [00:04:00] about the moon. The science was a little bit dodgy, but it was in the 1970s it was a good show, and although I, and I, and it was repeated a lot on television in the UK through the, my entire childhood.

So, it's constantly watching this story about people living on a kind of renegade escaped moon. But anyway, before we start, because we can't get to, we're just going to get horribly sucked into talking about the moon and science fiction. But before we start, it's just a reminder that everything we talk about here will be linked today in the slow, in the show notes below the episode. As I said, before we get into the sci fi discussion, because that's gonna basically take up all our time, Jo, do you want to tell us a little bit about the Digital Humanities Climate Coalition? Just to give us all a little bit of a context.

Jo Lindsay Walton: Yeah, I'd love to. So, the Digital Humanities Association of UK and Ireland launched a few months ago, and the Digital Humanities Climate Coalition, the DHCC, is one of its community interest groups. But we have been doing things more informally since about the time of COP26, [00:05:00] and we began really out of that kind of sense that, as Margaret Atwood puts it, climate change is everything change.

So every field, every domain should be exploring climate impacts and climate actions. Everybody should get their own climate coalition, and this one is ours. So, what is the Digital Humanities? It's an eclectic mix. One of the things the Digital Humanities loves to do is talk about what the Digital Humanities actually is.

Um, you've definitely got some kind of brilliant research software engineers, some very technical people, and then you've got less technically proficient people, including me, I should say, um, who've maybe come in via a history angle or a literature angle from the kind of more traditional arts and humanities and come to the tech from that direction.

And I think that's really our niche, is that there are all these fabulous new tools and methods appearing all the time. And hopefully we can signpost those and maybe build some bridges for the less technical users. So the DHCC's mission is to help everybody, and [00:06:00] especially arts and humanities researchers, to understand and improve the climate impacts of our use of digital technologies.

And it's community-led. You mentioned the link will be in the show notes, and I'd really invite your listeners, especially if they have any interest in widening participation in sustainable digital tech, you know, creating those on-ramps for different levels of experience, I'd really invite them to get on the mailing list in the GitHub and get involved.

As well as that side of things, we're also really keen to equip users to reflect on the big picture of climate change. So the Arts and Humanities loves to think about politics, ethics, about the social and cultural. features of the decisions that we make and the perceptions that we have. And when you work in tech or use digital technologies, it's very easy to get excited about this or that solution or optimization and maybe lose track of the bigger picture of climate change and climate policy.

A key thing for me is that the planet has a finite [00:07:00] capacity to generate green energy and to absorb carbon, growing, but growing at a finite rate. So there are these hard trade-offs there about how we use resources up until 2030, up until 2050 and beyond. Yes, it's complicated by innovation, by actions that might stimulate demand and investment and so on, but those trade-offs are there, and a particular legal entity might be net zero or better, but if it's using green energy, if it's bagsied some of our carbon absorption budget, then that means that's not available for other things.

And part of what we like to think about in the DHCC, in the resources that we provide, are these climate justice angles. Can anybody seriously think that we shouldn't prioritize food security, healthcare, transport infrastructure, disaster management, sanitation, biodiversity, things like that, especially in the global south, where the needs are greater and where the responsibility for climate change is so much less?

So encouraging that kind of critical scrutiny is something that we're really keen to support as well.

Anne Currie: That is very interesting. Yes, and of course, you've mentioned the [00:08:00] links to the DHCC toolkits in this notes before. That's all great. So I had a quick look at the DHCC stuff and it is really interesting stuff and an immediate thing that came to mind was something that, uh, I think is the is the key issue when you start to talk about climate and climate change and using sci-fi or, uh, literature to change people's minds and move people's, move people forward, which is that... Uh, and, and Joe, this, um, you're gonna know more, you, you probably know similarly, you probably think about this a lot, as I do, which is that fiction, and driving things forward, and getting people involved in things, is often about individuals, because there's no story without a protagonist, so literature tends to be about individual action. But, climate change, there's a big battle at the moment between individual action, which we know doesn't work, and we know, and I don't know if you've read Michael Mann's 'The New Climate War,' about, [00:09:00] it's not a sci-fi, it's, he's one of the, yeah, Michael Mann is one of the, the, I think it was the

Jo Lindsay Walton: I actually, I bought that book yesterday, coincidentally, but I haven't read it.

Anne Currie: Yeah, it's a good book, it's well worth reading. So Michael Mann was the inventor of the hockey stick on climate change and everything's going to go horribly wrong, we need to do something about it. And in his new book, in his latest book, which is well worth reading, The New Climate War, it's about disinformation and propaganda against climate change through, and not just climate change, but all change.

Big business propaganda tends to be about trying to steer people onto individual action, which doesn't really, for these kind of huge scale changes, doesn't really work. So it's a distraction. It keeps everybody's, "eh, don't drop any litter. Look over there." So yeah, it's litter dropping as a distraction to various things in the past that big business has not wanted us to be looking at. These days, you know, turning down your thermostat, we should all be turning down our [00:10:00] thermostats, but it's not in and of itself going to move the dial, ironically enough, on climate change. But, in fiction, you have to have a protagonist, you have to have a story, you have to have individual change, otherwise you don't have much interest.

I'm quite interested in your opinion on that, and also, I think somebody who tried to tackle that bit, with loads of issues in the book I would say, but nonetheless did attempt to tackle that, was Kim Stanley Robinson in Ministry for the Future. Um, I don't know if you want to talk about that at all.

Jo Lindsay Walton: So, yes, this is the book that comes up a lot, doesn't it? What Kim Stanley Robinson does in that book that's very interesting is throw everything at climate change, and then actually withholds judgement about what's been effective and what hasn't. He makes some judgements, but there isn't a kind of overall narrative that says, "these were the key drivers, these were the secondary drivers, and these particular measures [00:11:00] were ineffective."

It's a, it's a very interesting book. I would definitely recommend it. One of the things that interests me is that it does seem that, like, paramilitary action is a big part of the relatively hopeful future that he paints, but it all happens offstage. Yeah, I was so interested in that book, I wrote two reviews of it, two, for two completely separate, uh, venues.

But your, your, your really interesting point about this question of individual action and systemic action, um, or systemic change, i, I agree, I think 90 percent or 99%, maybe 100%. I might frame it slightly differently when that dilemma comes up. When we think, "is this about individual action or is this about system change?"

I tend to like to prioritize individual action, but I frame the individual action as saying, "you need to find your collectives. You need to find your alliances. You need to found your, your coalitions, [00:12:00] work within larger organizations, work within your employment context, within activist contexts, within NGOs."

So it is still your individual action, but you're, you're looking to drive that bigger systemic change. Because I also think that while individual action can be a distraction, so can complaining about the distraction. That itself can become a distraction. And just to bring it maybe a little bit to software.

I think software and design is a really interesting space for thinking about how individual agency meets that kind of systemic plane. So, I observe myself doing carbon intensive things on a daily basis. I now don't use a thesaurus, I just go over to my tab and ask chat GPT to give me a bunch of synonyms.

But these are design questions, they're not just questions of individual responsibility. There are ways of adjusting the structures and [00:13:00] incentives so that individual desires are manifested in different ways and perhaps in more sustainable ways.

Anne Currie: Yeah, it's, it's interesting you say that actually, cause I, one of the things that I noticed on the DHCC website was the quite correct point that should developers be developing in Python, which is a hundred times less efficient than C, for example, which is something that I used to talk about years and years ago, and it's certainly true because I used to be a C developer and Python's terrible compared to C, but I can see why people moved over to, to Python because C is just so much more difficult to write and it is certainly isn't low hanging fruit. You could bash your head against a brick wall there. But having said that, I used to rail against it myself. And now I rail against people who rail against it, as you say. But the Python development team have now produced tools that will compile Python to C, so you can write in C and get the performance characteristics of writing Python, nice easy language, [00:14:00] get the performance characteristics of C. Now. That's the perfect solution for this. That is a good foundational strategic solution, which means that you don't have to change what you're doing. You can still write your code in Python. You get the really great performance out of it. But would it have happened if we hadn't all been moaning about how unperformant Python was compared to C? So, so to a certain extent, individual action isn't effective, but moaning about it often is effective.

Jo Lindsay Walton: That's really interesting. Do you know, by the way, anything about the sustainability of this Mojo character that's just popped up?

Anne Currie: Mojo? No.

Jo Lindsay Walton: The new programming language apparently combines the usability of Python and the performance of C.

Anne Currie: I mean, it's entirely possible because, really, you, you're, what you write in and what actually runs are completely separate things. There is no difficulty at all, not that no difficulty, there's a lot of difficulty, but it is entirely possible to compile something which is [00:15:00] incredibly verbose, like Python, or presumably this Mojo language might be even more verbose, and compile it into something that's just assembly language, it just runs and doesn't, that is the purpose of a compiler. So you just need compilers that optimize for performance. But more and more compilers are doing that, which is really good. That's the solution we want. We don't want people to change their individual behaviors, we want compilers to get better.

But what we should probably do is get back to the actual thing that we're supposed to be talking about today, which is the moon and, uh, data centers on the moon and also data centers in orbit. Now I have, as, as our usual host, Chris Adams would say, I have a lot of reckons on this subject, so both good and bad. So just to give you a bit of context on this, as a, as a listener, back in May, we published an episode of Environment Variables called 'Data Centers in Space,' which I was on, which discussed the possibility and the real, very real possibility of building a data center in [00:16:00] space to mitigate power consumption and pollution and various other things. And again, I've, I've done a lot more thinking about that in the intervening time. And we focused on the ASCEND program, which is basically a space cloud for Europe with an awful lot of finagling around acronyms to turn it into ASCEND. And basic, the idea is to move data centers into orbit. And today we've got a link in the show notes below, it's a blog post from Western Digital written by Ronni Shendar which discusses the idea of, just a very real possibility. Not necessarily a possibility for tomorrow, but a possibility for at some point for building a data center on the moon. So just to give you a rundown of the, of the blog post, it talks about a startup company in the U.S. Called Lone Star Data Holdings, which wants to revolutionize data storage by building uh moon based data centers uh by using the, uh, lava tubes on the moon where you've got some kind of effectively, although, [00:17:00] although there's going to be an awful lot of demand for these lava tubes, because every plan for the moon involves using the lava tubes, how, how many lava tubes are there? But anyway, lava tubes on the moon to give you a kind of built in warehouse with stable, relatively, which actually is mostly about shielding you from space rays, which are pretty horrendous outside of the earth's atmosphere. Not just the atmosphere, but the, but the magnetic shield around the earth. So everything's terrible out there. But the idea is you build data centers on the Moon. And Chris, our excellent editor for this. So you, you use the reader as a listener will never encounter it, or will seldom encounter it, but Chris is marvellous and he does all our prep for us for this, and he's asked us some questions that we should discuss about the idea of Data Centers on the Moon, and the first question that he's asked us to discuss is how much energy could this really save in, for example, cooling compared to earth based data centres? And what impact [00:18:00] might that have on reducing carbon emissions? And what would be the issues with polluting the moon? Uhm, Jo, if you have any thoughts on that. I have loads of thoughts.

Jo Lindsay Walton: Um, I'm glad. So I, I asked actually, um, yesterday, my friend and collaborator, Polina Levontin about this, because I read the article, I'm not qualified to comment on the science, and she is a scientific one, and she just gave the wonderfully poetic answer that, "have we not always already stored data on the moon?

Our dreams, our forebodings, our utopian desires." So maybe that kind of speaks to the point about polluting the moon. Maybe it's the idea of this pristine wilderness that we don't want to spoil. In a very unscientific way, off the top of my head, and you know more about this than I do, the pros are that it is cool, both figuratively and literally cool,

and you've got plenty of sunlight, and then maybe some kind of co-benefits of a permanent lunar presence, [00:19:00] a staging post for Mars missions, an opportunity to do science on the moon. The cons would include lag time. The moon is over a light second away. Obviously, lifting a lot of mass and the energy and embodied carbon implied in that.

I don't know if they're, are they planning 3D printing and stuff in situ? If not, or even if so, there's a big carbon cost to putting stuff on the moon in the first place. And then remote maintenance. I would like to see you do this in Antarctica, under the sea first. A lack of legal framework as well. And then just broadly, the con of uncertainty.

Does a data center in low gravity in a vacuum with just a soupçon of atmosphere, no magnetic shielding, does, is the data going to behave differently over the years? So basically, I think it is completely bananas. I think they should absolutely go for it, but I'm definitely, I'm one of the haters that they need to prove wrong.

Put, put, put a data center on the moon. I think, uh, it's maybe slightly more probable, but [00:20:00] only slightly more probable than putting a data center in Narnia. Um, the White Witch's Curse of Eternal Winter also creates very favorable conditions for, for data center cooling.

Anne Currie: Yeah, oddly enough, my views on it are really very similar to yours, and I

have done a reasonable amount of research and it's, yeah, it's that I really want to see a moon base, I want to

Jo Lindsay Walton: What? Yeah.

Anne Currie: I, I, uh, so, um, Jeff Bezos, oddly enough, has some quite good thoughts himself on this, which is, in answer to the second question of, uh, won't it pollute the moon? Bezos's, uh, position, and I tend to think he's probably right, is "yes, good." Because actually you want to move the pollution that goes alongside industry from the earth to the moon. It's, uh, that is the purpose of industrializing the moon is that you get the pollution happening up there rather than down here, and we love the idea and the hopes and dreams and that sunny, [00:21:00] and I love to wave at the smiley face of the moon, full moon. But we know that in 1000 years time, that's going to be completely built over. If we survive, that's going to be completely built over and the ideal would be that the earth is better and the moon is a bit of a, a rubbish tip for Earth and that's not a bad thing. That is a, that is better than, than stuff polluting the biome. But you, you're totally right. And we said this in the last podcast. For climate change, it's, it's of no use to climate change whatsoever. The, the timescales are way too long. And you can get all of the benefits that you would get from a moon, a moon data center, much as I love the idea, and I really wanted to have them at some point, through, Greenland and Antarctica eventually will have constant 24/7 power through water, hydroelectric power runoffs from melting glaciers. We've got limitless power there if we were willing to use it. If we were willing to be bothered to put a data center on Greenland, [00:22:00] which has, has issues. But much, much fewer issues than building a data center on the moon. And we, and we, uh, even Microsoft are already building data centers for under the sea, which they find actually is very good for cooling. And if you don't poke around with them because there aren't people around, then they last longer. So you get better on, you get better use out of your embodied carbon and things like that.

Yeah, I totally agree. Climate change wise, it's a crazily stupid idea. It's a distraction. Although I love the idea and I really want this to happen.

Jo Lindsay Walton: Yeah, it's an interesting paradox, isn't it? That, that, that we do love the idea, even though we, we know it's a terrible idea.

Anne Currie: Yeah, we love the idea. We've got to have a Moonbase. We've got to have a Moonbase.

Jo Lindsay Walton: I think, I think it's quite a common thing, isn't it, that amongst sustainability leaders, amongst environmentalists, you get this understandable, and I have a lot of empathy with it, animosity sometimes [00:23:00] towards space travel and space exploration. And I can see where it comes from. It comes from these completely unscientific imaginaries where we can mess up this planet and simply escape to another one and it comes from, uh, you know, for example, within degrowth discourse, which is a very big conversation, but which I think captures some important aspects of the climate crisis that are not well articulated elsewhere. Within degrowth discourse, I think there's an association between space travel, space exploration, and the sense that there will be an infinite plentitude of resources for us to continue to keep expanding into if we just find the technological solutions. So I can see where that animosity comes from, but at the same time, earlier in the episode, I gave that kind of big list of things that I would like to see prioritized when we use our carbon budget.

Basic things like food security, transport infrastructure, [00:24:00] social connectivity, disaster management, etc. I would put space science in there as well. I think this is something that is exciting, inspiring, worth doing one of the, kind of, something that you wouldn't regret doing, something you wouldn't regret spending resources on.

So I'm interested in knowing if there are ways of separating that positive vision and association of space and space exploration, of separating that from the environmentally catastrophic set of discourses that it's been meshed with. What do you think?

Anne Currie: Yeah, it would, it's, it is a shame that the, the degrowth movement is never going to sell anyone because it's a bit hopeless. It's, it smacks of regressing to a, to a world where, it's, we talked, we talked about the Kim Stanley Robinson book, The Ministry for the Future. And, uh, one of the things in there was, it was talking about, oh, well, no, there are no mass holidays anymore, but there are still these lovely [00:25:00] holidays in which people go to amazing places on, on, hotter in hot air balloons and airships.

And the thing is, those are really crazily expensive. I can see why people resist the climate movement, because it really played to that thing of there won't be holidays for everyone, but there still will be holidays for an elite group of people, men, that's, you're not in it. We've got to keep technology that gives something good to everyone and doesn't just mean that there's, like, super stuff for a tiny number of people and terrible stuff for the majority of people and they can't go on holidays and that.

We have to come up with a solution that is in some way inspiring. If we get rid of all inspiring stuff, we're never going to sell anyone to get started. Although, having said that, I know we, we, we slag off progress against, on climate change, but we've made a lot of progress.

Jo Lindsay Walton: Huge amounts of progress. Yeah. And there's a, there, there are a number of perception gaps in terms of the kind of progress that has been made, [00:26:00] and the risks that, that we face. The IPCC science is not well understood, uhm, not broadly understood.

At the same time, there are also a huge number of kind of scientific uncertainties that are not well understood. Perfectly normal scientific uncertainties, a perfectly normal kind of part of scientific practice, all good science produces uncertainty. But these are not well reflected in contemporary climate policy.

And particularly, I think, in some of the more techno solutionist visions of the future. I think you're right that degrowth has a branding problem, and I'm interested in seeing some of those same ideas appearing under different rubrics, under different titles. I think often it's the way that the most kind of interesting, fascinating, and hopeful ideas somehow appear, with the absolutely worst possible labels attached to them.

But definitely, if [00:27:00] you drill down into a lot of degrowth discourse, you'll find a variety of opinions, but you'll certainly find ideas reflecting what you're saying about a climate transition being, needing to be just, and needing to be inspiring, something that has something for, for everybody, and realizes co-benefits in, in, in everybody's lives, and is not just about an, an elite enjoying a legacy of luxuries while the rest of the world kind of wanders around in hair shirts self flagellating.

Anne Currie: Yeah, because it struck me, say, in the Kim Stanley Robinson book, that was, he'd obviously made some effort to not write that, and yet he'd still written it. It's really hard for degrowthers to think about how they're going to pitch the message, I think. And I think it's a totally pitchable message, but it's also very difficult for them.

Even things like the 15 minute city, you'd think. Who would possibly object to the idea that you'd be able to,

like, get to the shops [00:28:00] with, with in 15 minutes walk, or a quick cycle, or a bus, and then we're regularly by... who could object to that? And yet, it's become a horrendous political hot potato. But we, we can't really, as Michael Mann put, points out in, in his book. "Don't underestimate how much money the other side have to put into convincing everybody to keep with the status quo." It's, you have to be a fantastic communicator to communicate change when there is an almost limitless amount of money arrayed to make whatever you say sound bad.

Jo Lindsay Walton: That's very interesting. The 15 minute city thing was astonishing, wasn't it? It got, as I understand it, mixed up in all these kind of conspiracy theories, where people thought they were going to be contained in these like urban oubliettes where they couldn't travel any great distance. Yeah, really astonishing, and the point about disinformation, about misinformation, about greenwashing is really interesting.

I think we're entering a time of [00:29:00] great epistemological uncertainty. I even wonder if the framing of greenwashing is adequate to cover all the sorts of instabilities of meaning and information that we're likely to be encountering. I wrote this kind of musical glossary of terms called, I think it was a greenwashing glossary or something like that, and coming up with other terms like greenwishing, for example, where you are doing something good, it is improving the sustainability of your practices, but you're also indulging in wishful thinking and you're not duly weighing the actual sustainability impact of what it is you're doing, and a bunch of other kind of terms like that.

Anne Currie: Yeah, it reminds me of the effective altruism movement, which is the kind of utilitarian charitable movement around 'you put your money where it's going to have the most effect rather than where you feel good about it.' Totally, that's had, a big proponent of that was the guy behind FTX. The [00:30:00] cryptocurrency collapsing thing.

So fundamentally, EA has been completely blasted away by, by the behavior of FTX, but they had a very good description of how you might think about doing the right thing and the wrong thing. And it was 'get your fuzzies and your utilons separately.' It's about what makes you feel good, you know, it might make you feel good to do certain charitable actions, but they might not actually be very effective.

In fact, there might be actively bad whereas there are other things you could do that you'd get no real potential, you get no internal strokes from, that would be very effective. And, but yes, you'll get your young, your fuzzies, which are about feeling good about yourself and your utilons, which is actually about having effective change and making effective change

in,

Jo Lindsay Walton: So,

Anne Currie: separate ways. Yeah, I, and

Jo Lindsay Walton: for you and me at least, building data centers on the moon is very productive of fuzzies, but not utilons.

Anne Currie: [00:31:00] Indeed, yes, indeed. Actually,

Jo Lindsay Walton: And I also feel like any file I saved on the moon, I would also want to save somewhere else as well.

Anne Currie: I think that would be sensible. It's not exactly your ideal disaster recovery location. And in

Jo Lindsay Walton: Maybe I would save it on the sun.

Anne Currie: all, all the disaster happens on the moon and the Earth's fine. Another thing where if you, if you're going to do DR, you really need to stop in both places. But so we better get onto this to the second bit.

Otherwise we'll just chat about this forever. Uh, and, and the second bit I think is even more of an interesting link than the first. This is about computing in orbit which is about doing more, having data centers, orbiting data centers. And there's a very good, interesting blog post about how we should all move into, move more data into orbits and you can analyze all the data. And it's a charming blog post about if you could process data that you are seeing in faster real time in orbits, you could monitor what whales are doing in even faster real time, [00:32:00] but it's, what it clearly is, it's a giant advert for Lockheed Martin and other American military companies, because it is the thing that you are doing, if you want to be processing data that you want to be looking at the ocean, processing data real time about what's going on there, that's entirely for military stuff, which I don't have any particular reckons whether that's good or bad, but there'll be a load of money going into it because China will be start doing it, America will start doing it, eventually India will start doing it, Russia will start doing it.

It is, uh, an arms race, I would say. Not a gre, this is greenwashing. You want a new greenwashing term for this one.

Jo Lindsay Walton: Yeah, I mean, I feel like, um, let's assume it does work. Let's say that the technology is solid.

Anne Currie: It'll work.

Jo Lindsay Walton: And under that highly hypothetical circumstance, right? My question is still, how does this fit into the big picture? This is something that we're interested in the DHCC. How does it fit into the big picture? Are these orbital data storage facilities, are they going to outcompete the [00:33:00] earthbound data centers that are using the dirty energy? Who actually holds the big picture of global strategy here, of addressing the urgent issue of climate change? Is it the conference of parties? Kind of, but they're mired in all these geopolitical rivalries. Is it the scientists?

The IPC? Yes, but they're constrained by the remit of political neutrality and face challenges around communication. Is it the finance, the markets, they're waking up to something, they're trying to incorporate climate into these risk management methodologies that they don't really play all that nicely with?

Is it science fiction? Yes, we're drawing in a really interdisciplinary way. We've talked about Kim Stanley Robinson throwing everything at climate change, but it is ultimately a story. I'm not really sure who does hold the big picture and if I was to try and summarize it in a crude way, it seems that we're hoping to adjust the rules of the game.

We haven't even adjusted them yet, but we're hoping to adjust the rules of the game so that goods and services and enterprises and value chains and industries and sectors and whole communities and regions that are incompatible with a, with a broadly livable planet are going to be [00:34:00] destroyed in the Schumpeterian whirlwind of, of creative destruction will, will crash and burn.

And I think there's a lot of emphasis on the creation side of that, building data centers on the moon or in orbit, but not enough imaginative, creative, realistic thinking about the destruction side of it. There's this expectation that enterprises are going to snitch on themselves. "Oh, we've tested for impairment, we're reporting against this particular standard, all our assets are stranded, we're just going to shut up shop, goodbye."

So I think I would be interested in more science fictional thinking about the potential pain of switching from carbon intensive activities to the sustainable ones. Not just the focus on the kind of shiny new possibilities, but also the focus on what it's like to shut up shop.

Anne Currie: Yeah, which, you should read my books.

Jo Lindsay Walton: will.

Anne Currie: Yes, yes, I completely agree. All the stuff we've talked about today, about sci-fi is marvellous, it's [00:35:00] lovely, it's fuzzies, but it's not green at all, and it will be no part of the climate solution, or very little part of the climate change solution. There's nothing here that is being suggested that couldn't be done vastly better on Earth. Now, I'm not saying that none of this stuff should be done, but it's not part of climate change, and it is being washed as if it is, and it's not. So, we, we, we have chatted for too long, and we have overrun all our, all our times today. We're now just having to, uh, zip through and do our, um, closing questions. Jo, if you had a data center in space, which fictional sci-fi franchise would you reckon would be best at running it? It's a good question.

Jo Lindsay Walton: Because we've been saying the word data so much, I can't get Star Trek Next Generation out of my mind. So, Data, Picard, Bev, Deanna, that lot. I think it would be hilarious in general because the captains always ride roughshod over the metrics that officers present them with. "Your download will complete in one hour" and they're like, "give it to me in 30 minutes."[00:36:00]

"Aye aye, captain."

Anne Currie: Yeah, I think the only, yeah, I

agree that Data would be excellent running a data center, but I think it would have to be Data on his own. I don't think anybody,

Jo Lindsay Walton: Aww.

Anne Currie: But you wouldn't need anybody else, would

you? You really wouldn't need anybody. But actually, I think the best people would be from the same franchise, the Borg.

I would just

Jo Lindsay Walton: Oh my gosh,

they are already a big data center, aren't they?

Anne Currie: They are a big data center, terrible customer support, but I think there are some major folk who be better at customer support than the Borg. And I will, I won't mention their names, but we all know who they are. Thank you very much indeed. We've come to the end of our podcast and all that's left for me to say is thank you so much, Jo, that was really great. Thanks for your contribution. And it was, and for our listeners, where can they find out more about you?

Jo Lindsay Walton: Thank you, yes, it's been really interesting. I wish we could talk longer. So, I think many of your listeners might be interested in the DHCC toolkit. Um, you don't have to think of yourself as a digital humanities person, [00:37:00] I hope some might be tempted to get involved and contribute. If you're interested in science fiction, I'd encourage you to check out the British Science Fiction Association, again, you don't need to be UK based, um, and our journal Vector, which I've been editing with Polina Levontin for the past few years.

If you're interested in climate communication and maybe some of the broader issues around the political economy of climate change, you can check out our Climate Risk Communication Toolkit, which is a publication of the UK University's Climate Network. And yeah, I think that's, I think that's plenty to be getting on with.

Anne Currie: Thank you again. So that's all for this episode of Environment Variables. All the resources for this episode are in the show description below, and you can visit podcast.greensoftware.Foundation to listen to more episodes of Environment Variables. See you all in the next episode. Bye for now.

Asim Hussain: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google [00:38:00] Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show and of course we want more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. Thanks again and see you in the next episode.

Decarbonize Software 2023 Preview with Adam Jackson

14 september 2023 | 31 min

Decarbonize Software 2023:

Decarb.greensoftware.foundation 👈 REGISTER HERE

Learn more about our people:

Adam Jackson: LinkedIn
Chris Skipper: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Electricity Maps [6:47]
Software Carbon Intensity (SCI) Specification Project | GSF [14:09]
Green Software for Practitioners (LFC131) | Linux Foundation Training [14:09]
Green Software Foundation YouTube Channel [16:49]
State of Green Software | Green Software Foundation [17:38]
Meetup Program | GSF [17:54]
Amadeus [20:15]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Adam Jackson: But really this year's event is about our community and the stories they have to share. We're interested in seeing what people have done with green software in 2023. But yeah, the theme is about, the theme of the event is about empowering software practitioners to decarbonize sotware and to build communities and we want to see what people have done.

Chris Skipper: Welcome to this episode of Environment Variables. I'm your host, Chris Skipper, and in this episode, we will be discussing the upcoming Decarbonize Software 2023 event with the Green Software Foundation's Community Project Manager, Adam Jackson. Hello, Adam, and welcome to Environment Variables.

Adam Jackson: Hi, Chris. I'm really excited to be here.

Chris Skipper: Great, good to have you, and Just before we dive into the meat of this podcast for our listeners, could you please introduce yourself?

Adam Jackson: Yeah, absolutely. So I'm Adam Jackson. I'm the Community Project Manager at the Green Software Foundation. Really excited to be here, as I said. I've been with the Green Software Foundation a few months now. I've also got my own little company helping tech companies connect their communities. And I've recently been involved in setting up a bunch of green software meetups around the world.

Chris Skipper: Awesome for those who don't know me, I'm Chris Skipper. I can't say my name apparently. And I'm the podcast producer of Environment Variables. And a veritable noob when it comes to green software, and in fact, all things software. Um, but I feel like I have acquired a lot of knowledge about the subject that would hopefully make me a good addition to any pub quiz team.

Now, before we dive in, here's a reminder that everything we talk about will be linked in the show notes below this episode. So, to kick us off, I've got a really simple question for you. Let's talk about what your role is like at the Green Software Foundation. I know you've said, how long did you say you've been part of the Green Software Foundation

Adam Jackson: Now, four, four and a half months now.

Chris Skipper: four and a half months, so relatively green to the Green Software Foundation, but then again, the organization itself is also quite young. What does being the community project manager at the GSF entail and what are your responsibilities and goals in the role?

Adam Jackson: Uh, yeah, sure. So it's, it's quite a broad role, um, because community is a very big part of the Green Software Foundation. Um, but essentially it's helping to connect people to the resources of the GSF. So, um, some of the things I'm involved in include the, um, running the events. At the G S F working very closely with my colleague Namrata, um, that includes our Decarbonize software event that's coming up in November, includes our hackathon that'll take place in the first half of next year, and a bunch of other events as well.

And then I run the meetup project, so that's about getting tens of thousands of people connected with local Green software Meetup community groups around the world. I'm also helping our organizations, we have member organizations in the GSF, um, helping them to build their internal green software communities.

And then finally I'm building a new program, which I can't actually talk about too much today, but hopefully that'll help highlight and celebrate some of the people around the world who are making a huge contribution to green software. And there's some really amazing people out there, so we want to highlight the great work that they're doing too.

Chris Skipper: Awesome. So this, uh, secret project, I guess you could call it, um, is more of a, I don't know, you can't tell us that many details about it, but when's the sort of, what sort of, what's the sort of target launch date for that? Little

Adam Jackson: Oh, Decarbonize Software. Oh,

Chris Skipper: Oh, that is

Adam Jackson: converges. Yeah, yeah, no, but the, yeah, yeah, we're going to be making a couple of exciting announcements at Decarbonize Software, so that's why I don't want to spoil it for our audience today.

Chris Skipper: All right. We'll definitely dive into more detail about Decarbonize Software. But first, let's just talk a little bit more about you. Where did your journey with green software start? And how did you come to find yourself at the Green Software Foundation?

Adam Jackson: Yeah, Chris, I love that you describe yourself as a green software noob because that's that's sort of me as well. Um, I'm very much reliant on, um, these a lot of experts in the Green Software Foundation telling me how these things really work, but, um, but I'm passionate about the environment, but, um, anyway, I, um, I guess my story in this sort of space really began in my previous company, so I've worked in Microsoft for about 12 years.

maybe a few years before joining the GSF. And my first job with Microsoft was actually an intern in Legal and Corporate Affairs, nothing to do with developers or software practitioners at all. But I started working with developer communities in 2014, back in the Windows phone days. Don't laugh too much about that.

That ended really well. I ended up being a program manager in Microsoft's Developer Relations team. And I also did a stint in developer marketing and um, yeah, I guess it was a few years ago, people really started talking a lot more about the environmental impact of software and of it more widely, um, as people started getting rid of their data centers and moving everything to the cloud.

At first it was just, it was a problem that had been taken away and then they started to think, oh, there's still a data center somewhere. There's still a network, there's still all these devices. And, um, actually I, um, I'd worked with. Asim Hussain, um, in the Developer Relations team. And many of our listeners will know that he set up and shares the Green Software Foundation.

So, um, he, when I, when I left Microsoft earlier this year, um, he reached out to me, see if I'd like to do something along the lines of what I did there, but for the foundation. Um, so I'm, I'm not an expert on green software. Um, I'm not a developer either, but I've spent a lot of time working with developers, but we always say in the foundation as well, it's not just the developer on the team, but it's also the team around them that can help decarbonize software.

So hopefully I, hopefully I do bring something to the party. And, um, yeah, I mean, I'm interested about the environment outside as well. I, you know, I want to see how I can decarbonize my life. We've just got solar panels, um, and they, I was going to say these, they do work in the UK, but I'm looking outside and we've got a thick fog this morning.

I'm currently getting about 60 watts out of them, so it's not, not great today, but, uh, solar in a battery system, you know, it's, it's really interesting. I've really got into looking at my dashboards to see, you know, what time do we draw power from the grid to top up those batteries? Um, am I doing that at a less carbon intensive time of day?

I'm looking, I use the data from Electricity Maps and it tells me what time of day I need to top the battery up. Is the greenest? I drive. Yeah, I drive an ev, so I'm trying, can be better, but I'm not, I'm not a hundred percent of the way there yet. So even though I'm not building green software myself, I'm also thinking as an individual as to what sort of impact I can have.

Chris Skipper: Yeah. That's awesome. I think having that mindset, even in your day to day life will give you the right approach to your job as the community manager, I think I would hope at least that most people that are part of the Green Software Foundation do that sort of thing. I know Asim has talked on this podcast before about his EV.

As well. And you're not the only EV owner that's been on the podcast , but No, that's good. And I really, I would like to afford one, but if I could, I would, but I, I unfortunately cannot. But yeah. Um, I probably won't put that in there. in the final edit.

Adam Jackson: they need, they need to come down in price quite a bit, to be fair. So, um,

Chris Skipper: I think they will eventually, I think when, when they become more commonplace.

And eventually fuel is phased out. Fossil fuels are phased out. They eventually will become more commonplace. But anyway, that's not what this podcast is about. Um, although yes, I do quite fancy them. I do quite fancy buying one. Um, yeah. So let's talk about creating a community that is focused on green software.

It's obviously a prime focus for the Green Software Foundation. But how would people go about creating a green software community inside their organizations? And what are the first steps someone could take to achieving this?

Adam Jackson: Yeah, I'm really glad that you asked that, Chris, because my job is focused on creating both external and internal communities, and internal communities inside organizations.

Yeah, building community at work, it's going to be different in every organization, but I've got a few tips. The first one is to really think about Culture and communication, bringing things together, who's the key stakeholders and champions that are passionate about green software inside your organization?

That could be developers, managers, executives, you might have a sustainability team if your company's big enough to have one. Um, what do you want to achieve and how are they going to be able to help you, um, get the word out? The GSF has some great suggestions for what you should prioritize, but what actually works for your company?

Um, you need to make sure that you've got a clear vision and also to ensure you make it worth it for the company too. You do want their support in building a community after all. Quite often, well crafted, efficient software is green software. Efficiency drives cost savings in the long term. So that's worth it for the organization and it's worth bringing people together to talk about that.

So set some goals you want to achieve, um, think about how you want to communicate that and think about how your community... Communicates and collaborates. And depending on the culture in your organization or where you are, how are you gonna bring people together? Are, are you mostly operating digitally or do you have a lot of people together in person?

Could you do some lunch and learns or some socials? I love the social, but if you are spread all over the world, you need to think about how are we gonna connect people and excite them and, um, and, and incentivize them to bring them together.

So, and then the finally, you, um, you need to think about. Who can help you communicate this?

So if you're a big company, you might have an internal comms team, or you might be smaller, so you might have to take it on yourself. But do you have company newsletters? Do you have company all hands? And if you're a smaller organization, how are you going to keep folks engaged? Because small company, you've got, you know, lots of different priorities.

Going on and you've got to try and handle them all it. It's different for every organization, but there's lots of help and suggestions online. And then don't forget to evaluate how your company is doing as well when you know how the community inside's doing, because that work's. Never done. Once you've set something up, that doesn't mean the work's done.

You've got to keep up that momentum.

Chris Skipper: Yeah, absolutely. I loved your answer about, uh, the setting up social environment or making having those sorts of conversations that you might have at the drinking fountain if you're in an office or perhaps via the, I don't know, nudging someone on Teams or something like that to talk. I don't use Teams, so, um, but, but yeah, um, that kind of reminded me of the saying that Asim had about creating avenues for green software or avenues towards green software can come through it.

Literally talking to one or two people in the organization that might have a passion for the environment or the passion for green software, more specifically. Um, I know we joked about EVs earlier, but it might be as simple as seeing someone who's an exec from your company with an EV in the car park and going up to them and being like, Hey, I have this idea for setting up this community within our organization.

And that could be the starting point for your community within the organization, if you don't have one already.

Adam Jackson: Yeah, we're all individuals after all, and I doubt that anyone is just interested in green software and doesn't have any other thoughts about sustainability. Think about what brings you together, and don't, we don't have to limit your communities, and we say this for external, for our meetup groups as well, you don't have to talk about software in every single talk.

Sometimes you might want to talk about something else as well, a related concern, and you're all individuals, you've been brought together by these. These shared interests, so let's talk about them.

Chris Skipper: Yeah, absolutely. Cool. Okay, so let's talk about the elephant in the room or the topic of this podcast, which is Decarbonize Software 2023, the event that'll be happening in November. In fact, the date is November the 16th, and it's a purely online event. Can you provide an overview of the event and its goals, especially in terms of promoting green software principles and practices, please?

Adam Jackson: Yeah, absolutely.

So the, you know, the event is all about showcasing the latest advancements in green software by the community. And we're inviting speakers to share their experience and insights. So I think, as I mentioned earlier, we are going to have a few announcements from the GSF. But really, most of the agenda is about featuring demos and lightning talks by our community.

So we're, uh, we're opening up, basically, to people to come and talk about things they're passionate about. And in terms of who can attend, the event is open to anyone who's interested in learning more. about green software and how it can address the global climate change challenge. It's a great opportunity to network with other green software enthusiasts and experts.

There'll be a big social around that. We'll have, uh, we'll have the chat open. We'll have people hopefully writing loads of stuff on, on LinkedIn and. Do we call it X now? X? Twitter X? And other social media platforms, of course. So it's not just about the event. It's about the community as well. But yeah, the event is free.

You can register now on the event website and we'll put that in the links. And if you'd like to speak at the event, you can also submit your proposal through our call for speakers. That's going to be open until the end of September. Um, and that's listed on the website. Uh, and don't forget, you can all, you'll, we'll also be.

Talking a lot about decarb on our social channels. So we are active on LinkedIn and X and we've got our newsletter. And yeah, we're basically, we're going to be talking about it a lot.

Chris Skipper: That's awesome. Yeah, I know you mentioned announcements, um, and I know last year there was a big announcement at, uh, Decarbonize Software, which was, uh, the announcement of the SCI, or the Software Carbon Intensity Specification, as well as the Linux Foundation Training Program. So, are there any, any little sort of sneak peek you can give us for announcements, or is it all going to be hush hush and you just have to register and you're going to have to find out?

Adam Jackson: Well, we are an open source organization. So if you look hard enough, you'll find some little hints here and there, but if you like SCI, then you'll like this year's event as well. I, you know, I think, I think you'll see some, some very positive developments in how to, how to measure. Inside your organization and how to make things easier and I think as I teased earlier, we'll be talking a bit more about some of the amazing contributors and experts in our community and how we celebrate those.

But really this year's event is about our community and the stories they have to share. We're interested in seeing what people have done with Green Software in 2023.

The theme of the event is about empowering software practitioners to decarbonise on software and to build communities. And we want to see what people have done.

Chris Skipper: Great. So on that theme of empowering green software practitioners, could you elaborate more on the types of topics or areas that you're particularly interested in receiving submissions for on the talks? etc.

Adam Jackson: Yeah, so I don't want to give too much guidance, because we really want to see what people have brought and their ideas. But, yeah, we were looking for lightning talks and demos covering topics such as how to measure and reduce the carbon footprint of software. Especially if that's using tools that GSF has built.

We'd love to see how people are using SCI and carbon intensity and things like that. We'd love to hear about how people have used the green software principles and patterns to optimize performance and efficiency of their software. We'd love to see demos, so if you've got a demo showing how your software is, is green or what your company is doing to measure.

I saw, um, some amazing demos recently with some fantastic dashboards and I thought, wow, this is amazing. This is green software in action and we can see it. I'd love to see things like that. And if people have some great stories about how they've collaborated and contributed to the community as well.

We'd like to hear that. But really bring your ideas. The one rule we're keeping is to keep things snappy. We're going to keep those talks and demos to about 10 to 12 minutes to allow people to ask questions. And when it's an online event, you've always got to keep things moving. We're looking really for short talks, not long demos.

Bring your ideas to the table.

Chris Skipper: Yeah, absolutely. I was thinking if people wanted inspiration, would looking at the Green Software Foundation's YouTube channel be a good way, good place to point them? Because I only say this because the CarbonHack 2022 event that happened last year, there was some really great submissions done via video.

And I thought the summaries of those projects were really nice and concise the way people do them. I think the videos were somewhere between one minute and I think the longest one might've been about 10 or 12 minutes.

Adam Jackson: Yeah, absolutely. Oh yeah, absolutely. I would definitely recommend looking through that and, um, of course we've also got all of our, um, podcasts on the YouTube channel as well, so if you want to listen to some of those as well, always a good

Chris Skipper: With the transcript, there's no, none of our faces, which is great. Um, you talked about the SOGS report or the state of green software reports as a central topic for people to talk about. What would be a good approach to leveraging this information in their talks during this event? And where could people find more information about the SOGS report?

Adam Jackson: Yeah, absolutely. Um, I'm really glad that you've, um, asked that. So, um, interestingly, we've actually prepared a bunch of resources based on the state of green software for our meetup communities, so, um, I've already done a load of slides. So if people want to nick my slides, then they're in our Meetup documentation, if people know where to look.

Maybe we'll put a link to that in there, in the list afterwards. State of Green Software is a wonderful resource for our members and communities. There's, I think there's about 30 insights on there. We've laid out our State of Green Software site in a really bite sized fashion, so you can look at the insights that interest you and then get some data on that. And that's a really good way of building a talk actually, whether it's for Decarb, or whether you're looking to do a community talk. So take a look and see what interests you, get some inspiration. If you want more information or some slides from me then you can always contact me as well.

But we definitely recommend some people going a bit more in depth into some of the insights that we've recommended on our site.

Chris Skipper: Yeah, so I think if you go to stateof.greensoftware.foundation, you can find the website, and the website is really easy to get around to navigate. I know this from having to research questions for guests about this, but each of those insights, if you click on them, you can just see such incredible detail.

And we actually have a podcast episode about the state of green software report as well, um, which was with Tamara Kneese. Tamara Kneese, who was the lead researcher, I think, on the SOGS report. And she did a really in depth episode, uh, with Chris Adams. So yeah, we'll put a link down to that one as well. Um, Cool.

Uh, so let's talk more about green software communities in general. Green software communities within organizations seem to be a significant aspect of this year's event. We also talked about empowering green software practitioners. Are there any examples of successful communities within organizations?

And if so, how were they built and fostered?

Adam Jackson: Yeah, I think, I think there's actually loads of examples, and some of them are really formal and others are informal, um, and one of the, um, I was really lucky to go to We Are Developers in, um, in Berlin a couple of months ago to talk about green software, and I saw a talk from Amadeus, which is one of our GSF members, and they, um, they, they did this amazing presentation with loads of dashboards, I love dashboards, so I got really excited, but I, they didn't have a formal community, they just had a group that they'd used.

I, I dunno whether they, I think they used Microsoft Teams or Slack and they just had everyone in together that was interested in these topics. And that was building, building sort of green software insights. So it doesn't have to be super, super formal. Um, but everyone does it differently. Um, recently we heard um, from Avanade, um, Avanade are a GSF steering member, um, they've been really consistent. They are a huge consulting organization, so they have to be. Um, they've put lots of effort into helping people build their own local communities within the organization. They've used the tools at their disposal, including they've used, they've used their SharePoint intranet sites to, um, put loads of information about Green Software and lots of links to Green Software Foundation resources - where people can find them. They've also looked at Microsoft teams and created channels for people who are interested and brought 'em together into one digital space. And then they've really thought about how green software aligns with their culture. Um, and what it means for their customers and they've, they've used green software to really improve their business.

I think that's what really makes it so powerful as it's not, they don't see green software as a cost. They see it as an asset that helps them reach customers, helps them do a better job with their customers, helps them find new clients along the way. Um, and, um, you know, and, and they've really empowered individuals to, they haven't said this is the way you have to do it, but they've empowered people to say, start your own local communities.

And they talked about how their team in Paris had decided to do a complete week full of green software activities based off some of the global resources that they'd shared. And I love that, just inspiring other people to take action.

Chris Skipper: Yeah, absolutely. It's interesting that you mentioned that, that people see it are now starting to see it as an asset in those companies. I'm curious as if to whether that should be more of a unique selling point, I know perhaps it's not the focus of the Green Software Foundation, but for those people who are developing green software and who want to perhaps build communities within their organizations, using it as a sales tactic, And perhaps, you know, uh, uh, as they, as the classic phrase always goes, always be closing a closing mechanism in sales to actually get those clients through the door.

Adam Jackson: Yeah, I think it's, um, it's going to be seen more and more, actually, over the next few years. It's not just, it's not just something that gives people the edge. It's going to become a requirement. You know, we're seeing already, um, if, you know, if you're a bank, um, or an, you know, regulated industry, you're already being asked about what your sustainability credentials are. So when you're looking for someone to help you improve your business processes or your software, you're going to want someone that can actually help you be greener from a software perspective. Regulation is catching up and Certainly culture is already there and people are, people see this as important now.

So I think, no, it's not just, it's not even an optional thing. It will be, become an essential part of selling your services in the future. And that'll, um, that will also be to end customers as well. We we're seeing, uh, I dunno, um, the B B C did, uh, a, um, TV show, just a one-off on, um, the carbon impact of streaming.

Um, and this was probably 18 months ago, and they talked about Netflix. I don't want to pick on Amazon Prime. We'll try and name as many as possible. Um, but streaming,

Chris Skipper: There are streaming services that are available.

Adam Jackson: absolutely, yeah, but these streaming services have a carbon impact, and people started realizing that, and it won't be long before consumers start realizing that any interaction they have with software, um, creates carbon emissions.

And they're going to start asking the people they buy services from, how green are you? Yeah, we're going to see, we're going to see a lot more pressure from the market, so to speak, in the future.

Chris Skipper: Yeah. And I think for people that perhaps are listening and thinking, oh, I've stumbled across this podcast and I'm a developer, or I work at software and this actually sounds something that's really could be an opportunity for you a either as a career um, prospect, or just because you're really passionate about green software and the environment.

Um, this, if you are, if you become part of the green software foundation, you are at the precipice of something that's effectively brand new in the space. So my, my suggestion is that you attend Decarb, Decarb software, Decarbonize Software 2023 in order to learn more about it. So with that in mind, where can people, how do people register for Decarbonize Software 2023? And what do they need to do in order to register as a speaker? Is there anything that they need to do that's special or any forms that you need to fill out? Anything like that?

Adam Jackson: Yeah, sure. So, uh, so registration is open now at decarb.greensoftware.foundation. And we'll add that to the links, of course, as well for this episode. Um, our call for speakers is open until the end of September. Um, and we've also got that, that link is on the main registration site. So if you go to decarb.greensoftware.foundation, scroll down, you'll see a bit more information about becoming a speaker. Um, and if you click on that, that will take you to a form and it will ask you what talk, what sort of talk do you want to do? What's it called and what's the content? Um, and yeah, we're really encouraging people to put their idea forward you don't have to put a huge amount of information so you don't have to have a you know a full presentation built or anything like that but tell us about your idea tell us what you want to talk about um and then um if you submit that i think it's the 25th of september is the deadline so you don't have too long um you know, so please put your, your idea forward sooner rather than later, and then we'll, we'll get back to you at the start of October, um, to, um, you know, and if we won't be able to accommodate everyone, if we, if we get loads of responses, um, Decarb, uh, the Decarb event is two and a half hours long, so there's only so many people we can fit in, but for the people that we can't fit in, we'll try and run some additional events over the following few months, and we'll fit all of those talks in as well.

Absolutely. Absolutely.

Chris Skipper: Yeah, absolutely. Or maybe you could become a another guest on the podcast if you're, if you don't make those cuts. Always looking for guests, always looking for

Adam Jackson: Yeah,

Chris Skipper: Awesome. So, that's it for this episode about Decarbonize Software 2023. It looks like it's gonna be a really interesting and enlightening event and we really encourage everyone to register for the event via the link in the show notes below.

Before we head off, we normally end the show with a closing question and for this episode, Adam, I wanna ask you, If there was one tool you could point people to that would really help them start a green software community, what would it be and why?

Adam Jackson: I'm really glad you've asked. Really simple, so head to our meetup guide. We will put the link in the show notes. Um, that is a comprehensive document that we've built that explains how we run our external meetup groups around the world. Um, and I'd also take a quick look at our meetup network as well. Um, I will put the link in.

Um, there is a map that shows where all of our existing communities are and what already exists. Our existing communities always need more support, so if you're in one of the global locations that already has a group, then get in touch with us if you'd like to be an organizer or if you'd like to help that group in some way.

If you are already organizing a green software meetup that's not affiliated with the Green Software Foundation, we would love to help promote your group and build, add that to our map and build our global family of meetups. And yeah, if you're somewhere that doesn't have... Any community activity, then please get in touch with us to, um, you know, to get, to get a group set up.

Read the guides, all of the contact details are in there. Um, we, you know, we want to hit the tens of thousands of Meetup members very soon and we need your help to reach people around the world. You know your local community better than anyone else. So, let us help you build those communities where you are.

Chris Skipper: Yeah, awesome. Thank you. So, we've come to the end of this podcast episode. All that's left for me to say is, again, thank you, Adam, for coming on. It was really great and really insightful. I am sure people are looking forward to Decarbonize Software 2023. Don't forget about those deadlines. September 25th is the deadline for talks and, uh, the event itself is happening on November 16th and you can register right now.

Registration is open. Thanks, Adam. Thanks for coming on.

Adam Jackson: Thanks. Thanks very much, Chris. It was great to come on and talk about community, talk about our Decarb event. Um, hopefully you'll have me again soon.

Chris Skipper: Yeah, absolutely. We would definitely want to have you more on this podcast. If we have more events, uh, for the Green Software Foundation, we'd love the community manager to come on and talk about them. Um, anyway, thank you. Uh, that's all for this episode of Environment Variables. All the resources for this episode are in the show description below and you can visit. podcast.greensoftware.foundation to listen to more episodes of Environment Variables.

See you all on the next episode. Bye for now.

Adam Jackson: Bye.

The Week in Green Software: Complex Carbon Accounting with Gaël Duez

7 september 2023 | 47 min

On this episode of Environment Variables, Chris Adams is joined by fellow podcast host of the Green IO podcast Gaël Duez. Together they will cover the complexity of carbon accounting, new patents around carbon aware programming from Microsoft, and the flight of climate nerds from Twitter / X.com or whatever we’re calling it these days. Finally they share some exciting events from the world of Green Software including some upcoming events and we find out exactly why Gaël is a real-life bond villain!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Gael Duez: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Exploring the complexity of Scope 3 emissions and the responsibility of the digital sector | Wholegrain Digital [5:28]
Introducing Digital Carbon Ratings | Sustainable Web Design [16:21]
How $1.3 billion in new contracts led Hewlett Packard Enterprise to train salespeople in sustainability | GreenBiz [24:46]
Elon Musk is killing ‘Environmental Twitter | The Verge [33:05]
Cloud Native Sustainability Landscape | CNCF TAG Environmental Sustainability [40:28]
Microsoft files patents for carbon capture and grid-aware workload scheduler | Datacenter Dynamics [40:46]

Events:

Apidays London - Sustainability track (September 14th, London on site [43:10]
CNCF Cloud Native Sustainability Week [44:10]
Decarbonize Software 2023 - 16th of Nov (Online) | GSF [44:36]

Resources:

Scoped Emissions as coffee | Chris Adams [8:55]
Net Zero Initiative — 2020-2021 Report [12:33]
Ecograder by MightyBytes [19:01]
Ecoindex.fr [19:11]
HTTP Archive: State of the Web [20:27]
Sustainable Web Design [21:30]
Call for consultation on the first specification for decarbonising use-phase emissions of connected devices | The Carbon Trust [22:01]
Carbon Emissions in Browser DevTools - Firefox Profiler and CO2.js | The Green Web Foundation [22:24]
Fairphone 5 sets a new standard with 8-10 years of Android support | Ars Technica [29:49]
Commown - Cooperative for long-life electronics | Circular X [30:49]
Mastodon.nl | Mastodon Green [36:03]
Climatejustice.social | Mastodon [36:25]
Mastodon.energy [36:42]
Bluesky [36:56]
Ketan Joshi [37:19]
Speakers Bureau | Green Software Foundation [39:59]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Gaël Duez: Financial accountants, they know for ages that one euro doesn't equal to one euro. If one euro is invested, or it's in your account ready to get used, or if you invest it in fees or in wages, it's not the same euro, and it's pretty much the same with CO2. And we tend to compensate everything. And you know, I love John Oliver's quote saying that we will not offset a way out of this climate crisis.

And this is exactly what is at stake here with this so called Scope 4, which is all about avoided emissions.

I'm your host, Chris Adams.

Hello, and welcome to another episode of This Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. In this episode, we're covering the complexity of carbon accounting, new patents around carbonware programming from Microsoft, the flight of climate nerds from Twitter or X.com or whatever we're calling it these days, and where we're finding our climate news instead. Finally, we'll be covering some exciting and interesting events from the world of green software coming up in the coming months. All right, before we dive in, though, let me introduce my guest and colleague for this episode of TWiGS.

With us today, we have Gaël Duez. Gaël, I'll hand over to you to introduce yourself. Thanks.

Gaël Duez: Hi, Chris. A pleasure to be here. Well, I'm Gaël Duez. I'm the founder of the Green IO Podcast, which aims to empower all responsible technologists, an expression I kindly borrow to our host, Chris, when he joined the fourth episode. So yeah, I aim to empower all responsible technologists within the tech sector and beyond to build a greener digital world one byte it at a time. So I guess it sounds pretty familiar to the listeners. And I'm also a former CTO trying to redeem the carbon footprint of its past IT operation, if I dare to say. I now help tech companies deploy sustainable strategies aligned with the Paris Agreement and beyond the carbon funnel.

I also contribute to our community, or at least try to, via public conferences and workshops on digital sustainability, and having the privilege of living in Réunion Island, I'm also the proud dad of a little daughter who enjoys hiking in its beautiful cirques, like we did last weekend, which is why I'm so energized this week.

Chris Adams: Oh, that's really nice to hear. I didn't actually know about that. So for listeners who may not be familiar with Réunion Island, maybe talk a little bit about whereabouts that is in the world, because it is quite a bit further out than I realized when I first heard you tell me where you were coming from in the first place.

Gaël Duez: Yeah, well, the truth is I'm Still mostly working in Europe and with European clients and colleagues, but I live in Reno Island. It's a small volcano island on the north, I would say north, northeast of Madagascar. So I'm based in Africa. But what is interesting is that people often think about it as the tropical islands, so you know, palm trees and beaches, et cetera, et cetera. And actually, it's a very, very mountainous island. There is a 3,000 kilometers high peak called Le Piton des Neiges. And 90% of the island is protected for biodiversity issues, or not issues, actually, because it's not issues yet, but for biodiversity reasons.

So that's pretty interesting island to live, even if we're a bit packed around the shore, obviously, because pretty much all the center is protected, but it's a beautiful place to hike and to do the mountaineering stuff, definitely.

Chris Adams: Wow, cool. Okay, we'll share a link on various mapping tools so people can see Gaël is actually talking about, because when I first saw it, I thought, "wow, that's amazing, it's like I'm speaking to a Bond villain," the first time I saw it. In a good,

Gaël Duez: I hope I'm a bit nicer than a Bond

Chris Adams: villain.

Bond villain a good way, a possibly benevolent dictator of an island, perhaps.

All right. Okay, before we digress too far, let's just provide a quick reminder of this podcast, what we do, and I suppose just the usual boilerplate. So this is a weekly news roundup show. And we're going to cover a series of news stories that caught our eyes that both Gaël and I basically put together over the last week or so.

I realize I didn't actually introduce myself. So my name is Chris Adams. I am the executive director of the Green Web Foundation. We're a non profit based in the Netherlands, working towards an entirely fossil free internet by 2030. And I am also one of the chairs of the Green Software Foundation Policy Working Group. So that's my involvement here.

And also, I am a regular host for the Environment Variables podcast and this podcast here. Okay, then. So we'll cover some stories, and there'll also be a set of extensive show notes with links to all the things we discover and discuss. Alright. So, Gaël, I think you've listened to the format before and you've submitted some of these and you've got a good idea what we talk about.

Is there a particular story you'd like to start with first so we can kind of get into the swing of the show?

Gaël Duez: Yes, indeed. I really enjoyed reading the article from Wholegrain Digital, the well known agency in digital sustainability, about exploring COP3 emissions and the responsibility of the digital sector.

Chris Adams: Yeah, this is the piece by, I think, Marketa Benasek. She's one of the writers at Wholegrain Digital. And this piece is called Exploring the Complexity of Scope 3 Emissions, Responsibility. And there's a couple of quotes which really caught my eye. Essentially, the whole thrust of this article is about trying to give people who work in technology an understanding of how organizations account for, essentially responsibility for emissions, both within their organization, but also outside of their organization. And this quote really leapt out at me. Basically, she's talking about how it's quite hard for you to get the header out. And the quote I like is this one here. So, "in the digital sector where products are often intangible and widely distributed, i.e. through data centers, telecom networks, travel, and so on, attributing emissions becomes challenging."

So she's basically saying, it's difficult to work out who's responsible for some of the emissions when you build a service, for example. She says, like, "many companies struggle to define the boundaries of their responsibility and accurately account for these emissions associated with what they do." And she basically outlines some ways of saying, this is how you can use some of the existing greenhouse gas protocols right now to think about responsibility for this, in particular, the eleventh part of Scope 3, which is related to like use of solar products. So this is one thing that is really interesting seeing agencies talk about this. 'cause typically they've said like, "no, it's not really on us to think about." And Gaël, I'll let you come, come in on some of this if, 'cause I think there's a couple of things that you might wanna share on this and then I'll come back to some of the other parts 'cause I realize you've had to wrestle with some of this stuff yourself as well in some of your work.

Gaël Duez: Yeah, absolutely. I mean, I always say Scope 3 is the mother of all battles, you know. And just to take a very recent example, I was reviewing with a client, a very large European tech company, it's greenhouse gas emissions yearly. So it's a yearly audit and as usual, more than 70% was Scope 3, including AWS solutions, of course.

So I know that we tend to focus in the digital sector mostly on scope two, or actually we want to have the greenest possible energy or sorry, electricity, because most of the time it's electricity. But the truth is, if we really want to make a move on climate change, we need to consider seriously the Scope 3 for everyone.

And you know obviously your Scope 1 and 2 is someone else's Scope 3, so it goes all the way up on the value chain.

Chris Adams: I agree. There's another part about Scope 4, which we'll touch on a little bit later, but it might be worth just briefly, I realize we've just dived straight into talking about scoped emissions, and it might be useful for me to just provide a bit of a primer for people who are new to this field. And like one way that I've used to describe this to nerds is talking about the way that people report emissions for any kind of service is usually in a kind of scoped system if you follow the greenhouse gas protocol, and you can think of it belonging to these lines broadly as Scope 1 is basically emissions from burning fossil fuels yourself, things that go into the sky, Scope 2 is emissions from greenhouse gases from generating electricity that you use, and then Scope 3 is this indirect supply chain emissions, basically all the other emissions that happen in your supply chain.

Now the way that I found most useful when speaking to other techie nerds is scoped emissions communicated through the medium of coffee. So, if you think of Scope 1, Scope 1 emissions is burning fossil fuels to make hot coffee, like maybe you burn gas on a stove to heat up water to turn into a delicious cup of coffee.

Scope 2 might be using electricity to heat up a kettle to make some coffee. And then Scope 3 might be you walking into a coffee shop so that you can have coffee, so you're not burning anything yourself, but other people are doing it on your behalf, so there's a whole supply chain associated like that.

And what we'll do, we'll share a link into the show notes with some helpful diagrams for this, because this was how, I believe, Simon, working on the Green Software Foundation CarbonWare SDK, presented this recently at the Linux Foundation. And it's a kind of relatively intuitive way to start thinking about some of this.

Gaël Duez: I love it. And just to add something, please remember that Scope 1 is not only about burning fossil fuels, they are also methane emissions. And just a quick anecdote, Starbucks' entire greenhouse gas footprint, 20% of it accounts for dairy production. And obviously dairy, it's not only about burning fossil fuels, but it's also the methane emissions from the cattle.

Chris Adams: This is right, yeah, I should have said greenhouse gas emissions, of which fossil fuels are a significant part, but you're right.

Gaël Duez: No, but I love your example. It is straightforward, but we tend to forget all the greenhouse gas and CO2. Obviously, CO2 is the main perpetrator here, so we should focus on CO2 first. But it's good also to remember that there are also players in the game, I would say.

Chris Adams: Oh, great. So now that we've spoken about what scoped emissions are, which is probably what we might have done before if we were gonna provide a kind of preamble for this blog post, there's another really interesting quote for me, which I found helpful, which is when Wholegrain themselves are talking about how they've been struggling with this, and this quote says, "calculating Scope 3 emissions is a challenge for us, ourselves, at Wholegrain Digital. Scope 3 emissions of the products we consume, such as software subscriptions, are really hard to calculate, but it's also not exactly clear whether we should take responsibility for our clients' websites during use." So while, technically, these emissions belonged to their clients, or their website's visitors, we also see it as our responsibility to assist in reducing the environmental impact.

They say, like, "digital agencies that make polluting websites should take responsibility for this." And the rest of the post ends up talking a little bit about ideas which are kind of beyond your value chain, and this is like the impact that you might induce, and I think they refer to this as kind of Scope 4, and I've heard other people talk about this as Scope 0, and this is a bit of a kind of wild west right now.

Because this is essentially referring to the idea that if you're building a website that makes it easier for people to, say, hire a cab or shop faster, then there's an impact from you speeding up that activity. And I think this is something that you've been thinking about as well, right, Gaël?

Gaël Duez: Yes, absolutely. And can you indulge me to be the villain here? Because if I'm a James Bond villain, I'm going to play my role. Please, please everyone forget about Scope 4. I really mean it. This is the worst possible naming convention that we could find. I'm really concerned about the discussion around this so called Scope 4, which actually is all about avoided emissions.

How the tools, the services you provide to your clients help them avoiding emissions. But, when we use Scope 4, there's emissions in the same bucket as Scope 1, Scope 2, and Scope 3. And to be honest, I am a big fan of the Net Zero Initiative, which provides a clear dashboard with its three pillars to how a company should contribute to the global objective of carbon neutrality and where tons of CO2 doesn't compensate, 'cause you know, financial accountants, they know for ages, um, that one euro doesn't equal to one euro.

If one euro is invested or it's in your account ready to get used, or if you, you invest it in fees or in wages, it's not the same euro and it's pretty much the same with CO2. And we tend to compensate everything and, you know, I love John Oliver's quote saying that "we will not offset a way out of this climate crisis."

And this is exactly what is at stake here with this so called Scope 4, which is all about avoided emissions. And if you deep dive a bit on the Net Zero Initiatives, I love their approach because it's a dynamic approach, not a static one. No company can reach net zero. That's not possible, because that's not scientifically agreed. What can be agreed is net zero in a closed environment, and the only closed environment we're talking about is planet Earth. So companies contribute to reaching global objective of carbon neutrality, and they've got three pillars to do that. And the first one, you beautifully described, Chris, is pillar A, which is reduce your own company emissions.

Then you've got another pillar, which is reduce others' emissions. And it can be either by helping your suppliers or your clients with your services or whatever solutions you want to deploy to reduce their own emissions. And this is where we tend to hear now this Scope 0 or Scope 4 approach. For me, it's really all about avoided emissions.

And of course, you've got also pillar C, which is removing CO2 from the atmosphere. And these three buckets should be counted and communicated in three completely separate way. And if you think about pillar C, it's a bit like the 1% for the planet initiative. Some company, believe me, marketing people, they will definitely know how to positively communicate on it. Could say, you know, "we allocate 1 or 2% of revenue, or whatever to financing climate technology to remove CO2 out of this atmosphere," but these tons of carbons, they will not offset anything.

And I think we really need to be cautious about using three different buckets to track how we contribute to global neutrality. Sorry if I'm a bit ballistic about it.

Chris Adams: That's okay. We have this podcast to have people with strong opinions and they are able to compete to share them and our listeners are able to decide how they feel or how they want to respond to that stuff. So you mentioned a couple of things about measuring the environmental impact of some of this. And I realized that you've also mentioned just before this call that there's some other groups looking at some of this as well.

And we're going to talk a little bit about that in a second with the next story. But the thing that might be worth just briefly sharing with people is that the GSG protocol right now is in the process of being updated and we've shared a link to basically an update from the World Resources Institute specifically about how they're planning to make some of the updates, because they've done a massive survey with thousands of responses from companies, non-profits and groups like that and we shared some links to basically the presented findings so far and also some of the early things talking about both Scope 1, Scope 2, and Scope 3, and how different companies and organizations are actually saying, "this is how they should be changed to more accurately represent the physical realities of what's happening in the world."

So should we go to the next story, Gaël? Because this one feels like it's tied quite tight to what you were just speaking about. There's idea of measuring this, trying to come up with some other ways of accounting for the emissions in a particular sector. And this link is from the Sustainable Web Design.

There's this introduction of digital carbon ratings that has come out now, and I'll just share a quote from the piece and then I'll have a bit of space for you, uh, to talk about some of this, Gaël. So the general idea is that, The quote I'm going to use is, "we propose a simple digital carbon rating system that follows the original principles of sustainable web design and aims to make website sustainability much more intuitive and accessible for a wider audience" and essentially the short version of this is that they're taking an idea of the average website or looking at a body of an existing data set that is generated by the HTTP archive to get an idea of how large and how small various websites are across this data set.

And they've created a kind of rating system based on where these fall in the distribution. So the fastest and the smallest sites are. Kind of graded at, like, an A or an A+ all the way down to an E, basically, or something along those lines. And this is intended to be used to provide some kind of rating, somewhat like an energy star rating, essentially, so that if you have a website, you can say, "well, we want to be building a, at least A website," or "we're at a D, we should be pushing to get ourselves to a B," for example.

I think I'm gonna open up for you to kind of have a bit to talk about some of this as well, actually, before we go into this in a bit more detail, because my organization was somewhat involved in this, and it's been something that the groups have been working on for a while, and I think there's lots of places this could go in, and it's the first time I've seen people really try to do this and create a kind of shared grading system for this.

So yeah, Gaël, over to you, man.

Gaël Duez: So I love this one because obviously, we need all those initiatives. But I mean, to be honest, I always feel a bit schizophrenic about the multiplication of those initiatives and ratings because we've seen others popping up around the world as well. And don't get me wrong, if you're a web developer based in a dark red state in the US and working in a pickup factory with a CEO watching Fox News on loop, you have my admiration and my full support if you manage to talk about this rating tool and to implement it somehow on your website.

So big, big, big kudos. And I think this is why this kind of initiatives are great. Still, it remains an awareness raising tool. I love the simplicity of the rating and the benchmarking with the HTTP Archive database because it could trigger some healthy emulations also. So really enjoy this part, this approach.

However. It's based on the single and highly debated proxy for energy consumption, which is data transfer. So for web professionals, I would rather advise people to use Ecograder created by MightyBytes, which has several components and not only page weight, or even better, the open source initiative ecoindex.fr, which also try to incorporate other environmental impacts like water. Now, what I believe is that all these initiatives, they're trying to fill a vacuum and this vacuum is the lack of commonly agreed and understood metrics when it comes to how carbon intensive or even how environmental intensive is a website. And this is why the job started with the W3C community under the lead of Tim Frick and especially Lucas Mastalerz, lead the metrics workgroups in this W3C sustainability committee is so important. We need to find Some common way to measure this different environmental footprint based on the latest scientific data available.

Until we do have this, I guess the more the merrier because you want to approach these issues under different angles. A super simplistic one like the one you just described in this article, Chris, and it will be very useful for some people in some situation. But other tools are needed for professionals to really deep dive on where they would have a big impact.

Chris Adams: I think that's fair. Because this is largely looking at one indicator that has been relatively easy to capture and put into a data set that can be made available, and the underlying data set from the HTTP archives. This is also used in the State of the Web report that came out last year, which had, for the first time, a really dedicated sustainability chapter.

So, in my view, I think this is really encouraging to see this and having some kind of rating systems is one way to make some of this a bit easier for people to understand. There's a couple of things that it might be worth briefly touching on for this because the actual grading is pretty, it seems pretty hard to get an A.

So if you want to have an A+, your website needs to be within the top 5% of all the websites that you have here. And, pretty much, it stops off at like E, which is around 50%. So if your website is the average, then you have a long way to go just to get up to an A, for example. And this current has been shared for feedback from people to see how people respond to this and see where they can go with some of this.

So I need to share that this is an early thing. There is a call to kind of get some more input from this and people can go to sustainablewebdesign.org to use the contact form to actually provide some feedback and share something for this. The other thing that I'll just touch on is that this isn't the only single way for understanding the environmental impact of digital tools.

There is also some work with the Green Software Foundation to come up with this metric called the Software Carbon Intensity Spec. This is one tool which is currently in use. There's also some work at the end-user side, which has been one of the contentious areas. Carbon Trust literally last night said they're doing some new work to come up with some standards for understanding and accounting for the environmental impact of end-user devices, 'cause typically this is one thing that's been very, very hard to use and they've got some large companies like Amazon and Meta already online, on board for that. So I suspect that's gonna be a thing that people see more of.

The other thing that we might share, so this is me from the small nonprofit that we work in, we did some work with the Firefox browser to essentially build some end-user carbon emissions specifically into that, and we've got a blog post that I'll share a link to this, and you mentioned ecoindex.fr, a French tool, and EcoGrader, which we've shared some links to there.

Now, as I understand it, Tim and the team at MightyBytes that worked on EcoGrader, they were involved in the creation of these digital carbon ratings. So they are involved in this. And there is an intention to kind of make this somewhat wider. But there is a tradeoff right now about saying, 'what kind of factors do you include and how easy do you make this for who to understand?' Because even just moving on from just thinking about money is quite a jump.

So when you start talking about carbon and water, and the resource depletion from the earth, and so on, it's a whole bunch of extra things which makes it really complicated. So, yeah, those are the things I might say as a response, that might provide a bit of extra context for this.

Gaël Duez: Fully agree with you here. It's really this dual approach, like, you've got communication and awareness tool, and this is super important that they are super simplistic, easy to understand, easy to grasp, because you still meet, on a daily basis, thousands of people who told you, "oh, really? My website pollutes? I wasn't aware of it. Oh, I didn't even think about it." And then on the other end, you've got web professionals who are already a bit aware of it and they, they're more like, "okay, but what, what can I do? Shall I reduce the JavaScript? Is it a question of image sizing? Is it a question of data transfer? Uh, shall I take into consideration, obviously, the obsolescence of the end-user tool?" Etc, etc. And it is a large spectrum, as you said, and we need to cover all of this. I think the main battle today is really about, you know, speaking the same language. And that will be awesome if all these tools at some point, hopefully, under the umbrella of the W3C, could agree on sustainable metrics that you, you know, kind of zoom in or zoom out, depending where you are on this scale.

And I fully agree with you that just moving away from money is a big challenge at the moment.

Chris Adams: And that actually is a nice link to the next story we had, because I was not expecting this, but this really caught my eye. So this is a story, how $1.3 billion in new contracts led Hewlett Packard Enterprise to train salespeople in sustainability so I wasn't expecting salespeople to be the kind of vanguards of sustainability in the technology sector this is basically a piece that will share a link to from greenbiz.com which is basically, it is a little bit kind of like puff piecey, but it's essentially, some folks at Hewlett Packard Enterprise, they're people whom we sell loads of service. They're basically saying, "we're training our sales team to talk about circular economies and energy efficiency and teaming them up with the sustainability team, because we found out that that's the thing that CIOs keep asking for and they're often not getting very convincing answers from this" and there's a couple of things that I thought was quite interesting is that, so salespeople typically tend to work on commission so they get a base salary and then they get a kind of chunk of their money in the form of commissions on product sales and there's a piece which talks a little bit about how they're compensating various staff for this or linking sustainability performance to compensation.

And this story talks a little bit about how the executive committee are, the compensation for them is tied to the company's performance against net zero goals. So this is something that is, in my view, kind of interesting because they're talking about things like energy efficiency, recycling content, stuff like that.

And uh, they've also shared a goal, which is they're trying to cut operational emissions by 70% by 2030. So this is relatively ambitious, but the operational emissions part might be the easy part to, actually, hang on. No, we're talking about people who make servers, that may not be the case. This very much is a case of where the big emissions tend to fall is whether it's in their supply chain or whether it's in inside the organizational boundary.

But this idea of actually building it in and actually having the salespeople talk about this gives you an idea of how, like, there is need or interest in having some shared language so that we can actually have essentially discussions outside of our little niche, basically. And I think this is something that you've got some experience with as well, Gaël, right?

Gaël Duez: Yeah, absolutely. Let me share you an anecdote. Last year, I was facilitating a digital collage online workshop for Evonex. Evonex is a pretty big IT company. They specialize in providing IT equipment, you know, to big companies and the attendees were mostly sales and marketing people scattered all over the world, I had literally people from four continents. And during the workshop, they started to get ballistic about it, like super enthusiastic, because the digital collage workshop focused a lot on embodied carbon footprints, as well as, you know, water footprint and material footprint, often called EMIPS, and they immediately could see the benefits, uh, in their sales pitch about, hey by the way, by renting equipment, by making sure that, you know, we will take care of, um, the end-of-life and we will reuse it over and over and over again, you are actually part of the, a virtuous cycle. You, you're getting closer of the much needed circular economy. And it was not even mentioned a link with their commission. It was just like, wow, that's a good sale pitch and I'm very happy to get all this valuable information because that will help me get more contracts.

Chris Adams: All right, you said something interesting about the model people are using, so basically, you're paying to have access to it rather than owning the actual tin itself, basically. That's what they're doing. Was that a trend that you saw, or was that a thing that people already are using right now in this scenario?

Gaël Duez: You know something, it's quite funny, when I started to deep dive in digital sustainability, everyone told me about the massive shift in business model which is needed from makers, like Apple, Samsung, etc, etc. And fun fact is, I started my professional career in the payment service industry. And one of my job was to run a small business unit, renting payment terminals, because, you know, when you're a merchant, In Europe, in 90% of the case, you rent your payment terminals from your bank, you know, there's kind of the absolute norms.

And the fun fact is, it provides a clear alignment of needs between banks and merchants. People want to have resilient and long lasting good bank, they don't want to have to send technicians to repair the device all the time. And you know, the truth is, everyone makes money with it, with this business model.

Because last time I checked, banks are not philanthropic institutions at all, you see. So, so I think, at some point, a shift from owning an electronic device to renting an electronic device will become more and more the norm, first in the B2B sector, and then at some point, why not, in the B2C sector as well. And that is a dramatic change because you close the loop. And when you design your product, you need to make them easily repairable and easily recyclable or reusable first.

Chris Adams: Okay. All right. Thanks for that. I didn't realize that was where you started out, actually, Gaël. You also made me think about some of the most recent announcements from Fairphone as well, because they announced recently they're pushing out a phone, the Fairphone 5, the newest one they're talking about. I believe they're talking about having a guarantee of between 8 to 10 years for a smartphone which is kind of mind blowing when you consider the kind of yearly kind of obsolescence process that you've typically seen before. We'll show a link to that because that's pretty wild and that's the thing that's quite interesting with Fairphone in this context is, they sell some of the devices but they also talk about some of the difficulties with managing both a kind of rental model where you're incentivized to kind of make sure that you capture the value and make it come back to also having a thing which allows people to kind of feel like they own it and they can fix it and they can do all these other things because different incentives come into play when you think about an entirely rental based model.

So that's something that we'll share some links for people who are interested in learning how other people are wrestling with some of this.

Gaël Duez: Yeah, I agree with you. Actually, I rent my, my Fairphone now from, from a company called Common because I really believe in this renting model, but it's more with a professional angle. Uh, and it's true that I think we need to be able to cover different needs from different people, and that's great. I mean, if you want to own your smartphone, what you've got the right to demand is to have it repairable, to have spare parts, to have, uh, accessible notice, to understand how to repair it, et cetera.

And if you want to rent it, obviously you want to be able to update the operating system and not, not to face a software obsolescence, et cetera, et cetera. So, I think it's not a one size fits all approach that we should embrace, and I think Fairphone is doing a very, very good job embracing different aspects of the spectrum.

They've got this five years guarantee on material, and now they claim eight years guarantee on software, which is mind blowing, as you say.

Chris Adams: Yeah, I'll share the link to the piece in Ars Technica which showed that, because I read it last night, and I was, when I was doing some research, I thought, wow, eight years, they've had to use a particular industrial chipset for IoT rather than consumer technology, because the assumption around consumer technology is that it won't last long enough for you to have this kind of warranty, but it's a good piece, and it really caught my eye.

All right, shall we look at the next story? Go on.

Gaël Duez: Just just just a side note Chris, you and I, we're not that young, unfortunately. So just remember that in the IT's in the 80s, sorry, just remember in the 80s, that it was very common to own for five years a piece of IT equipment. Actually, the average lifespan was close to 10 years. So, you know, it's, maybe it's getting back to what used to be normal and what used to be a sensible thing to do when you know how much energy and materials and water has been used when you build those equipments.

Yeah,

Chris Adams: Make it in the first place. All right, okay, that is, um, thank you for reminding me of the gray hair in my beard, Gaël, I appreciate that.

Gaël Duez: No, sorry. I don't want to be the villain in this episode, I'll stop. I was very positive here. I've got only nice things to say, and that's going to be the same for the rest of the show. Sorry.

Chris Adams: I can dream of going into becoming a silver fox, Gaël, that's my dream. All right, shall we look at the next story? Okay. So this one is from theverge.com. This says, "nearly half of environmental users went inactive after Elon Musk's takeover of Twitter took place, research finds." So this one is a story partly because there have been some questions about, okay, where do you get your news around climate these days and Twitter, and there was a real term called Climate Twitter.

The quote that I'll share with you is it's basically "almost half of environmental Twitter has vanished from the platform that's now called X, new research is showing. A wave of environmentally oriented users abandoned the site after the takeover, according to a study published this week by the journal in Trends in Ecology and Evolution."

And uh, I share this because we have seen an uptick in essentially climate denial accounts on this, but I figure this might be a nice way to talk a little bit about, okay, well where are they all going? Where do you find the news? 'cause I used to use Twitter a load to keep up with lots of news in this particular field, and I found it a bit harder and I figured, I wonder if you might be having the same experience yourself, actually, Gaël, and maybe we could talk a little bit about where we are looking instead. So if people listen to this podcast, they might find other things that catch their eye, or just talk about some other experiences of what we've seen.

Gaël Duez: I must admit that I've never loved Twitter. I tried, and just the idea of having to describe something complex, most of the time systemic issues in a few hundred words, characters, sorry, I've always struggled with it. So I was a very reluctant Twitter user, but I'm not proud to say that today I'm a very intense user of LinkedIn.

Okay, I know it's not necessarily the best platform ever, but I recall that I follow a lot of thought leaders in sustainability, in green IT, etc. on LinkedIn, and there are a lot of people doing a very decent job crafting very in depth articles, sharing resources, etc. I'm not the most happiest person on earth on the LinkedIn algorithm, obviously, so you need to do a lot of fine tuning to make sure that it's not a post about pack of wolves and how agile your organization should be. But I'm using LinkedIn quite a lot, like a million times more than Twitter. And then, of course, I use a lot of newsletters and other community. I could mention some of them if you want.

Chris Adams: Twitter's loss is LinkedIn's gain, basically, in this scenario here, yeah? So it's not particularly cool, but it is useful, and you get the information that you want to there, right?

Gaël Duez: Absolutely. And, you know, we need to take a bit of time to think and write when it's about climate change or environmental crisis. So I better like the long format that you will find most of the time on LinkedIn, rather than super short tweet and then all this ego battle, etc, etc. But don't get me wrong, you've got plenty of ego battle in the LinkedIn comments as well.

Chris Adams: Yeah, that's what I was thinking about as well. So I'll share some experiences I've had. I've been using Mastodon, probably I started using a bit more of it in maybe October, November, and I've been on mastodon.social and there are some really dedicated instances like versions of something like Twitter, so there's a Mastodon Green, which I know that quite a few people have moved to who I used to see being active on Twitter.

There's another one, climatejustice.Social that I've seen a few people being active on as well. This is one thing that's kinda nice, is that because it's federated, you see different groups that you didn't even know existing, or like little communities, that part is really kind of highlighted rather than it just being like climate Twitter, for example.

I also am experimenting with an account on a place called mastodon.energy, which is where lots and lots of really hardcore energy nerds have been moved to. So the people who I used to follow to kind of keep up with the insights there, I've seen a few people there. The thing that really surprised me though was how strong the turnout uh, on Bluesky has been for loads of climate people, so loads of the people who are not necessarily like super climate techie people, but talk about the kind of climate in the widest term, a bunch of people have moved to Bluesky, but because you need an invite to get on Bluesky, it's actually quite difficult to see any of that stuff.

And when I realized, "oh, that's where a bunch of them are," it really, really blew my mind. There's a bunch of other things that I think a kind of interesting, I haven't really used it very much yet, but this whole idea in Bluesky where you can pretty much create your own algorithms and there is an easy way for people to kind of create algorithms themselves that you might opt into to follow is interesting because there is a Greensky feed maintained by one Ketan Joshi who is a relatively well known climate writer, which is also worth looking at.

But there's also a few newsletters as well though, I think you mentioned before as well, and it might be worth just briefly talking about some of that because there's one or two that I found super helpful in this context.

Gaël Duez: Well, I'm going to mention two because in the first one, you will obviously not mention it, but the Green Software Foundation newsletter is gold. And I would say that the Climate Action Tech newsletter and community as well is gold. The Slack workspace of the Climate Action Tech community is where I find maybe 50, 60% of all my resources.

So big kudos to them. And I think it It's worth having a look at it. The issue I've got with these newsletters or these Slacks, I mean, it's not an issue, but it's, once again, all the feeds that you've mentioned, the Mastodon.green, the BlueSky, et cetera, the problem is it's very easy to fall into information bubble.

And don't get me wrong, that's very convenient. I mean, if you want to have scientifically supported information on energy transition or something very specific, ah, you don't want to enter a debate with some, you know, die hard, climate denier, whatever, et cetera. You just want to be with your, you, you know, with your people, with your folks, and then you will have a very in depth discussion.

Still, I also believe that we need to have these discussions happening in the open space. And today, this is why I was mentioning LinkedIn and some people are still using Facebook or Instagram a lot or YouTube even for these reasons that it's different because this is where like everyone is.

And this is why I believe we should still have some activities going on, on the main platforms, whether we like them or not. So it's really, I would say, two sides of the same coin. And the last one, which is very related to LinkedIn, Facebook or whatever, is where do professional people meet? And they meet in conferences.

And this is also where more and more, I mean, this is what I love when in this podcast you share at the end the link to various conferences is that in every professional conferences, we should be talking about sustainability, we should be talking about climate change, and once again, I'm going to say, instead of you, because it will sound a bit less self promoting, but the big kudos to the Green Software Foundation Speaker Bureau to make sure every professional events worldwide has access to speakers that will be able to talk about climate change, digital sustainability and all the environmental crises.

I think it's very important to be, also, where non truly aware people are.

Chris Adams: I think that's fair. I think you do need to find a balance between those two things.

So there's one thing I'll share just very quickly. We'll share a link to the cloud native sustainability landscape. That's kind of helpful in my view, because this is one place where a bunch of this research has been put into a kind of publicly accessible place and it's a nice roundup of all the stuff that's happening in this field.

We'll share a link to that. There's one story we have left and we're gonna do a quick roundup of the actual events we have coming up here. So we'll talk a little bit about patents, uh, Microsoft filing for patents around grid-aware carbon computing and ware computing specifically. Gaël, do you wanna briefly touch on this one here?

Because I think long and short of it is that we've been talking about carbon aware software for a while and there is a. peace in data center dynamics talking about how Microsoft have recently filed a patent specifically for this and I figured give a space for you to kind of provide some of your reckons on this as well because this in my view shows that okay people aren't just doing it just because it's a nice thing they think there's actually some value inside this and I think this is something that you were talking about briefly before as well.

Gaël Duez: Yeah, I totally agree with you. You know, we need to make the circular economy and soon the regenerative economy attractive for investors. So, hopefully, investors in the short future will truly embrace the triple bottom line because of new regulations or pressures from their stakeholders, whatever, etc.

But, you know, still, in the triple bottom line, there is still the planet and people, but also P, the P of prosperity, which remains so it, it will require investments to be viable. So it's a very positive sign to see climate tech being patented. Actually, I would rather have it fully open source, but this is the world where we live in.

So I think it's a very positive sign that, you know, you can make money by doing good things for the planet or the people. And the only caveat in this specific story that we shared, is making sure that the impact happens over the entire life cycle, and not only during the usage phase. So it is not that what we see sometimes, what I call climate tech distraction.

Oh, we're gonna remove CO2, but at the end, manufacturing and using the device emits more CO2 than what is removed from the atmosphere. But once again, there is a very positive trend to all this lifecycle analysis and I know that people in climate tech are more and more aware of it and take care of it.

Sometimes even multi criteria lifecycle assessments.

Chris Adams: Okay, thanks for that, Gaël. For people who are curious, we'll share a link to the article, plus the patent applications for this specifically, because yeah, I didn't know about this until seeing, "oh, that's why they're talking about a bunch of this stuff." So, Gaël, I believe there's a couple of events. Do you want to talk about the first one that's on this list?

Gaël Duez: Yeah, oh, absolutely. Apidays London, and especially the Sustainability Track. So first of all, Asim Hussain, the Green Software Executive Director, will be a keynote speaker. So I'm super proud of it. And I'll have the pleasure to host the Sustainability Track for the entire day of the 14th September, with la crème de la crème of UK green IT experts and climate activists.

And yeah, some names are pretty familiar to the people listening to the podcast, but we'll have Tom Greenwood from Wholegrain Digital, Sarah Hsu from the Green Software Foundation, Sandra Pallier from Climate Action Tech, Sandra Sido from the Climate Peach, Robert Price, Mark Butcher, Arwel Owen, and many more.

So I hope that I will see many of you there. It's a great event.

Chris Adams: Oh wow, I didn't know that Mark Butcher was on that as well actually. He's a really interesting person to follow on LinkedIn for catching some of this.

Gaël Duez: I do, I love his LinkedIn posts.

Chris Adams: Okay, alright, there's love for you, Mark, going out. Okay, the other few things I'll just draw people's attention to briefly. Cloud Native have a Sustainability Week taking place in October. This is actually a distributed remote event. There's a CFP open, so if you have a talk prepared, then there's still space to do it and it's happening all around the world.

We've shared a link for that. So there isn't, isn't one particular date that's happening in October. And then finally there's an event in November that I'll let you talk a little bit about here actually, 'cause this is one from the GSF, uh, Gaël, do you got this one?

Gaël Duez: My pleasure. So it's Decarbonize Software 2023. So it will be the 16th of November. It's an online event. And I think that the registration is open and it's really the annual event by the Green Software Foundation showcasing the advancements in green software by the community. So I'm really looking forward to watching this one because, you know, I don't know if you remember in 2022, it was an incredible event where the Green Software Foundation announced the Software Carbon Intensity Specifications, the new Linux training program, etc. And actually, if I understood well, the last week episode of the Green Software Foundation, the SCI specification is about to be ISO compliant.

So I expect some big announcement in this 2023 edition.

Chris Adams: That's good. I'm expecting some good things out of this as well, actually. Thank you, Gaël, for covering this. Gaël, this has been loads of fun. I really enjoyed you coming on, and I really appreciate you providing all the actual kind of insight that you did have for this. So, thank you again, man. It's really nice to catch up with you again, and this has been loads and loads of fun.

Gaël Duez: Thanks, Chris. Yeah, it was awesome. It was good to be on the other side of the microphone and a true honor to join your podcast, you know. I can die in peace now. I've been on the environment viables.

Chris Adams: All right, well, thank you very much for that, and... I'm going to let you go to enjoy your paradise island for the rest of the day, OK? Take care of yourself, mate.

Gaël Duez: Take care.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again, and see you in the next episode!

We Answer Your Questions Part 2

16 augusti 2023 | 46 min

Host Chris Adams is joined by executive director of the Green Software Foundation, Asim Hussain as they dive into another mailbag session, bringing you the unanswered questions from the recent live virtual event on World Environment Day that was hosted by the Green Software Foundation on June 5 2023. Asim and Chris start with a discussion on the complexities of capturing energy consumed by memory, I/O operations, and network calls in the SCI. They explore real examples of measuring SCI on pipelines of CI/CD, showcasing projects like Green Metrics Tool and the Google Summer of Code Wagtail project. The conversation shifts to the carbon efficiency of GPUs and their environmental impact, touching on the tech industry's increasing hardware demands. They also address the potential for reusing cooling water from data centers, considering various cooling designs and their impact on water consumption.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Asim Hussain: LinkedIn / Twitter

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Questions:

SCI is not capturing energy consumed by Memory , I/O operation, network calls etc. So what is your take on it? [3:27]
Does the GSF have any real examples of measuring SCI on pipelines of CI/CD? [7:15]
What is the carbon efficiency (or otherwise) of GPUs, say, onerous compute vector search? Is that good for the environment? [23:40]
Can the cooling water for data centers be reused? [36:28]

Resources:

Software Carbon Intensity Specification | Green Software Foundation [4:14]
CO2.js | The Green Web Foundation [6:20]
Wagtail CMS | wagtail.org [9:57]
Green Metrics Tool | Green Coding Berlin [11:09]
Eco CI | Green Coding Berlin [11:56]
Wagtail 5.1 gets a bit greener and leaner [16:06]
Kubernetes Power Manager | Intel - Marlow Weston [19:47]
Intel Power Optimization Library | Intel - Marlow Weston [20:59]
Reducing the Carbon Impact of Generative AI Inference (today and in 2035) | Hot Carbon [27:53]
Beyond ChatGPT: The Future of Generative AI for Enterprises | Gartner Report [35:01]
The AI startup outperforming Google Translate in Ethiopian languages | Lesan [35:36]
With Google as My Neighbor, Will There Still Be Water? | AlgorithmWatch [42:30]
The mounting human and environmental costs of generative AI | Sasha Luccione | Ars Technica [43:27]
Branch Magazine New Edition [43:51]
Green Software Foundations Discussions on GitHub [44:50]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Asim Hussain: We couldn't have done this two years ago. I feel like so many pieces of the puzzle are now coming into place, where people can really very easily, with an hour's worth of work, measure the emissions of a piece of software. Basically, the dream world I have is in six months time, thousands of open source repos all over the world just drop a configuration file into the root of their repo, add a GitHub action, and they're measuring an SCI score for their product.

Hello, and welcome to a special Mailbag episode of Environment Variables. This is our second installment of the format, where we bring you some of the questions that came up during the recent virtual event hosted by the Green Software Foundation on World Environment Day back in June. If you missed our first episode from this mailbag format, feel free to jump back when you'll see some of the other questions that came up and some of our eloquent and possibly not quite so eloquent answers as we ran through that. Today, we're going to run through a few more questions. And as ever, I'm joined by Asim Hussain, Executive Director of the Green Software Foundation.

Hi, Asim!

Asim Hussain: Hi Chris, how are you doing?

Chris Adams: I'm not too bad. A bit grey outside over here in Berlin, but otherwise not too bad really. Okay, before we dive into this, the questions we'll run through. If you're new to environment variables, every time we record one of these, we show extensive show notes with all the links to the papers and the sources and the things that we do have.

So if any of this has piqued your interest, there will be a link that you can jump into to basically continue your nerding out about this particular subject. And I think that's pretty much it. But before that, actually, maybe we should introduce ourselves, actually. Asim, I've introduced you as the executive director, but I suspect you might want to say a bit more about the Green Software Foundation, what else you do when you're not working at the GSF?

Asim Hussain: Thanks. Yes, I'm the Executive Director of the Green Software Foundation. I'm also the Chairperson of the Green Software Foundation, so I hold both roles right now. Yeah, I've basically been thinking about software and sustainability as Chris for quite a few years. Outside of the GSF, I'm also the Director of Green Software at Intel, where I try and work through an Intel strategy regarding, you know, greening of software and helping there.

Because, you know, the only people who buy stuff from Intel are people who run software.

Chris Adams: Thank you very much for that. We'll have this and better revelations and more insightful revelations coming ahead.

Asim Hussain: It gets better than this,

Chris Adams: Yeah. Yeah, my name is Chris Adams. It's a little bit Monday this morning, it seems. I work at the Green Web Foundation, which is a non profit based in the Netherlands, focusing on reaching an entirely fossil free internet by 2030. And I'm also a maintainer of a library called CO2.js, as well as being one of the chairs of the policy working group inside the Green Software Foundation. I'm also the regular host of this podcast specifically. Should we dive into these questions for the mailbag?

All right.

Asim Hussain: Let's go for it.

Chris Adams: All right.

So the first question that came through was one about the SCI. The question is, this SCI is not capturing energy consumed by memory, IO operation, network calls, etc. What is your take on it? This is a question from the World Environment Day thing. This might be a chance to explain what the SCI is, because as I understood it, it does capture

some of that stuff,

Asim Hussain: yeah, my answer on the day would have been like, huh? Yeah, it does. Or something a lot more eloquent than that. But yeah, this is Software Carbon Intensity is a specification being built by the Standards Working Group in the Green Software Foundation. It is almost in ISO. That is our goal for this year is to really go through that process.

Chris Adams: And just to jump in, ISO is the International Standards Organization.

Asim Hussain: Yes, that's the one. Yep. And what it is, let me just very quickly say what it is. It is a method of measuring software carbon intensity, which is a rate. If you listen to a podcast, it'll probably be carbon per minute of the listen. It's a rate rather than a total. Other kind of really in a standout aspects of it are that it's been designed very much by people who build software.

And so it's been designed by people who actually build and measure software to act as a good metric to drive reduction. So make sure that inside it is included aspects so that if you did things like move your compute to a greener region, or you move your compute to time when it's greener, or things like that actually would be recognized in the calculation.

Whereas, for instance, if you use the GHG protocol, oftentimes stuff like that isn't factored in and you can do carbon air computing to the, to the cows come home but it wouldn't really affect your GHG score. That's some of the aspects of the SCIs, very much built that way. Now, what I will say is if you actually look at the SCI equation, it's very simple.

You basically per hour, so it's always what we call per hour, so per minute might be the hour. Or per user, user might be the hour. So per hour, you have to figure out how much energy Is consumed. You have to figure out how much, what we call embodied carbon, so how much hardware is being used and if you're, if it's per minute, then you figure out how much energy consumed per minute.

If it's per minute, you just try and figure out how long is this piece of hardware normally used for and divide it by and obviously you get per minute. Then the other thing you also factor in is thing called I, which is the grid emissions factor. So how clean ditch is your electricity, any factoring or what?

Whatever it is for that period of time with electricity. And the key thing there is that's it, and so therefore, It includes everything. It doesn't exclude memory, or I/O, or network, because it's just energy, hardware, and grid emissions, and so as long as you've got some values for that, for your memory, for your I/O, for other things, you can do it. What I will say to answer, I think maybe, I don't think this was in the spirit of the question, but I think it's clear to it, measuring is hard. It's really hard. Like Chris, you've got co2.js And that does a great job of kind of network, but even then you have like multiple flags if you wanna use it in this mode or this model or this assumption.

Like, I love, I use it all the time these days. What did you say, like, all models are bad, some are useful? Yes, I do think that calculating an SCI score, which includes memory, IO, network calls, all the other factors in software is challenging, and I will acknowledge that, but it's also something that a lot of people are working on, and I think we're working on that with things like the impact engine in the foundation, and Chris, you're working on it with the co2.js.

Arne is working on it from Green Coding with those models. Yeah.

Chris Adams: with GMT, the Green Metrics Tool.

All right,

Asim Hussain: metrics, oh yeah, yep.

Chris Adams: Hopefully that should give plenty to refer to. I'll add a couple of links to what this SCI is to make that a little bit clearer, so for people to understand what that might be for that question. Should we jump on to the next question actually, Asim?

Asim Hussain: Yeah, sure,

Chris Adams: Does the GSF have any real examples measuring the SCI on pipelines of CI/CD? That's a soup of different letters there, but as I understand it, the GSF being the Green Software Foundation, SCI being the Software Carbon Intensity is a way to measure the carbon footprint, and CI/CD being continuous integration, continuous delivery, like automating the process of getting software out for people to use, all

Asim Hussain: mm hmm, yep,

Chris Adams: All right, so now that we've explained what the question meant and unpacked some of those, all those TLAs, three letter algorithms, do you want to have a go at this one? Because I can add a little bit myself with some recent work that we've been doing in my day job.

Asim Hussain: Yeah, so definitely, I'd say there's two things, is that A, a lot of work that goes on is also just behind closed doors as well, and that's one of the things that I find interesting about this space is that sometimes you'll just never hear of it. So, in terms of real examples of measuring SCIs, so there's a project called the SCI Guide, which has a number of case studies inside them, where organizations are really trying to document what they're doing and revealing the numbers.

Revealing numbers is very challenging for a lot of organizations, I can attest to it. You have to go through so many levels of approval to reveal your number. So there's, we've only got a couple of examples of those, but there's definitely tooling that we're building to make this a lot easier. So we're building something called the impact engine framework, which is a framework, which is what CarbonQL is now called the impact engine framework.

So if you've heard me say the word CarbonQL, it's now called the impact engine framework, and it's a tool with a manifest file and you can use it to calculate the emissions. And you can say, I wanna use co2.js I wanna use cloud carbon footprint, I want to use green metrics, and you wanna use whatever.

And it helps you measure an SCI score. And where we're starting to think now is we'd like to get to the point where, there is a GitHub Action, basically, the dream world I have is in six months time, thousands of open source repos all over the world, just drop a configuration file into the root of their repo, add a GitHub Action, and they're measuring an SCI score for their product.

It's been two years now in the making of even the specification. We couldn't have done this two years ago. I feel like so many pieces of the puzzle are now coming into place where people can really, very easily, with an hour's worth of work, measure the emissions of a piece of software, and that's where, so yeah, the CI/CD thing is coming, I would say, in six months time, at least from our side.

And it sounds like you've already got some work anyway from the green coding, green coding landscape,

Chris Adams: yeah,

I actually didn't know about the impact engine. That's, that's new to me as well,

Asim Hussain: yeah.

Chris Adams: The thing that we've been using, so with my day job, one thing we've been doing with a open source project called Wagtail, we've been working with some of the core developers there, and on the Google Summer of Code, a couple of early career technologists who have basically been, who I've been mentoring to introduce some of Essentially like green coding features into Wagtail itself.

Now, the last release of Wagtail came out, uh, in beginning of August, actually the end of July. Now, Wagtail is a content management system, a bit like WordPress, but unlike WordPress, it's written in Python and it's actually written on top of a, a, a software library called. Django, Which is what our own platform uses. Flagtel was used by a number of websites with NASA. If you visit the NHS website, you're using a Wagtail website. There's a number of ones that it's in using. And what we've been doing is we've actually We got chatting to the folks at Green Coding Berlin, which is pretty self explanatory, what they do, they do green coding, and they live in Berlin, we got chatting with them about this, because we were trying to understand, okay, if we're going to make some changes, are we going to be able to understand the environmental impact of, are we making progress? They also have a very literally named tool called The Green Metrics Tool. Can you guess what the Green Metrics Tool does, Asim?

Asim Hussain: I don't know, man, it's hard with these, these terms. Does it, does it generate green metrics in a tool?

Chris Adams: Oh, dude, it's so German. I live in Germany. This is like, to see

Asim Hussain: What's it say in German? Say it in German.

Chris Adams: no, I should, we don't actually have,

it's, it's, you know, the Green Metrics Tool is what it is in

Asim Hussain: Okay, all right,

Chris Adams: So, I think GMT is what we end up referring to it,

Asim Hussain: Oh, that's quite funny. Greenwich Mean Time. Greenwich Mean Time as well, yeah, yeah.

Chris Adams: We've been using that and. The thing that I think is quite interesting about what, uh, the folks at Green Coding Berlin have been doing is they've realized that, okay, there's a bunch of open source tools, op open source software in the world. So they've been basically forking a bunch of open source tools running this.

And then whenever there's a kind of CI run, they've been measuring some of this and, uh, they've actually got a project called Eco CI, which basically is like a GitHub action that fig, that measures the power used when you do a kind of, run as it were, a CI run to, to test something. So they've got some of these figures here and the thing that they've been doing, which we found quite useful as well, is they've been using a tool which allows us to run through common scenarios.

Like I go to a website, I browse through a few places. I search for something, I submit a form, I upload, something like that. We've got a set of journeys that we follow and we're using those as the kind of sample ones to as our kind of baseline to see. Is the work that myself and Aman, the student I've been working with the most, is the work that we've been doing there, has it been helping or has it been not helping? Because the particular piece of work that we've done recently is introduce a support for a new image file format. Called A V I F instead of just using like JPEGs and massively reduces the typically halves the size of any, any of the images that you do use. But there is a bit of a spike in energy usage compared to what you would normally would use both on the server or on the browser.

So we're now actually trying to run this in various scenarios to see is this actually an improvement on this? Because even though it results in a nicer experience, we're trying to make sure that we're going in the right direction. So that's one of the things we have. There's a couple of things we have going on as well.

But that's the kind of most concrete example that I might refer to. And there's a couple of links to both the output from this, but also the open source projects, because you can mess around with some of this stuff. Pretty much right after this podcast, if you really decided.

Asim Hussain: So this is the stuff that is using direct measurement. So you're forking it, running it on like a special rig that is like measuring it. Yeah, I think that's, it's interesting. I feel like this is like something that's been in discussion with the SCI as well, but we never landed on some good terminology for it.

I think we use measurement versus calculation. And we try to say the word measurement like direct, like what's happening in green coding, like direct measurement uh, something from counters or from a power meter or something like that, whereas we use a calculation is when you are just taking some sort of, we, we call it now the impact observation.

You take some observations about the system and you're passing into a model and getting an estimate of emissions. So I think we, I think the language here has gotta get a little bit more specific. I remember on the calls we were even asking, academics, whether there was like specific language around this and it wasn't.

Maybe the, maybe one of the listeners can say, actually ask him what you're describing is the word for calculation is X and the word for measurement is Y. This is, this is where we're getting to, and I think this is where the conversation is in this kind of generally metrics area. One of the reasons I'm exploring modeling is actually for a very interesting use case, which is once you model, you can simulate.

So once you've got a model, you can then tweak the model and say things like, so one of the things we're exploring is like, what if you were to change some aspects of the system, you've got a model, so can you then model that change, and then estimate the emissions reductions. And that's where like modeling has an advantage or modeling has a real disadvantage In the fact that it's a model and you're not really going to get a great actual measure.

So I'm not too sure, we don't have the answers. I just think this is an interesting question. It's like measurement versus calculation and I haven't fully formed my thoughts on this yet as well. But I think it's going to be an active bit of discussion for a while. Maybe it has been an active bit of discussion.

Maybe I'm just really late to the conversation.

Chris Adams: I'm not sure myself, to be honest, but we'll need to

see. The thing I think should be relevant, so when we were using this to figure out whether we're making things worse or better inside Wagtail, I asked Arne about some of this, okay, how are you actually coming up with these numbers? And they basically do things.

Yes, they have a rig, they've got like a bunch of machines that they have where they're reading the data directly from that. But they've also been doing a bunch of work with some of the underlying data that's published by various chip manufacturers. Something called the Spec Co. The

Asim Hussain: Best spec power? Yeah,

Chris Adams: yeah, And the, I've shared a link which basically goes into stultifying amounts of detail about what they do. They've talked, spoken about, okay, this is the tool that's used by green pixie, by cloud, carbon footprint, by TEEDS, like a French advertising company who've been trying to figure this stuff out, and they've. Basically share their modeling of it, which could presumably be consumed by Kepler as well. So they're trying to build these models because they don't have access to the underlying data. And this is something we spoke about in the last episode and the previous episode before that, about why it's a real challenge to get these numbers from especially large hyperscaler providers who. Basically, we'd really like to have much more control over the language. And in many cases, they give honestly quite good reasons for saying, look, share these figures. They are citing reasons like commercial confidentiality or an attack vector. This is why I'm quite excited about the Realtime Carbon project, because it's a chance to finally

Asim Hussain: the values.

Chris Adams: of that.

So you can actually have some meaningful numbers. So you can say, are we making it better? Or are we making it worse? Because even now, in 2023. Getting these figures is a real challenge if you're not running your own hardware.

And I guess, I assume, now that you're working at a company that makes the hardware, or makes much more of the hardware, that's a different change for you now, you see more of it from the other side, right?

Asim Hussain: Yeah, I do get and I speak to a lot of people now. And in fact, actually, one of the things that maybe would be useful to have a deep dive on spec power, if you want to have an episode, I can definitely bring some people is one of the people in my team, she's been spending a lot of time really getting into the weeds.

And it's fascinating working with people who build CPUs their entire life, because it's a different like, You think, Chris, we just write some variables in a Visual Studio code every now and again and claim to understand technology. Once you really get under the seat, there's a lot going on. That we are so abstracted away from and like one of the conversations happens all the time inside Intel is like how do we close that gap between what developers are doing versus what the hardware can do to be more efficient.

And I think there's the, there just sounds like there is just this chasm of opportunity here, which we're just not taking advantage of. A lot of the stuff that's happening on the intel side of the equation is just making people optimize their code. That just, but like using standard kind of optimizations that have been available for ages and a lot, there's a lot of just understanding that I don't even understand how a CPU works sometimes, like the energy curves just do not make any, any sense to me.

I'm not going to go into depth as to my lack of knowledge of what CPU is, but I could definitely bring people in who are much more knowledgeable than me. And then maybe let's have a deep dive into that. I'd be fascinating conversation, like really get into a chip.

Chris Adams: Yeah, because the thing that we've, the thing we're seeing from the outside, or the thing I've noticed from the outside, and I've seen other people also referring to, is the fact that- do you know how we had this thing back a few years ago where engines had like defeat devices where if they're tested, they're gonna work a certain way and they really are. It turns out that you often see some patterns a bit like that whenever you have benchmarks. 'cause if you design for a benchmark, you might not, it might not be designed. You, you could, there are scenarios where a chip will work a certain way that will make it look really good in the benchmark. Uh, and that might not necessarily be how it actually works in the. In the real world basically. You've got that happening a lot, lots of cases. I would really love to deep dive into that because this is the thing we struggle with and it's weird that say most chips are most efficient, like at two thirds capacity between two thirds and three quarters, right? Rather than, so you might think like you got, if I turn it all the way down, that will turn all the power down. No it doesn't work that.

Asim Hussain: It doesn't. Yeah.

Chris Adams: And there's all these other incentives about where you move computing jobs as a result, which has this kind of knock on effect.

Alright, we've.

Asim Hussain: There's actually really interesting work around like when we talk about moving compute around different parts of the world, there's actually a really great project being open source project run through Marlow Weston, who's one of my colleagues at Intel, and she's also one of the chairs of the CNCF environmental tag and I'm going to get the name of our open source project wrong. I think it's Kubernetes Power Mode. And what it does is it does like load shifting across cores on the same CPU. So normally when you, like, you want to max out one core before allocating work to the other cores. That's the most efficient way to go up the curve.

But most like allocators will just allocate them across all the cores on average. And so she's built this kind of, uh, Kubernetes, uh, scheduler, which basically will max that one core at a time. So you get to the top.

Chris Adams: Wow, I didn't know that was possible. That's a bit like how cars, so certain cars would be, if you've got a car with maybe a V8 inside it, there are some cars which will basically just run on four of the eight engines, eight cylinders firing all eight for fuel efficiency. That sounds like the kind of cloudy equivalent to that idea.

Asim Hussain: But there's also, but she's, she's actually got a second Kubernetes project I'll get the link to, which allows you, to change the clock frequency of your chip at the application level, so with the intention of; if you can change people overclocking, you can actually underclock, and underclock actually does this amazing thing where you get much more efficient from an energy perspective because everybody's looking at like reporting what is the like peak level efficiency but if you can just say look i'm willing to run at 20 less clock speed you actually gain more than 20 energy efficiency improvements but you lose that on the performance.

So if you can dynamically change the clock frequency, which happens a lot on like laptops and mobile devices, it does not happen on the cloud space. It has lots of negative consequences as well. Lots, yeah. You really can't just do it without knowing like how an entire stack works top to bottom. It's a very advanced piece of thing, but if you can take advantage of that as additional efficiencies again, reducing that chasm between what we developers think we know about tech and the hardware versus what hardware actually does is I think one of the frontiers of this space.

Chris Adams: This was actually something Arne explained to me, he was looking at why some of the figures that say, we spoke about a project called Scaphandre last week, he says that one of the reasons that, one of the things that's difficult about this is that, yeah, like you said, the clock speed can go up and down, and he, the kind of mental model that I ended, left the conversation with was a bit like, revolutions per minute in an engine, so you can have it red lining to go load really, really fast.

But if you scale it right back down, then you can be somewhat more efficient, but there's going to be impacts. I didn't realise that you had that kind of control with a software level itself. Actually, you could deliberately- I thought you could only just ask the CPU for work to be done rather than say, can you do a bit, cus that's that's not like nicing something. That's a different level of

Asim Hussain: That's a whole different level. Nicing is probably... No, it's not like nicing something. It's a very different level of hardware control. Yeah.

Chris Adams: All right. Wow, we went really deep. Not expected enough. Okay. Okay. Bye. Okay, so hopefully that should help the question that asked,

are there

Asim Hussain: even the question? What was even the question?

Chris Adams: there examples of measuring the SCI in pipelines?

Asim Hussain: We went off!

Chris Adams: Yes, there are examples of it. There's lots in the open. The work from Green Coding Berlin is probably some of the stuff that's really in the open. But there's also work done behind various corporate firewalls that you might not be able to see, or you might probably can't see unless you employ all kinds of industrial espionage, which I suspect you're probably not going to do that if you are good at that. Anyway, okay, let's move on to the next question, it seems because we're burning through our time.

Next question was about the carbon efficiency of GPUs. This seemed to be a question of basically saying what's the carbon efficiency or otherwise of GPUs when they're used for like owner respect search and stuff like this, and is this good for the environment? This is the question that I got, and I assume this was a response to people talking about the fact that with this new world of generative AI and LLMs, you use lots and lots of specialized chips, often, which look like GPUs or sound like GPUs. Do you want to have a quick go at this assume, and then I could probably

bounce on some of this, because I just, yeah.

Asim Hussain: Let me say two things. A, If you're using the generalized CPU, which is specifically for generalized and for anything else, so it will be more efficient on an energy basis. I would say the point though is when you start using GPUs and you start using specialized hardware, each of them has an idle power amount.

And so if you've got a GPU and you've got a whole series of them, or all this is the specialized hardware and you're not using them, that's actually bad. And so it's very important when you have this specialized hardware, like you're thinking through and you're thinking, I've got it, I'm using it. That's why I've got it.

Obviously, if you're in the cloud, it's a different equation, right? Maybe not, actually, if you can just order a GPU and not really use it. And the other thing I would say is, is, and I've seen this conversation go a little bit wonky as well is when oftentimes the total power of a system increases. 'cause a GPU consumes more power, and then people just say, oh, it's just, it's less efficient, it's consuming more power without factoring in that like a job will run faster and therefore the total energy will be less.

If that makes sense. I've seen conversations get into confusing territory and people have confused energy and power. 'cause power is like just the Watts per second, whereas the total energy, so if you're using so that, that's another way

Chris Adams: You're

Asim Hussain: about carbon efficient. Yeah. Was,

Chris Adams: being that you might have a GPU, a graphics processing unit, which is extremely energy intensive, but it runs a job for a short period of time and therefore it could be turned off or could be scaled back down. Right? That's the thinking. That's what you're saying, right?

Asim Hussain: I dunno if they can be turned off, but I think they're always on, aren't they? I don't know. Actually. I have no idea. But yeah. Are the ones that turn off?

Chris Adams: You can see there is there, there's a definite, uh, impact between something running a hundred percent and running and when it's idling, there is a change.

But I'll be honest, I'm outta my depth when it comes to figuring out how many compute, how many people who run data centers switch them off on a regular basis.

I suspect the number is very low.

So,

Asim Hussain: close to zero.

Chris Adams: yeah,

I was actually going to answer this differently.

Asim Hussain: Oh, go on then. Yeah.

Chris Adams: say that if you're asking, if you want to talk about the carbon efficiency of GPUs compared to like CPUs or something like that, it's worth understanding that the emissions will come from two places when you're thinking about this.

There's emissions created from making the actual computer, and there's emissions from running the computer. And when you make something which is specialized for the GPU, for example, that's going to be pretty energy intensive. And in many cases, you have a bit of a trade off, right, where if you, if you basically had a bunch of CPUs compared to GPUs, if the GPUs are more energy intensive to make, then if you don't use the machines very much, then you don't have much usage to amortize the kind of cost.

So that, so in that case, GPUs are going to be pretty inefficient, they're going to be pretty carbon inefficient. But for the most part, because these things are so incredibly expensive, they tend to get used a lot or there is an incentive to use them as much as possible. And even if you're not doing them, to make them available for free, uh, for people to use these or at least try, try and grow a market.

And that's what you see right now with, um, things like, uh, various tools like chat GPT and stuff like that, which lots of us are not paying for. The use of that results to a massive amount because you want to re receive a to achieve a certain amount of utilization, so you can actually get any kind of return on this.

The thing that I would actually draw your attention to or thing that might be worth looking at is recently we had the conference Hot Carbon, and there was a really cool paper which was specifically called, which addressed this, the title of the paper was called Reducing the Carbon Impact of Generative AI Inference. There's a number of people who are named on this. So Andrew A. Chien from University of Chicago and Argonne National Laboratory. Hai Nguyen, Varsha Rao, Tristan Sharma, Rajini Wijayawardana from the University of Chicago, and Liuzixuan Lin, I think, right? This was a really interesting talk. I think because it was basically looking at the environmental impact of tools like, say, AI, and saying, okay, we've got this whole kind of trend of employing LLMs, and large language models, and generative AI in searches and things like that.

What does the impact look like? And they basically looked at, say, the usage figures that were published for ChatGPT in March 2023 and that was like 1. 6 billion, like users. And then based on that, they, they they modelled the likely inference cost, which is the cost from using it, and the training cost.

And the thing, there was a few kind of takeaways. First of all, we often talk about the training cost as the big thing to be aware of. And they said no, like the training was 10 times the impact. And they said if you were to scale this up to say, Google's usage, then even if you had a training cost of about, that's going to have a ginormous impact basically. So we should be really thinking about the inference part, and in this case here, having something like a dedicated fast machine that does the inference, compared to a bunch of CPUs, for example, is really cool for a bunch of other reasons.

Asim Hussain: Yeah, and I just want to say, I think two things with the increased adoption, interest, usefulness of AI. Influence is going to go through the roof, as you said, it's on and the only place it's going to go is higher. The interest is going to go as higher as the years go on. As I've said before, nobody invests billions of dollars into AI if there's not a growth sector.

People aren't going to use it and more people are going to use it. That's inference. That's why inference is very interesting. That's going really high. I just want to say, I just completely forgot about the Hot Carbon Conference this year. I watched every single talk in the Hot Carbon Conference last year.

And let's put it in the show notes because I think last year's program was amazing. I watched every single video. I made copious notes on all of the, all of the talks, and I'm, I'm looking forward to going through it again this year and doing what you did. Sales and just listening to all of 'em.

Chris Adams: Yeah dude we had some of the people, we've had the speakers from the previous talks because there've been so many really good ones. The thing that I really liked, I just wanna come back to this one because I think there's some really nice things that came from this. This talk in particular in this paper. One of the key, key things was, is basically saying, let's assume you're gonna have this massive increase in usage. And I think the comparison was, they said if you were to scale the usage of chat GPT up to the kind of modeled usage, In, in this paper for say

Asim Hussain: Oh,

Chris Adams: mainstream search engine, a 55 times increase in use. If you were to scale it up that way, you might think, oh, crapes, that's 55 times usage. Assuming this is like in 2030, and then ev this, they basically tried to project this forward into 2030 and say, well, okay, what would the look, would it be that in 2030 we would've 55 times a carbon footprint if you did this? They basically projected, they took some trends and extrapolated them forwards. One of them was that you're probably going to see an increase in energy inefficiency over time because we have seen in moore's

Asim Hussain: sorry, you said energy inefficiency, did

Chris Adams: So energy efficiency. So they basically said, let's assume between now and 2030, you see a 10 times improvement inference, and that's based on what we've seen so far in terms of things keeping, keeping getting more efficient. Let's look at the carbon intensity of the grid will also be decarbonizing over time and they took some from current trends and what's actually especially been coming in with changes in policy and they basically said with these numbers is it possible to do something about these figures and what would the figures be if you were looking at this in 2030 in the next six and a half years and they basically modeled some of this and they modeled- they, they did this as a way to figure out the actual savings possible by using things like carbon aware programming, and one of the key things they said was that because inference isn't super latency sensitive, because of the actual on the machine in the actual chips in some distance, say machine doing a bunch of inference, then piping the results to you. It's not so latency sensitive and that means that you can quite easily run this in lots and lots of greener regions, even if you're accessing it from a place where the energy is not so green, let's say. Using this versus what we have right now. They, they we're probably not gonna have a massive increase with, I think the figures that I saw

Asim Hussain: Oh, so they,

Chris Adams: versus, yeah.

they basically said, based on this, if we were to employ, let's say we, let's assume you're gonna have machines becoming more efficient anyway, and you scale up this much usage, if you were able to carefully run the inference and serve the requests

Asim Hussain: Oh.

Chris Adams: the greenest regions.

Asim Hussain: But that's the assumption. The assumption is that you have to actually be green, do green software to decarbonize a software. If you actively, so it sounds like if we did everything we're asking you to do, we'll be flat. Do they have a number for what if people didn't do?

Chris Adams: Yeah they basically said, assuming if you didn't have any energy efficiency improvements, they said 55 times load will be 55 times a footprint. They said if assuming you have the efficiency improvements increasing at the same rate as they have been, you're looking at maybe With an uplift of 55 times the usage, you'd probably be looking at 2.6, two and a half times the

energy usage, I mean, of the emissions from the grid, right? But they said, if you were to actually use the learn,

Asim Hussain: Carbon

Chris Adams: programming like this, they brought it down to like, the ideal scenario would be you're looking at 1.2, which

Asim Hussain: But that,

Chris Adams: kind of mind blowing...

Asim Hussain: well, it's mind blowing, but I think it shows how important the work that we're talking about is. It's like, actually, it's one of the really great talks from last year's Hot Carbon, which I loved, which was, I've forgotten, I've got to apologize. I'm not going to remember which one it was.

But it was talking about how projecting forward kind of compute growth and how green software was a way of being able to handle the additional usage and load of the cloud without actually having to build more servers, because fundamentally we are constrained at the rate with which we can actually increase the cloud, but the growth is growing significantly as well, so like being more efficient actually allows you to deal with growth. You have to be green, you have to use green software if you want a realistic chance of generative AI being as ubiquitous as you want it to be.

Chris Adams: I mean, the other thing is, you don't have to assume that they have to be there, like, yeah, you don't, maybe, like, the option is, don't, you just don't need to buy all this equipment in the first place. These will never be a replacement for actually having better data.

Asim Hussain: What if they're just humans in a building that's answering your question? Is that more efficient? There was a Gartner thing I saw recently which is that the total amount of energy used by AI by 2025, so Gartner report, will be higher than the total amount of energy used by the entire human workforce in the world.

Chris Adams: I, I, I, I would, I don't know enough about that. And I feel a little bit worried about referring to that. But the point I was going to get to was the fact that you're seeing examples where Actually, just having good domain knowledge, it turns out to be much, much more effective than having loads and loads of compute.

And the good example that I've linked to here is actually, there's a company called Lesan, they're based in Berlin, and they do machine learning specifically for Ethiopian languages. And they outperform Google Translate, they outperform some of the large providers, because they've just got access to the actual benchmark data sets from the first place. This is the thing, having quite high quality data is another way to reduce the amount of compute used. And this comes up

Asim Hussain: true. Yeah, very good point. yeah,

Chris Adams: and this is also when you bear in mind that even just the whole tokenization that you have when you're, it's based around English language and so even another language is gonna have, we're gonna gonna need more tokens for the same amount of sentences. So there's a whole bunch of issues there that we might refer to.

Alright, so we, we dived quite far into an efficiency of GPUs and we might think about that. I think we've got time for maybe one more question left before we have to wrap up Mr. Hussain.

Asim Hussain: Okay. You pick it.

Chris Adams: Okay, so this one is, this is a question about water usage. Can the cooling water for data centers be reused?

And this is a question because people...

Yeah, actually, I think one of the worries is that people actually... In many cases it just gets pumped back into rivers when the water is that much hotter, you're basically just cooking the fish, which is not...

Asim Hussain: Sorry.

Chris Adams: not very helpful.

Asim Hussain: it depends if you like eating, I suppose it depends if you like eating fish,

Chris Adams: don't think it's good. I don't think the fish enjoy this, right, but basically there is- that's one of the issues, but I think this is more actually a case of this is speaking to the fact that in many cases, 1 of the big things that's come up is basically people talking about the water usage with compute, and in particular data centers where, which are very heavy on, uh, generative AI and things like that. And there's a really good example that we might refer to that I learned about, which is Google and some of their data centers in Chile over the last few years. There was a whole thing where you. So in Europe, for example, where there's lots and lots of water, you don't necessarily, or there's parts of North and Western Europe where if they're cold, and they already have lots of water around them and lots of rainfall, then it's not so much of an issue.

But if you were to put a data center where there's loads of drought that uses lots and lots of water, the examples, there's a company called Algorithm, organization called Algorithm, which we spoke about some of this, because you see protests against data centers. One of the key things was You find some data centers using something in the region of 169 liters per second. Now, if you run that in a place which has drought, maybe not the most equitable use of a scarce resource, especially for the people who rely on that water to live and survive. There are other examples where large companies have come in where they've ended up using significant amounts of water. The thing that was interesting about Chile was that Google wanted to deliver a deployed data center here. They had a bunch of pushback, but then they ended up choosing to use much, much less water intensive technology as a result, like I think it's adiabatic cooling, which is essentially a kind of closed loop system, which doesn't rely on evaporating water than getting rid of the water as a way to cool things down. This is one thing that came up and I've I have added a couple of links to both Algorithm Watch talking about this, as well as the actual organisation, the activists in Chile, talking about ok, we had a victory for this. The fact that, yeah, they are issues around it, but it's also a case of companies, they can make these choices, but a lot of the time, they might not choose to, because it's a little bit more expensive and here you feel like if companies could be making a huge amount of money, and Google spent 60 billion on share by buying its own shares last year, they're going to have fairly efficient, less water intensive cooling in a place where there's that's suffering from drought. This seems a fair thing, like these things we should be asking for and should be setting as a norm. There are other organizations doing this too.

Asim Hussain: What do you think, one of the things, I've got nothing to back up, one of the things that was hinted to me the other day, I think it was Sarah Bergman who might have mentioned on Twitter, that there might be situations where it's mutually the opposite. Being more carbon efficient might actually make you more water intensive.

Like for instance, doing things that reduce carbon emissions might require more water consumption, and which is why I think it's exciting that we're actually all starting to have this conversation right now, because I think we're so focused on carbon, and we're optimizing for carbon, but actually, the landscape is much more complicated.

It's much more of a surface where you're trying to minimize the environmental impacts of your choices. And you might have to make trade offs versus one versus the other. If there's a water scarcity right now, you might have to increase your carbon emissions. I'm excited that this is where the conversation is evolving to.

Thank you. Because once we add water to the mix, we can add other things.

Chris Adams: You see a trade off for sure, but in also, lots of these, ultimately, it comes down to capital expenditure.

Lots of the

Asim Hussain: it can be an And

Chris Adams: very, like, yeah,

like,

Asim Hussain: an and. Yeah, yeah.

Chris Adams: you are seeing this, but it's also worth bearing in mind that when you're looking at this, impact comes from the energy generation in the first place, because let's say you're going to burn a bunch of coal to heat up a bunch of water to turn to, to generate some electric is a huge amount of water being used there.

In fact, freshwater usage in energy generation, I believe it is actually the number one source of water usage in America. So we, when we talk about this, it's also worth thinking about the entire supply chain. Yes, there are absolute things you can do at the data center level. Also, if you look through the supply chain, there's also other areas, but typically with data centers, it tends to be very localized. So there may be water being used, but if it's water being used in a place where that people are depending on for drinking water in the same town.

You can understand why people are a bit miffed, basically.

Asim Hussain: it's like, we don't really think of data centers like coal power plants, but like, it's almost just the same. Like we treat, we treat, we treat, we treat them as very different. But at the end of the day, like water is a, is in this, in this case, could be a pollutant.

Chris Adams: Yeah.

Asim Hussain: If you're pumping hot water out, I don't know, I do not know enough.

Please don't quote me. I don't know exactly what happens here. I do not think that data centers are like, maybe they are like squirting like hot streams of water into rivers or something like that. But I'm just pointing out that you often feel like some things are like abstracted away from a mission so much you don't really associate it with the entity.

But like with a coal power plant, we just so associate it with emissions that we know what to think about it, how to think about it. But like a data center in a way is it generates emissions. I'm sorry if it is. Putting like hot water into rivers and streams. Isn't that a pollutant?

Chris Adams: Well, yeah. THere's all kinds of pollutants that you have. There's noise pollution as well.

There's very, that you might need to take into account when someone's citing big pieces of infrastructure because this is industrial infrastructure.

That's the

Asim Hussain: is. Yeah.

Chris Adams: Like there are cases of the. people having a really hard time with just the wiring and the noise pollution from data centers crypto mining rigs

Asim Hussain: really, you can you hear, if you live now, you'd be able to hear whirring

Chris Adams: I'll share a link to an example from um there there's there's an interesting case with amazon uh specifically where there's a there's a bunch of people who are basically complaining about the noise pollution um in i believe it's I think it might be West Virginia,

who are,

Asim Hussain: Yeah. There's semi

Chris Adams: where they basically hear this because it's loud enough, but you also see this with cryptocurrency mining in New York State, there's been lots of cases where you have typically the really quiet, serene places, where the calm has basically been punctured by the incessant whirring of,

Asim Hussain: like

Chris Adams: of all these things, yeah, exactly, so there's various dimensions that you would need to take into account that go beyond just thinking about carbon and carbon tunnel vision, but let's be honest dude, like, Most of the time, organizations struggle with just thinking about carbon as well as cash, right? So it's, it

Asim Hussain: Let's add water and noise to it though, Chris. Let's give, let's give him everything. Yeah.

Chris Adams: and the, what I'll do, I'll add another link, because there's some really fast, fantastic work by Sasha Luccioni, who's the climate lead at Hugging face. She wrote a really good piece in Ars Technica, talking about all the various things you need to take into account with the environmental and social impacts technology and specifically, um, AI. It's a really nice way in. And, oh, I should actually share, um, my organization brought, published a new thing, uh, this week, A new issue of Branch has come out and it's got a bunch of stuff talking about this from a, from a Tamara Kneese. She wrote some, she wrote about some of this, but also Dr. Theodora Dryer, she, she wrote a piece about's, also an expert in. We'll show a link to that 'cause that that would be fun for some, for some people as well.

Oh, blindly. We've gone way over actually Asim.

Asim Hussain: That's good. That's good. Great episode.

Chris Adams: We answered those questions, or at least we've peppered this, uh, these show notes with huge amounts of links to people who might wanna learn more about this and hopefully we've get add added some tantalizing hints. Asim, I think we're actually at our time, we've got through four questions this time around. I think there are some more, but in the meantime, I think I'm gonna have to say, Thank you for coming on and wandering through this with me. Yeah, this was fun, man.

Asim Hussain: Yeah. It's good to see you guys. I love these, I love these mailbag episodes. Let's do more of them.

Chris Adams: Yes, I want to ask you a bit more about the Impact Engine next time as well, because I didn't know about that.

Asim Hussain: Give us, give us a month and I'll, and I'll, and I'll be able to get into a lot more detail about it with you. Yeah,

Chris Adams: Okay, cool. Also, if anyone who's listened to this is curious and has questions of their own, please feel free to at us in various places or even come to the new discussions. The new Green Software Discussions website. I might ask you to point to this because otherwise I'm going to podcast.greensoftware.foundation

Asim Hussain: We'll put it in.

Chris Adams: address that we normally use. Is it visible? Is there

Asim Hussain: do you know we should create a short link? We, we should create a short link, but there isn't, if you actually go to our GitHub organization, there's just a tab called discussions. But you're right. We'll, we'll put it on our website and we'll make sure it's more prominent in the future here.

Chris Adams: Okay, in the meantime, go to https://podcast.greensoftware.foundation. Most recent discussions where you can ask some questions and then we may if we can fit them in the list, we'll add all of them so we can add other things coming through.

All right, that was us. Lovely seeing you again. Hope the mushrooms are well, and yeah, see you on the flip side, okay?

Asim Hussain: See you then, buddy. Bye.

Chris Adams: Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

The Week in Green Software: Green Kernels

10 augusti 2023 | 45 min

In this episode of TWiGS we delve into the intricate world of measuring software energy consumption, a topic vital for reducing our carbon footprint. Despite the strides in greening software, knowing how much energy software consumes remains a challenging puzzle, especially in the cloud computing era. Joining host Chris Adams are guests, Aditya Manglik and Hongyu Hè, graduate students from ETH Zurich in Switzerland. With their expertise in improving energy efficiency in systems, particularly operating systems, microarchitecture, and machine learning, we embark on a captivating journey to understand why quantifying software energy usage is intricate and what innovative solutions are emerging. Stay tuned as we amplify the geek factor to 11 and uncover the complexities of this critical field.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Hongyu Hè: LinkedIn / Website
Aditya Manglik: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Topics:

Tracking Energy Consumption and Why it is Important [4:09]
Tools for monitoring Linux energy consumption [19:49]
How to monitor energy consumption in a virtual environment [24:06]
Adrian Cockcroft’s Proposal for a Specification for Real Time Carbon Intensity | Green Software Foundation [30:12]
Monitoring the energy consumption of data centers [33:34]

Resources:

EnergAt: Fine-Grained Energy Attribution for Multi-Tenancy | Hongyu Hè HotCarbon [3:18]
Measuring Carbon Footprint of Personal Computing | Aditya Manglik LFE Summit [3:29]
Measuring Your Application Power and Carbon Impact (Part 1) - Sustainable Software | Scott Chamberlin [5:09]
PowerTOP | Linux [20:24]
Scaphandre | GitHub [20:57]
Cobbler | GitHub [24:48]
EnergAt | GitHub [41:40]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Aditya Manglik: At the end of the day, what we want to tell people is, okay, computing is great, but we have to be sustainable. And right now, data centers consume 3% of all global electricity. This number is only going to grow, right? Especially after COVID, we have had a massive increase in Digitalization, and now with the large language models coming in, like ChatGPT, it's going to grow exponentially.

So we have to be sustainable, and the first step to being sustainable about energy use is to understand where is the energy going.

Aditya Manglik: Hi Chris, please feel free to call me Adi. It's a pleasure to be here on this podcast. And, yeah, I'm a longtime listener of this podcast, so it's very exciting to be here. I'm a graduate student at ETH, where I work on improving the energy efficiency of systems, especially operating systems and microarchitecture.

And I previously worked on building a very nice, very complex energy attribution system in Linux as a Google Summer of Code student with the GNOME Foundation.

Chris Adams: Cool. Thank you, Adi. And in addition, we have Hongyu also. Hongyu, I'll give the floor for you to introduce yourself as well.

Hongyu Hè : Yeah, thanks. Thanks very much for having me on, Chris. And thanks for inviting me, Adi. So yeah, I'm also a graduate student in computer science at ETH. My research includes both software and hardware. And I'm currently interning at Apple, working on machine learning research.

Chris Adams: Cool. Thank you Hongyu. So for those who are interested, we featured both of, uh, the work from both of these two researchers. In the last episode, we spoke a little bit about Hongyu's, uh, uh, paper, uh, what one of the papers Hongyu was a contributor on at the Hot Carbon Conference. And we'll share a link to the talk presented there.

And we've also shared a link to Adi's talk at the Linux Foundation Energy Summit in Paris earlier on in June. If you're new to this episode, to this podcast, my name is Chris, I'm the Executive Director of the Green Web Foundation, and I'm the Chair of the Policy Working Group at the Green Software Foundation, and as a final reminder, we're going to cover a fair few papers and links and resources, and what we'll do is we'll add all of these to the show notes so that you can do your own research later on as you run through this. All right then, I think we're all sitting comfortably, so shall we begin, fellas?

Aditya Manglik: Yeah, I look forward to it,

Chris Adams: All right. Okay. Adi, I think I might start with you first. We've spoken a little bit about tracking energy consumption and why it's an important thing. Maybe you could just give a bit of a kind of overview about why this is important, what the state of play is in the different systems, because we know that computers run on, say, Linux. Lots and lots of machines run on Linux, but we also know computers use Windows and macOS. Maybe you could provide a bit of background, then we could talk about what the options are for people using these systems.

Aditya Manglik: absolutely. I've been working on this problem since almost five to six years now and it's an absolute pleasure to be talking about it. Well, I often like to say that you cannot improve what you cannot measure and that is where the problem starts. We don't know how to measure the energy consumption of our systems.

For example, if I ask you, how much energy does WhatsApp use? Or when you send a WhatsApp text to your friend, how many CO2 emissions did that message take? Learn Can you give me an answer? No, that's what makes me so excited to get out of the bed every morning and then try to figure out, okay, how much energy is WhatsApp using?

So it turns out that people at Microsoft, Apple, Google also care about this and they really tried to solve this problem and Microsoft has this very interesting kernel system called the Windows Energy Estimation Engine. It is running on all Windows devices. Android has a very interesting service called PowerMetrics.

You can think of it like a daemon. A daemon is a magical service that runs in the background of your system that does all the stuff for you and you don't know that it exists. PowerMetrics on macOS also collects all possible data about the energy consumption of your applications. Now, what about Linux, right, we are, we love open source, and Linux is a very important operating system, right, all servers in this world are, majority of them are running Linux, but we don't know how to measure the energy consumption of these servers, especially from the software, right, people often think to measure energy you need these hardware devices, or you need these electrical engineers to come in and plug monitors and then tell you, oh, this consumed 5,000 joules.

No, we want to solve it using the tools that we have, and I think we can solve it. I believe we can solve it. And that's what I'm working on.

Chris Adams: Okay, so you just mentioned two things, first of all. So one, first of all, you said that if you're using a Windows machine, there's existing tools that you can tap into and get readings from. And if you're using an Apple machine, you've got access to those kind of figures. But it's a somewhat murkier situation with Linux right now, there isn't a kind of common tool that is actually universally used.

That's One of the key takeaways I'm getting. And Hongyu, I believe this is what you've been finding as well, and you've been looking into some of this as well,

Hongyu Hè : Yeah, exactly. I think there are tools, but there is no common thing that everyone uses. And the standards of those tools are varying quite a lot. That's also, as you said, one of the reasons why we contacted the research in the first place.

Chris Adams: Okay, alright, so if I understand this, given that the majority of servers are now running Linux, basically not having some tools for the most common kind of operating system is one of the things which makes it difficult to come up with some of these numbers. That's what I think I'm understanding from here. We're naming this episode Green Kernels, and I figure it might be worth actually just talking about this idea, because this sounds like a relatively low level thing that's built into systems themselves, actually. Adi, could you maybe talk a little bit about this part here? Because I think that you've spent a bunch of time looking at this low level part of an operating system like Linux, like this kernel part. And before we dive into that, maybe you could actually explain what a kernel is and why that might be somewhere that you actually track some of this. 'cause not everyone may know what a kernel is when you're thinking about computers in the first place.

Aditya Manglik: Chris, first of all, I love the name Green Kernels, like when you talked about, to me, talked about this podcast and when you named it Green Kernels, I was like, yes, I came to the right place. Okay. And yeah, what are the kernels? I think our audience is really smart and even smart people sometimes just need to quickly jog their memories.

So what we're going to do is quickly jog their memories. A kernel is the core of an operating system. What does that mean? Okay, so for us a computer is just a computer on which we log in and do something, but what we do is an application, we use Microsoft Word, PowerPoint, Excel, these are all applications, and Windows that is running these things is the operating system, and operating system comprises broadly two parts, a kernel, which is the core, that you don't see, which handles everything for you and the user space, which is what you interact with.

So you know that start button, that you click that is part of the user space and that start button goes behind the curtains and does some interesting stuff that comes back to you and yeah, you see the effect of your action. So the kernel is the primary response, primary entity in any operating system that is responsible for managing the hardware, the applications, the processes.

Chris Adams: Alright, so this kernel part is the thing that. So far, for Windows machines and for Apple machines, there's something in there, but for Linux machines, you don't have that same ability to read information yet. And this is some of the work that you've been doing to look into to basically make some of that readable.

Is that the case?

Aditya Manglik: That's a great question, Chris. So, energy is typically thought of as an electrical engineering topic. And it's difficult and it's fancy. No, people typically don't include energy monitors and systems. That is the fundamental reason why we are trying to do this. You can measure the performance of your programs.

You can measure how much time it took and this is possible because your system tells you how much time your program took. Your CPU tells you how many clock cycles your program executed for. But if you ask it, okay, how much power it consumed or how much energy it consumed, I think things fall apart. And that's why you need to do a lot of modeling and build entire systems to figure out this information.

Now, Linux also has this information, but the models, right, so you can have all the data in the world, but until you know how to make sense of that data, it is useless, right, and that is what the model does for you, and these models don't exist in Linux, they exist in Windows, they exist in Mac OS, Android, iOS, But I'm not so sure if they are existent or if they're good enough for Linux.

That's what I think, but if you know about it, let me know. I would be very happy to know.

Chris Adams: All right, and when you're talking about a model here, maybe you could just elaborate on that, because I'm not sure I quite follow when you talk about something being a model like this. If I'm, let's say, you mentioned the example of WhatsApp, for example. How would I go about figuring out how much energy is actually attributable to, say, WhatsApp, for example, on a computer or something like that?

Maybe if we were to look at that example there, then we can say, okay, we could talk about some of that, then we could see how that becomes more difficult if you're thinking about things like cloud computing, because, as I understand it, The assumptions you might make when thinking about a desktop computer might not be the same as working with a cluster of computers, for example, and I believe this is some of the work that Hongyu you've done most recently and spoke about at HotCarbon.

Aditya Manglik: That's another great question, Chris. Okay, let me quickly dive into it. Imagine this as a car, okay? You are driving a car. Now, you decide where you're going to go. but its the engine that burns the fuel right? you're not burning the fuel, you're simply deciding oh I want to drive to London but your engine is what's going to consume the fuel now when you want to send a message to whatsapp what you do is you write out a message and you hit the send button and behind the scenes what the kernel does or what is actually happening is the kernel converts your message to a bunch of packets And it sends these packets over the network and along this way of converting this message into packets and sending it, you are using your device's CPU, memory, storage, network, screen, maybe the Wi Fi interface, right?

So, you see all of these hardware devices are immediately turned on as soon as you hit that send button, and that's where the energy consumption comes from. And what would a model look like to build such a model? What you would do is you would take in the amount of power of the CPU and multiply it by the time that the CPU was running.

Similarly, you take the amount of power for the, for the network interface, that would be the wifi card, and you multiply it by the time, by the amount of time that it was running, and then you accumulate all of these data points. And that gives you the energy of sending a WhatsApp message on your device.

Okay, we're not even talking about the energy that the servers consume, the energy that your friend's device consumes.

Chris Adams: Is that giving some pointers? And so the idea would be that if you can't get the figures from each of these pieces of equipment themselves, like this, like a CPU, like a screen or something like that, you might use a model to

come up with some numbers for that. Okay.

Aditya Manglik: Yes.

Chris Adams: All right. And that is based on the assumption when you're looking at a single machine, using a single program.

Now, on you, I think when you are, I think it's somewhat different with, there's assumptions might not always hold true. So maybe Hongyu, you could explain how this gets a bit more complicated in the cloud, or some of the parts there, perhaps.

Hongyu Hè : Thanks for the question, Chris. Yeah, I think Adi brought up a really good point about hardware and the model. One thing I'd like to add on at this point is the key reason why we need a model. Adi has introduced the concept of kernel. So kernel is basically a cushion, if you like, between users and underlying hardware. And the hardware is ugly, because they have different interfaces, it's really hard to interact with directly, so there are multiple challenges, and one crucial challenge that we have been facing is the lack of support from hardware, so if the energy attribution is there, so if I'm using WhatsApp and the hardware is telling me, okay, WhatsApp is using this amount of energy.

Then why don't those kernel, the cushion reports this to me, right? That's the key point. So here we don't have the crucial hardware support that we need. That's why we need the model to collect, uh, if you like, the proxy data, like utilization, the time you're using to calculate, uh, the amount of energy from the user side instead of relying on solely from the hardware. And speaking of like classic computing, I think also Adi mentioned a great point about multiple hardware, different devices that really makes the life more difficult. Because we need to take into account different kinds of devices, especially in the cloud, we have heterogeneous devices, CPU, DRAM, GPU, etc. Yeah, it makes things more challenging and much harder to calculate the energy consumption because of the distributed nature as well.

Chris Adams: So if I understand what you've been saying here, so there's one issue, which is a case of attributing the energy to a particular program, for example. And then one of the other issues is basically the fact that across all the different kinds of computers, not every single And Device that draws power will have a consistent way of reporting how much power it's drawing. So if I understand it, there are some tools that we do this. So lots of intel processes have a thing called running average power limit, for example. But it may be the case that if you're using maybe. If we were to step away from our WhatsApp example and say, I'm doing a really big machine learning job using a bunch of very powerful graphics cards, they might not expose the same information about how much power they're using.

So you would need to either model that or you would need to have some other way of getting that information back. Is that the case?

Hongyu Hè : Yeah, that's a very accurate summary. Thanks, Chris.

Chris Adams: Okay. All right. So that gives us some pointers here. And I'll just ask you one thing about this as well, because this is something you touched on in your paper. The example that Adi gave was being in a single computer, where you can be relatively confident that the hard drive is attached to the same computer and the screen is attached to the same computer, like a laptop, everything's in one place. This assumption might not be true when you're looking at cloud computing. Again, I understand it's a little bit more complicated. Is that the case?

Hongyu Hè : Yes, indeed. Yeah, that's, that's a great question. So in a cloud, for example, computing resources like CPU memory that we've been talking about are increasingly shared among many tenants or users, or, you know, for example, the organization like a university or company are using a class of servers. And this really makes The attributing of energy consumption really hard. And also this is quite a sensitive topic as well, because we don't want to point fingers arbitrarily without a very precise model there.

Chris Adams: All right, Adi, I assume this is similar to some of the work that you've been finding as well, because I understand your research has been focusing on the desktop part more than the cloud computing part, right? That's where some of your research has been, or have you been looking at the wider, somewhat more wider than that, for example?

Aditya Manglik: I've been focusing on the desktop for now, but I agree with Hongyu that it's tricky to correctly point fingers at people for consuming the energy that they're using. Yeah, so at the end of the day, what we want to tell people is, okay, computing is great, but we have to be sustainable. And right now, data centers consume 3% of all global electricity.

This number is only going to grow, right? Especially after COVID, we have had a massive increase in digitalization, and now with the large language models coming in, like ChatGPT, it's going to grow exponentially. So we have to be sustainable, and the first step to being sustainable about energy use is to understand where is the energy going.

And, yeah, this problem becomes more tricky because with the growth of cloud, you don't know who exactly is consuming how much. I'm very curious, and I keep talking to Hongyu about his work. He's doing fantastic work in this direction. Yeah, let's just say that we are both very curious about this.

Chris Adams: Okay, so if I understand it, it's a bit easier to get some of the numbers from a computer you have yourself. Right, when it's on your own computer, but, because increasingly we're moving computing workloads away from just the desktop into a kind of wider set, it ends up going from other places. So maybe you could actually talk a little bit, let's say that you do want to actually start measuring this, or you do want to start understanding what role you could play or what, or how you're able to at least measure this so you can start optimising it. Let's say you're working with servers right now and you're using a bunch of Linux computers. What are your options at now? So Ade, I'll start with you actually. I'm on a Linux machine, it's just one machine, and I want to understand the environmental impact of a particular service, or a machine learning job, or any kind of thing I'm about to do. Maybe you could just talk to me about what my options are right now, for the most part.

Aditya Manglik: Yeah, sure. There are a bunch of tools that I know about. The first tool that comes to my mind is something that I've looked at quite some time back, but it's a tool called PowerTOP. Just like you have the top utility in Linux based systems, this tool is called PowerTOP, and I think it is It used to be supported by Intel, and what this tool does is it tells you how much power each process is consuming on your system at any given point of time.

Now, sometimes those numbers are a little shaky. But it does a decent job. Post PowerTOP, I came to know about this interesting tool called Scaphandre. Scaphandre actually goes in and gets you the energy consumption. So Scaphandre monitors, I think Scaphandre has built some high level models for taking in all the information that we talked about in the earlier questions and actually calculating.

The energy consumption for a process. But the problem is that we assume that all of this is okay, is that if you talk about a desktop machine, it is not virtualized. We assume that all the number that you, the numbers that you see, they can be accounted to a single entity. Whereas if we go to the cloud, you have multiple entities running on the same hardware.

That's the fundamental premise of cloud, right? You want to reduce hardware costs by sharing the workloads. And that's where things get murky because we simply don't know how to separate out the energy for each entity. I think Hongyu would be able to shed more light on this.

Hongyu Hè : Yeah. Thank, thanks, Ali. I think those are really great points, especially you mentioned a tool called Scaphandre. I'm not sure if I'm pronouncing it correctly, but yeah, that's, that was one of the baselines of our paper. And indeed, as Chris has mentioned, there are tools available on Linux, it's not a thing people can use. There are tools and, for example, PowerTOP. But the models they have are coarse grained, meaning that, yeah, they are not computing by a fine grained energy attribution per user, per thread even, and we can talk about that, why that's important later, but in the cloud, as Adi has said, virtualization is a crucial technique, if not the most important technique, that enables users to share resources, but for energy attribution, actually, It's a key enemy, I have to say, because in order to get accurate energy attribution, we need to get access to hardware counters that tell us the statistics, the runtime statistics we need for our model to calculate the energy attribution. And that's, yeah, as Adi said, makes things much trickier.

Chris Adams: So if I can just take a step back for a second. So we spoke about, you've got a machine running, and, a machine will be running a series of, we might call them programs, but you might refer to them as processes, and within a given process, there might be a series of threads that's running, that kind of granularity is quite difficult, so if it's just my own machine, and it's just me using a computer, then you can attribute all of the figures to me, essentially, but when there's multiple processes or multiple programs for multiple people, working out who to share the kind of responsibility for the emissions, that's the difficult, that's the part that gets more complicated.

Hongyu Hè : Exactly, yeah, that's a fair summary. Thank you, Chris.

Chris Adams: Okay, we were talking about some of the tools available, so, and Adi, you were talking about Scaphandre as one of the tools which has become quite popular for this, for tracking some of this, but I understand, Hongyu, some of the research you did was, you've been using some other tools to help address some of the problems that you've come up against this when you're looking at basically working in a cloud like environment, for example, where you don't have absolute access to the computer yourself, for example. This was my understanding of some of the work that you're presenting at Hot Carbon, correct?

Hongyu Hè : Yes and no. Yes, it's because we are looking at how to accurately attribute energy in a multi tenant environment. And no, it's because attributing energy consumption in a virtualized environment is still an open question and we haven't solved it yet. And it will be very interesting to see future solutions to that. But indeed, we've compared with multiple tools like Scaphandre, and also the famous Cobbler, for example, but we explicitly Compared with those tools that run that target non-virtualized environment because in a virtualized environment, I think it's a fair game and no one knows exactly the wrong truth. What we found that is that existing tools are too course grain, meaning that when we use them to measure the energy of your application, for example, they will. Also mix in the consumption of other applications that run on the same server as your application does, which is a very common scenario. And we found that this could really lead to about 50% overestimation and over 90% underestimation.

Yes, in our paper, the main objective is really to measure the energy consumption of your application and only your application as precisely as possible and exclude the consumption of other applications.

Chris Adams: I can't help asking, when we start talking about these tools, is there an overhead from measuring your own footprint when you're trying to do this? Because as I understand it, this was This is one thing that has come up a few times, is that, for example, Scaphandre is written in Rust, so it's designed to be a very small, fast, lightweight program, which has some overhead, but I understand that there is going to be an overhead from tracking some of this in the first place, correct?

Hongyu Hè : Yes, that's a really great point. Thanks for bringing it up, Chris. Indeed, as you said, Scaphandre is written in Rust, and Rust is really an efficient language compared to, for example, Python and the mingle of energy attribution is to really have the precise knowledge of the energy efficiency of application so that we can improve and optimize our code accordingly. But as you said, yes indeed, there's a inherent trade off between the preciseness or how fine-grained our model is and the cost right in both terms of performance and and energy. And so our model takes into account, for example, the underlying hardware and to collect a more fine grain stati runtime statistics in our model.

But indeed, the overhead could be larger and mitigate the overhead we use conditional probability to do reasonable estimation whenever applicable, instead of trying to capture every single event per millisecond, so to speak. So that would be really costly. This part is a bit, you know, intricate, and we have more detailed mathematical formulation available in our paper.

But yeah, that's that's the high level idea.

Chris Adams: Okay, if I understand what you've been saying so far, so there's one option, which is to use like a fast programming language, which moves quite quickly, or there's another approach, which is to take A kind of sampling approach so that you are not having to, if you are using something which is maybe a little bit slower, like Python, you don't read quite so much.

And another option is to basically use something which is even closer to the metal, like in kind of the kernel space rather than in user space. And Adi, this is what I think you were talking about when you were talking about kernels. Is that, is that the case?

Aditya Manglik: Yes, that is the case. I think there's a very interesting data point that I read in some blog by Microsoft and what they told is basically, so there's three ways to measure energy. First is that your hardware directly tells you that, okay, I used X amount of joules and that would be a 98% accurate number.

It's not, it's still not 100%, right? Because of thermodynamics, but you still have 98% accuracy that, okay, this is the energy that this particular hardware device consumed. The next best step after that would be a kernel level measurement and a kernel level measurement would be, if I remember correctly, they pointed as 85% accurate.

And that is why it would be great to have something in the kernel, and that is why MacOs and Windows put these systems in the kernel to monitor the energy consumption. And finally, if you have something from the user space, now, it's not that accurate simply because it has visibility into a very small subset of the information that you need to get high in enough accuracy, and I completely agree with your point, that the more accurate you want, the more measurements you make, the more energy you are going to consume, right? So it's like a, it's like a catch 22 situation. I want to calculate something, but in the process of calculating it, I'm increasing the load on the system, and by increasing the load on the system, I am increasing the energy consumption.

You need to find out a good balance. between hardware and software based measurement mechanics.

Hongyu Hè : I think Adi mentioned a really great point, so I think the trade off is not necessarily in the programming language itself, but it lies in the model itself. So as Adi has said, the more fine grained your model is, the more costly it's likely to be. And I think we really need to strike a balance between how detailed you want your measurements to be and yeah, the cost it comes with it.

Chris Adams: I see, okay, and maybe this is a chance for us to zoom out a bit because, as I'm aware, one of the projects that the Green Software Foundation is currently involved with is this project called the Real Time Carbon Standard, the idea of creating some of this as something like a way to report these kinds of figures. As I understand it, one of the tools it seems to be standardizing on, and this is a project which is led by Adrian Cockcroft, who is a former VP of Sustainability and Cloud at Amazon and has basically a 20 plus year background working in this field, I believe they're settling on one tool called Kepler, specifically which ties into kernels to provide some of these numbers, but even then there is an ongoing discussion about, okay, how do you make sure that you have access to, how do you report numbers that are actionable, that developers or designers can use? Without actually disclosing too much information that might be a, a possible source of attack, like a kind of side channel attack, for example, and also what kind of resolution is necessary. Now, as I understand it, I think one of the things that people are pushing for there is the idea of going for minute level resolution rather than millisecond level resolution.

So at cloud level, that would al already be way further than what we have right now, but that might in theory, give you enough to then get an idea about what kind of impact you choosing to use, say, a computing job in one place might be compared to another, or at least give you something to optimize for carbon at that point. This idea of actually exposing the energy being used at this kind of level, I think there's a term that was mentioned in one of your papers about Energy Aware Computing or Energy Aware Cloud Computing? I'll ask you a little bit about this because I know that this was something I had to take away from you, but Adi, I'll come to you on this afterwards actually, because I think this is something that you've actually been speaking about at the LF Energy Core Forum as well actually. So. Maybe you could actually explain this idea of one energy aware cloud environment might actually be Hongyu.

Hongyu Hè : Yeah. As you mentioned that there's a great tool called Kepler. And I think, yeah, this kind of tools is very instrumental. as to, um, what kind of information they can give to both the users and the cloud. operators. And by Energy Aware or even Energy Intelligence, which is another level, is that we can make our decision based on, for example, the energy statistics we collected, for example, using those coupler or energy altogether tools to make decisions that optimize for, not only for performance but also for energy efficiency. And the reason for that is because data centers itself, or even networking, has huge potentials and they have great, you know, energy flexibility and we can use this kind of elasticity to do great things. For example, using data centers as energy storage or energy power bank for the smart grid. Yeah, I think that's one of the ideas, but there are, definitely a huge number of challenges we are facing in order to achieve this kind of energy aware cloud or energy intelligence cloud.

Chris Adams: All right, that feels like it's going in a somewhat different direction. So that's basically, but all that is necessary, in order for that to be possible, you can't be driving blind if you want to have this kind of awareness of what the grid's doing right? That's one of the ideas behind that. Okay, so maybe we should touch on why are we doing this in the first place? Because we spoke a bit more about yes, energy is coming from burning fossil fuels. We're not going to entirely transition our fossil fuels tomorrow so as long as we're burning fossil fuels to provide power there are going to be carbon emissions associated with this.

Adi I'd just like to speak to you about why you got involved in this, why you got excited about this in the first place or why you did do this? Because there must have been some process before you decided to try presenting a, like an energy conference and talk about personal computing in the first place.

I, I was quite surprised to see it, but I was very pleased to see someone actually talk about this and talk, talk about making some of this measurable.

Aditya Manglik: Chris, it's a personal story. It goes a long way back when I was an undergrad and in my undergrad, in my, I think, junior year, I had a laptop, which was not the best, and my battery had started to die out. I had exams to prepare for and my battery was acting up and I could not figure out I just charged in the morning why is it dead in 30 minutes so as a very simple minded engineering student my mind immediately went to the problem okay the battery is working good what is consuming energy let me kill off the applications that are consuming energy and that's how I got into the question Okay, I need to figure out which applications are consuming energy in order to kill them correctly, right?

And that's where the entire journey started. I could not figure out. And then it grew on me that, oh, how do we figure out? Because if we can figure this out... We can do a lot more very interesting things. So for instance, I think Hongyu mentioned really great points about energy aware scheduling in data centers, and I see a lot of effort from these hyperscalers to schedule workloads when renewable energy is available.

So when you talk about solar or wind energy, one of the key characteristics of these sources is that they're not 24 7 available. They're available in abundance. At a fixed point in time, and then they fluctuate a bit. So what you would want to do is you would want to maximize the utilization of these green sources when they are available. And if you can schedule your workloads at the right point in time, you can really decrease your carbon emissions. You can really decrease your utilization of fossil fuels while also maintaining your service level agreements with your customers. And that's a win situation for everyone.

You see how this simple problem of not being able to find out the energy consuming applications on my system turned into trying to save the world by reducing the energy consumption of data centers? I don't know. Yeah, so it's been a fascinating journey and I would love to keep going on this. But yeah, thanks for the question.

Chris Adams: Alright. Okay. So there is, um, I think what you might be referring to here is this notion of carbon intensity changing, depending on how abundant renewables are on the grid, for example. Is this something that you touched on as well when you were doing the work for your research on what you were presenting at Hot Carbon?

Hongyu Hè : Yeah, thanks for asking Chris. It's a great point that as we've discussed like how to use data centers as a utility or power bank, but I think our work is mainly targeting a user level optimization. And as you previously said, I really echo with The concerns from AWS so you know, the amount of information you're exposed to the user and the security concerns that, uh, come with it. And I think we need, really need to strike a balance, uh, between the two because users really need the information to optimize for energy efficiency. But on the other hand, you can't really, you know, expose too much information to the users because of the potential security concerns. And that's really, you know, a, a base, the, um, the virtualize the goal of virtualization, for example.

So it's tricky. But yeah, I think we need, uh, at least get something out for the user to optimize for energy efficiency.

Chris Adams: Okay, all right, so this is where some of this kind of cloud computing might be actually heading towards. Adi, you mentioned something about this idea of being able to control or adjust the carbon intensity of electricity by choosing certain times of day when there's an abundance of power in the grid. As I understand that, that's basically one of the reasons why you might do that, is because is that because there is the assumption there's more power than can be used. Maybe you could talk a little bit about how timing power, timing your usage when there's more renewables on the grid actually does help. Maybe you could expand a little bit on that because I know there was some useful research and I'm trying to find a link for it to bring into this, because there's a really nice model that's actually written in Python that actually demonstrates this and I found some pretty eye opening figures for it, but I figured maybe you might expand on or touch on some of this yourself because it seems to be something that you have an interest in as well.

Aditya Manglik: Absolutely. Let's take a look at two points and I think that would really help make this clear. Majority of us are working in the day, right? We go to our offices and we go back to homes. So majority of us use our devices during the day. And that's when we introduce a lot of work for the servers, right? So the data center operators like to call these patterns as diurnal patterns in which the usage spikes during the day and then dips during the night because people are sleeping.

And let's take a look at the second point. Second point would be, for example, the availability of solar energy. So solar energy, as you can guess, would be much more plentiful during the noon. and let's say less available during the evening. So what you want to do is you want to maximize the use of solar energy when it hits the peak.

But it turns out that people often maximize the usage of these devices after lunch, right? So what you do is if you have a surplus of energy available, you use it to schedule batch jobs. What do we mean by batch jobs? These are long running jobs. For example, training neural networks. During the time when solar is available, and you also keep serving your users and your customers using different sources of energy as and when they're available.

I really hope this example drives home the fact that careful balancing of our work as well as the availability of energy to do that work well, it really makes things happen for everyone.

Hongyu Hè : Yeah, so one quick point I have regarding what Adi has just mentioned is that actually I've done my bachelor thesis on energy procurement and modeling of energy in data centers. Actually it's quite surprising that loads of green energy is being dumped. And actually, uh, the, the, the smart grid is rejecting those green energies because, as you said, some parts of the world, uh, have a lot more excessive green energy than other places, for example, uh, Virginia, and, and I think it's a two-way bridge. By exposing more information to the users, on the other hand, cloud providers can also get more information about their workloads. And this can also benefit to their operation as to how they operate their data centers more efficiently and to participate more in the energy grid.

Chris Adams: I'm just gonna round up for the last few minutes. And I was just going to ask, if the people are interested in this kind of work and this kind of projects, how do they start, or what kind of tools would you suggest we look at? For example, if I start with you, Hongyu, then Adi, I'll come to you next. Hongyu, let's say someone, they've got some servers, or they're running some computing, and they want to start experimenting with these figures here. Where do they find out more about this? Is there a project that you would draw people's attention to, to look at on GitHub, or is there a thing you can pip install, for example, if you're running a computer, something like that?

Hongyu Hè : Actually, we implemented a prototype for our theoretical model called EnergAt, which is available on GitHub. Because we want to evaluate, uh, our theoretical model, uh, experimentally. And yeah, it provides users with both a command line interface and a Python API. So you can just download it by, uh, just pseudo pip install EnergAt, so E N E R. G A T. And sudo is very important here because we need the root permissions and it's being validated so you can find the details of our experiments in our paper, but in a nutshell, it can really precisely measure the energy consumption of our applications, even in a multi tenant environment. But it's not perfect.

As Chris has mentioned, If you want to contribute, there are plenty of opportunities. So, for example, we need a secure and efficient hardware software interface for energy reporting. And also, attributing energy in a virtualized environment is still an open question. And we might want to support more devices and more fine grained accounting as well. Yeah.

Chris Adams: Okay, cool. Thank you, Hongyu. And Adi, I think I'll leave the last word with you. If there's any projects or links you would direct people to, if they have an interest in any of this and would like to learn more.

Aditya Manglik: I think that's a very good question because people need to be aware of this. I think our audience would be using diverse devices. So please go to your device. If you have a Windows device, do pseudo parametrics and see what you get. If you have a Mac OS, go to the activity monitor and see the energy impact, okay?

Just, just see how much each process is impacting your battery. If you have Linux, please download Scaphandre and see how much Chrome or Firefox is using. And if you're really technical, please come and talk to me and Hongyu and we would love to dive deeper into more and more tools and help you solve your problems.

I really hope that gets people started. You can also look into Android and iOS because both of them report really good data about what these processes are using in terms of a battery. And once we build up enough awareness, I think then we can go deeper into. How to make these models better and how to reduce it, right?

Chris Adams: Okay, cool. Thank you for that, Adi. All right, so we've got options across all of the tools you might have there. And there's at least one thing people can start playing with. All right. Okay, gents, I think that takes us up to the time that we have available. And, yeah, thank you very much for coming on. And I quite enjoyed nerding out, plumbing the depths of finding out how to actually understand the energy used by various parts of our computing. Alright, cheers folks, thank you very much for your time, and yeah, I'll see you on one of the future episodes, alright? Take care folks, thanks.

Aditya Manglik: The pleasure was all ours, Chris. Thank you for having us on this call. I really enjoyed it.

Hongyu Hè : Thank you very much, Chris.

Chris Adams: Hey everyone, thanks for listening! Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please, do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode!

The Week in Green Software: Carbon Aware Spatial Shifting

26 juli 2023 | 39 min

Host Chris Adams is joined by Niki Manoledaki of Grafana and Ross Fairbanks of Flatpeak in this edition of TWiGS focused on Carbon Aware Spatial Shifting. They dive into Amazon's 2022 Sustainability Report, highlighting 19 AWS regions powered by 100% renewable energy, and explore videos from the Linux Foundation energy summit (links below). They also discover the importance of measuring carbon footprints in personal computing and IT, and learn about Kepler Power Estimation and the PLATYPUS Attack. Plus, they share some exciting upcoming events from the CNCF and some interesting Barbenheimer inspired portmanteaus from the world of Green Software!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Ross Fairbanks: LinkedIn / Website
Niki Manoledaki: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Amazon 2022 Sustainability Report — 19 AWS Regions are 100% renewable | Adrian Cockcroft [4:38]
Videos are out from the Linux Foundation energy summit - a selection [9:12]
HTTP Response Header Field: Carbon-Emissions-Scope-2 | IETF [21:28]

Resources:

Boavizta Project | GitHub [6:44]
Data insight: Heavy industry gains ground in Europe's corporate renewable PPA market | Energymonitor.ai 8:00]
Measuring Carbon Footprint of Personal Computing | Aditya Manglik LFE Summit [9:30]
Measuring IT Carbon Footprint: What is the Current Status Actually? | Tom Kennes [11:18]
Kepler Power Estimation [17:47]
PLATYPUS Attack [20:12]
Utilization is Virtually Useless as a Metric! | Adrian Cockcfoft [20:56]
Reducing data center energy usage with Grafana: A green IT success story | Bertrand Martin, GrafanaCon 2023 [22:21]
Optimizing Full-Stack Sustainability in a Real World Data...- Chen Wang, Hua Ye & Fan Jing Meng | GrafanaCon 2023 [23:15]
HotCarbon’23: Bringing Carbon Awareness to Multi-cloud Application Delivery | Dityaroop Maji [28:10]
Join Cloud Native Computing Foundation on Slack [37:01]
Climate Action Tech [37:31]

Events:

CNCF Global Week of Cloud Native Sustainability October 2023
New Working Group in the TAG: Green Reviews August 2nd 2023

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Niki Manoledaki: When you have that kind of data at your disposal that you didn't previously have, it can really tell a story that you can show to someone else and say, Hey, look at this dashboard that you just see how the energy consumption and temperature and CPU usage correlate with each other. And I think it's fascinating, and I hope we see more of these visualizations.

I'm your host, Chris Adams.

Hello, and welcome back to the Week in Green Software on Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. Today, we're diving into Amazon's 2022 sustainability report. And we'll be exploring Carbon Aware Spatial Shifting with Karmada, Kubernetes, and a new real time carbon footprint standard.

And we'll also be covering a few future events with Green Software. But before we dive in, let me introduce my guests for this episode of This Week in Green Software. With us today, we have Niki Manolodaki. Hi, Niki.

Niki Manoledaki: Hi, it's so nice to be on this podcast. I'm a long time listener, so I'm very excited to be here.

Chris Adams: And we also have Ross Fairbanks. Hey, Ross.

Ross Fairbanks: Hi everyone, I'm also another long term listener, so yeah, excited to be here.

Chris Adams: Cool. All right, before we start, I guess maybe we should do a quick round of introductions for what we do and what we work on, and then we'll just get right into the format of running through some of the news stories that caught our eyes and sharing a few kind of lukewarm to hot takes, depending on how we're feeling.

Okay, Niki, are you okay with me just handing over to you first?

Niki Manoledaki: Yep, so hi, I'm a software engineer at Grafana Labs and I'm working on also the back end of Grafana itself. I was previously at Weaver. It's where I did work on EKSCTL, the CLI for Elastic Kubernetes service. So excited to talk about um, the progress with AWS today, and I'm also a maintainer of Kepler, which we'll talk about very soon as well, and part of the CNCF TAG, the Technical Advisory Group of the Cloud Native Computing Foundation for Environmental Sustainability.

So we have a couple of things coming up there as well. We have the global week of sustainability in the second week of October where we'll have a bunch of local meetups on the world during the same week to talk about sustainability in cloud computing and we have a new working group in the TAG. That's we'll talk about as well.

Chris Adams: Cool, exciting. Alright, thank you for coming on then, Niki. And Ross, I know that we've worked together a few times, but for listeners who have not been tracking the repos that we end up messing around in, maybe you can introduce yourself and provide some background as well.

Ross Fairbanks: Yes, yeah, I'm a developer at Flatpeak Energy currently, but I've also worked with Chris at Green Web Foundation on various projects there. The main ones really would be Grid Intensity Go, which is a Go library for carbon density metrics, and also has a Prometheus exporter. I've also worked a bit on Scaphandre as well, which I think we're going to talk a bit about later as well, on some of the Kubernetes integration there. I'm trying to learn some Kepler Kepler to learn it from Niki back there.

Chris Adams: Cool, I guess learning from the horse's mouth, as it were, or whatever animal metaphor we're going to use for this. Okay, folks. For, if you've never listened to this podcast before, my name is Chris Adams. I am the executive director of the Green Web Foundation. And I'm also the policy chair for the Green Software Foundation.

I'm also one of the maintainers of co2.js, a library for calculating the environmental impact of digital services. I help organize an online community called ClimateAction.tech, where a number of climate aware techies tend to hang out. So if you haven't listened to this podcast before, the general format is we run through some of the stories that have caught our eye over the last week or so. And sometimes this will be suggestions from the actual guests themselves.

And, uh, I think we're just going to start off with one of the big ones, which has made some news in the last week or so, is the Amazon 2022 sustainability report.

So this, this was released a week or two ago. And, uh, there's a few kind of relatively large like, findings from the report that comes out each year. And what we've linked to is a summary of a blog post by the previous VP of Cloud and VP of Sustainability there, Adrian Cockcroft. There's a few highlights.

One of the key things is that Amazon are now claiming that 19 of their regions are running on 100% renewable energy, which is a increase. a significant increase from the 13 from the year before. They've also done something interesting in that they are now being much, much clearer about which regions are running on what they count as 100% versus over 95%.

You can see a few new regions in both India and China, which is a real shift and we've got one in Spain as well now actually as well. So Spain and Zurich. The other thing that it might be worth sharing is when you look through this report, this is the first time that you've actually seen Amazon show a reduction in emissions year on year.

So this is It's actually one of the largest companies in the world shifting. So this is actually a really significant view here. Now, the other thing that it might be worth talking a little bit about is that when we talk about renewable energy here, Amazon is using this market based method and the blog post we've linked to talks a little bit about how there are different ways of measuring the environmental impact of electricity, whether something is location based, where you look at the energy.

from the grid specifically, or you use a market based approach, which takes into account that you have seen significant investments in renewable energy by various organizations to speed up a transition. Ross, I know that you've had a chance to skim over there, and in the context of working with Scaphandra and trying to expose metrics.

Is there anything that caught your eye here?

Ross Fairbanks: Yeah, so what I find interesting was the part about as we will use more renewable energy, the scope 3 emissions become more important and how it's really hard to get data for that. And I think it's really interesting to what the Bovista project are doing, where they're producing an open data set of the embodied carbon in devices, using the data they get from the manufacturers kind of cryosourcing information. And I think as scope 3 emissions become more important, these sort of projects will become increasingly important.

Chris Adams: Okay, so you've used a couple of words that I think we might need to just break down when we talk about this. So we spoke about scope three emissions here, and these might be considered like supply chain inside something. So while there's a carbon footprint from obviously burning fossil fuels to generate power, to generate electricity, you might consider that scope two here, or if you have to burn, say, fossil fuels to run a generator, then that might be scope one in this scenario.

And actually this is one thing that is mentioned in this report is a shift to using biofuels rather than fossil fuels for running backup generators. Scope three, if I, as I understand it is all the supply chain. So that's all the emissions caused from making a server in the first place. That's what you're referring to in this case.

Yeah.

Ross Fairbanks: Yes, so it's scaling the emissions for those hardware devices.

Chris Adams: Okay, cool. Thank you for clearing that up. All right. Okay. There's one other thing that I just might draw your attention to that really caught my eye on this is, this is not Amazon's report specifically. There's a kind of corresponding link from this for, from a website called energymonitor.ai and, uh, they're basically quantifying the amount of renewable energy being purchased in various sectors. And for the last, say, 10 years, one of the big things has been that technology firms themselves have been basically the largest investor in renewables. But we've seen another shift in the last year in that we've actually seen heavy industry moving to actually eclipse this.

But even now, despite that. Between 2023 and 2022, Amazon is still making up for like 20% of all the renewable energy being bought in 2020. And this is the figure that kind of blew my mind was two thirds of all the investments in renewable power right now is coming from Amazon. And this complicates the matter somewhat because for a long time we've generally seen Amazon as being one of the kind of laggards here.

But one thing we see from here is basically that it's more a function of the size because they're so large and there's so much to be moving they can still be investing a significant amount and still not be as moving proportionally as far as some of the other companies that gives you an idea the size of the change we need to actually be making.

And we spoke a little bit about Kepler and Scaphandre and stuff like that and I wanted to just just see if we can jump into the next story from this actually. So we spoke a little bit about We've mentioned at previous, uh, episodes, we've spoken about the Linux Foundation Energy Summit. And, uh, there was a bunch of really interesting talks given there.

But the recordings of these talks are now online for people to see. And, uh, there was one talk, which is a particular reference, which is one from, uh, person called Aditya Manglik at ETH Zurich. He was talking measuring the carbon footprint of personal computing. Now, I don't know if you've actually seen any of this, but this one really caught my eye because this was someone basically saying, look, we need to have ways of reporting the environmental impact of software at a kind of computing level.

He was talking about okay, Windows has all these tools, and OS X has all these tools, but what we really need is something to run that path for all the servers in the world. And when I spoke to him, he didn't know that much about Kepler at the time, but that was a new thing for him. He's now looking into this.

And I figured this might be something that might be in your wheelhouses, folks, because, as I understand it, Kepler is one of the projects which this person was actually essentially calling for. What we need is something that works at Linux's level to actually start reporting these numbers. And, uh, Niki, is that somewhat related to what Kepler does?

Niki Manoledaki: So what Kepler does is it leverages EBPF to look at the kernel level syscalls and performance counts and it's attributes of those with Kubernetes resources. So looking at the energy processes, for example, in RAPL in the kernel is not something that is necessarily new and there are other tools such as Scaphandre that also do this.

What's new with Kepler is that this attribution of the energy consumption with workloads running in a container. So that's really what's changing things for at least in the cloud native ecosystem is this part and to add to this, I would like to mention this one really interesting study called Measuring IT Carbon Footprints What is the Current Status Actually? Which came out in June of 2023 on Tom Kennes' sorry. I mispronounced your last name, Tom, but he's very active in the TAG for environmental sustainability and what is interesting to notice what we just discussed previously with reporting carbon through AWS is a top down carbon monitoring, whereas what Kepler does and what the talk that you just mentioned, what it focuses on is bottom up carbon monitoring or energy monitoring first, because that would be the first step infrastructure.

So that bottom up approach to energy and by accent carbon monitoring is much more useful for engineers. So it's really talking about the persona in observability of who are those metrics for and what are they used for. So we see top down carbon reports useful for carbon accounting for the center operators for perhaps CFOs or whoever is reporting, whoever is using those reports, but for engineers who are optimizing low level software, Kepler is much more useful in those use cases.

So, Okay, cool. Thank you for this. And I just want to ask, Kepler, yeah, that's a reference to the astronomer from a few hundred years ago, but Kepler is also an acronym, right? I can never remember what it is. Is it Kubernetes? Yeah, help me here, Niki, because I always, it always sounds cool when I hear it. It's super nerdy, but

yes, it's a great acronym. It's the Kubernetes based efficient power level exporter. So it exports the data to Prometheus. So you can then visualize those data, that data on uh, Grafana dashboard. And there are some talks out there and there are some really interesting data visualizations that you can gather that way.

It's a really interesting setup and you can really tell a story through that data. And that's, again, coming to the point on personas of who is this data useful for and for what? Is it for like a platform team that is doing cost and performance optimizations? Is it for SREs? Is it for software developers themselves who want to monitor the energy consumption of their software, like on a release, from one release to another, and how these have changed?

So really thinking about the persona in the story.

Chris Adams: okay, cool. And Ross, I understand that you've done a bit of work with Kepler as well, right? And you've also done, you've contributed some code to Scaphandre and some of the other ones here.

Ross Fairbanks: I've looked at it from that angle. I haven't looked at Kepler yet. But I think because the REPL measurements are at the CPU socket level, being able to assign those, well, first to the process, and then to the container, and especially in Kubernetes, namespace, then, like Niki says, you can provide much more context on what is this process actually doing. It's also one of the challenging parts as well because with Scaphandre and I think with Kepler as well, we have the individual process, but then we need to use the secret file system, um, to then work out which container was this and then can get up to the pod level. So that kind of mapping is quite difficult, but that extra context is really useful in those situations I think.

Chris Adams: So I can't code Rust, but I try to at least write the documentation for how some of this works. And if I understand what the two of you are saying is that tools like Scaphandre or tools like Kepler, they essentially allow you to figure out what share of a machine's usage should be attributable to a particular program.

If it's using half the power, then you can say half of it should go to that and that's how you might track it across a fleet of computers. And I think you folks also used this term REPL or RAPL. And I forget, this is a reference to the fact that certain computers, some have chips on them, which will basically share information about the actual energy being used.

So if you know that, say, the computer is using maybe 40 watts of power, and it's using half of it, you might allocate half of that 40 to one program. Is that the general idea that these things use? Or that's the kind of approach they tend to take?

Ross Fairbanks: Yes, yeah. RAPL is an Intel technology, and so that's the most commonly used. I think with Cloud Protocol, there's also an estimation model that can be used in cloud settings. This is one of the strange things where actually it's easier to do this on bare metal because then you can access RAPL, whereas doing it in the cloud because you haven't got access to the physical machine, it starts to get a little harder.

Chris Adams: Ah, okay. All right. Thanks for the sharing the extra nuance. I didn't know that Kepler could do that. That's really cool, to actually do that without having access to the computer under the hood. Okay, so, Ross, you mentioned, so I know that you've done a bit of work with Scaphandre and other tools like this, and I've been trying to understand how some of this works as well, and I think, as I understand it, these tools will basically, you've got two kind of parts here.

You have one part which essentially measures how much of a. machine is being used for a particular process, a particular program, and then there's this combination with this thing you mentioned before, RAPL, which I think is it running average power limit or something like that, and that essentially tells you what power is being used.

So if you know that a process is using half the compute in a computer, and you know it's using maybe 100 watts of power, then half of that 100 would be 50 watts, so over maybe a couple of hours, you would attribute half of the power to it like that. That's how RAPL works. And Nikki, you mentioned that Kepler does something like this, but it also has a model as well.

Niki Manoledaki: It has a model and I think also because RAPL is not accessible in a lot of cloud platforms in most of the workload types. For example, on AWS, most Institute instances don't give access to RAPL, and only the bare metal instances do, which also, side note, bare metal instances on AWS are more expensive than other easy to instances.

So there is a little bit of a catch there. And the Kepler power estimation model helps to limit some of that and estimate some of the power consumption. And we'll dig some of this documentation in episode notes.

Chris Adams: it. Okay, so that's how I understand the role that these two things play. And now that we understand that there's been an issue about actually having access to the power usage, because you might have an idea of some of these tools will tell you we're using 100% or 50%. But if you don't know what the actual number is, you're like, 50% of what?

Or something like that. That's one of the things we're struggling with. And as I understand it, this is probably some of the impetus behind some of this new work that we've seen with the real time compute standard from Adrian Cockcroft, where he's basically been saying, look, if we don't have the concrete numbers for electricity, it's gonna be really hard for us to work out the footprint of any of these tools.

And therefore, we need to have something like this. And this seems to be one of the new projects that was based around Kepler for this. That's, I believe that's my understanding, but Nikki, I wanted to ask if you've been exposed to any of this because I think there have been some conversations with people in the Kepler community about some of this or about figuring out where to go, is that correct?

Niki Manoledaki: I'm wondering if this is we have a demo from I think it's a scene from the Green Software Foundation on the 2nd of August. In the CNCF TAG, and we're going to be talking about Specification. I wonder if it's going to be about this because I haven't heard of it until now.

Ross Fairbanks: Yes, one of the things we can talk about with Kepler in there, the plan is I think to use Kepler for the attribution part. This part we were talking before about how we can go from the socket level to process and then to container and then up to pod. To use Kepler for that, because it's already performing that task when it's getting the metrics from RAPL. I also found it really interesting from the proposal because it goes into some of the security parts on why it's blocked on a lot of the cloud providers. And it's because if you can get very accurate energy measurements for like decryption algorithms, you can start to break the decryption. But I think the proposal has a really elegant solution, which is to expose all the metrics at one minute intervals, and if you've got per minute data, that's fine for doing carbon awareness, but people can't use it as like an attack vector.

Chris Adams: Okay, and coming from someone who's basically worked for Amazon for the last N years or Netflix, you would assume that there's some weight carried behind that, saying, yes, it is okay to provide minute level things, you're not going to get everyone hacked, yes, it's okay to use these tools.

Niki Manoledaki: I think it's called the platypus attack where some secrets can be inferred from power metrics. It's the platypus attack, if I'm not mistaken. Great name.

Chris Adams: Sounds about right, yeah. So yeah, there's a bunch of these as well, actually. I know there's one where people realize that you could actually use the flashing light on a disk access drive on a computer as a, if you know when it's flashing, that's an indication of when you're reading from a disk. And that is actually, that has been enough for people to carry out some attacks to break some encryption before.

So you can see why someone is going to be a bit reticent of this, but to actually then have someone say, I understand about security. One minute resolution is sufficient for us to keep people safe while still allowing people to report on meaningful figures is actually a big thing. And bear in mind that when this is coming from someone like, we'll share a link to a link, I think from 2007, where Adrian's writing about this, he's writing, there's a paper called Utilization is Virtually Useless as a Metric, talking about all the different things you need to take into account with cloud back in 2007.

So if almost at least 10 years later, we've actually got someone talking about this. That suggests that it has some substance to it, and we've actually got a real chance to come up with some meaningful metrics for this. Alright, we went down a massive nerd rabbit hole there, I think, folks. The next story proposed here was actually, this... There's some work in the IETF for people who are looking, who are curious about this. So the IETF, I believe it's the International Engineering Task Force. There's a current RFC, which is basically a proposal for creating a kind of carbon footprint header in HTTP requests. So this is currently being discussed.

And as I understand it, this was also an idea that was proposed. And there was even a talk. By at the Grafana Con recently, Nikki, I haven't seen this, but I wondered if you might know anything about this or if this has come up on your radar, because I know that Grafana ends up being used as the defacto dashboard in lots of places here.

Niki Manoledaki: This is the HTTP header that containing CO2 emissions has been on my radar for a while. And I only just realized that it was connected to Sentry software. So Bertrand Martin did a talk at GrafanaCon on reducing data center energy usage with Grafana. And so that's a really interesting use case. Again, looking at data center as a whole, where you have access to RAPL, you're not in a public cloud provider, you do have access to all of the data is at your disposal, and so there was, I think they reduced that at the data centers electricity usage by 15%.

Also, yeah, the temperature was increased from 18 degrees Celsius to 27 degrees and a lot of The power savings were achieved through this, and it's a really interesting use case. There was another talk that was featured at GrafanaCon, which was a talk by Chen Wang at IBM. She's also in the TAG. They were using Kepler to measure some of the workload's energy consumption.

And they also achieve, if I'm not mistaken, 75% power savings in their data center, some incredible numbers. And what both of these talks have in common, they do use Grafana dashboards to visualize those metrics. So I think there's a really interesting book on the power of storytelling. When you have that kind of data at your disposal that you didn't previously have, It can really tell a story that you can show to someone else and say, Hey, look at this dashboard that you just see how the energy consumption and temperature and CPU usage correlate with each other.

And I think it's fascinating, and I hope we see more of these visualizations.

Ross Fairbanks: I think just the part on cooling I think is really interesting. I went to actually talk at one of the KubeCons where there was someone, I think from the Open Compute Foundation was looking at it. Because also for waste heat as well, I think there's lots of potential things we can use for waste heat, for like district heating, those type of things as well. I think like heating, cooling as well as water usage, a couple of things that aren't sometimes looked at, we focus a lot on energy consumption. But there's other aspects as well I think are really important.

Chris Adams: Okay. That's quite a nice graceful link moving through to the fact that, okay, you can talk about energy efficiency all day long and uh, it sounds like there are ways to actually get access to this. And we've seen examples of talks about, okay, these are the things I can do by reducing the energy usage from this.

But there are other levers specifically around, effecting the carbon intensity of electricity if we're only going to look at carbon intensity without before looking at like changing the life cycle of hardware and stuff and Ross I think the next one is actually it's a link to a post that you shared here that I think helped explain some of the differences between the approaches people are currently taking when they do try to shift the carbon intensity of computing by either moving it through time or or moving it through space.

And you've been doing some work with a tool called Karmada that might not be that well known because most of the work happens outside of, there's a very significant community in China or other parts of the world, right?

Ross Fairbanks: So Karmada is a CNCF project, um, that does multi multi cluster scheduling for Kubernetes. So it's effectively a federation. So you have one Kubernetes cluster that's your control plane cluster, and then you can join multiple member clusters to it. And those, especially for carbon intensity, those member clusters could be in different regions, using different electricity grids. And they could be different cloud providers. And so the work that I was doing was creating a Kubernetes operator called the Carbon Aware Karmada Operator that gets a list of the clusters that are available and gets the carbon intensity for each of those locations. And it actually uses a Liquid Intensity Go project that you and I have worked on at the Green Web Foundation to get the metrics primarily from electricity maps that are used in their free tier. And then once you have the carbon intensity of those clusters, It then looks at the workloads, and you can say, I want to run these workloads in the two clusters, say out of three or four, that have the lowest carbon intensity. So that's the kind of high level of how Karmada works, and the operation of just adding carbon intensity onto what Karmada can already do.

Chris Adams: Okay, and I am aware there's another operator that was published by Microsoft which focused on moving things through time, not moving things through space. Is that correct?

Ross Fairbanks: Yes, this is what kind of referred to as temporal shifting rather than spatial shifting. And that's, temporal shifting is something I've been interested in for a long time. It's for jobs you have that aren't time sensitive. So the classic example of it is when you upload like a YouTube video, Google needs to transcode the recording, but it doesn't need to happen straight away, unless there's people actually waiting for it.

You can actually delay that, maybe even up to 24 hours, and people won't actually notice. And what the Carbon Aware Keda Operator does is it gets the carbon intensity forecast for an area, and then it sets the maximum number of replicas. So it's actually doing demand shaping. It's saying, depending on the carbon intensity, we want to run more or less of this workload.

Chris Adams: Okay, so that's one. And Karmada is doing space now. Now this sounds a little bit sci fi. Are we already doing time and space at the same time, or is this like the next frontier as it were?

Ross Fairbanks: Yes, this is the for the next frontier and the current for the work I'm doing with Karmada. It's a very simple kind of scheduling algorithm. It just uses the lowest carbon intensity. But what you could do is look at the forecast and say, actually, for the next two hours, I know the carbon test is going to be low, so I'll move this thing here. Whereas if you know from the forecast, the carbon density is about to increase. Maybe this isn't the right region. You can put it in another area. So I think as we get more into this topic, people will start doing more sophisticated scheduling.

Chris Adams: Okay, maybe this is a nice time to just jump into or refer to some of the things we saw in HotCarbon in that case Because I believe you shared a link to some work by the recent HotCarbon conference, which has its videos now visible I think there was a person called Dityaroop Maji this was related to the VMware stuff.

Maybe you could just expand on this one here, because there's a couple of other really nice talks from HotCarbon that it'd be nice to just refer to.

Ross Fairbanks: Yeah, so this is doing spatial shifting, but just applying it at a different layer. Yes, so this paper is from a team that were looking at the VMware global load balancer. And what they were looking at was, by default, the load balancer will route traffic to the closest data center. But they were also adding a carbon intensity module to say, can we actually reduce the emissions by routing it to a different data center? What's nice is the algorithm they're using also considers the location. If actually you're moving the data too far and it's going to impact performance, it takes into effect both carbon density and the location, which I really liked. And it's similar to areas we've been looking at, but just applying a different layer in the networking stack.

Chris Adams: Ah, I see. So, I've got a request coming in to visit a web page. Please generate the web page, but whoever's the greenest and closest to do this, so I can do it within a time limit, right? So it doesn't look like I'm slowing everything down. That's the general idea that it's doing it, right?

Ross Fairbanks: So you can include kind of the distance to pack acid travel as well, and I think considering the performance but also reducing emissions, and I think it was about 21% they found in the paper they could reduce the emissions by introducing this module.

Chris Adams: Without having any impact on, basically, people's endu- So it's essentially a free, in terms of user experience, there's no perceivable change, and you reduce emissions by 20%. That's the idea, in the paper.

Ross Fairbanks: Yes, yeah, although I should just include the caveat, it's a kind of a prototype that they're working on at the moment, but I think there's a lot of potential to use it in this area.

Chris Adams: Cool. All right. So this talk here is the first time I've seen someone speaking about getting rid of the assumption that you're looking at one computer and it might be that the actual resources you're using, like a disk or memory, might be physically a machine somewhere else because you've got a kind of disaggregated approach to data centers these days, rather than just having a single variant of a kind of desktop machine.

That's the key thing that I saw from it. Okay, so that concludes our deep dive into the wonders of cloud computing and Kubernetes. And if you have made it this way through, thank you for staying with us. We're just going to do a quick roundup of coming events that may be interesting to technologists who are looking at this.

Niki, I know there's a couple of events that you mentioned on the radar for you. Any chance you could refer to those or just give it add a quick reminder for people for these ones here?

Niki Manoledaki: There are a few events that we are organizing in the CNCF TAG. One of the main ones that we're preparing for at the moment is in October. We're planning the Global Week of Sustainability. So that's going to be events all over the world. I think we have a couple dozen cities represented at the moment.

Happening in the second week of October, I'll be talking about cloud native environmental sustainability in our local meetup groups, that's the CNCF meetup groups and yeah, find one near you or feel free to organize an event. We have a guide for local meetup organizers and that's very exciting. Another thing that is coming up is, uh, we do have demos and talks in the CNCF TAG, Environmental Sustainability Regular Meeting. So that's on every first and third Wednesday of the month at 5 p. m. Central European Time. And we do have a talk on the 2nd of August. By Asim [Hussain] from the Green Software Foundation, and we're going to be talking about some of the specifications and around measurements for carbon during that meeting.

And lastly, we do have a new working group for brain reviews that I wanted to give a shout out for and we're going to be meeting every second and fourth Wednesday of the month. Those meetings are open to everyone and in this working group we're going to be looking at evaluating the sustainability of various projects.

So Karmada and KEDA that Ross mentioned, for example. And so we're going to be looking at how to use Kepler on infrastructure that is available through the CNCF and how to set up those pipelines for measuring the carbon intensity of cloud native software and doing those assessments of cloud native tooling.

So that's very exciting.

Chris Adams: Cool. Thank you for sharing that. I will be showing links specifically to this so that if this is caught in one's eye, they'll see where to go to next. All right, so we've covered some of the events. We've gone into a super nerdy deep dive into the wonders of cloud computing, Kubernetes, and all the various ways you might measure that.

I think we just have to round up with some of the closing questions now. Chris, our producer, he throws these curveballs every single week. And this week there's been a bunch of hype in the news about the term Barbenheimer, this kind of portmanteau between Barbie and Oppenheimer, releasing on the same day.

Now, we've seen a few other portmanteaus, I know that Adrian Cockcroft has been pushing for DevSusOps, and if you look at the sustainable web movement, there's this term SustiWeb that's floating around. I wanted to see if either of you have any portmanteaus that you either love or hate in this field that might be worth sharing with others while we're here.

And I know there's at least one that's been shared here, so i'm not sure whose creation this one is, but maybe one of you might explain what hemigration is perhaps?

Ross Fairbanks: Yes, that would be me, yes. Staying in this rabbit hole we've been in today. Hemigration is moving applications between hemispheres. This is actually an idea that's in the GSF Carbon Awareness Docs, and it's about moving your applications to the hemisphere that has the most daylight hours to make the most of the solar power that's available.

And yeah, if you can move your workload to move it, I just like the idea of this, your applications moving with the seasons.

Chris Adams: Of course, it's like the opposite of chasing the moon, which is what people were talking about 10 years ago because we figured Because it's colder at night, you won't need so much heating. So the flip side now is, yes, it's warmer, but because there's more sun, the energy is going to be cleaner, right?

Ross Fairbanks: Yes, exactly, that's it.

Chris Adams: Okay, cool.

And I see another one which is Green Ops from... Okay, Niki, this is your suggestion, right?

Niki Manoledaki: Yeah, I don't know how common GreenOps is as a term. I haven't really heard this term be mentioned in the podcast. So far, correct me if I'm wrong, but GreenOps takes its name from DevOps and FinOps. So operations related to development or operations related to cost optimization. And the idea is to apply some of the strategies.

of FinOps for optimizing around carbon emissions and energy consumption. So that's GreenOps. It's a very loosely defined term in terms of what GreenOps looks like, what practices exist. Usually the idea is that if you reduce your resource utilization and if you implement FinOps practices you may be reducing the carbon that you emit through your infrastructure.

Chris Adams: That's your one, yeah? Yours is Green Ops. Okay, I think as I understand it, Google and ThoughtWorks are big proponents of this Green Ops term, and you'll see it in a bunch of their marketing, and their writing literature. I'm afraid I actually don't have a really good one myself, and I think, now that we actually have Fetch, I can't even make a joke about making Fetch happen.

I think I'll spare you, any of my particular kind of dad joke puns for day. But I, what I will say is thank you so much for coming onto this. I really enjoyed diving really into the depths of some of the specifics about how different tools make it possible to understand and optimize for carbon and optimize for energy use, like you mentioned here.

So yeah, thank you so much for coming on you two. I guess I'll see you folks in either the working groups or in the Slacks or in various other places. So just before I do go, I just want to check if people were interested in any of the things that you've discussed, Niki, where would you suggest people go?

Is there, if people want to find out more of the stuff you're doing, is there like one or two links that you would really draw people's attention to?

Niki Manoledaki: I would love to see people join the CNCF Slack channel for the TAG for environmental sustainability. That's where we have most of the communication. And we post a lot of links and blog posts that Ross shared and we organize through that channel. So that's our main form of communication.

Chris Adams: Cool. Thank you. And Ross, if there's anything that you would point people to, what would you point direct people's eyeballs to for this?

Ross Fairbanks: Yes, yeah. I direct them at the climateaction.tech Community, which I think Chris and I, you're both, we're both there as well. Especially the Green Room for Channel, which gets a lot of these kind of discussions, and it has, I use it, I find it really useful for researching these topics as well.

Chris Adams: Brilliant. I think that takes us to the end. This has been really fun. Thank you one more time. And that's all for this episode of the Week in Green Software. For all the resources in this episode, you can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables and see all the links that we mentioned and all the sites that we found.

See you in the next episode. Thanks a lot and bye for now.

Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

The Week in Green Software: Open Data with Fershad Irani

19 juli 2023 | 33 min

In this episode of Environment Variables Chris Adams is joined by Fershad Irani, an independent web sustainability consultant and maintainer of CO2.js. They discuss topics like open data on the greenness of power, the wonders of HotCarbon, new projects from the cloud native computing foundation and the Green Software Foundation, and gzip.ai. Fershad shares his experiences working with the Green Web Foundation and the growth of the open-source carbon estimation library, CO2.js and there’s a cameo from Fershad’s cat!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Fershad Irani: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Electricity Maps Open Data [4:02]
HotCarbon 2023: 2nd Workshop on Sustainable Computer Systems | HotCarbon [9:27]
Articles by Assaad Razzouk | Thought Leader Renewable Energy | Angry Clean Energy Guy [17:51]
Quinbrook pops up in Grok’s camp at Sun Cable, deal close | Financial Review [19:05]
Adrian Cockcroft’s Proposal for a Specification for Real Time Carbon Intensity | Green Software Foundation [20:55]
Data centres & networks | IEA [25:14]
Graduated and Incubating Projects | Green Reviews Project | Cloud Native Computing Foundation [28:36]

Announcements:

The Green Software Foundation is Hiring [30:45]

Resources:

Observable [6:10]
The Internet of Tomorrow Must Sleep More and Grow Old | Romain Jacob [10:10]
Green Networks | Environment Variables episode with Romain Jacob [11:12]
OVH weathermap [12:21]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Fershad Irani: Pardon the pun, but this turns the heat up on those or has the potential to turn the heat up on those big cloud providers and gives people a chance to, like you say, compare them on their carbon footprint, they might even need to start competing on carbon footprint because that's going to be important in the future.

Chris Adams: Hello and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Hello and welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. In today's episode, we're looking at open data about how green the power we use is, the wonders of HotCarbon, some cool new projects from the cloud native computing foundation and the Green Software Foundation, and a really cool technology called gzip.ai. Finally, we have some fantastic opportunities for you to be part of the Green Software Foundation because they're hiring. But before we dive in, let me introduce my guest today, Fershad Irani, an independent web sustainability consultant and maintainer of CO2. js. Fershad, I'll hand over to you to introduce yourself, if that's okay.

Fershad Irani: Cheers, Chris. Thanks, man. I know you've been trying to get me on this podcast for a while, so it's exciting to be finally here. Yeah. Hi folks. As Chris mentioned, my name is Fershad Irani. I'm a web sustainability consultant and I live in Taipei, Taiwan. Most of the time I spend these days is working with Chris and a bunch of other amazing people at the Green Web Foundation.

I think I've been over... It's over a year now, hasn't it, since I've been there, Chris? Or close to.

Chris Adams: I think it has been indeed. Yeah.

Fershad Irani: Yeah, during that time, we've, we've done a heck of a lot, I think, um, doing a bit of writing and a bit of coding. Chris did mention co2.js, which is where I've spent a big chunk of the last year. It's an open source carbon estimation library.

I think Chris has mentioned it once or twice on this podcast, just snuck it in there. It's been really cool watching that project grow over the year and now it's being picked up by some other quite large projects itself, such as the Mozilla Firefox browser and web page test, which is just mind blowing to think that some code I've written is in those projects.

But outside of web sustainability, I do have a bit of a life. I help organize and play in a local touch rugby, or touch football for all the Aussies out there. We have a group here in Taiwan and we play weekly and try and send teams internationally whenever we can.

Chris Adams: Cool. Thanks. Thanks, Fish. Oh, and I'm calling just for context when we work together, Fershad said Chris, you can call me Fish instead of Fershad. So if I call Fershad Fish at any point, it's, it's basically just, uh, a, a shortened version of the, of his name that he's comfortable with us using. All right.

So that's who I'm referring to when I ask. What do you think, Fish? I do not have any actual Fish in, uh, the podcast today.

Fershad Irani: Cat's out of the bag.

Chris Adams: Yeah. Yeah. Actually, is the cat also in the room as well? Like we, we gonna have a incursion from her today.

Fershad Irani: She likes her video calls, so she might jump in on this one. Eventually, it's food time at the moment.

Chris Adams: All right. So we may have a third guest as we record today. All right. I should just briefly introduce myself before we start. I mentioned my name is Chris Adams. I am the executive director at the Green Web Foundation, which is one of the member organizations of the Green Software Foundation. The Green Software Foundation, I work there at, as the chair of the policy working group, and that's basically the thing I'd probably share with you now, actually.

So I think Fish. I guess we should probably start looking through, you're familiar with the format. So should we do the usual thing of running through some of the stories that caught our eye and then basically share a bit of context for them?

Fershad Irani: Let's go, mate.

Chris Adams: All right. Okay. So I think what's the first story that we have here was actually a story from Electricity Maps.

They're one of, they're another member of the Green Software Foundation. And earlier on this month, they released some open data and like a significant amount of open data, actually. Fish, I might let you start on this one because there's, I think it's worth people understanding why this is interesting and it might be worth you sharing some of this because you've worked with a number of different providers of carbon intensity data now.

Fershad Irani: Yeah, this data that Electricity Maps has released is just a huge data dump from 2021 2022. You have almost 55 countries in their data set and it's just such an awesome amount of historical data with so much granularity, not only yearly data, which we're used to working with most of the time, but now to have monthly, daily, hourly historical data available for free.

That's something that's really going to be handy for a lot of people building out tools and analysis around carbon emissions and all that type of stuff. Yeah. Until now, for the most part, we have been working with annual. grid intensities, like the data that we've got from EMBA and we've put that into co2.js in the last year.

I can't wait to play around with, with this data set and see, um, what we can get in there in the future from monthly, daily, even hourly figures if we're able to. They do say that they are going to release 2023's data. I'm not too sure when.

Chris Adams: So as I understand, this is basically a push to essentially increase the floor of data quality that is available in the public domain for people to use so they can, Oh, there's our cat coming in. Yep. The goal is to increase the level of data quality as a floor so that rather than only having to use annual data, which often occludes and hide some of the information to provide a much higher resolution. So for example, you can see if you're going to like decide to move computing jobs to different parts of the world at different times, you can see the impact of this. The other thing that's also interesting about this is that it's actually released using the open database license, which means that you're able to build on this commercially or use it in all kinds of projects.

Now, what I have done when I found out about this is I had a go at this and I've used a tool called Observable, which makes it really easy to build little tools, little exploratory notebooks. And we've got a couple of links to essentially the hourly carbon intensity data for a series of countries that we found.

So we've got one for Germany and Finland, but basically they have one for almost every single sub grid in America, which is what people might refer to as balancing authorities in America. And this is cool because I think one of the things that I realized when I started playing with this data is that this lets me see, okay, if I did a bit of, if I had a computing job last year, where else could I run at the same time?

Or where could I move that to? over the whole kind of geographic space and time last year to see how I could have reduced the emissions for that. That's something that I haven't really been able to do before. And it's nice to see this. The other thing that is worth bearing in mind is that this, there is a commitment from the organization, Electricity Maps to publish on a yearly basis every time going forward.

So at the end of 2023, they'll be publishing the data from 2023 as open data for anyone to use. And you're able to use the data real time from them. Basically as a commercial product, right? And that's essentially what you can see this being used for. It's a way to increase the data quality used for the last year.

But if you want to do something real time, then you may need to use electricity maps or what time or some of the other tools, depending on what your specific use cases, but this being out is a really cool thing. I'm really happy to see it.

Fershad Irani: And just on that last point, like, we do need more of this and as much of it as possible to be open source in terms of monthly, daily, hourly emissions data. And if that can come from governments or from other private entities, that just helps all of us in this space. Like it, it helps drive decisions like you were saying about.

Carbon aware computing and stuff like that that also helps improve the accuracy and transparency of the carbon estimates that we're producing it that we're going to start relying on for reporting and other legislation that comes in the future.

Chris Adams: Yeah, like, I feel this is quite useful because this publication, if you're looking at the carbon intensity of anything you did last year, you've basically gone from, essentially you made something 8, 760 times more accurate because you've actually got hourly figures for this stuff, which has been really hard or really expensive to get access to previously in this kind of way.

So that's cool. I don't think that's worth bearing in mind is that you have to ask yourself, how many times do you need to pay for this data? Because you may not be familiar with how kind of literature grids work, but in many parts of the world, there is a small levee that's put on to that's built into the kind of basically hourly rate you pay for any power is usually between 20 to 10%, a lot of the time, which is essentially allocated to potentially funding a transition to renewables for this stuff. And this information is collected anyway for this. So the idea that it's actually visible and that it's available in the public domain is a really good thing and really long overdue. So it's nice to see this. So yeah, good news story. We've shared a link to the data portal and it's free for anyone to use and fetch the data from. And hopefully we should see this turning up in places like the CarbonAware SDK and any other tools like CodeCarbon and so on, so you can start making more responsible decisions about when and where. You run any kind of computing jobs.

Let's look at the next story. The next story is about HotCarbon, which is a online conference, online and hybrid conference that's initially based in America, but the cool news is that it's basically sustainability, ICT, nerd Christmas, there's a bunch of really good papers that have been released, and there's also now the recorded videos of all the talks from this.

And if you are trying to find out what the state of the art is. in discussions around digital sustainability. This is one of the places to look for the kind of technical discussions about this. And there's a couple of talks that I think both Fish and myself have really caught our eye. Fish, I'll let you go first.

And then I'll come in with my one actually, because I think there's one that you. Quite liked, right?

Fershad Irani: Yeah, my one's actually from last year's HotCarbon. There's actually a paper from Romain Jacob and Laurent Vanbever, both from ETH in Zurich. And in last year's HotCarbon, they published this paper, which is just, it's got just a beautiful name, if anything else, The Internet of Tomorrow Must Sleep and Grow Old.

And that was, I think the very, probably the first, if not one of the first times that I've personally really started thinking about with the data transfer is the best proxy for website carbon emissions and how we calculate them that kind of began a rabbit hole for me and I'm still going down that rabbit hole as I think a lot of us are. But it was really interesting and presented some of the the ways that networks operate and function presented that really clearly and the video for that is is a really good short 20 25 minute watch I think it is but they've also got a paper this year with kind of a less pretty sounding name, but Chris, you want to talk to that one?

Chris Adams: Yeah. So first of all, before I talk about this one in particular, I'll just let people know that last year we did an interview specifically with Romain Jacob about the paper that he shared last time. So we will share a link to that to go into more detail about it. But the general thrust of the paper from last year.

Was that the internet is basically provisioned in its current state for availability above all else, which means there's lots and lots of the time we've massively over provisioned for it. So big, it's like having the biggest possible computer you can imagine just for when most of the time it isn't actually used that much.

This time, he's actually, Romain Jacob is the, one of the lead authors, along with Jackie Lim and Laurent Venbever, I think, from ETH Zurich. They're talking about, are there ways to do something about this? And it's not such a poetic name. But the general thrust of this paper is that given that we know that most of the time we're not using the entire capacity of the internet, is it possible to kind of power down parts of it as it were, is it possible to make parts of the internet sleep so that you can make meaningful reductions in the energy usage and as a result, the carbon footprint of this stuff.

And the argument basically is that yes, you can do some things like this. There are savings in the order of tons based on looking at a open data set from OVH called the weather map data set where OVH, which is another cloud provider have basically shared the traffic that they have running inside their own networks.

And they basically explored this and said, given what we know about how the internet is used and what kind of usage patterns we have, is it plausible to selectively power down parts of the internet and still maintain like the same level of quality of service basically. And it's super nerdy. But it's a really nice cool paper and it's a fun read.

It's one of the first times I've seen people actually work with real data from a real organization, because one of the thrill struggles you have is actually having access to this information. So this is really cool to see this. There are other more ones. There are many more papers as well, but I think what we might need to do is run through the list and see if we can get some of the people from HotCarbon at 2023 to speak about this, because there was a number of really exciting looking papers and there is 20 videos and 20 papers to read through.

So if you want to. Basically see what's happening at the real cutting edge. That's a place to look.

Fershad Irani: Just to be sure that HotCarbon's already happened, right?

Chris Adams: Yeah, it happened a couple of weeks ago, but the videos were literally published. I think last night or two nights ago or something like that. So there's a bunch of, that's the place to look.

Fershad Irani: So hot off the press.

Chris Adams: Yeah, absolutely. HotCarbon, hot off the press indeed. And if you want something even hotter, there is a mailing list called the E Impact mailing list, which I'll share a link to, where there are ongoing and robust conversations about all this stuff here.

So Fish, you spoke to this idea about, okay, is data transfer a good proxy for understanding website carbon emissions or anything like that? That's the place that I am usually following to see what the conversations are, going back and forth on that stuff. And it's a really useful place to learn from essentially world experts for free about what's happening there.

Fershad Irani: Do you want to do a spicy take and give an answer to that one? Is data transfer the best proxy for website carbon emissions? Chris Adams.

Chris Adams: I'm not going to have a spicy take on this one yet, because I'm still trying to figure this one out. Because I feel that there's lots and lots of evidence that basically shows that the network part doesn't change all that much based on what you send over the wire. So you can make the argument, rather than thinking about it like a kind of road and cars driving, it might be more useful to think about networks as like a cycle lane where you have people using it.

So. You know, if you in aggregate, look to all the people cycling on a cycle lane, you might see a small change in usage, but you're not going to see a massive changes if you had like loads of cars driving along it. And I think this is an issue of us having mental models or not the correct mental model when we think about this stuff.

That's about as spicy as I can really take. Cause I don't think I know enough about it, but Fish, we should probably share a link to your piece, because this is one thing that we've had. Bunch of time talking about with both implementing the sustainable web design model in co2.js, but also because there's a whole separate discussion about this, both at a regulatory level, but also in inside industry with actually the sustainable web design model specifically, there's a whole bunch of work going on there that I suspect you might have some reckons on or something you could share on there actually.

Fershad Irani: Yeah. And I think it's also worth noting that there are other methodologies for estimating website carbon emissions or digital carbon emissions out there that don't use data transfer necessarily as their proxy. And they use other things like time on device or they try to measure the actual usage of a device, which is something that you can also do these days in.

The Firefox web browser, which is super cool. And I'm with you. It's something that we're all learning as we go. And there's more research coming out about it for now. Data transfer is the best we have, but with what's in Firefox, hopefully other browsers can implement that type of technology as well. We can start to see some real world data that we can then base some of our estimates and assumptions off.

And we can then work with that.

Chris Adams: There is one thing that I would wish for, if we could see something like this for HotCarbon 2024, this whole paper here is based on the willingness of one organization to share some data about how a network is working so that it can form basically a public understanding of where the real impactful decisions and interventions are possible can be made when we think about greening software, right?

We know that browser makers like Firefox and Chrome and Microsoft Edge, they have all this telemetry information about how their browsers are being used because they use it to improve the products, right? If there was a way to share a suitably safely prepared data set, which was a representative sample of how websites and things were used, it will be so useful for us to actually understand this.

And now that we've done a bit of work with say Firefox, for example, we understand that these numbers, they can be collected and they can be used because. If you're using Firefox now, you can basically turn on the Firefox kind of profile and you can see right down to the process or thread level, what the energy impact of various parts of the page are.

And we know that some of this stuff is essentially presented in telemetry to inform product decisions. If you had organizations sharing some open data around this, it would be such a help for understanding what the things are. What the most effective interventions would be for impacting website for carbon figures but right now we don't have that yet, but it's the thing we could hope for. And who knows, there's a year now for it. So fingers crossed, eh?

Fershad Irani: HotCarbon 2024.

Chris Adams: Yes. All right. Should we move to the next story fish?

Fershad Irani: Let's go.

Chris Adams: This one is a story from a character, someone called Assaad Razzouk, who is, I think he's actually based in Singapore.

And he's one person who runs a podcast called the Angry Clean Energy Guy, but he's actually has a background working in this field. I basically wanted to share this cause I found this really interesting specifically because when you speak to people who are thought leaders in the kind of world of cloud and sustainability in cloud, one of the recommendations that you'll hear people say is, please don't run things in Southeast Asia right now.

Because the energy is really dirty and it's really hard to do that and because it's so hot, it also means that even the computing that you do run, there's going to be a massive amount spent to keep the computers from glowing red and overheating rather than actually doing your computing. And this is the first time I've seen where someone saying, no, there's actually some changes taking place there's been massive investments, particularly from Singapore in some of the surrounding areas, to make some changes to this. So while we've seen the energy transition move quite quickly in China and to an extent, Europe and America specifically with the IEA, you're now seeing some signs of this in Southeast Asia as well, which hopefully means that computing will be getting greener over time.

And Fish, I know that you initially came from Australia. So I figured I'd share this link here from Grok Ventures and Quinbrook, basically the story about connecting Australia to Singapore to provide a punch of clean energy through this actually.

Fershad Irani: Yeah, I'll, I'll be a bit cynical, as any good Aussie should, and um, just say, this is something that I've heard mumblings of doing something like this for, I think, over 10 years? Since, yeah, before I moved here to Taiwan, and for the last 25, 30 years people have been talking about high speed rail along the east coast of Australia, and that's still not there.

This is a really cool idea, and something definitely that, when you look at a place like Singapore,

Chris Adams: Hmm.

Fershad Irani: It's small, they've got land constraints, they can't just suddenly put up a whole bunch of solar, they can't really put up a whole bunch of wind because it's a major shipping channel and a lot of planes come through there as well.

They need to be looking outside to import energy, and they've got Indonesia, Malaysia pretty close by. It's good to see that Singapore is doing some investment outside of their own borders in clean energy. As someone who lives there, Asia, has got a way to go in terms of being green. But the potential is there. We sit on this thing called the Ring of Fire, and it's an active geothermal hotbed. I've got hot springs 20 minutes by car from my place. There's potential there for, for geothermal beyond just using solar and, and wind. So Asia does have that possibility of being a, a green hub for digital sometime in the future.

Chris Adams: Do you know what I actually totally forgot about the whole Pacific Rim Ring of Fire stuff, because there was an announcement, I think two weeks ago or last week from Microsoft, them basically breaking ground on a massive geothermal project for some of the data centers in New Zealand, specifically for this.

So yeah, that's actually a useful, interesting perspective. I didn't think about that actually.

Fershad Irani: Let's move on to the next story, which is from the Green Software Foundation and one of the brainchilds of Adrian Cockcroft. It's about introducing a specification for real time carbon intensity. Chris, I think you'll be able to speak a bit more to this, but from my understanding, what this is all about is aiming to set a common way for data centers to report on energy and emissions, preferably in real time.

And I think that's something that would be useful for a tool like Cloud Carbon Footprint, wouldn't it?

Chris Adams: Yeah, first of all, it's really cool to see this proposal go ahead because essentially one of the struggles you have is even if you're using, say, Microsoft, Amazon and Google, you're running, you're trying to run the same computing load between these three, it's almost impossible to have any kind of meaningful comparison between these things because they all measure carbon in slightly different ways and include different things, whereas other ones don't now, what It's basically been proposed here is there's actually two things.

So first of all, there are different ways of measuring. And also the figures that you see are not particularly actionable a lot of the time. So the resolute, the information you will usually come a few months later rather than in real time or anything, or even the same half hour, basically. Now what's been proposed here is essentially a way to talk about minute by minute metrics that a cloud provider would make available so that you can actually make informed decisions about when and where, or what kind of computing jobs you choose to schedule, or even which providers you're going to choose to use compared to other ones. Now I've read through the proposal and it's really well thought through and one of the reasons that people have said that they can't share this information before is that cloud providers basically will usually will say. We can't share this data because there's a security issue related to this. And Fish do you remember when we did some work with Firefox, we had something like this because one thing we learned when we were trying to get some high resolution figures for the browser, one of the solutions was we could get these figures, but you would need to run Firefox as root, which might not be a good idea for people to be doing that.

And essentially what the thrust of this points to is that if you keep the resolution at minute by minute level. Then you're no longer disclosing any kind of dangerous information that might help an attacker, but it also provides sufficient resolution for you to make much more informed scheduling decisions as an operator.

But also you actually get some consistent ways to make comparisons between different providers of these services. So this is my view is something that is really overdue and to see someone who's actually fleshed it out quite well, and actually thought about lots of the issues and how this relates to some of the weird aspects of how people count energy is green with certificates and so on.

There's really good news and it's also. Interesting to see that you've got groups like the cloud native computing foundation getting involved or have it expressing interest as well. I think this is long overdue and you're right. Tools like cloud carbon footprint could presumably could in theory consume this kind of information if it was exposed by the providers, because right now they have to use models and guesswork based on the billing data, which is much less useful than getting direct figures.

It also means that any other cloud provider who, which is not the big three could also share this information. So you could finally have some meaningful ways to make meaningful comparisons between them.

Fershad Irani: And that's something I didn't think about when I first read it, but it actually really good like pardon the pun, but this turns the heat up on those or has the potential to turn the heat up on those big cloud providers and gives people a chance to, like you say, compare them on their carbon footprint.

They might even need to start competing on carbon footprint because that's going to be important in the future.

Chris Adams: This is exactly it. This makes some of this possible. And it also means that new entrants can actually start sharing these numbers. So you could compete on transparency to provide these numbers as a way to help customers make the responsible decisions that are currently really difficult to do. Or you could even plausibly build this into some of the tooling so that it's just part of how Kubernetes works or part of how maybe even Docker might work for example. This is actually, in my view, really exciting. And I'm really curious to see where it goes next, actually. All right. I think we've spoken about that quite a lot.

Should we look at the next one. This is the IEA. So the IEA, Fish, I'll let you speak a bit about this one here. Cause this is the International Energy Agency.

They've updated their data set, their, their information about data centers for 2023, this, this is the resource that is almost always cited as the authoritative figures on what the environmental impact of the tech sector is or how much energy it uses. And if you want to cite any numbers, these are peer reviewed and generally pretty reliable numbers you can refer to.

They're safe ones to use. And yeah, they're pretty eyeopening. Aren't they Fish?

Fershad Irani: Yeah. Firstly, it's good to see this data being updated. It's not so good to see some of the figures that are coming out of it. But like talking about data centers, the big three plus plus Meta. One of the things that struck me from this report was that from 2017 to 2021. So that encompasses some of the COVID years.

The report says that there was a doubling in the amount of energy consumed by those four providers. It also then goes on to say that it expects there to be moderate growth for the next few years. I really hope that their definition of quote unquote moderate isn't another doubling because then we're going to be in serious trouble on the data center front because that's a lot of energy to be consuming.

I think in the report it says somewhere around 1. 3% of total global like energy use or something. And that's without including cryptocurrencies, which is a whole other ball game. I think they've steered clear of it in this report.

Chris Adams: Yeah. As I understand it was broken out separately because it's generally considered not part of the existing economy for this part. And also we're not going to talk about cryptocurrencies on this because the less said about them, the better. But generally speaking, this is one of the first times you've seen these figures broken out like that, because typically what you've had people talking about is the actual energy usage staying more or less about level for the last, say, 10 years or so, but what this really highlights is that this has stayed level because we've had a massive concentration of usage to a very small number of providers, as opposed to having a large number of maybe less efficient providers.

There are some good signs of that in terms of in absolute terms, the figures are not growing as much as they could be, but it also means that we've got this massive concentration of, we've got all this consolidation, which has other impacts in terms of, okay, how easy is it to then pass all kinds of policy as a result for this, to move things away from being level to going down rather than going up.

And this is the thing that we'll see coming forward, basically.

Fershad Irani: And I think on that thing, just like one thing that I can't possibly see it going down in the future is like just the amount of volume, the amount of internet traffic that is there. There's a number in that report for 4.4 zettabytes of internet traffic in 2022, which is, I don't even know what that number is, man.

Like it's just mind bogglingly big.

Chris Adams: a zettabyte, right? I'm just, if I can find the figures for that, it's. Good Lord. So there's 21 zeros behind it. Yeah. If a million is like three, six, that's seven. So yeah, that's 20. That's a. A very large number. That's an incomprehensibly large number, but

Fershad Irani: That's mental.

Chris Adams: yeah, that's one of the issues that we struggle with.

Okay. So this at least gives you an idea of where the most recent current data is that you might refer to. Okay.

There's maybe one more story. Then we'll look at what else is going on in terms of jobs and things going out there. Fish, this is one I just want to point people to, because I've seen quite an interest in the Cloud Native Computing Foundation.

There's a new thing called the Green Reviews Project, which has come up. And, uh, I'll just read the kind of blurb on this because this, in my view, looks cool. Basically, the Green Reviews Working Group helps CNCF projects assess and improve the cloud native sustainability footprint. So the idea of this, as I understand, is to start integrating sustainability reviews into how projects are maintained and run so that you get an idea of just bringing up the floor of competency on projects.

So people have some way to talk about this and think about it. And essentially consider these as requirements in the same way that you might look at other things as requirements. This is interesting in my view. I was quite excited to see something like this. And there's a couple of links of what this looks like in practice with, I believe, the Falco project.

And a couple of other ones there. So yeah, interesting to actually see something like this happening. This looks like it's going to be merged in the next week or two and a working group, the kind of technical architecture working group for this. And yeah, I was quite excited to see this actually land.

Fershad Irani: And that's a really good way of making sustainability or sustainability considerations a regular part of a process and a way of doing things. That's rather than it being its own separate silo that might get looked at, might not get looked at. If it is part of the regular process that everyone has to go through, you're going to see more traction, more movement in the right direction, which is good to hear.

Chris Adams: Yeah, I think it'd be really useful to actually have a chat with some of the CNCF folks on this, maybe they can come on the podcast and talk about a, how this happened and what this looks like, because we are now seeing various open source projects or groups starting to essentially start, create their own groups for this.

So WordPress has one, Wagtail has one. This is one, which is, seems to be across some, a number of all the projects in the CNCF, the Cloud Native Computing Foundation. So there's a bunch of stuff going on there. So yeah, this, in my view, this is quite exciting, actually.

All So the final thing, this is a little bit like we shared last week. If you are looking for work, the Green Software Foundation is actually hiring for a technical project manager and a content project manager. So these are funded positions that are available. You look at and it's, and you can apply with links that they have there.

So there's, that's what's going on there. Okay, so we're just coming up to the hour for this show, and, uh, we normally have a kind of easy question to round this off. Now, Fish, I know that you've been doing a bit of travel away from Taipei and you've just come back, so I figured I'd ask, what's the first place you, you try to go to, to get some food you can't get anywhere else or as good as anywhere else when you are back in Taipei?

What's your first place you're thinking of

Fershad Irani: We got dumplings. We had dumplings the first time, first night we got back, which is quintessential Taiwan. I adid find myself that when I was on the road, I was traveling through Australia, mostly where I grew up. And I did find myself craving instant noodles, which is a bit weird, but there's just a dearth of choice.

There's hardly any choice in Australia for instant noodles. And then you come back here and you've got mind blowing

Chris Adams: cornucopia of ramen in packets?

Fershad Irani: Oh yeah. Yeah, so it was dumplings first and instant noodles a very close second.

Chris Adams: I was not expecting that second answer. I'll be I'll be real. Okay for me when I come back to Berlin It's all about falafel for me There's a really good place called Lausanne when you come back to Kreuzberg and it's probably the best falafel in at least five square kilometers if you're going in anywhere near Kreuzberg. So that's all for this episode.

All the resources and links will be shared in this podcast episode, and you can visit podcast. greensoftwarefoundation to look at some of the previous episodes that we've actually referred to a few times. And finally, Fish, thanks for coming on. Really, I really enjoyed hanging out and chatting with you again.

So everyone else, see you on the episode and Fish, bye for now, I suppose.

Fershad Irani: See you folks.

Chris Adams: Cheers, Fish. Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing.

It helps other people discover the show. And of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware. foundation. That's greensoftware. foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: Data Center LCA with Stani Borisová

12 juli 2023 | 43 min

This episode of The Week in Green Software, features guest Stani Borisová; Expert in Life Cycle Management at IVL and former researcher at RISE Sweden. Host Chris quizzes her on her expertise in data center LCA and they discuss interesting news tidbits to share from a global tour of Singapore, Norway, Germany, and America. They discuss how data centers might be unnecessarily using too much heat to cool themselves down, how Germany’s Energy Efficiency Act has perhaps not gone far enough and how Norway’s investment into oil and gas affects renewable energy resources for data centers. Finally we have some fantastic opportunities for you to be part of the Green Software Foundation!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Stanislava Borisová: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

"Continue AI development – for the sake of the environment" | Journal of Extracts (extrakt.se) [9:49]
IMDA Launches Sustainability Standard for Tropical Climate Data Centres | Data Center Storage ASEAN [15:14]
Germany to pass Energy Efficiency Act, demanding heat reuse in data centers | Data Centre Dynamics [19:01]
Government Norway approves 15 Billion Euro Investment in Oil and Gas Industry | Datacenter Forum [23:02]
Fairphone 4—the repairable, sustainable smartphone—is coming to the US | Ars Technica [28:11]

Announcements:

The Green Software Foundation is Hiring [39:31]

Resources:

Open-source solutions are essential to greening software and ICT | SOGS Report [5:39]
Carbon Aware SDK | Green Software Foundation [7:27]
Digital does not equal green | SOGS Report [7:45]
Decarbonization alone cannot make software green | SOGS Report [8:47]
Climate Policy Radar [13:48]
Global Stocktake Explorer [14:35]
The Week in Green Software: Code Green and Clean Power | Nina Jabłońska | Environment Variables Podcast [22:40]
Green Rocks Newsletter [27:27]
Mining.com [27:27]
Techbuyer [36:22]
Optimizing Server Refresh Cycles: The Case for Circular Economy With an Aging Moore's Law | Rabih Bashroush, Nour Rteil, Rich Kenny, Astrid Wynne [37:12]
Regenerate! Board Game [40:16]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Stani Borisová: We talk a lot about CO2 and climate change, and it's very important, but at the same time, there are other aspects of the world, such as, let's say, the water ecosystems or the depletion of minerals and such, and they happen at the same time as climate change happens, so they're inseparable.

Chris Adams: Hello and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field, who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. In today's episode, we have some interesting news tidbits from a global tour of Singapore, Norway, Germany, and America.

Then finally, we have some fantastic opportunities available if you fancy working with the Green Software Foundation, because yes, they're hiring. But before we dive into this, let me introduce my guest today. Today we have Stani Borisová from IVL, Swedish Environmental Research Institute. Stani, I'm gonna let you introduce yourself here.

So yeah, the floor is yours and maybe that's if I've mispronounced your name. Please do tell, please do help correct me because I am not sure if I've got it correct.

Stani Borisová: Hi, Chris. Thanks for having me here today. So you've pronounced my name very nicely and properly. Uh, and I work as a consultant in Swedish Environmental Institute, as you've mentioned, specifically in the area of LCA or life cycle assessment. I also do life cycle assessment of data centers. And recently I've also been involved in the development of PCR.

Uh, which stands for, um, product category rules, or basically a template for a standardized life cycle assessment of electronics. And among other projects, also a heat reuse about mealworm farming from data center excess heat, or an EU project on waste reuse in process industry. It's great to be here.

Chris Adams: Cool, thank you. So before we dive into the meat of the show, if you, if this is the first time you ever listen to this podcast, we will share all the links that we, for every article that we discuss and anything else that comes up that's interesting, we'll do our best to share the links to this as well.

Okay, so, Stani, you said a couple of things that caught my interest here. I didn't know there was any link whatsoever between mealworm farming and data centers, and I wanted to dive into that a little bit first, actually. Could you maybe expand on what was going on there and what the mealworms were for?

Because, yeah, this is the first I've ever heard of it.

Stani Borisová: Yeah, of course. This was a project I worked on before at RISE Research Institutes of Sweden, where we were focusing on sustainability of data centers and looking into different industrial symbiosis opportunities based on this excess heat. And one of them was a very nice, very circular project where we got in touch with the local brewery and used their spent brewers to actually grow mealworms.

And the idea there was to actually feed them to the chicken in a village nearby and basically farm chicken and close the loop and reuse both, both the heat, but also the spent brewers and look into how this excess heat from data center, whether it's. enhances or increases the rate of growth of these mealworms or not.

And we're actually just finishing up publishing this article about it, so hopefully you'll be able to read about it soon.

Chris Adams: So that's when I can find out if mealworms like the excess heat, if it's good for them, and they enjoy it, or they don't like it very much and they don't grow quite as fast,

Stani Borisová: Yes, exactly. But spoiler alert, they do like the heat.

Chris Adams: Great. That's interesting, especially in the context of some of the news articles we'll be referring to somewhat later on, actually. Okay. And as I understand it, I think you've done a bit of work with both the OCP, the Open Compute Project, and also with the SDIA, some of those working groups, the Sustainable Digital Infrastructure Alliance.

You've had some involvement with those over the last couple of years as well, right?

Stani Borisová: Yeah, that's correct. To be honest, it's been a, it's been a few months since I've been involved, but previously I have worked in, in these groups, mostly on a life cycle assessment in data centers and especially advocating for data transparency, because I would say that nowadays the biggest problem with data center sustainability is that we know too little to actually know how sustainable or not they are.

And the reason for that is that there's just not enough data from these very long value chains. And otherwise I've been trying to get people to look into broader sustainability, not just CO2 emissions, but looking into other impacts that one has on the world, on the environment, because of course, we talk a lot about CO2 and climate change, and it's very important, but at the same time, there are other aspects of the world, such as, let's say, the water ecosystems or the depletion of minerals and such, and they happen at the same time as climate change happens, so they're inseparable.

Chris Adams: Okay, so that's a nice link to the State of Green Software report that was published by the Green Software Foundation. I didn't introduce myself before properly. My name is Chris. I work as the policy working group, one of the chairs in that working group. And one thing that we did over the last year was get a report commissioned to make it easier for people to start basically getting into this field and understanding that yes, there is an environmental impact associated with software from this.

And one thing we've been doing with various guests has been sharing this report ahead of time and asking them which of the insights, which of the 30 insights the report is comprised of, seemed of interest to them that they might want to talk a bit about. And Stani, you've identified three of these here, open source solutions, essential to greening software.

Digital doesn't equal green and decarbonization alone cannot make software green. Is there any of those that you might want to start on just to touch, just to see what, just to share why you found them interesting so other people who are coming to this might get an idea of what kind of content there might be in here.

Stani Borisová: Sure. I think maybe we can start with the first one. The open source solutions are essential to greening of software and ICT. And the reason for that, at least in my opinion, is that you need to be able to understand what you're dealing with in order to evaluate it and then identify the hotspots and then improve whatever the issue seems to be.

So without the proper understanding, Which can only be achieved by seeing and getting the data and information. You can't really improve anything. You can be hoping and blindly aiming at something, but more is needed.

Chris Adams: Okay, I think I could go with you on that as well. With my green Software Foundation hat on, there's a bunch of open source tools specifically for this. Uh, so there's things like the Carbon Aware SDK, a software development kit specifically designed to make some of this available. But also with another hat on, the non profit I work for pretty much everything we do is open source specifically for this reason, because you need to be clear about some of the assumptions you're making if you want people to trust in the stuff you're doing. So I very much agree with you on that. And this second one, digital does not equal green. Maybe this is worth talking about from your perspective here as a life cycle assessment specialist.

So maybe if we could just briefly touch on that one as well.

Stani Borisová: Yeah, actually, I would say that there has been a lot of movement for switching to digital, especially during the pandemic, because people suddenly started talking about the benefits of not having to commute and not having to potentially rent office buildings and so forth. And just assuming that switching to digital would always be better, but that's not necessarily true to a certain extent, digitalization can make things much better, but of course we need to understand what it means to become digital and especially with certain things such as cloud services, I think lots of people don't understand that those aren't really located somewhere in the sky, they are actually somewhere on the ground in an actual physical location and that also has impact.

Chris Adams: And that is probably a nice segue to the next one. So decarbonization alone not making, cannot make software green as an LCA specialist. This is one thing that we do while we look at carbon a lot. You've just mentioned both water and depletion of natural resources. I'd like to get your take on this one because there's a couple of other stories where we speak about that in a bit more detail actually.

Stani Borisová: Decarbonization, it's an interesting name, but it very much depends what you imagine as decarbonization. You could say that You have your software, which produces emissions, for instance, CO2 emissions, and then you just purchase some credits and offset these emissions, which could make you, for instance, net zero or carbon neutral, but it would not make you green.

Green would assume that you're doing something good for the environment, and that is not just then the climate change aspect. The environment, as I mentioned before, are so many more topics. So. Yeah, I think we need to look at things with a broader perspective.

Chris Adams: More than just carbon, then. Okay, and should we jump into the news now? Alright, so the first story that you shared was actually one that was initially in Swedish, I believe. And I'll be honest, my Swedish isn't brilliant, but I was able to just pass it through some automatic translators to get the general gist of this.

And the key thing that I found interesting was... This kind of quote that came from, and I'm not sure if as someone who can speak Swedish, maybe you can tell me if I've got it more or less, but essentially the kind of thrust of this article that you shared the link to was basically saying, AI experts and business leaders were seeing people talk from like open AI and organizations saying, we've got to stop doing any development on, on AI.

And we're going to have to like, make sure we do, we bring in regulation, but. For, for way in the future, nothing to do with what we have right now, all right? There's this idea of pausing development to keep things as they are. Now, there's a researcher at KTH, which I, as I understand, it's an institution in Sweden with a decent, with quite well known for actually pioneering work in sustainability and digital, digital for the last 10 years at least.

They're basically saying, no, we don't need to stop doing this. We should actually be using this because there's a bunch of. Places where this actually is very helpful and there's a number of specific use cases for this. I might ask you to maybe help provide a bit more on this because I realized I haven't actually described what KTH is and I forgot what the K in the TH is.

So maybe you can help me there and then we can talk about some of this.

Stani Borisová: So the KTH in English is basically Royal Institute of Technology. It's a university located in Stockholm. K stands for Kunglig, I think, which means the royal. And yes, it's a very surprising news. I myself was pretty surprised to find this in my newsletter because just talking with colleagues during coffee breaks, you hear lots of concern about privacy and where all of this could happen.

We've been testing chat GPT at work and playing, seeing what we can do with it. And it's very impressive, but also very scary. So I understand this. Impulse to want to stop things, to want to put it on ice and take a break. So I've been reading a book recently called the best of times, the worst of times futures from the frontiers of climate change.

And some of the issues that it discusses are actually these models that we as humans make, for instance, for climate change. And lots of these models are based on our economic models, even though these are very different issues, the environment. That we're dealing with rather than economic issues that we are so used to tackling.

And one of the the biggest problems that this author identifies is that in the economic models one tends to discount time. So basically time is money or what happens later in the future is worth more than what happens now but that's basically the exact opposite for, for the environmental problems, because we need to tackle them now, since in the future they will become much more serious.

And that is one of the biggest pitfalls of the current models that humans are developing, especially if they're taking some simplistic way of modeling based on economics. So, I think that what these researchers from KTH are suggesting, to basically let AI take... A look from all these different viewpoints and design something better could really help us tackle something that we alone cannot because we're very, we have our subjective opinions and we see things through our own filters and potentially AI could get some more objective view.

But of course, one could also polemize how objective AI can be when it's very much modeled based on our own opinions and what we feed it.

Chris Adams: Thank you for that. So following on from this one thing that may be of interest because when I was reading through some of this initially there was this idea that yes there are all these use cases where it does make sense to put it in the hands of a wider set of people. One example I think it's actually worth people paying attention to is Climate Policy Radar.

It was a nonprofit that was launched maybe a couple of years ago. What they've been doing is they've been basically taking all of the existing climate policy all around the world to put into a single model, to see what some of the kind of features of policy that gets passed in one place has, or what some of the kind of good practices might be regardless of the language.

So with the idea being that when you're at events, say COP 27, 28, or something like that, people are able to essentially compare some of the policies that have worked in certain places and see where the actual practices are able to be used regard, without having to actually speak that particular language, because we're aware that there are absolutely language barriers here. And these are some of the tools that people have been using.

There's also in this year 2023, there's a recent thing that's come out from the same organization who've been doing some work with something called, I think it's the, the stock, the Global Stocktake Explorer, which again is taking this information, which has thousands of pages of different policy of essential all the countries have been doing.

to meet their own kind of climate goals and then put it into a tool which makes it possible to skim through this and actually pull out some of this information because expecting any single person to read thousands upon thousands of pages is a bit of a tall order, right? This is more like an example of this stuff.

Stani Borisová: Exactly.

Chris Adams: Okay, all right, that sounds pretty cool. We'll add some links to that. That's a nice link for the next part. So this is a story from, I think, the Infocom Media Development Authority. So basically this is a story about green software in Singapore, actually. Singapore have started to create some actual standards for greener data centers in their regions.

Now, this is interesting in my view, because for the longest time, you may have seen say, countries like say Sweden, where you're in, or to an extent, Germany as well, or some parts of say Northern Europe, or even parts of North America, to have a relatively clean grid. Which means that running infrastructure is actually, there are steps you can take.

But when it comes to Southeast Asia, it's been actually quite harder and there's been a lot of actual advice basically saying, if you can avoid running infrastructure in Southeast Asia, it's probably worthwhile doing because the grid is so kind of fossil fuel based. It's going to have a greater environmental impact running work over there than other places.

This is problematic because there are lots and lots of people in Southeast Asia who need to use this stuff, who need to use services, and why shouldn't they be able to access this too? And this is actually something related to the way that some people are realizing that you can actually run data centers at different temperatures to actually change the kind of cooling you might actually need.

Maybe you could come in on this one actually, Stani, because I have never had to run a data center myself, and this sounds interesting in my book.

Stani Borisová: Yeah, I think it's also very interesting. So the idea here is that most data centers cool their servers quite a lot to let's say temperature of 22 degrees. And one of the biggest reasons for that is the so called ASHRAE envelope. And the ASHRAE envelope is a combination of, I think, five different factors, which you have to adhere to in order to keep the warranty of your servers.

And some of those factors are temperature and humidity and so forth. And what's interesting is that quite a lot of people I would say in the data center world who understand that some of these conditions are very strict. And potentially stricter than they have to be, but at the same time, since all of the manufacturers based their warranty policy on the ASHRAE envelope, no one dares to get out of the envelope and operate differently.

So that means that people are unnecessarily cooling their data centers too much, data centers that don't have to be as cold, that could operate just as fine at, let's say, 2, 3, 4 degrees higher temperature. So what's happening in Singapore is that they decided to support the gradual increase in operating temperature to 26 degrees.

I think that's fantastic. I'm hoping that this will inspire even other parts of the world where maybe the countries are not so hot. The climate is maybe colder than the Southeast Asia, but still it could be applied throughout the world. And thus they're aiming to potentially save cooling energy up to 2 or 5% for every one degree increase.

So yeah, I think it's very exciting news.

Chris Adams: Okay. Cool. So, I'm really glad you mentioned the ASHRAE envelope in some of this because I honestly thought when the figures for running datacenters unnecessarily, particularly cold, might just been a human comfort thing rather than a kind of warranty thing or something equally arbitrary if basically this is not actually tied to the actual performance of the servers themselves.

I think if you're able to increase the temperature by two or three degrees, and if it's between two and five, that's like 15, 20% savings, which is nothing to be sneezed at given that this is actually a significant draw of both water when it's used to cool it down and also energy to actually cool things down as well, as I'm understanding.

Stani Borisová: Yeah.

Chris Adams: All right, following on from this, now from Singapore to Germany, where I'm based. So this is another one related to heat reuse as well, actually. This is a story from Data Center Dynamics. Germany surpassed the Energy Efficiency Act, demanding heat reuse in data centers. I wanted to share this with you and get some of your take on it because there was heat reuse, which we've now learned is good for mealworms.

But also it's one of the greatest, providing space heating is one of the significant drivers of emissions in Germany. In fact, I think it's one of the largest ones. It looks like it's going to be landing in September. And, uh, there's a few things which caught my eye. First of all, there was this kind of mandatory kind of tightening of efficiency requirements, so that data centers have to be more efficient, uh, and basically use more of their power to actually run service rather, rather than actually just be trying to cool things down through better, better design.

But there was also... A few interesting things about a shift to renewable energy. So the idea is that this act would require for the data centers to basically be using 50% renewable energy by 2024, which is pretty soon. And then 100% by 2027. That's impressive, actually. But there's a couple of caveats that I figured might be worth discussing.

So yeah, I'd like to hear your take on this one here as well, actually, because it sounds like it's good for the mealworms, definitely. But there's also some other climate implications for some of this.

Stani Borisová: Yeah, I think it's an interesting article and I think with these kind of things the devil lies in the details and it sounds very good, I have to say, 50% renewable now. If we assume that some of the data centers in Germany don't use any renewable energy is a great increase. already next year. But as you said, indeed, this can be met using certificates.

And additionally, I wonder what's going to happen with renewables as we've seen recently in the EU taxonomy, natural gas being classified as a renewable. That makes me very concerned because basically you could just operate a hundred percent on natural gas and claim that's renewable energy doing us all a very big disfavor.

And at the same time, I think This proposed law had a very big potential from my point of view. The idea was to reuse, I think, 40% of the excess heat. But there was so much effort to just keep it easy, start slowly, take our time, that as far as I understood it, next year it's going to only require that 10% of all excess heat is reused, which unfortunately is very little.

And if I remember correctly, I've spoken with a previous colleague of mine about this and how we use energy to compute, but of the energy that we use, maybe 99. 9% does not end up being in the compute power, but ends up being heat. So that just tells you. The enormous amount of heat being generated. And it's a real shame to then only reuse such a small proportion.

And from what I've seen and heard, lots of data centers are trying to advocate for their placement in the society as potential sources of heat. Obviously we need data centers. They're crucial and we have systems based on their existence. So we can't get rid of them. But at the same time, I feel like the skeptics or the conservatives have won this battle a little bit and made what could have been a really revolutionary step, something very mild instead.

Chris Adams: Okay, I really appreciate you giving that extra perspective on this because I did mention there was some of these caveats and for people who've listened to this podcast a couple of episodes ago we had Nina Jablonska from Energy Tag talking specifically about some of the issues related to using credits from other parts of the world to mark energy as green like we said here.

Technically this would mean that energy in Norway should No longer be counted as green. But whether that always happens is another matter. But let's just move on to the next story, which I believe you shared here. I was surprised to see this mentioned from a data center publication, actually. So basically, this is the Norway government approving significant investment in the oil and gas industry.

Now, I wanted to ask, do you know why this might have showed up in a data center? Publication for this, because it feels like it's an energy story rather than a data center story. And I was a bit lost on this one. So maybe you might be able to shed some light and then we can talk about some of the other things that have been going on in Norway that are also interesting in this kind of transition technology kind of field.

Stani Borisová: Sure. Yeah. I was also slightly surprised to see it on, on the data center forum, but the two main reasons I think for this is first of all, data center world, I would say is very energy focused whenever it comes to just talking about sustainability. I would say 99% of the focus is energy and at the same time, what I just mentioned with green natural gas being classified as renewable could suggest that there are some players who then will use natural gas and its abundance and its increased abundance from Norwegian supplies to claim that they are operating on renewable power.

Chris Adams: Okay, thank you for sharing that because we touched on this in a previous episode about okay, when you have significant drivers of demand, like when you're building full of servers full of GPU cards, which are extremely energy dense, it may be that the power draw you're needing is actually greater than the grid itself is able to provide, just like we have problems with the transmission of networks, there's also transmission issues related to energy.

There's also another kind of investment or a real kind of change that was actually announced in Norway in the last week or two, this massive deposit of phosphate, one of the transition materials that was, it's used for batteries and it's used for fertilizer.

And this felt like a kind of in my view, really interesting, because you do see things like oil and gas being a big thing in Norway. While Norway itself uses a very clean grid, it's one of the key places that people point to when they look at the migration to electric cars and things like that. Most of the oil and gas is exported into a massive sovereign wealth fund, which is used to basically, in many ways, provide some kind of base to set things up in the future.

Now, this discovery of at least 70 billion tons of phosphate. First of all, this is larger than any other deposit that's ever been discovered of phosphate, which is interesting, which is important for farming, but also transitions, but also it suggests that this is another route away from relying on oil and gas for a sovereign wealth fund.

And I wanted to get your perspective on some of this actually, because we do see things like data centers taking on much more batteries and things like this as another way to provide the necessary kind of way to meet the demand for power.

Stani Borisová: I thought it was an interesting piece of news. It was very exciting. Indeed, as you mentioned, most of the phosphate rock is used in fertilizers, but there is a proportion being used to produce batteries. So I think this maybe could also help the opposite direction, not just the oil and gas, but instead, as you mentioned, the renewables, maybe some on site power generation with potentially European cheaper batteries.

So I find this to be a very optimistic news. What is also important to keep in mind is the mining and the emissions related to that and the health and environmental concerns from that. But it seems like the Noria mining, they keep that in mind. So hopefully they are planning to do the apply carbon capture and sequestering as I, if I remember correctly.

And then another maybe important aspect is to make sure that this phosphate is being recirculated and that it's not being emitted somewhere into water because that can actually cause a lot of problems with eutrophication, which basically means this phosphate, which basically is very nice as a fertilizer, ends up in water.

Which sounds nice because then plants have more food to eat, but what it does if it happens in two big quantities is that you suddenly get a lot of biomass that is growing and expanding, and as it decomposes, it prevents the ecosystems and the fish from getting their oxygen, so then they die. So there are all these different aspects to look into, but I would say that generally it's exciting and hopefully we'll have some better and more accessible batteries for renewables.

Chris Adams: And ideally, no longer needing to export quite so much oil if you have, if there's another massive natural resource that people might be using instead. So I think you just spoke about, I think, is it an algal bloom? The algae growing, using up all the oxygen, then suffocating all the fish. That's the, essentially the runoff effect of phosphate fertilizer there.

That's the other kind of flip side of this.

Stani Borisová: Exactly. Yeah.

Chris Adams: Okay, we'll share a link to a really good newsletter called Green Rocks, which is specifically about the environmental aspects of mining. And we'll also share a link to mining.com, which also provides another kind of industry view on what's happening here. All right, we're just coming up to the last story now, actually.

So this one, I shared this because... This really caught my eye, and because you have a focus on lifecycle, I figured you'd probably have some opinions about this one here. So this is a story in Ars Technica about the Fairphone coming to America. And I understand that you're familiar with the Fairphone, so maybe it might be useful for you to explain this to the uninitiated who may not have purchased or had to own a Fairphone themselves.

Stani Borisová: Sure. So Fairphone is a Dutch company that is selling modular phones. And I would say that they're doing their best to really look into all of the supply chains and reuse as much material as possible. As a coincidence, one of my very good friends works at Fairphone. So I do have a lot of information from them and they are really in touch with the suppliers, even traveling to the local places.

Here in Europe we might think that you bring your electronics to the dedicated separation place and therefore everything is fine and everything is solved and you did your part. But in reality, oftentimes these electronics and dangerous parts of them end up somewhere on the streets in Africa where people don't understand which parts are dangerous, which are not, and children are playing there and such.

So it's very good that someone has this in mind. And another great thing is that it's a modular phone. So if anything breaks, you can just. Get another piece and keep on using your phone as long as possible. So I think it's a very good news that a Fairphone is coming to the U. S.

Chris Adams: So this is one thing that I wanted to ask you about because I owned a version, one of the original Fairphones, which I'll be honest, I loved the idea. The idea of using it was probably better than the experience of the first iteration of the product. And I got the second one as well, which was nice enough to use.

And I really did appreciate the modular aspect because I actually was able to eek out the life by literally just upgrading a camera. It cost me 40 euros to swap out. One of the old cameras for new cameras, and the rest of it was still more or less working, actually. And I know that Fairphone was initially set up largely almost like to prove a point, rather than actually be an enterprise initially to set up to make a bunch of money.

For example, as I understood it, Fairphone initially came out of organizations who were essentially campaigning for labor rights in electronics and basically said, Look, this needs to be changed. They were so sick of people pushing back saying we can't possibly treat people fairly where they've decided to make a phone themselves just to show that it could actually be done.

And I wanted to ask you, have you seen any examples of this being adopted in industry or some of these practices filtering down? Because in my view, at least early on in the industry, before they started growing. It was very much like a kind of demonstrator of a company compared to other things. Just showing that this could be done and to raise some of the bar.

And I know that we have some laws coming forward, which seem to be informed by some of what's happening here. But again, I don't get to speak to a life cycle and that assessment specialist all that often. So I figured I'd ask you, have you seen any of this percolating down into other fields or other examples in the industry?

Stani Borisová: Maybe not in as much in depth, I would say, but there are lots of different initiatives where companies have to look into their entire supply chain and look also into the social aspects of their business. There are lots of different standards on how it's done. And then there is something called social life cycle assessment, which is a sister of This traditional environmental life cycle assessment, where one really looks into all the different shareholders and stakeholders involved, and even interviews the communities and sees how things are done to once again, point out the hotspots and show what can be improved and how, and potentially even compare that.

So I would say that there are some initiatives. Oftentimes it's because of the regulation rather than. because of some extra incentive internally from the company. And hopefully we'll see more of that. What we also see, for instance, the social LCA, that's mostly used in different EU projects. So that's where it's being used quite often.

At the moment, I have two colleagues in Portugal where they're part of a social LCA. And they're interviewing local communities about cement production and how that impacts them, but also talking with the workers on site and, and such. And of course, the society should be considered as well when it comes to sustainability.

So hopefully we'll see more and more of that.

Chris Adams: So not just carbon. So this is one thing I was going to ask you actually, because, so you mentioned before you were working at RISE, which is a research institute, and now, rather than just doing work for the kind of the state as it were, you're working in a company which is essentially helping organizations like maybe corporates understand some of their own responsibilities or what the impact of their products actually have. Is it different when say a researcher asks for data versus a company asking for data in their supply chain? Maybe you could share some of that because we've spoken before about how data is really a real problem and I figured I should ask you about some of this as well.

Stani Borisová: Yeah, I think that's a very good question. So indeed, I was working as a researcher before. Now I work as a consultant. The biggest difference I see, for instance, two years ago, I was trying to do a life cycle assessment of a data center in Buden in the north of Sweden, and the project was already at its end and we were asking from the former suppliers, which we had collaborated before with, but it was just not interesting. And at the same time, it felt like we were a research Institute. We didn't really have any purchasing power to leverage, to get some data and potentially buy more in the future. And that's something that is very different now. I would say when I work with private companies and when they are doing their data collection. I would say that they have a much bigger access to data. Of course, you would want to start collecting your data as you're, let's say, building in the initial phase. As you're, as you start with your project, start also with data collection for future sustainability purposes, because of course, once you finished your data center and sold it to someone else, of course, for that new person, it's much harder to then contact people and convince them to send something, but still you are a company that is a potential consumer, potential future consumer that comes back and purchases more. And that really helps getting the data to the consumers. That of course, doesn't make the data open to everyone, but it allows for hopefully some future benchmarking where we know that the results that the companies are presenting to us are really based on the real data and very accurate data.

And then instead of comparing PUE, we can start comparing actual data center, let's say climate impact or I don't know, depletion of resources, impact and such and start to understand. What is good and what is not good?

Chris Adams: Okay, thank you for this. So, following on from this, I'd like to ask you about the role that software plays at the data center level for this, because we spoke a little bit about the Fairphone, and one thing that I found really interesting about the Fairphone was they have quite a long warranty compared to other places.

So they have like between five, some cases even seven years of basically a commitment to have it to support it and uh, we've seen essentially the lack of software support in many cases updates are inducing people to move away from what would otherwise be functioning hardware. Do you see any patterns like this in the data center when people are working with hardware in a data center kind of context?

Because my intuition would think yes, but I don't really know enough about it and once again it sounds like maybe this is actually a pattern we see more, uh, occurring in other places as well.

Stani Borisová: Yeah, absolutely. Actually, there's a British company called Techbuyer that basically works a lot with refurbishing old hardware. And they've also done some studies on comparing the performance of servers. And let's say taking two generations older servers compared to the new ones refurbished and such, and it can be done and the performance can be almost identical. And at the same time, this allows for immense emission savings in terms of CO2, in terms of everything else. I would say, even if you don't have enough power to influence your energy supply and where it comes from, just reusing the hardware for longer than you would intend to initially can do so much difference.

Chris Adams: Ah, okay, I think I know the paper you're talking about, and the nerd in me loves this paper. This was Optimizing Server Refresh Cycles, the Case for a Circular Economy with an Aging Moore's Law. This is by, I think, Rabih Bashroush, Nour Rteil, Rich Kenny, Astrid Wynne. This was a really cool paper, I thought, because there was a really eye catching stat that I remember, like, when I read through it.

Yeah. According to this study, From Eureka, which is a research institute. So they did some research of 300 data centers in Europe. And the thing they basically said was 40% of the deployed servers are around older than five years old. And they were consuming 66% of the facility energy, but providing only 7% of the compute capacity, right?

Okay. So just flip that around. That means that there's the other half is providing 90 plus percent for using less than half of it so if you wanted to reduce emissions swapping out those obviously would be the thing to do having a kind of cash for clunkers kind of thing would be an immediate climate gain but later on in the paper they do talk about this idea that yes if there would be over more than five years old yes but some of the newer ones They don't need to be that recent to still be quite effective, like you mentioned.

This is something that blew my mind when I saw it, to have both of these extremes in one paper, actually.

Stani Borisová: Yeah. And I think also the emissions from the energy itself are so dependent on the source of energy. So it could happen that you're located, let's say somewhere in Scandinavia and you're using very clean grid. And in such case you could obviously optimize the energy efficiency, but you could achieve a much bigger impact in that particular situation just by prolonging the service life of your hardware.

Chris Adams: cool. All right, we'll share that link to that paper for people. I would actually love to know if there's any more recent studies, because this paper that was published, I think last year, it was based on a study from from 2018, which had these crazy, this wild differentiation between older than five years is terrible.

Younger than three years, not so bad, right? And it'd be really interesting to see if that is the case or how the kind of fleet of infrastructure has and what kind of changes you would target if you wanted to reduce the environmental impact from using software for this. All right. Stani, I think we've covered most of the stories here.

Normally we do like a set of announcements and the thing I might share is that at the Green Software Foundation, they're currently hiring for a couple of roles. The first one is a kind of technical project manager role. And, uh, a content, uh, project manager role, both of these are available and we'll share some links to that.

So if people are interested in working in this field, they, that's an option. But I think the last thing I want to say before I thank you actually is this closing question from our producer, Chris, are there any particularly green games that you've played that have a real kind of focus or message that you would like to share or that you've enjoyed that people might know about?

Because we mentioned Doom and stuff before, but there are also other games which might not be quite so violent and might be a bit more, kind of, friendly to mealworms, if nothing else.

Stani Borisová: I do have a recommendation, but it's not a recommendation to, uh, Uh, video game, but rather, um, a board game, uh, it's called Regenerate, a cooperative resource management game, where one is trying to regenerate the environment. And what I really like about this game is that if you just go to their website, you can actually print everything out on your own at home, along with all the instructions, all the parts, and just play and enjoy the good feeling about making something good for the environment, at least in the game.

Chris Adams: So basically you just, you can just download the PDFs and print all the things that way, right? So it's a bit like a... Ah, open source applied to board games,

Stani Borisová: exactly. Yeah.

Chris Adams: That's pretty cool. All right. I kind of wanted to say something like SimCity, and I believe there's another game. Is it like Planet Zoo or something that's in this field, which is a little bit like a kind of computer game like SimCity?

Stani Borisová: Yeah. Planet Zoo is very fun as well and you really get immersed in this nice good vibes of a good zoo and you're trying to take care of your animals and make them happy and such. Yeah. That's also very fun.

Chris Adams: Cool. If people want to be inside to get away from some of the heat this summer, then maybe those are two things to catch people's attention. All right, that's all for this episode of This Week in Green Software. All the resources for the episode will be available in the show description below at https://podcast.greensoftware.foundation And you can see more episodes available for you to peruse and listen to at your leisure. Stani, I've really enjoyed having you on here and I've learned a huge amount and I think some of our listeners probably have as well. So thank you very much. This has been loads and loads of fun, Stani.

Thank you.

Stani Borisová: Thank you so much as well. It was great to be here.

Chris Adams: All right. Take care of yourself and have a lovely day. Ciao, Stani.

Stani Borisová: You too. Bye.

Chris Adams: Hey, everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit https://greensoftware.foundation That's https://greensoftware.foundation in any browser. Thanks again and see you in the next episode.

We Answer Your Questions!

5 juli 2023 | 56 min

On this episode of Environment Variables, host Chris Adams is joined by Asim Hussain as they dive into a mailbag session, bringing you the most burning unanswered questions from the recent live virtual event on World Environment Day that was hosted by the Green Software Foundation on June 5 2023. Asim and Chris will tackle your questions on the environmental impact of AI computation, the challenges of location shifting, the importance of low-carbon modes, and how to shift the tech mindset from "more is more" (Jevons Paradox). Chock-full of stories about projects implementing green software practices, and valuable resources, listen now to have your thirst for curiosity quenched!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Asim Hussain: LinkedIn / Twitter

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Questions in the show:

What computation is needed for AI, the explosive use, and what impact will that have on the environment? [7:17]
Regarding location shifting - is the foundation concerned that when everyone time shifts to the same location or the same greener grids, that can increase the demand of those grid's energy, which could increase fossil fuel burning to meet said new demand? [18:50]
Why not just run low-carbon mode all the time, not just when the carbon intensity is high on “dirty electricity”? [34:35]
Given the Jevons’ Paradox, how do we change the thought pattern that more is more in tech? [38:15]
Are there any notable examples of organizations or projects that have successfully implemented green software practices? What can we learn from them? [49:00]

Resources:

Peeling The Onion’s Layers - Large Language Models Search Architecture And Cost | Semi Analysis [8:51]
Simon Willison: LLMs on personal devices | [12:40]
2023: These Are the World’s 12 Largest Hyperscalers | Data Center Knowledge [13:13]
Energy Market Trends at the Layer Below the Internet Stack | Chris Adams [20:55]
Carbon Intensity API | Carbonintensity.org.uk [25:05]
Don’t follow the sun: Scheduling compute workloads to chase green energy can be counter-productive | Adrian Cockcroft [27:15]
Ecovisor: A Virtual Energy System for Carbon-Efficient Applications | Abel Souza, Noman Bashir [30:06]
Branch Magazine | climateaction.tech [35:08]
Green software requires a holistic approach | State of Green Software Report [38:46]
ICT industry to reduce greenhouse gas emissions by 45 per cent by 2030 | International Telecommunication Union (ITU) [45:35]
Windows Update is now carbon aware | Microsoft [49:24]
Xbox Is Now the First Carbon Aware Console, Update Rolling Out to Everyone Soon | Microsoft Xbox [49:37]
Sentry plugin for Grafana | Sentry Computing [51:55]
Storj [52:58]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Asim Hussain: We're talking about a cultural change that will take generations is I think what it would really take. I don't think this is gonna happen in our lifetimes. I think the world that you've described is a beautiful world. I hope to, I hope my dream it will exist, and I think it'll only exist if the culture changes and the culture changes worldwide and dramatically.

Chris Adams: Hello, and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host, Chris Adams.

Welcome to a special mailbag episode of Environment Variables. We're thrilled to bring you the most anticipated questions that arose during a recent live virtual event hosted by the Green Software Foundation on World Environment Day on the 5th of June with over 200 passionate practitioners participating from all around the world, the event featured an expert panel consisting of influential voices in the green software movement.

Our panelists at the time included Asim Hussain, who's here today. Hey, Asim.

Asim Hussain: Heya.

Chris Adams: We also had Anne Currie, the Greentech advocate at Container Solutions and Community Chair at the GSF. We also had Tamara Kneese, the UX research leader and strategist and lead researcher of the State of Green Software Report, as well as Pindy Bhullar, the ESG Chief Tech Technology Officer at the bank, UBS and the PhD in her own right. During the event, they introduced the Green Software Foundation, and unveiled all kinds of insights from the recently published state of Green software report, which sparked all kinds of engaging discussions. But we didn't really have enough time to cover all the questions that were coming in from the people who are asking there.

So you can think of today as a bit of a kind of roundup of some of the questions that seemed particularly interesting and felt like they probably need a bit more time to actually delve into them properly. So that's the plan for today. We're gonna look into some of the questions that we didn't have time to answer or I wasn't actually there so I didn't have a time to answer, but that's what we're doing.

So today it's myself, Chris Adams of the policy chair at the Green Software Foundation, and I'm joined by Asim Hussein, the chair and executive director of the Green Software Foundation and lover and grower of mushrooms. And Asim. I'll live some space for you if you wanna talk about anything in particular, cuz I have some interesting mushroomy factoids I'll share with you after this.

Asim Hussain: It's just the way you said lover of mushrooms was a little bit, um, risque. But yeah, I only love them cuz I like to eat them. But yeah, labro, I, I am a mycologist. I grow mushrooms and currently actually very desperately trying to rescue two bags of a lion's mane mushrooms, which I think I've let rot in our bags a bit too long.

So hopefully I'll have some lion's manes mushrooms this time next week.

Chris Adams: That sounds cool. Can I share my fact about mycelium, which I think is really cool. So I listened to a podcast called Catalyst with Shayle Kann, and he recently did an episode where he was talking to one of these researchers in South Africa who published some research about what I referred to as Microrisal-

uh, The fungus. Basically it's the kind of fungus that you can think of being attached or interfacing with the roots of a tree. And they did some research for the first time to get an idea of how much CO2, how much carbon is actually sequestered by this. Cuz there is a kind of symbiotic relationship between this particular kind of fungi and the trees.

So basically the trees they. They make sugars and stuff like that, whereas the fungi are really good at leeching out all these other kind of nutrients, and they have a kind of, they're basically swap basically. So the mushrooms give the trees all the kind of nutrients, and the trees give sugar to the mushrooms by comparison.

And as a result of this, that answer meaning that there's a bunch of CO2, which basically is drawn into the trees and then fed to the mushrooms and the figure that they calculated for the amount of extra carbon that's stored in the soil. It was something like 13 gigatons, which is a third of all of the carbon dioxide that's emitted by all us burning fossil fuels.

This was so cool. I had no idea that shrooms were basically doing all that extra work under the ground, and it made me think of you Asim. I thought, yeah, as he would like, he would be proud of his little guys. When heard about this,

Asim Hussain: I am proud of my little guys. Yeah. No, read the same. I don't think I, I didn't listen to the podcast. I actually didn't dig into it that deeply. But yeah, I read the similar article recently and it's mycrorrhizal.

Chris Adams: Thank you. Yeah. Sorry.

Asim Hussain: Yeah. Mycra is a mushroom or the fungal component. Rhiza means root. Basically a type of fungal relationship where it works in symbiotically with roots and yeah, there's a really good book called Wood Wide Web, which talks about this wonderful relationship between trees and just everything around you actually. And they've actually even shown that kind of mushrooms and the kind of, these networks are really large and they actually act as not even

Chris Adams: a network.

Asim Hussain: like a network, but also like a controlling network.

So you can have two competing trees. But, and if one's not surviving well, the microrrhizal network will then force that negotiation for sugar, for carbon in a way that benefits one of the trees, because actually, like a break in the canopy is really harmful for everything. So it's like, there's lots of really intelligent stuff like that happening and um, yeah, it's really, really clever.

Chris Adams: That's kind of cool. I had this idea that they're a bit like the kind of shroomternet, but didn't realize there was this kind of extra, almost what diplomat, I see your diplomat role you're playing as well to get various real entities to play nice with each other. Wow. That's cool.

Asim Hussain: really cool. And there's just li- for the listeners. There's a really interesting guy out there called Paul Stamets, and if you've not heard of him, just look at, look him up on online. And he does some really, he's been in this space for a long time now. He's an amateur mycologist who became famous and he is been really pushing this world and he's got really lots of great TED talks on why mushrooms are all kind of a sustainability solution. It's got books written on it. And if you're actually a fan of Star Trek,

Chris Adams: I knew we were gonna

Asim Hussain: you're even gonna head there. So there's a new Star Trek series called Star, star Trek En, no, hang on. What's it called? It's just called Enterprise actually. And the lead engineer is called Stamets and it's actually based off of this Paul Stamets. And the engine they have is called the Spore Drive, and it can travel to anywhere in the universe through a spore network of stuff. And I just thought it was amazing that even my one, my favorite Star Trek, even that has been mushrooms have integrated even into my favorite sci-fi show, which I just love.

Chris Adams: Oh wow. Is it Enterprise or is it Discovery?

Asim Hussain: DIscovery. Sorry. You're right. It's

Chris Adams: that was the

Asim Hussain: Yeah, yeah, yeah. Enterprise was that other one that didn't.

Chris Adams: Oh my word. We've gone full nerd already. Oh. At least. That's hopefully a kind of entertaining kind of what's segue before we dive into the mailbag. All right. Okay. Should we start with a mailbag then? Asim. See what the first question is.

All right. Okay. I, I watched the recorded World Environment Day kind of panel that you had and we'll share a link to that. So there will be some things that we're not gonna cover because they've already been covered in that recording. Okay. So the first one touches on at LMMs and things like that. So the question as I read here was what computation is needed for ai, this explosive use, and what impact will that have on the environment?

That seems to be the question. Asim, I'll put this to you and then we can have a bit of a, kind of take it and turns on responding with that one actually, if that sounds good to you.

Asim Hussain: Yeah, sure. So I think obviously AI has come up quite a lot even on this podcast. Many times I think the. There, there's been couple of papers. Now there's been one, I forgot what, I think it was about five times. I think that the, we, we'll make sure we quote the, the actual paper in the thing. There's one original one when the, when ChatGPT three came out, which suggested that the energy consumption of a Google Bing chat, uh, search ChatGPT search was five times more. Am I getting that right? Does that sound familiar, Chris? Five times more than, than a normal one, which I'm, now I'm starting to wonder even if that was a, a understatement because actually like. You have multiple chat conversa because there's a few, a further paper which talked about the water consumption, uh, from a chat.

And that one was interesting cause that talked about the whole conversation. Not one question to ChatGPT, but like the series of conversations you have to get to an answer. Then that got to half a liter of water. And I'm not too sure if that original paper was talking about one individual question or the series of questions you've gotta get to to get to your answer, and there's lots of kind of evidence.

I don't, I'm not too sure. I think there's a significant amount of compute being used in LLMs and there's not a lot of transparency on it right now.

Chris Adams: This is actually one of the problems that we do have. Right? So the thing that I might share for you is like you could do a kind of bottom up get based on a look at this by looking at, say what a Nvidia H100 is. Make an assumption of how many might be in a given data center and work out okay if a H100 from Nvidia kind of graphics card or ML card specific pulls this much, then that times this number gives you some kind of number.

And you do see it particularly. Big figures like you see people throwing figures around in the kind of hundreds of megawatts or gigawatts of power, particularly in America right now, for example, you do see figures like this, but I've actually struggled to find particular like specific figures on this cuz.

It's also worth sparing in mind that right now the thing that we could actually point to is some of the existing research that you see from some of the kind of cloud providers right now. There is a company that called Vantage, I believe, and they do cloud billing analysis, and they recently shared some information about, okay, what proportion of spending is actually coming from, say, is being allocated to GPUs versus CPUs and stuff like that.

And I'll share the link in a document for us. But the general argument is that yes, AI is large. I think that the figures on the front page that for their most recent report, shows that you've got something like what you might qualify as AI in this compared to as a share of what people spend on typical cloud like Amazon, elastic Compute Cloud, E C two, and the figures are between.

Six and 8% of all E C two spending is now going towards this, but it's growing extremely quickly. That's the thing that we are seeing for this. And uh, I'll share a link to this cause I suspect there'll be a new version of this clo- cloud cost report, which gives us some numbers about what people are using on billing.

But this is not the same as what kind of forward investments people will be making for this. And one of the problems that we see is that, you basically are running up against the limits of the grid right now because compute, cause data centers are such dense uses of electricity. You have scenarios where a data center full of these kind of cards will use more power than a grid is able to actually deliver to it.

So you're constricting factor right now is the capacity of the grid to feed into the data center more than anything else. And that's one of the key problems that we're seeing come up again and again at the moment.

Asim Hussain: I hear all the arguments and again, like I think there's a real lack of data right now to really get clear answer, but I think I. Then, then you have to look at other kind of economic and proxy arguments. And I think there's one factor which no one can ignore, which is the amount of interest and investment which is heading into LLMs and AIs in this space is outstripped anything that existed before. And that could not be true if this was not a at least perceived as a massive growth opportunity for organizations, which I think that there would be like a knock on effect in kind of emissions and something like that. Whether that's right now is a good point.

Whether it's right now or whether it's like everybody's seeing that this is the future and this is what the putting effort into, which means that this is gonna be a big growth error in the future as well. I think that's this aspect of this, that it's just true and we can't ignore, there's a lot of interest in this space, so I think that's the thing to think about as well.

Chris Adams: So there's one thing that it might be worth bearing in mind, is that I don't think it's related to the fact that in many places, that people are constrained on the supply, being able to actually meet this demand for it. You are seeing a real kind of, crop of new, smaller, much more efficient LLMs being created specifically because there is an interest in being able to not be dependent on either a singular provider of this or just being able to run this on, say, your own hardware, for example.

So I'll share a link to a really nice post by Simon Willison, who's been talking about some of the most recent open models that are designed to basically run on a laptop that can in many cases give you results which are comparable, if not indistinguishable from some of the really expensive LLMs and expensive generative models that you see right now.

So there is a kind of shift for this, and I actually dunno what direction you're gonna see because like you said before, because we don't have access to what kind of percentage the AI is really making up right now in terms of future investment. We don't really know. And like when I've, I've seen what I've done.

I've shared a link in the show notes to. This chart from data center knowledge, it basically gives an idea of projected growth by hyperscalers over the coming few years. And we can see figures of maybe say Google like 3000 megawatts of assumed capacity right now, Microsoft at similar figures around the 2000 megawatt mart.

And same with Google. And you see that there's projections to double that over the coming few years. And I dunno if these are before or after the decisions that people have been making for this, because you've gotta remember that.

Asim Hussain: right.

Chris Adams: We've only just seen in less than a year did you see the Nvidia H1 being released and not just that Now AMD have come out with their own ex, the equivalent, they're competitor to this.

So you now see many more things available. So the question is, where are these gonna go and how are they gonna be powered? Like the worst scenario would be that people end up. Basically setting up data centers and then finding non-G grid ways of generating power to actually run these machines, like using diesel and stuff like that.

I'd really hope that doesn't happen, but I can imagine scenarios doing things like that or I. Possibly opening up some of the existing generation people that have shut down from cryptocurrencies. And since we saw this crypto collapse, and this is one thing that Tamara mentioned, she said it's worth looking at the role that LLMs are placing in the kind of public discourse.

They're very much filling that same role that the Metaverse was supposed to fill in or the other NFTs were supposed to fill maybe a year ago, for example. It's worth saying is that yes, we know that this is being seen as a drive of demand. Whether there are actual numbers that are reliable right now is another matter, and like we can point to various figures for demand and what people are looking at.

But at best, these are all currently like trailing indicators, like the Vantage report, which shows these figures. That's only based on what's already been set up. And that doesn't really tell us a story about are people stuck with buildings full of graphics cards that they can't plug in and sell AI for right now.

Because they can't get this stuff connected to the grid, for example.

Asim Hussain: I just also think that we just can't ignore the fact about the amount of money that is flowing into this space. And I hear the arguments for open source models and my heart really wants them to win out, but when it costs like a hundred million dollars to train up something like GPT three, which I think was the estimate.

And the real benefits come when you like compute even more and more. I just don't know if, if the open source models will win out because obviously people are spending money because there's an advantage to doing so. People wouldn't normally be able to peel off a billion dollars from a company just for no reason.

So it's, I think that there's lots of data we don't have. That's the data we do have, and that's telling us I something at least that there's a lot of,

Chris Adams: Yeah, I guess this is one thing that you currently do not have. Again, it's really hard to get a decent number about where things are going with this, for example, because I think I'm in the kind of camp where actually there is a lot of interesting stuff happening with open models where people are basically defining where they're gonna compete and try and come up with alternatives to this.

Like when you look at some of the purchases being made by companies which aren't just Microsoft and Facebook and Google, right. I think the example of Databricks is a really good example in my view, Databricks. They published a bunch of open data specifically to help build this competing ecosystem, and they recently purchased a company called Mosaic ML.

They published two in openly licensed LLMs. One's called MPT 30 B, and the other one called MPT seven B. These are the ones that you can run on our laptop right now. And they're large, like 19 gigabytes in size and you need a relatively chunky laptop, but you are seeing this and they are, if not comparable, you do see, I think there is, you do see a bit of an arms race right now, and it's interesting where things will go because we've linked to this whole Google has no moat. Sorry. Open AI has no moat and neither do we memo that I shared from Google before, but right now we're not quite sure. I don't know if it's a function of organizations just having loads of access to cheap money and doing this, or people seeing it like, oh, this isn't actually defensive in the long term.

Because now you see all these open models coming out. It might be comparable. It might be the case that if you can just get to the workloads already or get to the workflows, people are not gonna care that much in the same way that you know how loads of us end up using, sorry, loads of people end up using Microsoft Teams, not necessarily because it offers

Asim Hussain: about it recently.

Chris Adams: offers.

The best user experience as someone trying to join a video call. Right. It's more the case that there's a workflow that people have, or there's a way to bundle some of this in. I wonder if that's where some of these network effects are actually more important than essentially the training staff and all and it's, and that's where the levers are more. So, yes, AI is interesting and cool, but it's actually much more about market structure, antitrust and stuff like that. That's probably gonna be more of the drivers perhaps.

Asim Hussain: Oh, interesting. I see. It's not how powerful your model is, it's how you can integrate it into your existing business models take how can mon, how you can monetizing or even don't even monetize it, how you can use it to strategically win versus your competitors. And that might not even necessarily be like anything really to do with the raw power of the model.

Chris Adams: We've gone a bit off the initial question, but Yeah. But yeah,

Asim Hussain: chance, that's, that's a risk you have when you ask an AI question to this, to this podcast.

Chris Adams: Yes to people who with possibly questionable data access to information. Should we move on to questions? So for the next ones, these might be ones which are more solid footings for us. Okay. So the next question was one about basically time and space, location shifting and time shifting when people are talking about the idea of carbon aware software.

And the question is basically regarding location shifting, is the Green Software Foundation concerned that when everyone time shifts, To the same location or to the same greener grids that can increase the demand on the grid's energy, which should increase fossil fuel burning to meet said new demand.

Now, Asim, you might need to unpack this, first of all, for people to understand what happens with the whole merit order for that, because it's not immediately obvious if you're not familiar with grid workings. Why lots of people using computing in one part of the world will lead to more fossil fuels burning other than literally just it's a part of fossil fuels, for example.

Asim Hussain: That would be, let's say, if you had a frictionless capability to move your compute to anywhere in the world at any given moment in time, and you just picked. And every, and everybody in the world had exactly the same capability of frictionally moving their computer. Whatever's the greenness right now, it's probably gonna be France or somewhere in the Nordics.

And then what would happen is, let's say in the next hour, it's France. There's the greenness in the world. Every single bit of compute in the entire world would just move to France. Then those data centers would then, or if they can theoretically handle that load, will then suck up all that green energy, which made them kind of the greenest grids.

And those grids still need to make an energy for the people in France to boil their kettles and do all the other things they need to do. And so all they can do at that moment is burn stuff, typically burn stuff, you know, coal and gas and these things. Those are batteries. Those are chemical batteries. So you, they're the things that you can spin up.

Gas, especially stuff you can spin up very quickly and so effectively that just burns more fossil fuels. And so that's basically, that's what would theoretically, this is very theoretical, that would theoretically happen, um, is if you did, if everybody did just move all their compute over to, let's say France, France would be forced at that moment in time to burn more coal and gas.

Have I explained that correctly? Have I missed something out, Chris, or?

Chris Adams: I think that's about right. The thing it might be worth us talking about or sharing in the show notes is an article I put together called Understanding Energy Trends. At the layer below the internet stack, which talks about this, there is a kind of really nerdy techno concept called like merit order.

With the idea being that different kinds of energy have different costs, so things like solar and wind, once you've installed them, because you are getting the fuel from the sun, you don't have to purchase that extra sun to run it. So essentially the costs are almost free. So that's very low, right? Now there's other things which are designed to work, which are really redesigned, which are again, expensive to install.

But once they're installed, the fuel is relatively cheap for the amount of power they get out. And like nuclear in a good example of this, where you can get loads out that way. Now you have other kinds of fuel like say coal and gas and so on. But broadly speaking, the higher the cost of the fuel you, you have a trade off where things can respond more quickly in response to demand, but they're usually dirtier.

So the idea, that's the kind of the idea behind this, and I think the argument being made here is that if everyone moved all their computing to one part of the world, we would in induce all this extra demand, which could only be met. By things responding to the extra load on the electricity grid, which would usually be met by people spinning up really dirty, gas fired power stations or extra coal or stuff like that.

I think that's the argument and that's essentially the question. Now that we've actually understood, explained the premise, I should ask you what is the kind of official response to this? Let's, is that likely to happen? Is this the thing we should be aware of and how do we respond to that then?

Asim Hussain: My answer to this one is always, have you ever watched the TV series called The Wire? So watch

Chris Adams: Yes.

Asim Hussain: watch The Wire.

Chris Adams: not sure. I'm not sure where we're going, but go with this. All right? Yeah.

Asim Hussain: with this. It's about gangs and police like in New York City, but there was one of the latest seasons, there's a gangster called, I think it's Marlo Stanfield. I remember one of the episodes, like one of his workers is telling him something.

He turns around and goes, that's one of them good problems. And that's what I think about this thing. So someone's telling me a problem and I'm like, this is a good problem to have. If we are ever even remotely getting to the point where demand shifting is affecting a grid, that is a level of achievement, which is excellent.

Yes. Okay. Yes, there are negative consequences to that approach, but we are not even remotely there right now. So worrying about that is I think, a little bit too hyperbolic at the moment. You shouldn't do something because if you take that thing to the absolute extreme, it will be negative, I think is.

What I would say to this argument, I would say demand shifting is never going to be the one solution you have in your pocket to reduce your emissions of your application, your architecture, I always describe it as one of the things that you can do. It's one of the easier things to do. It gets you started on the much more challenging journey of energy efficiency, hardware efficiency, reducing the amount of energy you use, reduce the amount of compute you use.

But it gets you there. And I think that's why a lot of people have been interested in carbon air computing. I always say it's not going to be the solution that solves climate change. It's nowhere near gonna be that solution, but it's a stepping stone on the journey there.

Chris Adams: I think I would've asked that slightly differently actually. Now see, because when I see this question being asked, it's essentially will this demand cause people to do this? I think this has some assumption that people who are basically trying to move computing here, they're doing it because they're looking for kind of greener energy right? Now, I think there is one way that you can solve this purely from a just information point of view, and if you are looking for the lowest carbon intensity and you can see that the carbon intensity is increasing, then

Asim Hussain: I see. Yeah. Yeah.

Chris Adams: you would just choose to not run it there. Right. So this is somewhat dependent on organizations having some of this information published and visible for people, but I think that's actually something that can be done.

And the i e, even in places where this is not public information right now. So for example, in the uk this information is visible on a really clear basis right now, like the UK there is, I think there's a website called carbonintensity.org.uk, which publishes things on a very permissive license for this.

France already has its data available that you can pull out, so if you're gonna do this, then I think you would just look before you deploy something, or you would build some software to check if you're gonna, if you're gonna make it worse. That feels like that would be one of the solutions there, but that feels like a thing that is something that will be made available to people in a number of different ways.

Asim Hussain: You reminded me of a very interesting conversation I had with somebody from Google, cuz it created a service, started with v something and you pumped into that service. What then? I think it was the next day. So the next hour's workloads were going to look like. And then what the next, I'm gonna say tomorrow I think it's gonna be, I think it's day by day? But what tomorrow's carbon intense, the grid's carbon test is going to look like. And you actually pulled lots. You pulled, you didn't just. Didn't just like where's those carbon intensity? Push it there. You actually had a thought. You actually need to run quite a lot and I can't push it all in that one place.

Let's, let me be more intelligent on where I put it. And the idea was thrown out there. What if this service like existed? What if we all collaborated? What if we all said tomorrow, and this is not an unlikely to happen, but what if you all like was so open that we said, do you know what I'm going to submit to this online database that I'm need to run this much compute?

I need to run it tomorrow. I want to run in the greenest region, and it's scheduled it for everybody. And so you run in France, you run in Germany, you run in Norway, and we get it all together. That kind of openness, the data, I think would solve this problem as well, but I don't see anybody, any corporation being that open regarding their workload, future workload.

Chris Adams: The thing that you just described there, Asim, was essentially how energy markets work, right?

Asim Hussain: Oh,

Chris Adams: And they're regulated markets where there is not one owner. You don't, where you, rather than just having only the Amazon Cloud or the Microsoft cloud, you have multiple things, right? So in order for that to be possible, you would need to have a.

Kind of different structure or you need to have people who played a part in actually making sure things can be dispatched to the kind of correct actors in the role you'd need, like a feed in tariff for compute or stuff like that. I feel this is actually quite a nice chance to draw people's attention to a really interesting proposal that I felt from Adrian Cockcroft actually, who a, he's written a piece about why the whole kind of idea of trying to schedule workloads can be counterproductive, which I'm not totally sure. I agree with all of it because I think that whether you actually can see people shifting, creating that demand is another. I'm not sure if we have seen that, but I understand the kind of thrust of his argument.

But the kind of real time carbon footprint standard that he's proposing feels like it'll go a long way to actually providing the numbers that people would actually have access to or need access to realize, am I gonna really doing this? But the other thing you could do is to just actually like price carbon into the cost of cloud, right?

We already have spot markets. If you had in your spot markets, there was a spot plus, which basically had the cost of carbon at say a hundred Euros, a ton or a hundred dollars a ton, and you looked at that, then that would actually be a really, in my view, a fairly simple way to make sure you're not shifting all the compute to the worst places.

For example, like again, like if you're gonna go like totally neoliberal and price based, then yeah, that's how you can do this stuff. But the other thing you can do is literally just go totally whacky, just have massive batteries and data centers the way people look like they're doing for other significant drawers of electricity.

So for example, if you're looking at say, electric car charging stations, lots of them now have lots and lots of onsite storage to deal with the fact that car driving. It's really spiky, so most of the time it's not being used, but then people come to it and they need to pour a huge amount of power very quickly.

Likewise, like high-end induction hobs, which have their own batteries inside it, to again, deal with this big spike in use. There's lots of kind of strategies you can use, which don't meet mean that you don't need to actually burn fossil fuels for this. It does involve nerding out about the grid, and that's one thing which is new to a lot of actual people who are working with this.

So to an extent, I don't see it in the same way that you might see it. For example, Asim Cause I feel this is actually one that can be addressed using various techniques, people using other sectors which aren't technology, for example.

Asim Hussain: I will just say it's probably clouded from having many conversations with people about carbon aware computing being shut down because of versions of this question. Yes, car- we like the, but if it was to be taken to the extreme, then it would destroy our whole infrastructure, which I think for me, I've had this question and in phrase as a response to me as a reason for why we can't even entertain looking at carbon aware computing.

So that's where I, you've probably triggered my default defense mechanism against this question, but I, there's, I think there's a very nuanced, very important topic. I think it's important for you and me to have different opinions cause that's how we get or share all this important knowledge, uh, with the world.

Yeah.

Chris Adams: All right. Oh, there's one thing I should actually go add a bit of a kind of plug for. So there's some really interesting work from, I think it's Abel Souza and Noman Bashir. They've been publishing some fantastic papers talking about specifically the likely impacts of carbon air computing, what the savings could possibly be.

And it is a kind of quite a technical paper. Actually, when I read, I was like, oh wow, there's a lot of numbers in these charts. But it's really good. I found it one of the most useful ones for informing my opinion about where this goes, and I'll make sure that we share some links to this cause I think it'll actually add some extra nuance to this conversation.

The other thing to bear in mind is that if you just have this, then. Literally, it's not like large Hyperscale companies are not making enough money to buy batteries, right? If they're able to spend 70 billion on share buybacks in a given year, they can probably afford that. Buy is literally hundreds of megawatts of extra capacity.

You could just have five or six hours of local battery storage so that you wouldn't even need to touch the grid. You just run it locally if you want it to be sure that your power is green, but that's a separate discussion. So I'm just gonna park that once again cuz this is a, I'm worried about sound like a bit of a broken record on this one.

Asim Hussain: You do. I don't know where all this money is cause it doesn't find itself into my, into my budgets. But yeah.

Chris Adams: There was a report from the Rocky Mountain Institute who were doing some analysis on green Bitcoin and things, and they'd said, with 115 billion US dollars, you could buy up every single coal fired power plant on Earth and replace it with renewables and.

Asim Hussain: 115 billion?

Chris Adams: 115. Yeah. Last year, the combined share buyback, so the money made by Apple, Google, Amazon, and Microsoft as in they had so much money that they just thought, oh, I'm just gonna buy my own shares.

That was more than $125 billion. So single handedly. That could solve it in a single year, but we've decided to spend it on, do you know the problem with climate change? Shareholders aren't getting enough money back. I feel that this is the thing that we need to be talking about. If when we're talking about green software, it's like where is this money going?

Cuz we clearly have the money for this. It's just a case of priorities and we could be moving faster if we really wanted to. But that's again, I'll stop now because I'm a little bit ranty. I'm a bit worried. Sorry about that.

Asim Hussain: I want, I want everybody to know that the entire time I worked for Microsoft Chris. Adams would always type whenever he is talking to me on chat, would always type m dollar sign, always m dollar sign for Microsoft. But I dunno why you don't do, I'm an intel now I suppose. I suppose there's no, there's, could you euro you could do in Intel, but I don't know, I dunno.

Maybe there's another, maybe there's a reason why you don't put Euro, a euro signal into there for Intel. But anyway.

Chris Adams: think it's, cuz this was actually something that I, when I used to read The Register, when I used to, when I first came into technology and they used to call, I think IBM was called Big Blue and it was, there was another one with the beast from somewhere. And uh, yeah, Microsoft wears M dollar. Like

Asim Hussain: What was it? So

Chris Adams: because they made so much cash.

Yeah. It wasn't me being smart, it was like, oh, total second hand whim, mate. Yeah. They are very effective at basically

Asim Hussain: very, they, yeah. They've

Chris Adams: shed-loads of money. Absolutely.

Asim Hussain: yeah. That's what com, that's what corporations are there for.

Chris Adams: Do you know what? In India, right, I didn't realize this, but in India, every large company as a condition of working inside India, which most populous nation in on earth, something like 20% of all the profits of all have to be like by law in invested in what India considers like priority areas. So specifically into renewables straight away, right?

So there are all these mechanisms that people actually do that mean that we can direct funding to places to speed up action on climate. And if we're talking about technology and talking about carbon awareness and stuff like this, then we really need to be prepared to think about and have conversations about how much in the way of resources do we really wanna allocate towards what the science is spelling out and how much do we need to make sure that share price goes up, because yeah, okay. It's nice that people have like pensions and things and all that, but also it would be nice to have a livable world and having just this much money available feels, come on, let's like get this sorted.

Literally one year would solve it, but that's another discussion. Anyway. Let's move to the next question because I think people listen to, if they're trying to here to learn about code, not about economics.

Asim Hussain: we've turned into a, we started off with opinions on ai. We turned into an energy podcast, and now we're talking about capitalism. So let's just go, let's turn into a politics podcast. Let's just do it.

Chris Adams: I guess it's everywhere. Let's move to, okay, next question.

Asim Hussain: next question.

Chris Adams: Okay. The question, this was one about this idea of some machines and some software running in a low carbon mode, and the question basically came out saying, why not just run low carbon mode all the time? Not just when the carbon intensity is high on dirty electricity for bits of software.

This is essentially one of the questions that came from this. And I think, uh, this might be a reference to like things like Branch Magazine or even with your CarbonHack thing. One of the winning designs was, uh, software kit that would show different kinds of versions depending on how dirty the electricity was to stay inside a carbon budget.

Asim over to you.

Asim Hussain: It's an interesting question. Well, a, the user, I presume I, the way I've always imagined and the way you, in fact, you can use Branch Magazine. You can just go on Branch Magazine and say low, assume it's high. I dunno how the terminology, sorry, but low carbon mode all the time. I've always imagined these kind of UI modes in your system as something you can select if you wanted to or something we should auto select based upon that aspect of how, how carbon it is. So I, I've always imagined it's, it is user driven and I suppose if you've got a product and they, and you force yours into low carbon and the competitors doesn't, and all your users move over to competitors, then you've gotta factor that in as well.

But I also think, Chris, is there like an argument here about, does that stop money from going into renew? I don't think it does. I think the money will go to renewables.

Chris Adams: No, this basically, this isn't really about, uh, I don't think this

Asim Hussain: Yeah. Yeah. I don't think it was finance. Yeah. Yeah. Okay. Yeah.

Chris Adams: Okay, so the way that we designed this in Branch Magazine, cuz we were like playing around with this idea, we basically made it a thing that was user definable so they could choose to override this, but we would set a default to kind of. It was as much an education piece as anything else because for a lot of the time people aren't even aware that most of the time you don't even think about where the power comes from. So the idea that this is being foregrounded and the materiality is being exposed to you was the new idea for us. That was why we were doing it, to really emphasize this, because we thought this is a nice way to park back to some of the ideals of threat being something that's supposed to be open for everyone accessible and everything like that. So we figured if you design a low carbon mode, that kind of emphasizes the fact that the grid changes, but also at the same time emphasizes the fact that when you're using something, it should be accessible for people who cannot be who, who may be partially sighted or stuff like that.

Then you can embed some of these other values in how you build things to communicate different kind of sensibility. So I think the general answer is a lot of the time people do quite enjoy having quite rich experiences and having a kind of sober or monkish experience all the time might not be particularly compelling for lots of people.

And I, I think that's okay to actually be explicit about some of that right? You, I don't think it's realistic to think everyone only you ever wants to see some of this stuff or even make all those decisions for someone else. I think that might be a little bit too paternalistic, but that was my kind of take for it.

But there are things you could do to kinda hide this. You could possibly design it so that. When you build something, there's certain things, there are ways to provide a rich experience whilst reducing the kind of resource impact in the same way that you can refactor code whilst reducing the amount of computation it needs to consume to do something.

And I think that's the thing for it. But I figure like the thing you should probably do is just put it into the user agent or if you're look at, using a browser so you can have like a user agent of change where they decide this stuff themselves. You already have, do not track low bandwidth please. Stuff I think, stuff like that I think would be cool, but we don't have any browsers doing that yet, but this may be early days.

Asim Hussain: Yeah, yeah. Yeah.

Chris Adams: All right, question four. This question is about Jevons Paradox. All right, so this basically says the question is given Jevons paradox insight number four in the state of green software report, how do we change the thought pattern that more is more in technology?

Now, it might be worth just briefly explaining what Jevons Paradox is. Before we dive into this question, Asim, I could probably do it, have a go at talking about Jevons Paradox, if you wanna get ready for answering this particularly thorny question. Should I do that?

Asim Hussain: you do it? Yeah. You do a better explanation of it.

Chris Adams: Okay, so first of all, I'm gonna point people to the fact that this is the fourth insight in the report to buy some time.

But basically, Jevons Paradox is a name given to the phenomena where when you increase the re the efficiency of a particular resource, using any resource, you can increase the absolute use, even though individually it's more efficient. So this initially came from hundreds of years ago when William Stanley Jones noticed that making coal fired steam engines more efficient meant that more people used coal fired steam engines in new places, which would lead to an increase in the absolute use of coal.

And he was so worried about this that he thought we would run out of coal. So he started writing all these papers about, please could we not do this? This is terrifying. We're we are gonna stop progress if we make everything too efficient and hundreds of years later. This kind of applies with things like cloud computing and stuff like that, or it's often used as a way to say, you can't just talk about efficiency, you need to talk about absolute figures.

So if you make something more efficient, you just result in more use. And the common example is cloud. So by making hyperscalers talk about cloud being much more efficient. But the flip side of that is because it's suddenly more efficient, more and more people have access to it, which increases the absolute usage of this.

And we have seen absolute increases in just kind of technology and compute use. And I think that's one of the things that. Is what's inspiring this. But you also see it in things like ride sharing and stuff. There are examples of Uber and Lyft and other kinds of companies. When you make it really easy to hail a ride, you result in more people driving.

You increase the miles driven in a given city because it's so much more convenient. It's also subsidized by venture capital as well, which makes it e uh, which makes it cheaper than other options, but that's one of the impacts you have. But broadly speaking, making things more efficient is said to have a kind of rebound, which can increase the total use.

So that's it.

Asim Hussain: Yeah, I just think it's, part of it is really just, it's about resource constraint. You're, cause we were talking about earlier on, weren't we? I remember what the context was, but we were talking about if you're resource constrained, you have to make different decisions and you don't use that resource as much just cause it's just, I think it was AI, just cuz it's not there.

And so if you're resource constrained, if there's only 10 of something in the world, you'll just make choices that only use 10 of something in the world. But then if you make it 10 times more efficient, you'll still use all of the resources that you had and just use more of it. So I think that the argument here is as you're making things more efficient, that natural resource constraint, which was forcing you to make these trade offs and be not wasteful, at the very least, disappear.

Then you can just, you just start being wasteful. So I think the solution here is there has to be a constraint some way of, it's not, you're not gonna stop. I don't think we're gonna stop Jevons power. We don't think there's any way we can really force the world not to make things more efficient, just because that's what we're just absolutely engineered to do.

But what we need to do is to enact constraints, whether they're artificial, whether they're regulatory, whether they're some other aspect of it. We need to enforce that constraint and that's how we do it. Like for instance, when organizations set carbon targets and another kind of like targets to achieve.

Right now we have this kind of, it's okay, you can, we just, we'll just carry on increasing. It's okay for now, but there needs to be a real extra constraint, which forces you into those actions is, I think that's the only way. That we're really going to deal with this cause I, I see it right now, even with ai, like as AI gets more and more efficient, we're just gonna use it more and more to solve problems.

Inefficiently, but to, but conveniently. Yeah. Yeah. That's my answer.

Chris Adams: Okay.

Asim Hussain: Do I win?

Chris Adams: I think so I'm gonna ho, I'm gonna wait for the jury to be out and we'll put that to the listeners. That's what we should do. So basically, I think the main thing I'm getting from what you said there is that you do need to be prepared to talk about absolute figures here, and that's one of the key things, so

Asim Hussain: Well, I wouldn't, I wouldn't necessarily agree with absolutes. I would just say there needs to be another for, or not even just one, multiple other forcing functions to force your usage down, whatever that is. I don't know what it is.

Chris Adams: Oh, okay. I guess this is a little bit like when people talk about carbon budgets on websites or carbon budgets on services you've got, that's a decision that people have made to go for that. I have a bit of a struggle with this term because when people talk about Jevons paradox, it's often used as a kind of way to say it doesn't matter that you're, you are talking about efficiency because you, you are just gonna make it back.

And there is a kind of subtext which basically says, why are you even trying? It does feel a bit kind of "okay doomer" and if we look at the last say, couple of decades, yeah, we have seen. If we follow like the IEA, the International Energy Agency, what the energy people who look at how much power is being used by stuff.

They basically say that over the last, say 10 years or so, we've seen, we've seen a massive increase in the use of computing. But if you just look at the energy usage, we have not seen the corresponding at the same increase in the energy being used so that we use more computing. But the energy's been more or less level. Now, you can take up issues with those numbers because when you look at numbers that include, say China and stuff, like the numbers look quite a bit higher than what were from the IEA, but that's not peer reviewed and we can't really use those numbers yet. So they, there is some contention there, but I feel this also just ignores the fact that people have been moving faster than Jevons Paradox to keep things better than they otherwise would've been.

And I feel like when this is rolled out, a lot of the time it's not rolled out in a way that says, is there a 10% impact? Is it, do you get 10% of rebound? So if you're doing 20% of savings, there's a net saving here. And like without these kinds of numbers, I think it's actually, if it ends up being quite an academic and difficult thing to engage with and like, I think that's one of the struggles I have when we talk about some of this, because it's often used to either disincentivize people trying to make like honest and effective changes in the efficiency of stuff, or it's being used to, I know almost as a bit of a gotcha to say it's still so it's still doing this and I feel like, oh, congratulations. Oh, I'm really glad you told me that. We should be thinking in absolute terms about the climate, right?

I'm like, okay, yeah, surely we've established this years ago. So that's the thing. But this doesn't actually answer the question of how do we change the thought pattern that more is more in tech? I suspect what you said Asim was this idea like, I don't know. My assumption would be that you do need to actually be prepared to think about absolute figures and how do you stay inside those and you consider those a constraint in the same way that you might design something that has to be accessible. You say it has to be staying inside these kind of targets that need to be improving each year. That's what the ITU, which is the International Telecommunications Union and the Science-Based Targets Initiative. All these folks are basically saying, Yeah, the absolute carbon emissions of the ICD sector has to be halved by 2030.

And I think that's one of the things you might need to do is actually have some narrative that says, oh yeah, we're gonna halve our emissions by this much, and then how do we fit that into how we work outside of technology? There are ways to talk about some of this, cuz this is essentially, this is a little bit edging into the whole discussion about do you need to, is it growth first or is it, do we have to, do we target growth so we can have nice things or can we just aim for nice things automatically, directly.

Yeah, because there's things like, yeah, donut, economic, there is like typical kind of economic thinking, which is the thing we need to do is get really rich. And we might poison ourselves along the way and endure all this damage, but because we're so rich, we can then undo all that damage and somehow, I'm not sure that somehow unmake the extinct, all these things which we made extinct along the way.

That's one of the arguments around growth. So you have enough wealth that should pay for things, but schools have thought like donut economics and so on, and they basically say there's a social foundation of things everyone should have access to. There's certain kind of overshoot and like we can target making sure that everyone has enough of a social foundation.

By while staying inside this zone, if we just target that stuff first, if we think about the things you want to have immediately, so rather than just focusing on get rich first, we can focus on, let's make sure everyone has access to shelter, has access to healthcare, stuff like that. But again, it's like using needs is what you might think about it.

Like what user needs can you meet here that you're targeting first, rather than trying to grow for something large there but,

Asim Hussain: Yeah, I agree with you, but I just think it falls into that same bucket of arguing, which is, here's a bunch of ways the world could be better, and if we did it this way, wouldn't it be amazing? We would just have all these issues would be solved, and I'm always like nodding my head and going, that sounds beautiful, Chris.

I would love that world to exist. I just have. Absolutely no idea how it's even remotely possible to get there, given the way the current engine works. Actually, the only way, I'll tell you, the only way I think this will ever work, and this is the only way, and I think I, and I do think this is not gonna be something that happens by 2030.

I think this is something maybe our children might be able to, we'll be dead. We'll be dead by the time this, this can change. We're talking about like changing the Overton window, talking about changing the Overton window, like we are talking about A cultural change that will take generations is I think what it would really take.

I don't think this is gonna happen in our lifetimes. I think the world that you've described is a beautiful world. My dream, it will exist and I think it'll only exist if the culture changes and the culture changes worldwide and dramatically, and it'll only change if every generation comes along and like just shifts it a little bit and say, I'm of the generation, it was all the money for us, so everybody just wanted money.

It was like, what job gets you the most money? Okay, let's take that job. And I think hopefully I'm gonna raise my children to be a little bit different, to think different, to think what is the most positive impact you can have on this role? How can you be a better custodian of the planet? I think these kind of changes to drive, I think it's gonna take generations.

I really do. And I think it's something we actually do have to do,

Chris Adams: Or hand power to people who are not optimizing for the things that we, you and me might have been optimizing for, perhaps. Yeah. Okay. Okay. Christ, I've gone into politics again.

Asim Hussain: Yeah. We've done it again!

Chris Adams: All right. All right. This is our last question in the mailbag, and I think we might have to have another episode to answer some of these questions if there's interest.

We don't know if there will be. So are there any notable examples of organizations or projects that have successfully implemented green software practices? What can we learn from them? That's the question. And Asim, I'll put this one to you cuz I suspect you've had a few conversations with people doing some of this.

Asim Hussain: Yeah. I think that in terms of things that have been published and have got large enough to have significant impact, there's some of the stuff that happened at Microsoft, so the Carbon Aware Windows is I think one of the bigger implementations of carbon aware computing and I would, now that's a direct line between carbon aware windows to carbon aware Xbox.

I can draw that line for, oh, we need, still need to get those people onto the podcast if we can. So I think those are really great examples. And I, and I was actually Scott Chamberlain, who's my lead now at Intel, he's the one that was driving a lot of the carbon aware. Windows work for ages until he got that out.

And so that's like Windows now does carbon aware updates. I know sounds small, but updates like the type of workloads are very carbon aware and shiftable. So I think that's a really good example. Again, like very similar carbon aware work from Google, like earlier on. That was almost like two, three years ago now.

They did similar stuff with their data center workloads. I think the rest of the work that's happening in this space, there's a lot of work still happening right now in the measurement space. I know there's smaller bits of consultancy work that's going on with the larger companies. I don't know the specifics of, I don't think there's anything that we could re talk about publicly and also that would wow people, cause it's just the, the guts of people, organizations, and the work they do.

But I think to me, those are always two of the big wins in use cases. And it's always interesting me that those are both carbon aware examples because the investment you need to implement carbon awareness on the scale of which you could reach with it in a very short period of time is very impressive.

Chris Adams: Okay. Alright. Thank you Asim. Alright, so that's carbon aware programming. So there are examples that you can point to using carbon aware programming that create measurable figures from this. And that's like ones which are not particularly latency aware, but are convenient that happened in the background, which don't result in a kind of poor user experience, but still deliver some carbon savings, yeah?

Asim Hussain: The one thing I'll add to that I think is that what will drive more of the work in energy efficiency and kind of hardware efficiency in all these other spaces, I think is measurement. I think is ubiquitous, making these things very easy to function. I think though, I think the work that you've done with Co2.js And really making the ability to measure this stuff very, very easy is what's gonna drive a lot of the kind of the next generation of changes, and it probably already has, I just dunno any of the success stories there yet.

There's probably the loads of websites. Yeah. But,

Chris Adams: So actually this is a nice way to talk about, or a nice segue for what I think is interesting and there's, there's some work by company called Sentry Computing. So a talk at Grafana Con about basically how us tracking all the metrics for compute usage meant that we were able to reduce our own usage by X percent.

That was a cool thing in my view. But the reason I'm talking about this is not only has he shared some of that stuff, But he also ended up proposing a kind of proposed, a new HTTP response header specifically to the IETF as an RFC to basically say, this is how HTTP should work. We should bundle these numbers into HTTP so that there is a header for like carbon emissions, like Scope two or something like that.

Now, whether that's the correct number to use is another matter. But I think that's one of the examples of focusing on the efficiency part and the kind of resource usage and energy level. If we use the Green Software Foundation way of thinking about this, which is carbon, uh, and it's hardware efficiency, energy efficiency, those are the things I, I think the ones that are worth looking at.

There's also this other thing that, I dunno if the numbers are big yet, and I would love to get a second opinion, but there's a company called Storj. Which is S T O R J. They are basically making claims about massively reducing the hardware footprint of providing object storage, like S3 style object storage by using loads and loads of unused capacity in data center storage.

Just the same way that Airbnb can use unused capacity in houses and hotels and things like that. And they're basically doing this.

Asim Hussain: unused if they're using it?

Chris Adams: So the idea is that let's say you've got a bun, a loads of service, and they have free space on hard discs, which aren't being used by anyone right now. So they do that and they use this technique called erasure encoding.

So you take it to file, you split it up enough points so that you don't have to replicate the same file like five times if you just replicate enough of the overlapping shards of it. And are they, I'll share the link for that. Cause I think it's interesting. I don't know enough about the, what kind of peer review have you seen for the numbers, but I think it's extremely clever and it's one of the few examples I've seen of people doing something on the hardware efficiency part of green software that I think is cool and it ends up being sub substantially cheaper than using object storage from some of the big providers, for example. So it's, we're talking like around 20% of the price and the figures they say is it's maybe 70% lower carbon footprint for storing a terabyte of data over a year compared to some of the big providers or a data center.

So that's the stuff that I think is interesting, but I need to caveat that with that. I don't have any independent verification of that stuff yet, even though I think it's super duper cool. All right.

Asim Hussain: Chris Adams, always a scientist looking for peer, reviewing his statements; Asim Hussain always like shoots from the hip, whatever stat comes to his mind. There you go.

Chris Adams: Alright, Asim, I think this has taken us till the end of our time. We have allotted for our mailbag episode. I really enjoyed this. So thank you very much for coming on and, uh, we should probably wrap it up and say thank you to everyone for listening and we'll have more of the regular programming with interviews with for more experts coming up in the coming weeks.

All right.

Asim Hussain: Alright,

Chris Adams: Thanks mate. Take care of yourself. Have a lovely week and everyone enjoy. For those of you who do celebrate it in America, happy free from us Brits tomorrow here.

Asim Hussain: Commiserations to the Brits. Go back.

Chris Adams: Alright, leave it like that. Torah.

Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit https://greensoftware.foundation That's Green Software Foundation in any browser. Thanks again and see you in the next episode.

Environment Variables One Year Round Up

29 juni 2023 | 28 min

Join us for a special episode of Environment Variables as we celebrate over a year of bringing you the best insights on Green Software! From open source's role in reducing software emissions to making green changes in organizations, carbon-aware computing, and more, we revisit the most captivating moments from our top 10 most popular episodes. Listen to snippets from our discussions with industry experts and dive into the world of Green Software, as we reflect on our journey, share valuable knowledge from our guests and hosts, and continue to raise awareness about the importance of Green Software.

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Episodes:

How can Open Source Help Reduce Software Emissions? | Ep 4 [1:40]
How do we make Green Changes in Organisations? | Ep 3
The Week in Green Software: Green Software Legislation | Ep 15
The Week in Green Software: Calculating Software Emissions with Navveen Balani & Srini Rakhunathan | Ep 20
Carbon Aware Computing | Ep 2
Fact Check: Sara Bergman & Software Carbon Intensity | Ep 13
Amazon's Customer Carbon Footprint Tool | Ep 1
From Carbon Aware to Carbon Intelligent | Ep 9
Green Networks | Ep 10
Community Clouds and Energy Islands with Dawn Nafus and Laura Watts | Ep 14

Resources:

Building Green Software Book Pre-Release | Anne Currie, Sara Hsu & Sara Bergman.
State of Green Software Survey Report | GSF

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Chris Skipper: Hello and welcome to this special episode of Environment Variables. For those of you who don't know, it's been just over year since the Environment Variables podcast was released, and we thought we would take a look back at the best moments over our first year. This podcast started off as an idea from Asim Hussain.

The director of the Green Software Foundation and has blossomed into a beacon for green software awareness on the internet. Now, just as a disclaimer, you may not have heard my voice before. I am Chris Skipper, and I am the producer of Environment Variables. I don't often get in front of the microphone and would say, I started this journey not knowing a lot about software or its effect on the environment.

However, I now have accumulated this assumed knowledge or full knowledge of a lot of the facts and terminology that surround green software, which has benefited my day-to-day life in the way that I use computers and the internet as a quote unquote layperson. It has also made me a great addition to any pub quiz team, but that's enough about me.

In this episode, we will be looking at the moments of our top 10 most popular episodes of Environment Variables, which started in April of 2022. This is purely based on listener stats, so we thought it would be nice to revisit some of these tidbits For your listening pleasure, as always, links to each of the episodes will be down in the show notes below, or if you want to listen to all of the episodes of Environment Variables.

You can as always visit https://podcast.greensoftware.foundation, preferably after this episode. So to kick us off, our first moment we're going to look at was from an episode entitled, How Can Open Source Help Reduce Software Emissions? Where Asim was joined by Chris Lloyd-Jones, affectionately known as CLJ, or sealjay as is his Twitter handle.

Head of open technologies at Avenade and co-chair of the open source working group at the Green Software Foundation. And Dan Lewis-Toakley Green Cloud Lead at ThoughtWorks and the other co-chair of the open source working group at the Green Software Foundation. Now, in this episode, they discussed the benefits of open source versus closed source, what opensource tools are out there, and how they can help reduce software emissions.

This was the fourth episode of Environment Variables, and the quote here features both of our guests. But with a particular tone of impending doom from CLJ himself.

Dan Lewis-Toakley: It's about engaging the community and growing adoption of this tool and similar tools rather than necessarily holding it tight to our chest and trying to have the secret source that, you know, that we want to provide to our clients or to partners.

So that was a, that's sort of a, a key aspect. Different license types, more restrictive ones can often be a deal breaker for some companies and organizations. To adopt software.

Chris Lloyd-Jones: Yeah. I dunno about you, but sometimes it can also feel like, you know, when you've got a toddler or a child is holding a suite and you have to kind of pry their fingers off it at times for an organization, it can be quite hard to make that decision.

And I say that because we, we've contributed some code as Avenade, the Green Software Foundation, to start that CICD pipeline tooling. But one of the reasons why I push for that is because I also think if you keep that secret source close to your heart, you run the risk of people, first of all, jumping over you because they just want to get things done.

They want to get things done quickly. What you do becoming out of date and stale, then you've invested a whole load of time in something which isn't compatible with what people are now using. You can also just look like a, a bad actor, particularly in fields like this where we're not, I don't know, making a search tool and comparison to like, you know, the whole open, open search, elastic search for raw rate.

We're trying to actually solve a problem here where fundamentally, if the world doesn't go a zero, we might all die in a massive heat test. There's good reasons to do it.

Chris Skipper: Moving on to something a bit lighter in tone, these next two snippets come from our third episode of Environment Variables, which has the title, How Do We Make Green Changes in Organizations? This was the first time EV regular Anne Currie joined us on the podcast. Anne is a tech ethicist at Container Solutions and one of the organizational leads at the Green Software Foundation.

Not only that, but she is a sci-fi author and a non-fiction author who is currently co-writing the new O'Reilly book, Building Green Software with other GSF members, Sara Hsu and Sara Bergman, which is currently in pre-release via the publisher's page, linked below in the show notes. In episode three, Asim and Anne talked about what are the real factors that drive organizations choices around increasing efficiency within their organization?

What needs to happen for senior leaders to make sacrifices for sustainability and can regulation push for real change inside organizations? They also discussed their love of ops people, developers, and the role of middle managers, which is what this first quote from Anne is about.

Anne Currie: There's been quite a lot of psychological thinking, organizational thinking about how do you make changes in organizations and because a lot of it came from changing the finance industry.

Cause it had to really change, uh, after the, after the big crash in 2008, 2009, and they did loads and loads of psychological research and organizational psychological research. And what they found was that really unless middle managers decided to do it, it all stopped. You know, they would stop it going up and they stop it going down, which is where I think the tech conferences are very good because they tend to be attended by people in the middle managers, senior architects.

If you can get them on board, then that's all that's really required. Top down, get stopped by them. Bottom up, get stopped by them.

Chris Skipper: I also wanted to include this next snippet from Asim, as it includes one of his key beliefs about green software, which is that it can only take a few people to make changes within an organization to set them on the right path to making green changes and green choices in relation to software-based decisions.

Asim Hussain: The first time I got the hint that this might not be the right direction or there might be a different direction, was when I, we started talking about regulation, when there starts to be hints of regulation on the horizon. And one thing I realized, I mean, just the conversation, just the thread of regulation opens more doors than anything else. Like for instance, one of the things I'd learned early in my years at, at Microsoft was that you really do have to find customers. You can't just like be waving around going, Hey, I'm really passionate about technology and I know my area and, and you know, like if we were to build this feature, you know, I, I, trust me, trust me, a lot of people would love it that no one's paying attention.

You've gotta come in with like, I've got five customers. They all want this feature. This is how much money they wanna spend. If they get this feature, we should prioritize this. Let's get this prioritized and be, okay. Look, let's, let's do this. That regulation surpassed that, opened the doors, surpassing that fear.

I, I, I used to work in investment banking. And one of the lessons I learned leaving investment banking was there's only two things people really care about. And that's fear, fear and greed. Fear about fear, I think is greater. I think fear and regulation is greater than greed of, of, of money. So I think that's a direction that would really help us out a lot is more regulation in this space.

Chris Skipper: Now leading on to a very serendipitous segue here. Our next snippet comes from an episode titled The Week in Green Software: Green Software Legislation. This was one of our first episodes of a newer format called The Week in Green Software, or TWiGS, which is a news and events roundup of everything going on in the world of green software.

On this episode, Ismael Velasco, who was one of the researchers that helped out with the recent 2023 State of Green Software survey again, links in the show notes below, talks about the dense legislative landscape around green software and technology and energy regulations. Everything from France's Digital environmental footprint reduction legislation to the UK's greening government ICT, and digital services strategy. This quote is about green public purchasing.

Ismael Velasco: The regulatory movement at institutions are driving extends also to what is known as sustainable public procurement, or green public procurement, or green public purchasing, or sustainable public purchasing. And it's the idea that more and more governments are choosing to develop environmental standards for purchasing products, and that movement toward environmental informed procurement is rapidly accelerated around software specifically, and this matters again, how governments choose to purchase counts. Governments around the world spend an estimated 11 trillion in public contracts every year representing approximately 12% of global GDP.

At a national level, this varies even more. So some governments will spend no more than two or 3% of GDP, but other governments purchasing represents up to 57% of GDP in some countries, so how governments choose to purchase has a significant impact.

Chris Skipper: This next quote comes from an episode that featured Navveen Balani of Accenture, also part of the steering committee of the Green Software Foundation and Srini Rakhunathan of Microsoft. Titled The Week in Green Software Calculating Software Emissions. Asim sat down with our guests to take a deep dive into the process behind Accenture's use of the Green Software Foundation software carbon intensity specification to calculate a measure to track and ultimately reduce the carbon emissions of one of its internal reference applications.

If you would like to learn more about the GSF's SCI specification, we have a whole Environment Variables episode on it linked in the show notes below. In this snippet, Srini talks about the use and value of the SCI specification.

Srini Rakhunathan: It's gonna change the way we look at carbon value, or how are we systemically capturing monitoring data? The intent of the project is to be able to provide a value, a carbon emissions value, which you can use. It's more for action for you to continuously iterate and figure out where are you at a particular milestone after you have taken some of the measures provided as part of the SCI specs. Which is around making it more carbon aware, making it more efficient or energy efficient or all of it.

So you need a way to tell whether you have progressively made it past your different milestones, whether you're continuously reducing or you're stagnant, or you're increasing because it's always possible that you need to pull all these parameters to make sure, because we are not building applications just to make it sustainable, right? We are building applications to make money for your business.

Asim Hussain: Add value to the world, let's put it that way.

Srini Rakhunathan: Exactly. And so you need a way to easily calculate across your different hosting infrastructure, whether you do it on the cloud or on-prem. You host your app on your laptop. The project aims to tap into the different data sets available and, abstract away the calculation algorithm and just provide you a value, most intelligent value. That's what we would say when we were kickstarted it, and I think it's going good. We should probably have something really cool coming out of this.

Chris Skipper: Now, this next episode definitely increased my green software vocabulary with a particular focus on terms such as time shifting, location shifting, curtailment, and many other terms related to carbon aware computing, which happens to be the title of said episode. Here, Asim sat down with guests, Scott Chamberlain, formerly of Microsoft and Henry Richardson, of Watttime they talked about how we can build sustainable software that reduces the impact on the environment and how these decisions may just lie in the hands of the developers instead of the CSR teams.

So if you were as confused about those terms as I was, I suggest you listen to this episode in full. But to give you a slice of just how these work, here's a snippet from our guests.

Henry Richardson: So this is a really interesting disconnect that we're seeing right now. And as especially in the near future, load flexibility will have a lot of emission savings potential cause we'll be able to shift out of those dirty periods into the curtailment periods.

But once we eventually attain those a hundred percent or near a hundred percent clean grids, the flexibility won't be saving emissions directly, but it will be enabling a hundred percent clean grid because you'll be following wind and solar. And if we didn't have that flexibility, we would have to do fossil resources. So like it's an essential piece of a clean future grid. But it's gonna be harder to quantify the benefit of it in the future.

Scott Chamberlin: That, that's a great way of putting it, Henry. It's, it's, we get to clean grids faster the more we have carbon awareness, because carbon awareness allows us to maximize the use of our renewables, whereas today, we're already curtailing them.

Right. I think that's, that's an excellent way of putting that.

Henry Richardson: Exactly.

Chris Skipper: And just over the halfway point, speak of the devil. Our next quote comes from that SCI episode. It's an episode of Fact Check with Sara Bergman and Software Carbon Intensity. Fact check is another format of episode on Environment Variables, where we take a deeper dive into the bigger questions in a one-to-one discussion with a special guest.

In this instance, Sara Bergman, senior software engineer at Microsoft and an individual contributor to the green software foundation's software, carbon intensity project discusses not only her own green software journey, but the software carbon intensity ISO standard, why it excludes carbon offsets and fact checking what that tells us about offset based green software claims from Google to blockchain.

Sara Bergman: In an ideal world, elimination and offsetting would be the same. We are not in an ideal world where in we don't have perfect technology, so they are not the same. If you eliminate something, it means you never emitted it. It stays in the ground, doesn't go up in the atmosphere. However, if you offset, there are several different ways of doing that. One's most talked about is forestation. So you plant trees, which in itself is great. It's good for biodiversity as well. It's good for oxygen that we breathe, uh, but there are a number of problems surrounding this. There has been reports of projects where trees were planted, for example, and then later they were deforested anyway. There are also other studies that should suggest that there isn't enough space to plant the amount of trees we would need to offset all of the emissions considering the rate emissions are growing at.

So there are a bunch of questions there. Same with something that more talked about recently as the carbon captures. It's like a giant vacuum that sort of sucks carbon directly out of the atmosphere. There are very few functioning examples of this, and they are extremely costly and it is betting our future on a technology which isn't really mature enough to hold up to this promise. I mean, hopefully it will be, but I think it's a dangerous bet.

Chris Skipper: We now go all the way back to April 11th, 2022 when our very first episode of EV was published. Amazon's customer carbon footprint tool. In this episode, Asim was joined by guests Chris Adams, now host of Environment Variables, Sara Bergman, and Danielle Erickson. As they discuss the impact that Amazon's customer carbon footprint tool is having on the green software landscape, how do services like AWS affect climate change and what are the effects on the environment of these huge data centers?

We also learned about how you can use your heat from greenhouses to grow tomatoes. In this particular snippet, we actually have a link to our last quote about the SCI standard, as Asim talks about how the open source nature of Amazon's customer carbon footprint tool relates to this.

Asim Hussain: And I especially like the fact that because cloud carbon footprint is open source, not only is your methodology public, but your data.

And the, the underlying data assumptions at a very low granular level are public. I can see what is the energy consum if I'm using this particular server, this particular load, that data is public and we're actually using that in the foundation, in the software carbon intensity standard. Where, where you leveraging that data cuz it helps engineers kind of calculate the carbon emissions of, you know, processes or estimate the carbon emissions of processes so they can then make those kinds of decisions. So it's kind of the openness of the data is I think, also missing with these tools. But I've also heard it's extremely difficult for Amazon and Google and Microsoft to make this data public.

And it's not only revealing competitive information. There might also be legal constraints. You know, if you reveal some of this information, you, the SEC might come after you because you're revealing proprietary information. There's actually lots of complications around that from what I've heard.

Chris Skipper: Next up, we have a pair of quotes from episode nine of Environment Variables titled From Carbon Aware to Carbon Intelligent. In this episode, Chris Adams is joined by Colleen Josephson of VMware Philipp Wiesner of TU Berlin, and EV regular Sara Bergman, as they discuss the opportunities with making first carbon aware and then carbon intelligent computing. This episode featured a true plethora of terms, variability, curtailment, disaggregation, 5g, 6g, delay tolerant networks, intermittent computing, IoT, and even a short segue about Raspberry Pi's, or make an appearance in this action packed episode. In this first quote, Philipp explains variability.

Philipp Wiesner: Variability can be quite dramatic. So in France for example, they have clean, not clean maybe, but like low-carbon energy because of all the nuclear power they are deploying throughout the day. So there you have barely any potential. But then there's regions like Germany for example, which are very interesting because they're super variable like Germany employees comparably much wind power as well as solar power. So at many times of the day, they manage to have large fractions of the grid provided by green energy. But if neither sand nor wind available, we burn brown coal, which is pretty much the dirtiest of all fossil fuels. This is why variability is really crazy. Like within a normal day, you can expect twice or like 50% fluctuations.

That could be that one kilowatt hour that you consume now is twice as dirty if you consume the same kilowatt hour a few hours later and within a few days, you can even see like the difference between the min and the max can be factor four or something. So one kilowatt hour can really vary from 100 grams CO2 up to 400 grams or or more 500 grams. Yeah.

Chris Skipper: And following up, we have one of the most memorable quotes from Colleen.

Colleen Josephson: The same thing comes in with upgrades. So if you look at upgrading hardware in data centers or telecommunications hardware, so 5G I think, got some bad press for how much power the base stations consume. But what's actually true about it is that the power consumed per bit transmitted has gone down significantly. So there's a good advantage to upgrading your hardware, but then you know, what about this hardware you're getting rid of everything that we produce has this concept of embodied emissions. It takes resources and carbon to produce this hardware, so you have to really carefully look at that sort of trade off.

It turns out that keeping our devices, especially smaller devices in use for as long as possible, is one of the greenest things that we can do.

Chris Skipper: Yes, so keeping smaller devices in use for as long as possible is one of the greenest things we can do as individuals. But what about bigger entities, corporations, organizations, or even networks?

That's where our next quote comes from In an episode entitled Green Networks, episode 10 of Environment Variables. Host Chris is joined by Eve Schooler, principal engineer and director of Emerging IoT Networks at Intel and Romain Jacob of ETH Zurich. They discuss how can we reduce the energy produced by networks? How could we leverage current research to make the internet more energy efficient? Two very big questions of which one answer is provided in this next snippet from our guests.

Romain Jacob: What if we were to redesign indeed those wired networks? So that reliability is not something we get rid of, but we modulate the requirements we set there and say, reliability is just one objective.

How much performance degradation are we willing to tolerate in order to save on energy? To give a very concrete and simple example, most traffic on the internet is driven by human activity, right? And human activity has a very serious seasonal pattern. We use the networks more at certain time of the day than not at others.

It's very easy to think that we could turn off part of this networks for certain parts of the day, because we don't need that much bandwidth, and if we do, we might be able to tolerate a bit more delay than at peak hours. It's very similar to turning off the public lights on the streets, you know, at night when nobody's driving.

Right. It's the same principle.

Eve Schooler: Or even in your home, right? The analogy of one's parents growing up, don't forget to turn off the lights. It's exactly the same analogy.

Romain Jacob: Yeah. It's the same idea, right? And there is no reason this cannot be done. We know we can do it. The question is how far can we push it? One limitation factor, one blocking factor at the moment is how quickly we can turn things on and off, right? Because switching on a router or switch takes as of today in the orders of several minutes, right? So it's not something that you can just do multiple times for hours or so. Because essentially your network will be completely inoperable. It can be changed. If we were to change the hardware, if we want to change the operating systems we run on those machines, we could improve on that. How far can we go? This is kind of an open research question at the moment.

Chris Skipper: We have come to our final episode on this list, which is Community Clouds and Energy Islands. In this episode, host Chris is joined by Dawn Nafus of Intel and Laura Watts of the University of Edinburgh. As they explore community clouds, data centers, energy regulation, and projects on the Islands of Orkney.

One of the key talking points in this episode was how we've moved from a more decentralized internet running on centralized power to a more centralized internet, running on more decentralized power. Is this the only computing model of the future? What could a decentralized internet running on decentralized power look like?

We see hints of what this looks like at the edge of the internet, but also the edge of the grid, which is what this first quote from Dawn is about.

Dawn Nafus: One of the things that's really been heating up right now on social media, you might imagine is with the recent changes, shall we say to Twitter. There are a lot of folks like myself who have moved over to Mastodon.

On Mastodon, we've been having a rip roaring conversation about what would it take to actually stand up a Mastodon server in a place like Orkney where stuff is in fact community run and where there actually is community benefit to how the energy actually works and how it's organized. And there are a million challenges to that that we can talk about, but that's that next step where, once you get beyond scheduling right, you can start to think about all these other social implications that are far deeper than just, you know, writing some scheduling code.

Chris Skipper: I also wanted to include this quote about regulations from Laura.

Laura Watts: Regulation is one of the biggest challenges to what we're talking about for energy and data, and thinking about how we do things like the ecovisor, and that is manage assets because you need to be able to have permission from the regulator to basically switch these things on and off or be able to have any impact on the grid because keeping the lights on is an absolute commitment. So if you are going to start changing the load, your data's gonna get into this space and thinking, how do we write code for using different amounts of energy sources that's gonna change the load on the grid?

And that starts getting into regulatory issues. And it seems like a dull thing, but actually it's a really important space to start talking about, because we can have huge impacts on what the grid looks like. What does a data electricity grid combined look like in the future? That's a regulatory and governance question. As much as it is a technical, how do we shov the data about and change what the software looks like?

Chris Skipper: So we have come to the end of this episode of Environment Variables. I hope you enjoyed this year roundup. So what have we learned from the first year of the Environment Variables podcast? Well, it is safe to say that when we started this podcast, there weren't that many stories on mainstream news sites that mentioned green software.

However, we are now able to produce not only a weekly episode that covers news topics related to green software. But we have a fantastic newsletter with amazing resources that will keep you up to date with the latest and greatest regarding green software that goes out weekly. Link in the show notes below.

We also learned that despite all the real and heavy facts regarding carbon emissions of software, there is even realer and heavier optimism in the community. So here's a great big thank you to you, the listener, for continuing to support this podcast. As I mentioned before, you can find links to all the shows in this episode below or alternatively, you can visit https://podcast.greensoftware.foundation to listen to all of the episodes of Environment Variables. You can also find out more about the Green Software Foundation, including resources and tools related to green software at https://greensoftware.foundation, that's https://greensoftware.foundation In any browser.

Thanks again for listening, and we'll see you on the next episode when we'll be having a very special episode of Environment Variables again. So keep an eye out for that. Bye for now.

The State of Green Software Survey with Tamara Kneese

21 juni 2023 | 44 min

In this episode of Environment Variables, we cover the State of Green Software Report with the Green Software Foundation’s very own lead researcher Tamara Kneese. She and host Chris Adams delve into the insights from the report including key topics such as the carbon footprint of crypto mining, regulations for generative AI, and the role of consulting firms in shaping emerging technologies. They also discuss how the results highlighted the impact of the tech industry, AI sustainability, and the need for responsible innovation. To find out just how interesting the results of the survey have been and everything in between tune in now!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Tamara Kneese: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Key Talking Points:

State of Green Software Survey Report | GSF [3:28]
Measuring carbon emissions is crucial for scaling sustainable AI | SOGS Survey Report [15:15]
Web3 rings alarm for green software practitioners | SOGS Survey Report [23:13]
Software legislation has quadrupled in the last decade | SOGS Survey Report [32:32]
New Green Software Developments in the US | Blueprint for an AI Bill of Rights | OSTP | The White House [37:15]

Resources:

Responsible AI is green AI | SOGS Survey Report [13:43]
Decarbonization alone cannot make software green | SOGS Survey Report [13:43]
Trustworthy and Responsible AI | National Institute for Standards and Technology [37:58]
Amazon’s Antitrust Paradox | Yale Law Journal [38:54]
Amazon workers press company on climate change response after Pakistan floods | NBC [39:05]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Tamara Kneese: I think there's much more of a sense of urgency from policymakers, from advocates, from activists, from researchers, scientists who are looking at the numbers and looking at the data and thinking, we really need to do something now. And I think it's all part of the same general wave of really trying to find a way to live with tech in a way that is more sustainable, in a holistic sense, in a way that is actually good for people and for the planet.

I'm your host, Chris Adams.

Hello, and welcome to another episode of Environment Variables, the podcast where we explore the world of sustainable software development. I'm your host, Chris Adams, and today we have a fascinating conversation lined up for you. Joining us today is Tamara Kneese, the lead researcher at the Green Software Foundation.

And today we'll be diving deep into her involvement in the groundbreaking state of green software report and at the Green Software survey for 2023. This survey aims to shed light on the current landscape of green software and the challenges and opportunities it presents. In this episode, we'll explore tomorrow's role in designing the survey, her research findings, and the implications for the future of sustainable software development.

We'll dive into the various insights provided by the survey that includes in identification of specialist green software tools, the progress made in decarbonizing software, and the potential of software to contribute to renewable energy infrastructure. So whether you're a software developer or an environmental enthusiast, or simply curious about the role of technology in shaping our planet's future, this episode is for you.

So that's me being talking for a while. Let's welcome Tamara Kneese to Environment Variables. Tamara, the floor is yours. Please introduce yourself.

Tamara Kneese: Thank you, Chris. So I'm Tamara Kneese and I'm the lead researcher at the Green Software Foundation. Very excited to be here with you. I'm an academic with a background in media studies, science and technology studies and gender studies. And now I'm about to start a new role as the project director of Data and Society's newly launched Algorithmic Impact Methods Lab.

Chris Adams: Wow. Congrats. That's a new development. This is how I spoke and that's, oh, I'm really glad to hear that actually.

Tamara Kneese: Thank you.

Chris Adams: Cool. You're welcome. Yeah. Okay. So if you have not listened to this podcast before, my name is Chris Adams. I am the executive director at the Green Web Foundation, and I work at the Green Software Foundation as the co-chair of the policy working group, which is where I met Tamara, and where we end up seeing each other on a more or less biweekly net basis. All right. Thank you for introducing yourself, Tamara. Before we dive into the State of Green software survey, I just wanna check anything we talk about will do our best to share links in the show notes like we do on a regular basis. And I think with that, we should probably go get into it, shall we Tamara?

Tamara Kneese: Yeah.

Chris Adams: All right. Okay, so the first thing, for people who've who are new to the whole Green Software report and the accompanying survey, could you maybe just give us an overview of the survey and the objectives please?

Tamara Kneese: Yeah, so I think it's worth noting that the entire process was actually quite lengthy. So we started doing preliminary survey design and other kind of necessary background research and work in Q3 of 2022. And we really were collaborating across companies and fields of expertise and knowledge. So basically a group of us who were part of the policy working group at Green Software Foundation came up with 25 questions that we thought captured really key information like demographics and other data to understand the level of awareness and adoption, points of friction, and also green software enablers from the perspective of software developers. So we were trying to go beyond the the surveys that are out there already of C-Suite and venture capitalists, and we were really trying to understand what was happening on the ground.

Because in order to make green software successful and scalable, we have to meet developers where they are. And we also wanted to leave some room for developer feedback, and I personally found this to be the most exciting and fascinating part of our findings because developers wrote in their own ideas and their own experiences and even frustrations, and through their feedback, we can really better understand the gaps between leadership or corporate goals and what developers are actually seeing and doing on the ground, and hopefully we can help to bridge that gap.

Chris Adams: And as I understand it, there was a fairly decent size chunk of the survey, went out to quite a few people and you had a substantial number of responses. That's my understanding of it, right?

Tamara Kneese: Yeah, so we had over 2000 responses in total and people from all over the world, which is great. And we also heard from developers from a broad range of companies. So we heard from people from the large companies like Microsoft and Intel and also from smaller companies and startups, and we actually reached a few people who are outside of our own core membership of Green Software Foundation.

We reached people through LinkedIn and through social media, even through Mastodon, which is lovely. So, And so people found us in a variety of different ways. So we really had a great range. And in order to kind of contextualize the survey findings, we did a lot of background research. And so while a lot of the insights are prompted by the survey results, I was also looking at the latest research reports from firms like McKinsey.

And also at leading academic journals, really sifting through what is the cutting edge kind of research on green software in the present moment. And we also did some background interviews and conversations with thought leaders from across the green software space.

Chris Adams: Okay, so I remember this being quite an undertaking actually, and uh, there was a number of people involved in this actually, who was, who involved, I remember being on a series of the week of the weeks, maybe it might be a nice time to just quickly name check some of the people. I think it was Zane who was involved on some of this and I believe shout out to Ismael as well, who's also doing some work as one of the researchers on some of this as well as basically various other people inside the policy working group. You said something interesting about the fact that you had like a decent number of people. That's a lot of free text to be like wading through, right? For all the kind of questions you had from there surely?

Tamara Kneese: Yeah, so that was really quite interesting because analyzing the survey results, particularly when you have a lot of write-in material to wade through for 2000 people and clearly not everybody wrote in lengthy comments. Some people just filled out the survey questions and left sections blank, but it was still quite a bit of material and it was really interesting because it provided more sort of nuance and context that kind of, and some people actually almost included a meta commentary on why they filled in results the way they did. And they were like, I'm answering this question with my organization's priorities, not necessarily my own. Just to really provide a little bit more insight into why people were answering questions the way that they were.

And yeah, it, this was also, like I said, deeply collaborative. So Zane was my graduate intern at Intel and he helped with the survey design. Ismael was very much involved with carrying out a lot of the desk research and also with the survey design process. And I believe that he also did interview a number of green software thought leaders as well.

And then we had feedback from a number of people on who were part of the policy working group. Chris, you were there, and Lisa and Elise, and of course Asim and many others was providing feedback all along the way.

Chris Adams: Okay, thank you. And maybe it's briefly worth just talking a little bit about the goals, what we were trying to achieve for this, cuz it does seem a little bit contrived to me asking this cuz I remember writing one of the proposals to try and get some of this involved. But I know that this was something that you had to take and run with quite a lot actually.

So maybe if you just might expand on some of the particular things we were trying to shoot for this, then I might be able to chime in on some of it before we start jumping into the meat of the report and some of the real particular interesting insights.

Tamara Kneese: Yeah, absolutely. So one of the main goals was really to raise the profile of green software, and I was really interested to see the percentage of developers who actually had some degree of awareness. And so it makes sense that a lot of the people who filled out our survey already were somewhat aware and already interested.

Although there were a number of people who replied in the comments that this was the very first time they had been exposed to green software. And so by putting out this public report that can be taken up by the press, that can be taken up by policymakers, that can be taken up by academic researchers.

It is a way of really, getting the word out about green software, thinking about the report as a mechanism for evangelizing green software is really part of what we wanted to do. And we also wanted to understand after knowing that 92% of developers we surveyed said that they were concerned about climate change and wanna do something about it.

So what do they need to actually make that happen? What resources, tools, and other forms of support do they need to take action? And another key element of this is reaching out to ICT industry leaders to the C-Suite who really wanna know how and why they should make green software part of their organization.

And really trying to emphasize the business case for green software from their perspective was another really key part of this survey.

Chris Adams: Yeah, I remember actually that being one of the things that we were speaking about, cuz one, one of the kind of key. The reasons that we would try to actually get some funding and add a time and money to actually do this research was that it just made it we want to make it quite a bit easier to have conversations with people or give something just a few links that you can use when you're making an argument and like you can see some of this manifest in the shape of the report.

So typically when you might have a report, you might have maybe 20 to maybe a hundred pages that you might like slap on someone's table or something, but whether people actually read that or actually engage with the content is another matter. And this is probably reflected in the design, the fact that rather than having a long narrative that you'd run people through, it was made up of a series of like smaller insights that you could share for this.

And this is one thing that was, I think it was, it was a decision made quite early on, but it's still quite a lot of content. Like for 30 insights, that was quite a high word count by the looks of things. Maybe you could expand on that part there before we dive into one or two of the insights that caught your eye and my eye.

Tamara Kneese: Yeah, at some point I started doing a word count assessment looking at all of our various Google Doc collection that we had created. And yeah, in total there were around 13,000 words, which is a pretty substantial report and pretty much equivalent to what a standard sort of report would be from something like McKinsey.

And I think the trick was that we really wanted it to be digestible. And so breaking it up into smaller segments and making the website a bit more interactive, which was as them's idea, by the way. And also huge shout out to Osama, the Web developer and also our project managers for keeping track of all of this.

So both Anita and Oleg did a huge amount of work and is managing expectations and making sure that we were able to get all of this content into segments that were understandable. And Namrata also had a huge hand in editing the insights to make them more public facing. So she had a really great eye for figuring out how to frame particular insights to make them appealing to different kinds of audiences.

So for trying to reach somebody in the tech press, if we're trying to really appeal to C-Suite, if we're trying to reach developers, versus academics. What are the kind of key words and terms of phrase that really appeal to different groups? So we were thinking very much about audience as we designed this as well, so,

Chris Adams: Okay, cool. For anyone who is curious, you can visit it at https://stateof.greensoftware.foundation. But don't do that just yet. Cause we're about to jump into talking about some of the particular. Insights so you can hear about it directly from the source, as it were. All right, Tamara, if there's a particular insight that you would direct people to first, which one would you suggest or which one is like the ones that you found found most interesting to, to look into, for example,

Tamara Kneese: So I thought the ones that we included around both responsible AI and about the need for measuring carbon emissions as part of scaling sustainable AI, and then also finally our insight around decarbonization not being enough and really thinking about environmental impacts and social impacts and tandem with decarbonization goals is something that I thought was really important.

And it's worth noting that when we created this survey and when we sent it out, it was right on the cusp of ChatGPT and all of the hype around generative AI. So we, we didn't have any questions directly asking people about that, but from the comments you could see it beginning to percolate and there were a few references also to Web three and blockchain.

What has been really interesting for me is watching as AI has really come to replace crypto in a lot of the discourses around the environmental toll of information technology and a lot of the conversations that we heard about NFTs and their environmental impact. Now we're hearing about the environmental impact of ChatGPT and other generative AI.

Chris Adams: And you see the same patterns with influencers in LinkedIn. Everyone's just switched. I'm an expert in metaverse and crypto to, I'm an expert in AI now. Yeah, so you do see that actually. Okay. Maybe we could dive into one of the particular insights here, the insight that that was a takeaway, and I'll just link specifically to the URL in the show notes.

Measuring carbon emissions is crucial for scaling sustainable AI. Now this one here, we, you were talking about the fact that the world has moved. You know, a lot has happened since we saw this, and I couldn't help notice that your new role is actually to look at a bunch of this stuff. Are there any things that you, dr- what, what do you think has been the biggest change since the initial research and what you're seeing maybe in. June, 2023. Cuz in my view there's a few, there's a number of really significant changes. We've spoke about how this influx of open source models has come in, but also there's a real push for transparency in a way that there hasn't been.

But also you see some of this regulat regulation like taking place or kind of taking shape. Now maybe you could like elaborate on some of that stuff actually.

Tamara Kneese: Yeah, so clearly we have regulation coming out of the EU especially, uh, that is very much looking at things like the carbon footprint of different models, but also thinking about social factors and relationship to things like emissions or water use. And what's funny is, so in one of the insights, the one that you just pointed to reference research by Hugging Face, and they were looking at.

The life cycle analysis of their bloom model, which by the way is still way less impactful from a carbon perspective than other models. But really thinking about it, not just in terms of training and the emissions connected to the training itself, which is often how this is calculated, but also looking at all of the emissions tied to the manufacturing process for the equipment.

Uh, that facilitates the training and production of AI and also thinking about deployment. What are the emissions connected to use in the real world? And I think that this is going to become more of a standard. So as regulation catches up to where innovation and technological production are going, I think we'll see much more of a demand both on the part of regulatory bodies, but also from consumers, and presumably also from developers that really want to know the numbers and who want to know what the true impact of these technologies will be.

Chris Adams: I'm really late to the party and I just spent this weekend looking at some of these, look at some of these tools for the first time in a while. I know that Hugging Face, they have got their own kind of equivalent. Something like ChatGPT with the idea being that you can swap in different models. But the thing that I realized, and I figured it, My, I could ask, while you're here I'm gonna ask you. I really don't know what to, what even units I would use when talking about this. Like for example, when you talk about a website, people talk about maybe the carbon footprint per gi- megabytes sent over the wire and that has all kinds of issues in its own right. But I dunno what I would even use when I'm talking about this.

For things like say AI, right? Do people measure carbon on a per token basis or per question basis? This is something that I'm out at sea right now cause it feels, I dunno if it's well understood enough to even have a unit yet or something representing this functional unit that you might use to talk about this.

Tamara Kneese: Yeah, I, I haven't seen anyone come up with a sort of industry-wide standard yet for how to even talk about measuring uh, impact in this way. And so that will actually be quite interesting to see if the regulatory push forces the industry to come up with better standards because that, that's been the way things have worked in general in terms of calculating emissions and thinking about scope three emissions and all the things that were always tricky and very hard to measure. And I think especially because we don't fully know how generative AI will be used. So obviously there's been a lot of sort of movement towards incorporating different models into things like search engines,

Chris Adams: Yeah. Okay.

Tamara Kneese: um, and so, yeah.

Chris Adams: alright, so with that, so maybe we don't end up with a particular unit, we just see like an uplift in the same way that, let's say you've got like a coal fired power station and you're gonna make that carbon capture and storage using like a third of it. You just have a multiplier. Maybe it's something like that cuz this is one thing that I believe was referenced pre previously. The idea that, let's say, and like you just mentioned just now, about integrating LLMs into search. The fact that I'm just speaking to a robot rather than actually seeing the underlying sources, that is, that changes my relationship to the data that I'm actually able to access, for example.

But there is often an uplift in, for example, if I'm using say, an LLM Enabled search compared to a regular search, there's gonna be a multiply of maybe 1, 2, 3, some kind of figure. I know that I think the wide has an article that we've linked, which has a fivefold increase, but it could be, yeah. We basically don't have the numbers for this yet.

Maybe that's the way we talk about it. Hm.

Tamara Kneese: Yeah. And that I, yeah, the, that is the number that I believe we even linked to it in one of the insights as well. But it is interesting when we start trying to quantify these things, and that's another interesting connection to what was happening with figuring out, say, Bitcoin emissions, where you would often find articles saying the Bitcoin mining industry is equivalent to all of the emissions of X country or Y country and really trying to figure out what the true number is can be, can be a bit tricky. And, and then there's also maybe a point at which nobody even cares anymore. Does calculating the exact number, does it help in terms of, uh, making people, yeah.

Chris Adams: is it good or bad? Because this is, you're right, this is the thing that comes up again and again when you've seen numbers used, I don't know, I've done this and I've done, and it's created 10 tons of impact. That does beg the question. So what is the acceptable amount of impact that you do want to actually have in order to actually, uh, benefit from whatever is being provided here?

And that question there is seemed to be a very difficult question that. I think it's very hard to engage with when you see these discussions a lot of the time, cuz a lot of the time we don't necessarily have the literacy outside of technology to even make a comparison.

Tamara Kneese: Yeah, totally. And then that was another interesting question that came up too, in terms of balancing different kinds of considerations, especially when it comes to things like AI that might take a massive amount of energy, but also a massive amount of water. How, if you're trying to optimize for less water usage, it might actually conflict with your attempt to optimize for less energy

Chris Adams: and carbon. Yeah.

Tamara Kneese: And how do you balance that and how do you make a call about what should actually be prioritized?

Chris Adams: This is especially the case when you think about the supply chain, right? Because let's say your example here, I'm going to not use something which is using water locally, and as a result, I'm gonna use a lot more energy to actually account for that because for basically the largest use of water in most countries is actually coming from cooling.

It's actually from the actual thermal generation, like burning fossil fuels to heat up water, to turn a turbine, to actually generate the power. You end up in many cases, just moving some of the wa- usage into other places. This is one of the things that is, one of the takeaways I found was actually it's very much about the locality of where it's taking place.

So if you are in a city and you have a data center which is ha- is using lots and lots of water and it's pulling from an aquifer that everyone is drinking from. Yes, that's gonna have an impact in the same way that it's gonna impact, say the cost of electricity, or it can impact the cost of electricity compared to something happening further out.

If you're bringing in power from say, across a national border or something like that. This is why, I guess it's, in many cases it can be quite complicated and why I'm glad there is an actual report exploring some of this. So maybe I wanna ask you, we spoke a little bit about AI and I suspect we might come back to that again cuz it's the topic du jour.

There was also another insight here talking about Web three rings, the alarm for green software practitioners. Now I figured it'd be good to ask you about this cuz you have done a lot of research and this, so you authored another report around this. So you've got like some form and some, some background on this.

So maybe you could explain, maybe expand on this one here cuz this one caught my eye and I think there's a few interesting talking points in this one too.

Tamara Kneese: Yeah. And with crypto we did see a lot of legislation that was very much concerned about the carbon impact of Bitcoin mining particularly, but also of crypto mining in general. And it is interesting that with Ethereum, the shift to proof of stake and really lessening the carbon impact by over 99%. This was prompted, of course, by a lot of the public opinion and news stories about the harmful effects of NFTs on the environment and how can you claim to be creating this kind of technology for good that will lead to empowerment of marginalized communities and decentralized payment structures if you're also responsible for, all of these carbon emissions? And what's interesting is that as certain countries or states banned crypto mining, the crypto industry moved into different locations, they found workarounds. They partnered with existing energy companies and infrastructures. And so with generative AI in particular, it will be very interesting to see what effects the new wave of regulation has on practices. So how will companies find workarounds? How will, how will they change their practices to accommodate new laws and regulations?

But will it change the landscape of how, you know, development is happening?

Chris Adams: That's a good point actually, cuz when you think about, just like from a really kind of operational level, one of the reasons people have been talking about some kind of proof of work, cryptocurrency, like proof of work mining was the idea that yes, it loads of uses lots and lots of power. But I can turn it off and it's really plausible.

That's the idea. And there are lots and lots of parallels to essentially the extensive machine learning you might use for training. That's not something which is particularly latency sensitive. So you can see a lot of the same ideas being applied to this. So maybe you do see something like this. Maybe that is actually a new role we have for, I guess like may, maybe you can end up seeing AI just move directly into that kind of slot as you suggest actually.

And I guess maybe it might be worth actually talking about the role that consulting firms play in the role of this. Cuz if 12 months ago we had the whole thing about how the Metaverse and Web three were gonna be like the next big thing, then they turn out not to be the next thing thing. And now they are.

And now they're not. And then maybe this is one thing that's, we might. Look into the roles for this, cuz like you said, there was a real push now and I wonder if we're going to see a flip again now that we've seen like Apple release, another take on something which might be like a metaverse after we've seen Facebook burn through literally tens of billions of dollars on this stuff.

Tamara Kneese: Yeah, absolutely. Uh, I think it's really interesting that McKinsey in particular had these reports on the metaverse and also on crypto more generally, and how they would transform the industry and they would be worth between. X trillion and y trillion dollars. You know, it was, it's always in the trillion range and nobody really knows how these numbers are calculated or where they come from, but there's a lot of enthusiasm and I do think it creates the environment for a lot of kind of poorly thought out businesses, decisions to be made.

And so what I noticed with Web three, particularly among a lot of large tech companies, is that they were all trying to keep up with each other in connection to crypto, and they were like, we don't really totally understand this thing. But we think it might be important. There is clearly a lot of money in it, and so how do we follow the headwinds and make sure that we are keeping up on this new crypto thing as much as possible?

How do we talk about Web three or decentralization within the context of the enterprise? Then suddenly after the crypto crash and all the various scandals, the collapse of Silicon Valley Bank, there's a moment of disavow or denial like, ah, we're done with that. Toss it out. We don't need to invest in that anymore.

And you know, that's not great if you're in business decisions based on McKinsey's honestly, very dubious speculations about how much money a particular industry will be worth in the future. I think we were all hit over the head with the metaverse as a trend that we should all really pay attention to and care about, despite the fact that it was really just meta pushing this very particular vision of what the metaverse would be.

And maybe I'm speaking too frankly here, but I do think the problem that I'm seeing right now with the AI landscape is that companies are doing the same thing, and so they're all trying to out compete each other and make sure that they're hiring a lot of people who are AI experts and really making a push for AI to be the next big thing.

McKinsey also agrees that AI will be the next big thing, but we don't know yet what the uptake will actually be or what it will look like in practice. And I think with Web three, one of the largest issues is that there were just weren't very many use cases that made a lot of sense. And things like the carbon offset market, or one place where Web three really had a hold and as we saw from various news reports and studies,

Chris Adams: There wasn't much there. There.

Tamara Kneese: There that was all fluff too. And most of those offsets were not valid and even attempts to create standards in the space did not necessarily protect against fraud. And so the worry is that with some of the potential uses for generative ai, perhaps.

They will not play out as McKinsey or other consultancy firms are saying right now. And perhaps there's a different way that they will be used, but we don't yet know. And so I think a lot of the kind of push towards financial speculation and investment could lead companies down a bad road if they overinvest in things that are actually not going to pan out.

Whereas we do know that regulation is on the horizon. We do know that. Things like ESG are really prompting, not just consumers, but also investors and shareholders to take more activist stances on things like the environment and so I think paying attention to the things that are real is actually more important than trying to imagine some sort of speculative future.

Chris Adams: So more science and less tech bro FOMO, right? Yeah. Okay, great. That's actually, that's a, that's, thank you for sharing that actually, cuz this is a nice segue into, I was, did wanna have a go whole discussion and point everyone to this piece about McKinsey and company and AI and the, there's a really fantastic piece by Ted Chiang Chiang who spokes about talking about the idea of essentially AI as the next, as taking, filling in the same role that a management consulting firm might actually come in.

So you'll come in to help you in many ways justify some of the decisions you might be making anyway, or to help pro help support some of your priors that you might actually have. So if you have tech mo, tech bro FOMO about LLMs, then there's, it's really helpful to find someone to say, yeah, that's gonna be the case.

Cuz if you, cuz we can see some of these elsewhere, but I feel that some of the things with LLMs and should we just call it applied statistics? Cuz in many ways that's how a lot of it does actually feel like it's, do you know how like you have like autocorrect on your type on your phone and you just press buttons randomly you'll come up with sentences? I don't wanna call it ice spicy or a spicy autocorrect. Cause that feels like it's, that's not particularly fair, but I think there is some truth between these two things of having this being this biblical new techno technology and something which is somewhere else. And you spoke a little about regulation coming in, cuz this feels like it's one of the big drivers right now and this is one thing that was actually touched on in the report, a particular thing we're saying legislation is quadrupled in the last decade. So like this big swell of new laws and I guess a new enthusiasm for civil society to be part of this discussion rather than it being just accepting what's coming in.

Maybe you could expand on that a bit more actually tomorrow.

Tamara Kneese: Yeah, I think in general the tech industry, obviously really benefited from the pandemic and a lot of, so companies that were already doing really well started to do even better. But for a long time now, the tech industry has really dominated the economy in a lot of ways. They. Tech dominates the stock market.

They have an outsized impact on the economy and they've become a major seat of political power. But regulation really has not matched the pace of change within the tech industry and how quickly the tech industry has really become a very powerful entity in a global context and I think that what we're seeing with regulation right now is an attempt to rebalance and figure out how can we keep the things that we like about what technology affords us, but how do we make sure that it's not doing anything really harmful that wasn't really anticipated when these companies started? And so thinking about one obvious example would be Meta's influence on political elections and the spread of propaganda that can lead also to violence. And so beyond misinformation or disinformation, but also thinking about real world effects.

Chris Adams: referring to Burma and Rohingya people. Some, some of the, the violence. Yeah. Okay. I see.

Tamara Kneese: Yeah. And so thinking about the ways that companies that are built for one particular purpose then have an effect in ways that were not really an intended output of the technology itself. And so, I think what we're seeing right now is a way of attempting to mitigate harm and the environmental toll has become a focus in light of the climate catastrophe that we see unfolding all around us and with new reports from the IPC and thinking about the very short window that we have in which to act.

And I think there's much more of a sense of urgency from policymakers, from advocates, from activists, from researchers, scientists who are looking at the numbers and looking at the data and thinking, we really need to do something now. And I think it's all part of the same general wave of really trying to find a way to live with tech in a way that is more sustainable, in a holistic sense in a way that is actually good for people and for the planet.

Chris Adams: So hopefully finding a way to avoid a second gilded age and avoid to go from there straight into a burning age. Yeah, if you've, okay. All right. So that's one. So that's one thing we've looked into, and I suspect, I know that we've covered this on pre previous episodes before about how we're seeing some changes with, in Europe, like the corporate sustainability reporting directive. Actually that's one of them. And some of the recent laws there about basically getting organizations to disclose energy usage and disclose their resource footprint in a way they haven't had to before. You do see this kind of in this shift here actually, which is not something that we've seen something before, but, and I dunno if we have enough time to really talk about things like the role of like antitrust and organizations moving in and talking about, okay, if you have these large organizations, where are we actually redirecting all of the surplus?

Given them that we're in the middle of a literal emergency in some cases actually. So maybe this is one thing we could briefly just like close clo close out on actually, because you mentioned that there is dry, there is interest from a number of organizations or there's a number of different places you said there's pressure from staff, there's pressure from investors and you, and there's pressure from the regulators. Alright, Tamara, so we spoke a bit about regulation and. I'm a European, despite sounding like someone who isn't a European anymore. But one of the key things that we see, or comes up again and again is how you really see this appetite in Europe and to a lesser extent, the UK with things like the competition and markets authority, talking about getting organizations to be a bit more explicit about green claims, but we don't really hear that much about what's happening in America. And I know there, there have been some changes from the investor level, like the SCC and organizations wanting to be somewhat aware of their climate exposure, which is why you might wanna where you find CIOs being asked about digital footprint and stuff. But I wonder if there's anything else that you've seen on the horizon in America that might also suggest that yeah, there's actually there are changes coming down in America that people might wanna prepare for if they're technologists and like thinking about how this new world impacts the world of software, possibly even AI cuz that's one of your special, your areas now?

Tamara Kneese: Yeah, it's a great question and I, I think certainly quite a bit of movement in the US. And I think it is in part an attempt to catch up with regulation that's been happening in Europe. And everyone I know who's in the AI policy space here in the US is certainly paying attention to everything that's happening in the EU right now.

But one sort of development was back in October, The Biden administration and Alondra Nelson, who was working for the Biden administration, put out a blueprint for an AI bill of rights. And so really thinking about what it means to assess the impact of algorithms on civil rights. We also have a, a new sort of report and area of inquiry from the National Institute for Standards and Technology.

And they're looking at trustworthy and responsible AI, and they're assessing AI based on a number of different factors, including things like reliability, safety, security, and resiliency, accountability, privacy, fairness. And what's been interesting is that sustainability isn't always mentioned in a lot of these different policy recommendations for responsible ai, and I would argue that is something that we definitely need to have on horizon, is really being able to talk about the impact of AI on marginalized communities in a climate context as well. And that's something that another sort of beyond the policy kind of landscape.

Another interesting development would be groups like Amazon employees for climate justice which the seat of power for that is largely in Seattle and in the US in general, but these Amazon employees who were really pushing for Amazon to pay reparations to Pakistan for all of the devastating flooding there because of Amazon's role in climate change, and also making a connection between Pakistani American workers who are on H1B visas and really thinking about precarity in a labor context in conjunction with the catastrophic effects of climate change on people in the majority world. And I think we're going to see more activism like that probably from people with tech companies who are working with climate activists, environmental justice groups, people who are really being impacted by technology, and that is something that I'm certainly seeing even in academic communities.

So there are a number of kind of academic conferences, or even at places like Mozilla Festival where you have academic researchers and people at different sort of AI related nonprofits and policy think tanks who are bringing workers into conversation or bringing marginalized folks into their research conversations and treating them really as co-authors, as co-researchers, not as like subjects.

And so thinking about having people who participate in Amazon Mechanical Turk and other forms of micro labor on platforms. Actually bringing them into academic panels and having them carry out research on their own, that can also be taken up and used to influence policy as well. So really working from a really bottom up understanding.

And I think that's what we were attempting to do also with our SOGS report, is think about including a more bottom up perspective. So instead of just listening to what McKinsey's saying, or just listening to what venture capitalists or a C-suite or saying, also really taking a hard look at what the pain points are for people who are either building these technologies or interacting with them on a daily basis.

Chris Adams: Okay. Wow. God, there. I thought there's gonna be a lot to take into account for future Green software reports. And when we talk about where the boundary of influence might actually be for this, rather, not just the operational impacts, but the other things that it might be enabling. Okay. I think we've, just taking us up to a time here, before I say thank you, I just wanna ask, there are 30 different insights in this report and we've covered a bunch of these things.

If there is one thing that you'd ask people to start with, which one would it be and why?

Tamara Kneese: So I, I think the responsible AI is green AI insight is very,

Chris Adams: one. Yeah.

Tamara Kneese: Yeah, and you know, because it really gets at this need for a much more holistic understanding of what green could be, and I think it really gets at the need for a lifecycle analysis and an examination of entire supply chains and the effects of technology supply chains on marginalized communities. And that's something that I think many of us at Green Software Foundation at Green Web Foundation, we talk about these things, but how do we prioritize working with communities and that kind of collaboration in the kind of work that we do, even within our own organizations, within our own companies? And that's something that I think really could use a lot of attention.

Chris Adams: Okay. Thank you for that. That's a definite food for thought and I, this is, I really hope we do get to look into some of that more in future reports that come out from this. Tamara, thank you so much for giving us the time, uh, on what ended up being a public holiday. I really appreciate you doing this, and next time we'll make sure we don't actually have it booked up.

Booked up for this quick break. That's all for this episode of The Week in Green Software. All the resources for this episode are in this show description below, and you can visit podcast dot Green Software Foundation to listen to more episodes of Environment Variables. Tamara, thank you very much for that.

I really appreciate you sharing your insights with us. I wish you a lovely rest of the day and yeah, have a lovely week. Take care, Tamara. Bye.

Tamara Kneese: Thank you, Chris.

To find out more about the Green Software Foundation, please visit https://greensoftware.foundation. That's https://greensoftware.foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: Code Green and Clean Power

14 juni 2023 | 42 min

Joining host Chris Adams on this episode of TWiGS is Nina Jabłońska, operations coordinator at Energy Tag and a master's student in sustainable energy systems. In this episode, we'll explore insights from the Linux Foundation Energy Summit in Paris, including Microsoft's urgent call for green coding and real-life examples of reducing computing emissions through cloud carbon footprint analysis. We'll also touch on employee activism at AWS, where tech workers stood up for climate action and better work-from-home conditions. Nina also tells us why sometimes she goes by Nina Jab%o%ska, and why Cara Delevingne and Keanu Reeves are the ultimate “carbon-free couple!”

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Nina Jabłońska: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Carbon-aware software is central to decarbonization | State of Green Software Survey | The Green Software Foundation [3:01]
Reducing cloud emissions by 60% | ThoughtWorks [24:26]
Microsoft Issues Code Green Alert | Analytics India Mag [27:09]
Corporate Amazon workers protest company’s climate impact and return-to-office mandate in walkout | AP News [34:43]

Resources:

Green Software Principles and Embedding Carbon Awareness in Your Applications | Dan Benitah & Szymon Duchniewicz, Avanade | LF Energy Summit [4:12]
Technical Research Fueling NextGen Actions | Hallie Cramer & Ana Radovanovic, Google; Iegor Riepin, Technical University of Berlin; Gailin Pease, Singularity Energy | PyPSA | LF Energy Summit [8:55]
GitHub - PyPSA/247-cfe: Explore the impacts of 24/7 Carbon-Free Energy PPAs [8:55]
The politics of data centers | The Green Web Foundation [19:27]
Cara Delevingne sprays industrial wastewater on her face in her latest beauty campaign | Vattenfall [38:20]
Sad Keanu | Know Your Meme [40:07]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:
Nina Jablonska: So it's just a matter of really wrapping your head around how you want to do it and what concrete actions you want to you want to undertake. But in the end, I think it just pays off to be green, and I hope it will only go in this direction.

Chris Adams: Hello, and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discussed the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Hello, and welcome to another episode of this Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. In this episode, we have some insights from the recent Linux Foundation Energy Summit in Paris, Microsoft issuing a code red for green coding, concrete examples of using cloud carbon footprint to reduce computing emissions. Before we dive in though, let me introduce my guest today for this episode of this Week in Green Software. Today we have Nina Jabłońska of Energy Tag. Nina, I'll hand over the floor to you to introduce yourself and say what you do.

Go for it.

Nina Jablonska: Hello. So I'm Nina Jabłońska. Nice to hear you Chris again. So I am the operations coordinator at Energy Tag. Super happy to be here. So privately, apart from working at Energy Tag, which is a nonprofit organization, I am also actually finalizing my master's studies in sustainable energy systems, which is why I am currently placed in Stockholm.

And recently I have actually taken up Spanish classes, which makes a lot of sense

Chris Adams: Wow, bueno!

Nina Jablonska: Yeah, my, my masters, the first year I spent actually in Barcelona and I started learning. A bit of Spanish just to be able to do my groceries, and since I found that in Stockholm, everybody speaks English.

So I figured I would just continue this journey of learning a third language online with, with a, um, with a teacher. But it's pretty exciting, a very nice thing to, to do some extra brain exercising in a bit different way than studying and working.

Chris Adams: Cool. All right. Thank you for that, Nina, and I should probably introduce myself as well. My name is Chris Adams. I am the the co-chair of the Policy working group in the Green Software Foundation, and I'm also the executive director at the Green Web Foundation, another nonprofit working towards an entirely fossil free internet by 2030.

If you are new to this show, the format is as follows. We basically have some people come onto the show. We look at some stories in the news that caught arise, and we just talk about them and share some notes and reckons basically. That's largely it, and we have a few stories ahead of us that I've foreshadowed already.

We're gonna start with the first story we have here, a one from the green software report. So what I've been doing for the last few weeks is I've been inviting guests to pick an insight that caught their eye. And Nina, you are no different today. I'm gonna give you an idea of anything that you, that caught your eye when you're looking at this, and it may or may not be related to what you do at work.

So yeah, I'll hand over to you. What looked interesting when you were looking at this?

Nina Jablonska: Mm-hmm. Yeah. It so happens that one of the stories is actually a bit related to what we do at Energy Tag, and that's the fact that the carbon aware software, so a software that would operate in the time and place where the grid carbon intensity is low, or it would optimize its work based on the grid carbon intensity is super important to, to enable further decarbonization. And we also heard a very diff, very interesting presentation from Avanade on this during the LF Energy Summit. I think the whole conversation around having software that really can optimize and choose when and where to operate, depending on what kind of sources are available in the grid is very important.

Chris Adams: It's about, yeah, the underlying kind of how dirty or green the energy is. Okay, so one thing from the actual conference we're both at the LF Energy Summit, was this talk by Dan Bonita and Szymon of- Nina, I think I might need you to help me with this, cuz it's a long polish name that I'm gonna struggle with.

Could you maybe help me with this? It's, is it Szymon Duchniewicz maybe? Okay. One more time, please? Duchniewicz. Okay, great. Szymon, so Szymon that from Avanade, that was the other deck we can share a link to that has them explaining some of the work they were doing and what they were using there. Okay, thank you. That was the only thing. I just wanted to make sure I got it pronounced properly and it sounds better from you than it does from me,

Nina Jablonska: The Polish surnames are, he doesn't have the weird, the weird letters though. You, for

Chris Adams: Diacritics and things.

Nina Jablonska: Exactly. You turned mine into just the regular l and n I can see here, but for example, on my Spanish student id, they just turned that into a percentage sign. initial student Id said that. My name is Nina. Jab.

Percentage sign, o, percentage sign, ska.

Chris Adams: Okay. This is why people should be using Unicode as much as possible instead of Ascii, yeah. Nina. Ya percentage nska is not the correct name. The, the.

Nina Jablonska: It is definitely not, but imagine going to a bar with my friends and they would check the IDs and I would just show my student ID because it was the first card that I got out of my wallet and the guy was just like, are you Elon Musk's daughter? And I just, yeah, I should learn to make a robotic sound of my voice and just try to like say back something funny in Spanish, but I could not just, it was.

Chris Adams: That's, that's probably gonna be easier than explain the problems associated with Unicode and Ascii code and stuff in the middle of a bar. You've mentioned that you're in Stockholm right now, right? So that's in Sweden? Yeah. And that's a relatively green grid by comparison, right? Because that's usually a hydro or wind.

What would it be?

Nina Jablonska: Yeah, yeah. The northern parts of Sweden especially are very clean. They're mostly based on hydro power, as you mentioned. In the south. There is still a bit of. Challenging situation with a bit of leftover gas generation and so on. But overall, yeah, overall Swedish grid is really one of the cleanest ones, which is why most of the challenges present in other countries are not really discussed here so much.

Chris Adams: Ah, I see. Whereas,

Nina Jablonska: could be a good, could be a bad thing, but they're really moving at the forefront of decarbonization every of every single sector, including transport, et cetera.

Chris Adams: I see. Okay. And so I'm in Berlin, so right next to the border, relatively close to the border of Poland. And you know, you are originally from Poland as well, right?

Nina Jablonska: I am hence my complicated surname.

Chris Adams: Ah, I see. Yeah, so that's an example of grid, which is relatively cold, heavy, and somewhat dirty. So the same program run in maybe say Sweden compared to say, uh, Warsaw or somewhere might be much dirtier depending on how you would run that.

Although that may be changing over time because opponent turns out to be one of the fastest decarbonizing countries as of last year. That was one of the stories that I learned of which I was surprised to hear from, given how coal heavy that grid is, for example.

Nina Jablonska: Yeah, it could be surprising, but it could also make a lot of sense just given the fact that decarbonizing the first 10, 20, 30% is way easier than decarbonizing the last. 30, 20 or 10% of energy demand. Of course, a lot of challenges come with balancing the grid supply, with introducing the technologies needed to cover the last really couple of kilowatt megawatt hours of demand in various hours.

So I think it is a good sign. Of course, I wrote from my country, I would love to see coal as soon as possible, but I also think there is still a lot to be done. Partly, it's also nice that maybe the technological development needed to really cover this last piece of demand is going to happen abroad. So let's say maybe people from other countries and the grid operators from other countries will figure it out for ourselves and we will just follow their example in this sense.

But,

Chris Adams: Okay, cool. So that's one of the examples from here. So we spoke a little bit about the difference. The carbon intensity can be different, and this was one of the ideas behind. And kind of carbon wear computing, for example. I guess it's one thing that. It might be worth asking is like, how do we know this is really the case?

And this is something that I found quite interesting that when we were both at the Linux Energy Foundation Summit, there was actually people talking about, okay, this is how you can build open models of understanding what kind of interventions you kind can actually make. This might be a kind of jump off point cuz one of the panels that we saw was specifically related to this.

There was a number of large companies, but there was also some work from, I believe, I think you might know Igor's surname, Dr. Igor. Do you remember? Can you help me with his surname? Yes. So he was talking about some work that the technical University of Berlin had been doing with Google to basically model a kind of way to decarbonize all of their infrastructure between now and 2030.

And the thing that kind of caught my eye was, The use of an open source modeling tool called PyPSA Python Power Systems Analysis. I think that was it. I think that's something you might be a bit more familiar with than I was. And uh, this is the first I've heard of this and maybe you might be able to shed some light on this cuz I didn't know that much about it before coming to this event, this conference.

And I thought the idea that you could essentially talk about this stuff in Python, it was cool actually. Yeah. I'll hand over to you because this is something that I think you mentioned that you've been doing a bit of work with as well yourself. Right.

Nina Jablonska: Yeah. Yeah, so I, as I mentioned, I am also still finalizing my master's degree. And of course, as the last part, I do need to deliver some thesis, some piece of my own work. And I actually decided to do it based on this open software and this openly available piece of code that guys from TU Berlin have written and come up with.

And this is indeed part of a bigger study that they perform, which is studying the effects of 24 7, carbon free electricity procurement on the system level emissions, but also on the emissions for those customers who decide to follow this 24 7 carbon free procurement strategy. And then also on the cost premium of doing this, instead of following the business as usual, let's say energy attribute certificate system.

So of course, the way companies and big consumers source clean energy from the grid is through buying those annual certificates where they match their annual energy consumption with the same amount of energy in those attribute certificates that are coming from, let's say, solar power plants or wind power plants.

And why this system doesn't work, obviously is because it doesn't really send these precise signals. As we mentioned, also, with the carbon aware software, it doesn't really tell you in which time and location. The grid actually has a higher carbon intensity when it has lower carbon intensity, and it also doesn't reflect probably most importantly, the actual cost and the actual effort that you have to put in decarbonizing every single hour of your electricity consumption.

So the study developed by the TU Berlin lab is essentially doing this, assuming that some companies follow this strategy of sourcing electricity in every single hour, but also from deliverable location, meaning for example, from the same country or for the same bidding zone. And also that this energy is going to come from additional assets.

So for example, from newly built solar power plants, instead of competing with the other grid users for the existing assets. And they essentially see how that would impact the grid and the great outcome of the entire LF Energy conference as well. For me, and the great thing that I could do is to meet Igor, who is one of the authors of the study and who has helped me greatly with using their piece of code for my own thesis.

And I just think it's so great, really, that it is available openly and we can just download it literally on our personal computers. I am not even a, an energy system modeler. I never created. Any model similar insights to this one may be a small one during some classes at university. So I think it's really a great thing that anybody could just download it and it is very much, I think, encouraged also to just provide them with feedback to challenge their assumptions, which is essentially what I'm doing for my thesis.

So yeah, it is very nice for me to combine this data and software openness that we discussed also with many other people, both of us, definitely at LF Energy Summit and this kind of new generation of clean energy procurement that is proposed.

Chris Adams: So there's one, there are two things that I found quite interesting when I spoke to Iegor about this. When he gave, when he spoke on this panel, he said one of the ideas from the work that was going on is yes. You can get an idea that basically moving from maybe an annual approach to, to an hourly approach helps address some of the existing problems people have and some of the accusations of kind of greenwash related to green energy wind.

People will say, I'm running all my infrastructure on all my servers on solar power, and they're saying this. Happens at night,

Nina Jablonska: Mm-hmm.

Chris Adams: may be somewhat inconsistent with how most of us understand solar panels to be working, for example. There's some stuff like that. But one thing that caught my eye was he was actually saying that, yeah, by doing this, we're actually ma able to model the impact of whether we can make the infrastructure itself responsive.

So rather than just having to make sure that we've got enough wind and solar or hydro or stuff to do this. They were talking about how they could model reductions in demand to need so much in the first place. And the same if you have maybe 10% of all the compute that's flexible, then you need less generation, which then will impact the kind of cost of actually us moving away from what we currently have right now, which is lots and lots of fossil fuels to something which is greener and more humane on the grid.

So that was cool. But you said something quite interesting just now about like location, like local and deliverable and what was is additional, is that other thing you just mentioned? Yeah.

Nina Jablonska: Yeah. Basically, if we want to put it in short words, hourly matching is not enough. So the entire study, and I think the first concept that comes into mind and that's discussed is how we want to match the supply of clean energy at an hourly basis with the demand. And as you mentioned, what's impossible in the real world to consume solar energy at night if you don't have a battery?

So hourly is the first pillar that we, we now call them the three pillars of carbon free energy. But the other two is, are actually equally as important as it turns out. The second one being the fact that you need to make sure that this electricity is deliverable, which means you not only have an existing interconnection between point of generation and point of consumption, but also that there is enough interconnection capacity to actually allow this electricity to flow instinctively. It doesn't make sense to import electricity when you're in Sweden from, for example, Portugal. The likelihood of

Chris Adams: I see.

Nina Jablonska: of your electrons flowing from a solar power plant in Portugal and actually reaching your, for example, my house in Sweden, if I had one, are extremely low.

And we actually talked to some people who are modeling this as well on l F Energy Summit. I think there's going to be a, a summer of many interesting reports coming. And the third pillar then, is the additionality. So really seeing how it's not only enough if you match hourly and location, but also you need to make sure that if you introduce especially new additional demand, you do need to make sure that you have additional assets to meet this.

And that is a huge thing, for example, with hydrogen, since hydrogen, especially with the ambitious goals that both the EU and the US have right now. It's really. An amount of electricity that you cannot just forget about in the eu. If we want to meet the goals for hydrogen generation by 2030, I think we will need somewhere around the equivalent of the annual consumption of France.

Chris Adams: Okay.

Nina Jablonska: So it's something.

Chris Adams: New generation, like new clean generation that would be coming in,

Nina Jablonska: Exactly, and then exactly nuclear generation. So that's the question. How do you ensure that, that this is actually clean and that it contributes to decarbonization? A study from Princeton University who are performing, we can call them sister studies, to the ones from to Berlin from Europe. So Princeton did a study where they analyzed what would happen if you didn't ensure hourly deliverable and additional electricity, and what they found is that the hydrogen you would end up producing would be twice as intense, twice as high in emissions as the one produced right now using gas.

Chris Adams: So that would be the opposite of what you wanna have. So basically if you, so if I, let me just check if I understand those two things, cuz at the idea of hourly, we kind of understand, like when people say they're running on green energy and they say they're running on power and they're running a server at night, that's very intuitive, easy to understand.

The deliverable thing that was like, When people say, oh, I am running green. I'm running my say websites or running a server on Green Energy. And what they're really doing is they, are basically running it on a normal grid, but they're buying some certificates from somewhere else in the world and then saying, yeah, because I've bought these certificates that my energy is green.

And the example might be if I was to buy the green certificates from Spain, but I was running servers in the Netherlands or Germany, then it's quite difficult to actually shift electricity all the way across that. So it's a bit of a kind of porky pie to say that's a bit, that's green energy, right? Exactly. Okay, cool.

And the final thing was about additionality. This is this idea that when someone's going to be saying they're running on green energy, they need to make sure it's adding new power to the grid. Cuz if they're not doing that, maybe you could expand on that one, because this is one thing that I think is.

Interesting in the world of technology for us, because it looks like the whole discussions about massive new amounts of demand, like AI and stuff like that, it seems somewhat comparable to the discussions people are having about saying, one thing we need to do to replace fossil fuels is have a way to make hydrogen from say, solar or wind, or stuff like that.

And you're saying that you need to be really careful that the power is coming from clean sources. Cause if you just pull it from the grid at certain times, it can be almost worse than basically getting hydrogen in the usual way, which is usually coming from methane gas or something. Is that what you're saying?

Nina Jablonska: Methane reforming. Yeah. Which is a, a process which I, I am not a chemical engineer or hydrogen expert, but this is basically a process where you directly use natural gas to produce hydrogen, which should not be the case anymore. We should really try to use electrolyzers. But if they are grid connected electrolyzers that are going to draw this clean power from the grid. Again, if we allowed them to, in a way, fight or compete with the current consumers in the grid for the clean power that is currently available, that would be a disaster because we wouldn't end up not increasing the clean capacity. We would end up in a way forcing the order grid users to report that their carbon

Chris Adams: there, there is dirty air compared to what they have. I see. Okay, so there's a concrete example I think of that we've seen. So one example this makes me think of is in the Netherlands we saw a bunch of new Microsoft wind turbines installed with some new data centers and you saw quite a lot of pushback from locals because there's this idea that, oh wow, there are these cool wind turbines that are in our neighborhood now does that mean we're gonna be run using green energy?

And if I understand it, that power, it was additional, but it was all additional going just to the data center. So it wasn't materially changing the kind of. Maybe the, uh, the carbon intensity of the power that you could trace to that organization, for example, even though it's forming into the grid, that was one of the things that was, can lead this kind of gap in perception essentially, that people were saying, I thought those wind turbines meant that my use of power is greener.

But no, that's not how it's actually being presented right?

Nina Jablonska: Yeah, that depends on which framework you use because there's, there's essentially two ways of reporting or accounting for the emissions of your electricity consumption. One of them is market based and. The other is location based. If you simply use the location based, you literally only look at the location where you're based and you take the average emission factor.

And then if you do have some additional wind turbines producing clean power, then of course you're going to lower your average emission factor, which is great. But would a lot of companies do, and actually many would argue that it's a good system to implement, is the market based one where. Through the use of those contractual instruments, such as, for example, clean energy certificate, energy attribute certificates, or other -miliar similar instruments in the market, you can buy them in order to influence the amount of carbon emissions that you report, which on one hand, again, allows you to reach 100% renewable or zero emission reporting with the use of annual certificates coming from no matter what, no matter when, no matter from which kind of assets. But on the other hand, this actually does drive additional installations and additional capacity of solar and wind. It's enough to even see what the energy attribute certificates that are existing right now, what they did to the grids they actually did when they were established.

Over two decades ago, they actually did encourage a lot of additions in wind and solar capacity all over the world in, in Europe, in the US and in many other parts. So I don't think we should cancel this entire system, but there is definitely a need right now to do the same, to provide the same extra financial incentive and to provide this extra something that's going to capture the cleanliness

Chris Adams: Get their recognition right?

Nina Jablonska: Exactly. And that's why we would need, for example, the certificates. That would be issued on an hourly basis in order to incentivize energy storage or in order to incentivize other clean firm technologies that are dispatchable, like geothermal, for example, or demand site management. Again, coming back to the carbon aware software, we need a tool that is going to capture the fact that if you choose to not consume one kilowatt hour at night when grid carbon intensity is very high, but you choose to consume the same kilowatt hour during the day when for example, this avoids curtailment of solar power, then that's a great thing to do and you should aim to do so. That's it.

Chris Adams: Oh, why? Okay. Wow. There's a lot to it then. So if I understand that correctly, in the example here, say maybe having giant company. It set up some wind turbines near me. On a location based basis, it might look a little bit greener, but from a market based basis, it might make the electricity that I get to say I get to claim as being somewhat less green while the other organization gets to say, because we've spent a bunch of money to deploying this, we are gonna claim some of the greenness of it.

Okay. That does explain the two different ways of thinking about this and

Nina Jablonska: Then again, just important thing to say is that of course we want the additional capacity, and of course it's going to do good globally for the emissions. That's also important to track what are the global emissions and emissions in the zone, for example, in the country and. It's important also to remember that if a data center purchases, let's say some wind power and additionally some storage, in order to make sure that they have this 24 7 hourly available clean energy, this extra power that's generated by this wind turbine is also going to be injected in the grid.

And then if they would report their carbon emissions at an hourly basis, they would basically. Sell this extra generation to the grid, so that would also globally decrease the emission. So there is a lot of really aspects to consider here. Of course, it's not easy to predict it, but yeah, definitely it's worth to keep in mind that the market and location based systems work a bit differently.

They can trigger different behaviors and trigger different results, and I think it's worth looking at both and really taking the time to understand what they mean and what, what are the rules behind them and what are the equations behind the numbers basically.

Chris Adams: Okay. All right. Thank you for that. We've gone really deep into the rabbit hole

Nina Jablonska: We did.

Chris Adams: of carbon intensity here, so should we look at the next story then, Nina? This one here was about cloud carbon footprint, about a, this is ThoughtWorks talking about how they've been able to use this tool themselves to basically reduce the emissions across a number of services by around, I think 60%.

And that more or less worked out to be around a saving of 46% in terms of actual cash money costs. And that seems largely I think the phrase they're using is identify zombie workloads and switch things off. So like we spoke about before, there's literally just a case of switching things off here, but I think there is some, there's a couple of things that kind of interesting here that might be worth diving into.

There's something about Scope three. Did, did that surprise you as well, Nina?

Nina Jablonska: Yeah, it is very interesting because. Scope three emissions are one of the, I would say, most challenging to track and try to estimate. So I would definitely say it's interesting to, to take a look at the methodology that they took to calculate this since the emissions that we were discussing mainly so far are scope two, so directly from the use of electricity that you consume, but scope three emissions are really the emissions around the final use of your products.

And then the final upstream and downstream activities around this. So I think it is really extremely difficult to, first of all, set targets, but then of course, measure your progress towards these targets. Most companies, when they do have some emission reduction targets, they are focused around scope

Chris Adams: Mm-hmm.

Nina Jablonska: So definitely anything that they, that they use in order to track those and to measure how their net zero goals could be achieved would be very interesting to, to look into.

Chris Adams: Cool. Now, one thing that's also interesting from with my green software hat on right now is that. The cloud carbon footprint has typically used one form of measuring carbon intensity called average carbon intensity rather than just marginal carbon intensity. Cuz the Green Software Foundation has a standard called the software carbon intensity standard, up until recently was primarily just focused on marginal carbon intensity and it's, we probably don't have the time to dive into marginal versus average here cause that's quite a complicated thing.

But essentially the key thing that caught my eye when I was looking at this was that, oh, you've got tools like cloud carbon, footprints. Theoretically, you should be able to use these tools to work out S SCI scores now for particular applications so that you can start tracking some of these and comparing them to each other, for example, in the same way that you might say share like an efficiency score or grading or things like that.

That's one thing that we've got some scope to move towards now for some of this stuff. Okay. Should we look at the next story then? Nina?

Nina Jablonska: Yep. Yeah, absolutely. The next one is about carbon again.

Chris Adams: This is, yeah, Microsoft issuing code green alert. So this one here is, there's two things that are interesting here, really. So one of these was basically about this idea of Microsoft's making a big thing about. Not using water so much in its data centers are trying to move away from this and shifting to this exotic way of calling called adiabatic cooling. And this is interesting in my view because a lot of, there's been a real pushback recently about AI models, not just in terms of carbon, but also about in terms of like water usage as well here. And one thing that surprised me with this is like, There. There was one thing which is hidden away basically saying, yes, Microsoft has its claim to be carbon negative, which is zero waste and be water positive by 2030.

And these were all made before there was a whole discussion about this cut and sudden craze in generative AI and large language models and things like that. And there's a question now like there's suddenly all this new demand coming onto the grid on all these other new discussions, and it feels very similar to the additionality stuff you just mentioned before about, okay, where do we go with this?

How do we fit this in?

Nina Jablonska: Yeah, there is a very interesting analogy between this extra demand from the AI related processes and hydrogen in the regular grid because both of these have this extra feature in them, and again, if we don't account for the extra demand that the data centers will have, I'm not talking about water, but electricity since I'm not an expert on water use here, but this really again, boils down to ensuring that this extra demand, no matter how big it is, which again, how do we estimate this and how do we even predict how much it can blow up in the future?

It is really crucial that it will be covered with clean electricity, and again, the three pillars of hourly deliverability and additionality of these assets will be really important. Again, to not turn this extra demand into a carbon bomb that's just going to blow up in our faces. Because the stakes are really high.

We, as we've seen with, with hydrogen demand, and I think AI and data center uses. Data centers use how much of global electricity right now?

Chris Adams: We can look up the state of carbon waste software to see what figures they have for that. That might be the best way to do it.

Nina Jablonska: it's around 1% maybe, but it's still a lot.

Chris Adams: Yes. That's still gonna be, so that's like half of aviation is the figure spec I've heard people use specifically just for data centers, not the rest of the infrastructure. I'll say networks and things like that. So that's still. That that's a medium sized European country, basically. So that's about uk for example, the UK is about 1% more or less.

Alright. So yeah, that's the kind of ballpark you're looking at basically.

Nina Jablonska: Yeah, so again, if we, if this is the base consumption right now and with AI, we don't really know where it's gonna go and how big it's going to be. And the pace edit at which it has been growing recently is really beyond any estimates and any guesses. So I think we may see a similar discussion actually coming up.

We've had this discussion about hydrogen. We do have the delegated act already requiring these three pillars in the European Union for the definition of in order to produce clean hydrogen and to define hydrogen as green and carbon free, we need to ensure hourly new asset, uh, no more than three year old asset than the electrolyzer, and then also the same bidding zone.

The same discussion right now is happening in the US and there is more and more discussion moving towards. Also implementing those three pillars in the requirements there for the tax credit. And I think we may maybe in a couple of months hear the very same discussion on how to ensure that the data centers and that AI is powered by clean power only.

Chris Adams: You actually raised a good point, Nina, because a few weeks back we looked at one of the stories of reports, uh, by amongst the authors was one, uh, woman, Sasha Luccione. And she was actually in the report that, so in the paper that she shared that we'll share a link to, there was this idea of saying, if you're gonna run AI or if you're gonna have these models, Then what you should probably be doing is actually figure out how many hours are you gonna be using this for, and then find a way to attract that because these are kinds of models which are not latency sensitive this way other the way other things might be.

So if we're gonna have a model that's gonna use this much power then and we realize it's gonna create this much value, then yes, we should find ways to make sure we can run that on an hourly basis. For example, always using green energy and because these kind of things are plausible. There are scenarios where if you are gonna take maybe 20,000 hours of compute to use this, you could plausibly put them at certain times where.

The sun's in the sky, or there's really easy access to green energy so that you can work out what the cumulative carbon footprint of that might actually be. That's quite similar to what you just said actually about the whole hourly thing that you mentioned with hydrogen there. I guess the question that we might have when we look at these two things is that.

With hydrogen, there's a whole discussion about we need to get our fossil fuels, so we're gonna need to have this much hydrogen. But we haven't really had that discussion about saying, do we need to have this much ai or do we need to, how much compute is enough for this?

Nina Jablonska: How much AI do we need?

Chris Adams: yeah, because there is, there's gonna be all these assumptions about do we just need to do less or do we need to meet these massive targets that we might actually have?

Cuz the ones you mentioned before was for hydrogen was what a another France worth of electricity, for example.

Nina Jablonska: Yeah, around around 500 terrawatt hours of electricity, which is more or less the annual consumption of France.

Chris Adams: Okay. That's a significant amount. Wow. All right, so that's, I guess we need to have maybe some of those discussions with data centers and talk about how some of that work, how, what the priorities might be for some of that then as well. Okay.

Nina Jablonska: Exactly. And also one more thing that just popped into my mind is I remember there were some people even pushing against the development of AI, and I remember one of the ideas was to tax AI companies like 98% or something from the revenues. In order to just really limit, I'm sorry, I don't remember who exactly said that, but there was an idea that if AI is going to really develop at this pace and gain so much revenue, then maybe they should just pay ridiculously high tax in order to limit this in a way. And this gets me to think that if AI is going to be a business where you can make a lot of money, why don't you make sure that this money actually goes in a good place and in this sense, and the same conversation again, goes into hydrogen. With hydrogen, you are going to have subsidies. You're not going to just generate hydrogen and pay 100% of the price on your own.

There is going to be large subsidies in both EU and US, and therefore there really is no excuse to not make sure that you have this additional deliverable, hourly clean power, and that you really don't use only clean power from the grid because you have the money, you have the technology. The technologies are available largely.

We do have, we do know how to do it. So the question is, yeah, where is the

Chris Adams: priorities that you want to have for this?

Nina Jablonska: and where either the legislation or the push maybe from the society is going to be to really make sure that it's doing way more good than it's going to do bad.

Chris Adams: I see. So that's like a discussion about, okay, if there is all the value being created, how is that distributed? That's the discussion that you, that needs to be happening there. Okay. Alright. Speaking of this, we've just got one more story I think to look into before we wrap up with a couple of, with a, the little question here.

This is actually a story from AP News about actually staff. This was happened last week actually. This says Amazon work workers. Basically they did a walkout about the company's climate impact and the return to office mandate for the walkout. This is the follow one from us having a discussion about employee activism last week.

I, I think it's interesting because we see a lot of the impact and lots of the action from large firms coming from employees a lot of the time.

Nina Jablonska: I think it's, I think it's an extremely interesting conversation in general, not only about remote or at in the office work, but also of the impact of both the pandemic and the climate crisis and how of the new generation of workers coming to the job market. I think the reality of what you're looking for in a job has changed so much, so much over the past two or three years,

Chris Adams: mm.

Nina Jablonska: and there there is power in the people.

I think main outcome of this is just really, if you want to keep your employees satisfied, it's not only enough anymore to pay them well. In fact, some people are willing. There was also studies done on this that people are willing to be paid less if they know that their companies are. Extremely climate aware and sustainable yeah.

Chris Adams: I agree with you. This one is actually this one thing that there's a particular quote. Okay. Today looks like it might be the start of a new chapter in Amazon's history when tech workers coming outta the pandemic stood up and said, We still wanna stay in this company and the direction of this company.

This is from Eliza Pan, a former employee and a co-founder of Amazon employees for Climate Justice. These are some of the people who have really been pushing for Amazon to be moving faster. And as an organization, which is as large as it is, this is a real significant driver. That's a lot of leverage that we'd actually have there.

And you see this in a number of places, and I, I think it's worth paying attention to some of this stuff here because, power, leverage and pressure comes from lots and lots of places, and today as we learn more and more about the state that we're in, I think as much pressure as possible will help us accelerate and get to where we need to be, especially for 2030 on this one.

Nina Jablonska: And also not to mention that sometimes I would. Even say, usually if you do the climate consciousness part right of your business, you may up even saving money, not only spending more, so it's just a matter of really wrapping your head around how you want to do it and what concrete actions you want to you want to undertake.

But in the end, I think it just pays off to be green, and I hope it will only go in this direction.

Chris Adams: So this has been interesting with AM Amazon specifically because they are already. The largest purchaser of renewable energy in the world now and last year, I think they've already spent something in the region of $60 billion on infrastructure already. And this gives you an idea of the kind of scale that we need to be moving at.

So this is the largest one saying you need to be moving even faster than this in order to do this. And I think it's, for many of us, it's difficult to get ahead around just the scale of these numbers and the change that we might actually be needing to do. Like you mentioned, another France worth of, of of electricity, for example, to replace fossil fuels in certain areas.

It's a dizzying times to be in, that's what I can say.

Nina Jablonska: Definitely super interesting.

Chris Adams: Okay, so we've covered through this, we spoke about the some of the green coding stuff from Microsoft. We spoke a little bit about the carbon oil sdk. I think the last question we have now is just a final question that we have. This one is, I think this is one from Chris, our producer.

If you could have any celebrity become an ambassador for green software, who would it be and why?

Nina Jablonska: Is it a question coming from the Cara Delevingne, maybe.

Chris Adams: Oh, that's a good one. Actually. I hadn't actually thought about that, but you might wanted to explain that one because if you, I'm not sure how many people know about this stuff. Actually, it's worth just explaining some of this.

Nina Jablonska: It is so good. Of course it has it's many haters, but haters gonna hate. Basically, Cara Delevingne participated in a, in advertisement from Vattenfall, which is a huge company here, a huge energy company here in in Sweden and in other countries as well, where she promoted an emission free face mist or perfume. I don't exactly remember what it was, which was basically a campaign targeted at advertising.

The fact that waterfall is moving towards carbon neutrality and they're going to, uh, create, oh, I don't remember actually if it was hydrogen or if it was just industry clean water. Anyway, it was a campaign that was aimed to raise awareness about the climate goals of Vattenfall and the way they're moving forward with with all of this, but I wouldn't pick Cara Delevingne for green software. I think for green software, and I'm so bad at names of actors and actresses, but the guy who played in Matrix?

Chris Adams: Keanu Reeves.

Nina Jablonska: yeah. Yeah. Because the software, the zeros and the ones in the background, they were running like green. It.

Chris Adams: also green as well.

Nina Jablonska: So I think he would like to the back move and avoid some carbon emissions and then, and then have these zeros and ones in the back.

And it's all green software.

Chris Adams: So it's literally green software against a black screen. Okay. So that's your person.

Nina Jablonska: For Keanu Reeves.

Chris Adams: All right. Okay. And also when builds, when things aren't working so well, we can always roll out the sad Keanu meme. So it's, it works out on low all these levels. Excellent. All right. Okay, so Cara Delevingne for Energy and Keanu Reeves for software.

Nina Jablonska: Carbon free couple,

Chris Adams: Carbon free car. Yes, indeed. Alright, that takes us to the end. I think that's all we have time for this episode of this Week in Green Software. All the resources in this episode are gonna be in the show notes as we normally do. You can also visit https://podcast.greensoftware.foundation to listen to more episodes of Environment Variables.

And, uh, finally, I wanna say huge thank you, Nina. Thank you so much for coming on for this. I enjoyed hanging out and chatting with you, and I wish you the best with messing around with Python Power Systems analysis.

Nina Jablonska: Thank you. I'll need some of that for the next month or so. But thanks so much for having me. It was a lot of fun. A nice way to start the week.

Chris Adams: Cool. All right. Take care yourself, Nina, and everyone else. Thank you very much and see you on the next episode, tara!

It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit https://greensoftware.foundation. That's https://greensoftware.foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: Open Source Innovations with Tom Greenwood

7 juni 2023 | 38 min

TWiGS host Chris Adams is joined by special guest Tom Greenwood from Wholegrain Digital, to bring you the latest news and updates from the world of sustainable software development. They discuss insights from The State of Green Software report, the cost reduction strategy of Amazon's Prime Video, Atlassian's sustainability program, Wholegrain Digital's Employee Activism Policy, the open-source Falcon LLM, and the innovative approach of heating swimming pools with servers. They also highlight upcoming events like the GSF’s UN World Environment Day Event (today!) and the London Open Source Data Infrastructure Meetup. Tune in for a deep dive into the intersection of technology and sustainability.

TWiGS host Chris Adams is joined by special guest Tom Greenwood from Wholegrain Digital, to bring you the latest news and updates from the world of sustainable software development. They discuss insights from The State of Green Software report, the cost reduction strategy of Amazon's Prime Video, Atlassian's sustainability program, Wholegrain Digital's Employee Activism Policy, the open-source Falcon LLM, and the innovative approach of heating swimming pools with servers. They also highlight upcoming events like the GSF’s UN World Environment Day Event (today!) and the London Open Source Data Infrastructure Meetup. Tune in for a deep dive into the intersection of technology and sustainability.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Tom Greenwood: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

State of Green Software | Green Software Foundation [2:54]
Scaling up the Prime Video audio/video monitoring service and reducing costs by 90% | Prime Video Tech [09:04]
Don't F&*% the planet. | Atlassian [17:58]
Activism Policy - Wholegrain Granary | Wholegrain Digital [18:47]
UAE's Falcon 40B is now Royalty Free | Technology Innovation Institute [26:28]
New data centre turns waste heat into warm water for swimming pools | TheNextWeb [29:17]

Events:

Resources:

Green Software for Practitioners (LFC131) | Linux Foundation Training [7:10]
From Carbon Aware to Carbon Intelligent | Environment Variables episode with Luis Cruz at TU Delft [8:03]
So many bad takes — What is there to learn from the Prime Video microservices to monolith story | Adrian Cockcroft [11:09]
Grok Ventures | Mike Cannon-Brookes [15:42]
Blueprint for climate activism policy | Business Declares [22:49]
Climate Clauses | Chancery Lane Project [23:36]
Learn About the Salesforce Sustainability Exhibit Unit | Salesforce [25:38]
tiiuae (Technology Innovation Institute) | HuggingFace [28:51]
W3C Web Community Group [34:44]
Curiously Green Newsletter | Wholegrain Digital [36:30]
Oxymoron Newsletter | Tom Greenwood

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript below:
Tom Greenwood: There's no point people wasting time and money and energy, like reinventing the wheel. Somebody's doing something that might be useful to others. Put it out there, share it, and then we can all stand on each other's shoulders and go a lot further, a lot faster. I think.

Chris Adams: Hello, and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Hello, and welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and in this episode we'll be talking about a few insights from the recent State of Green Software Report. A brief survey of developments of the environmental impact of large language models, unexpected news about monolith versus serverless for green coding, employee activism, policies for the workplace, and finally some events and opportunities for development from the world of green software.

But before we dive into this, let me introduce my special guest from Wholegrain Digital for this episode of this week in Green software. With us today, we have Tom Greenwood. Hi Tom. Why not introduce yourself from here?

Tom Greenwood: Hi, Chris. Yeah, I'm Tom and I'm co-founder of Wholegrain Digital and been the big proponent of sustainable Web design for a number of years. I wrote the book. Sustainable Web design and yeah, into all things sustainable business and sustainable technology. Really so keen to be here.

Chris Adams: Cool. Thank you. Also the website Carbon Guy as

Tom Greenwood: Not the website carbon guy. Yeah.

Chris Adams: Yes.

Tom Greenwood: Not many people know that I'm involved in that. I sometimes, I'm walking around in like tech conferences and I hear somebody talking about this website, carbon calculator, and I'm like, yeah,

Chris Adams: Yeah,

Tom Greenwood: I was involved in that. Right? Yeah.

Chris Adams: And for folks, and if you are not used to, if this is your first episode, my name is Chris. As I mentioned before, I am the executive director of the Green Web Foundation and the policy chair for the Green Software Foundation. I'm one of the maintainers of a software library called CO2 js, and we also work at the Green Web Foundation, where we run various checkers and tools and open source software for we're understanding the environmental impact of green software.

So if you are new to this show, the general format is that it's a roundup of some new stories that we've seen this week that we thought were worth talking about. So what we'll do is we'll share a link, have a bit of a chat about each of these, and then run through until we run out of time. Should we have a look at the first one together then?

Tom Greenwood: Yeah, sure. Let's dive in.

Chris Adams: Okay, so the first one here is a new report that was released last week, which is called The State of Green Software released by the Green Software Foundation. This has been a bit of a labor of love for the last. Me nearly a year trying to get some of this together. And, uh, it went live last week. And if you go to https://stateof.greensoftware.foundation/, you can see the report and all of the findings in there, hyperlinked glory.

Tom, there's a couple of things that caught your eyes on here. What, what should we talk about first in this one?

Tom Greenwood: Yeah, sure. First of all, someone's been very busy doing all this research, which is fantastic, and I was really excited to see that there was so much in it, which is a big change from several years back where there was not a lot going on in this space. And the thing that jumped out at me specifically on that level was one of the items around they'd found 2000 specialized software tools related to green software, which is a huge number, does that, to think that there are that many projects going on, however big or small it might be, it means there's a lot of people interested and are starting to actually work in this space.

Chris Adams: That's true, and if you think there's maybe what, and a few million developers. That's actually a surprisingly high proportion of developers and projects. Assume you have a one-to-one and there's not no one person making a hundred of them. That's actually pretty impressive, actually. I think the research that we have here.

Let's have a quick look. Cause I can't remember if this is just looking at GitHub or if there's been a look across all of the different tools. Yeah, this is mainly looking at GitHub actually. So given there's been a kind of explosion of other tools like GitLab or giti or other things like that, there may be more like you folks used GitLab as well,

Tom Greenwood: We use GitLab. Yes. So we are not included in the statistic. Yeah. So that's

Chris Adams: be more than that case.

Tom Greenwood: it could be, yeah. And they've been quite cautious in their numbers, I think, in the way that they've laid it out. So actually, if anything, it could be more.

Chris Adams: So we have some forks, but including the forks, it's at least 2000, which is a number which is significantly higher than zero, which probably what it was maybe five years ago. So that's encouraging. Alright, anything else catch your eye on this one actually, Tom?

Tom Greenwood: Yeah, the other one was around it, what was called decarbonization alone Cannot make green software, and it was basically an article around the fact that we just need to think about more than just using renewable energy and making things energy efficient. There are obviously two really important components, but there are other environmental factors from digital technology, basically they focus specifically on water. But I think like the message is really, let's think more holistically about the environmental impact of digital technology rather than just being about like energy, weather and carbon emissions. And I think that's really important because like water is obviously like a key resource fundamental to human life, especially clean drinking water.

And it's limited, but it also gets, if you take this more holistic mindset, then you also think about things like electronic waste as well and how these things factor. Um, I'm really glad that they've highlighted that. Cause I think that. When we're talking about these things, we do tend to focus, be a bit tunnel visioned on the kind of the energy piece.

Chris Adams: You used that spec specific term, a tunnel visioned on the carbon. We'll have to share a link to that diagram of the person's eyes. You know the one I'm talking about where someone's only. Looking at carbon and missing all of the other kind of parts of this big circular forms of impact that we actually have.

Okay, cool. I'm glad you mentioned that. Well, we work at the Green, Web Foundation. We talk about the environmental impact of software in a number of ways. We talk about efficiency, which is one thing we are used to, but we talk about intensity of which carbon is one, and you can also have water in intensity and.

Various other minerals being drawn out the worth and also toxicity and things like that. So yeah, this is absolutely a place to be looking at. All right. There's, can I give, can I share one, because there

Tom Greenwood: Yeah. Yeah, go

Chris Adams: my eye. So this one here was this idea that developers want to have a positive impact.

And there was one really nice study or stat from this goes from according to this survey, which is over, I think it's just under 3000 developers took place in the survey from this, 35% of the practitioners shared they and their organization never measure the environmental impact of software. They said that they really want to though.

So they found that say only 8.5% have actually taken any green software training, although more than half of them are looking for stuff. So this suggests that there is an unpinned demand and or an unmet demand for this stuff. And I guess the thing we should probably do at this point with my hat on is basically say that or share a link to the fact that the Linux Foundation does actually have some free training that can give you, I'm not sure if I'm allowed to call it accredited or certified, but there is some form of recognition that lawyers allow you to use the note word for that.

Let you say that you've done this and get a bit of a grounding on this. We'll share a link to that because it's actually quite useful. And it was based around the Principles Green website from a couple of years back that Asim worked on

Tom Greenwood: that's brilliant here. It's obviously disappointing in a way that like people haven't been able to necessarily do the training or find the training that they want, but the fact that like more than half of software practitioners would like to. Is amazing cuz it's just a matter of time. Then before they do find it and they do take that training and they start to embed it into their works.

That's exciting.

Chris Adams: There is also scope and hope in my view for things when people are just starting to come into the industry. So previously we did a podcast interview with Luis Cruz at TU Technical University Delft, where he was talking about an open source syllabus that he was working on for students doing a master's at his course.

And last week I went to South Southern Germany to Frey Borg to, for the first ever software engineering course, which was specifically aimed at sustainability first ahead of the actual software engineering part, which I really enjoyed. It was really cool, and it was a project run by the European Commission where they basically issued something like 20 scholarships to all these young students from say, Bangladesh or Nigeria or Germany or Indonesia.

Lots and lots of places which are outside of just the North America and Western Europe where people who are often, many cases you would associate with the people on the sharp end of a, lots of the changes of climate. You had a bunch of those people coming along and learning and talking about, okay, yeah, this is how I wanna build this into my work when I graduate.

It was really exciting and really inspiring. I'll have to share some more links to that one as well.

Tom Greenwood: That is amazing.

Chris Adams: All right. Let's look at the next story then. So this is one about, I think it's called Scaling Up the Prime Audio Video Monitoring Service, and Reducing Cost by 90%. It's a bit, it is a pretty, pretty dull title, but basically the thrust of this story is that it did arou, it did the rounds.

Recently Amazon shared a blog post, uh, about their Amazon Prime video service. And a lot of us are used to this idea of serverless software being the kind of trendy thing that turns itself off and it's seen as one of the most efficient ways to run infrastructure. The key thing from this was the team at Amazon basically saying we moved away from using serverless to using boring, old and busted monolithic parts of the infrastructure and we saved 90% of our infrastructure costs by doing this. This caused an explosion of hot takes across the internet with everyone saying, oh, monoliths versus serverless, everything like that. And I find it interesting in my view, cuz this goes against the narrative that we typically do have where everything has to scale down to zero and everything like that.

Anything that catches your eye on this one, Tom?

Tom Greenwood: Yeah, I get asked this question a lot, people asking me like, oh, shouldn't everything be serverless and I'm not gonna claim that I'm an expert in this but, but my answer is always, it depends like everything. And I think my view has always been that we shouldn't be dogmatic about technology. Like it's all about context and what you're trying to do with it.

And every technology has a really valuable use case, but equally, every technology has its flaws that mean that the wrong. Application is not necessarily the best thing. And I think this is really interesting cuz they basically, they've started off, they've used the hot trendy thing and it has been the right thing to build a prototype and get it going and.

Like an MVP to demonstrate the principle, but then they found that actually we wanna scale this. We wanna make it really robust and efficient over the long term. Then actually the sort of monolith approach is actually really what we need. And I think it breaks down some of the dogma and I think it just just demonstrates actually we need to just assess each use case on its merits rather than being dogmatic and allying ourselves to one solution for everything.

Chris Adams: Yeah, I agree with you on this. What we'll do is we'll share a link to a really nice piece from Adrian Cockcroft. Basically, it's called So Many Bad Takes. What is there to learn from the Prime video microservices to monolith story, which expands on this in a bit more detail. Also, just for context, Tom, you folks use WordPress and PHP as like one of the main things that you folks use, right?

Tom Greenwood: We do. Yeah, that gets a lot of criticism.

Chris Adams: This is the thing that is, in my view, entertaining because the actual programming model for PHP, if you think about the things that people like about serverless, like you run something and it scales back down to zero. The actual programming model used for PHP, where you just load a script. Bootstrap everything, server response, and then go back to, to nothing Again.

That's basically how things like PHP tend to work and how they're designed to work. This is how the whole shed hosting thing, for which it may be maligned, but this kind of approach has been essentially the mainstay for a bunch of infrastructure for 20 years. So when you actually think about this, if you squint basically WordPress and bhp, a bunch of this stuff can look kind of serverless in this way.

So yeah, that's the thing that I just have, I'll share with all of you. Okay. Should we look at the next story from this one?

Tom Greenwood: Yeah. Yeah.

Chris Adams: All right, so I'm not sure I'm allowed to say this without ruining the language on this, so I'm gonna just spell it out the, there's a really nice piece from Atlassian called Don't F Sharp at bang percent the planet.

I think it might be don't, I'm not sure that's what

Tom Greenwood: Yeah, we know what to say.

Chris Adams: in, yeah, exactly. So I was born in Australia and Australians can be known for colorful language, and this is a. Quite an Australian way to talk about don't f the planet. Basically, this is Atlassians talking about their most recent work on basically net zero and them sharing an actual report about how they did it, how they went about, what steps they did, what was easy, what was not so hard, and how other organizations can follow this themselves.

Have you been following any of these, Tom?

Tom Greenwood: Yeah, it's really interesting. It's good to see. Firstly, just that like you've got a big name company in the tech space that's not one of the big three. Going down this path and saying, look, we're taking sustainability seriously and we're taking it seriously on multiple levels and it's not just a marketing thing.

But then it's also nice to see how they're laying out really transparently in a way that a lot of the big tech companies are a bit vague. And instead Atlassian has said, look, here's a nice report that tells you like how the journey of like. Why we're doing this, how we got here, where we've got to the things we found difficult, and then obviously they're gonna be reporting on that moving forward, which is really good.

One of the things that jumped out at me was the fact that they set themselves this target for 2025 for having everything using a hundred percent renewable energy, but they then found that actually just switching to renewable energy provider and then using credits for the rest is, is like such a low hanging fruit.

They did it almost immediately and then the question was like, oh, okay. Like how do we make this more robust now? That's the question moving forwards. But the low hanging fruit was already there and they didn't even know it until they looked into it. So it's nice to see things like that where I think they used the phrase Go fast and then go far, which I think is a really good way of thinking about it.

Do something to take it like a step forward and then keep moving forward because there aren't no any fruits. And I think a lot of people are put off by just feeling, oh, I dunno what to do. Especially things like net zero can sound really complicated and scary as like such a, like a big, lofty goal. And I think it's nice to see how the, they've taken this approach of let's just start moving in the right direction as fast as we can and then find the blockers along the way.

Chris Adams: Yeah. I have to say I'm a bit of a fan of Atlassian in general, and I, we use Trello at where we work. All right. And I know people have opinions about things like Jira and, and Confluence and things like that, and, but it, it's really nice for an organization to be so transparent about the infrastructure as well.

In their report, they're one of the few organizations that basically say, here's a breakdown of all of our infrastructure, how much we're using, and ev every single data center. And they provide this reporting, which is almost impossible to get out of other organizations, so it's really cool to see them doing it.

Yeah. The other thing that I think is quite interesting is that there's a kind of stereotype of like tech billionaires being generally terrible people. Right. There's something really interesting that I think from Mike Cannon-Brookes, who's one of the founders of Atlassian. So the funding he's using rather than I know, turning large social media websites into kind of havens for right wing climate denial.

One of the key things that he's been doing is basically aggressively buying up the biggest source of carbon emissions in Australia, which is the biggest power station they have. And then finding ways to refinance it so they can shut it down and replace it with wind and solar basically, or primarily solar.

So this is what they're doing and that's one of the projects called Grok Ventures that's doing all this stuff. So there's all this stuff here, and then there's like activist investing to accelerate this transition away from fossil fuels. It's like really cool to actually see someone talking about some, something about this and using funds in a kind of, In my view, a very kind of pro-social and progressive way, but also somewhat techy and boring basically, is okay, you will need to do some boring refinancing of this stuff rather than only looking at the shiny things.

It's cool.

Tom Greenwood: Yeah. Yeah, it is. Yeah. The things like financing can sound really dull, but I think it highlights how typically you'd look at something like sustainability in a business, big tech company, as being like, okay. Yeah, let's do some offsets and we'll reduce our energy consumption and so on. But actually like looking at how that company has resources and influence that it can use more broadly, that actually can have really big impact is super interesting and it's really nice to see that they're pulling that lever.

It reminds me a little bit of also, like when I lived in Australia for a bit, there was a company, big mortgage company that was looking at introducing solar powered loans, and the idea was basically that they could refinance people's homes in order for them to buy solar panels and then once they got solar panels, that would generate an income that would actually mean they'd pay up for their mortgage faster than if they hadn't done it.

And it's things like that where you think actually, like it sounds really dull when you like about refinancing things for um, sustainability, but then when you actually look at what you could achieve, oh, actually that's pretty clever. It was a bit of financial wizardry.

Chris Adams: Yeah, absolutely. And uh, really good example in the UK is actually one of the writers of Love Actually. His new thing is actually this thing about divest moving your pensions, basically moving your pension outta fossil fuels into renewables. Cause he basically said there's 51 trillion. Pounds or dollars of money invested in stuff and you can either, by default it's usually invested some chunk of it in fossil fuels.

So one of the things you can do is actually just take some of this. You're not having to give any money, you're just making a change. So it's not doing the bad stuff and going into the good stuff. It's really cool. They said like they've run, they started the pro- the campaign just a few years ago, and they talk about how they've been able to redirect something like more than a trillion pounds of investment away from fossil fuels into renewables already, which is gonna make a significant change, right? Yeah. Once you find the leverage point, it's pretty impressive. And this is going back to the hot waste, what we're talking about. This is one of the reasons why it's interesting to work at developer because there aren't that many developers.

This is the argument that Asim keeps making. It's basically because there aren't that many developers, if you can impact some of the developers to make changes there, then you can have some relatively high leverage changes that cascade through the actual supply chain, basically. So that's one of the things.

But sometimes you do need boring policy things, or not boring, but whole need necessary, but sometimes need to do the work on the policy front for this. And, uh, this is something we spoke about ages ago actually, when you folks published your sustainability policy, and then you released it with creator comments. That was really helpful cuz that was directly relevant for our organization. We ended up essentially using that as a template ourselves for this stuff. And that's probably a nice segue into the next story that we see here.

Alright, so this one is about Wholegrain Digital. We've introduced an employee activism policy, so this is talking about some of the other things you can do as a technologist, which aren't just about coding.

Cuz while coding is fun, we are also citizens as well as just consumers or. I don't know what's, I don't have a word for a contractor. Yeah. We're more than just contractors and consumers I suppose. Maybe you could talk about a bit about this one, because there's a bunch of thought that's got into this and I was really pleased to see this go live actually.

Tom Greenwood: yeah, sure. So it actually came from an event called Good Fest that I attended last year. Good Fest is. Like a, a sort of creative conference for, for making the world better essentially held down in Cornwell every year. It's amazing event. And there was a talk there from a guy at Patagonia and he was talking about how Patagonia in the US and he is like really supportive of their employees taking part in activism.

Immediately after that, had a lunch with a guy called Viril who is involved with Just Stop Oil and it was a really interesting conversation where he started talking to me about how actually a lot of the barriers to activism are employment related. Activism can come in many forms, but a lot of people are in.

Either nervous to get involved because they're worried that it might reflect badly in terms of what their employer might think of them, or they are struggled to get time off work or they can't afford to get time off work, or they're worried that, like what happens if I get in trouble and like I get arrested or something and then I might lose my job?

Or what happens if something happened and maybe I got glued to a. Bus or something. And then I had to miss a day off work and, and people think, oh, I don't really know how I can fit this into my life as an employee in a company where I've got responsibilities and I dunno how the company will look upon it.

And so on. And we started chatting in about, surely like companies could introduce things that would basically try to mitigate as much of this as possible. So I said about trying to figure out like what might those things be? And then along the way, Ben Tolhurst from Business Declares, which is like a nonprofit, organization that gets businesses to commit to net zero and playing a role in trying to tackle climate change.

He heard that I was working on this, got really interested because he's really interested in the sort of activism side of things himself. And also he's got a lot of connections with people in other businesses that are looking at what they could do from a climate change point of view. So it quickly evolved into Ben helping write the policy as well as hooking in people from other companies who were, who were intrigued by this idea, or maybe this is something that that we might be interested in doing as a company, we then realize that we really need some lawyers to tell us whether we're doing something stupid. So we involved Bates Wells, which is a B Court law firm, and who are basically like a bunch of hippies that have got law degrees. And I'm not sure if they'd like me describing them that way, but I think that's why they're so brilliant.

They, they care about the outcome rather than just being like all about risk mitigation and

Chris Adams: Principled legal professionals who like granola.

Tom Greenwood: Yeah. Yeah, that's, yeah, they're great. And they were really helpful in basically going through it and rather than literally tearing all of the heart out of it, which it did get run by another law firm who literally did that and came back and said, just don't do it.

Bates Wells came back and they were like, look, here's all the risks we can see. Here's our thoughts on how you can mitigate them. You go decide how much risk you're prepared to take. And so the outcome of that is that we crafted a policy for our company, Wholegrain Digital, which basically means that people can take time off to get involved in various forms of activism. If they do get arrested, they're not gonna get fired. So there's security of employment. If they do get arrested, we'll contribute to the bail money if necessary. So there's some things that we could do that are quite tangible that basically say, look like if you are passionate, we're not gonna tell you to go out and do anything.

But if it's this sort of comes intrinsically to you and you feel this is important, it's not just climate change, it might be some other kind of social or environmental issues and you want to go and stand up for it, then. We really wanna back you because we need people in society like that. And here's what we can do.

And then in return, it basically says, here's the things we ask of you if you are going to do that, and you want some support from us as a company. And it basically is saying, look, just be careful. Try to keep yourself safe and uh,

Chris Adams: have your back. Basically,

Tom Greenwood: we have your back. Yeah, exactly. So we published ours and then business declares then published a variant of it, which is open source so that any company can base, and it's got guidance notes in the template as well, including some of these things that the the spoke to us about.

So that as a company, you can basically take this template and say, okay, how might this fit for us? How much risk are we prepared to take? And you can craft your own policy. Gonna get your own legal advice, but hopefully more and more companies will see this as a way of lowering the barrier to entry for people who would like to get involved in more activism, but maybe feel like there's some things that are holding them back.

Chris Adams: Cool. I'm really glad you shared that, and I really like the approach that you've mentioned about it almost being a kinda like modular approach. So you're able to see how far you're able to go because this somewhat reminiscent of the work. I believe the Chancery Lane in the UK has been doing around writing climate clauses into kind of commercial contracts and things like that.

Cuz I know that you folks have spoken about things like having a carbon budget on a consulting project, or if you're building something, you'll do things like that. They have a bunch of existing mechanisms like that, which are easy to put into standard form contracts. So when you are doing some work.

These are the things you can include, and I assure you there is a kind of link to WordPress for this, which is why I'm where I'm getting to with this. The way that the people at Chancery Lane explained it to me was basically this idea that in the legal world there's like Lexus Nexus and there's a few organizations which have these kind of standard form pro contracts.

They're built like the WordPress of standard commercial contracts that you do this stuff and they realized this and they said, okay. We can take some ideas from open source and we can apply that to the legal kind of world. So there is now a website called Climate Clauses, which is from the Chancery Lane Project, which is, you know how like WordPress plug-ins extend something to carry in new directions.

They've basically taken the same approach to standard form contracts that people use for entering business deals with other people. For example, they said, here's how you can include like the module for net zero or the module for a budget or the module for environmental performance clauses. It's really cool

Tom Greenwood: really good. Yeah.

Chris Adams: Yeah, we'll share a link to that as well because when I first heard about that, when I had a lawyer explain it to me and say, yeah, dude, this is basically no, we, we saw what you folks were doing at WordPress, we figure we should have that too. And I thought, wow, that was such a cool idea. Cuz it's so different from, it's very easy as a tech you to just think, oh wow, we are the only people with this special tech spec technique.

But it turns out that no other skills and other industries can be inspired by some of the things that we, we are doing here.

Tom Greenwood: Yeah. Yeah. And like you said earlier about the sustainability policy that, that we shared with you guys. I guess a few years ago. There's no point. People wasting time and money and energy, like reinventing the wheel. Somebody's doing something that might be useful to others. Put it out there, share it, and then we can all stand on each other's shoulders and go a lot further, a lot faster, I think.

Chris Adams: Yeah, I agree. And just to round this story off, the example of these contracts, I just came back, there's a project called, there's a contract called the Salesforce Sustainability Exhibit. So Salesforce large company, they basically used this, they used the chance relaying climate clauses as the basis for all their stuff they do with all of their supply chain now.

So it just like open source, it always comes up in weird places. So it turns out that a bunch of people working on a side project ended up having some of the basis for one of the largest companies in the world to be, for them to use is their basis for essentially building sustainability and climate awareness into how they do business with pretty much the entire supply chain.

Tom Greenwood: That is amazing. Ripple effect, eh?

Chris Adams: Yeah, indeed. Alright, so speaking of open source, let's look at the next thing that we have here. So this is Falcon, an entirely open source LLM, which is a large language model you can run on your laptop. Tom, this was, this might feel a little bit left field and uh, I, I'll give the introduction and then I'll let you come in on this one actually.

So, We are used to large language models coming from organizations like Facebook or Microsoft or Google, for example, or not necessarily coming from, but us being able to use the results of, and this one here is, okay, weirdly or unexpectedly, United Arab Emirates. Have basically published a entirely open source, royalty free, a machine learning module.

And there's a few things which are interesting. In my book, when we looked over the website, they, it's designed to be more efficient than say, GPT three by the significant amount. And it requires maybe a fifth of the computer inference time. So when you've used, once you've trained it, you are using to try to get numbers back when you speak to ChatGPT, is that kind of thing.

That's what they're using and this is designed to be open for. Anyone to use. So rather than having to only get it from say, OpenAI or own you from Google, anyone is able to run this even on if your laptop's fast enough, your own laptop here, and I think this is one of the first and largest and highest kind of performing open models for this.

And it speaks to the idea that only a few years ago, but only less than six months ago, these were millions of dollars to train and only available from $3 trillion companies, and now you can train and run this stuff on your own hardware for less than a million tens of millions of dollars. Right? Amazing.

Right.

Tom Greenwood: A huge leap forward and it, yeah, it feels a bit ironic that, that this is not coming from OpenAI. It's coming from coming from the Technology Innovation Institute in the United Arab Emirates as an open source option, which is amazing and brilliant that they're doing that. And it's almost a.

Chris Adams: that like open night, it's not open source, it's not really open data, but it has the word open and you not, you don't typically associate open with lots of the kind of news that you might read about the, you about say the UAE for example. Especially when you think about things like say COP 28 and stuff.

And yet here you have an open model being released, it gives you an idea of just where the stuff comes from open source. It's really hard to predict basically. So we'll see what happens with this. And

Tom Greenwood: is. Yeah.

Chris Adams: folks who are curious about this, we've shared a link to Hugging Face, which is a kind of GitHub for machine learning stuff where there is a bunch of really interesting work by, it's one of the only machine learning and companies I know of with a climate lead.

Who specializes there and maybe one day we'll get, we'll be able to get them onto the show. Sasha Luccione, she's been creating some really good papers and yeah, this is one of the things that she's been doing actually. So we'll share a link to that for folks to look at. All right. Should we look at that last story then, Tom?

Because this one, one that you shared that, I think it's a really nice one actually. So this is Heating Swimming Pools with Service And this one, maybe you can talk about this one actually, Tom, cuz this was the one that you brought, brought along.

Tom Greenwood: Yeah, so this one really caught my imagination. About a year ago. There was, there was a company that was in the Netherlands that was building sort of small data centers on farms to heat greenhouses and that. That really caught my imagination as like a great way of doing things. And then suddenly this one popped up in the uk, this company called Deep Green, installing tiny cloud data centers at leisure centers, basically.

And it's this beautiful kind of symbiotic relationship between data centers that need cooling and. Swimming pools that need heating and particularly we've got this energy crisis, like energy prices have gone through the roof. A lot of local councils in particular are like really struggling with money.

Some swimming pools are being closed down just cuz like cost of heating the things. And then you've got this company that comes along and says, actually, like we could put a little data center in your leisure center and heat your swimming pool essentially for free. We'll pay for the electricity to run the servers and give you the heat.

It. It just seems like such a brilliant solution where you've got this huge tank of water that needs to be kept constantly warm, and you've got these servers that need to be kept constantly cool. It's one of those things where when I saw it, I was like, oh my God, how has no one thought of this before?

Chris Adams: Yeah, is there's a pleasing circularity to this and a kind of this term called free calling, which is usually around air, and this feels like the same idea. I'm a big fan of this as well. Actually. It also asks questions about what data centers should and shouldn't look like because we're used to data centers being, well not, the common narrative for big data centers or when you hear a data centers machine learning, you're thinking of like a football pitch full of machines, which is almost like a kind of big box out of town Walmart style warehouse full of things, like a bit barn. And this is the further opposite. It's integrating it into the fabric of the urban environment, for example. Yeah, this is a really nice story actually. Thanks for sharing this one.

Tom Greenwood: it's okay.

Chris Adams: All right, so we're just coming up to the half an hour mark and we are just gonna look at some of the events now, actually. So I guess the thing we should probably share is this is the 5th of June time of recording. And if you aren't aware, today is the UN World Environment Day and uh, today you probably. Us recording this now is probably gonna be a little bit late for people to know about this, but there is an, there is a virtual event taking place later on today, which will be recorded, which is by the Green Software Foundation called Green Software Revolution, where there's a number of people including the.

One of the head, the, the chair of the community group, Anne Currie, Asim Hussain, he of Principles Green Pindy Bhullar ubs, the at the bank, who is also a PhD specialist in sustainability. And Tamara Kneese, who was the lead researcher for the State of Green software report. So there's a bunch of stuff there and we've online for people to stream or look afterwards, after this.

And what else have you got? They've got this open source infrastructure meetup on in London on June the 14th as well. We have that there. Is there anything that caught your eye on this one, Tom?

Tom Greenwood: I guess my curiosity is more to see what's happening in terms of sustainability in this open source world that they're gonna be focusing on AI and deep learning for enterprise, and I guess we're at this space now where the whole AI world is like kicking off in a big way this year and there's, in the sustainability world, I guess the question is just, oh my God, how much energy is gonna be used by these things?

I think it's fair to say a lot of us are in this space where we're like, torn with these like amazing, exciting opportunities of the technology as well as some of the potential threats, both from an environmental point of view, from a like societal point of view. So for me, the, I'm just curious to see what the op open source world is bringing to this conversation, because obviously that's where a lot of kind of conversations around tech ethics often happens in the open source world.

Chris Adams: Yeah, I'm with you on this as well. It's easy to get really caught up on some of this stuff, especially if you start playing with some of these like chat tools, but, You're right. There is a non-zero footprint associated with this stuff, and even now it's actually quite difficult to find some of the numbers for this.

We have shared links to give some estimates of this, and now that you have an entirely open stack, presumably you could start coming up with some numbers, and yet I'm not aware of any services that still provide these numbers on a kind of per request basis yet. Like how Website Carbon has done, or even some of the stuff that we've done with like.

co2.js or so on, but it does feel like it's needed. So you are aware of when you're speaking to someone, what the actual impact of that when you're speaking? Not someone, when you're speaking to spicy auto complete, like it's not a person. We've gotta remember that.

Tom Greenwood: I, when I've tried these tools, I keep having to tell myself, don't say thank you. It's a slippery slope.

Chris Adams: I know what you mean. But if you, British, you're taught to, you're taught to apologize when someone stands on your own foot, let alone saying please. And thank you. So

Tom Greenwood: but I feel like saying thank you to ChatGPT, GPT is the slippery slope to ex machina.

Chris Adams: Could be. Alright. Alright, Tom, I'm gonna, I'm gonna park that there before we go down that scary rabbit hole, but I'm just gonna, we're just coming up to the end, so I'm just gonna come up with one of the questions. Are there any open source projects that you've seen that you might direct people to or anything that you'd like to direct people's attention to?

Whilst I have you here on this call before we head off?

Tom Greenwood: I guess the thing that would be great to direct people to is the W3C Web Community Group, which is, to be honest, I haven't been. Anywhere, like as involved as I, I would've liked to have been for personal reasons, but they're doing amazing work as a community, really exploring kind of all of the facets of what goes into creating a more sustainable Web for the benefit of the wider web community and producing some guidelines to help everybody.

So it's just something I'd, if you're interested, go and have a look at it. Get involved. There's some exciting stuff happening there.

Chris Adams: And W three three is the Worldwide Web Consortium, W3C and SustiWeb is the sustainable Web group.

Tom Greenwood: Exactly. Yes.

Chris Adams: Yes. Awesome. Okay. I think that one of the few groups who are really good at using Wikipedia or using Wikis to share links and things, cause I think there was a link shared about their massive list of resources that they were working on at the moment.

Tom Greenwood: Yeah. Yeah. They're pretty good at docu documentation. It's quite impressive.

Chris Adams: All right, we'll share some links for that for anyone who's so interested in that part there. Tom, I think this brings us up to the time that we have and I wanted to say thank you so much for coming on. I really enjoyed chatting with you again mate, and it's lovely hanging out again because now that I'm no longer in the UK it's much, much harder to come visit you, folk.

Come visit you in London, so, or wherever you are now. So, Once again, thank you very much for coming onto the podcast and I wish you the best. And just before I go is, where should people follow you or where should people look if they want to hear? If they like the sound of your voice or found what you had to say?

Interesting. Is there a newsletter or a website you would point people to?

Tom Greenwood: Yeah, so my company's wholegraindigital.com. You find that you look me up on LinkedIn. There are lots of Tom Greenwoods who run, and some of them run Web design agencies. But if you find the Tom Greenwood that runs Wholegrain Digital, then that's me. And then I've also got two newsletters. So there's the Curiously Green Newsletter, which you can sign up for at www.wholegraindigital.com/curiously-green/, which is a Green Web newsletter covering things that's going on in the world of sustainable Web design, but also Greentech more broadly.

And then I have another newsletter on subs called Oxymoron, which is about exploring the confusing world of sustainable business and how we reconcile the aims of creating a more socially, environmentally friendly world with the world of business.

Chris Adams: Cool. Thank you very much. I'm subscriber to both of those and I really do enjoy them. So thank you for writing them and once again, thank you for coming onto Environment Variables, Tom. Take care mate.

Tom Greenwood: Thanks Chris.

Chris Adams: Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit https://greensoftware.foundation/. That's https://greensoftware.foundation/ in any browser. Thanks again and see you in the next episode.

Fact Check: Colleen Josephson, Miguel Ponce de Leon & AI Optimization of the Environmental Impact of Software

31 maj 2023 | 42 min

This episode of Fact Check we ask the question, can AI always help us optimise the environmental impact of software? Host Chris Adams is joined by VMWare’s Colleen Josephson and Miguel Ponce de Leon to tackle this from their unique perspectives within the industry. They also talk all things sustainability in virtualization and networking and how this begins with green software. They also give us insight into how VMWare is tackling decarbonization within their own company.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Colleen Josephson: Twitter / Website
Miguel Ponce de Leon: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

O-RAN SC Projects | O-RAN [6:34]
Deep Cooling Solution in Data Centers | Office of the CTO Blog [19:45]
RIC Radio Access Network Intelligence Controller [27:28]
Kubernetes used in the O-RAN open source AI/ML Framework | O-RAN [29:32]
HotCarbon 2023 [38:06]
European Green Digital Coalition [37:26]
Evaluating Coupling Models for Cloud Datacenters and Power Grids | Adrien Chien [39:35]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:
Miguel Ponce de Leon: And the thing is, we're right in this maelstrom, this tornado of activity that's just got underway and just seeing how they fit together, it's not a perfect fit. I would say there. I couldn't give you, this is exactly the time horizon and this is how it's gonna happen, but I can tell with the level of funding, both from governmental agencies, from companies themselves, from research institutes to lots of public bodies and developers in their own time.

It's a great time to be in and around this space of developing software, but specifically for the delivery of green technologies as we see it.

I am your host, Chris Adams.

Hello and welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and in this episode we have two very special guests for an episode of fact check on Environmental Variables. From VMware, we have Colleen Josephson, and Miguel Ponce de Leon. Hey guys.

Miguel Ponce de Leon: Hey Chris. Hey.

Chris Adams: Okay, so for our listeners who are unaware, VMware has been a member of the Green Software Foundation since January, 2022. And we'll be talking a little bit about AI and the environmental impact of cloud. And we figured VMware spent a lot of time working in this field. In fact, they're synonymous with virtualization, but they also work in a number of other fields.

One in particular is networking. So this is where our backgrounds come from. Before we dive in, let's just actually give a chance for our guests to introduce themselves properly so they don't, so we know who they are. So I think, Miguel, if you just introduce yourself and what you do, we'll hand over them next to Colleen afterwards.

Miguel Ponce de Leon: Very good. So my name is Miguel Ponce de Leon. I'm director of Distributed Edge Intelligence in the office of the CTO here at VMware. So it's very much looking at research and innovation that happens from cloud. Towards enterprises and towards telecommunications. So networking, which is a topic that we're gonna talk about here, and very much looking at what it means and the impact of having the edge of the network and sustainability and connectivity to the cloud have had an impact on our products and services.

Chris Adams: Thank you, Miguel and Colleen. I'll give you a bit of space to introduce yourself as well. For folks who missed your inaugural podcast last year.

Colleen Josephson: Thank you very much. Yes. I'm Colleen Josephson. I'm a senior research scientist at VMware. I'm on the same team as Miguel, and a bit of news is that I'm actually transitioning to a full-time academic position at UC, Santa Cruz, where I'll be continuing to. To research sustainability, particularly in the space of low power in distributed systems.

Last time I was a guest on the podcast, I was the org lead for VMware, and I'm very pleased to share that Miguel, uh, I've passed on the torch to him, so he's very qualified and excellent, so I'm really excited to be doing this podcast today with him.

Chris Adams: Thank you, Colleen. All right, so for folks who are new to this format, Fact Check is a kind of format we use where we basically take a statement that people. Put into use in discussions around sustainability and software and we basically dive into it a little bit more to actually examine some of the assumptions because a lot of the time it really helps to really understand what some of the nuances behind this.

And today, the fact check statement we are looking at is this one, can AI always help us optimize the environmental impact of software? And Miguel, I might invite you to talk a little bit about this part first actually, because this is something I think you've heard at least one time before.

Miguel Ponce de Leon: More than one time, actually, I just as part of the introduction, just again, to say, look, I've been in and around telecommunication systems over the last 25 odd years from analog systems to two G, 3g, 4g, 5g. I know all these Gs that people hear about, but essentially it is and about looking about optimizing the way the networks are going to be deployed and used in the future, and very much we're seeing huge uptick in the use of artificial intelligence and machine learning to help with this optimization, and it is true to say that it can help, but I think there's a number of factors in order to ensure that what you use for putting the model together, for deploying that AI model and that, that you optimize the network, that you are getting the totality and that end-to-end sustainability and energy usage from all of that.

I think there's a lot of factors that we need to drop into it when we discuss that.

Chris Adams: Okay, so we've got a series of factors here and maybe if we start with one or two dimensions or key things that have an influence here, if we start there, maybe that'll help us frame some of our discussion. So maybe if we start with the first one, what's one of the bigger levers, Colleen, maybe if I hand over to you, if there's one that you would pick, what might, might it be?

Colleen Josephson: Training. Training is a very expensive process nowadays. Whether it's in the telco space, cuz again, VMware, as you hinted, we have a long history of virtualization in cloud, but that also has become very relevant to telecommunications. We need to train the models that we use to make these decisions to try and save energy and the process of having so much data.

And training it. It can be really power consuming. So I think one of the things that stands out to me is what is your anticipated energy savings once you've deployed this model? How long do you anticipate that this model will be good for? And do you need to retrain it? All of those you wanna have some idea of so that you can calculate whether or not it was worth the energy to train this model in the first place.

Miguel, do you have anything to add?

Miguel Ponce de Leon: Yeah, just, oh, I absolutely agree when it comes to, to train the model, but what I'd love to really highlight to everybody, even here listening is that these systems are being more open for everybody to get involved, more developers to, to get involved. Again, I talked about all those Gs before, but they were very much closed to certain vendors and certain companies that built it.

We now have Open-RAN this open radio access network, which means that we can use. More AI models and people's training of those models within systems and deploy them more readily. But it also means that we have to have newer understandings about, again, when we talk about wireless, we talk about power amplifiers.

We talk about controlling power amplifiers with AI models, and again, how it is that we're going to make sure that we efficiently train those systems in an energy efficient way before we even talk about turning on and off those radio head ends to save energy more often than not to save it so that, again, when we don't have so many people within an area, and therefore the radio doesn't need to be pumping out that signal the whole time.

Again, immediately you can see what the benefit is. But if we end up spending so much energy trying to train the model in the first place, then again, have we achieved our actual overall goal?

Colleen Josephson: For the listeners tuning in who might be less familiar with the structure of wireless networks, it might be worth giving a little bit of background on this power consumption. For telecommunications specifically, the biggest energy challenges are the radio access network or RAN, as Miguel was just talking about, and also data centers.

And the way that I like to think of it is that wireless communication is basically shouting energy into the void at high enough power so that the receiver can decode it. It's inherently consuming a lot of energy. So a lot of the challenge in this space is trying to look at how can we use various tactics, whether they're AI informed or not, to save power for this radio access network.

And also increasingly data centers.

Chris Adams: So there's one thing that you mentioned there, Colleen. So you spoke about most of the kind of areas where people are focusing on are either the data center or the radio access network, but not really the kind of pipes between data centers or the kind of backbone. Is that one thing there where there's already a fair degree of efficiency or where you don't see that much change at that part?

Colleen Josephson: I think the pipes between data centers that's inclusive of data centers. I'm sorry if I was not clear of that. Yes, so that the kind of the inter data center communication is definitely a part of the energy consumption end to end system that we need to consider.

Chris Adams: Okay, but there's a chunk about radio, but there's one area which is relatively new to us or relatively new, like this radio access network where cuz we've seen so much growth in cellular and things like that, and without growing so fast. This has been one thing that we are now looking to use tools like AI or something to make them work a little bit more efficiently than what people would manually be switching on things on and off or what, what, maybe we could explore some of the levers you actually have here, because it's not clear to me why you might be using AI in the first place for this part specifically.

Miguel Ponce de Leon: I might say a little something about that, Chris, because again, what you're gonna find with radio, what you're finding today with 5G, and you'll find it in the future as well, is that there will be more aerials. They will have smaller power outputs, but there'll be more of them. And with more of them, it means that you have to network more of them.

And with networking, that many nodes, you're going to have a, an optimization system in order to decide where to place them, when to place them, and when to leave them on and when to actually turn them off. Because if you have less of them, what you're actually gonna do is you're more or less, what's happening is you're leaving the actual radio head ends and the amplifiers, you're leaving them on constantly sucking up all that energy in juice. Now what we have is we have more aerials, lesser coverage, but because you have more, essentially it's more of a complex item than having a, a couple of network engineers monitor them themselves monitoring the issues as they potentially break or they have to be modified.

So it's a much wider range of, let's say, input variables that you actually have to, you cover off on.

Colleen Josephson: Just adding to your comment, Chris, earlier and Miguel's, absolutely right. We have a lot of input Variables, but. You were talking about kind of data center, network consumption and kind of the back haul, and I think I wanna revisit this topic of radio access network and wireless communication. And it's gonna get a little bit down into the physics of it, but if you have a wire going from one end to the other, fiber optic, those have much less loss.

So you can consume much more rapidly, get much more throughput for the same power consumption with a wired network than compared to a wireless network. With wireless you have all sorts of types of loss. The channel conditions are changing drastically all the time. So this wireless aspect is really one of the things that makes it, the radio access part of the network's a higher power consumer because you have this signal that you're sending out into the air as opposed to this little cable that has much more controlled conditions, you're gonna see something they call path loss. When you're going from the transmitter to the receiver, you're gonna see something called multi-path, which is where there are multiple copies of the transmitted signal arriving at the receiver. And these wireless networks have to be designed to overcome some of these challenges, and that's where a lot of kind of the radio access network consuming more power comes from, if that makes sense.

Chris Adams: I think that does. So if I was to maybe take a step back for folks who might be familiar with 3G and 4g, when we talk about some of this, if that was a model where you have one or two very large kind of transmitters or receivers, that's a shift to 5G or possibly even six G is many more smaller ones possibly with this.

And as a result, you have a kind of explosion of complexity. That's the, that's the thing that you have to manage, that you didn't have to manage before. And maybe the, this other thing I should ask you about then is that, it sounds really basic, but one of the ways when people talk about, say, 5G or six G being potentially more efficient or greener is just because it's easier to turn some of the system off rather than just having things blasting the entire time, 24 7. That's one of the assumptions that you're looking at?

Miguel Ponce de Leon: That's exactly it Chris. And just to explain the complexity as well, there is the possibilities. We're looking at them again with next G and six G systems. We'll, we're basically, there'll be a small aerial connected to your home, for example. But again, what you'll want to do is making sure that is controlled efficiently.

That it's looked after optimally. So for us, all of that is causing this need to self-organize the network in some way. That's the way we use terminology. There's a self-organization of where you put the frequencies on each one of the antennas and things like that,

Chris Adams: Ah, okay.

Miguel Ponce de Leon: having that kind of control remotely can be interesting and it could be.

For example, that your home aerial today could be with operator number one, we won't name the operators, but you can keep in mind that most, in most countries we have 2, 3, 4 operators in country, but an operator could pay you for that aerial that you've put on your home and could be optimizing it for your six G signal tomorrow.

It could be operator B that's using your aerial in some way, shape or form, and they may have a different way of optimizing. That same aerial for the signal that it's producing and sending out there and paying for your time to use it. So again, that's why we see a much more open system, a kind of an open ecosystem of how telecommunications will be actually provided in the future.

Colleen Josephson: And adding to that, there's some really great opportunities for, I like to talk about the power of open systems and virtualization, software definition, rapid prototyping. This really gives us an opportunity for that that we haven't had yet quite in this space of the traditional monolithic vendor stacks.

By opening it up to more members of the ecosystem, people can prototype these new clever applications, whether they be AI based or otherwise, and deploy them and get much more quick feedback on how much energy are we saving by trying this and rapidly iterate on those sorts of developments.

Chris Adams: Ah, okay. I'm really glad you brought it back to the AI part here for the initial fact check. So basically, faced with all this complexity, there's been this assumption that if there's something complex, we're just gonna throw AI at it. Like how Google did when they bought Deep Mind and then threw AI at their own data centers to twiddle the knobs instead of actually having humans do this.

And it turns out that maybe that isn't the best way because there's a significant impact. From actually training in the first place, and that may be larger than the inference in this case, for example. That's where the complexity lies in some of this, by the sounds of it.

Colleen Josephson: Yeah, definitely.

Chris Adams: I think that helps break down some of this actually.

The idea being that, yeah, there is an inference that we need to be thinking about and a training part that we need to be aware of. And if you spend your entire energy budget on the training part, then you need a huge amount of use to make up for that saving that you might actually have. And that's something we're not quite sure whether we'd actually see here.

And you also mentioned something that was quite interesting Colleen about a kind of decomposition of what might have been a kind of quite monolithic stack into a larger number of small look moving components, for example. So rather than just having one vertically integrated system, there might be a number of different players involved, or some of their work might be done in a cloud somewhere else or something like that?

Miguel Ponce de Leon: Yeah, so I will have to say a little something about how the networks are changing towards cloud native, right? So how communication service providers, or telcos, as we might call them as well, are moving towards more, and I'll say something, technologies like containerization, containerizing, the software within those containers, they're actually developed by different entities, by different developers, by different companies.

And the integration of them all still provide you with an end-to-end telecommunication system. And what I think is really interesting here, For, and especially for the Green, Software Foundation is around the carbon impact of the software that's developed and how that's tested in a CICD a continuous integration, continuous deployment development environment to see the impact of the overall delivery of these individual players who have been plugged together to provide a communication system as we go towards next G. And this is something, again, a lot of research is going into certainly around communication service providers in Europe, they're saying, okay, I can get vendor A, B, and C. I'll plug them together, I'll put them in a containerized orchestration environment.

But they're also asking their question, not just performance, not just security, but what is the sustainability and energy impact? And if B is not written in an efficient software way. The software way is not energy efficient and not secure. Well, I'm afraid vendor B will have to pop out and an alternative to that is, is being looked at how you pop that in.

And again, all the factors around performance, security, and sustainability are important factors with these products actually going online. So again, that's the type of research that we're seeing happening at this moment in time is how do you do that? How do you measure the baseline around that, especially in a cloud native world?

How do you get the baseline and then how do you take actions Because now that it's so pluggable and playable, like I give the example, I can put A, B, and C, but now I can take out B and put a replacement in. We really have to be cognizant of the delivery of, of that service too.

Chris Adams: Okay, so it sounds like you're implying there are for want of a better term, as long as you are honoring a particular contract of an API. The idea would be that if you have a stack of technology, here's a chance to swap out one part of your stack to make, to replace it with a greener part of your stack.

And hopefully that will result in a kind of more diverse, healthy ecosystem that you'd be working with here, for example, where there's a kind of chance we would compete on transparency and compete on sustainability in this instead of just actually on performance and cost, for example. Cuz in many cases that's the world we we are living in now, right?

Miguel Ponce de Leon: Big time. Big time. And I do know of operators that now are looking at can they provide their clients with a energy sustainable service? Like they'll allow their customers to actually choose. So you can choose your service and it has A, B, and C, or you can choose your service with A, Z and C. One is more green.

One is using energy in a more efficient way. And is that what your company decides to do? Then they can offer it in that way. So that's also what I see from a research perspective, what we're seeing at companies and entities around the communication space looking to, to address. But I know Colleen has some examples too, around this, right?

Colleen Josephson: Yeah, so bringing back the data center thread, one thing that I thought was really interesting in some of the work that VMware has contributed to is that we saw the data center portion of a network's power consumption double between 4G and 5g, and we're expecting that trend to continue and become even more pronounced going into six G.

So that's why we say that the two big things to think about are the radio access network and also the data centers. And that brings to mind one of our data center success stories where it comes to AI and energy savings. One big source of emissions is power that's drawn for data center cooling in particular.

And to tackle this, we've actually partnered with Intel and a company called Clark Data on a solution called Deep Cooling. It uses big data and AI to model various physical parameters in large data centers. Things like power load heating, required cooling, and it uses insights gained from this modeling to predict the results of changing computing workloads and then automatically adjusts the equipment parameters to optimize the system cooling, and it's implemented right now in several large Chinese data centers, and it's been effective at helping customers significantly improve power usage effectiveness, and reduce carbon emissions. I think the figures I have here saving 18 to 25% of electricity for cooling.

This is one example of kind of an AI success story. But again, you always have to think about when I go to train this model, when I go to use it, what is the story going to be when we consider end to end, not just the immediate usage.

Chris Adams: So this maybe might be a chance to talk about some of the metrics you might use with this then, because one thing we spoke about in a previous episode was that there are various researchers who are talking about AI and they're talking about. The idea that you might attract the energy embedded into that model.

Just the way we talk about embedded energy in, say, building physical hardware, there might be an idea of like energy embedded into a model before you actually use it, for example, as ways of listing this stuff. And when people talk about that, people talk about, say like the energy usage, but people also talk about the carbon impact of that part as well.

And this kind of speaks to the idea there's maybe another lever, not just energy itself. Is that maybe something you might wanna talk a little bit? More about the fact that it's not just energy, it's the kind of energy or how green or dirty the energy might be, or what levers you actually have there to affect that.

Colleen Josephson: Yeah, that sounds like it's getting a bit into some of the green load balancing or carbon aware workload migration that we talked about last time, and I'm pleased to share that our work on that has been progressing. It's still very much in the research phase. There's not much new that we can publicly share yet.

What I can say is that the calculations from the Mobile World Congress work that we did a year ago, which found that you can have carbon emission savings of up to 50% by more intelligently placing your workload. Depending on where the municipal power is greener or less green, they appear to match our prototype results.

So we're preparing to submit some research writings on this work. So stay tuned. Hopefully much more will be publicly available soon. And we also have some exciting collaborations on this front, looking at how carbon wear load balancing interacts with energy grids and making sure that data centers that begin to implement these novel solutions, they remain good energy citizens and don't unintentionally negatively impact our energy systems.

Chris Adams: So this sounds like we might be thinking about AI in a few other places then, because there's this phrase that I haven't heard people use that often. Being a good energy citizen as a data center, could you maybe explain that bit more because most of us, we know that data centers use energy, but there's more qualities to the energy than you might have there, and you might not know about the density of demand or load, for example that might be worth explaining.

Miguel Ponce de Leon: So one of the things I can mention is that we are working with, uh, grid utility. In Ireland and with that grid utility that also hosts a data center. As it so happens, we're also working with an accelerator program, a program that is helping startups to look at how you can not just link, but actually be able to take the correct measurements from the green sources, the wind farm locations, and the usage within the data center for its workloads. So again, here it's about leveraging, not just the research we'll say, that would come from research performing organizations or from the offices, the CTO of VMware, but also looking at startups and startups within the space and to link this. And that is helping the utilities understand what type of usage.

And imagine it's a utility that has their own data center. So it's helping them be a good citizen, even within their own environment. But being able to measure it and then being able to take action on it, right? Because that's the important thing is, okay, you've got your baseline, but what can I change about what I'm delivering within that data center, even down to the containers?

How can I move my clusters and pods? Overt and maybe consolidate some of the pods. We're even moving some of that research as well to look at, even with the pods being available, how many of the CPUs are they using within the cluster? So again, it's about being able to help data center owners being good citizens around that space.

Chris Adams: Okay, so there's one thing that came out of that. We spoke before about how previously with three and 4G you might have basically a series of very large antennas blasting stuff out all the time, but now you'd be shifting to a lot, a kind of constellation of smaller antennas, which you might have to spend some time coordinating and time and energy in terms of coordination costs to figure out which ones to turn off so you can provide things working quickly, but also things being more geo efficient.

It sounds like there's something like that on the data center level as well, like where we might have had data centers, which have been a steady 24 7 load, but there's actually scoped scale it down or up a bit. Is that what you're proposing?

Miguel Ponce de Leon: That's exactly it. Chris, you're painting a great picture here of the interconnectivity of it all. But yes, cuz you know, again, as Colleen was saying earlier, we have two main parts to the network. There's the radio side and then there's the core side. And that what we're doing here as part of the VMware team, as well as collaboration with a number of other companies in it, is attacking it from both sides.

And again, looking at how you can really look at that end-to-end element of actually delivering those potential energy savings in order to reach some of the goals. As Colleen was saying earlier, I could say the telecommunications world is really looking at reducing by 2030 and even beyond then by a number of factors from where they are today.

You need to look at all facets of how that's delivered. So yeah, that's, and that's what you hear from what we talk about when we're looking at the startups. We're looking at how to link both the wind farm energy to the actual data center energy that's used.

Chris Adams: Okay, so I can see why people might just say, this is so complicated and there are so many moving parts. I don't wanna think about it. I'm just gonna let AI think about this. And that's why there's this assumption that, yeah, that's gonna be doing the optimizing, but there's an impact in its own right to do that.

And there may be other ways of doing this. Maybe we could talk a little bit about some of the projects that are either in the open or in the world that people might point to allow people to start playing around with some of this stuff for experimenting. Cuz Miguel, you mentioned open RAN. So my assumption is there's an open standard or some open source projects that people might play with and I know that we've spoken about in previous episodes, some software or some tools on the kind of data center side. Maybe Miguel, if I spoke to you about some of the open ran style, things like that people could point to and look at, maybe to experiment with themselves. What projects on GitHub or GitLab or things like that might you point people to if they were interested in this kind of very dynamic new network and data network world?

Miguel Ponce de Leon: Sure, and I'm sure we can give some links as well. So folks, and it's always the easiest thing, right? We give the show notes and we'll give some links off to it. But there is again, to help control the radio access network, there's a thing called the RIC; the short name for it, but it's radio, the radio access network intelligence controller, and there is an open source version of this.

And the RIC uses a thing called, uh, cube flow. So this is a way of being able to host your machine learning model in a way that will be able to interact with the radio network. So there's a couple of open instances there where if you have a Kubernetes cluster, once you have the, uh, open source rig from the ONF the Open Networking Foundation that you can develop, again, some in-house terminology here, but we have what's called the X app and the R app.

So the X app is this realtime application that can immediately, uh, basically turn on and off the radio head ends to help with that energy saving. Or we have the near realtime app, which is more or less, you can spend a little bit more time considering, given the complexity of the number of aerials that are out there, about how you'll deploy and which ones you would turn on and off.

And that's somewhere, again, a number of easy programs written in Python that if you wanted to get up and running and in doing so, you could have an impact on a future well known operators network in your area. Because the whole system is becoming far more open and the app that's developed on the open source projects I've just mentioned, you could then put them on things like the VMware RIC.

We offer a one that's very much telco grade gets deployed in the network, but the app that you've developed in ai, that the model that you've developed can be deployed in the same way. You don't have to wrap it up much more differently than to do so there. There are some relatively easy touch points to, to get involved here.

Chris Adams: Okay. And Colleen, we spoke before about junkyard data centers and I think last time you came on we, you were talking about some of the research that you folks are doing with the VMware to start tracking and measuring the savings, the carbon savings before. So if there, are there any kind of data center projects or orchestration projects you might point people to?

Because I think I've spoken about things like eco visors and stuff before, but I'd love to know what else is going on here actually,

Colleen Josephson: Yeah, I shared some overlap with what Miguel was talking about, to be honest, and I think containerization, Kepler very important project. Yeah, so Kepler is this energy monitoring and tool for Kubernetes containers and you can hook this into some visualization systems. You know, that's one kind of open source project for monitoring data center, energy consumption that I'm aware of.

Very important in that area. Not AI specific, but those two areas are a frequent topic of conversation among people who work in cloud and data centers.

Chris Adams: Yes, Kubernetes efficient power level exporter. I found the link for it actually. That's what it stands for. There's a project. Where there's actually an ongoing conversation and some of the kind of Green Software Foundation I can't remember which repo it is, specifically where Adrian Cockcroft has been mentioning this idea of Kepler as one of the mechanisms to allow kind of minute by minute reporting at a cloud level so that you can actually get some of the numbers to optimize for carbon or optimize for energy usage.

Cuz this is one thing that you don't always have for all your providers, and that's one of the kind of underlying piece of technology used to expose these kind of resources usage figures for each of these kind of pods or clusters of computing and things like that. So we've got about five or six minutes time left, and I did want to leave some space to talk about some of the kind of wackier stuff that we didn't get a chance to talk about last time.

Just before. Towards the end of last episode, we spoke a little bit about things like junkyard data centers, which were like data centers made of various end of life computers and things like that. And Colleen you mentioned different kinds of either zero power or low power. Things like soil powered batteries and stuff like that.

And seeing as we've gone into all this kind of industrial level stuff, I figured it might be worth an interesting to look at some of the other level, like the really low, ultra low power stuff. Is this what you're gonna be going to study or is this some of this work you've done before? Because I think it's gonna be interesting to some of the crowd here, realizing that things happen at the bottom end of the scale as well.

Colleen Josephson: Yeah, yeah. This is the bridging of the two worlds and the data center and these big monolithic or not so monolithic anymore systems are really important to consider, but. We use these telecommunication systems to hook into much smaller devices, tablets, smartphones, and ever increasingly IoT. And what is really interesting about these IoT and smartphone, smaller user devices is they are special because they've been designed to be power efficient.

So the carbon footprint for them is significantly larger in the manufacturer phase. Compared to the device use phase. So they, they have a much higher embodied carbon footprint proportionally than the energy consumption that they use. There's some really interesting work going on here for how can we lower the embodied carbon footprint of some of these massively, or we anticipate them to be massively deployed miniature systems.

And one of the ways that I've worked on that's out of the box is batteries. So we have some ultra low power communication devices that, you know, we can. Begin to use to do something called simultaneous sensing and communication. And one of the bigger footprint aspects of some of these systems are the batteries, traditionally speaking.

So if you can minimize or eliminate the need to have a battery, then you can significantly reduce the embodied carbon footprint. So one of the things that I've looked at is can we actually harvest energy from the soil itself? So this is really early stage research that we're starting to look at UC, Santa Cruz, and it hooks into something called intermittent computing, if you've ever heard of it.

And it's this idea of computing systems that don't constantly have power available and the paradigms that the system operates at. We design data centers, we assume that power is always going to be available or we did. And so now if we have to be much more dynamic and on our feet about when power is available, we have to be able to very rapidly save progress.

Go into power shave mode and then rapidly spin back up again when there's power available. So the intermittent computing community has been really active at connecting the ultra low power and ultra far edge and hooking it into our core networking and traditional communication systems. I can add, if we're looking for off the wall ideas.

Hot Carbon. The first inaugural hot carbon workshop was last year. I'm pleased to share that this year there will be a second iteration of this workshop. One exciting development is that it's tentatively going to be sponsored by ACM Sig Energy this year, in addition to VMware's continued support. So I'm actually working as the publication chair for that workshop.

The submission deadline was yesterday, 5 21. The workshop itself will be on July 9th. Just shout out to those of you listening. So tune in for what's sure to be a very interesting cutting edge work in the sustainable software space, and it's going to be a hybrid workshop. The physical location is in Boston Mass, but now if you go to the website, hotcarbon.org, up very soon should be a registration link so that you can sign up to attend virtually or in person if you happen to be in the Boston area.

Chris Adams: Cool. Thank you for sharing that. Colleen, for anyone who is on the fence, I virtually attended Hot Carbon last year and I basically plundered that list of people for guests for this podcast cause there's loads of really interesting projects going on there. There's also some really nice stuff I. Colleen I just wanna ask about this idea of intermittent computing, cuz it sounds like it's like really almost super serverless. So the idea that there's, it's almost like battery-less, basically the idea that you do the things scale right down to zero and you just basically don't really work until you've got energy coming back in again.

That's the idea behind that.

Colleen Josephson: I think there, there are definite connections to battery-less. An intermittent computing paradigm can work whether there's a battery or not. Trying to work around how charged that battery is and turning off when power is reducing. But yeah, it's a very prominent area of work when you consider battery-less computing and just to connect everything end to end.

You might have these very low power sub microwatt, in some cases, devices at the ultra far edge, but you need to have something. That brings that data back to the cloud, and this is where you have more traditional edge computing, like maybe a server that's at a farm or some things that people are starting to think of are edge data centers that are potentially even mounted to drones, or edge 5G that's mounted to drones.

So lots of really near edge and far edge paradigms.

Chris Adams: Okay, so it sounds like as we have moved from monolithic, gigantic computers to things becoming smaller and more distributed, there is a coordination cost, which is why people often talk about, this is so complicated I'm just gonna hope that AI solves it for me, which is where some of these ideas came from, that of course, you can use AI to automatically work out, erase the environmental impact from computing.

I hope that one thing we're taking away from this conversation is that no, it's a bit more complicated than that and, but there's lots of exciting rabbit holes to drive down. Folks, I've really enjoyed this conversation actually, and I think we're coming up to the time that we have here with the la Do you folks, if in the last few minutes that I might actually ask, are there any projects or things you might point people to that we haven't spoken about that you'd like to give a shout out to before we wind up?

Miguel Ponce de Leon: Okay, because I'm based in Europe, right? There is a coalition that's working together with a number of working groups, really looking at how all the things that we just talked about, but with more specifics. Cuz I know you, you wanna, do you use the word fact checking around what we're trying to achieve here.

So there's a number of working groups in, in Europe where companies are coming together and really looking to do this Now. One of them is the European Green Digital Coalition, right? So the EDGC I'll send you on a link and around it as well. But thi this is a space where, We're having to look at, cuz at the end of the day, we will standardize around what way you're going to measure these changes.

What way It's gonna have an impact on business in what way? When you do offer this fantastic green telecommunication service, what you're saying is energy aware. But there's gonna have to be standardization around whether or not that is actually the case. And so there is a good bit of work in and around this.

Again, I almost seem to use the word research and it's happening. And the thing is, we're right in this maelstrom, this tornado of activity that's just got underway and just seeing how they fit together, it's not a perfect fit I would say there. I couldn't give you, this is exactly the time horizon and this is how it's gonna happen, but I can tell with the level of funding, both from governmental agencies, from companies themselves, from research institutes to lots of public bodies and developers in their own time. It's a great time to be in and around this space of developing software, but specifically for the delivery of green technologies as we see it.

For me, that would be the big pointer, and I'm hoping that someone, one or two of the topics that we've mentioned here, would give Annie developer an opportunity to actually, like you were saying earlier, Chris, have a look at a GitHub project. Have a look at being able to develop a small model, some code, and have somewhere to actually apply it where it will have an effect on your own services in the future.

So certainly that's what I'm excited about and why I'm working hand in hand with colleague on this particular topic.

Chris Adams: Okay. Thank you, Miguel and Colleen. Just as we as we wind out, what would you draw people's attention to apart from hot carbon, of course, which is freaking awesome.

Colleen Josephson: Yeah, I've got two, two things. For those of you listening who might be interested in doing a bit of a deeper dive on telco sustainability, here's, there'll be a i, I just shared a link to our VMware Telco sustainability white paper, so that goes, Into more detail on some of the challenges of the radio access network, the RAN and the data center, and then also coming back up the stack to this idea of data center being good energy citizens.

I want to name our collaborator Adrien Chen at the University of Chicago. He's been active with us in this area and in some of these collaborations and this paper here. Evaluating coupling models for cloud data centers and power grids, that that work is really great for showing some of the problems of how data centers can disrupt the grid.

So I encourage people to go check that out if they're interested in that topic as well.

Chris Adams: Cool folks. We began talking a little bit about a fact check. We. I think we've realized that you can't automatically assume that AI will automatically reduce the environmental impact of everything. And we realize there's quite a lot to it. But I've really enjoyed driving down all these rabbit holes with you folks and uh, yeah, thank you so much.

We'd love to have you folks come on again. Yeah, folks, have a lovely morning or afternoon wherever you are in the world and yeah, see you around. Take care folks. Tira.

Colleen Josephson: Thanks again, Chris.

Miguel Ponce de Leon: Thank you Chris

Chris Adams: Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get to your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation In any browser. Thanks again and see you in the next episode.

The Week in Green Software: DIMPACT with Andri Johnston

24 maj 2023 | 29 min

Host Chris Adams is joined by Andri Johnston, Digital Sustainable Lead for Cambridge University Press and Assessment as they talk about using DIMPACT to calculate the carbon impact of digital publishing as well as news from the world of green software concerning one acronym; ESG and one portmanteau; LightSwitchOps. They also cover some upcoming events and we learn about Andri’s love of books!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Andri Johnston: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

ESG as a lever for sustainability impact | Elise Zelechowski Thoughtworks [2:49]
Using DIMPACT to calculate the carbon impact of digital publishing at Cambridge University Press & Assessment | Andri Johnston & William Pickett [7:31]
How Would the Business Benefit from Your Greener Java Application? | Holly Cummins at Devoxx UK [14:53]

Resources:

Thoughtworks.com/socialjustice [3:33]
ShareAction [6:16]
As You Sow [6:43]
The True Climate Impact of Streaming | Netflix & Dimpact [8:49]
Virtual conference carbon footprint revealed | Cambridge.org [11:23]
David Hsu on mastodon.energy [13:53]
The Baking Forecast UK [21:20]
Effects of Internet-based multiple-site conferences on greenhouse gas emissions | Vlad Coraoma

Events:

Green Infrastructure Meetup GfK [x] Green Coding Berlin - 31 May 6:30-8:30PM | Green Coding Berlin [23:19]
UN World Environment Day: The Green Software Revolution (Virtual Event) - Monday June 5th | GSF [24:02]
LF Energy Summit 2023 (June 1 – 2, PARIS & Virtual) | Linux Foundation [24:46]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript below:

Andri Johnston: We have no control over what device people are using or what energy grid they're using, but we do have control over how we're building our software and making it as energy efficient as possible. And in my experience, the moment we were able to show that those numbers to board execs and that kind of level, it was like, oh, okay, so we actually can do something.

I'm your host, Chris Adams.

Hello, and welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and in this episode we'll be looking at the news from the world of green software that involves ESG and a portmanteau LightSwitchOps.

Not only do we have this, but we'll also have some exciting events for you to attend. As always, we usually try to introduce our guests before we get started, and this week we have Andri Johnson joining us. Andri say hi.

Andri Johnston: Hi. Thank you so much for getting me on the podcast. So yeah, I'm Andri Andri Johnson. I am the digital sustainability lead at Cambridge University Press and Assessment. And a little bit about me. I'm a very keen runner and gardener, however, that's taking a backseat right now cuz I am 36 weeks pregnant and I've deep dived into the world of sustainability and babies, which is a very weird world, but really exciting.

So, yeah. That's me.

Chris Adams: Cool. Thank you. Andri, you seem to be continuing our trend of guests who are just about to have kids before they disappear onto the other one. Yeah. Sara. Sara Bergman who was on before. She was also just coming to the end before she popped a sprog as well. And uh, yeah, uh, nice to hear. I wish you were the best for the second half of this actually.

Okay, so for everyone who is, who listened to this first time, this is basically a news roundup show. So what we do is we look at some stories that caught our eyes, we share a couple of records about them, and then. Basically move on from there and if we discuss anything or any links come up that might be interesting.

We'll make sure we add it into this show notes, so if there's any acronyms or anything like that, we'll work to point to some either definitions or lists for further reading. Okay. Andri, will that further ado? Should we look at the first of these stories?

Andri Johnston: Yeah, sure. Let's go.

Chris Adams: Okay, so this first story is ESG as a lever for sustainability impact by Elise Zelechowski at ThoughtWorks.

Elise is actually the co-chair of the policy working group, and this is a blog post that she's written about some of the approaches that ThoughtWorks is taking around esg. ESG for the uninitiated stands for environmental, social and Governance, and it's a kind of set of terms or a kind of framework that people use to assess a company's performance in terms of sustainability and across a number of dimensions.

ThoughtWorks has been talking about some of this for a while, and they've been probably one of the, in many ways, kind of leaders in this particular field. They've literally got a url, which is thoughtworks.com/socialjustice, to give you an idea of where they stand on a bunch of this stuff. And Andri, Andri, you had to look through this piece.

Is there anything you wanna add on this one?

Andri Johnston: Yeah, I found it really interesting and I'm really encouraged by the fact that especially on the tech side, more companies are starting to look at ESG. Not just reporting on carbon emissions. So at Cambridge University Press and Assessment, ESG is very important to us because we are linked with the University of Cambridge, we have to comply with a lot of social issues as well.

And our environmental team and our EDIB team work very closely together. So for me, this was really interesting that Elise was talking about how companies need to evaluate themselves on all of these factors, and also working with your suppliers and that kind of thing, because that's, that's difficult.

Like it's one thing, getting your carbon emissions from them, but really making sure you align with their ethics and it aligns with your ethics. I think that's really important. So I found it really interesting and very encouraging that more companies are starting to do that. Yeah.

Chris Adams: I wonder how much might also be tied to, so ThoughtWorks went public this year and uh, there are over the last year or two, you have seen. A real uptick in ESG as a kind of buzzword in investment circles and in particular, say Europe for example, things you might consider kinda like green investments are able to get access to capital in ways that other groups don't.

Particularly with something related called the European Taxonomy, which is specifically about saying, okay, these things which are green, Which we have decided society needs more of, get access to lower interest rates so they can borrow cheaper compared to other ones. And, uh, this is, there's a real push from investors asking about this in many ways.

Sort of getting an idea of what they're exposed to from a climate point of view, but also from a kind of governance point of view and things like that as well.

Andri Johnston: I guess there's also something around how this sits with your employees, because for us, we're technically a not-for-profit and for us, a lot of people come to work. At UP&A because of the ethics around it, the university, a couple of years had to de divest because of connections with big oil companies.

And for us, it's the same. Our internal colleagues will ask us questions, who are we partnering with? What's their ethics? How does that align? So I guess there's both the investment part but also thinking about your colleagues internally and how they feel about your sustainability as a whole. So using this as a measure.

Chris Adams: Yeah. This is definitely a thing. Actually, there's a few interesting kind of jump off points here that may be going outside the remit of green software, but a problem might be of interest. I know in the UK there's an organization called ShareAction that's been doing a. Bunch of really interesting work on employees and where their money's invested, because a lot of the time if you're working for a company where that you feel that you believe in or you want to be spending some time with, you'd, you'd like the the money that's being put aside for you to also be doing things that you also believe in, rather than propping up fossil fuels or doing things like that.

Especially when the science is spelling out that we really need to not be involved in that as well. Okay. And there is also an organization called, I think As You Sow they've done some really interesting work with shareholder activism, specifically about putting shareholder resolutions to get a large organizations to move more quickly on sustainability.

There's some really fascinating work that they did with Microsoft actually to basically really push for some more circular electronics. And what we'll do is we'll share a link specifically about how basically some engagement at a kind of shareholder level using the kind of shareholder resolutions mechanism was used to basically say, let's actually look at the environmental impact of say things like, say Microsoft Services lasting longer, and is that actually a net gain for us?

And it turns out that it was, it's a really nice and interesting story because yeah, it turns out there are lots of ways that you can push for things rather than just coding. And I think that's gonna be the next thing we might touch on actually, which is the story about using DIMPACT. So DIMPACT. Or maybe you can help, how to pronounce this.

What's

Andri Johnston: I think we just say DIMPACT, but yeah, it's digital impact, so it can be either, but yeah, DIMPACT.

Chris Adams: All right. Okay, so this is using DIMPACT to calculate the carbon impact of digital publishing at Cambridge University Press and Assessment. This is a story about the use of DIMPACT, which is a model that was developed by a consortium of different organizations, largely to understand the environmental impact of basically digital services and Andri just before we started the call, you were telling me a little bit about some of the origin stories of this and how some of this came about, and this is some of the work that you've been involved with, right.

Andri Johnston: Yeah. Yeah. DIMPACT was created by a consultancy company called Cornerstone based in London, in conjunction with the University of Bristol and their computer engineering department. And it basically, the first kind of version of it came from the BBC wanting to understand what the carbon footprint was of their online advertising.

And from there, the tool is built with each different company using different model, and for example, there's a digital publishing module, which we use, but there's also a video streaming module, which Netflix, for example, used last year, Netflix and the Carbon Trust with Cornerstone published a white paper on video streaming, and it was really interesting debunking a lot of the myths around video streaming and where the majority of the carbon emissions lie.

But the whole tool is based on the same kind of methodology. It's just different workflows depending on what type of business you're in. Advertising, streaming, or publishing like us. Yeah.

Chris Adams: I see, and you spoke about this idea that there's maybe one model that's been published that, that we can, we're gonna link to, this is open for people to look at, but the idea was that get based on the use case, you might want to use a model slightly differently. So streaming might use a digital infrastructure in a different way to publishing, for example, because there's maybe a different environmental impact from watching something or streaming some files compared to dynamically generating a page every single time or something like that.

Andri Johnston: Yeah. Yeah, exactly. Exactly. And I think what's fascinating and what I find really good about the tool is the granularity it can go into. So on a publishing side we publish the majority of our content online, so all of our customers access our books, our journals, our textbooks online, but there's also a content production kind of side to it, and that's what I find really useful about the tool.

It doesn't just look at the hosting and the accessing of your webpage. It also looks at your entire workflow of your content production and storage, which I think is quite unique. And that's why we started using it because it allowed us as publishers to look at our complete workflow from the start of producing your content for book, for example, all the way through, let's see, someone accessing it online.

So yeah, that it gives you a very good overview of your complete carbon footprint.

Chris Adams: Oh wow. So that's quite a large boundary. So rather than just saying, I'm just gonna look at the website you're talking about, okay, if there's maybe another batch job, or there's another whole edit editing subsystem, you're keeping that inside your kind of boundaries at where, right. Okay.

Andri Johnston: Exactly.

Chris Adams: All right. Okay.

What we'll do, we'll share some links to some of the underlying assumptions and ideas behind how these models work, but there's one thing I also wanted to actually talk about, cuz there's another link inside. This was this idea of using some of these related tools to understand the environmental impact of virtual conferences as well as the publishing part.

Andri Johnston: Yeah. Yeah, because we'd partnered with DIMPACT to do the publishing side when I was doing the article that the will link to it was a pilot project that we did in the academic division for at CUP&A and in the academic publishing world conferences are huge and we had a lot of people coming back and saying, no, we should go back to in-person conferences.

But at the same time, a lot of pushback and saying, no, we should do it online for environmental reasons. So we actually really wanted to understand what the impact was of a virtual conference. So we partnered with DIMPACT again and they. We're developing a module for virtual conferences. They're similar to the publishing and advertising module.

They developed this module for virtual conferences and we took a two day online conference with about a hundred participants per session. We had eight different sessions and we had people from all over the world joining us. We had about 500 participants in total over the whole time. So it was quite big.

Um, and we actually found that. Over the two days, the conference emitted between 15 and 20 kilograms of CO2. For the entire conference. We would've had at least 70 to a hundred people fly in from the US to the conference. It was just really fascinating to do that calculation and to see what the impact was and was never to say.

In-person events are bad. It's just so you have the tools and you have the knowledge to be able to say, this is what the impact is of our online conference for this specific conference. We made the choice to do it because of this. So yeah, I think it's great to have that knowledge.

Chris Adams: Okay, so in this scenario, Streaming is not the new flying. Flying is still the new flying.

Andri Johnston: Flying is still the new flying. Yeah.

Chris Adams: Okay. And what, you said 15 to 20 kilos of co2? Right. Okay. So if we assume, typically this is a kind of common factor that I've seen shared, and we'll share a link to. Typically if you have a cappuccino or a cup of coffee, you're looking at maybe between two to 400 grams of co2 just for that cup of coffee here.

So basically the coffee break for a hundred people was roughly the same carbon footprint as the entire virtual conference, just for the assumptions are more or less in the same ballpark here. Okay. All right. That's a useful thing to bear in mind when we're looking at some of this then. Wow. Okay. Also, there's one thing I might share related to this, now that you've actually just spoke about this.

There's a chap on mastodon.energy, David Hsu. He's done, he's actually been sharing some really interesting work about that's been happening at MIT about their approach with flying and basically academic. The whole kind of idea of, okay, if I'm an academic or I'm doing a bunch of, I'm involved in lots of conferences, where is the environmental impact?

We'll share a link to that deck as well. Cause it's really fascinating and it shows a bunch of the ideas that are being used in MIT to basically get handle on the environmental impact from the whole life cycle of creating what we might refer to as knowledge products like this. And you talk about lots of measures, like say, applying an internal carbon levy the way Microsoft does, but they also talk about things like, say literally just public tracking of this stuff. So you get a rough idea of, okay, at a team level, these are the miles flown for this kind of work and things like that to basically make these things more visible so you can create, uh, an awareness of social proof around this stuff. Okay. Thank you, Andri. That was great.

Andri Johnston: it sounds fascinating. Yeah.

Chris Adams: All right.

Should we look at the next story? Okay, so this is Holly Cummins at Devoxx UK. How would you or the business benefit from your greener Java application? So we spoke about a talk by Holly Cummins at Red Hat about greener Java applications. And this is a talk that Holly was speaking about and this is where this notion of LightSwitchOps has come up.

I think she's also Holly's on a bit of a roll, cause she's mentioned she's come up with a bunch of really quite memorable terms like cloud zombies and things like that here. Is there anything that kind of caught your eye when you were reading this stuff?

Andri Johnston: I think I really liked the idea of making colleagues more aware of greener Java and greener coat because, so taking it back a little bit to the work we did with the DIMPACT tool, as we know in the majority of digital products at the majority of our carbon emissions lie with our end users, and there's still a little bit of a back and forth whether we should be reporting on that, whether we shouldn't, is that part of scope three?

But what we realize is that's the one place you can actually start making a difference in the way that you. Build your software and that, and especially the way that you are creating the content and these pages that people access. And I think this is a kind of thing that you need to provide to your teams as a tool to say, we can make a difference in how we're building our website.

So for me that was a really easy way to say, here's a tip, here's a small thing that we can do as digital product people to make our websites more energy efficient at the end of the day. So for me that was really good. It's that, found it really interesting. I was like, I definitely wanna share this article with my teams.

Yeah.

Chris Adams: So I have a question here about borders, cuz I just, before this, Cost. I did a bit of research on the model, and there are various ways of modeling the environmental impact of a digital service. And the actual end user part is often one of the things which is can seen as a kind of stickler. Like for example, Mozilla, when they report on their emissions, 98% of their scope three, their supply chain carbon footprint came from the device of end users of this.

And basically, this is an organization which reported their carbon footprint of around 800,000 tons. Where 98% of it came from end users, whereas large other organizations like say Google and Netflix, they haven't included these numbers themselves for some of this, and Digital Impact does, but it includes the usage, not the kind of embodied emissions that went into making the

Andri Johnston: Yes, that's correct. Yeah. Yeah. But I think something that we need to take into consideration and why, I think for us it's important to report and to calculate those emissions is because, and Holly points to the carbon emissions of different regions for data centers, which is very true. But also carbon emissions are different for different end users depending on where they're based in the world.

So we know that over 50% of all our customers are not based in the global south and don't have access to grids that are gonna greenify anytime soon. It's really important to understand that if that's where the majority of our carbon emissions lie, we have no control over what device people are using or what energy grid they're using.

But we do have control over how we're building our software and making it as energy efficient as possible. And in my experience, the moment we were able to show that those numbers to board execs and that kind of level, it was like, oh, okay, so we actually can do something. So yeah, I think it's just important to keep that in mind and not just say, we're not gonna report on it because it's not our problem.

Because I do think in some ways it's our responsibility too, to still work on it.

Chris Adams: Yes. There's also one thing you just mentioned actually just springs to mind how there's actually a paper from, I think the Limits conference a couple of years ago that was, that came up in some of the discussions of the Green Software Foundation. When we're trying to figure out, okay, where should the boundary be for some of this, there are decisions that can be made at publishing level that will basically induce people to upgrade or use one device over another device.

We might know this is premature obsolescence, but a lot of the times you might see it when if you use Slack or tools like that, you might see how. You stopped being able to use certain browsers with Slack, for example, or even if you tried to use a browser, like say, Firefox might be the browser that I use by default.

I can't use Firefox on some tools because they assume the use of Chrome or certain tools like that. And there's a whole thing about, okay, well how far are you gonna actually be supporting a, and what happens as a result of you choosing to support certain devices over another one? All right. Okay. There's also, the other thing that we're gonna talk about with this was this idea of LightSwitchOps.

I really like it. It cuz it feels like it captures, and we've spoken about, there's an idea in tech where people talk about things like serverless, whereas there's an idea of things being switched off that when you stop using them. But the idea that we don't have service is a bit of a kind of, It's a fiction that is cool, but let's be real.

We're not really, we know their service still there. Whereas LightSwitchOps, the idea of switching things off, it makes it feel much, much nicer and it feels easier to say than scale to zero. So once I kind of get your take on LightSwitchOps as a kind of low tech and, but a friendly way to talk about some of this stuff here.

Andri Johnston: I think that's exactly what it is. It's a friendly way to explain to people how we can build our products and how we can make sure that they're energy efficient. So just to give a bit of background, I don't come from a tech background at all. I come from a publishing background and I taught myself all of these things and I got fascinated by it.

And it sometimes gets quite overwhelming when you are talking about, especially things like around cloud hosting and serverless, and it's becomes very techy. But the moment you can simplify it, Then it's almost like you can say, oh, okay, yeah, I can make, I can do this. Anyone can change the way that we're creating our products.

So yes, I think that's exactly what it is. It's simplifying tech for everyone to understand and digital sustainability, because it does still seem very farfetched in, in some ways. If you're not super techy.

Chris Adams: Yeah, I'm with you on this. Are you familiar with the baking forecast, by the way? Andri. Okay. The baking forecast is the way I talk about carbon intensity to people who are not really already really into this stuff. At the baking forecast, you, I think it still is a Twitter account that basically will tell you, when the electricity is gonna be particularly green in the UK, so if you're gonna bake a delicious cake or loaf of bread or anything like that, it'll be a particularly green cake, which means that you'll feel particularly good about yourself.

And as opposed to if it's a really fossil fuel, heavy, heavy moment of the grid, maybe on a wait a little while, probably don't bake today or bake tomorrow, and they just provide little forecast just like we have the shipping forecast, which is a well loved. Institution in the uk you now have the baking forecast to communicate the idea of carbon intensity of electricity.

And that seems to be pretty intuitive to a nation that is a fan of the British, the Great British Bake Off or anything like that. It seems like a really nice way in to talk about something which is gonna become more and more of a kind of staple or a regular constant. Uh, and as we move away from a kind of fossil fuel kind of based grid to something which is more in tune with the natural cycles and rhythms that we see of like sunlight and wind and things like that.

Andri Johnston: Yeah, it feels more natural in any case, isn't it like to live with nature in that way?

Chris Adams: Yeah The approach that I found as well, like when you talk about grids having a kind of cycles, then it's a little bit like really sped up seasonality for food. Uh, that's how I try to explain that to other people as well. Cause once you've. People have some notion like, yeah, okay. Things come into season, there's cycles you can use there.

And if you think about grids and electric, that's like the equivalent for people who work in technology is basically like chefs might have seasonality. We have grid intensity for what we do.

Andri Johnston: Yeah, that's a really good way to see it. Yeah.

Chris Adams: Okay. Thank you. Alright. Should we look at some of the events here to see what we've got coming up?

Andri Johnston: Yeah.

Chris Adams: All right. Okay, so there is an event that's listed below here. This is the Green Infrastructure Meetup by Green Coding Berlin, and it's about, there's a guy, yeah, Arne Tarara. He, he works at the imaginatively titled Green Coding

Berlin. They're basically a a bunch of people in Berlin who really into green coding basically, and they're presenting and running an event that's happening in Berlin on the 31st of May. So that's what's happening there. And there's a, there's some talks that they have, Arne Tarara and his little gang of people, they have a bunch of open source tools, which they make available for everyone to use for free and to basically try using some of this stuff.

And the other thing we have is UN World Environment Day. This is on June the fifth as well, actually.

Andri Johnston: That looks really interesting. I was looking at that. I was like, oh wow. Like I really want to attend that one.

Chris Adams: Yeah, and the people speaking here. So we've got Asim Hussain, who's a regular on this podcast. Anne Currie, she of Space Death Rays and data centers in space. There's Tamara Kneese she's actually one of the lead authors on a report about. Basically cryptocurrencies and the environmental impact of grid cryptocurrencies for the Linux Foundation, but she's also doing a bunch of work with the Green Software Foundation on upcoming green software report and Pindy Bhullar, this CTO for ESG at UBS and a PhD researcher.

This is an online event for anyone who's interested on the 5th of June. The other thing they don't have mentioned is the Linux Foundation Energy Summit that's taking place in Paris where it's a two day long conference with a bunch of events that I'm hoping will be shared afterwards because, well, I'll be going there in person to see, and I've seen a bunch of tools and talks coming upcoming, specifically about various kinds of open source tools that you can use to quantify and understand the environmental DIMPACT of the digital services that we use in on a daily basis.

Andri Johnston: Sounds really interesting. I have to say, I'm definitely gonna share the UN one, the online one as well, because I am a bit biased towards online events. So having, coming from South Africa, I always had this feeling that there's not enough opportunities to go to things like this. And one thing that I really love is how there's more and more online events like this for people who can't travel all the way.

Yeah. So I, I think that's really great to have more of these.

Chris Adams: I know what you mean. I'm really with a bit of luck. What I'd love to see as a follow on from some of this stuff, we've seen people understanding the environmental impact of entirely virtual events, and we see a significant amount of information about the impact of in-person events where we see that most of the time, the road to the impact is around 80%.

Basically of flying people to and from an event a lot of the time. But this idea of kinda like hub and spoke or kind of hybrid events, I haven't seen that many published reports or things to help us understand if there is a way to have some kind of interim here. Because there is something to be said for high bandwidth interactions with other people, but there is also a significant environmental impact associated with that.

And, uh, there have to be some alternatives to this. What we'll do, I'll share a link cuz anyone I know about is a piece by Vlad Coraoma. He's shared a paper about this from a couple of years ago, but beyond that though, ah, Andri something we could, it'd be really nice to see. So if few folks are doing that anytime soon or doing any kind of hybrid events, do please let us know and it'd be really lovely to hear about that.

Andri Johnston: Yeah, it's definitely something that we're interested in looking into more, so I'll keep you in the loop.

Chris Adams: Cool. All right, Andri, thank you very much for this. We're just coming to the end of this, which is time for the kind of closing question. I suppose you work for one of the most well known publishers in the UK, and I guess I should ask you about books in that case, where do you tend to get your books? Do you buy 'em secondhand in shops or do you read on Kindle, or is there some particular channel you into?

What would you recommend?

Andri Johnston: I think I am a bit of a book snob. Previously I worked in trade publishing in Penguin, so I'm a little bit of a book snob if I'm just reading something. Quickly, then I get it on my Kobo because I'm also very aware of the actual impact of a physical book on the environment. But when I do buy physical books, I usually go to indie bookstores and buy the hard cover version because I know how much goes into producing it as well.

I don't buy that many books, which seems shocking, but working in publishing, but it's just because I've become very mindful of what I do buy. So if it's a quick read, I read it on my Kobo, but if it's, it's like a book I really want in paper, I'll go to a very indie niche little bookstore and support that way.

That being said, secondhand, old secondhand book stores that are very dusty are definitely one of my favorite places in the world. So yeah, I'm definitely a little bit of a book snob, but that's to be expected.

Chris Adams: That's so good. The problem, I think that's fair. If you work with books all day long, then you get to have opinions about books. All right, Andri, thank you so much for coming on for this little session today and uh, yeah, I've really enjoyed this chat, so thank you for coming on for this week in Green Software and hopefully we'll have you on again sometime soon.

Andri Johnston: Yeah. Thank you so much for having me. It's really great and thank you for the great podcast. I really enjoyed and I always learn a lot, so thank you so much.

Chris Adams: Cool, thanks Andri Andri. See you around. Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation In any browser. Thanks again and see you in the next episode.

The Week in Green Software: AWS & Scope 3 Emissions Data

17 maj 2023 | 40 min

Host Chris Adams is joined by the GSF’s Asim Hussain on this episode of The Week in Green Software. They discuss some interesting news about Amazon, AWS and their scope 3 GHG protocol emission data. We also find out how Python has got its Mojo back and we have a very exciting tool from Catchpoint WebpageTest for measuring site’s carbon footprint. Finally, some great green software events that you can be part of!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Asim Hussain: LinkedIn / Twitter

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

AWS confirms Scope 3 GHG emissions data will be made freely available to customers in ‘early 2024’ | Computer Weekly [3:17]
Amazon’s SusScanner for CloudFormationTemplates | Charles Roberts on LinkedIn [13:22]
Mojo may be the biggest programming language advance in decades | Fast.Ai [16:28]
Measure & Improve Your Site's Footprint with Carbon Control from Catchpoint WebPageTest | Catchpoint [22:14]

Resources:

Picture of Coffee Analogy with GHG Scope Protocols | GHG Protocol [8:24]
Survey on Need for GHG Protocol Corporate Standards and Guidance Updates | GHG Protocol [10:25]
Adrian Cockcroft’s Monitorama Talk | Vimeo [24:41]
Carbon Impact | Dynatrace Hub [24:58]
Is data transfer the best proxy for website carbon emissions? | Fershad Irani [28:23]
Learn.greensoftware.foundation | GSF [31:55]

Events:

Ottawa GSF Meetup (May 24 at 9:00 am EDT): | GSF [31:00]
SDIA Hackathon on the Environmental Impact of Software (May 24 at 1:00 pm CET, Berlin): | SDIA [32:47]
London GSF Meetup - Anniversary Special (May 25, 6:00 pm BST) | GSF [36:04]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:
Asim Hussain: I think in an organizational perspective, scope three is turning out to be quite an amazing lever to drive change cuz by calculating the scope three, they're also applying pressure to their suppliers and saying, reduce your scope three, reduce your emissions, reduce effectively your scope one and two and your three or our go to another supplier.

I'm your host, Chris Adams. Welcome to another episode of this Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and in this episode we have interesting news about Amazon and AWS. A programming language with a lot of Mojo and some exciting events coming across from the world of green software, as well as some interesting news with WebpageTest

before we dive in though, let me introduce my esteemed guest and colleague for this episode of this week in green software, which we sometimes call TWiGS, today we have Asim Hussain. Hi Asim.

Asim Hussain: Hiya. So my name is Asim Hussain. I am the executive director and chairperson of the Green Software Foundation. I'm also the director of Green Software at Intel.

Chris Adams: Cool. All right. And most recently, learning about throat singing to go along with your mushroom collection.

Asim Hussain: I'm also becoming quite musical, so I've actually bought, I've actually got two flutes since we've met Chris. I've bought two flutes, a guitar, and, wait, what else? Oh, there's another musical instrument I can't quite remember. Anyway, yeah,

Chris Adams: That's three musical instruments

Asim Hussain: my voice. My voice. That was it. I've been taking singing lessons. I knew there was a third I.

Chris Adams: that's pretty cool. My wife is a trained musician and she's been teaching me the spoons, so that's basically all I have forgot to go with.

Asim Hussain: My wife is also a classically trained musician. She speaks it very hard to be somebody who's learning music inside a house because yeah, anyway, there's

Chris Adams: Wow. I did not know.

Asim Hussain: yeah, yeah. Both quite musical. Yeah. We've got, our partners are quite musical. There you go.

Chris Adams: Okay. And if you are new to this podcast, my name is Chris Adams. I too am one of the directors of the Green Web Foundation, a small nonprofit working towards an entirely fossil free internet by 2030. So before we dive into the rest of this show, it's worth sharing. All the links and all the discussion we do, we will share it in show notes with this.

And the general format is to basically look at some stories in the news relating to green software and share a few reckons on them. And you're very welcome to come chime in with your comments, uh, afterwards as well. Okay, Asim, should we start?

Asim Hussain: Yeah, let's go for it.

Chris Adams: Okay, so story number one is AWS Amazon Web service confirms Scope three GHG emissions data will be made freely available to customers in early 2024.

This is a story from Computer Weekly, and as Amazon is one of, is basically the largest provider. This feels like a fairly big deal Asim, especially when you bear in mind that this may bring AWS's Customer Carbon Footprint tool up to kind of parity with some of the other providers like Google and Microsoft.

Asim Hussain: Yeah, I remember when AWS first came out with their tool. Probably there's a lesson learned here for cloud providers when you come out with carbon measurement tools, make sure it's got scope three in it, because almost all the news are like, great, but where's scope three? Because it's so essential and it's such a large figure cloud providers.

I've heard an argument for smaller cloud providers where they don't own, physically own the data centers with which they are providing services, that an argument can be made that it's so much more challenging to obtain the Scope three data there. But when you own your own data centers, the expectation is that you're going to be able to have to provide that, that data.

Cause it is such a significant number.

Chris Adams: Hi, I'm so sorry, Asim. I realize we've just dived straight into a jargon without even just telling it all what scope one, two, or three might actually be. So I'll just quickly, for those who are new to the subject, or folks who have never heard of the GHG, the Greenhouse Gas Protocol, essentially this is a way.

The kind of defacto standard for measuring the carbon footprint of any organization or any activity. And, uh, you typically split it into three kind of buckets of emissions. And because we are nerds or developers and drink coffee, we can use hot beverages as the mechanism from standing. The difference between scope one, two, and three, you can think of scope one, which is from combusting fossil fuel.

That's a bit like. Turning on gas to heat up water so you can have a nice cup of coffee. Alright, scope two. It's like turning on an electric kettle so someone is setting fire. Something to heat up some water somewhere to generate electricity so that you can heat up a kettle. So it's all the emissions associated with electricity that you might purchase, for example.

Now Scope three is a little bit like walking into a Starbucks or a third wave coffee shop, and then. Buying a cup of coffee. So you are not involved in actually farming beans or burning anything, but there is definitely a supply chain associated so that you can have coffee. So these are the three kind of scopes, and typically Scope one and two are quite the common ones that organizations tend to report on. But for these, for lots and lots of organizations, scope three can make up 80% plus of the environmental impact. And this is why we've been talking about it as being quite a big deal because if you do not have 80% of your reported numbers, they may look somewhat different to the other providers.

Asim Hussain: And I, that's a really great, is the word analogy or metaphor? A

Chris Adams: Analogy I think is you.

Asim Hussain: yeah. And I'm gonna use that as well. That's wonderful. And I also realized, as I was saying that I think I may have given the impression in my previous statement that Amazon just made the casual choice not to include Scope three, and I don't think that's, that was the point.

Scope three calculation is hard. It's extremely hard. It's extremely hard to get right. There's a lot of error bars. It's really obvious a lot of the time what to choose, what values to choose, what to input, and so that's why perhaps Amazon's taken this long to come up with the Scope three data because they've been spending a lot of time making sure they wanted to provide Scope three data that they were comfortable with providing.

I just wanna state that it's quite complicated cause as you can imagine, going into Starbucks and trying to figure out. How much a coffee with all the components you can imagine that come up to delivering that coffee to Starbucks, the shipping, the person in the farm, making it, growing it, all of that stuff is someone needs to calculate all of that stuff and it's very, very hard.

Chris Adams: This is indeed true, and even when organizations are reporting Scope three, there's sadly 15 separate subcategories beneath it, which makes it even more complicated a lot of the time. And there is another thing which makes it even harder, is that when you're trying to record Scope three, it's one of the parts of this protocol where.

There isn't the same concern about double counting in other places, cuz initially when the GHG protocol was actually put forward, the idea was that you would use your own scope three figures as something that you wouldn't necessarily compare to another provider, but you could compare to your own performance over time as a way to track your glide path to something which might be avoiding climate apocalypse at an organizational level. However, this is one of the things that has actually made Scope three quite difficult for people to understand because this double counting issue is prevalent in this kind of scope and not so much in some of the other parts.

So yeah, that's one we can dive into. And what I'll do is I'll share a link to that picture. Cause I've got a nice diagram for the coffee one I just shared with you. Yeah.

Asim Hussain: that's the go. But also I think you raised an interesting point with the whole idea of double counting, cuz I wanna dive into that just a little bit more, think it's interesting, so for instance, it's quite easy with like scope one. Like you, if I have an oil drum in my front yard and I'm burning something in it, No one's double counting that.

That's not in your yard. We know that's not in your yard. We know it's in my yard. It's very easy when I'm buying electricity because of the way that you have to trust the system works. The database is actually like allocating that kilo hour to me it only goes to one other person. Whereas with supply chain is quite interesting cuz your scope one and two, if you are a manufacturer, your Scope one and two will actually be another.

Organization's Scope three.

Chris Adams: Indeed.

Asim Hussain: So I believe. The theoretical idea is that if every company, an individual in the world calculated the scope one and twos, that would all sum up to this wonderful total, which is equal to the total of carbon emissions in the world. And you're right, scope three is just, just nice to have.

But I think in, in an organizational perspective, scope three is turning out to be quite an amazing lever to drive change Cuz by calculating the scope three, they're also applying pressure to their suppliers and saying, reduce your scope. Three, reduce your emissions, reduce effectively your scope one and two and your three or our go to another supplier and it's providing that pressure, which I think is really, I dunno if that was intentional or an accident, but it Yeah, I can see the, I can

Chris Adams: This was one of, this was one of the, I think this was one of the principles initially, and there's some very explicit principles designed for the kind of GHG protocol. The other thing that's worth sharing that we might refer to a little bit later before we dive too deeply, is that there is actually a whole process of redesigning how people measure this stuff.

And what we can do is we can share a link to a kind of summary of some of the responses to the World Resources Institute saying, Hey, we're thinking of updating how we measure carbon emissions. Cause there are some problems with the current approach. What do you think? So there's some stuff there that we'll share to, but we'll share a link to, but we won't be able to dive too deeply into it because Asim, I think the two of us are getting outta our depth pretty quickly.

Asim Hussain: that depth. Yeah. Um, but anyway, one of the, Part to this title that my mind zoomed in on. It says AWS confirmed scope three GHG emission data will be made freely available to early 2024. Why was the word freely added? It seems weird. A census could work without it. So the fact that freely is there implies that, can you pay for it now?

Chris Adams: The reason is that if you are a publicly traded company and you need to do your own reporting, there's actually been a while where you. If you spend enough money with that with Amazon under an nda, you can get these numbers. Alright? Now the thing is that basically means that only people who are spending above a certain figure ever get to have an idea about this.

And it also means that if these figures aren't in the public domain, then it becomes very difficult to have a data informed discussion about where we're moving with any of this stuff. And this is. Important when you have the largest provider, which has an organizational carbon footprint of 60, 70 million tons each year, which is, this is like a small European country level basically.

So this is one of the things that has been problematic. So hopefully this may be a reference to saying it becomes available for everyone so we can finally have some understanding around this. But as opposed to just only the people who wanna do an NDA on that, cuz there is a kind of prisoner's dilemma aspect when you basically only get your own numbers so you can get your own reporting, but you make it difficult for anyone to have any kind of effective policy interventions on this at a kind of more wider and societal level.

Asim Hussain: Yeah. Yeah, very brilliant. So basically my personal website is hosted on Amazon s3. So I mean, I was always a major custom of Amazon's prior to this, so now I'm able, but even, I didn't even, I didn't classify, uh, to getting the data, but even now I'll be able to get, anybody can get it. Okay.

Chris Adams: Yeah, next year, eventually. Yeah, so

Asim Hussain: year, sorry,

Chris Adams: we early 2024 basically just in time for the law to make sure it's absolutely essential anyway, so you do feel like, alright, organizations have to report this in 2024, so this May, maybe there's some link between that and all these new laws landing, which have reporting deadlines in May, 2024.

Possibly. We'll see where that goes.

Asim Hussain: it's weird. Weird cause and effect here, isn't it? I wonder. Yeah,

Chris Adams: Oh, we'll see. It's good. It's good to actually see the progress and this does make it easier for any responsible professional to start understanding some of the impacts associated with their use of digital services. Alright, next one. Next story. Up.

Asim Hussain: Yep, yep. Yep.

Chris Adams: This is another one continuing our kind of Amazon tip, which is Amazon's SusScanner for cloud formation templates.

This was shared by Charles Roberts, senior Security consultant at Amazon, and from what I can tell, this is now an open source tool, which can basically scan your cloud formation code to give you an idea of where you might make some improvements. And it's, this is largely referring to some of the kind of pillars, architected pillars, and sustainability from AWS.

And assume, I think you folks might have done something like this in your old Microsoft days about having some recommendations and pointers for this, right?

Asim Hussain: yeah. Maybe I'll just take a step back and talk a little bit about the well architected kind of framework itself, which is, Amazon's got one and Microsoft has one that's also called the Well Architected Framework.

Chris Adams: Oh,

Asim Hussain: I believe Amazon's came first. I used to think all that the well architecture framework was, cuz if you go to the website, you'll see well architected framework and they have I think four or five pillars secure to your reliability, this and that and the other.

And it's advice for how. If you want to build a reliable cloud application, this is how you should build a reliable cloud application. Now, for the longest time, I just assumed it was just advice on our website, but it turns out it's actually a scoring system. And so what Amazon Cloud consultants do and so does Azure ones, is when you work with the customer is it's the scorecard.

You go through and you ask 'em questions about their infra, about their system. Based on their answers, it gives you a literal score and you get a number afterwards. And that number can indicate how much work you need to do to rectify. And so when they added the sustainability pillar, what they also did was they added a bunch of questions.

And if you don't, if you answered no or however it was structured to those questions, you got a yes or no, you got a certain score. And so from my understand from this is cloud formation, is there is what you call it now, infrastructure as code? Is that what Cloud I believe, yeah. It's infrastructure as code.

So it's textual description in configuration files for how your application is defined, and it effectively runs it against that, those scores, and it sees what's your number, and it basically gives you a sustainability score, which is really cool. That's to say, yeah, automatic sustainability score. Yeah.

Chris Adams: this is true. I think there are a number of tools that start, do start doing this, but having something riches in part of the ecosystem. Okay. It's better than not having this. So yeah, if you cut live it a Python, or if you ever have to manage anything related to some infrastructure as code and you're using cloud formation instead of Terraform or some other tool, then yeah, worth a look all.

Asim Hussain: I think from when I, I had an original chat with the, with Charles. I don't believe Charles was the actual person who authored this, but he's the one who shared it. Currently. It's automatically scanning stuff and comparing it to AWS world Architected pillar rules, but you can create your own as well, so you can create custom rules for it, perhaps to your makeup or perhaps other people can come up with their own rules for what makes a good sustainable application, add it to that framework as well. So yeah, it can be

Chris Adams: Wow, I didn't know that. Cool. All right. Next story coming up is Mojo, possibly the biggest programming language advance in decades. This is a link to fast ai. Yeah.

Asim Hussain: It's quite a headline. Quite a headline. Mojo may be the biggest programming language advanced in decades. Come on, let's talk about it.

Chris Adams: Okay, I'll, I, I actually read through this and I, I'm actually pretty excited about this and I shared this before because I do a bit of coding in Python as my kind of main working language, and Python is often maligned for being a slow language, even if it is a relatively pleasant language to be using.

Asim Hussain: Mm-hmm.

Chris Adams: But the general kind of gist of this story is that it combines the ease of use of Python, and it's designed as a kind of super set of Python so that you would actually have all the syntax and all the kind of ease of use and the familiarity of using Python. But you have a really smart compiler, so normally.

Like with kind of, you have different flavors of Python. So for example, there's like maybe Sea Python or Pi Pie or stuff like that. These take this and come up with some kind of much, much faster representation of that code, and there's limits to what you might have there. Now, Chris, what's the guy's second name?

Who's behind this?

Asim Hussain: Latner.

Chris Adams: Thank you, Chris Latner. Yeah, so LVM is known for creating a kind of where you might have things that kind of create assembly or stuff like that. It creates what's referred to as an intermediate representation. So this is like a piece of language, which is easy to turn into, really fast code for hardware.

One of the kind of innovations was that a project which he worked on was called MLIR, so a machine learning focused intermediate representation. This is particularly interesting because it means that you can have that same ease of use of taking something which is relatively easy to write and make something which is really easy to run fast on GPUs or TPUs.

I forget what TPU stands for, but it's a transformer processor unit perhaps.

Asim Hussain: Tensor processing unit

Chris Adams: Thank you. Yeah, tensor. So basically we fast AI essentially. And this essentially means that you get the speed of these really low level languages with a lot of this. And typically you could do bits of that. Like you might write something in Rust, which is like what the COR kids do, and then use some kind of bridge language.

But this idea is that there's like a subset of just extra functions you might type. So rather than typing deaf my function, blah, you just do. Fn my function, which is somewhat Rusty, and then your compiler knows that this part can be super duper fast and they're promising hundreds or thousands of hundreds or thousand fold speed improvement on this one, which is mind blowing in my view.

Asim Hussain: My initial first thoughts were, I've been a user of Python for a long time now, on and off, but it is been aligned for a long time for being slow, and I think that's unfair. Because you would never normally write the things that need high performance directly in Python. Like that's why Python is still used for machine

Chris Adams: Mm.

Asim Hussain: because what you end up doing is you import NumPy or, or even TensorFlow or something like that, and then you actually, your code is executed using those libraries.

Those libraries are written in C and c plus.

Chris Adams: Correct.

Asim Hussain: So a, I think it was off times maligned. And so I was like thinking to myself, is this really necessary? Because. You can still get that performance improvements. But I was just reading it again and I realized that it was a really important point here was that it's actually really hard to debug that.

Chris Adams: Exactly. Yeah.

Asim Hussain: And that's the really, I, as soon as I like read that, I was like, oh, of course. And now you can actually just debug Python code. It's not just gonna make a call to some external thing, mysterious black box that does something fast and you dunno what it's gonna do. You can step through, step by step.

And I think that's very interesting. That's right. Cause that was a learning experience that's really gonna help people. Build performance systems. Yeah.

Chris Adams: I am hoping this means that I don't need to learn how to try and learn Rust or something, to be honest, cuz there's lots of things which seem nice. But this basically is an alternative to having to learn yet another language because if it's gonna take me 10 years before learning Rust, then I'm not gonna be very far from retirement before I'm any good at coding in these new languages.

So yeah, that.

Asim Hussain: I never thought about it from that perspective, but you're right that a lot of people are gonna hate me for saying this. This could be a Rust killer. This could be like, there's a lot of people who know Python already and if this is gonna give you effectively Rust level speeds, which it would, it will do.

Cuz it's a system level, it comp compiles onto system level. That's very interesting.

Chris Adams: See now you see where I share that? See, kind of be exciting, right? Yeah. Alright. But for folks who are fans of Rust, it is the, I think, the most popular, most lovely programming language. This doesn't mean that you won't have a job. There's lots and lots of work for Rust, and I think Rust has actually been adopted in favor of C plus for a number of Microsoft projects now, cuz there's a lot to be said for.

Yeah, absolutely. So there's plenty of life in a Rusty world. Absolutely. And this isn't really out yet as well. We're not even sure if we're gonna have this as open source yet, but it looks like it might be in all the previous projects like LLVM and so on were open source so we can hold our fingers and hopefully Asim, we can stay relevant for maybe a few more years before we are consigned to the aging programmer trash pile.

Okay.

Asim Hussain: know. There's, there's still people paid to code in a cobalt, so I think we're, I think we're all right.

Chris Adams: You're right. Okay. That's great. Let's hope that stays there and we don't get replaced by some form of machine learning in the future. Alright, should we move on from that one? Because that feels like it's about to just, I can see it yawning open ahead of us.

Asim Hussain: you mentioned AI. Let's close it down quickly and move

Chris Adams: Yes. Okay. So the next story is from an organization called Catchpoint.

And in particular there is a project called Webpage Test, which is an open source, a Web performance tool, specifically designed that's used by governments and lots of Web performance specialists to basically analyze pages somewhat like how we just described the well architected thing for Amazon Cloudworks.

So this is interesting in my view because webpage test is one of the most well known uses like Web performance tools. They've started incorporating the carbon measurements inside this, and it's a really nice quote from. I guess what I'll refer to as the godfather of cloud or what, what would you call Adrian Cockcroft?

Sustainability is becoming a higher priority for organizations globally. Not only is our ethical responsibility, but there are new regulations that will require companies to monitor and manage their environmental reporting, said Adrian Cockcroft, tech advisor and sustainability advocate. Carbon control is making it easy to take the first step by measuring the carbon footprint of a website, a Web application, whilst also providing actionable recommendations on actions that could result in improvements.

So that was like major praise. I was pretty excited to see that actually, because this is, this was like one of the former VPs of cloud and he's very much someone who cast a significant shadow across industry. So yeah, I saw that showing up in LinkedIn. I was like, Oh sweet result.

Asim Hussain: interesting. I remember he also mentioned, cuz Adrian was a year ago, he spoke, he gave a very, a really excellent talk, a very inspirational talk as in, I don't mean inspirational as in go and do it, but as in opening people's minds to the idea of monitoring as an action for this, it really helps inform something of me.

So I think I remember part of the post was, The creator of Carbon Control saying that the idea for it came from meeting at that Monitorama conference like a year ago and how that idea kickstarted this whole thing, which, yeah, which makes me, I always say the most powerful thing you can have as an idea.

So it's just goes to show, just having one one talk can lead to, yeah.

Chris Adams: The thing I might share with you is that success has many mothers. And this is also using a library called CO2.js, which includes some of these numbers, some of the kind of conversion practice. So if you have an idea for how much, what kind of resources a webpage might be using, this is what it converts into carbon figures.

So one thing I've shared a link to the Monitorama talk, cuz it's a really good talk and this kind of process from monitoring tool to carbon tracking tool. Dynatrace is another organization that does something like this. So there is a, it's a real kind of trend in my view, and it's really encouraging, I think.

Asim Hussain: Has Dynatrace added some sort of energy carbon tracking? I haven't seen that

Chris Adams: Yes, last week Max uh, Schulze from the SDIA, he referenced this and I didn't know about it before there. I'll share a link into their show notes. But yeah, they have their own carbon impact figures as well now. So.

Asim Hussain: There's so much stuff happening in this space. It's so amazing. Just, yeah.

Chris Adams: So who knows, maybe Datadog will do it and let's hope they don't charge 65 million per year for the privilege. Sorry, that's a nerdy joke about Datadog's recent investor reporting, realizing, and they mentioned that one of their providers was paying 65 million us.

Asim Hussain: million for.

Chris Adams: it looked like it was Coinbase.

People weren't paying attention, and suddenly the numbers went up. And when Coinbase realized there was a 65 million hole in the reporting and they had to explain what happened, and they said, yeah, someone realized that they weren't paying attention to it. So if you ever feel bad about cloud spend, yes, they,

Asim Hussain: So they literally, they, because of Coinbase's growth, they just hadn't factored in how much the observability was adding to the whole thing, and it just added to

Chris Adams: I think that was the idea. So this was like the, I think the canonical example of sometimes cloud can lead to people not paying too much attention to expenditure. But see, I've never done 65 million of spend before and I'm not sure I will, but that's now my kind of benchmark to make me feel better about myself if anything I have is not very efficient.

Asim Hussain: that, that Datadog account manager is driving around in a new Ferrari, I reckon.

Chris Adams: You hope so, or maybe not, because that's gone now. So they probably had a comfortable disc, had a discussion and said, Hey folks, are you sure you wanna be spending 65 million a year with us on tracking your logs and metrics? So there was actually something ongoing there. So there was some proactive outreach to say, folks, I'm not sure if you mean to do this.

Are you sure you wanna be doing this? Apparently,

Asim Hussain: We're just bringing it back to Catchpoint trying to, because you've been a Major S CO2.js, which is the Green Web Foundation's project. I did put the Green Software Foundation website through and fingers crossed and it, and we scored pretty well. I think there's still room for improvement. I thought believe we did score green, but why don't you tell people what does it tell you?

Chris Adams: So the main thing that webpage test does is it will look at your page and analyze it. And like we mentioned with Well Architect Checker, it'll basically. Tell you some things that you could improve based on what it's seen about your page. So if your pages are very large page and sending a bunch of JavaScript over the wire, which would result in a kind of poor experience for someone waiting for it to be loaded, it will say, maybe you shouldn't be sending such massive payloads over the wire, because it's not gonna be very much fun.

And it also is gonna have a impact on your end user's battery as well. Now what it actually uses is inside the library that we maintain called CO2.js. There are a number of different models, and one of the models that is in use is called the Sustainable Web design model, which is based on some peer reviewed literature. Basically saying for this amount of usage, which is right now is basically the data center of the wire. It basically makes some assumptions about how much energy use happens on the device, in the servers and in the networks. And this gives you some idea of what the actual missions might actually be.

So that's how it works, and you can link back to it and we can share a link specifically to see some of the assumptions for this. There's also a really nice post by one of the people who was actually advising on this and helping get this implemented, Fershad Irani, he's written about, okay, this is the things you need to make.

These are the assumptions we've had to make here, and these are the alternatives we might use in future for this. So this is designed to be a first step that you could then start improving this, cuz as we know Asim, all models are wrong, but some models can be useful.

Asim Hussain: This is the post that Fershad wrote about basically asking the question is network bandwidth the only metric we should be using and

Chris Adams: Yes,

Asim Hussain: between? Yeah. I thought it was a very interesting, yeah, very interesting thought. Cause I think, not how CO2 JS works, but I presume it just says JavaScript is equal to this,

Chris Adams: pretty much.

Asim Hussain: Yeah. So he was thinking about what if you could split and you would give the same carbon waiting to a kilobyte of a JPEG image as you would give to a kilobyte of a video image, but maybe there's a difference.

Chris Adams: Yes. So this was raised by Mike Gifford, who was. So he was a real kind of like sustainability and accessibility advocate based in Canada. He actually opened this issue in the CO2.js repository to talk about some of this stuff. So it's really worth looking and we'll share a link to that. But what we've done is we've shared a link called Is Data Transfer their best proxy for website carbon emissions?

Where he explains this and talks about where this is good and why this is bad. Because very much, a lot of the time, the tools you use to understand an environmental impact of something, it'll often be impacted or influenced by what data you have available because not everything is instrumented to provide the kind of levels of num uh, levels of detail that you would like to have at the moment, but we're getting there.

Asim Hussain: Yeah, like I think my response to him was like, like when it comes to models, as we just said before, like they have inputs and they have outputs, and you tweak the inputs to optimize the output. And so if the only input you have is bandwidth, that's the only thing you would tweak. If you separate it out bandwidth for image and video, and you saw that video is so much higher than image that would change the decisions that you would make, which I think is an interesting thing here cuz I, there's a balancing point between making something useful and ubiquitous and so everybody finds value out of it versus getting enough fine grain information. So the behavioral choices afterwards are the right or better people claim better choices.

Chris Adams: Yes. This is actually a nice segue to some of the events we might be discussing. Actually assume So should we look at some of these on then?

Asim Hussain: Let's go for it.

Chris Adams: the first one is Ottawa, the Green Software Foundation Meetup on the 24th of May.

Asim Hussain: Wonderful. Yes. I'm so excited about this one.

Chris Adams: Abhishek's talking. This is actually exciting cuz Abhishek has been quite involved in this as an

Asim Hussain: unfortunately he had to, for personal reason, had to pull out one at the last minute. It's, Henry Richardson is now giving a talk there, which is just as exciting cuz he's the, I dunno what his title is, at Watttime is Researcher.

Chris Adams: Lord of beards.

Asim Hussain: Lord of beards,

Chris Adams: does have an impressive beard here.

Asim Hussain: lord of beards, emperor of electricity, carbon emissions.

That's what he is by anyway. He'll be giving a talk and I believe also, yeah, Tajinder Singh from GitHub will be giving a talk on sustainable DevOps. Oh, we never heard that one. Have we yet? Su No. Yes, we have SusDevOps. We've heard that one before. No, we haven't. We've heard DevSusOps.

Chris Adams: no. Is this like the people's front of Judea? It does feel like it.

Asim Hussain: one of them will win out in the end.

Chris Adams: Oh, speaking of things with ops at the end, so Google and ThoughtWorks have a thing called green ops. That's their particular term that they use for this and on in Berlin, Google had an event talking about green tech and assume, do you remember your principles.green stuff?

Yeah. I'll have to share a picture. They saw them sighting your principles green and their own internal stuff. It was pretty cool. I've got a really blurry photo. I'm not sure if I'm allowed to share it, but I could definitely tell you. It was definitely in the kind of presentation that was shared with various people at it.

Asim Hussain: Ah, brilliant. That actually I just checked now. We haven't actually deployed it. We're on the verge of deploying a change to principles.green. So actually all those pages are now going to get forwarded

Chris Adams: Oh.

Asim Hussain: to the new learn.greensoftware.foundation, the new, I forgot it suddenly, can't quite remember what we called it now, but our green software training.

But if the evolution of this page be.

Chris Adams: All right. Here's the elegant segue I was gonna be mentioning. So the SDIA has a hackathon on the 24th of May. Also in Berlin, where this is actually some work with the German Environment Agency, the equivalent, the Umweltbundesamt. I think they are actually hosting this event. With the SDIA, A Green Software Foundation member to do a hackathon about trying to understand the environmental impact of software.

And they have actually a whole set of tools and a hack day specifically for improving the environmental impact of open source tools. So there's a competition there. The reason I was mentioning this is because this was actually where the initial work with Mozilla to start exposing some of these numbers.

It initially took place. I met some folks at Mozilla and they explained how they've been able to reverse engineer the energy usage figures for visiting a page. So you can actually get some of the numbers, like the environmental footprint for JavaScript versus videos or pictures and things like that. So maybe here's to actually extending it some more so we can make that more available to more people.

So yeah, that's the 24th and available for everyone.

Asim Hussain: I've gotta say, that's the wonderful things about hackathons. It's not even really what happens on, it's not even necessarily who wins on the day, but it's the work that happens. It's the connections you've make and happens after us. I did not know that that may work that you've done with Firefox.

I don't think we've really talked in any great detail about at all. I think we definitely would love to be less deep dive on that, or even deep dive on CO2.js one day.

Chris Adams: We should ask some of the folks at Mozilla because there's a bunch of other things they're doing and there's a bunch of really cool stuff they've been doing with telemetry that I think would actually fit into this. Cuz I think there's a chance to create a kind of public data set, specifically be used on actual observed data rather than the model data that you see, right?

Asim Hussain: because they have data

Chris Adams: Yeah, exactly.

Asim Hussain: not in any nefarious way, but they must be collecting data.

Chris Adams: Anyone who runs a browser, they gonna Google collect this, probably Edge, have this as well. They've got a rough idea of, cuz every single organization would have to optimize this and try to reduce the kind of costs imposed on their users.

Asim Hussain: if it's anything like how it was at Microsoft, like you'd actually ask the users of the canary version of Edge and you actually, you would have a pop-up saying, do you want to give your data? And I go, 0.01% people say, yes, but that's enough to get like a significant amount data. So I imagine the firefighters doing the same out of, I just wanted to just make sure everybody's clear.

There's a hackathon in Berlin 24th of May. There's a prize of 1,700 Euros. That's Space Shack Berlin, which sounds amazing. Chris will be there, Max will be there.

Chris Adams: Yes, I am definitely gonna be there. Some, some of the Green Web Foundation, some of the SDIA folks will be there and I suspect some other people will also be around as well. So there's a nice group of people now said doing stuff in our little town, and I really like it. Actually. I'm very much enjoying it

Asim Hussain: our little town of Berlin. Yes.

Chris Adams: compared to London, where I moved from.

It does feel, yeah,

Asim Hussain: I always thought Berlin was huge. I've never really,

Chris Adams: the population is definitely lower, but it feels a bit more spacious.

Asim Hussain: Berlin. Anyway.

Chris Adams: All right,

Asim Hussain: Great stuff. So you

Chris Adams: back to London.

Asim Hussain: Oh, back to London. There we go. Always comes back to London. Yeah, so there's a London Green Software Foundation Meetup happening the day after on the May 25th at 6:00 PM UK time.

That is actually, I believe it's coming with the UBS offices in London, which have very cool offices actually. And it's also a special anniversary special. It's actually the two year anniversary of the birth of the Green Software Foundation, yet I will actually be there myself. There'll be networking, drinks and pizzas. Will there be a cake? We should definitely have a cake actually now that realized, yeah, there will be a cake. I dunno. Maybe there might be a cake.

Chris Adams: If you've got 10 days, alright, you who knows, you might even have cake pops. Easy for people to eat.

Asim Hussain: Great idea.

Chris Adams: Cake on a stick is the future my friend.

Asim Hussain: Cake on a stick. There we go.

Chris Adams: Yes. All right. Okay, so I think that's it for our news Roundup and list of upcoming events. This is the part of the show. We have a short show closing question to ask to our guests.

This is what we see. We've seen a number of meetups happening recently. If you could travel anywhere without too much impact in the environment, where'd you like to see, uh, meetup and why? I'll put that one to you. Asim.

Asim Hussain: Oh, that's a great question cuz you know, I would actually really like to see meetups happening in places where it is typically been very hard for, not for us, but for people to discuss. Other, other people with an interest in sustainability and find each other. So I often find in, especially in Asia, it's in the larger cities, it's usually better.

We've had meetups in Japan and some of the larger Indian cities, but I'd love to see, oh, I could travel. Oh, I probably wouldn't travel myself because there's a little bit too much impact, but I'd love for other people to go and travel locally to their local Asian GSF meetup.

Chris Adams: Okay, cool. I'm gonna be really boring here. I'm gonna say something like Vienna. Because never been to Vienna. Sounds like a cool place. And what I've been told is that Vienna is one of the cities where if you had like electric scooters and things like that, you had dedicated places to put them rather than putting them in the middle of the pavement.

So there are car parking spaces dedicated for that. And there's even an app in Vienna. So if someone has parked it in the wrong place, you can take a photo. Send it to it. And then the people who are allowed to operate the scheme, they have an SLA to maintain, so they have to get it moved within four or five hours, otherwise they get fined.

This feels like a really interesting use of public space and I feel like, yeah, I quite like using some of these scooters, but I don't like how if you are in a wheelchair they can get in their way and it doesn't feel like it's the best and most equitable use of space. And this felt like a really nice a way to address some of those issues.

Asim Hussain: That sounds lovely. Yeah.

Chris Adams: Yeah, so Vienna, that's what I would say. Alright. Okay. That's all for this episode of The Week in Green Software. All the resources in this episode are in the show description below and you can visit podcast.greensoftware.foundation to listen to more episodes of our podcast. Finally, I think huge thank you.

I've really enjoyed chatting with you again, Asim, nice to see you. So yeah, take care of yourself, mate. Lovely seeing you. Too-da-loo. Ta everyone.

Asim Hussain: Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: IaaS, PaaS, SaaS!

10 maj 2023 | 35 min

Host Chris Adams is joined by Max Schulze from the SDIA (The Sustainable Digital Alliance) and they discuss three stories from the worlds of IaaS, PaaS and Saas! While these three acronyms are more than likely ever present in most digital people’s lives, we might not know about the environmental impact that they have. Chris and Max cover stories from the CNCF, Google, CIODive and OpenJS as well as upcoming events in the Green Software community.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Max Schulze: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

CNCF: Sustainable Software Projects Breed Eco-Sustainable Computing Systems: / Forbes [6:52]
Google: We have no moat, and neither does Open AI the story of open source vs openAI / Simon Willison [14:37]
Technology pulls enterprise green ambitions within reach: / CIODive.com [20:57]
OpenJS Foundation Receives Major Government Investment from Sovereign Tech Fund for Web Security and Stability / OpenJS [30:41]

Resources:

Microsoft scales cloud-native workloads with carbon awareness / SDX Central [7:02]
Kepler / Red Hat Emerging Technologies / [7:33]
ECO-Qube / SDIA [9:25]
Environmental Data Agent (EDA) / SDIA [9:55]
Carbon Aware Scheduling on Nomad / Green Web Foundation [12:37]
Vicuna: An Open-Source Chatbot Impressing GPT-4 with 90%* ChatGPT Quality / LMSYS Org [20:40]
Unilever completes cloud-only migration to Microsoft Azure / CIO Dive [24:21]

Events:

Rise of AI 23 (9 - 10 May, BERLIN ᐧ Hybrid) / Rise of AI [28:15]
SDIA Hackathon on the Environmental Impact of Software (24 May, BERLIN) / SDIA [28:45]
LF Energy Summit 2023 (June 1 – 2, PARIS & Virtual) / Linux Foundation [31:47]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

Transcript below:

Max Schulze: It's important that we recognize that this is happening on more than just Kubernetes. That's why I think it's a movement that's happening.

Chris Adams: Absolutely. Yeah. Diverse ecosystems are healthy ecosystems.

Hello and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Hello, and welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software and software development. I'm your host, Chris Adams, and in this episode we're gonna be covering three stories about green software and the acronyms, PaaS, SaaS, and IaaS.

We'll also be discussing some exciting upcoming events in the world of Green software and talking about moats and open ai. Before we dive in though, please let me introduce myself, my guest for this episode for of this working green software to today we have Max Schulze. Hi Max.

Max Schulze: Hey Chris.

Chris Adams: Okay, Max, uh, although we've known each other since about 2019 when we first went to some Green Cloud procurement events back in Brussels many years ago, you may not be familiar.

Some other folks might not know too much about you. So if you introduce yourself, then we'll kind of jump into some of the stories.

Max Schulze: Thank you Chris, and thank you for having me as your guest today. It's always fun to to chat with you and debate with you. I'm Max. I'm the director of and founder of the SDIA Sustainable Digital Infrastructure Alliance. That's why we never say that. We always say SDIA cuz it's unpronounceable.

Very excited to be here. I also talk about digital infrastructure all day. I write about it. I think about it. I talk about it, and recently I've been very involved in driving data centers to be more sustainable and more transparent through like the European energy efficiency directive, but also the code of conduct for data centers.

Chris Adams: This is something you'd be doing in not just German, but also in English. Is that correct?

Max Schulze: Yeah, I'm one of the very few Germans that can also speak very fluent English, so it's not a problem for me.

Chris Adams: Okay. All right, Max. Okay, so we've spoken about the format here and for folks who are new to this, basically what we're gonna do is look at a few stories that caught that kind of came up across our feeds and came up on our radar This week. We'll do share a little bit of reckons and share some extra context that you might not be too much to aware of, and then we'll look at some coming events.

So Max, the first story we have here before we actually go too far, we've used these phrases, PaaS, SaaS, and IaaS. Let's just briefly just define these, cuz these might come in handy and we might refer to this a little bit later. So Max, Hey, let's start actually, because ERs is at the very bottom. Let's start there then.

Work your way upwards actually. All right.

Max Schulze: Yeah. So yes, for me is the commodity. It's compute. Storage and network capacity essentially is what AWS started with EC2 VPC and s3. Those were like the three primitives that you can build almost any software from and abstracted in a way that you don't see the server, you don't see the data center.

So it's really like digital resources as commodities in the highest form of abstraction.

Chris Adams: And IaaS for most people stands for infrastructure as a service, right? So you pay for things on a kind of monthly or hourly basis, almost like metered electricity. Yep.

Max Schulze: Yep.

Chris Adams: Okay, cool. All right then, so that's IaaS or IaaS, PaaS, the next one up. That's platform as a service, which as I understand it, you might not want to be working with all these low level things yourself.

So there are companies that provide higher order so. Functions on top of this. So for example, you might choose to purchase cloud storage or some computing power from a hyperscaler or maybe a French company, say Scaleway or something like that. But you might also pay for someone to just give you a place to put a Docker container or a piece of program, run a program, and they take care of keeping it fast and secure so that you can just focus on writing some code.

So that's platform as a service. Do you have a go at describing what SaaS or software as a service might be then Max.

Max Schulze: To me it's the most ironic one because past, like a database as a service is also a software as a service, but it's tools that developers use to build more software. So it's funny. And SaaS is when you then glue a lot of different kind of open source components together into a product. For example, this product that we're using to record the podcast is also a software as a service product.

The most well known right now is probably Office 365 or like Microsoft Teams or now SaaS based product. So usually subscription based, usually Web based, but can also be desktop applications. But it's more about the fact that you don't have to maintain, run, or install any software. It's just available to you at any given time as a service.

Chris Adams: So we've defined these somewhat confusing terms and Max is almost like a taxonomy that people actually need to be armed with for them to understand any of this stuff. And it might be worth just briefly touching on that before we dive into one of the stories. Cause I know this is some work that you've done when you speak to policy makers about them struggling a bit to try to work out where one layer ends, another one begins, and where you need to purchase these and how you actually purchase these.

Max Schulze: I think in tech or IT, we often lose simplicity to understand things or to make things understandable, because we want to be so specific, we want to be 100% accurate. And a lot of people get lost in that process when they hear a hybrid cloud setup or on-prem versus public versus, yeah, something else. Or a VPS, a VPS machine, or a dedicated machine, and you hear these things, it's like, what is this?

And I think it's, what I've really come to realize is that you have all these words, and think about it, public cloud. Five years didn't exist five years ago, didn't exist as a term, and now we are using it every day and every developer knows what it means. But outside of our bubble, very few people actually understand all these different technologies.

And ultimately it's software, hardware, some form of infrastructure, networks, data centers, and energy, and that's it. This is the whole thing, but we have so many words to describe different configurations of those that I think it's really important that we have some kind of taxonomy. You also know that we wrote a paper on this sometime ago to propose something, but I think we need very simplified explanation models so that we don't lose policy makers with our complexity.

Chris Adams: All right, thanks for that, Max. So now that we've got a little bit of kind of background for what some of these terms mean, should we look at the first of these stories here? So this is one in Forbes from the CNCF, which is the Cloud Native Computing Foundation. They're talking about software projects and e sustainable computing systems.

We've touched on some of this, the idea of kind of Kubernetes or low Carbonetes and the idea that there are existing products and projects that could be adapted to basically tread someone more lightly. On the planet as it were. And this is the first time I've actually seen people in the CNCF really talking publicly about this and really talking a lot about their special interest groups, where they are essentially trying to get gaggles of nerds to talk about some of this stuff.

And there's a few particular projects of interest that have been catching people's eyes of late. One of these is Kepler from Red Hat, which stands for Kubernetes Efficient Power Level Export-er, right? And that's been one tool which has been used to essentially expose some of the energy numbers when you run little clusters of compute so that you can start to quantify and understand the environmental impact from some of these tools.

Max, I know that you folks did some work with similar project or related projects around this, and it sounds like it will be quite simple, but I think you can share a few war stories about some of this stuff and where some of the difficulties might actually be when you're working with this.

Max Schulze: So I think first of all, it's great that the CNCF is actually talking about this. They've had the sustainability working group, I think for almost a year now, and I'm glad that they're now going a bit more public with it. I think there's a lot of value in Kubernetes and what it can do in terms of distributing workloads across different physical locations and machines.

I think the worst story that you're alluding to is that. I think we've seen a lot of technology solutions that enables load shifting, which is the one of the holy graves. Like I can move it where the energy is greener. I can move it where there's under utilization and things like this. The problem is that the infrastructure, so it doesn't matter if it's a virtual machine or the data center doesn't actually expose the.

These metrics that you would need to make those decisions. Right. So we see a lot of workarounds where you're trying to guess which data center you are in, and then you're getting the emission factor of the whole grid around it, which doesn't really exist, and there's so much data missing that. I think the problem to solve is not necessarily the at the highest level, but rather in the levels below, making these APIs available, making these metrics available to enable load shifting, not to solve load shifting.

Chris Adams: Okay, thanks Max. This is also something we might touch on a little bit later. When. We talk about some of the events that we have upcoming here. I think you also did a bit of work with a project called ECO-Qube when there was some discussion about actually exposing some of the numbers from data centers, because as I understand it, one of the reasons people basically say we can't share these numbers is we don't have access to these numbers, but I understand that you ended up spending a bit time looking into this specifically to see where the real pain points and what the possible solutions might be for people at the data center level to expose some of the numbers for the rest of us further down the food chain as it were, so we can actually optimize for carbon.

Max Schulze: Yeah, we did. I actually, we even released, it's gonna go open source in I think a few months. It's running in three pilot data centers. It's basically a monitoring system like Prometheus and Grafana. But what is surprising that even though we in it monitor a lot, right? The underlying infrastructure doesn't have a very sophisticated monitoring system.

You can really imagine it like software from 20 years ago. It's not that you can just query are the diesel generators on or not. It's the cooling system at 40% or 50% load. And so we wrote basic in experimental piece of software that collects all that data. Makes it available as one unified API so that you talk to the infrastructure, like you talk to some database system and you just say, what's the energy mix right now?

Is the diesel generator running? Is the, I dunno, is there wind park running nearby? And you get all these signals and you can respond to those signals. And I fundamentally believe that when the signals are available, when the transparency is there, then software will adjust to the available parameters.

Chris Adams: Okay, so we will share a link to this. What was the name of this particular project that we should be looking for, that we should be searching for if we do look

Max Schulze: the environmental data agent, EDA EDA, if we renamed it, I think to EDS, but EDS is already like European data something, uh, or European Space Agency, something related to that. So it's not the environmental data system. But the environmental data agent, because it is really a physical box that sits in a data center that has two network cables connected.

One from the IT side and one from the infrastructure side. And it also acts as a firewall because you don't want the data center, physical infrastructure connected to the internet either that, um, the security risk, of course.

Chris Adams: When you talk about connecting that in that way, you are, you are referring to the idea of maybe you don't wanna have the cooling systems exposed so someone can switch them on or off, resulting in. We already have heating problems as is right now, so making

Max Schulze: You don't want it to be hacked. Let's put it like this. And the safest way for cybersecurity, for everybody listening who loves cybersecurity, the best way to protect a computer is to not connect it to the internet.

Chris Adams: Okay. All right. Yeah. Okay, so that gives us some pointers there. And there's something to be talking about some of the CNTF projects that people might be interested in. So we spoke to Kepler, and we'll be touching on a proposal by Adrian Cockcroft to a formula. VP of Sustainability, Amazon, who's been talking about some of the actual proposed metrics that you could actually use, and he'll be explaining where some of the problems are when you do try to actually work out these numbers from existing providers.

So that's one thing we could talk about, but maybe in the future episode, we might touch on it a little bit later on today. Yeah, go

Max Schulze: what about you? Didn't you guys do something with Nomad?

Chris Adams: Yeah, so my organization, the Green, Web, Foundation, we did a bit of work with Nomad because we saw a bunch of people using Kubernetes as a way to orchestrate all kinds of computers to run software.

And we have a kind of tradition at the Green Web Foundation, where we look for the people who are not the dominant providers, because they're often doing some really interesting work. So we did some work with Firefox as, for example, took a carbon calculation library into that particular library as well into that.

But last year we did a bit of work with the folks at Hashicorp, because we use Nomad ourselves to run our kind of internal infrastructure. And there's now a separate Fork of Nomad, which does have this kind of carbon oil computing inside it right now. So this is on one thing that we did a bit messing around with, and also what the other reason is that I'm quite a big fan of a service called fly.io, which is one of these PaaS.

It's a kind of way to manage software without having to actually be. Maintaining all your kind of Amazon accounts yourself, for example. And I think the idea of like green fly sounded kind of cute. So we were just doing some work. We're there to use that basically. So we did some work on this, I think last year with an organization called Ripe that the people who allocate IP addresses to the world basically.

So we did some work there. But yeah, that's the exposure that we have so far since in the last six months. What we've seen is Microsoft basically donates their own. Schedule and a bunch of their own open source works specifically for demonstrating how they do carbon air computing. So there's a bunch of things around here, but that is probably the thing to look at if you are using Kubernetes.

But if you are looking at Nomad, then I'm very happy to talk about this and I'll share a link to a blog post where we explained how we would go about doing this and where we're moving to next. Cause it's got a lot easier.

Max Schulze: I think it's though, it's important that we recognize that this is happening on more than just Kubernetes. That's why I think it's a movement that's happening.

Chris Adams: Absolutely. Yeah. Diverse ecosystems are healthy ecosystems, as we say, where I work. Okay. Next story. Let's talk about moats. So this is a story which is partly blogged by developer Django co-founding. Nerd, I suppose, and now a AI specialist or AI researcher, Simon Willison. He linked to this leaked paper from Google, which says, we have no moat and neither does open ai.

This is basically a story which in many ways, kinds of goes against this kind of narrative that we've seen over the last year or two, where lots and lots of the. Advances in kind of machine learning and AI have been associated with ex ever larger amounts of compute. And while we did do a bit of research and point to some papers a few weeks ago where there is less of a link between absolute brute computing power and the actual kind of accuracy of models, there's now a really interesting paper talking by a, a kind of nameless person inside Google.

Basically saying, since a bunch of open models have been released, The kind of gap between million pound uh models and literally a hundred dollars models worth of training is really narrowing quite now. And the quote that I think is probably good for setting the scene is this one. So this was talking about comparing models like say, GPT3 or Google's bard, for example, with some of the more recent ones based around, I forget this, Lama, LMA, and Lama and alpaca.

I'll refer to these later on. While our models still hold a slight edge in terms of quality, the gap is closing astonishingly quickly. Open source models are faster, more customizable, more private, and pound for pound more capable. They are doing things with a hundred dollars and 13 billion parameters that we struggle with 10 billion than 540 billion parameters, and they're doing this in weeks, not in months.

Max Schulze: I think the biggest risk here is the rebound effect because. Yes, it's now more efficient, so to say, to put AI models. It's basically everywhere and everybody can have their own and like I can have one for my notion space. I can have one for my Basecamp, I can have one for my Wiki. I can have one for my tickets.

That just means that same as with LEDs, we will get exponentially more AI models embedded into everything that's probably in aggregate, we'll use still much more infrastructure, much more energy, much more GPU power than ever before. I think doesn't matter if you have 10 big ones or a billion small ones, the effect is probably almost the same, if not even worse with the small one.

So just looking at it from the environmental perspective, what do you think about it?

Chris Adams: I think that automatically reducing the amount of compute needed for this, in my view, is a good idea. And if you just have a larger number of smaller players who are playing rather than just an oligopoly of three or four. I think that means that when it comes to actually regulating and being able to have civil society involved, I think you get more people able to talk about this.

And you don't just get to have innovations coming from a very small group of people. So I think that you end up with, um, which would be more representative of society and therefore probably coming up with a. Uses and ideas, which are probably not quite so full of some of the kind of gaps and some of the kind of blind spots we've seen previously.

So I think this, by having a larger number of people doing this, I think is a good idea. I think that makes it more likely that you have one or two people who are pushing for, say, The idea of these models being as a part of you training, you basically just say, I'm only gonna run this on green energy, for example, or I'm going to disclose the information about how this has been run.

And because you have more people who are actually able to do this, I think you're gonna end up with. Greater transparency in people being much more explicit about both the providence of the data and the actual cost in terms of environmental impact that's come into it in the first place. So I think this creates scope to compete on transparency and compete on the fact the data can be trusted and has actually mean created in a more equitable fashion.

So I think that's good. And there's another really key idea that I saw here was basically people saying lot of these new open source projects, they're achieving wins, not by having loads of data, by having much, much better curated data. And I think this is actually a much more promising direction to be going in than what we've been seeing so far where you just throw ever larger amounts of compute because if you are, say Google or Microsoft, you've made so much money that you don't know what else to do, then spend 70 billion on buying your own company's stock, right? That makes you think that it'd be nice if there's just more people who are able to use this rather than a small vanishing number of people put involved here.

I think that's actually useful from a governance and an environmental point of view, basically.

Max Schulze: So also from a society perspective, you basically say, well, we had this property theft problem, right? Then also larger datasets, open source dataset, address that. I agree with you. I do think just for completeness, I want to say that we do still need rules for bias and lots of other problems. I think we, from a societal perspective, so the third component of sustainability, we do need still rules.

I think we need rules, some basic ground rules and principles for what's okay to do with AI and also what problems are okay to solve with AI. There was a great article by The Economist on a war game that they play with nuclear weapons and basically the end of that every. Minister and government person that plays that game says, I want to automate the decision of when to deploy retaliation strikes, and like everybody can tell.

And also my neck hair went up and I was like, no, you don't wanna give that decision making power to a machine. That's the whole point. And I think we need those kind of principles and rules of what you can do with AI and whatnot.

Chris Adams: Okay. This is actually one thing that gets touched on a little bit later, is this whole explosion of much, much clearer data sets where you actually have a good idea of what the provenance is and how that's been created. Cuz this has been one of the problems that we've touched on in previous episodes.

When you are playing around with LLM models, if you look on Twitter, you will see lots and lots of people telling you about how your job will be destroyed if you do not learn to use OpenAI, but there's also. A flip side to this about, well, some of the labor rights associated with this, and whose labor is actually going into these models and being obscured that we touched on.

Okay, so this is the last story that we have. This is technology pools, enterprise green ambitions. This is from CIOdive.com, and this story is largely talking about some of the providers of IaaS infrastructure as a service. And what's some of the mechanisms you might actually have available to you to reduce the emissions associated in your supply chain?

So this is aimed at telling CIOs. Okay. There is this thing called Scope three, which is like a kind of way of thinking about the emissions in your organization's supply chain. And let's talk. It talks a little bit about which companies are doing better than others, and we've seen some new updates from actually Amazon of all people.

So for the last year, there was the kind of initial rush to get an early version of say, Amazon's cloud computing. Sustainability dashboards out the door. And then no updates have happened for 12 months. And then we've seen some updates again. So it looks like folks are starting to pick up on this, but that is not the whole story.

And I think Max, you might have some records to share on this one here actually.

Max Schulze: Ayayay Chris, I'm gonna be very diplomatic here. I admire that Microsoft is very transparent about their Scope three and their data reporting. You have to. You with a grain of salt, because if you go one layer deeper and you look at the available data from, let's say, how much emissions is in a data center building, how much is in a server, even HP and Dell's reports are.

Let's say vague. So let's say if you look at it as like a chain of data that you need to collect, I think that it's starting and it's good that the big ones are really saying we want to be more transparent, and then they realize, oh, we don't have the data from our suppliers. I think that's a good thing.

I think that it's always tricky because even the GHG protocol is made mostly by corporations. They make their own reporting rules. So I think there is a lot of work still to be done to include all environmental effects. I do agree with the article also talks about third party vendors and consultants.

I think there's a lot of tools being built to help with this. Our formulas have been integrated in Dynatrace monitoring tool. So in Dynatrace, you can actually now at least calculate the energy of your AWS systems and you get very much larger numbers than AWS is reporting to you. So yeah, again, comes down to rules and standards.

And I think in tech, this is the first time somebody's doing like a holistic inventory of all environmental impacts of digital technology, and we are lacking so much data at every corner. It's just not there. So a lot of these things are really brutal estimates right now.

Chris Adams: This is true. We spoke about this change. We were surprised by some changes coming through, and I'm afraid this is a bit spotty and nerdy. This energy efficiency directive that was basically rooted recently that is seen as a transparency win for a few people, basically that to provide a lot more transparency at a data center level that you haven't seen previously.

Max, I think you, you had some exposure to this or you saw how some of the, what they say in gentlemen, how the sausage gets made, right? Maybe you could share a little bit on this one here.

Max Schulze: Yeah. Yes, we were very involved in this process. It was very politically loaded. It's essentially about, so even the big cloud companies often rent data center capacity at co-location facilities, which are, uh, You can think of it like WeWork for your servers, right? You get everything power cooling included in your rent, but you have to bring your own servers.

And some of these companies are very intransparent about your energy consumption, the emissions of your energy consumption, the embodied emissions. And this law essentially forces them to both make it public and then also attribute it and give it to their customers. And that's quite the game changer, especially because the law, the first reporting interval, you pointed this out as well, is already in May, 2024, which is very short notice and will drive a lot of data centers to now really quickly scramble together a reporting system.

Right now, the most used spreadsheets and uh, that, that's why also what you mentioned earlier, the, our EDA project of course, can help with this. That's why it's open source so that every data center can deploy the monitoring tool and then release that data as quickly as possible. And I think that will also increase the accuracy of what the hyperscalers are reporting, but also what you as a developer have access to.

I hope it's, of course the, all your Digital Oceans and AWSs have to still pass through that information, but I, the law really sets the stage to get the data in place to begin with.

Chris Adams: I see, and there was one thing that surprised me when I read through this, was this real focus on heat reuse, basically. Now that's a key thing. It's starting to be warm again, at least in Western Europe compared to other parts of the world. But, uh, we do know essentially space heating or heating things up here is one of the big sources of emissions for this.

And, uh, I think Max, you mentioned some of the idea, like one of the reasons some of the transparency stuff might have come through is because there was a almost disproportionate amount of interest in making sure that heat gets reused by various organizations. Is this one thing that you saw?

Max Schulze: Yeah, absolutely. I think heat recovery has been something the SDIA has been talking about for five years as like, why on earth are we putting a hundred megawatts into a system that produces 100 megawatts of heat and then not use it because you will know your computer generation. Every electronic process generates that heat and it's just silly to not use it.

And because you mentioned it, this idea for example, that in summer you can't use the heat is complete bollocks because storing thermal energy, right? Putting it in an underground tank to store the heat, you can store it for a whole year until winter and hear this, to store energy one kilowatt in the lithium ion battery.

Cost about $180 per kilowatt hour storing one kilowatt of thermal energy using any form of tank, $1 per kilowatt hour. So

Chris Adams: Okay.

Max Schulze: not storing that heat recovered or not, we have to use it. It is like literally we put all these green electrons, right? All this green energy into the data center, and then instead of reusing those green electrons, we just throw them away as heat.

And it's the rarest commodity we have on this planet right now is green electricity, and we should use it as much as we can in as many times as we can. And yeah, I'm really glad that the directive suggests that, or basically forces the data centers to at least consider heat recovery and show proof that they have considered it.

Chris Adams: Now, and this is something that I understand has been so outside of this world, I'm a bit of a kind of heat nerd because I know that Denmark is actually one of the, one of the countries which has a long history of storing heat for long periods of time just like this. And I think I might have spoken outta turn about just when it's hot, you don't, you might not want to put the heat somewhere, but you're absolutely right.

There's various parts of the world where they store things in significant bodies of thermal mass, like pools underwater and so on, so that you can pull the heat out and as people end up do using things like heat pumps and so on, you're able to move it around to other places to make better use of that actually.

Okay. All right. We podcast about green software and not just heat. So I'm just gonna look at some of the events as we run up now. So we have, I think, two or three events on the horizon that are coming up here. The first of these is this rise of AI 23. This is happening in Berlin on the ninth and 10th of May.

This is a hybrid one, and basically the Responsible AI crew starting to look at carbon neutrality and trying to understand. The actual leverage points on a project to work out where you can make meaningful savings. Here, this is a free thing to join to see some of the talks and there's another event taking place later on in May that I think, Max, you might have something to say about here cuz it has your organization's name on it actually.

Max Schulze: Yeah, so we are hosting a hackathon with the German environmental agency to measure the environmental impact of software. This is really about measuring and we build a test system where people can basically upload their code to a GitHub account, and we have a special CICD runner that people can use and that runner is so to say, energy aware and also carbon aware, and it measures everything. So what we are gonna do, hopefully, is take some open source projects like a Django or like some noJS library, and gonna basically see what can we change in the code and then run all the tests again. So without. Reducing the functional scope without removing any tests. How can we make it more energy efficient?

How can we use less energy or use less server capacity, and also measure in between different versions of software. That was an idea that somebody brought to us that I thought was really exciting to say, you know, when you release a new version of a piece of software, can you basically do a diff of the energy use versus functionality growth and basically say each new version should not use more energy than the version before unless it adds like significant amount of functionality free to attend.

Chris Adams: Oh, cool. All right, so a bit like how cars keep getting bigger and bigger and some of the things you might care about, say crash safety, but some other things like cup holders you may be less excited about or there's maybe things you are. The idea is to provide that level of transparency to make some of that more visible to people.

Yeah.

Max Schulze: Yeah, what you can measure, you can change, right? And so I think if we make people aware that their software keeps growing in terms of environmental impact, then I think there's more incentive to reduce it.

Chris Adams: I see, and this is part of the project, I think of the German Comp SoftAWERE, the project with the Federal Environment Agency, the Umweltbundesamt who are doing this

Max Schulze: Very good pronunciation, Chris. Very

Chris Adams: to get some practice in. All right. Actually there's some, I'm I, there's another story we didn't have time for, but I'm just gonna share cuz it caught my eyes.

Basically the Sovereign Tech fund is essentially a fund in Germany and they've basically made a donation of 875,000 euros or $900,000 to the openJS Foundation. So the people, the foundation that maintains like noJS and various tools like that. This is the first time I've seen a government make a direct donation in this kind of, Fashion for quite a substantial amount for something like this.

Actually, this is like the focus on security rather than sustainable software. But it caught my eye cuz I was not aware of just how much, I guess the German government is involved at the software level now.

Max Schulze: Funny. Yeah, I think it's interesting because. That also means that soon the open source community will finally get political because once you take the money from the government, you also have to have a position.

Chris Adams: This is very true actually, and I wonder this might be something we touch on in a future episode actually. Okay, so we've got that happening on the 24th of May. That's the hackathon there that people will be going along to, or that you can go to. Final event that we have listed here is the LF Energy Summit taking place in June.

So this is happening at window. In LA Force, Paris, France, and this actually has a few people from the Green Software Foundation presenting both the Green software principles and uh, the Carbon Aware sdk, which is an open source software development toolkit for people who are building projects, uh, using the kind of Microsoft stack, using C Sharp and so on, and trying to basically reduce the impact there.

This is something that I think we also touched on, Max, you spoke about the idea of Linux appearing in various parts of the kind of. Energy sector now, and this to my knowledge, is where you see a lot of people in the kind of open source world now looking at lots and lots of proprietary scarda infrastructure tools and say, maybe we can use some open tools to make it easier to maintain and manage this stuff.

Because once we go down to the data center level, it can become a bit harder to get the numbers or basically work with open source software the way that you might see it at an infrastructure level really closer to the metal.

Max Schulze: Generally a bit worried because I think. The energy system works, right. Our computers right now are running. The light is on. There's no no problem to fix. What we need to do is to scale renewable energy as quickly as possible, and the technology for that. We have 14 megawatt wind turbines now is really there, and a lot of the problems in renewable energy is about.

It's about permitting. It's about, I don't wanna win park in front of my house. I don't want solar field next to my house. And I don't think that the solution is digitalization right now. I think as a tech sector, as digital people, we need to look at our own stuff and not go bring our stuff to other industries right now so much we need to make sure that we are not going to be the bigger problem at the end as the biggest energy consumer left or something like this.

I'm always trying to put us back on track. Like tech can solve tech. We don't need to go solve energy system right now because energy system is already on track to decarbonize itself and let's just let them run with that. If they need help, they will call and we do us.

Chris Adams: Okay. Okay, Max. I think that's the wise words indeed. Max, thank you so much for coming on to this show and this conversation. I really enjoyed chatting with you and I hope we can do this again sometime soon.

Max Schulze: Me too. It was, it's always a pleasure to talk with you, Chris, and I always feel like we have to rush so much, but we could do a lot. We could do four hour podcast episodes.

Chris Adams: Maybe in the future we'll do that. Okay. Alright. That's all for this episode of The Week in Green Software. All the resources for this episode are in the show description below, and you can visit podcast.greensoftware.foundation To listen to more episodes of Environment Variables, the kind of larger name for this podcast.

I'd like to say thank you very much again, Max for coming on, and folks, see you next week on the next episode. Bye for now. See you around Max.

Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get to your podcasts.

And please do leave a rating and review if you like what we're doing. It helps other people discover the show, and of course, we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation That's greensoftware.foundation in any browser.

Thanks again and see you in the next episode.

The Week in Green Software: Data Centers in Space

3 maj 2023 | 39 min

Host Chris Adams is joined by Anne Currie on this episode of The Week in Green Software. They discuss the potential for data centers in space and how the use of potential death rays might be the way forward in powering these! Not only this, but sweeping changes in Reporting Law, and making Kubernetes clusters into Low Carbonetes clusters are covered too. Anne has a special report on her upcoming book and Chris finds his own variation of Boaty McBoatface!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Anne Currie: LinkedIn / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Data centers move into space to mitigate power consumption and pollution / El Pais [2:41]
Building Green Software Book Preview [14:43]
Microsoft Scales Cloud-Native Workloads with Carbon Awareness / SDX Central [21:32]
Solicitation for Public Comments on the Business Practices of Cloud Computing Providers / FTC [26:32]
The SDIA welcomes the deal of the European Council and Parliament on the Energy Efficiency Directive / SDIA [30:16]

Resources:

Community Clouds and Energy Islands with Dawn Nafus and Laura Watts about Data Centers in the Orkneys [6:26]
White Paper that Anne’s Book is based on [14:43]
Designing for Sustainability by Tim Frick [20:43]
https://github.com/Azure/carbon-aware-keda-operator [22:00]
The Green Web Foundation’s Guide on Carbon Aware Scheduling w/ Kubernetes and Nomad
Ross Fairbanks [22:18]
Kube Green [28:17]
Boaty McBoatFace
Culture Ship Randomizer

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript below:
Anne Currie: Data centers in Greenland are an obvious thing because there's tons of free energy, green energy from ice melt runoff.

Chris Adams: Yeah.

Anne Currie: But one of the issues there is that nobody lives in Greenland to man the data center, but even few people live in space demand the data centers. So I would, I would hope that you would solve the Greenland issue first.

Use all that enormous amount of energy before you,

Chris Adams: Okay, so our site, so before we reach for the sky, let's sort out things down here on Earth. Yeah?

Hello, and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Hello, and welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams. This week we'll be talking about data centers in space, sweeping changes in reporting climate laws affecting the digital sector.

Making Kubernetes clusters into low Carbonetes clusters and a set of interesting looking coming events. But before we dive in, let me introduce my esteemed guest for this episode of this week in Green Software. This week we have Anne Currie. Hi Anne.

Anne Currie: Hello.

Chris Adams: Anne, would you just introduce yourself?

Anne Currie: My name is Anne Currie and I am currently one of the community co-chairs for the Green Software Foundation. I'm a software engineer. I was a software engineer for many years, many years and 30 years I've been in the industry. And so I date from the time in the nineties when we used to build software using the same kind of techniques that we might, that we are thinking about using to date for green software because machines were very weak then, and we had to handle that. So that's my perspective on this.

Chris Adams: Cool. Thank you, Anne. So if you are new to this podcast, my name is Chris Adams. I am the executive director of the Green Web Foundation, a nonprofit focused around us reaching an entirely fossil for internet by 2030. And I'm also the policy chair, the chair of the policy working group in the Green Software Foundation.

Each week we do a run through of stories that caught eyes, or that might be fun to discuss. And everything we do discuss, we share all the links that we can find for you to dive it down your own little Wikipedia kind of holes after this session. Alright, so Anne, should we look at the first story that came up here?

It's data centers move into space to mitigate power consumption and pollution. So this is the story from El Pais, a Spanish newspaper, and uh, they published this story I think of stem. It's about this program called Ascend, which is Advanced Space Cloud for European net zero emissions and data sovereignty program.

Anne Currie: That's not, that's not a little bit contrived at all.

Chris Adams: Yeah, exactly. And the European Union has selected fails Acun space, a joint venture between Thales group in France or Thales possibly, I'm not quite sure if I'm pronouncing that correct. And the Italian defense conglomerate Leonardo. And the plan is to see if you can create space data centers here. And I think the plan here is to try to address some of the energy issues related to data centers on the ground. I found like the initial press release for this, but Anne as a science fiction writer, I figured you might have some records here before we dive into this a little bit more, actually.

Anne Currie: Yes I do and I read the piece and it's, it is an interesting piece from my perspective. I am also a science fiction writer and I have written a series called the Panopticon Series and three of them are set in space and address the technology in space you mentioned is all about Arthur C. Clark. And interestingly cuz Arthur C. Clark was a physicist when he did his. For example, his lunar based novels, he put realistic technology in it, and he often had patents. So he had the first patent on an electromagnetic cannon in space, which uses electromagnetic fields to fire stuff around. Or it was a, an idea of delivery mechanism for getting stuff from the surface of the moon into orbit.

Chris Adams: Whoa, whoa. Did you say that Arthur C. Clark patented the rail gun? Is that,

Anne Currie: did? Yes,

Chris Adams: oh my God, my mind is blown. Gone. Please do. Go ahead. That's, I'm never gonna think of Quake another way that's changed how I think of Quake. I.

Anne Currie: But thinking, I don't think he was thinking about it in the form of a rail, but it might be quite specific patterns. Often quite specific. It might be specifically for electromagnetic canons for the delivery of stuff from the surface of the moon to to lunar orbit. But anyway, he did have the patents on that, which now expired cuz it was quite a long time ago.

But anyway. So that all that stuff does all work. And in fact, there's loads of interesting things you can do with rail guns in space as a way of, as a transport mechanism or a power transport mechanism as well. But anyway, there's, that's a side there. In terms of data centers in space, obviously you've gotten a lot of power that can potentially be generated in space room solar because you've got nothing in the way between you and the sun, and the panels can be in a hundred percent light.

So, but it is a very. Interesting idea. And along those lines, China and India have been coopering for a while. I dunno if they still are, but they were cooperating a few years ago on a space based solar power system. So space based solar power is the idea that you have a giant laser in space and you capture the energy from solar panels and you beam them down, beam it down

Chris Adams: down to a panel on the ground, right?

Anne Currie: Yeah. Which has some kind of giant death ray connotations around it. So it's not, it depends how you do it depends whether you use light lasers or microwaves to get it onto the ground. And then there are lots of, but it is perfectly doable and I think that's a very plausible idea for getting power, using the same kind of idea of using solar and space to get power down to

Chris Adams: Okay.

Anne Currie: re-usable space.

But yet it's, so the idea of data centers and spaces that you build them out there in orbit somewhere, probably quite a long way out. Cause they don't necessarily need to be a near earth orbit. And that's quite busy. And yet you could just be powering it directly. The difficulty with that is always that it's gonna be very hard to maintain that data center, but it did remind me slightly of a story that came up a few years ago, and it's definitely true, which is that Azure have been experimenting with undersea data centers in effectively the size of

Chris Adams: Yeah. Around the Orkney's underwater. Yeah.

Anne Currie: Yeah, and those have similar issues in that you put them in and then you can't maintain them. That's it there. So the idea of having a self-contained, smallish data center that's no one can subsequently touch is not a new one. So it's not utterly, utterly impossible. And of course, Starlink has got the cost of getting stuff into orbit down quite low.

Yeah. So it's not impossible. It's

Chris Adams: It doesn't seem impossible. I, so I think I struggle with some of the numbers on this. Because in this press release, we see something saying, okay, we want to install data centers in orbit, powered by solar plants generating several hundred megawatts of power. Now, several hundred megawatts is a very large data center, so like hyperscalers are between 20 to 50 megawatts of power.

So you're looking at something like that and then, I also, let's just look at say, okay, the International Space Station, they've got solar arrays. Right now they have maybe 120 kilowatts of power coming through, but they're old, and that's two and a half thousand square meters here. So more or less, if you are looking at something which is, I don't know, what is that roughly, that's maybe that's for a hundred kilowatts of power.

You're looking at maybe what, 2000 square meters per a hundred kilowatts. That means. For a single megawatt of power, you're looking at 20,000 square meters, and if you're looking for hundreds of megawatts, that's gonna be 20,000 square meters multiplied by hundreds. That's a lot of solar to have in the sky.

This is the thing I was struggling to get my head around. Things might have got more efficient in the last, say, 20 to 30 years, but surely that is gonna be a heavy thing to get into the sky in under any circumstances, will it not.

Anne Currie: Presumably that is a massively heavy things to get into the sky, but launch has really come down in price a lot. And of course it doesn't have to be particularly co-located with the data center cuz you can use those space-based solar power death rays to the subject of my last science fiction novel called Death Ray.

So you can, you don't have to be right by the thing, you could have those arrays literally millions of miles away in space and beam it back. You do get dilation on the beams if they're too far, but you can keep relaying them.

Chris Adams: Okay, so we could have our data centers in some or, and then the solar panels further out. So they're far away from there. Okay, so that's one thing. Then you mentioned that there, there's different kinds of orbits, right? So as I understand, There's kind of low earth orbit like LEO, which is Starlink, and that's maybe 2000 kilometers above the ground.

And then would that mean you're hidden from the sun so that it's dark for your satellite sometimes?

Anne Currie: Yes, I think it does. I think you have to be reasonable distance out

Chris Adams: Yeah. So geo stationary I think is like maybe bit further out where it looks like you're not moving because you're that much further out. Right?

Anne Currie: And you've asked me a question, I dunno the answer here. I do not know how far you have to go out to be constantly in the sun, but to be honest, it's just less busy further out. So if you can be further out, there are loads of reasons why you might prefer to do that. And yeah, it's just a matter of then beaming the power back.

Chris Adams: Now I'm with you on this, and then this feels like latency's gonna come up at some point, right? Because I'm curious about cause. In the LEO, like low earth orbit 2000 kilometers. We already use CDNs for like to have things closed. So if it's 2000 kilometers, that's one thing, but if it's something, I think geostationary is something like it's either 20,000 kilometers into the sky or 40,000 kilometers into the sky.

So that's gonna be, I don't know if speed of light is what, 180,000 kilometers per second. That's gonna be a significant chunk of latency no matter what you do. And that's even if, if it's just you going straight up and down, if you're going around the world, that's gonna be even harder, surely.

Anne Currie: Yeah, lower orbit latency isn't too much for an issue. It depends, but it, the further you go out, the more there it is. If you had your data center on the moon, latency is about a second each way? No, it's half a second each way, but it's, it is a total latency. You say about a second, which is obviously it wouldn't make for a very good podcast or a Zoom call, but it depends on your use.

And it depends. Just depends whether the latency is an issue or not, because sometimes bandwidth is more of an issue with than latency, so.

Chris Adams: Oh, okay. Yeah.

Anne Currie: Yeah. It all depends what you do, what you're doing with it, and where it's going and how much. I mean, I, I would guess that the whole point was a lot of the things they're talking about, like in that article they were talking about, data that they gathered in space, being analyzed in space, using the, a big array of CPUs, and then boomed back in a more compressed form back to earth.

In that case, latency is not an issue in any way, but if you wanted to move all data centers into space, then latency would be a giant issue. As you say, CDNs and stuff on the edge.

Chris Adams: Yeah. And the final thing, we'll stop on the space part cause there's other things we're gonna talk about. But the final thing that really kind of, cuz I scratched on my head about this, cuz last week we spoke all about using different kinds of ways to keep computers cool. Right? Now when you're in space, one of the arguments seems to be that because it's so cold anyway, you don't need to worry about cooling.

I don't think that's how I understand physics. As I understand it there are three ways to cool things down. There's radiation, convection, and conduction, and I'm not familiar with that many cool breezes in space, so I can't rely on conduction. Maybe convection, not very much. So that just leaves radiation.

And those pictures of the space shuttle with its doors open, it's open to radiate out heat. Because it's got so much heat still. So I feel like if you've got this issue where data centers generate lots of heat and there's no way to get rid of them, this feels like a problem that I don't see how it's gonna be solved by putting things into space that people haven't really taken on board yet.

And Anne I'm struggling with this, maybe you've got some pointers or maybe it does sound just bonkers.

Anne Currie: Not just bonkers, it's, it is, you're completely reliant on radiation and they have quite good things where they have little radiating shapes and stuff that can radiate off heat more quickly. But it's not easy. So it's not easy. It feels to me like you'd be able to do better on the moon because at least you're in contact with something that can conduct heat away.

Depends on how conductive moon dust is and or moon rock is, and I dunno that.

Chris Adams: Heat up the moon until it glows red. Okay.

Anne Currie: But yeah, you're right. It's not a no-brainer that we could just go in. It's not like that under sea ones. People go, oh, that's great cuz you don't have to worry about cooling if you've got a data center under the sea.

And that's true because it could just conduct into the sea and that's fine. But space is not the sea, you can't do it. So yeah, it's not trivial.

Chris Adams: So we have latency, death rays and uh, heating some of the challenges that may face us if we try to put data centers into the sky. But this is one potential proposed solutions to the issues around energy crises or the energy supply for, or sustainability issues related to data centers by the sounds of things.

And thank you for sharing all this, um, about the, provided the science fiction pointers on this, cuz yeah, this blew my mind when I first saw it and I, I think that you've actually shared a lot of useful things on this.

Anne Currie: I think it'll happen. I think it'll happen, but I think there are other things that data centers in Greenland are an obvious thing because there's tons of free energy, green energy from ice melt run-off.

Chris Adams: Ah, yeah.

Anne Currie: But one of the issues there is that nobody lives in Greenland to man the data center, but even few people live in space demand the data center.

So I would, I would hope that you would solve the Greenland issue first. Use all that enormous amount of energy before you,

Chris Adams: Okay, so our site, so before we reach for the sky, let's sort out things down here on Earth. Yeah. All right. Okay, cool. Thank you Anne. Alright, next story is a sneak peek about a new book coming out. Anne, I think this is, this is your thing. It's a coming O'Reilly book called Building Green Software. We were quite excited about some of this cuz there's a couple of co-authors who also been on this here before.

So spit on this podcast. Anne, I'm gonna hand over to you to talk a little bit about this cuz you are far more familiar with it than I am. And yeah, you know better than I do. So please do tell more.

Anne Currie: Yes, so this is an O'Reilly book that we're working on, the O'Reilly book called Building Green Software, which is gonna be there. It's not the first green software book they've done, but it's the first kind of full picture as opposed to there are quite a few good niche ones out there for things like Web development, but this one is all the things.

And there will be me and Sara Bergman, who is a key part of the Green Software Foundation and Sarah Hsu, who is also a key part of that Green Software Foundation. And so we are writing all together and the idea is to net down the thinking that. We've all, as a community, come to agree on about what's the right way to do things.

So it's all based around the idea that there's three things that we need to be good at. We need to be good at carbon efficiency, hardware efficiency, and carbon awareness. So that's what we'll be talking about in the book. So we'll be talking about carbon efficiency in terms of code efficiency, operational efficiency, plus design efficiency through carbon awareness, how designs that allow you to shift around what you're doing.

And we'll be talking a little bit more about that later, I think in this podcast and hardware efficiency. So don't cause everybody to throw away their phones every time you produce a new version of your software, cuz I dunno, you might know the answer to this, Chris, but what has the most embodied carbon per gram of anything in the world?

And my guess would be a chip. And in terms of consumer devices, my guess would be hands down a mobile phone.

Chris Adams: Do you know what? I've never actually thought about that in terms of Okay. A single kind of consumer good in terms of post embedded energy inside it. So it's true that there's a significant amount of power that goes goes into turning sand into silicon and all the other kind of materials there.

Anne Currie: But also the operation, cuz silicon fabs are unbelievably difficult to make.

Chris Adams: Yeah. And of course, and this is actually one thing we should probably talk about in a future episode. When you look at where lots of the really high-end chips are currently made, a lot of them are in Taiwan, which has a very kind of fossil fuel heavy grid. So even if the stuff is really efficient and even if they're just using electricity, that's gonna be one of the problems.

But even then, when you are making these, because most of the ways that you achieve the high levels of heat, don't rely on electric kinds of power they rely on, but literally heat from combusting fossil fuels. You've got an issue there. This is actually something that's changing. There's a really fascinating paper by Doctor Sylvia Medadu who's talking about some of the advantages in heat pumps.

You can now get heat pumps up to the high hundreds of degrees lts basically. So there are lots and lots of things that can be decarbonized now, but. For you to reap those benefits, you actually need to have decarbonized electricity in the first place. And Taiwan is struggling a bit there because it's not a really big place with lots and lots of land and it doesn't actually have much in the way of surrounding kind of shallow water for creating, say, offshore wind or things like that for the time being.

So that's gonna be an interesting one ahead of us. But yes, you're right. I guess,

Anne Currie: need the power back beamed in from space on giant lasers.

Chris Adams: yeah, maybe what they need is a death ray. Yes. Uh,

Anne Currie: in handy in all kinds of ways. If you were Taiwan as well, I would imagine.

Chris Adams: Yeah, let's leave that one there before we get taken off the internet by a advanced persistent threat. Alright. Okay,

Anne Currie: But, but anyway. Oh, that's, that's an aside. The book. The book. So the book, we are beavering away at the moment writing the book. We've submitted quite a few chapters already, so it's all going well. And the idea with an O'Reilly book, the way they do it is it's you as a writer, you writers, you submit the chapters and as the chapters are at least vaguely polished.

Vaguely, okay. They'll go live for people to read in a kind of advanced read on safari. So that will be,

Chris Adams: shortcuts thing?

Anne Currie: Yeah, so shortcut so people will be available, will be allowed to read these things. So the introduction has already gone out and it's not live yet, but we are expecting it to go live quite soon. So we will let everybody know through this when it's live and also the co deficiency chapter.

And after that we've got the various other chapters. But they'll, they'll be available in Safari quite really quite soon. Then the book, the book actually gets physically published and we get an animal. We'll have an animal, but we dunno what the animal's gonna be. So at that point we'll find out what the animal is and the book gets published.

And then, uh, so it'll be available to buy in physical form if you so choose. And also at that point it will also be available. One of the things that we agreed with O'Reilly is that it will also be available under a Creative Commons license at that point.

Chris Adams: Boom.

Anne Currie: even need to buy it or have an O'Reilly subscription.

And because if this is all stuff that is, hopefully by the time we get through this and everybody's reviewed it, it will be, this is just what we want everybody to be doing. This should hopefully be a baseline.

Chris Adams: that's super cool. I did not realize about the actual Creative Commons licensing for that. That's really helpful. That means that brings the barrier right down.

Anne Currie: Yeah, but it'll be a while cuz it takes quite a while for the book to actually come out. So I'm imagining that first quarter, 2024 will be when that's available, unless we really get our skates on and get it done much more quickly than that.

Chris Adams: I have one question, if I may before we move on from this one. I haven't used Safari and I have never written a book, but I have heard horror stories about working with publishers and emailing Word documents back and forth. Is it still that process or is there something like GitHub or what does it look like to write a technical book these days for a technical provider?

Anne Currie: For O'Reilly, you've got quite a lot of different options, and one of them, the one that we are using is just Google Docs, and so that's super easy because Sara is in Norway. Shira, our editor, is in. And I was on the west coast of the us. I'm in the southeast of England. Sarah isn't too far away from me.

She's in London, so that's quite easy to do. But fundamentally, Google Docs are pretty good for that kind of thing.

Chris Adams: Wow. So Sarah, yourself and Shira, it sounds like it'll be

Anne Currie: Sarah and Sarah. Yeah. It's really quite hard.

Chris Adams: Yeah, and this is interesting cause there are a number of existing green software books. So there's one called Designing for Sustainability by Tim Frick, who's

Anne Currie: Oh, yeah, yeah. Which is very good actually.

Chris Adams: Yeah. And then Tom Greenwood from Whole Grain Digital.

Anne Currie: yeah, yeah, that's also

Chris Adams: he had his. Yeah. And I think there are a couple of other books that I've seen come out as a number of other ones, but. The first time I've heard of one of these books, which is actually written by guys who aren't just men basically. So this is actually quite inco. I think this one book may have actually righted the gender balance in the sustainability book canon.

Yeah.

Anne Currie: hopefully, and there is method in our madness on this in that we wanted to make sure that we got on stage to talk about it as well. And we are three women who are very good public speakers, so really we should be able to make a little noise about this.

Chris Adams: Good. I wish you the best and I'm looking forward to some of the shortcuts for some of this. In that case, should we look at the next story?

Anne Currie: Okay. Absolutely.

Chris Adams: All right. Okay. This is Microsoft Scale's workload with Carbon Awareness. Now the actual story is links from SDX Central. As far as I can tell. This is basically a kind of press release talking about Microsoft and Cloud network stuff.

But the thing that was really more specific is actually some of the GitHub issues that we've linked to inside the show notes here. Basically, there is a carbon aware operator for Kubernetes to add in a bit of kind of carbon awareness into it by the looks of things. So if you go to github.com/azure, then Carbon Aware KEDA operator, there's an open source operator that you can plug into Kubernetes to do this.

And I think this is something that we've both discussed before, but I suspect you might have some records on this because I joined a mutual friend of ours, Ross Fairbanks, did some work in this field a while ago as well, actually.

Anne Currie: Yeah. Ross and I used to work on a startup called Micro Scaling Systems, which was all about cluster scheduling. And one of the thing that we always had in mind was adding carbon awareness to cluster scheduling, so moving jobs so that they, uh, get, wait until there's green carbon available. There's green electricity available on the grid.

Now Google have been talking about this as well. They aren't offering it as a service like this Kubernetes scheduler, but the idea, they've been doing it internally, they've been trialing it internally as a way of shifting workloads in time. So, so that they consume green electricity, more assiduously than they would otherwise have done.

And this is the same idea now. As far as I'm aware, this is mostly about being able to compress what's on your machines so you can turn machines off. It's all kind of bin packing on machines, cuz all the machines you wanna compress them in so that there are few machines running because they've got loads of containers running on those machines differently shaped, and they're all squeezed onto a smaller number of machines at times when there's no green power available.

And some machines got to get turned off. And in order to do that you have to have jobs that can wait. So this is not just merely a matter of scheduling, and it's two sides of information here. You need to know. What the current mixes are on the grid and what it's likely to be, which a lot of that goes around weather and grid load.

So it doesn't matter if you've got great weather, but you've got high grid load, then maybe you're still not gonna have any green power. But if you've got low grid load, maybe you've actually just got too much power and you want to be using it. So it's not just about following the sun or following the wind.

You need this information about what the grid is like, what the weather conditions are on the grid, so to speak. And you also need to know what jobs that are running in your data centers are non-time sensitive, so they can be moved around forward and backwards in time. So the same kind of things that might be running on a spot instance, for example.

Now Google pointed out that with their stuff, they're pretty good at labeling their jobs internally, so they're pretty good at labeling jobs and saying this is a low priority. You can just wait if this has to wait. 12 hours fine. Things like video transcoding for YouTube. Sometimes that happens very quickly.

You might notice as a user, sometimes it happens very quickly. Sometimes it happens and it takes quite a long time, and that's because Google. Just go. It's not a high priority thing, so something needs to wait that it will be video transcoding. So you need jobs that are non-time sensitive and are labeled as non-time sensitive.

So I say one of the things that Google pointed out that they struggled with a little bit on this is that they can do it internally where things are very well labeled, but they find it very difficult on the public cloud where VMs are just black boxes and they have absolutely no idea whether the contents could wait until there's green power available.

So for a scheduler, you need both the information on which to schedule, but you also need the information about the jobs to know which ones are schedulable. There's work to do as a user, as well as just install the scheduler. You will need to start labeling your jobs.

Chris Adams: And this is presumably something gets touched on in both books and patterns about the idea of decomposing maybe a particular monolith or a single big program into a number of smaller programs where some bits have to be. Really low latency, responding quickly, any other parts can be moved around so you can make use of either carbon or cost savings presumably.

Anne Currie: Yes, absolutely. Yeah, cuz it, cuz you won't be moving these jobs, I would imagine. You won't be moving them from data center to data center cuz you know, data gravity and all that kind of stuff. But you will be moving them in time. There's really no downside to moving things in time. And so there's no data gravity download.

So it's, that's where the wind tends to be.

Chris Adams: So just, can I check with, so you mentioned an interesting concept here, data gravity. So data gravity is the idea that one data is in one place, you are not able to move it, it's difficult or expensive to move to another provider. Is there like a technical reason for that or what's the thinking behind that?

Anne Currie: Yeah, it's, it's network, it's bandwidth, it's all, and it takes time and blah, blah, blah. But there's an awful lot of data. Gravity is one of the re one of the ways that

Chris Adams: So we're referring to egress fees here. Yeah. So paying money to get things out of your cloud storage. Oh, just by the way, if anyone, you. There's a whole FTC kind of inquiry right now about oligopoly and competition right now at the cloud sector. So this may be something that if you feel like you would like to be able to do more stuff with in terms of green computing, maybe this is a thing that you might want to respond to the ongoing FTC basic kind of inquiry into this stuff.

Cuz I feel that maybe it'd be better for us to actually be able to move things to more than just two or three clouds. Cuz data gravity seems primarily to be a kind of business constraint rather than a technical constraint.

Anne Currie: Yeah, it probably is. Really? Yeah. Technically it's diff, it's difficult, but it's doable. You can get, you could have copies of your data in multiple places. And yeah, you could move it at night. You could copy it on, you could do the whole snowmobile thing, copy, copy it all out once to a bunch of disks and drive them across the country.

It's not an insurmountable problem no matter how big the, the data is, but it is unbelievably costly. So yeah, that is fairly insurmountable because you don't, there's nothing much you can do about that.

Chris Adams: Okay, this is true just like the cost of transmission in some places actually. Alright, so we spoke about the Azure carbon aware KEDA operator. So I think maybe we should actually explain what KEDA says. Cause there's something in the briefing here. Yeah. Kubernetes Event Driven Autoscaler. So the idea would, being that this would automatically scale Kubernetes up so you have more computers or more pods, and then scale it down again in response to various activities, that's what it would be, right?

Anne Currie: That sounds plausible. I dunno, but that sounds plausible. And actually then you just use your normal scheduler and presumably your normal, however you label your pods normally on how many of these do I need to keep alive at any point? So the ones that are less important to you can just all get shut off.

Chris Adams: Yeah. There's also another related project to this called Kube Green, which is a project by some folks in Italy. Actually, this is early on. It doesn't do quite the clever kind of carbon air scheduling stuff, but if you want to dip your toes into this. It literally turns off your pods when you go home. So basically all your staging devices and your developing developer machines at 6:00 PM they switch off and go to sleep just like you might choose to go home and go to sleep.

Also, in the show notes, we've got a link from the Green Web Foundation, where we talk a little bit about this using both Kubernetes and Nomad a while ago. But the stuff from Azure looks really complete and it looks really quite exciting actually.

Anne Currie: That is, so I, we ran like a sustainability track at, uh, QCon, London, uh, a couple of weeks. Sco big conference and it was very successful. One of our, I think our top rated speaker was a woman called, another woman called Holly Cummins, who is a really excellent speaker from Red Hat. I don't know if Holly, but she spoke about that.

Her dream was that we'd have effectively light switch operations, so it should be as confident. Turning off machines on in, in the cloud or in private cloud or public cloud or wherever. As you are turning the light off with the light switch, because when you turn the light off with the light switch, no one thinks, oh, I won't turn the light off just in case it doesn't turn back on again. The aim is that you feel that confidence about all of your systems that you could just turn them off. Because you don't need them overnight knowing that you could turn them back on safely in the morning, and her light switch analogy was excellent. It's like you don't leave all your lights on at all night just in case you can't turn them back on again in the morning.

That would be madness, but that's what we do with computers. One of the best things that you can do with your systems is invest in making sure that you can turn them off and then on again.

Chris Adams: I think that's a useful piece of advice to make sure you can turn your things off and on again, if you'll. Is like a low stakes, but I think you and me, we've been on various projects where we've been afraid to do that. So I'm glad that someone is spelling it out that it's really needs to be this basic.

Alright, next story. This is one that really caught my eye cause this is the SDIA who are a Green software Foundation member, the Sustainable Digital Infrastructure Alliance. The kind of headline is the SDIA welcomes the deal of the European Council and Parliament on energy efficiency directive. This is super like legal blood gump, but basically there is some really, in my view, quite far reaching stuff inside this.

Essentially, there has been a whole bunch of laws being thrashed out about transparency around energy usage for data centers, and this seems to have snuck through in the first quarter of 2023. And there are some headlines which are in my view, which go much further than we are right now. So I'll just read some of this stuff out.

So owners and operators of data centers above 500 kilowatts will need to make the environmental performance public at least once a year. This includes annual energy consumption, power utilization, temperature, heat utilization, use of renewable energy, as in how much renewable energy you're using and where it's coming from.

And we haven't mentioned it here, but it's also water usage as well. Now these are figures which. I, and you've tried to get, I'm sure you can talk about how easy it is to get access to these figures

Anne Currie: Almost impossible. Yeah.

Chris Adams: and now they're like, it's law. Basically this is coming in. People need to be delivering, do their first reporting in March, 2024.

So things, something which a large providers have been pushing back against and saying, no, we can't possibly share of this stuff now it's. Basically gonna be part of the law in all around Europe. So if you're outside of Europe, you still may be okay, but this is quite a precedent to be setting, in my view, cuz yeah, this is something that a lots of us we've been asking for and really pushing for and now you've essentially got one block saying no, this is a condition of doing business in this part of, cuz how on earth are we gonna know if we were on target or not in harming our emissions by 2030, for example.

Anne Currie: Yeah, and in fact, this was discussed on stage at the coupon conference as well this time by Adrian Cockcroft, who is the retired VP of Sustainers. He can never remember what the titles of anybody are at aws, but anyway, who's the big cheese? Of sustainable architecture at aws and he was saying, if you're American, you might think this might, this won't affect you cause it's just an EU directive and your data centers in the US who, who cares?

But it is basically the GDPR of Green. The EU is such a big block. They have so much clout and they exert their will that this is the same way that everybody ended up having to do GDPR. This will be the same. Everyone will have to comply with these things, even if you think that you're in the US and they won't touch you.

The reality of the situation is that this will all spread out in a GPL style until everybody is forced to comply in the same way that we've had to comply with the GDPR.

Chris Adams: That's a win for transparency by the sound of things, but it's probably gonna be a headache for a bunch of people who have to start reporting in less than 11 months for the first reporting deadline for this. There's also something that I call my eye here is that any data center exceeding one megawatt of power, they need to recover the waste heat.

So basically they need to put it to good use or prove that they, it's either technically or economically unfeasible for them to be doing. So this isn't a really interesting one because within Europe at least, and I'm gonna speak about Germany where I live, like 40% of the energy demand is from gas heating things up.

So if you have this being put to actually addressing one of the other big demands for energy inside Europe, that's actually quite a far-reaching one and one megawatt. That's likely to impact pretty much every hyperscaler, cuz hyperscalers tend to be 20 megawatts upwards in size. And as a kind of, I was trying to do some like rough figures, like 500 kilowatts if you're assuming maybe 15 to 20 kilowatts per rack.

So that's, I know between 20 to 40 racks of service based on how efficient your data center might be. That's not that big. That's like a lot of data centers. This is gonna be impacting basically,

Anne Currie: Yeah, so we're gonna see an awful lot of public heated pools

Chris Adams: I hope so. Yeah, absolutely. I

Anne Currie: ringing every data center anywhere, everywhere in the world.

Chris Adams: Maybe this will change how we think about how you build data centers. Like when you build a data center as a kind of big box out of town, Walmart style thing, then it's really difficult to use the heat. But if you're able to integrate the data center into the kind of fabric of the urban environment, then there is

Anne Currie: really want to have, but you don't. But that has issues of its own. You do not want that generally because in the open environment, you want people living. And also you don't want the draw on the grid, cuz often those cities, the grids are already overloaded. So it would be counterproductive to have a whole load of data centers now suddenly located in urban environments just so that they can have a local pool that's heated up using their excess path.

I would say that's counterproductive, but.

Chris Adams: they're providing or generating any of their own, any of their own power on site. That's another thing that some of the new providers are doing. They're basically looking at using batteries on site as a way to act as a kind of anchor customer, but also to provide use. Cause if you have this case where you're scaling machines up and down, there will be times where you should be able to be a kind of active participant in the grid.

Just like how having a kind of read write energy grid, just like we have a two-way internet and you could have a two-way grid, but that's a, another podcast I suspect.

Anne Currie: Part of the grid balancing solution, which is absolutely required, particularly when we struggle with grid balancing at the moment, and that's when most of the grid is powered by stuff that is utterly predictable, like gas or coal. When he starts adding a whole load of comparatively massively unpredictable solar and wind into the mix, then grid balancing is a major problem.

Chris Adams: It gets more complicated depending on how much of a grid island you might be. So if you are connected to other things, You can get stuff from neighbors, but if you can't then it's a bit more complicated. Now there is some interesting news related to that. I, I assure you, were not an energy podcast. There is basically an energy England to Dutch interconnect just announced in the last week and there's a bunch of similar stuff happening around this field, but we probably need to discuss that another time.

And I think we're coming up to the last few minutes of this and I think there's been a question being that's been posed to us that I think Chris, our producer, shared. If we were to launch one data center into space, what would you name it and why? I am. You can have a bit of a thinker like, and I'm gonna go for the dataface out answer that people tend to use when faced with this stuff or what English people tend to use when they get the chance to name things.

Boaty McBoatface, the well-known research vessel, was doing some absolutely fantastic work in the field of climate science. I'll share some links specifically for that because yeah, both face or Richard Attenborough or the Sir Richard Attenborough is its official name. That's a thing. So yeah, that's my example.

That's my answer. Data McDataface. What about you, Anne? you call a data center?

Anne Currie: I dunno, but I can immediately say what I would choose as my naming convention. I would give them culture ship name conventions, the Ian Banks Culture Series, all the AI spaceships name themselves, with some slightly tongue in cheek name.

Chris Adams: Of course I love you one and

Anne Currie: Oh yeah, exactly. Yes. Yeah, so I would, I would give them culture names.

So that's up. There's an exercise for the listener to come up with a whole load of, in fact, I believe there is a cultureship named generator online that you can, it will automatically, or to be a perfectly honestly ChatGPT, but almost certainly supply you with culture ship names that it has made up. So I would defer to Ian banks and in the, and,

Chris Adams: And a generative AI large learning model. For naming our servers. I guess that's a circular of nothing else. All right, I think that takes us up to the time we have here. Okay, that's all we have for this episode of The Week in Green Software. All the resources for this episode are in the show description below, and you can visit podcast.greensoftware.foundation to listen to more episodes of this particular show.

Thank you very much, Anne for joining us, and hopefully see you on one of the future ones. So bye for now. See you around Anne.

Anne Currie: Goodbye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation In any browser. Thanks again and see you in the next episode.

Fact Check: Ola Fagerström and Microsoft's Surface Emissions Estimator

26 april 2023 | 45 min

In this episode, we have a very special guest for an episode of Fact Check on Environmental Variables, Ola Fagerström from Microsoft Surface joins host Asim Hussain to talk about his work on the Microsoft's Surface Emissions Estimator - an important tool that helps measure the carbon footprint of the device. Ola talks about how difficult it was to take into account everything from materials used, to manufacturing, to packaging, and even end-of-life disposal to give an accurate estimate of the emissions produced by each device; and how these principles can be applied to other areas of green software development.

Learn more about our people:

Asim Hussain: LinkedIn / Twitter
Ola Fagerström: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Resources:

Microsoft’s Surface Emissions Estimator Announcement
Ola’s Post on LinkedIn about Microsoft Surface Emissions Estimator / LinkedIn [7:46]
Energy Star Calculation / Energy Star [15:51]
Microsoft Edge Green Tabs / Microsoft [23:13]
Internal Carbon Fee / Microsoft [37:12]
New EU Laws on CSRD / European Commission [41:59]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcription Below:
Ola Fagerström: When is the first company going to start to say, we only allow eight tabs open? Because if you start to have, I'm just making numbers up, 16 tabs, you might run to your boss and say, Hey, I need a device with the 32 gigs of memory because my memory is constantly filled. Yeah, sorry. We put a policy that you can only have eight tabs open because that will save on the memory, and therefore we can buy cheaper or devices that are actually greener.

I'm your host, Asim Hussain.

So welcome to Environment Variables, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Asim Hussain. In this episode, we have a very special guest for an episode of Fact check on Environment Variables from Microsoft Surface. We have technology specialist Ola Fagerström.

Ola Fagerström: Hi there, Asim. What a nice, uh, way to get introduced as a special guest and.

Asim Hussain: Probably worse ways of being introduced isn't there? Not so special guest anyway. Ola, like it's great to have you on the show. Obviously we were colleagues at Microsoft, we're both circling kind of the sustainability space. I'm not at Microsoft anymore. So to give our listeners some context, could you please introduce yourself?

Ola Fagerström: Yes, I'm Ola Fagerström. I'm based and live in the fantastic country of Sweden, where probably some of you listeners have heard of a small girl called Greta. Which is also helping, pushing the environment forward and doing a lot of stuff. So part of, I can say part of that inspiration, but I've been with Microsoft for in, oh, what is it this week?

11 years.

Asim Hussain: Oh wow.

Ola Fagerström: So quite some time I've been with the devices and Surface family for the last, what is it? Eight, nine years since we launched Surface Pro Free back in the market. Way back in the days. So, uh, that is sort of my day-to-day job and then working with sustainability almost on a daily basis to make sure that we can.

Help and tell our customers what we actually do, and I report back to our dear engineers what they have to do better

Asim Hussain: Ah, yes. Yeah. Yeah. That

Ola Fagerström: as well. Sit sitting a little bit that in between of, uh, explaining difficult things to people on the simple way on one end and the other way around as well.

Asim Hussain: Yeah, and you have to explain the sustainability difficult things and even more difficult. So thank you for that. Yes, we've really great to have you on the podcast to talk about your work. So to just explain how this works to our listeners, this is a type of episode which we call Fact Check. It's slightly different from our typical Environment Variables or this week in Green Software TWiGS episodes.

We're first going to ask you a few questions about your background and experience in the world of sustainable software, on the world of sustainability, and then we're going to do a bit of a deep dive into your work with Microsoft and the Microsoft Surface Emissions Estimator, and then we'll just see where that takes us.

Ola Fagerström: Sounds like a good plan, I think. Yeah.

Asim Hussain: Yeah. Um, just a, before we dive in, it's just a reminder for everybody that everything we talk about will be linked in the show notes below this episode. So here's my first question. Can you tell me more about your background and how you became a device sustainability specialist at Microsoft?

Ola Fagerström: Yeah, so one of the things, as I said previously, coming from a region or a country where sustainability has always been on top of mind. Of our customers, people living here, it's been a, a constant topping, bringing back to our dear friends, sitting on the other side of the pond to ask them to do more. And then at one point in time, then you realize as well, okay, if they can't provide you with everything, then you actually have to start to dig in yourself.

And help to do more as well. That is creating the thing that didn't actually exist. And that started quite some time ago, years and years back.

Asim Hussain: So that's cuz like you're in the surface space and imagine you're getting asked a lot of questions. Is that I bet

Ola Fagerström: Absolutely. And going back 4, 5, 6 years ago, bringing that back to maybe a market where sustainability was not the top of the agenda. That was in many cases, yeah, but we have some legal information you can find online. We can send you a PDF and things like that. But that was not so much maybe what my, my customers and I wanted to know when we started to think about those things.

Asim Hussain: Yes. And that's really good. I think Anne Currie, who, who's one of the co-hosts of podcasts, talks about this quite often, which is if you are a customer of an organization, asks questions cuz that's what gives Ola and people like Ola kind of the imperative to do this work that we're describing here right now. So just before we really dig into kind of the service missions estimates, I know you mentioned that you live in a country where sustainability is a key, a very important thing. But would you say there was a particular moment or turning point in your career where you decided to focus on sustainability?

Ola Fagerström: I think one of those key things when I really felt like, okay. Even though I think this is important now I know that the company feel that this is important. And one of those key things was of course, when Microsoft in 2020 went out and said that we are going to be carbon negative, we are going to be zero waste, we are going to be water positive.

All of those things. One of the fine lines in the announcement said as well, that one of our focus things is around our product and devices. And I was like, don't really know all about that. Let's see what's behind that.

Asim Hussain: Yeah. I love that answer because I've said this before and I, cause I was at the same company at the same time, and when the leader of a company comes out publicly and makes a statement, it gives freedom to the people working inside that organization to then follow through on those actions. Cause then when you're having meetings, you can just say to people Sachin, he said it on stage, so you know, you better have this conversation with me and let's talk it through.

Whereas prior to that public announcement, that kind of conversation is more difficult, isn't it?

Ola Fagerström: And, and the interesting in this, and then we'll probably discuss this a bit further on as well, if you are listening as a, as a customer and you have a company that has said something publicly, in most cases, the one who has signed off on that is in many cases the board of directors or the CEO. So you can quite easily start to hold them accountable to that.

Okay. But you said this. We have set up this call. You and your dear friend at the table has actually signed up from this, so how are we going to do it?

Asim Hussain: Yes. That's the wonderful thing, about making a public commitment. Whether it's telling all of your friends, you're gonna quit smoking, or whether it's a public sustainability commitment, there's a certain amount of peer pressure to keeping you honorable. So let's dig into the work that you've been doing.

Firstly, can you give us a quick overview of what the Microsoft Surface Emissions Estimator is? And you know what it does? Am I saying it right as well? Is that the official title?

Ola Fagerström: Yeah. Yeah, it's a very long word. And of course we have legal friends that's making sure that we use exactly the right wording. So therefore, it's not the calculator, it's an estimator and things like that. At,

Asim Hussain: Mm. Mm-hmm.

Ola Fagerström: so it very much came from that announcement that we said that we are going to be as a company, carbon negative. When you start to dig into the details of that, you are also saying, see for example, that we are publishing what we call from our side an Eco report. So you can see, okay, what is the estimated impact from purchasing, a, surface Pro Nine, let's say. Those are public things. That's a PDF that you can find online, but then it quickly starts to come and you start to nerd, as I did around those things because in all of those PDFs, there are some fine lines saying, oh, the numbers in this PDF comes with the fixed assumptions of whatever it might be.

In our case, it says that the numbers in this case is based on the assumption that device is used in the US for example, for four and a half years.

Asim Hussain: Mm-hmm. Mm-hmm.

Ola Fagerström: Brilliant. If you're a customer that's using it for four and a half year in the US but if you start to think, yeah, but I'm a UK customer and I'm going to use that device for two years, is that really accurate for me then that number? So I start to then to dive into, okay. But we probably can calculate that in a much more granular and much more accurate way for all those customers who don't fit that four and a half year.

Asim Hussain: Hmm.

Ola Fagerström: The surface emission estimate is all about to make sure that you can understand what is the footprint from me with the assumptions that I say in the tool versus assumptions that someone has made for me.

Asim Hussain: Hmm. Okay, so you're just trying, so you, you, I dunno if it's this, if it's just the Swedish way of saying it, but you called it a eco. Is it LCA report you're describing before?

Ola Fagerström: Yeah, so in that, in that eco profile, you then see that the numbers are then based on what we call an LCA calculation or a life cycle calculation. If you don't know what an LCA is, basically you collect all the information that you have about the product that you intend to manufacture, and you can do some estimation of what is, for example, the environmental impact from producing that.

Asim Hussain: Because effectively what an LCA sounds like, an Eco Report is just another, or a summarized version perhaps, of what an LCA form sounds like. Maybe something that's formalized for devices. It perhaps is, it sounds like. Yeah.

Ola Fagerström: And you can find those eco reports for any given product. It's basically an LCA is not something specifically to the pc. It's done in the car industry and it's been done for a very long time. But we are on the path of where that data becomes super crucial for customers to understand.

Asim Hussain: Because, okay, I see. So like for with a typical LCA report, what you're doing is you're just basic, effectively just writing a PDF because the technology has not moved on since 1994 about the ability to disseminate knowledge to other human beings. Just the PDF is the pinnacle of data dissemination in the sustainability arena.

But the challenge with the PDF with something digital is you have to make a ton of assumptions like what you just made because it is just a static bit of data. So like, how would I write a static bit of data? About the carbon emissions of my laptop. You have to make a ton of assumptions. You have to assume what my user profile is.

How much do I use it? Where am I? How long am I gonna keep it for? But that's all statically written down in a PDF. From what I'm hearing, from what you're saying is the estimator like a dynamic LCA, where you can put your real values in and it gives you your real estimated values back out.

Ola Fagerström: Yeah. Cause what we have done now in the estimator is basically we have taken that LCA and broken it apart because in a normal LCA then, or in the output from that, you say, okay, how much in this case, carbon emission is associated with the production of that device. How much comes from transporting it to the location where it's going to be used?

How much carbon emission is going to come from using that device in the PC, and that's how much electricity is that going to use in the location where you might be. And then the last one comes from the end of life treatment. So what we do in this case is that we say that use the production part be because that's very static.

But then the transportation part is dynamic because we know, for example, then by our own logistics, what does it take to ship a device from our factory to maybe a central warehouse? To a central warehouse to the end location where the customer might be? The other thing as well is that. The energy grid in the UK might be very different from the grid in Sweden or in India or somewhere else.

So we need to start to think of how much energy or how much CO2 is associated with, for example, one of use in the UK versus Sweden versus India. So that's what we also then bring into the tool to show that, okay, how much energy is that device actually going to use? And then we have. Done some clever stuff in it as well, where we actually start to show actual usage or telemetry data in that to make it even more granular.

Asim Hussain: I believe the estimators designed a, probably a, and you tell me if I'm wrong, obviously tell me if I'm wrong, but for organizations that perhaps would buy a rather large amount of surface devices and manage like a fleet of devices, and so then typically if I was in that organization prior to the estimator, all I would've ever been able to do is take the LCA report.

I've bought 10,000 services. Multiply, whatever that number report by 10,000, and that's just my carbon emissions. Now I can then put in some real values. And you mentioned transportation, so I presume what you can, one of the things you can say is 1000 of those machines were delivered to our France offices in France, 1000 in Germany, 2000 in Singapore, or something like that.

So the estimator sounds like that's one of the variables it takes into account. So also sounds, from what you said, it also takes into account grid intensity of electricity to also take that variable into account. Are there any other variables, like what other variables

Ola Fagerström: can I factor in?

Yeah, so you can also say for how many years I'm going to use it. So you can say, I'm going to use it for two years, or three or four years instead to take that into account as well. Okay. How long are you going to use that energy in that location? But the cool stuff in the tool as well. Like we said, we're going to link to the tool so you can look it up yourself.

Is that on the used part? We actually have two bars in there that shows. One, how much energy is that device using according to the Energy Star standard? Because all the devices say that, oh, we are rubber stamp. We meet energy star standards. But what people don't realize, and I have to share a little bit of a, a hard moment when I did my digging out about this tool, is that when you are reading the formula that Energy Star is using, To measure the amount of energy a device, tablet, or phone is using, they are saying in the formula, okay, you as a manufacturer has to measure how much energy is that device using and the grand prize for you Asim, if you know the answer to this, how much energy is your device using when it's in, in off mode?

Asim Hussain: Ah,

Ola Fagerström: When it's in what they call sleep mode and what they call as long idle and short idle, but what are we missing in this calculation?

Asim Hussain: Uh oh. Well, it's idles. It just sounds like effectively off states, isn't it? So it's just like idles, sleep, off. Not actual. Also perhaps as a mirror of what I am actually like when I'm in front of my computer, which is fairly idle and sleepy, but doesn't really like factor in like as sometimes on a rare occasion I'm also quite active.

So it sounds like it's not taking that into consideration.

Ola Fagerström: So the funny is that when you explain that to people and say that, imagine that you would go out to the garage and look for that new car, and then they would have in the fine print and say, oh, by the way, the amount of fuel this car is going to use, it's measured by you parked a car in the parking lot, you put in the key in the ignition.

You turn the key and let the engine run for a little while or for a little while longer, but you don't leave the parking lot.

Asim Hussain: Right.

Ola Fagerström: And I was like, yeah, but that can't be right. I must be missing something in their calculation. So I asked one of our engineers that's doing the Energy Star certification, ask them.

Is it really right that Energy Stars actually don't calculate for actual usage? He said yeah, there's a small formula for it, but you have to think of where does Energy Star come from that was created to make sure that you lower the energy usage from refrigerators,

Asim Hussain: Uh,

Ola Fagerström: that, air conditions, which has a very fixed on and off or idle, but they don't actually take into calculation the actual usage of a tablet or a pc.

Asim Hussain: interesting.

Ola Fagerström: So when you start to think of, oh, we have a energy saving goal in our company, great. As a starting point that you ask for energy star certified devices. Of course, I'm not arguing by that. But it won't take you to your energy reduction goals because you need to start to measure actual usage. So what you will see in the tool is one bar saying, this is the amount of energy that energy stars are requiring to meet Energy star certification.

Then we have a green bar saying, this is the amount of energy that we see collected from devices being used out in the field. For example, the Surface Laptop studio that might be used for CAD work and stuff like that. Of course, it's going to use a lot more energy than just to log into windows and don't do anything.

It's like measuring the amount of fuel a car with a turbo is using based on that you're just parked is on the parking lot. Of course, the turbo is never kicking in. If you're just standing there and looking at technician.

Asim Hussain: So, cause I imagine what people perhaps have done in the past with the absence of an LCA, maybe with one is take the energy star, I'm sure energy star as well as giving you A, B, C, D, E, F, or whatever it is. It's probably giving you an energy consumption per minute or something like that. And if you were just to use that and multiply it by the number of minutes you're using it, but it would not reflect cuz that's effectively saying like that's how much energy your compute would use if it was idle.

What you are reporting in the estimator is actual usage. Isn't that a bit, I'm sure like that bar is so far high than the energy star bar, isn't it just a pretty depressing like thing to look at all the time?

Ola Fagerström: No, actually, it's interesting because in some cases that bar might be higher. In some cases it's actually lower because some cases we also see that the device is used as intended. For example, the Surface Go, which is more sort of an iPad format. People might use it, just log in, check their emails, not doing anything more, so the energy usage will be actually lower than what we see when we did the testing.

So this is also then seeing and giving the customer the deep insight because again, the nerding part of this, and this is the fun part of listening to podcasts, that you always learn stuff. When you are reading those eco profiles from the other large manufacturers, the sort of little fine line saying in there, the use of this device is calculated according to the Energy Star standard.

So that means that, oh, we say that you're going to use 15 kilos of co2. Okay. But that's just based on that me logging into windows and not doing anything. So what we now do in the tool is actually using that telemetry data to saying that a hundred surface pro nines are going to use, I'm just making up numbers now, 900 kilowatt hours.

Asim Hussain: So just want to clarify one quick point before we dig into this. So is that information available kind of per customer, or is this something that you're aggregating up and just making available yourself? Or, if I'm a customer and I'm using the emissions estimator, and I've got 10,000 machines, would you tell me the average consumption across all my machines?

Ola Fagerström: No, it won't be across your machines, but it's across the sort of Windows estate. But it's based on that specific skew and product. So it's for a Surface Pro 9 i5, whatever it might be, coming from the data reported back to Windows and Microsoft.

Asim Hussain: So this actually could be quite useful information for a lot of people out there in the world. Like they could actually go to this and then see, forget energy star. You've effectively got a real world model of usage, real human UX, user experience of devices and energy consumption that goes there.

Fascinating.

Ola Fagerström: So, of course inside of today, that's only a long green bar saying this is the energy usage. Then we have to start to think of what you are doing on your daily, which is making sure that the software that runs in that green bar, Because then we can start to slice that up and say, oh, what is, for example, Edge using as part of that, or whatever kind of software I'm running on that device.

Asim Hussain: Can you reveal? Cause I've, I think I've heard data from various sources. I don't think I've heard anything official from a Microsoft perspective, uh, or just stay quiet if it's true. But is it true that browsing or browsers take up, is it 60 or 70% of the energy consumption on most kind of laptop devices?

Is that something that is too far away from what you think is true?

Ola Fagerström: I think the interesting in that question is that if you look at what we have done in Edge, where we now can start to put the tabs into green mode, which is actually quite interesting because then you can start to measure those things as well to see, okay, how much are we actually saving by using that kind of green mode or whatever kind of green or energy saving that you have put in your software.

And then start to think of how are we collaborating in that sense with the OS that device is running on.

Asim Hussain: That was the feature of Edge that dragged me onto Edge was that that was, it was both useful from a sustainability perspective and also just from a human perspective. Cuz I'm one of those people that just opens a ton of tabs and it's just way more performing.

Ola Fagerström: Yes to nerd on that a little bit. It's kind of interesting as well when we start to talk about in this world to make sure that we save on the environment. So when is the first company going to start to say, we only allow eight tabs open?

Because if you start to have, I'm just making numbers up, 16 tabs, you might run to your boss and say, Hey, I need a device with the 32 gigs of memory because my memory is constantly filled.

Yeah, sorry. We put a policy that you can only have eight tabs open because that will save on the memory and therefore we can buy cheaper or devices that are actually greener.

Asim Hussain: That's a very interesting angle to take on this whole world. You're right, actually, yeah. You constantly install things on your computers. They slowly degrade over time. And then all you have to do is reinstall everything and start from scratch and everything's really fast again. So yeah, there's a slow degradation which forces you into action choosing tooling.

I think that's also another factor, like tooling should be,

Ola Fagerström: Absolutely. And. It leads me also sort into the discussion where another thing that people constantly talk about now as well is of course circularity in in, in those things as well. And where I, cuz I'm in a lot of discussions with customers and they often ask, well, yeah, but what kind of questions should I ask for the, yeah, probably the first question you should ask for, what is the residual value of the thing that I'm going to purchase?

Asim Hussain: When you say residual value, would you mind, would you mind kind of clarifying what you about residual

Ola Fagerström: Yeah, of course.

Asim Hussain: Yeah.

Ola Fagerström: Cause when you start to think of when you purchase something, you are probably then doing it with the intent of either using that thing until it falls apart and you're going to recycle it, or you're going to purchase something with the intent of, I'm going to use it for X amount of time and then resell it. We know for a fact something with no value at all will most likely just be thrown in the bin. Something with a high value will for sure be sold over and over again, regardless if it's a PC, a phone, or whatever it might be. And I often refer to this as well in calls and say, nobody would think of taking their three year old iPhone and throw it in the bin.

Everybody knows, Hey, I can just go on eBay and get $400 for it and it will definitely be sold again. Exactly the same thing we see with Surface. My daughter is still using my Surface Pro free that I got. What is it now? Eight years ago,

Asim Hussain: Yeah,

Ola Fagerström: a couple of weeks ago when we checked on eBay, that was still worth $150.

Compare that to something. Made of cheap plastic that nobody cares of, that would've been thrown away a long time ago.

Asim Hussain: It's a very interesting point as you remind you. Cause I, I got a, a electric car a couple of years ago and I leased it. And what was fascinating about that experience is that they, they still look arguably very expensive for a car. However, the thing that was fascinating about with the lease price was actually quite low.

Ola Fagerström: Yep.

Asim Hussain: after I've finished leasing it for two years, its value remains high so they can lease it for less, if that makes sense. If that math makes sense. Yeah.

Ola Fagerström: interesting to this is again that key question. When you purchase and you want to come into circularity, how are you going to enter that and how long are you going to use that? Whatever you buy until it falls apart and you're going to make sure that you repair it yourself. Or if you repair something, send it back to repair centers and things like that.

Or are you going to do, just as you said, making sure that working with the lease partner that's actually going to make sure that electric car's getting sold again and again?

Asim Hussain: That's very interesting. Yeah, I think the world is gonna have to move towards that secondary market model. And I think it's interesting cuz in the tech space, I would argue that four or five years ago, like it really mattered getting, I felt a very significant difference getting the latest model in my life.

I would argue these days, maybe I just don't do as much hardcore as I used to, but like for most of the work that I do, which is having a meeting, browsing, writing a document, like there's almost no difference. Now if I was to get a new machine in terms of any performance, like a machine that's like 4, 5, 6, 7 years ago, would do just as fine for me as a machine right now.

Which maybe indicates, and I used to this, the nature of devices is changing.

Ola Fagerström: And I think also when we start to think of the other thing that you used to work with it, which is of course Azure, back in the day at Microsoft. How much power do you need in your device versus how much can you actually start to use from somewhere else when you need it?

Asim Hussain: Hmm.

Ola Fagerström: Without your own organization need to, uh, purchase 10,000 servers and have to build all of those things where you today can then buy it from Windows 365, which powers next generation of windows.

So, do I need to buy a laptop with. I nine just because I'm making some CAD drawings for a couple months, or can I just buy that as a license when I need it for X amount of time?

Asim Hussain: The There are virtualized desktops as well. I think that's, that's kinda the thin client argument, isn't it? Like we could get away now, perhaps a lot. I think a lot of people will probably get away with very thin clients and then just a lot of the compute on the cloud and then that can be hopefully have greater efficiency, efficiencies of scale

Ola Fagerström: I think so, definitely too. And something that I'd like to point out, what is a little bit special with the tool or actually the methodology that Microsoft and the eSign team is actually using now? To do that LCA, which stands out as very much as the gold standards for everyone else, is that when you're doing that LCA, you can do it in two ways.

Either you can use a simplified tool, which is based on an either an algorithm or you can use a tool that then takes into account your own supply chain.

Asim Hussain: Oh.

Ola Fagerström: That means I've made this little simple analogy. So if you are listening to this and you've imagined in your head that you work in IT, and then our dear friends at the communication department comes to us and ask, Hey, you work in IT.

You can always help with everything. Hey, of course we do. We do it all the time. We are going to run a super expensive ad campaign. Okay. We need you to find a new picture supply for that. Okay. Also, I can do that. Okay, great. The only requirement we have is that it needs to be a picture of an orange and gray super sport bike.

Okay. I know exactly what I'm going to ask for. Shouldn't be a problem to find a new supplier for it. The first supply took them about 20 minutes to do the job, it only required one person. They used a simplified tool where you just put in a few keywords today, orange and gray, super sport bike, standing inside your studio.

Voila! You have a picture of an orange and gray bike. The tool that you used may not give you who took it because nobody took it. Can you see when it was taken? Probably not. Is it a hundred percent color accurate? Most likely not either. The other supplier, it took them three bloody months to do the same job.

They used a super expensive tool to do it. They used a super high resolution sensor. Came back with the most astonishing picture you can ever imagine. You're zooming to every screw. You can see who took it when it was taken a hundred percent color accurate. Could you compare those two? Of course. Cuz Hey, you just asked me to find a supplier of orange and gray bikes.

Could you really compare that? Probably not. So what most manufacturers today are using is a simplified tool because their catalog of devices is so large,

Asim Hussain: Right.

Ola Fagerström: so they put in a few keywords like. What is the size of my screen? What is the material that I might use? What is the thickness of the motherboard?

It takes about 20 minutes to half an hour. What it gives you is sort of a half blurry picture because it's not the display that you exactly are using or exactly that disk that you are using. What Microsoft has done very differently now is that we have gone out to our own supply chain. And collected exactly the measurement of what does it take to produce that specific screw that we use in a Surface Pro 9, if I just make something, or that specific motor board that is used in just that device, and that takes all the way from producing that raw material.

To getting it to the next supplier, to the next, to the next, to the next. So of course, that spreadsheet is, you know, thousands and thousands of rows of processes, and then we produce that life cycle for that specific skew using the bill of materials or the bomb for that specific device.

Asim Hussain: So I'd always assumed, actually that's how everybody does it. So what you are saying to me, that's how Microsoft does it. But if I'm another manufacturer, there are tools out there. I'm some sort of Web interface tools where I can just answer a series of questions roughly describing my head. I've got a head, I'm holding headphones right now.

It's these headphones that I've manufactured. And it will roughly generate cause it's gonna have, it's gonna be so averaged out now, isn't it? Yes. I've got a bit of metal. But what metal is in this? How much, you know, within 20 minutes it'll gimme a very rough report. But you are saying there is multiple levels of quality of LCA reports that are out there.

Is there some way, as a consumer I can know that this report was built using not, yeah.

Ola Fagerström: what you need to start to ask, especially if you are an organization and purchase something, the first key question then, okay. The data that you provide me, or this number that I saw in your eco profile or whatever it is that they call it, is that using what the nerd calls primary data means data collected from your own measurement and supply chain.

Secondary data might be data that is publicly available in some sort of database. What we could see when we did our own measurements, we could see, for example, that the assumption of how much energy was used when you, for example, molded out the chassis of a surface device, and we could pair that to the measurement from our actual supplier of it.

We saw that the energy usage was 20 times higher.

Asim Hussain: So you actually created this manually yourself or some other mechanism through in-house manufacturing. You created your own chassis or what case you said, and then you calculated that was in a controlled environment, so you calculated the energy you needed to do that. You went to your suppliers and then asked your supplier for the same information, and then number was 20 times less than your number, is that what you're saying?

Or the other way around?

Ola Fagerström: No. So what we did before was we used one of those industry available databases that said that if you are producing one kilo of aluminum into a CNC chas- chassis, you're most likely going to use this amount of energy. So we could say then, so don't hold me accountable for the numbers, I'm just making some up.

Let's say that to create 500 grams of surface chassis, you may be using 15 kilowatt hours of energy. But then when we actually compared that and went to the supplier who did that molding for us and ask, okay, how much energy are you using to create 500 grams of that chassis? And they said, oh, but that's 150 kilowatt hours that we use.

Asim Hussain: Oh, so suppliers were telling you like a greater amount than these existing kind of emission factor models.

Ola Fagerström: Exactly. Because when you start to look down to where those model comes from and the assumptions of that, in some cases, those models and assumptions might be 20 years old.

Asim Hussain: Wow, I suppose you are actually trying, cuz you're trying to model everything in the world. Everything that could possibly be modeled in a supply chain.

Ola Fagerström: The interesting to this is also what many people are not aware of. So inside Microsoft we have what's called the internal carbon fee. So all parts of Microsoft needs to pay a carbon fee depending on how much they add to the organization.

Asim Hussain: In terms of carbon emissions?

Ola Fagerström: Yeah so as taxpayer, we all know we want to have control of our numbers to make sure that we pay as little or, as much tax as we have to. Of course, to be able to do that, you need to have measurements from the people that you are purchasing things from.

Asim Hussain: Yeah, I know. I remember for the Microsoft announcement, that was one of the announcements was one of the pressures that Microsoft put on its suppliers was to provide that data. Which is one of the hidden aspects of that, but you had to dig a couple of layers deep into that announcement to get that. But yeah.

Ola Fagerström: And that, of course, then starts to play into account. Cause what we do then in the estimator is that then we are both showing very accurate numbers based on that specific skew that you are looking at. Which comes from the measurement of our own manufacturing and supply chain. And then couple that with measurements from the logistics and then we take measurements from actually using it device based on telemetry in the location and for the years that you're going to use it.

So when we start to compare that there is, okay, here we are providing you with a Hasselblad picture for you picture nerds out there, you know what it is and the image quality. Versus the simple picture that I got from Dall-E or using Bing Create. Very different in when it comes to quality and control of everything.

Asim Hussain: That sounds really fascinating. So you've got better numbers, you got more refined. Numbers more accurate, more precise numbers. How are these numbers being absorbed back into Microsoft in its sustainability? Are they being factored into the sustainability report and kind of Microsoft commitments or, yeah is this a separate measurement?

Ola Fagerström: So of course being able then to both model your supply chain, then you can start to take informed decisions when you are selecting what you want to do with your suppliers to make sure that you meet your sustainability goals. That could be things like, okay, which suppliers should we invest in to make sure that they have green energy?

Are there some suppliers that adds more carbon footprint to our devices than others? Where should we put our effort and investment? If you don't have that kind of measurement from your supply chain, then it's easy to say, oh, we are just going to go out and plant a couple of a hundred trees and hope that we are fine with that.

Okay, great for the planet. Maybe not so great. If you're going to meet your sustainability goals. Don't want to cackle down on people saying that we're planting trees, but or for them who plant trees. It's a good thing you do, but you need to start with the basics.

Asim Hussain: Yes, and I can see actually the drawbacks then of using those simplified models would be that every single manufacturer that uses that simplified tool to generate it, it would give you poor information for what to optimize for. Cuz if it's just an emission factor for thickness of motherboard rather than constituent, then that's just gonna apply pressure to have a thinner motherboard.

Because you're assuming a thinner motherboard. Cause that's the only variable there that you can tweak. Whereas the actual surface landscape that you wanna optimize for is very different. So that's a really fascinating insight. I think we've run, we're getting close, closer time. So just wanna ask one final question.

You know, what advice would you give to other companies or individuals looking to reduce their carbon footprint? Other people in similar roles that you perhaps in other companies.

Ola Fagerström: I really urge people to say or ask exactly what I said. Ask your supplier, how did they get to their number? Are they actually measuring that from their own supply chain? And I mean, that could be if you are just going to purchase a new office desk, find the eco profile for that office desk and ask that supplier.

Have you measured that from your own factory? Because it will tell you both how much they have actually invested in people like me and you to actually do this work and make sure that our customers can actually get that kind of data. Because if they even don't know, it'll just give you a rough idea about what it is.

Yeah. It tells you quite easily what it is.

Asim Hussain: Mm, very interesting. Yeah, ask them if it's primary or secondary data.

Ola Fagerström: Yeah. And again, if you are working at the company that has gone up and said, okay, we are going to meet this goal, and you are actually manufacturing something or writing code, or whatever it is, are you actually measuring that and how are you providing that to your customers? Because with the new laws coming in EU with CSRD and things like that, I was told in one meeting that organizations might need to report on a thousand data points when it comes to sustainability.

Asim Hussain: Good.

Ola Fagerström: That is the work that's coming just ahead of us.

Asim Hussain: Yes. Yeah. And so asking lots of questions and you have to ask the questions to get that data to be able to report, and

Ola Fagerström: And then also be able to say, no. Okay, dear customer, if you can't provide me all supply, if you can't provide me with that, I'm going to go to that other manufacturer and I have this other little analogy that I use. If you would go to the car or to the garage that we talked about car before, and you would say, okay, how far can I drive in this electric car?

And they would say, oh, oh, you can easily get a thousand miles on every charge. Brilliant. I would buy that car directly. And then you ask that little hard follow up question. By the way, have you tested that yourself? Oh, you know, no. No. That is based on that. It's an electric car and it's a sedan. Oh, so you haven't invested in testing that yourself?

No. That would be way too expensive and required way too much work for us. No. Maybe in the future we will do. That's the work that Microsoft has already done because we set up that work years back again. When we started to say, okay, we can't select you as a supplier if you can't provide me, and we will make sure that we can provide our customers so they can actually feel secure when they buy our products to understand what is the footprint of things.

Asim Hussain: Wonderful. Full. Yeah, great information, great insights from somebody like yourself who's been at the forefront into really calculating these numbers. And I can really see, and I, I knew you from before, but I can really see that doing this properly is really important to you and making sure that there's a truth to those numbers and a usefulness to those numbers.

So thank you very much. Ola, thank you so much for your time, for time today and spending time. And thank you so much for the work you've been doing at Microsoft.

Ola Fagerström: Thanks a lot for having me and fantastic to know. It out a little bit over over an hour as well.

Asim Hussain: There we go. So that's all for this episode of Environment Variables. All the resources for this episode are in the show description below, and you can visit podcast.greensoftware.foundation to listen to more episodes of Environment Variables. I would like to say huge thank you to Ola again for being on the podcast today.

Thank you for sharing your unique input to the world of sustainable software, and thanks again for coming on this show.

It helps other people discover the show. And of course, we want more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation . Thanks again and see you in the next episode.

The Week in Green Software: The Hidden Cost of AI

19 april 2023 | 41 min

This week host Chris Adams is joined by Asim Hussain and Environment Variables regular Sara Bergman to discuss the hidden costs of generative AI. What’s really at the tip of this iceberg and how far down does it go? They also discuss just how thirsty AI chatbots really are and developments in platform engineering. Finally, we share some opportunities for development from the world of green software.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Sara Bergman: LinkedIn / Twitter
Asim Hussain: LinkedIn / Twitter

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

AI Chatbots Guzzle Enormous Amounts of Water, Study Finds: / Evening Standard [3:16]
Two-phase cooling will be hit by EPA rules and 3M's exit from PFAS "forever chemicals" / DCD [9:35]
The Mounting Human and Environmental Costs of Generative AI: / Ars Technica [15:02]
ChatGPT: Mayor starts legal bid over false bribery claim / BBC [20:29]
Free Dolly: Introducing the World's First Truly Open Instruction-Tuned LLM / DataBricks [25:19]
How Platform Engineering Makes Software Sustainable: / Devops.com [30:43]

Resources:

How much water do data centers use? / David Mytton [8:10]
Making AI Less “Thirsty”: Uncovering and Addressing the Secret Water Footprint of AI Models / Back Market / UC Riverside & UT Arlington [9:28]
Jevon’s Paradox / Wikipedia [14:24]
The AI Iceberg / Ars Technica [22:54]
Simon Willison’s Blog about Dolly [25:37]
Breaking the code of silence: what we learned from content moderators at the landmark Berlin summit / Foxglove [28:03]
Holly Cummins from Red Hat’s Speech at QCon London 2023 [32:34]

Events:

Meetup on How to measure energy consumption of software (April 24, Virtual) / Green Coding Berlin [35:11]
Microsoft India’s Green Software Development Hackathon (March 21 – April 24, 2023 • Virtual): [36:13]
GreenTech Southwest Meetup (April 20, 6:00 pm - 8:00 pm • Bristol & Virtual): / Green Web Foundation [37:00]

Transcript Below:
Asim Hussain: I'm talking to people in my family, in fact, who are like thinking, will I have a job in two years time? Will I have a job in three years time? And like as historically, we have ignored in the just transition the other side have created a lot of very unpleasant noises, which has forced us to deal with that.

I think the same thing's gonna happen here. I think there's gonna be a lot of noises, and I would love for people to start really talking about how do we make that transition fairer.

I'm your host, Chris Adams.

Welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams of the Green Web Foundation. And in this episode we have some interesting news about how thirsty AI chatbots really are, and we uncover the hidden costs of generative AI.

And finally, we share some opportunities for development from the world of green software. Before we dive in though, let me introduce my guests and colleagues for this episode. Today we have Sara.

Sara Bergman: Hi. Glad to be back and in this format as well. So my name is Sara Bergman. I'm a software engineer at Microsoft. I also work with the standards working group here in the Green Software Foundation. I do some conference speaking. I'm actually speaking at conference later this year on this topic in particular. So it's very near and dear to my heart.

Chris Adams: Cool. Thank you, Sara, and Asim.

Asim Hussain: Hi, I'm Asim Hussain. I am the executive director and chairperson of the Green Software Foundation. And I also am director of Green Software at Intel. Excited to be here. Am I supposed to say an anecdote?

Chris Adams: This is when you normally talk about mushrooms Asim.

Asim Hussain: This

Chris Adams: Yeah.

Asim Hussain: I actually bought a robot lawnmower yesterday and it's currently mowing my lawn, so that's a weird anecdote. But anyway.

Chris Adams: That is quite a special anecdote. I don't quite know where to go from there, so I'm just gonna park it and then maybe the,

Asim Hussain: let's cut it out

Chris Adams: yeah, the.

Asim Hussain: robot. It's a really shitty robot. The every few minutes I'm going outside and pulling it out of a ditch. But anyway.

Chris Adams: Okay. And if you're new to me, my name is Chris Adams. I am the executive director of the Green Web Foundation, where we are working towards an entirely fossil free internet by 2030. So before we dive in, there's a reminder. Everything we talk about on this show will be shared as links in the show notes that are published.

So if something did catch your interest, please do follow the link to the podcast.greensoftware.foundation link, and you'll see all the links that we do actually have plus some extra commentary. Let's begin with our roundup of the news then. So this is the first story from the evening standard actually, AI chatbots, guzzle, enormous amounts of water study finds.

So we're starting this episode off with some news from a mainstream news source, the evening standard in London, in the UK. It's actually covering a report from the university of Colorado Riverside and the University of Texas Arlington, where some researchers are working to estimate the water consumption figure for AI chat models such as Google Bard, and ChatGPT.

The water consumption required to train advanced AI models such as Google Bard and ChatGPT is potentially staggering. With training GPT and Microsoft Data Centers requiring something in the region of 700,000 liters of clear freshwater according to our paper from these two universities. The operation used super computers with 10,000 graphics cards and over 285,000 cores.

And this study calls for AI model developers and data center operators to be a bit more transparent about the water usage, as well as suggesting some steps that people could actually do to make better use or incentivize people to use chatbots during cooler, more water efficient hours.

Asim Hussain: Hmm. I think in the parlance of, uh, World. The consumption of water in data is, no one was surprised. No one in the data center space went, oh, water. But it's actually a term, I think, is it? W There's a term called PUE for power. There's a term for water called WUE, am I saying that right? Water Utilization Effect?

Sara Bergman: Yep. Water Usage Effect.

Asim Hussain: And this has come up multiple. This is something I remember when I was at Microsoft. It was one of the, Sara, do they have a target for w I

Sara Bergman: Yeah, for 2030, replenishing more water than they used. But all hyperscalers do have targets for water usage. So it, as you say, it's definitely a known problem in the space.

Chris Adams: This is one of the next ones, so maybe I should ask, Asim. Why do data centers need so much water in the first place? Is it cuz it's when you initially think about it? Yeah.

Asim Hussain: Very thirsty ops People see ops people. They're so big. When the ops people walk around, they just need to guzzle a lot. I think it's for cooling. I think it's for cooling, just cooling service. When you actually look at kind of compute, the challenge of compute is the challenge of cooling and trying to get the maximum from efficiency as possible is all about kind of cooling chips.

So, I'm not sure the mechanism of cooling where it's probably just like typical HVAC systems and things like that, but it's just cooling aspects of it. And I think also from my understanding, it's not just the fact that it's usually, cuz you don't just wanna take put river water in your HVAC system. It's like good quality clean drinking water, which could go to everybody.

But it's also the fact the other end of it, you're pumping out hot water into natural like rivers and stuff like that. And I'm not a hundred percent whether this article or paper covers it, but there's like an impact to nature as well. Some of this stuff, I'm unsure. Maybe both of you have got an idea for this.

Like I'm unsure like why is this, why are we running out of water?

Chris Adams: So in the study there is actually a map mapping out water stress in various regions, and you'll often see overlaps between areas of water stress and data centers typically because data centers as critical infrastructure get to get the first bite of the cherry, as it were, just like with power. Last week, Aerin Booth was talking specifically about this, how when you are considered critical infrastructure, essentially priority goes to storing data about people rather than the people.

That was the phrase he used, which was quite memorable.

Sara Bergman: I think this is a fascinating topic, first and foremost, and also what you can use for cooling is essentially air cooling. So you can like pump in cool air interior data center and use that as a cooling. That works great in some parts of the world for the majority of the year. So where I live in the Nordic this is great. This works great. Other places like Arizona who has a lot of solar power, it does not work so great. Arizona is also water stressed. So you get these multiple climate factors on top of each other, which becomes a problem. And the second part, which makes water very interesting is that it's so incredibly localized.

If we think about our energy grid, that is also localized, but we can send electrons way further with way less waste or loss compared to water. We don't have the infrastructure today to send water over vast amount of stretches of land. We have very localized production of water. There is like even some places in the world where you look at, there are some islands around here in the Nordics, they have like incredible amounts of water facilities because it's just not centralized and cuz there hasn't been a point or they don't have the, a large body of water for which they could centralize, et cetera.

So that makes it an even more localized issue and even more connected to the communities who live in the immediate vicinity of the data center.

Chris Adams: Okay, so this is one of the issues. It's the locality, not just actually the wider thing. There's also a link from here from David Mytton who's a researcher in this field, and he's been one of the people contributing to a number of the Green Software Foundation projects. He's written a bit about how much water do data centers use.

It's really worth looking into because one of the immediate takeaways that you don't really see from this paper is while we are talking about the water used by the data centers themselves, the majority of the water in most cases right now is actually coming from the energy sources, not necessarily the data center themselves.

So you've got water being used to cool, say the generation, like a big fat thermal station, like a coal fire power station, or a nuclear station. So there's a chunk of the water usage there. Then there's the more localized water usage in the data center itself, where it's pulling it out of, say, an aquifer or a like subsurface store of water.

So there's two places to look at, and this is another one of the levers you might actually have. There are tools, as Sara mentioned about different kinds of cooling, like adiabatic cooling, which use ideas from almost hundreds and hundreds of years old, as well as actually the effect at the generation part.

And this is where moving away from burning fossil fuels, I'm gonna keep bringing that in, is a way to reduce the water impact. The paper here actually talks a little bit about having metrics so that you might get an idea of the embedded water in a model, just the same way you might have the embedded carbon in a model, for example.

So it's worth a look, and we'll share a link to the underlying paper as well as the original evening standard piece that was shared. Sara, I spoke a little bit about adiabatic, but you've shared a couple of links which look interesting as well here about the different kinds of cooling available in the world of data centers, which is not really my specialization.

Sara Bergman: I won't say it's my specialization either, but I think it's been some interesting conversations. So this has been a known problem for a long time, which means there have been a lot of people thinking actively about this, which is always a great thing to be in or a great situation to be in. And one of the things that have been experimentally used by several hyperscaler providers is what's called two-phase cooling, which is basically where you use a type of liquid, it's a chemical, not water based.

And then you use this to cool your server acts. It looks wild, but it's apparently very effective. But as of February, this year, the European, oh, let's see if I can find the full name for that. E C H A.

Asim Hussain: Yeah. European Chemical Agency, there's no H there. Where's the h?

Sara Bergman: Where's the chemical? Maybe, I don't know. Yeah. This is how you know we're not chemists in case you tune into the post calls wondering our profession. Yeah, so they, but I think also the American EPA, they both ruled or have concerns about the type of liquid that was used for this cooling, because it's what's called a forever chemical, which we have seen in other areas.

Its not great on, on the environment and the people who inhabit it, but because of that, 3M basically stopped producing these PFAS type of substances, which were used for cooling. And now people are saying that will likely slow down the process of innovation of this two phase cooling.

Asim Hussain: So what I understand about forever chemicals is that these are chemicals that once they're in our bio circle, they'll go into the oceans and the animals will eat them, and then we'll eat them and then we will pee and they'll go in the ocean and it'll carry on going and going. And so all we're eating is like the sludge of chemical and they're all, they're quite damaging.

That's, yeah, that's very scariest. Forever. Forever. Chemicals.

Chris Adams: These are things in like, non-stick frying pans. So every single one of us, we have a little bit of a non-stick frying pan inside us forever now. Thanks guys.

Asim Hussain: looking at my non stick fry pan, probably a lot of non-stick frying pan is inside me right now. The, so, um, Sara, so what I, one thing I've never really. Never really started. One of the questions I've always thought about liquid cooling, cuz I've seen the videos, they look really cool when you've got like a server and it's, it looks like it's boiling water and inside, but it's not.

It's, it's the special oil and chemical. The reason you put servers inside these liquid cooling things is so that you can put more electricity into the server and get more power out of it. So it will make the cooling more efficient. Maybe I'm talking about Jevons paradox type thing, but it'll actually mean there's greater amount of electricity going in. I suppose maybe the trade off is overall better.

Chris Adams: Yeah, these are typically used. You'll see liquid calling and liquid immersion cooling in particular. Is that simply because liquids are much, much more efficient at moving heat than air, right? So this is a really good way to get the heat you don't want somewhere else, and that's a tool that's been used in many cases.

However, it's often quite expensive upfront. A lot of us will usually default to be using air cooling in a lot of places. But yes, as the power density and racks increases, then people are reaching for more kinds of cooling, just like how cars used to be air cooled and now increasing your water cold. You have the same thing happening with increasingly industrial servers.

So yeah, this is something that we could definitely talk about and it's definitely one of the mechanisms that people do use to move some of the heat around. But once you've got that heat, you still need to figure out where it's gonna go and what you're gonna do with it.

Asim Hussain: So just to clear, so with immersion cooling right now, it would allow us to use less water because fundamentally it's more of an efficient mechanism of extracting heat. And so right now it would allow us to reduce the water, but we'll just be back at the same problem in the near future anyway, as the power density increases.

Sara Bergman: Possibly, I'm not sure, but it should also clarify that there are two kinds of liquid cooling so we have the two phase one, which is the forever chemical, and then there is the one phase one which uses a water based or an oil based. And it's, as I understood it, less effective, but it potentially still uses less water.

But yeah, it's always a scale problem, right? Because our industry has to grow. So as we come up with solutions, we outgrow the solutions as well.

Asim Hussain: Yeah. We're in a constant race to, to increase efficiency faster than consumption increases. Yeah. Yeah.

Chris Adams: Outrun Jevon's Paradox,

Asim Hussain: Out outrunning, Jevons Paradox is our challenge. Yeah.

Chris Adams: Okay. And for those who are not initiated, Jevons Paradox is this notion that as things become more efficient, the absolute use tends to increase. This was first noticed when William Stanley Jevons in hundreds of years ago, noticed that making steam engines led to an increase in the use of coal that was being burned because people end up using them in lots of new places.

Just the same way that making cloud can be more efficient. But if it makes more people use cloud, then we still have an increase in absolute usage. So this is one of the things that we are currently wrestling with as an industry basically. Should move on to the next story because it feels like this is a nice segue?

Okay, the next one up is the mounting human and environmental costs of generative AI. This is from Sasha Luccioni, one of the researchers at Hugging Face, but I believe also was working at MLA, a Montreal Institute for Gen. I think she's a colleague of, or was a former colleague of Abhishek, one of the people working in the Green Software Foundation Standards working group actually.

So this one here is a story from Ars Technica and over the last few months, the field of AI has been growing quite rapidly, as we know. And we see all these new waves of new models being used like Dall-E, which is specifically for generating images, or GPT four, which is, you may be familiar with if you've heard of the term ChatGPT every week brings a new promise of new and exciting models, but it's easy to get swept up in the waves of hype. And these capabilities, these shiny capabilities come at a real cost to society and planet. So this piece basically outlines some of the key areas that we need to be aware of. And one of them is actually the environmental toll of mining rare minerals to actually create the GPUs in the first place.

And there's also a bunch here about the training costs increasing over time, how these have been growing larger and larger, and I think the figures we were looking at was actually the economic accessibility of this. So the training cost of GPT three has listed around 5 million dollars, 175 billion parameters, which basically restrict who gets to create these models in the first place. That's the idea behind this. And this also talks a little bit about the climate emissions as well. Citing a study from Professor Emma Strubell, talking about the environmental emissions from an earlier model called Bert, b e r t, which had a figure of around 280 tons of emissions dedicated to the training of this one here.

There's also talking about some of the other larger models and how they're growing over time. So these figures, I think this one refers to things like GPT. The estimated figures would've been if, depending on what the energy source would've been, they were looking at around, assuming you're running this on coal and natural gas, like one of the typical things on the grid, the figures for the actual emissions associated with training are in the region of 500 metric tons of carbon emissions, which is what they've listed here.

Asim Hussain: Yeah, I just think there's a lot of stuff being said about AI. There's a lot of noise. There's a lot of noise about this, and I see a lot of people trying to play down the impacts of AI Both. I'm just gonna talk to sustainability. I think the societal spec, or which I would be happy to talk for hours and end about, but like for instance here, it was just mentioned that just the cost of training GPT-3 was $4.6 million.

That's one training run. How many times have people actually. Train that you tweak it, you run it again, that doesn't uncover inference. And this is also GPT-3, which I believe is an order of magnitude less complicated than GPT-4, which is an order of magnitude less complicated than GPT-5. And when you actually look at the way.

Like the business value of AI models, the emergent properties are coming through vastly more compute and vastly more data. This is just an indication of where we are right now. The future is like this nonlinear curve upwards of even more and more. So I think that's the thing I think to think about and that some of the things people have said to me have been around the lines of, even though it's this big, it's not as big as the airline industry.

It's not as big as other things. Whereas my answer to that is actually one of the reasons why we pay so much attention to software space is so few of us like a 2, 3, 4%, but it's relatively small number of people, the number of people who are involved in training things like GPT is minuscule. The number of people that need to be influenced to reduce those emissions, I think is very small, considering the impact they would have, which is why I would argue it's very worthwhile having strong sustainability conversations in this space. That's my rant on this whole topic. Yeah.

Sara Bergman: I agree. I plus one the rant, but also something that I think is interesting in this space because I really like to read research. And this is a topic that I've been speaking about for a few years now and what I see is there's conversation and research around the training cost and there's quite a lot there and it has been for quite like the paper you mentioned, Chris, that.

From 2019, it's a very well written paper, so I would still argue that people should go and read it. But this has obviously been talked about and I don't know if it's because it's easy to research on or because it's fun, but what I see less obvious inference cost, and when you have these huge production companies, I can only guess that the inference is much higher than the actual training, and any number of times, like a billion is gonna be a big number, but even more in the shadow than inference is data collection, where at least I cannot find any good research on what is the cost of this? How do people do this? And when we have these larger and larger models, well we need larger, larger datasets to feed them. And how do we gather this data and where does it come from? And I don't know, maybe we're awesome at reusing datasets here. I don't really know. But without research, it's hard to say.

So if someone's listening, have a great article, please send it my way. I'd be curious to read it.

Chris Adams: So we'll speak on the labor part for a second, cuz that's definitely one thing that's, that was highlighted in this essay. But Sara, you mentioned something quite interesting about where the data is coming from because there's actually a really interesting scenario that came up literally just this month. If you are using ChatGPT.

Try typing in the name Brian Hood. Ask who is Brian Hood inter ChatGPT right now. Tell me if you get something back.

Asim Hussain: Who is.

Chris Adams: Brian Hood. So b r i a n Hood?

Asim Hussain: It's taken me a long time. I think, mayor?

Chris Adams: Yep. Okay.

Asim Hussain: Okay. An Australian mayor prepares the world's first defamation case against ChatGPT is that what it is?

that what

Chris Adams: Yeah, this is exactly it. So this is a story. Basically, this gentleman in Australia essentially started a lawsuit against ChatGPT, cuz this gentleman was, he was a whistleblower in a significant financial, essentially a FIA financial fiasco back in 2010. And for a long time ChatGPT was basically listing him as the person who was the cause of this.

So, when you have models, which can lie and not tell the right thing, you had this baked into the model. So if you would ask this, it would basically say this person was, it would switch this stuff around. Cause we know that ChatGPT can be a little bit, not entirely truthy, basically, because it's very small auto complete, not actually really intelligence.

And this now brings up the question, how do you actually then retrain the entire model Now to get this person out, you can't just remove the line in a database because the model will break. So there's now a whole set of new research onto retraining and relearning because this person here has a pretty valid request of, Hey, can you please not be defaming me inside this?

There's a human right associated with this, specifically in Europe, for example. This is the kind of things that are currently, we are wrestling with right now. Have to manage either the retraining of this or to allow a sense of redress and accountability. When you are looking at this stuff.

Asim Hussain: Sara are the, well, you are the expert of the AI like cuz we just read that it takes $4.6 million to retrain GPT-3. I presume it doesn't take $4.6 million to edit out this Australian mayor, but it wouldn't be the same as like Google just de-indexing page

Sara Bergman: No.

Asim Hussain: I imagined be significantly higher. Yeah,

Sara Bergman: Yeah, because we have transparency into those systems, right? We don't really have transparency into what makes, yeah, it's like a brain. You don't go in there and just remove a line like, yeah.

Chris Adams: It's an interesting one to think about that, right? Because if we don't have this, I think the essay does refer to some of these ideas. It presents this idea of a kind of, AI iceberg is the kind of model that's shared inside this with things like the benefits generate really realistic images, do your homework for you, answer questions, and then there's all these other costs and things we need to take- think about like the raising barrier to entry, the actual emissions, the tons of carbon emissions. If it's a 50 tons of CO2 to remove someone's name from a 500 tonne model, then that's quite an expensive refresh cycle when you're developing. If we ever worry about continuous integration, this feels like it's in a different league, for example, that we might need to be somewhat aware of.

Asim Hussain: Or if you wanna delete your data, if you like go with a delete data request, how does it, that has to like factor all the way through. And then the thing is, like we, we've had Anne on. Me and Anne have spoken quite a few times about this idea that really these days it's about developer velocity. Like we have abstracted so many layers.

Like one of the big paradigm shifts in computing was adding like a level of programming abstraction onto hardware, which birthed like an entire explosion of software being written and applications being built because we just made it so much easier to build, arguably inefficient. We weren't focusing on efficiency, we were focusing on speed of developers and speed to getting stuff out to market.

And I'm seeing more and more stuff in the generative AI space, which is a similar idea, which is, it's not really about building efficient stuff, it's just about getting stuff out there as quickly as possible. What if you, instead of writing a really efficient API call, you just wrote like a ChatGPT prompt as an API request, and it gave you like 50,000 words and you stripped 49,000 on the way cuz you needed one. But that was just the quickest way to get a solution. How many times have we built software like that? Have you seen software like that? That's I think the danger of the world that we're heading in is cuz it's all just about that getting a valuable thing out on where does the ef- where our world is about efficiency.

Everything we talk about is about efficiency. So where's the efficiency in this? That's where I'm concerned at.

Sara Bergman: Yeah, no I agree. And I think I called the AI a brain and I already regret it because it is generative, right? It doesn't like, it reads the world and then present it in a slightly different format. I've seen people say, oh, AI has the same biases as us. Like, yeah, because we modeled it after ourselves. At least for now that's how it works. So that's another aspect I think is interesting.

Chris Adams: So this is one thing, we'll just touch on this and then move on to the next story. Have you folks been looking at their new stuff, coming with, I think Databricks published Dolly, which was an open source LLM, and they've also created their first entirely open Creative Commons training dataset. Specifically, they've got a bunch of their own staff to work to create this dataset where there's very clear provenance of where the data's coming in.

So unlike with some tools where you don't know where the data's come from, there's a lot of visibility on this, and there's a chap Simon Willison. He's been blogging all about how he's been able to run these things on his own local machines, but also there are people running these on Raspberry Pi's. Now, they're not very fast, but this is the direction that things are moving to where you see the open sourcing of this.

Because right now, while OpenAI is, we don't actually know where the data is coming from, what data is being used for this, which has all kinds of interesting issues when, like we just mentioned here as well.

Asim Hussain: It's not very open Is.

Chris Adams: We said we were gonna talk about that. There was also one other thing that was touched on here was this, the idea of when you're building an LLM and you're creating this training data, there's also an issue of like hidden labor inside this.

Uh, there are people who are actually paid to be working on this. And I think the piece by Sasha Luccioni refers to actually this time article about essentially Kenyan gig workers being paid less than $2 an hour to essentially examine all the messages for OpenAI. And this is very similar to the kind of content moderation stuff that you see in lots of other places.

I think this is actually really interesting cuz it raises these issues like what do we actually do? Or what can we done to address this issue of like unpaid laborers or underpaid people working in AI model development. Cuz that's not necessarily the people writing the code, but there's definitely people involved in doing that just like we have when other platforms, for example, too,

Asim Hussain: I think this is where things get very interesting with, and I'm gonna stray very far beyond sustainability right now, but we're talking about why are they getting paid so little is because it's unskilled labor, right? It's unskilled labor, and that's why you can get away with paying people $2 an hour.

When the AI is a de-skilling the workforce anyway, right? So the de-skilling the workforce and like I saw somebody post the day, like maybe the future of software developing is just prompting ChatGPT, which is an interesting idea, but I'm thinking like, how much are you gonna pay somebody to enter prompts into ChatGPT versus-

Chris Adams: As a prompt

Asim Hussain: engineer.

As a prompt engineer.

Chris Adams: So there's an issue of underpaid or very low paid labor involved in the AI model development that we don't really have really well addressed right now. And this is something that definitely needs to be addressed. There is actually one thing that may be worth looking at in other fields where you do see this, where you have the use of gig workers involved in this, like content moderation.

There's lots of parallels to this kind of work where you have people who are considered outside of a firm who have to do this work. And typically what we've seen there is actually honestly workers organizing and talking about their, the conditions they need to be part of. There's a link to Foxglove, the UK law firm, who are doing some really interesting work specifically about representing the power and representing essentially people working in these kind of moderation or below the API kind of roles because they are very clearly part of it.

And depending on when you look at the cost structure for a particular, say, organization, like if you look at, say, Facebook for example, Facebook might have made something the region of over a hundred billion of revenue last year. And this project, the actual moderation, which is one of the key parts, there is a single contract to a consulting firm for 500 million dollars for that part there, which is less than half percent of the revenue is being dedicated to this part here.

And that's an issue about, okay. How are you representing these costs when we're creating this and how do we actually make sure that there is a just and a correct way to actually recognize this kind of labor, especially to address some of these issues? Cuz it doesn't seem particularly fair to me right now to do this.

Sara Bergman: And I think it's also interesting because this is not the first time in history this is happening, right? Machines have been taking jobs from humans since the Industrial Revolution, but we want it to be for good. We want the machines. Whether that's a mechanical machine or a digital machine to take away the dangerous, the boring, the damaging jobs so that we as humans can focus on the fun, create a positive things of labor.

And here you have a paradox, right, where this is potentially improving the working conditions of one group of people, but forcing it for another group of people. And how do you represent that and how do we solve that? Yeah.

Asim Hussain: There's a lot of parallels here cuz we're talking about this in the energy transition as well. We're talking about just transition. We're talking about if we wanna move everybody over towards renewables, we have to think about the people that are in other industries because one of the reasons, cause they vote and they'll get quite upset if they lose their jobs and don't have an option.

So it's always been like this thing as towards just transition and I'd love for there to be conversations about that happening right now in the AI space. Cause I'm talking to people in my family in fact who are like thinking, will I have a job in two years time? I'll have a job in three years time. And like as historically, we have ignored in the just transition in the other side have created a lot of very unpleasant.

Which has forced us to deal with that. I think the same thing's gonna happen here. I think there's gonna be a lot of noises, and I would love for people to start really talking about how do we make that transition fairer?

Chris Adams: I think this is one that we need to dive into in a bit more detail in a future one. In the meantime, Foxglove is probably one organization that's doing some really interesting work in this field at the moment, and I believe there is some work coming from AI now in this field as well. Okay, so the next one is, Platform engineering, how platform engineering makes software sustainable.

So this story here comes from devops.com and the kind of thrust of this story is platform engineering seems to be essentially an idea of providing self-service tools for software teams themselves. So they are able to say, spin up an application or access, spin up a database. Ideally in our kind of green software world, spin up a cluster that turns itself off at the end of the day.

Because we do know that one of the big issues we have is that people just leave things on. Cause it's easier in terms of people's time to leave, loads of things running rather than turning them off. So this is the idea that by incorporating tools like Cube Green or some of the metrics tools, uh, I think Spotify had been doing, you can essentially set some more sustainable defaults for teams rather than relying on them to remember to do all this stuff themselves.

That's the general idea for some of this. Those are the benefits. Being proposed was, yeah, less friction. Reduction of cognitive load for developers. Basically consistency across the board. That's one of the things that seems to be pushed in this angle here.

Asim Hussain: Yeah, I'm seeing it's got two aspects to it from a sustainability perspective, and I like the way they phrased it. They're talking about observability and optimization. You know, observability is like, how do you. Having that visibility. Most people don't really know. They don't think we have. I think we should tell 'em that the secret Sara about how people use cloud?

Like most companies haven't got a clue like what's going on. And they're like sending out emails going, why are we spending $50,000 on this thing? And somebody goes, oh, I forgot to switch that off two years ago. That's like a level of visibility which exists right now. And so like any kind of like surfacing of that observability.

And when I was at Microsoft, I got that email like every, two weeks we hired someone just to email everybody, to tell them to switch things off cuz it was costing so much money to keep it running.

Sara Bergman: And actually at QCon in London last month, Holly Cummins from Red Hat, she actually had a whole talk on cloud zombies and she told a lot of these funny stories. So if you had the chance to catch that, she covered a lot of it because yeah, it's harder even with on-prem because then you might have a physical server.

That you don't even know where it's located in your buildings cuz you run a university and you have lots of buildings or whatever. And uh, yeah.

Asim Hussain: I have definitely remember, this is a long time ago and it was not Microsoft. So before you think I'm saying a Microsoft story, I remember. I was doing audits and then discovering machines that hadn't, we just completely forgot about, completely, utterly forgot about whole machines.

Chris Adams: So the company was making so much money that they didn't have to care, or they didn't realize that it

Asim Hussain: You can look through my LinkedIn and probably guess. Yeah.

Sara Bergman: but I think this is also interesting because if we look at data centers. We have seen on-prem data centers that are managed by one corporation for their own sort of benefit. They're typically way less effective than the hyperscaler cloud platforms, and that is because they can spend their money as in engineering resources to make them so efficient, because it's worthwhile the investment for them just in pure monetary cost because that's their primary business, whereas, Company running an On-Prem data center.

It's not their primary use case. They're maybe never gonna have enough people to do asset management in the most effective way possible. And it obviously depends on how you do it, but this internal developer platforms platform engineering situation, it could be the same thing, right? It all depends on how you do it, but potentially if you allocate a small group of people who are responsible and sustainability is one of their primary goals, then they have the incentive to spend the engineering capacity, and I'm saying engineering, but really any kind of practitioner that you are to make it more sustainable instead of having 10 different teams all have to fight for the same change. So it has potential, but it's possible to say the outcome without really getting into the nitty gritty.

Asim Hussain: I loved your explanation there. I'd never really heard it phrased that way, but yes, like Microsoft and Azure, the team, you're making money through Azure, so you're gonna make Azure more efficient. But if you are a consumer, you're making money by selling shoes or something.

Sara Bergman: Yeah.

Asim Hussain: Yeah, that makes a lot of sense.

Yeah.

Chris Adams: Alright let's talk about some of the events that we have listed inside this then. We have just a few meetups and events, which might catch your eyes of some people. These are all virtual events. So the first one is a meetup on measuring the energy consumption of software. This is some work by a group called Green Coding Berlin.

They're a small but mighty little outfit working on some very literally named tools. Like we can probably guess what the Green Metrics tool might do, right? As you can imagine. Does that, and they've got another tool called Eco CI, should we guess what that one does?

Asim Hussain: Continuous. I'm trying to think of something funny for I, but I can't think of anything. Integration.

Chris Adams: So, yeah, this is a small team of Berlin based developers who've been building a bunch of entirely open source tools specifically for this. The thing that I might share with you that might be interesting is that some of this work looks like it's probably gonna be fitting into the work with Wagtail at Jango based CMS on the Google Summer of code that we mentioned last week, where the goal is to actually start figuring out what the environmental impact of running various CI tools actually is, but also to measure where the, where the savings might actually possibly be inside open source tools. So you can pull out some of these patterns cuz we have things like patterns in the Green Software Foundation. Some of those might be applicable specifically for this.

There's a Microsoft Green software development hackathon from the 21st to the 24th of April.

Christ, that's this weekend, isn't it?

Asim Hussain: Oh wow.

Sara Bergman: Yeah.

Asim Hussain: I thought this was like the internal Microsoft Hackathon, but this looks like an external one. It looks like Microsoft India is having a green software development hackathon and Wow. 1,945 registrations, so good luck if you're applying. You've got a lot of stiff stuff. I got a lot of competition there.

I dunno what the prize is, it doesn't say.

Chris Adams: I don't see actually that much in the prizes listed here either, but I'll be honest folks, I dunno about you, but. I'm gonna sound like an old fart now, but like when I first got involved with this, they weren't called hackathons. They were called hack days. With the idea being that you'd like work on something, you're not trying to win a prize, you're doing it because actual hacking on something is interesting in the first place.

You figure out what can actually do rather than to win something to kick off your startup, basically, I. I'm a bit confused about this part, but uh, yeah, it looks like it's happening this weekend. Maybe if something comes out of it, we might feature it.

And then finally there's the Green Tech Southwest Meetup in Bristol of all places.

This one is run by Hannah Smith of the Green Web Foundation, and it's featuring two people. Adam Turner, who's the head of digital sustainability at DEFRA. That's the Department for Environment Food and Rural Affairs, who are the kind of defacto environmental department for the UK government.

Asim Hussain: Thought it was DARPA.

Chris Adams: Oh, not that's, that's America and that's a slightly

Asim Hussain: Oh, is that the me? Okay. Isn't that the weapons one? Any whatever.

Chris Adams: No, that's, yes.

The, there there is DARPA and darpa E. These are the folks. I'm out by death here. I'm not gonna pretend to, to know what DARPA is or what it, or what the significance would be for this one. I'm actually quite excited about this because this guy, Adam has actually been writing quite a lot about where the challenges are, and he's also been talking about, okay, these are the things that we're doing as a government with legally binding targets that we need to be meeting, and this is what our sustainability strategy looks like.

And the nice thing is that. If you are not sure about what to do, because the UK government publishes all their stuff, you can kinda copy their homework to get an idea of what you might apply internally for yourself, and they're actually got a really quite well-developed team and there's a lot of really interesting stuff to be looking at for both just on-prem and in cloud and things like that actually.

Asim Hussain: Mm, very interesting.

Chris Adams: We've spoken about the news and the events. I guess there's a question. We spoke a little bit about this generative AI iceberg and some of the things that should be on this list. Is there anything that isn't there that should be there? Maybe Sara, I might start with you. If there's anything that we should be putting into this AI iceberg as things that people should be aware of that will share a link in the podcast for.

Sara Bergman: Yeah, so one thing that I saw that wasn't there was lack of diversity in research. They said lack of interest. So maybe the same thing, but I think with the increasing cost, if really only the really large universities and or the really big corporation can participate in this sort of bleeding edge AI research, we're missing out on emerging research for people who probably have great ideas and couldn't do this incredibly well, but simply don't have the monetary resources.

Chris Adams: That's a really good point to mention here cuz when you see a lot of these papers, they tend not to be coming from that many different places so far, and this was something that was raised in an earlier discussion, an earlier paper, basically making the argument that, yeah, when you only get models coming from Western Europe and North America, you're probably gonna miss out on a bunch of extra context that we are currently missing.

In the same way that in lots of other places, when you only have modeling from certain parts of the world, you get an incomplete idea of what modeling the world might be looked like or what kind of policy options we do have to meet the climate challenges that face us.

Sara Bergman: Yeah, because talent is equally distributed opportunity's not.

Chris Adams: Indeed.

Asim Hussain: I love that.

Chris Adams: I think we'll leave on that one as a challenge from this. So that's all for this episode of The Week in Green Software. All the resources for this episode and papers and links will be available on the Green Software Foundation podcast website, which is podcast.greensoftware.foundation. So thanks again for listening and we'll see you on the next episode.

Thanks, Sara. Thanks Asim.

Asim Hussain: Thanks, Chris. Thanks. Thanks everyone.

Chris Adams: Bye.

Sara Bergman: Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation In any browser. Thanks again and see you in the next episode.

The Week in Green Software: Generative AI and Cloud Zombies

12 april 2023 | 26 min

Chris Adams is joined by cloud sustainability advocate and founder of Cloud Sustainably, Aerin Booth in this episode of TWiGS brought to you by the Green Software Foundation. On this episode Aerin provides his insight into the cloud to discuss cloud zombies, the effect that generative AI is having on the environment and exciting developments from Xbox (including a list of some of Aerin’s favourite nostalgic games!). We also touch on GreenOps and the future for green software developers.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Aerin Booth: LinkedIn / Podcast

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Generative AI and cloud zombies: Raising the alarm about global climate impact: / Silicon Angle [2:48]
Report on ChatGPT Model’s Emissions Offers Rare Glimpse of AI’s Climate Impacts: / Truthout.org [2:48]
Xbox’s New Energy Measurement Tools are World Changing: / Xbox [6:23]
Wagtail and the summer of code / GitHub [14:32]

Events:

Resources:

PS5 Power Consumption / ecoenergygeek [10:58]
The UK’s Digital Markets, Competition and Consumer Bill / Linklaters Sustainable Futures [20:16]
The Carbon Reduction Opportunity of Moving to Amazon Web Services / 451 Research [21:02]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:
Aerin Booth: When I think about anything we choose to do, not only in terms of carbon and IT, but in our life, if it doesn't have purpose, it's almost a waste. And we forget that we're not really building things for ourselves in technology. We're trying to build services for one, helps people in their day-to-day lives and hopefully save the freaking planet in the next upcoming climate change catastrophe the rest of our lives.

I'm your host, Chris Adams. Hello, and welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and in this episode we'll be discussing generative AI worrying impact that I could be having on the environment.

We'll also cover some exciting news from Xbox and some events for you to take part in as part of the world of green software. Before we dive in though, let me introduce my guest with us today. We have Aerin Booth the former head of Cloud at the UK Home Office now turned cloud sustainably advocate joining us.

Hi Aerin.

Aerin Booth: Hello. Nice to meet you. Thank you so much for having me on. So my name is Aerin Booth. I, like Chris said, former, I say in name in, in some ways product manager for Public Cloud at the UK Home Office. But while I was there, I signed 130 million pound contract, was part of negotiating the memorandum, understanding between the UK government and the cloud providers, hyperscalers, AWS, Azure and Google.

Um, and did a lot of other stuff for the cloud community across the UK. And then moved into my own consulting. So I've been running a consulting company for the last few years. Not to go too salesy or anything, but it's called Cloud Sustainably. Just helping people rethink how we consider carbon emissions in it, because it's on the rise.

And obviously we all know and care about this. We wouldn't be listening to this podcast otherwise. And yeah, I think more people talking about it does the world a little bit of good. And yeah, I've been on an interesting journey to say the least.

Chris Adams: Okay, thanks Aerin. So if you're new to this podcast, I am Chris Adams. I am the policy chair, sorry, I'm the chair of the policy working group in the Green Software Foundation and the executive director of the Green. Web Foundation. Every week we do this, we will basically share any of the links that we discuss and do a roundup of the news.

So that's generally of the plan. Today, it's gonna be a bit of a short one cuz it's Easter, so we're gonna keep it short and sweet. And I suppose Aerin, should we look at what stories have come up on our radars today? What's the first one here? There's one about generative ai and in particular, the environmental impact from generative AI and cloud zombies.

Aerin. I think given your background, with cloud, the Cloud zombies one might be an interesting one for you to start with actually.

Aerin Booth: It is really interesting. Yeah, cloud zombies, I mean it's, we could probably use any term we want really. I did a talk on being ethical in tech at Reve last year in November, and just before that I did a talk, I think it was titled DALL-E oh God, what was it called? Now? Is DALL-E Ethical? Something around like the use of ai?

What's the purpose?

Chris Adams: This is DALL-E, the AI model, not the former modern artist, right?

Aerin Booth: Yeah, yeah. There's a good comparison, isn't it? Names of such power and sort of representation. In terms of what were open AI going for, when they were creating this, and I see the power of AI. I use it myself, like when I'm on Instagram, on TikTok and all these things, and you see these, I don't use filters, but I do use like generative images.

I had one recently, which was really cool, but. When I think about anything we choose to do, not only in terms of carbon and IT, but in our life, if it doesn't have purpose, it's almost a waste. I saw a lot of apps on Twitter ever since I did that talk, and just keeping on what's going on in the world of AI stuff like, oh, change your hair cuts to these five different 10 different styles, and it cost like 30 quid.

Because it costs that much energy effectively. Like this article and some of the notes here are talking about the carbon emissions training a model. So what was it? Chat GPT three. It was 502 tons of carbon, which at the end of the day, what is that? A little bit in Norway for a little bit. But if we're all using this all of the time, and now what we're seeing is stuff like Bing and another search engines integrating, using this model as every single transaction on the internet.

And there's a lot of transactions on the internet. It's bigger than we realize.

Chris Adams: Yes. Okay. So that's the stuff that you get to use. I think this term, cloud zombies is particular being referred to this idea of basically, long running jobs and basically, essentially cloud stuff that no one is even using alone. Some of this stuff here, you could probably make an argument that most of us aren't necessarily asking for kind of a neat kind of AI features inside Bing when we're just doing a search.

But I think this one was specifically referring to this massive amount of waste from people generally not really turning things off or just it being easy to leave something on. Then, then to turn it off and face the consequences of things being turned off. I think that's where the phrase came from, and I wondered if this is something you might have something to speak to given some of the background with Cloud that you saw.

Aerin Booth: Definitely. I mean, you think about working at the Home Office, none of this is not public, by the way, that this is in ministerial statements in Parliament, but the percentage of AWS spend as terms of cloud and we had the definition of cloud is always hazy, a bit dreamy and up in the cloud, but we estimated it was about 94 to 97% of all of our spending was with AWS when we at the Home Office. And when you think about that, is that a problem? Yes. Now, but we started using the clouds in that department in 2016, and it's now 2023. Think about the journey cloud's been on since it was invented, invented, or whatever else we started using there.

It has gone through generations like we, we used to string together S3 and VPNs and build things ourselves. Then we have managed services, or Kubernetes came along in the middle, like all of these different generations of the cloud and a lot of enterprise organizations are now carrying all of these generations of the cloud and sometimes losing the skills.

As new people come in, they've not always got the history of skills or long-term experience. They've been taught whatever's being taught today, and a lot of the time it is new services. So it's like you've got two ends of the spectrum. Let's say AWS is a labyrinth designed to trap anyone in any decision they could ever make and fuck it up, and then charge you for the privilege.

And then a consultants in to say, oh, we'll help you with FinOps or whatever. Go calm down like you're an idiot by the way. When you need help from others, lemme charge you some money. And then you've got like the opposite end of the spectrum with GCP where in my view is let's just turn everything off every other day and it's okay.

That makes sense. If things can be migrated easily. But if you are asking a company to every single year, start again. Whatever they're building on the newest platform, okay, they're keeping up with the skills, but what about developing new features instead just re-platforming because you are whim on what you are going to turn off and there's two ends to the spectrum.

And Microsoft, I wish I could say they're doing better, but honestly, the horror of active directory and stringing together some of this stuff when it's supposed to just work. It's pretty gnarly. Like you, you wouldn't wanna be going in there as a startup these days. I'd recommend something like Digital Ocean or some of a new cloud SMEs, particularly for cloud.

And yeah, here we are, next generation of cloud computing. It already exists and a lot of people don't see it because we stare at AWS, Microsoft, and TCP all the time.

Chris Adams: I see. Okay, so this is one thing. So there, so it, so I guess one of the questions I should probably ask you here is, given that there was a significant amount here and uh, we're to. Talking about being able to switch things off. Basically. In your experience, how did people manage to keep track of which things were running and which things were not running and these kind of experiences?

Cuz you need to have people to have some understanding of which things are, where you're able to scale things back. In a lot of cases, and in many cases one of the kind of ideas behind switching to something like cloud, which might be more efficient, is that it's supposed to be much easier to manage and it's easy and there, there are supposed to be some co cost savings for this surely, which would result in energy savings?

Aerin Booth: Yeah. Uh, well, Let's say energy savings, for example. So this is something I've been finding very frustrating. So I started caring about sustainability in the cloud, let's say 18 months ago. It was probably like November, ReInvent, two years ago. Yeah, about 18 months ago now. So I've not been doing this for a long time, but I've been in tech for a long, longer time, 10 years or so, or at least doing my own stuff.

And it's not so much spend, cuz spend comes as a result of you doing an action. Having something that is created to have a purpose like spend is secondary to whatever is you are building as a team or whatever features are coming down the pipeline. And a lot of the time, because the pace we all move at in terms of new features, new releases, management direction, all of this sort of stuff, we're on the hamster wheel.

Basically just saying, okay, I've gotta add this new feature. Or the tails wagging the dog. You do user research, get two conflicting pieces of advice, and you're throwing on a new feature rather than making whatever's there better. So you're always moving forward. You're never really stopping to deal with tecta, you're never really stopping to keep up on platform.

Restructures best decisions here and. I wouldn't say it's by design, but no one makes it easy. No one is really saying, okay, here's what you need to do. How about you try this? Like Amazon, you'll get hundreds and hundreds of emails about all sorts of stuff just randomly announced to every single account, and you pay your ID as an enterprise.

Not very helpful, especially when you've got account teams who will help, but they've got their own sales targets as well. Because at the end of the day, the cloud is all about selling like it's rent-tier capitalism of technology. Like previously, you could own your own data center, manage it like even, and a lot of those can be cheaper for the right size organization.

Just the cloud doesn't always make sense. Let's say, because one of the things we've always talked about is like a cloud first policy, but the cloud doesn't just mean AWS. Like I'm saying, Azure and G C P, there are loads of different cloud options and even SaaS services managed services. These are all in the cloud.

IAS, PAS, SaaS and whatever else, that's how the cloud's moved on. A managed service from AWS is a SaaS service in a different form, and people just get a little bit mixed up sometimes and it advice for anyone. Just stop and think a bit more in terms of what you wanna achieve before you just start building stuff.

And I dunno, products management and delivery management are the two key parts. Like doing them well is about shaping the team in the right direction when we're building stuff. And that's where people forget. They just follow the rules and don't understand why you should be doing these things. But anyway, quite off topic for the original question.

So,

Chris Adams: Alright, let's park that one there and come back to a little bit later cuz there's an event later on with the head of Digital Sustainability, who's speaking at Green Tech Southwest. And we can touch on some of that a little bit later. Next story we see here, I see one about some of the new ideas in Xbox.

Xbox has some new energy management tools, which they're basically using to, by the sounds of things, reduce environmental impact from gaming specifically.

One of the key things was this idea that by essentially optimizing Fortnite, the game, people have been able to identify something, the region of 18 megawatt hours a day of power, and at the same amount as basically an entire wind farm in Sweden, and basically remove that by making some optimizations to the actual game itself by removing, say, excess use, which people aren't using, for example, or removing some of the really expensive computing when it can't be perceived so easily.

This is one thing. Didn't realize that. This is actually quite interesting in my view, cuz A, we forget just what the impact of gaming might be when you have all these machines, which are about maybe half a kilowatt of power, or hundreds of watts, for example, but also the scope for actual optimization here.

Aerin Booth: Yeah, and, and I really like what Microsoft's doing to be honest, in terms of Xbox and the direction for their technology because they clearly have a vision that's further ahead than other providers, I'd say in some ways, especially when it comes to sustainability and technology and just even connect. I remember Microsoft Xbox Connect, like that was pretty cool back then and it just got dropped off slightly and I think PlayStation's a bit head in in VR with the new VR 2 coming out, but, I like to think about this again as a generational problem of gaming.

Gaming is one of those industries which just derises it a lot in the media and people assume like hardly anyone's a gamer. I probably bet now especially was listened to the podcast. You game a lot more than you realize. Like people do it on Candy Crush, do it online betting. It's all gaming. It's all gaming theory.

It's all basically around, okay, what we getting out of this? What's the purpose of this game? It either gives you fun or, or takes money off you, pick your poison. And I've got an Xbox, I've also got a PlayStation and Xbox is doing some really interesting stuff. I think one thing I might have read, maybe it's not in this article, but it's like just doing updates and downloads at a period where the carbon intensity of your energy grid in the country that you live is low.

So scheduling overnight or whatever it is, cuz it, it doesn't follow the sun per se. The way we consume energy, especially when renewables are coming in the day and it varies, especially Europe way. We're all connected. So it's, I don't need to care about that. I don't need to worry, go around the house and unplug things and do all these smart home setups if it's baked into the technology that we're delivering.

And Microsoft obviously just deciding to do that, whereas PlayStation, and especially, I'm gonna say rather than PlayStation, Activision, blizzard, like Activision and COD. Have you seen the size of some of these games? GTA 5. They are like almost terabyte games these days. Can you think like they're always getting updated?

People are going up and down, and especially when you almost have to do a fresh install somehow, or some reason I've had to do it once or twice over the years, and that was really unnecessary. And as much as we had like games flying around disks years ago, and everyone's saying do digital now, it's like with a game, I can keep that passive to a friend, take it back to a shop and get a refund from CEX or a computer exchange store.

Whereas now if I get a digital download, what I'm getting is a license to rent something off you in the future and continually have to ask for permission to download it, which you may ban me one day who the hell knows. So it's okay, we're solving the technology carbon element and now we Microsoft's going, but didn't we just do that anywhere with disks?

Recycle your disks. It's mostly plastic and glass and cases and whatever, and yeah. Do you know, there's an interesting fact and, and Corey Quinn mentions this one, the fastest way to transfer data around the world is on a hard drive on a plane. That's the quickest way to send data halfway around the world because the fiber network, speed of light is the limitation and there's only so many open routes.

So it's like, okay, we can still ship more data in the world, but it's just like, yeah, put it on the back of a plane, fly it round. It's, that's what we used to do with disks anyway, so it's, thank you Microsoft kind of thinking about this, but what's the point of all this and like these days, especially with ai, do games have to get any bigger, like I'm pretty.

Most games I love Legend of Zelda, Wind Waker, it's cell shaded. It's timeless because of its style, not because of it's trying to chase realism of the day. Cuz we always get better at doing realism. So we always date a game by being realistic.

Chris Adams: So this is actually a nice segue because one of the ideas for this story here was people are basically talking about getting between 9 to 16 watts per user of savings per player, basically, which I know sounds okay. And then you gotta think about how does that relate to, say, the power usage of say, maybe a PS 5 or Xbox.

The numbers that we, I just did a bit of Googling for this beforehand, and we can see some numbers for like when a game is in full use. So with a PS5, the numbers. We see from say, okay, I'll be honest, this is pretty short. So the citation required from ecoenergygeek.com, PS5 power consumption gaming figures.

You're looking at around 200 watts with outgoing, up to 350 watts of power. Now, when you compare that to say, min Nintendo Switch, Which is basically, let's say you're using something full-time gaming. You're looking at maybe between 6 and 12 Watts of power usage here. Now this is something which is 10, 20 times, and I, you've gotta ask yourself, is it really 10, 20 times better the experience for this when you have this kind of trade off here? This, there's one thing that we're not really so aware of when we are looking at the gaming we might actually use here, because the savings we're seeing here are basically the entire usage of some other smaller devices like you just mentioned.

Aerin Booth: For me, I've got Steam Deck for example. They have Steam do gaming and like, you know, they've built community, they've built the steam store. Like, you know, that in itself revolutionized gaming. Like the games died off for probably creative differences and gotten bored and all fighting rather than playing together, but, What they created is a storage.

Very good. The steam deck and the innovation and hardware is really good, so like they add a bit of a dodgy controller, but I love the steam deck. Not only can I stream games to it, so I can have it in my house, connect to my wifi and stream Microsoft Cloud straight to it. I can remote play to my PS5 to it.

I can emulate other games that are legal to emulate if I own copies elsewhere, blah, blah, blah. Disclaimer here. Yeah, it's a brilliant device and to be honest, I've not looked into the energy consumption so much, but gaming, we shouldn't worry about gaming if I'm a gamer. You shouldn't sit here and worry about our energy consumption of, okay, I've left my TV on and I've had a game on all night.

I remember when I was a kid, I didn't have a memory card for Final Fantasy VII. I had to play with my PlayStation and never switch it off and say to my mother, never turn this device off because we don't own a memory cars and you can't save games. So I'm like trying to play Final Fantasy VII and Crash Bandicoot with like never stopping it, which is quite an interesting one.

Not a good attempt. I tried playing Crash Bandicoot recently as the remasters and oh my God, that game is frustrating. I can't believe I even bothered as a kid. So at the end of the day, gaming is about connecting people World of Warcraft. Best game for me in my life in a lot of ways because of the people and the connections I made.

I have four godchildren because I met a friend in a guild. She married the guild leader. They had kids. They asked me to be the godfather. I went to another wedding, second marriage of hers. We've been lifelong friends ever since. I met her when I was 13, playing World of Warcraft as a old blue warlock, a female warlock, which is quite interesting.

That's all my online personas have ever been women. Which is now, I look back at it, I go, that's really interesting. But yeah, it's good fun and like, yeah, it brings communities together it's healing and Tetris great example. If you play Tetris after a serious accident or incident, whether that's a stress ambulance, blah, blah, blah, it reduces PTSD because you give your brain something to do while you're trying to process all those thoughts, and it actually helps you not get them stuck and can do with your hands and spend more time processing at the same time and literally prescribe someone 30 minutes on a Game Boy of playing Tetris or on the phone after an accident or after whatever else. And it'll really help them in the future. So gaming is not until worry about when it comes to carbon.

I, I think, What is to worry about is just attitudes from gaming companies around their impacts and how they run things, and that's their choices rather than the console manufacturers, some licensing deals here and there, but yeah, it's open Wild West out there these days.

Chris Adams: So don't blame the gamer. Blame the. Way

Aerin Booth: Activision Blizzard,

Chris Adams: Oh,

Aerin Booth: I, no, I find it interesting. Activision Blizzard because in all honesty, There's been uproar online around the culture of that organization. It was started by a big group of men, basically provided over quite a horrible culture for a long time, over decades. They had very good stories like Green Jesus, we all love them.

For all rest in peace, you're back from the dead, who knows? And you know it, it lost its path slightly. But what I see with the newest expansion is it because they really did a change of their culture? They started to actually focus on, on, on being more loving in the environment. The game itself is much more interesting now.

You have queer characters and dragons and all this stuff. A lot of people are really happy about it these days and it's like this. The thing about boycotting anything, if you boycott something. Like how is anyone supposed to get better? Okay. Tell them what they did wrong, accept them. Make changes like removing people, which a lot of these companies have been doing.

So removing these people who were bad for the environment and then that's it. Like I know especially, and I'm not saying this to people who were harmed directly, I can never. I can't comment on that. That's your own stuff and your own opinions. You can choose to just never interact with them again. But to say to half the world or most of the world, you have to do this way or you're a bad person.

What's that gonna get us More fucking sad people in the world.

Chris Adams: Yes. Okay. All right. Should we jump onto the next story?

All right. Next story we have here is wagtail and the summer of code. This is quite an interesting one in my view. Basically, Wagtail is a CMS, just like WordPress is a content management system used by significant part of the internet. Wagtail's used by companies like say, a number of charities, Google, lots of well-known blocks are actually running on this, and this project is about the Google Summer of Code. There was a joint project. Basically start embedding some greener coding practices into Wagtail itself. So the idea here is to do things like introduce some kind of green modes or also think about, okay, ways that you can create a different architectures to make this scale down to zero in various places.

This one actually, I think, is. I have to say I, I am somewhat involved in this because this is a joint collaboration between Torch Box an agency and the Green Web Foundation. We've been doing some work for this, but there's a number of really promising directions for this to go in, and this was using some of the tools from an organization called Green Code, green Coding Metrics, largely because when you do use some tools like say a cms, it's not obvious where to make these optimizations.

And if you're making able to have something open for people to start implementing some of these pattern, The idea is that you can possibly adopt these in other places, so I might ask you actually. Yeah. Any reckons at your end on this one Aerin?

Aerin Booth: so it's really great in terms of anything that is up the scale. Wagtail is effectively a content management system same as WordPress. I understand it in terms of that. That's what I've ever heard of. I might take a look afterwards. I had a little look then, but yeah. Great. Do carbon reductions at a platform level and anything hack around it.

Like this is the great thing about open source projects and whatever else, like if you solve an open source project. As in terms of it has this capability now and anyone can contribute to, if you solve the problem, then we can come back like that as a team, as everyone in the community to say, okay, we care about this.

Now we're gonna show it some love and make it better. That's the whole focal point about community-driven development, which is open source communities. This is the sort of thing which we could almost stop and do anyway. And you see this at Kubernetes, especially these days. I think Adrian Cockcroft did a good talk at QCon just talking around obviously stuff he learned at Netflix, but you know what we need to do with Kubernetes?

Effectively, Kubernetes is a zombie of its own. It was open source to encourage people to be able to migrate between clouds or just have a more generic platform layer. You put a box and a docker, whatever to cube and it goes there and we can move trade cubes around. But the reality is it helped people get onto the cloud quickly.

Okay, I can write all these things, but at the end of the day, there's always services that are connected different ways. So it's like I've got my Kubernetes cluster with all my nodes and whatever else. And then I'm relying on, uh, SSL toys. Ssl, isn't it? Like all of these things which are either cloud providers you host from AWS or you buy a SaaS service and you've then think it as chemical reaction.

You're slow as the weakest reaction or whatever else, or bond is as weak as the weakest chain link. And that's where a lot of the time when I see outages in the cloud and have experiences them, it's never really the whole of Amazon going down though that has happened. It just happen a lot more regularly these days due to thermal events.

That's why we care about carbon emissions. A lot of data centers can't actually handle variations in temperature than designed for that, but that's where it actually goes wrong. Like you don't even notice it as well. And I think you asked me something earlier, which I didn't finish off, and I'll come back to it now.

Enterprise organizations, and I'm talking generically here because I know it's gonna be a problem everywhere because to be honest, they all run in very similar ways. And the government, civil service, massive organizations have worked in all the biggest ones. Ministry Defense, DWP, Home Office, HMRC.

They're very similar. We don't even know what we build. Like enterprises, generally things will get built. Innovations, money will come in cuz this is it. Money arrives a decision maker and it something gets built underneath. So when you're an enterprise organization, you've got all this money flowing out through cash cent-, cost centers, whatever you call them, bloody accountancy things and stuff just starts getting built, which is fine, but people are building their own fiefdoms.

People move on. Things get passed around. I would bet any money that people who've gone to use Service now, especially service now because it's very self directed. You have to do a lot of manual work, and I think they've got better things these days, but again, it's hard to connect everything across the generations.

I would be very surprised to find many organizations that have a very full record of every single IT service, it's service name, it's in service Now. All of it's onboarding for live service and operations because things just fall through the cracks and they just exist. They are zombies themselves, and you have a service that just works.

Even look at the internet itself, there's so many open source projects that sit on underlying all of our core open projects that I think there was one a couple of years ago where, I dunno if it was protest or it just went down, but it caused half the internet to fall over. Yeah, we're all building crumbling towers when we go too big in terms of tech and it, especially enterprise, because sometimes it's like, oh, you know, you're a bit of an old man now Microsoft, can you really do everything in its dog like, especially when startups are just doing it better and faster Digital Ocean, Genesis Cloud, Leaf Cloud, all of these places in Europe.

Chris Adams: Okay. Alright, we'll pick that thread when we come back to this. Let's look through the last few stories that we have here. Cause we're just coming to the end of it from here. There's a few events coming up there. Linux Foundation Energy Summit in June in Paris. There's a few people speaking specifically about tools like Carbon Aware SDKs, and if you're in Paris, it seems like it's worth going.

I know that I'm actually going along to see some of the talks, cuz it looks like one of the most interesting places to. Essentially find out what's happening at this intersection. One layer down where we work with at the internet, for example. There's that there. There's also a upcoming event is Green Tech Southwest at the meetup on the 20th.

This looks interesting cause so if nothing else, you've got Adam Turner who is the head of digital sustainability at DEFRA talking, basically providing a bit of a way in for people who are new to the idea about. Apply sustainability to the digital sector, but he'll also be talking about the UK government's digital sustainability strategy themselves.

This I think, is actually quite interesting cuz this is one of the UK's probably further along than are a number of other governments right now and they're at least very quite public on this. I might ask if there's anything that you wanna add onto this one here, cuz we're coming up to the last few minutes for you Aerin?

Aerin Booth: Yes. I'll take it over for a couple minutes and just kind of add my views in here. So I'm a former civil servant. I was independent, impartial. Following the government, blah, blah, blah, blah, blah. Always did my job to the most first honour and respect of that role. I was part of a lot of these negotiations and sustainability was always a touchy subject.

Like it's a very difficult thing to get anything straight out of a lot of companies, because we're all playing with assumptions and numbers right now with the reality of it, like Scope 3 hardly exists in all the cloud providers, especially AWS, they don't show Scope 3 at all in any of their online tools or any of their reports.

They have an awful report. 451 did this report about the carbon reductions you can make by migrate into the cloud and they say you can have 88% energy savings in the cloud if you migrate to the cloud, and therefore you'll have carbon reduction. Okay. I said something very specific there. 88% energy reduction if your migrates to the cloud.

If you read the report itself only covers scope two carbon emissions. It doesn't cover scope one, which I thought was quite interesting considering there's so many diesel backup generators in all of these places. If you think about Puerto Rico, for example, when they had the hurricane several years ago, the only thing that didn't lose power on those on the island was the data centers.

The people of that country waited, what, nine months, year, 18 months to get power restored to everyone. Data centers never lost today because we prioritize data of people over people themselves. That's a crazy thing that's going on here. We're like we shipping these boxes of ones and zeros. Rather than thinking maybe I should do something better with my energy on this island and help people out for a little bit and take the loss and turn off some hard drives, put it all in disc or tape storage and turn the damn thing off for a bit.

And there's so much we really need to think differently about because yeah, that report, that's all over 80 versus sustainability pages, it's literally the top link. I went to an event in Ireland, I flew over for Amazon's first ever sustainability event organized by the wonderful public sector and marketing team over in Ireland.

And they basically said that report, and then they also said, oh, look at our wonderful carbon reporting tool, which by the way, everyone only reports Scope one and two. Scope one and two is about 7% of carbon emissions from AWS. 93% of all their carbon emissions comes from their own supply chain. So when I'm making a decision about my cloud and I'm looking at these wonderful graphs, I am seeing 7% of the a hundred percent of the big picture and thinking I'm making an 88% carbon reduction.

So what effectively Amazon has just said. You can reduce your carbon emissions by 6% if you move to the cloud. Sound a bit different now when you really analyze the facts and read the report more than two pages or get through their first blue blog post and sustainability for me cuz I've worked with Amazon, I'm a community builder at Amazon I don't really care about not pulling punches or whatever else because I've honestly tried to work with them quite a lot and their PR team consistently always pulls the plug on sustainability conversations. And I'm not even joking. I was supposed to do a Twitter space or someone invited me and I got pulled, I'm not saying it was me, I just say they have a problem talking about sustainability.

They have no idea how to do it. Cause they think. Oh, if we admit this is a problem, everyone's just gonna run away and not join. It's like maybe, but you can't just carry on what you're doing just because you want to make some profit at the end of the world while the rest of us like have to deal with the climate crisis.

And yeah, I almost not given up on Amazon, but I won't be helping them directly anymore. I consult generally probably between like a bit like Corey Quinn. 10 to 15 different people or teams who reach out and ask for my advice at Amazon and all of it's unpaid, like I'm giving unpaid labor to one of the world's richest companies for no thanks or credit.

Not to say thanks. I was like, oh, I deserve this fin. But to be blocked or just have events disappear randomly because of PR decisions, when all I've ever been trying to do is help is like, well, now the time to just do things differently, which is why I've got my own podcast. Why I come onto podcasts like this.

I do obviously talk about sustainability with my rights hat on AWS channels when I do get on. And yeah, we need to think differently about the cloud. And for me, me and Adrian Cockcroft have been talking about this for a long time, and it'll be coming out soon and I might as well throw it at the very, very end.

I've been recently asked to propose a book to the British Computing Society on Green Operations, and I'll be writing that hopefully with some co-collaborators and figure out how can we really reframe this as a cultural issue. Okay. We are DevOps because even with DevOps, what we prioritize development and developers, we're thinking about ourselves when we're building services, try thinking about the people and the planet.

Chris Adams: So this is the dev suss ops thing is.

Aerin Booth: Not DevSusOps, that was Adrian's term and we're not gonna use that. In some ways it has its own purpose, but again, I don't want people thinking about developers or su sus like it. It just doesn't even make sense. Like what you read that I have no Green Operations is about thinking people and the planet when we're building services.

If you put those two things at the top of your priority list, okay, what is my priority? It's for people on the planet, not users. By the way, people. And then you go, okay, how do I build this? I'm gonna build it with diversity and accessibility at the front of my mindset and open source and reusability. And if honest to God, you just do those things.

Think about people on the planet, build accessibility needs and usable parts and, and sharing and SaaS services. People just want to use it because it is actually a good service. You don't have to just build everything from scratch. You just need to think about things differently, and it's always chasing the tail or the money or the next feature, whatever else.

And we forget that we're not really building things for ourselves in technology. We're trying to build services. For one, helps people in their day-to-day lives and hopefully save the freaking planet. In the next upcoming climate change catastrophe the rest of our lives. So green operations hopefully be published by the British Computing Society once I submit the full manuscript later this year.

And yeah, it'll be a good little thing to sort of, you know, think differently when it comes to sustainability in the cloud.

Chris Adams: Okay, so green operations, you heard it here first. I think that takes us up to the time we have. This actually Aerin. I'm gonna say thank you very much for joining us for this episode of this week in Green Software

Aerin Booth: Can I just add one more thing? I always love doing this at the end of my podcast. One I is just in terms of, you wanna find me? My name's Aerin Booth, but my online persona is Aerin Clouds, A E R I N C L O U D S. That's on most social media. It's really easy to find quite a unique name, but. What I always try and say to people is like, if you're listening to this podcast, if you finish it and made it this far, thank you so much and well done.

But do something nice in the next week. Take some time off you. Were gonna do, spend some time in nature. Don't forget that there's other things out there, rather than staring into a box on a screen and working for cloud companies or technology companies. When you know you shouldn't really be traveling away from everyone all the time, you shouldn't always be on the go.

We need to balance these things out. And that's Green operations. If you stand in nature, you're gonna think about nature a lot more when you're making decisions, if you're always away from it, if you're in a city, you're on the tube, getting back home, sitting down, having a takeaway, it doesn't really cross your mind.

And sometimes we just need to have a little bit of fresh air and it really helps us just to do some of this stuff. I'm basically a digital hippie. So let's go and let's do this together. Peace and love.

Chris Adams: So the secret to Green Ops is to get out in the green. All right. That's a nice point to end on. That's all we have time for. All the resources and links will be added to this. If you have any feedback, go to greensoftware.foundation in your browser. Uh, And if you did enjoy the show, please consider leaving a review on Spotify or Apple Podcasts or wherever you get your podcast.

Your feedback is valuable and helps us reach a wider audience. So thanks again. Thanks for listening and seeing you in the next episode. See you next week, Torah. Take

care, Aerin. Bye.

Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

Thanks again and see you in the next episode.

The Week in Green Software: Netflix, Refurbishment and Anti-Greenwashing Laws

5 april 2023 | 44 min

On this episode of The Week in Green Software, Chris Adams and Asim Hussain discuss the latest research on streaming emissions from Netflix and DIMPACT, the environmental impact of refurbished tech from Back Market, The European Commission's Right to Repair Law and their proposal for an Anti Greenwashing Law which is being echoed across the channel with the UK’s Digital Markets, Competition and Consumer Bill. Asim also discovers an alternative to central heating with his hot TV! The usual exciting resources and events in the show notes from TWiGS, Environment Variables and the Green Software Foundation.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Asim Hussain: LinkedIn / Twitter

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

What the Latest Research on Streaming Emissions Tells Us / Netflix [2:39]
The environmental impact of refurbished tech / Back Market [13:21]
The European Commission's Right to Repair Law / European Commission [17:58]
Microsoft wants to export 'grid-interactive' Dublin datacenter setup / Microsoft [23:31]
The European Commission's Anti Greenwashing Law Proposal / European Commission [36:20]
The UK’s Digital Markets, Competition and Consumer Bill / Linklaters Sustainable Futures [38:06]

Resources:

Table that Chris refers to from ADEME [14:19]
ADEME Study on the Impact of Refurbished Electronics / ADEME [14:19]
The Restart Project Podcast [20:27]
US Right to Repair Law from 2022 / Reuters [21:21]
Balancing Power Systems With Datacenters Using a Virtual Interconnector Balancing / Conor Kelly [24:31]
Ecovisor / Ecovisor [26:21]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:

Chris Adams: Turning bauxite into aluminum is incredibly energy intensive. It's in terms of density of load versus the area used. The only thing that is greater than it is data centers.

Asim Hussain: Oh, alright.

Chris Adams: Yeah. Or maybe Bitcoin mining, but you can, they probably count as a data center as well. But basically, yeah, incredibly dense load, which is why you see this, and this really spelled out to me just how big a player some of these large companies are now.

I'm your host, Chris Adams. Welcome to The Week in Green Software, or TWiGS, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and I'm joined here with Asim today. Hey, Asim.

Asim Hussain: Hiya

Chris Adams: Oh, see, before we go further, we should probably introduce ourselves, shouldn't we?

Asim Hussain: yeah. Sounds good.

Chris Adams: Okay, so my name's Chris. I am the policy working group co-chair in the Green Software Foundation, and I'm also the executive director of the Green Web Foundation. I'm also an organizer at climateaction.tech, and I think that's enough things for me to explain what I do. Asim, I'll hand over to you next.

Asim Hussain: Yep. Hi name's Asim Hussain I'm the executive director and chairperson of the Green Software Foundation. I'm also a director of Green software at Intel and uh, ex-organizer at climaaction.tech, which is where me and Chris, we didn't meet then, but we first started working closely together.

Chris Adams: Where we started working together. We met at OMG Climate, another conference based in London and Berlin, which was a successor to OMGDPR, which is a conference all about GDPR, which was considered quite an earth shattering thing to be thinking about back in 2018.

Asim Hussain: I definitely shattered the earth a little bit, didn't it?

Chris Adams: Yeah, a little bit. It definitely had an impact.

So today, if you're familiar, if you're not familiar with the format for this show, we generally run through some of the stories that have come up and share some of the commentary and sometimes we invite guests on to talk about some of these. I think we've got a few stories here from link, stuff from Netflix, the environmental impact of refurbished tech from Back Market, and some interesting news from the policy point of view with both the European Commission and stuff going down in the UK as well.

Should we start Asim?

Asim Hussain: Let's start with Netflix, should we?

Chris Adams: Yes, what the latest research on streaming emissions tells us. So this is a piece published by Netflix and in collaboration with the folks at DIMPACT, which I think stands for Digital Impact. So this has a number of real kind of heavyweights in the field. Dr. Daniel Schien, Dr. Jonathan Koomey, Jens Malmodin and a number of other companies are sharing data along this like BT, Orange, TalkTalk, Spotify and Netflix themselves. Uh, they did a bunch of literature review about what the state of the art is for this, and they used some of this to work out some of their own figures themselves. Asim, I think you've had a look at this as well.

What caught your own when you saw this?

Asim Hussain: Well, what caught my eye was I've been hearing about DIMPACT. Is it DIMPACT or D-IMPACT?

Chris Adams: I'm not quite sure

Asim Hussain: DIMPACT for a while and Daniel Schien is Dr. Daniel Schien is a member of the foundation's, been active member of the standards working group amongst other things for a while. So we've been talking for a while and just really, and yeah, it's.

Think the papers as well as the Litera literary Review does it in such a way where actually creates recommendations or what's it call 'em? Principles. Yeah. Which is a good way to go. It gives very direct feedback advice to everybody else for what to look at. See, they've got four principles. Should we dig into them?

Chris Adams: Yeah, you can run through. These are the ones for the government. So as I understand this is basically the company saying, Okay, government, this is what you need to do so we can do our reporting properly. That's largely it. I think with the idea being that one organization is saying it's too big for any one of us to solve by ourselves, so you're gonna have to have government involved for this part here.

That's the argument they're making, at least as I read through it. And yeah. Should we run through these cuz there's four and they're relatively catchy. Do you wanna start with number.

Asim Hussain: yeah. Principle one, expand access to shared contemporary data that is no more than one to two years old, and which does not compromise competitive and proprietary information, which is interesting cuz that's actually oftentimes the feedback I hear from organizations regarding being more transparent with around data.

It's that worry that you are going to be revealing yeah, competitive and proprietary information. And I don't know, I will say now I'm gonna be opinionated. I would say it's not even with the greatest understanding that you will, it's the worry that you might, because if you're an exec inside an organization, it's far easier.

You're not gonna have a huge mistake by saying no to revealing some data. But the worry is that you are gonna say yes to something and then something will get revealed.

Chris Adams: And then you publish your cost structure. Is that it?

Asim Hussain: Accidentally. You don't realize if you divide it by five, it's your cost structure. Like you just don't realize that.

Chris Adams: Fun factoid. 10 years ago I worked at a company called AMEE, Avoid Mass Extinction Engine. One of the ideas behind some of the carbon accounting was to work with organizations because if you are upstream and you have organizations sharing their carbon emissions to you, It does indeed give you some idea of what the cost structure is likely to be, and that gives you an idea of who you should be speaking to first in terms of trying to achieve some carbon reduction.

So on one level there is this idea about a cost structure thing, but there's also this idea that if you have deeper supply chain engagement, then there's greater chance to have some mutually beneficial up collaboration there. And there were was one example of a very large soft drink manufacturer then working with their suppliers and they would basically say, A huge chunk of our supply chain emissions is from you guys having old fridges.

So why don't we actually just agree with you to get so you buy better fridges. Cause it's gonna make us look better and you look better so you can you please change the fridges? That's literally how it worked and this is why I think it's quite interesting because it works both ways and uh, there is a kind of mindset shift that may be necessary for this.

Anyway, we've got three more principles to look at here cuz this is quite exciting. Okay, next.

Asim Hussain: You do the next

Chris Adams: one.

Yeah. Okay. Ensure appropriate modeling for decision making. This is through continued research to avoid oversimplified and biased results. I think this is actually a reference to the fact that in many cases there will be models which you say let's you looking at something like streaming, for example.

This one here, if you look at the research, Netflix basically say, okay, for what we do, and we're looking at about 1% of the emissions come from the data center, 10% from the network, and nearly 90% come from the device manufacturers like at the end, which is like your tablet or your big screen, or your router, or like the wifi on your house, for example.

Or maybe a TV or digital video box, whatever settop box you might actually have. And that's a different modeling from what you might see in other services. So if you have something which is entirely web based, where they're not doing so much streaming of video, then you might have a different setup because each request has a lot more work going on.

You're not sharing the same thing. Cuz the whole point about things like Netflix in many cases is that everyone gets to see the same video, right? It wouldn't work if every single video was different, right? That would make for a pretty ropey shared experience.

Asim Hussain: are they? But I think that's also speaking to the fact that, and I think there has been some pretty simplified, they call it oversimplified. I would also say there is a need for a simplified models here as well, but there are like simplified models. I don't wanna name a name, any names here, but there are simplified models that people have used for networking in the past, which have come under some criticism.

And I can imagine if you're Netflix, those models would overestimate your emissions just, I know you just mentioned about the end user divide, but there just overestimate your networking emissions.

Chris Adams: It's gonna make you look awful.

Asim Hussain: for Yeah. Whereas the reality is, I think that's, what're talking about basically use good models.

Chris Adams: Yeah, I think the, so the thing that's worth the thing,

Asim Hussain: Revolutionary statement there.

Chris Adams: The useful statement to reach for this is that all models are wrong, but some models are useful. So depending on what your resolution might be, it might be useful to use a quite high level model, say sustainable web design for Web design. But if you tried to use something like the sustainable web design model for Netflix might not give you particularly useful answers.

So that's the second principle. Okay, number three, go on. This one is you. I think.

Asim Hussain: Principle three. I love principles. Principle three, institutes energy efficiency requirements for devices and infrastructure, TVs, data centers, internet networks, home devices, et cetera, energy efficiency requirements. So it's like an energy star aspect.

Chris Adams: Yeah, it does sound a bit like that, and I think this is, I think this is interesting for two reasons. So first of all, various bodies like the GSMA, I think, or GESI, which is G E S I, and I think the site's best Targets Institute. They've issued a press release in maybe 2020, basically saying if we want to hit two degrees of warming, we need to basically half the energy consumption of the entire sector, um, by 2030, and that's about a 7% reduction year on year.

So that's what you have there. And uh, we don't have any kind of regulation for this kind of stuff. And in many cases there's a cadence for which when new bits of technology come in to allow you to scale some of this stuff. But back, for example, so like routers for example, when we look at this helpful diagram from Netflix, you can see a significant chunk is actually stuff which is on the subscriber's premises. So that's like your router in your house. And because they are always on and there's no real kind of sleep process, there is basically no relation there. And when that's taking up, maybe say a quarter of the impact they see here, that's like a, that's a space where you really would really would be helpful to have some kind of agreements on this.

The good news is there is actually some of this described in the next, uh, draft of WiFi. I'll have to share a link with this to allow things to scale down.

Asim Hussain: Oh really? That work is now actually becoming part of the standards.

Chris Adams: It's a really early draft, so there I've seen some discussions where people were saying, yeah, maybe we should have a way to scale some of this back when we're not push sending data over the wire.

You see the same thing with the deep connections as well. For some of the really fast connections, what you essentially have is when you have nothing being sent over the wire, to make sure that the system is very responsive. When data does come down the wire, you basically have messages saying, I've got nothing to send.

I've got nothing to send. I've got nothing to send. So you're still sending stuff even when you're not sending stuff. So again, it's because people haven't prioritized energy usage for this. So there's a bunch of scope here. So yeah, that's a fun one to talk about.

Asim Hussain: To drive that work, there needs to be kind of requirements for devices to be energy efficient. There's also a software, I just wanna be clear, like I don't think this is purely a hardware story. This is a software story. A lot of that story you just described it, uh, it could be implemented in the hardware, but it could also be software components to that standard as well.

And there's a lot of stuff about switching devices off like this. This stuff is a software driven aspect of it. And there's also like your TV is oftentimes taking in a stream of data and un-decoding it and putting it onto the display. That's a software. We're in that world where you actually, like the boundary between software and hardware is blurred because you could actually like implement a lot of the stuff as a hardware aspect of it.

But I think there's stuff there as well. You can implement energy efficiency stuff in in software as well. And just another random, my TV's hot every now and again. I go close to it with my hand and I'm like, wow, this is actually radiating a fair amount of heat. So I think that's something to think through.

But anyway.

Chris Adams: Yeah,

Asim Hussain: don't

Chris Adams: so,

Asim Hussain: of your TV as a radiator, but mine acts like a radiator for my house.

Chris Adams: So you heard it here first. If you want to reduce the environmental impact from your Netflix habits, use a smaller screen or turn on the big hot television that's attached to Asim's wall.

Asim Hussain: Watch more Netflix shows in winter and reduce your energy

Chris Adams: bills.

Yeah, there could be.

Asim Hussain: in summer. Get out in summer and enjoy the sunshine and stop watching Netflix in summer is basically what- that's advice from Netflix apparently, according to the Netflix report.

Chris Adams: Should we move to principle four? Assume before we get in trouble. Okay. Prioritize broad availability of low carbon and renewable energy for companies that operate large scale infrastructure and consumers since most streaming emissions come from inside the. Basically what they're saying is you need an entirely fossil free internet, which is what my organization cares about more than anything else in the world.

Yes.

Asim Hussain: A fossil free internet broad of low carbon renew- and that's so important. I mean, when we talk about the pathways to achieving the only viable pathway to achieving the goals of 2030, very fast decarbonization of the electricity grid is pretty much the only path these days, isn't it?

I don't know if anybody talks seriously about another one.

Chris Adams: There is a bunch of really useful work that's been published by Transition Zero and Ember Climate about this. We can share links to this, but let's not dive in too deeply because we end up being an energy podcast rather than an a Green software podcast. All right, so next one, the environmental impact of refurbished tech.

This is a story from Back Market and uh, I think this is quite interesting. I've shared this Asim because Back Market, yes, they're in the business of selling refurb technology. So if you wanna buy an iPhone or an iPad or something, rather than buying it directly from apple.com, you can buy a refurb one from them instead.

So,

Asim Hussain: I've actually never heard of them before, so it's really, yeah.

Chris Adams: Yeah, they're available in quite a few countries, but they're not the only people. They work to look, work a little like a kind of two-sided market. So they speak to smaller shops that do some of the refurbishing and they connect buyers to this. So they're like right in the middle. And I've ended up getting into habit of basically purchasing most of my electronics through them now simply because there's a really interesting set of stats and research on this, which basically shows that the carbon savings are really quite substantial. So this isn't just like a puff piece. This is actually work from the French from ADEME, which I'm not gonna try to pronounce in French, but more or less translates as the French environment and energy management agency.

They published a study last year to work out just how much of a impact having refurbished or circular electronics might actually be. And, uh, I've shared the table here from this report, from the actual link that's in here. And generally speaking, say, let's say you're gonna buy a secondhand smartphone and hold onto for two years versus hold onto a, bring a new one and having it for three years, there's every single saving here in terms of CO2 or water or e-waste, it's above 80 to 90%.

Asim Hussain: So it's just so I understand this number correctly. I'm seeing a table, and for one cell it's saying smartphones, and then the other

Chris Adams: Tablets and laptops and desktops. Yeah.

Asim Hussain: from a smartphone carbon perspective, I'm seeing the percentage, 91.6%. So what I believe that's telling me is, I will save 91.6% of the emissions. Why wouldn't it be a hundred percent?

It's a refurbished or something a little bit confusing in my mind.

Chris Adams: There's still an environmental impact from taking something and refurbishing and doing some work on it.

Asim Hussain: right, right. Okay. Yeah. So if I was to keep that, why'd they say two years and three years? That's a,

Chris Adams: I think the reason this is because that's what they typically look at when people purchase a new machine. They might not hold onto it for so long. There's actually a link to this really detailed report from ADEME, the French agency, where they go into do this like 200 pages long, where they talk about every single possible scenario of one year ownership versus five year ownership.

Are you buying a secondhand one that's two years old versus five years old? It's like every permutation you could imagine. But the high level stats here are basically the key takeaway is that buying something refurbished is a really effective consumer way to reduce the environmental impact of the electronics you have.

Or conversely, how guilty you feel about your gadgets habit, basically.

Asim Hussain: and I think it also speaks to the, I won't say fallacy, I will say the misunderstanding of the effectiveness of recycling. because recycling you really will not get anywhere close. I don't know what the statistic is, but it will be in nowhere close to 91.6% for a smartphone. Reselling something or just giving it to somebody else rather than throwing it away is one of the best things to do.

It's actually, I still, that's why I love the, I've gotta really gotta check out black market. Sorry, I just called it, I just, I called it black market cause that's what I first read when I saw it. It's the black market, but it's back market

Chris Adams: As in take back or give back or refur-. Yeah.

Asim Hussain: it black market accident many times from this point forward, but I think that's a really important aspect.

Cause I think one of the challenges people have for refurbished stuff is a lack of trust.

Chris Adams: Mm-hmm.

Asim Hussain: I buy stuff from eBay all the time, but I have it with this. Is it gonna, what's the mo- what I'm gonna get? Whereas getting some form of kind of trust out of interest do they offer guarantee or something when you buy it from Back Market

Chris Adams: they do. And where are we of turning this into a advert for them? Because we, cuz basical. I purchased all these. I think I've pretty much got my phone, my an iPad and my work laptop through this one company. They're not the only company doing this. In the UK there's an organization called Tech Buyer.

There's plenty of other ones, but I've been pretty happy with this and I've got a year warranty from this and basically because. Computers have, their speed is not improving at the rate they used to be. It's okay. So I've got a machine from 2020 that I bought secondhand and it's banging. It's really good.

It's, I'm really happy with that. And that's enough to last be for the next few years.

Asim Hussain: you just need a Web, what is it? 60, 70% of all apps is just Web. You're just browsing on the website. We've solved that. You've got a laptop that's good enough for that, and it'll be good enough for it for a long time.

Chris Adams: And it's just as well, because there's changes in the law coming down in particular, this is the next story that is somewhat related to this. One of the reasons that has been a problem with getting these things like devices lasting longer is that if a single thing broke it, you basically didn't have an easy way to get it repaired or replaced or anything like that.

And there's been some interesting new laws with a new right to repair law, which will require hardware makers to provide fixes for up to 10 years. From new, new electronics. So that's so much further for this, uh, basically the, it's still being drafted to some, when it comes to actually being implemented in different countries.

So just because it's plaster, either European, European level, you still need someone to implement the interface, as it were in Germany or France or stuff like that. But generally speaking, yeah, things like, uh, hardware, cell phones, tablets. The goal is to have things at least five years and up to 10 years of there, and also providing clear access to all the bits that you might need to repair these.

Asim Hussain: This is kind of like, I think iFixit. Is it, are they a non-profit or a for-profit? I can't remember, but this is, this is kinda what iFixit has been trying to do, which is figure out how to repair things. When manufacturers have provided no information about how to repair it, and they're just like smashing up 50 iPhones just to figure out how to repair an iPhone, and then they're publishing the information online.

But this is saying, this is actually gonna have to be law. Oh, it looks like there's still some negotiation that needs to be done here, but it'll be law for organizations to affect even beyond the guarantee period. It same between for five to 10 years. Five to 10 years. You have to make something, somethings repairable. That's amazing.

Chris Adams: I know it's pretty cool, right?

Asim Hussain: This is, this is how the world used to be. The world used to be you. Anything. I don't feel old. I do not feel that old, but I do remember repairing things. Repairing things like a normal electric thing was a normal thing to do. You went to repair shops and you got things repaired and you brought them home.

And these days it's almost impossible to repair anything. And the times when you do go and try and repair something, the cost of repair is so much higher. Not higher, just almost the point where you like another 20% more, I can buy something new and then you get that world. Whereas I love the idea of whole cottage industry of repairing things has almost been lost.

And now that generative AI is gonna take the rest of our jobs, maybe we can just, maybe our jobs will be, maybe the only jobs will be left will be repairing the machines, which the AIs need to survive. Maybe we'll be in service of the AIs to repair the machines for them. Okay. There we go.

Chris Adams: Topical. I like it.

Asim Hussain: we go.

Chris Adams: Okay. This is the thing. It's, we'll see how it shapes through. Before. For folks who are more interested in this, there's a really nice podcast from the Restart Project because last month there was London Repair Week where there's a bunch of interviews with people who actually are doing repairs of electronics and talk about how it's changed or what some of the trends might be.

We should share a link to that cause it's quite fun. It's quite a nice kind of uplifting and happy story. When usually a lot of things around climate and technology can be a little bit hard work. All right.

Asim Hussain: This is interesting. Do you think that there's? As Moore's Law and all this other stuff, it's not slowing down. I don't know. There's other things that are happening. The law's not that simple, but as technology moves, is the fast paced nature of technology, the thing that's made things harder to repair and now that maybe technology will move slower, or you just mentioned your laptop's gonna, it gives more room, more breathing space for people to try and repair things.

Chris Adams: I don't think it's that. I think it's more a case of business models. Right. So even one of the things you did see, cuz in America recently passed a recent right to repair law themselves and organizations which have been moving, who have been pushing back against this one of the tactics. So if. Basically said as well, okay, what we're gonna do, we care about the sustainability of things, but you can only ever return things to us.

So therefore we, we are gonna capture the entire value chain ourselves rather than share it with anyone else. And that's very much a deliberate decision that some people have made to say we are gonna be, and if we get devices from anyone else, we're gonna either withhold the parts, which means you can't create a whole kind of secondary market around this.

You'll see things like that. And I think this is one of the issues. It's very much a case of. How people designed things. Because even if you look at, say a hardware point of view, there are examples like the framework Laptop in America, which is essentially a laptop designed to be the opposite of say, an Apple MacBook, which honestly I own cuz I'm stuck using Apple devices.

But basically this is, this is the thing.

Asim Hussain: You're in the ecosystem and that's how it's designed.

Chris Adams: Exactly. So these are, this is very much a function of the, in my view, the business model for this. So even if I wanted to have some things which were not the same, the fact that I am locked in using the same operating system means that I'm not able to do that. And that kind of integration isn't really addressed with this right now.

And I feel like in many cases it's case of which feudal landlord do you want? Do you want Microsoft? Do you want Google or do you want Apple? You can do everything yourself, but then you're open and vulnerable against all the bandits and everything like that. But then you have to realize, except that, yeah, maybe the feudal landlord has shareholders to return to, and the priorities are making sure those guys are happy rather than you are necessary happy.

So that's some of the things you have to worry about really.

Asim Hussain: Well, I think it's interesting. I think there's parallels here between the kind of the closed source and open source ecosystems as well. It's like those are huge, like what you just described about you can only return products to us is a closed source system. I think there's definitely cases where the open source model has been more successful than the closed source model. I'd also probably argue this case where the closed source model has been more successful than opensource model as well

Chris Adams: I don't think we are gonna resolve this one on this call. We see not this one. I think there's plenty going either way for a bunch of this. Should we look at the next story of this one here?

Asim Hussain: go on. Yeah.

Chris Adams: Okay. All right. So Microsoft wants to export grid interactive Dublin data center setup. So this is a story that I think is interesting from the perspective of kind of green software developers because it's worked by Microsoft who have basically working with a power management specialist called Eaton to build kind of grid interactive UPSs into their data centers. So this basically means that rather than just having backup energy, which just sits there doing nothing, the idea is, is that the backup battery could supply it into the grid if necessary.

So this basically allows you to kind of smooth out, say spikes in demand or stuff like that. And you can see it as a kind of compliment to renewable energy sources being somewhat variable and at times intermittent. And what they're doing is they, this is about them saying, we've got this setup. One, we're gonna do it in lots of other places, and we're gonna start with Ireland because Ireland only gets maybe 35% of their power from renewable sources, but there is a really aggressive growth in data centers plant in Ireland, or that we've seen over the last few years. So this is actually quite an interesting one. I think that the person that might have been related to some of this work a chap called Conor Kelly, he published a paper about this idea of balancing power systems or data centers.

I think Asim this is my, uh, you used to work at Microsoft. I reckon you might have some reckons on this.

Asim Hussain: Yeah, I know Conor. In fact, we should probably reach out to him, see if he wants to come on the podcast. Conor, if you're listening as you as of course you always listen to our podcast. Welcome to, to come on. Yeah. Conor's, he's been in this, I wouldn't be surprised this, this looks like it's got Conor's fingerprints all over it.

Yeah, so I think for, for my understanding, so just so I'm gonna break it down for everybody, so. Data centers have a lot of backup power supplies. Sometimes actually diesel, but they sometimes actually are battery powered, sometimes actually hydrogen batteries, all, all sorts of stuff. But this is a battery powered backup.

And the point here is can you reverse the energy back out onto the grid? And therefore be like effectively like a short term battery act as a battery for the grid and therefore make money. Which I think is a really not make, necessarily make, maybe, probably do or not make money, but I'm sure they have other relationships with the utility providers like a, I dunno, reduced fee or something like that.

That's interesting. And I think it's also interesting cause I remember I was talking to somebody ages ago in this space about carbon aware computing and I was talking about shifting compute. And they raised the point, well if all of our data centers have batteries, like why not just shift the energy? Cuz if you shift compute, it's like if you take the opposite side, it's like you're shifting energy.

So we've got batteries, shift the energy. And I was like, huh, that's actually quite a great idea. Actually just shift the energy from France to, well, wherever.

Chris Adams: So this is the work that I think is most interestingly demonstrated by the work on some, I think Ecovisor is the name of this kind of series of projects where the idea was that rather than just having a rather software just knowing that there's power coming in, software has some idea about what quality of power that might be saying, this is the grid power and this is the carbon intensity of this. Or here's renewable energy that, that you might have attached to a data center. And this is what the carbon intensity of this is. And this is how much battery power is available and how many hours of battery is available.

And also what the kind of intensity, carbon intensity of that might be. Cuz if you were to charge that battery when there's loads of wind on the grid, then you've got really green energy, which in many ways may be greener than the grid power that you have. So if you wanted to say optimize for the greenest possible power, you might choose to only run on say, battery when the grid I is particularly high. If you are able to figure out where you are pulling the power in from. If you're saying, don't feed me grid power, but please feed me power from the battery and from onsite solar onsite renewables, then you are able to control the actual power going into the system.

And there is a, the thing that's really cool, I did a talk about this cuz I was so excited about it when I discovered it at FOSDEM. This idea of kind of virtualizing and different kinds of power, I think is one thing that if I wasn't doing what I was doing, I would try to build a service and build a company around it basically.

Asim Hussain: Virtualizing. Describe what you mean.

Chris Adams: So, you know, we have virtualized compute, pay for compute, storage, and network. Yeah, it's a big physical machine, but it's exposed to you in the form of a computer, which is just the right size. Just has enough memory. Just has enough of this right. Now if you know that, say your computing job don't need to happen right now, they're not time sensitive. Then you could say, okay, I only wanna be fed on variable power, for example, renewable power. I don't need it to happen, come from the grid because I value that it's greener and it's cheap more than it being available all the time and dispatchable. And I think that people who actually have batteries inside data centers, I think people will figure out how to turn that into a product that you can sell as make available inside this.

Because I think that's a kind of value added thing that you could quite easily add to cloud compute to just say, by the extra green stuff inside it, which you know for sure has been come from the power that stored at certain times. Yeah. You could segment now the power that way.

Asim Hussain: Cause I've always wondered about that, about battery, cuz if you're a wind farm. When I sell electricity, I can then sell like a rec and I, that's like how you signal that my electricity was green,

Chris Adams: Mm-hmm.

Asim Hussain: but if I have a battery on the grid, I'm currently, I know the grid is 50% green, 50% fossil, and I'm storing the energy, like the, like I know that electricity is like half green, but no one else would know.

That's what I mean by virtualized. So if I was to then send that back out onto the grid, I could then give. like a half-rec here somewhere, or, I dunno how that would work actually.

Chris Adams: You just don't decouple the greenness from the power. That's a whole kind of silly market construct that only happened because for historical reasons, right? If you actually just treat the power like it really is, yeah, then you could totally do stuff like this. And that is the premise behind this Ecovisor concept, which I think is super exciting and even has an API to implement.

Asim Hussain: Oh, eco. Yeah.

Chris Adams: Yeah, exactly. If you virtualize the compute, why don't you virtualize the power? And that allows you to purchase things differently because I think there are people who would basically say, okay, I'm using a bunch of cloud services. If there's a way for me to just purchase a kind of greener quality of power from this for certain parts than I would.

Cuz that allows you some more tools as a designer of services, for example, if you know that like then maybe I can pay for say, eight hours of definitely green power, for example. Then I can redesign the rest of my time to either, I can redesign my compute to work within that budget, or I can say, I know I don't even have to see what I mean.

Yeah. You could do a bunch. You could come up with all these new system designs. Yeah.

Asim Hussain: completely not thinking about recs in the slightest

Chris Adams: Yeah.

Asim Hussain: just, thinking about the whole, just like blank. Yeah. Interesting.

Chris Adams: This is also why the discussions and if you look on the kind of, you look at the trends for grids, there's a huge amount of battery storage looking to be connected to the grid, both in the UK and in America. For example, masses. And it's growing so fast. So I think.

Asim Hussain: As in it's there, but it's not physically connected to the grid. What do you mean? Or do you mean?

Chris Adams: It's being connected or people who, cause you need to apply for permits for a bunch of this stuff. It's basically being permitted and being fit into the grid. So traditionally you might have had relied on say things like open cycle gas turbines, which have really this stuff. You as batteries come into this, that means that the kind of marginal intensity was you you don't have the same signals, basically, so you can't be sure that it's gonna be open cycle gas turbines that are powering that marginal power now. So a lot of the assumptions we make about marginal intensity may not be the same, which is why many cases open some of this up. And if you just look at the location based amount, then I think it's actually an exciting new horizon opening or opening up to us. And I reckon Conor's probably got something to say about this, cuz I, yeah, this paper I read, I thought, this sounds super cool cuz it basically uses data centers like CDNs, but for power essentially in the same way that CDN relieves pressure on network by rather than you congesting commonly used channels with lots of the same things being sent over, you're just getting it from somewhere nearby. So you can think of transmission in the same way. If you have a way to reduce the need to fetch power from somewhere else, by getting it from somewhere nearby, then you've essentially taken this idea of a CDN, but you've applied it to grid services.

And I think that is actually, there are so many places, there's so much overlap in this stuff. There really is, and. I think more people should be discussing this cuz cloud is utility. Once you start thinking about these things as utilities, then yeah, all these ideas which have been developed for decades in other fields become applicable in our world as well.

Asim Hussain: I'll just say one thing, and I think we were talking about gang Adrian Cockcroft in, but I listened to his talk at QCon last week and one really interesting statistic he said, which I thought was fascinating. But if you add up all the power from all the major clouds, it actually becomes one of the top 10 energy utilities in the world.

Chris Adams: Well, it's not just that there's a crazy figure I saw recently. So if you look at, say, which companies have been purchasing all the kind of corporate green energy, power purchase agreements in Europe, right? Amazon is responsible for 19.9% of all the PPAs, the capacity. In the last 10 years, so 20 to 13, Google is 7.4% and that's like the next two largest organizations are Alcoa, which makes aluminium and Norsk Hydro, which is basically, it's really eye-opening seeing these numbers.

So these are the stats from Wind Europe, and I will find a link for this. I didn't realize that nearly 30% of all of the PPAs, the power purchase agreements for renewable power has come from big tech firms.

Asim Hussain: Which is just to be clear, you're not saying that's 30% of all energy or 30% of all. Energy, but 30% of all like this, these what these things called power, which we won't go into, which is still a significant amount just to go to some tech company perhaps how belittling of me just to go through a couple of tech companies.

But you know, aluminum I've always heard is like quite significant.

Chris Adams: Yeah. Turning bauxite into aluminum is incredibly energy intensive. It's in terms of density of load versus the area used. The only thing that is greater than it is data centers.

Asim Hussain: Oh, alright.

Chris Adams: Yeah, or maybe Bitcoin mining, but you can, they probably count as a data center as well. But basically, yeah, incredibly dense load, which is why you see this, and this really spelled out to me just how big a player some of these large companies are now in the kind of shifting of the grid and how that might impact what we do as developers and people building digital services.

Asim Hussain: why like things like grid, what are we calling it? I've forgotten the term Grid

Chris Adams: Interactive is what they.

Asim Hussain: Yeah, grid interactive batteries. Yeah. That's why it's so important. Yeah. That's why it could potentially could be quite

Chris Adams: significant.

You say potentially one of the reasons you can get the idea that people are doing this just outta the goodness of their own heart, right? And the linked story basically says they're not just doing it out of the goodness of their own heart. They're doing it because it makes financial sense a lot of the time.

So when you are a larger organization, you do a big power purchase agreement like this, you're gonna get power way cheaper than other people because you're buying in so much bulk. So yeah, you get to say that you are green and everything like that, but

Asim Hussain: I gurantee if you are creating utility scale kind of grid interactive batteries, you are getting a better deal. There is definitely a financial, and I wanna say like one of the things that's quite surprising to me actually was to find out that the interrelation between gas and renewables, which is an unfortunate kind of temporary, in the decarbonization of the grid, you do need to be able to create electricity very quickly when the wind dies down.

And currently for a lot of places that's now gas. Whereas that's why kind of battery solutions are so important, cuz we don't want that to be gas. We want that to be non fossil solutions.

Chris Adams: Yeah, basically if you can find a way to avoid burning fossil fuels to quickly respond to changes on the grid, you need to generate power quickly or reduce power quickly, and this is what some of this stuff makes possible, essentially. So you can either reduce demand from data centers or feed power in instead of the gas.

Yes.

Asim Hussain: know how, how, how much we try every single episode to not be a podcast about the energy system. We end up being a podcast talking about the energy system.

Chris Adams: We're talking about distributed systems in my view, and like the internet and the grid are, there's lots of similarities between these two things. So anyway, we can move away cuz we're talking about the idea that if you are prepared to be flexible on this, then you can get paid quite a lot of money for this.

And that doesn't mean like it's okay for you to be doing this stuff. But if you say that you're only doing this for the goodness of your own heart, there may be changes in the law that mean that you're not really allowed to say that now. And this is some of the new stories. There's next story actually.

Asim Hussain: The EU Commission's, anti-green washing law proposal.

Chris Adams: Yeah, so this was published, I think end of March, was the draft version of this, which basically says that, uh, later on this year, pretty much all the, there'll be moratorium on new kind of certification schemes, if anything, marking something as green. And also you'll only be allowed to use a certain set of really explicit, like the greenhouse gas protocol and stuff like that.

They're gonna say every single claim has to match up to this stuff here. So there's gonna be some really much more stringent stuff, and there'll be like injunctive things saying that if you don't, we can basically force you with the full force of the law to stop you talking about power being green, for example, or things being presented or things being ocean friendly, for example.

Asim Hussain: Is that driving things more to certain well defined terms? Like academic terms? If you say the word when you say you're a hundred percent powered by renewables, I'm like, okay, let's break that down for a second. What do you mean?

Chris Adams: Yeah,

Asim Hussain: You know, like what does that

Chris Adams: so the early version I saw in the stats basically is quite nerdy in terms of it's using all the kind of recent technical language, and there is probably gonna be a challenge in terms of how people communicate that because the good example of this is Ireland once again, cuz we're talking about Ireland anyway, right?

What we saw before this was that the advertising agency in Ireland, Basically banned energy companies using the term a hundred percent renewable for power in Ireland, because Ireland only has 35% of its power coming from renewables. So therefore it can't possibly be a hundred percent right. And it has implications for all the people running data centers in Ireland, right?

So suddenly where people have been talking about, oh yeah, our cloud super green, a hundred percent green. Now you've had the laws basically saying, no, that's not allowed to, you can't make those change. And this go next story has a similar thing to this cuz you are seeing a similar story in the UK the Digital Markets Competition and Consumer Bill is going through law, and this is a bit like the GDPR.

The idea is that if you are making misleading green claims and you continue to make them, when you're told to stop, you'll be liable to finds of up to 10% of global turnover for misleading green claims. So this might explain here.

Asim Hussain: say anything about green at all from this point

Chris Adams: I think this is the thing that was interesting cuz back to the world of cloud recently Amazon used to have a hundred percent sustainable as one phrase that was used.

But in the last year there was a change to say up to 90 x percent renewable instead. And I was wondering why they made these changes, cuz Google say we're a hundred percent renewable. Microsoft says we're a hundred percent renewable. Amazon has been really weirdly coy about this and I wonder if it's because they saw this lawsuit coming through. Realize that even if you are following the letter of the law and the way that you know, if you purchase enough renewable credits, you get to say your stuff is green. If it's seen as misleading to consumers, then you're still not allowed to. It may be that like organizations, they were being a bit careful about this stuff because yeah, there's a real shift in this stuff happening basically.

Asim Hussain: Well I welcome this. I think one of the challenges that I see in our space, and it's something that I've railed up, talked about thing before, is the different languages that organizations use. Like I was picking something up the other day. I think it was some food and it had written on it carbon negative at the top, which is a term which has no legal definition, carbon. And underneath it in smaller writing, it wrote climate positive. And I was like, okay, so it's carbon negative and climate positive. What do both of those things mean? And it's like so much left to interpretation. Whereas if we landed on, you know, like we work in standards in the foundation, if you land on a very standard definition of this stuff, I think that's really beneficial to the end user.

I think yes, it might take some time for them to learn the language that we are talking about, but they will bother to learn it cuz finally when somebody says something, they'll understand what it means.

Chris Adams: Yeah, I agree. I think this is positive. I think it possibly suggests a problem for yourself as a director of the Green Software Foundation, and myself, a director of the Green Web Foundation. So we, we have to might have to end up with a much less catchy sounding name of our respective organizations, Asim.

Asim Hussain: Oh my word. Yeah. Okay. Whatever. I'll take it.

Chris Adams: Okay.

Alright, so let's, we're just coming up to the end. Should we wrap up? Are there any events or things that we should be pointing people to or is there a list of up coming conferences we might tell people about?

Asim Hussain: Probably as we get close to Earth Week, there's a bunch of meetups being launched through the foundation. People are running a bunch of things around Earth Week, but yeah, we'll talk on, talk about it at a future episode.

Chris Adams: For next week. All right, then. That's it for our news and events roundup for this part, all the resources in this episode and more about the Green, Software Foundation are in the show description below. If you're looking at this podcast, and you can also visit podcast.greensoftware.foundation In your browser.

And if you did enjoy this show, please consider leaving a review on Spotify, apple Podcasts, or wherever you get your podcasts. And your feedback is very valuable. It helps us reach a wider audience and hopefully helps improve the content of this show. So thanks again for listening, and we'll see you on the next episode.

So bye from me.

Asim Hussain: And bye for me.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation In any browser. Thanks again and see you in the next episode.

The Week in Green Software: Calculating Software Emissions with Navveen Balani & Srini Rakhunathan

22 mars 2023 | 36 min

Host Asim Hussain is joined by two of only a handful of people to try and calculate the emissions of software; Navveen Balani of Accenture and Srini Rakhunathan of Microsoft. In this episode of TWiGS Navveen does a deep dive into the processes behind Accenture’s use of the SCI Specification to calculate a measure to track and, ultimately, reduce the carbon emissions of one of its internal reference applications. Asim also quizzes Srini on the upcoming CarbonQL project by the Green Software Foundation which Srini is leading. We also get some links to great resources (particularly for UX folk) and some exciting event news!

Learn more about our people:

Asim Hussain: LinkedIn / Twitter
Navveen Balani: LinkedIn
Srini Rakhunathan: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

How Accenture Implemented the SCI Specification Score to Track Software Emissions / GSF [2:59]
CarbonQL / GSF [19:49]
Powershell Module For Watttime Emission Data / Watttime [25:15]

Events:

GSF London Meetup (Thursday, March 23 at 6-9pm GMT) / GSF [30:12]

Resources:

Circa by Andy Wood (part of the Carbon Hack ‘22) / GSF[26:56]
Sustainable Design Toolkits for UX Designers: / Vitaly Friedman [27:33]
Sustainable UI (part of the Carbon Hack ‘22) / GSF [28:27]
Green Jobs Network / greenjobs.net [31:52]
Branch Magazine / climateaction.tech [34:04]
lowwwcarbon.com [34:32]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:

Asim Hussain: Oh my God, that's an interesting way of looking at what we're doing. Like we can't actually, like we're not doing the work ourselves just piling onto the DevOps people to fix all of our coding problems

Navveen Balani: it's a, I call it as a journey from DevOps to GreenOps, so finops and so on. Yeah. So integrate all of these and then get a highly sustainable software

Asim Hussain: as long as it's not my ops, as long as it's somebody else's ops, it's not my problem only joking.

Hello and welcome to Environment Variables. Brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software and your host Asim Hussain.

Welcome to The Week in Green Software, or TWiGS. Where we bring you the latest news and updates from the world of sustainable software development, I'm your host, Asim Hussain. Today we'll be discussing the latest development from a bunch of people, but on the call we've got Navveen from Accenture. We're gonna talk about some of the work that they've been doing calculating carbon emissions are using what's called the SCI specification, which we've been developing, the Green Software Foundation, and we'll also be sharing some valuable resources and some exciting events to do with the world of green software. Before we begin, let me introduce our guest for the episode of TWiGS, with us today, we have Navveen.

Navveen Balani: Hi everyone. I'm Navveen Balani. I'm a chief technologist with the Technology Sustainability Innovation Group at Accenture. Very excited to be here. I look at innovation and creating assets at the intersection of technology and sustainability. I'm also a Google Cloud certified fellow and a published author with several technology books.

Asim Hussain: Wonderful. Yeah, I did discover just how many books Navveen has been writing. You just, you go to his Amazon profile, there's quite a few there. I know how hard it's write one book, Navveen, so I'm quite impressed. Oh, it's you.

Srini Rakhunathan: Hi, I'm Srini. I'm with Microsoft and I'm working for the sustainability division of Microsoft. Sustainable Software is something that, that, it was just a spark about three years back in January when the goals around carbon emissions were announced, and I immediately wanted to do something within the consulting space for which I was part of at Microsoft, but I'm happy to be here.

And then Greens Software has started and I've been with them since I think the beginning and as far my interest with green software is going, I like to write a lot of blog articles because they're small and probably can finish it within a couple of weeks. A little lazy that way. But yeah, it's been exciting working with GSF, the different groups and I just look forward to learning a lot from this initiative.

Asim Hussain: Wonderful. It's great. It's great to have you both here and you've both been involved in green software since before the foundation were born. You've both been involved since day one of the foundation, so yeah, really appreciate it. So let's go with the news. Let's start off with the first piece of news.

So this is an a piece on our website, which is called How Accenture implemented the SCI Specification score to track software emission. We were gonna cover this last week, but we figured it made more sense to wait till this week so we could be joined by Navveen because Navveen, you work at Accenture and you are one of the authors of this post.

So yeah. Why don't you tell us people a little bit about what the article's about, and I'd love to know what were some of the challenges you faced while trying to calculate the SCI specification. Just very quickly, just explain to everybody, and the SCI is something called a Software Carbon Intensity specification, and it's a specification being developed by the Green Software Foundation to measure carbon emissions of software.

So yeah, over to Navveen.

Navveen Balani: Thank you Asim. The SCI provided us with a practical methodology to baseline carbon emissions offer software application included embodied emissions and reducing the same. And when we started realizing the SCI specification, we found out getting accurate data for energy and embodied emissions was quite challenging.

And as we were deploying the application on cloud, many of these details were abstracted by the cloud vendors. . So we had to change our focus from getting accurate data to probably approximations and work towards getting a baseline for our SES code. And once we had the SES code, we basically have a value to track.

And as long as you use the same approximation, you could easily identify what caused the reduction in GI code, for instance, did you change the code to be more energy efficient? Or you did hardware optimizations and based on hard hardware optimizations, the energy estimation got reduced. So with SCI and using this SCIis a score, we were able to in our carbon emissions and once we had the score, we went about with various reduction techniques.

Asim Hussain: So you basically, you calculated the SCI score using a type of ar, like a reference architecture. Could you describe the type of application that you were trying to calculate a score for?

Navveen Balani: Yeah, so our reference architecture was basically a three tier application, which was deployed in the cloud we had using the interfaces create created using Web application. At the backend, we have developed all the application code was available as a set of containers, which was deployed on a container management.

We had an API gateway, which all communications happened to the API gateway. And at the backend we had a few databases and no SQL and SQL databases, which stored the data for the application as well as alt-text information. And then we have other cross-cutting concerns like security logging, right, which is provided by the cloud vendors.

Asim Hussain: So it's basically like the, this is, this is about 95% of all applications in the world. Is this basically right apart from a database, an api and some compute using one mechanism or another, virtualized machines or something else. And you mentioned earlier, earlier cause you, because I think the, the thing you mentioned at the start is the thing that comes up all the time, which is a data problem.

It's a data problem. I think it's a problem that we've, we three have discussed for quite a while now. So what were the solutions like, how did you solve the data problem to be, to be good enough for you? As in like how did you solve getting good enough data for you?

Navveen Balani: Yeah, getting data was a challenge because. Most of the data right, is currently not available, or I will say really available. So we have to look at certain approximations. For instance, the first was around energy calculation and the various approaches to calculate energy curves. Some are based on coefficient values, like how many Watt Hours it would take to run a virtual machine.

Some are based on statistical methods and some provide methodology through a set of APIs. And most of the energy calculation strategies are based on certain approximations, so there is no single source of truth. And the intent here is to basically use the same methodology for any future calculation. So you could actually have a baseline and then you use the same approach again and again to find out the SCI score.

Asim Hussain: So when you say an energy curve, you are talking about trying to calculate how much energy is actually consumed by a CPU because it's a curve, right? It's not linear. It's a curve. Yeah. And what did you land on in the end? So what, which solution did you land on in the end? You mentioned a couple of solutions there.

Navveen Balani: We went ahead with a combination of a statistical based method and uh, APIs, which are provided through various third party sources, and we found out that we have to pick up one of the values. So we went ahead with a statistical based method, which provided in our instance, a better accuracy, I would say.

So our concern was to find out, accurate method will we end up using? Right. So statistical based gave us a good, I would say benchmark, but even if you have picked up the API method also, it was not about more about accuracy, but I think it's more about getting a value, right? So that we could unblock ourselves to arrive at an energy curve.

Asim Hussain: And so that's basically just a, just a sim- I'm, I consider myself the simplification. I'm the five year old on the call and you have to explain it. I'm asking the five year old questions. So you're basically talking about being able to provide some model with some inputs, be it. I'm guessing CPU utilization, maybe RAM usage, maybe.

It depends how much telemetry you can gather. If you can gather as much telemetry as possible, I'm passing that into a model which then just pumps out for you. Given all these inputs, we guesstimate your energy consumption is that amount because you actually don't know, like I can tell you from like Intel's perspective, I'm with Intel now, so I can tell you from Intel's perspective, like we have certain tools, like for instance, Intel's RAPL, but Intel's RAPL only really gives you data at the socket level. The mechanical mechanism of being able to calculate electricity, we can only give it you at the socket level just cuz that's the pipe coming in and there's lots of model-

to get that to the process level, you have to do lots of modeling to try and figure out how much of that incoming value do we partition off to all the processes running on a machine. So at some level you always have to use statistical modeling, even if you've got like a direct measuring capability. I'd say yeah.

Navveen Balani: So, yeah, as you pointed out, right? So it was as long as you can get telemetry data, whether it's CPU, Memory usage, storage utilization, network input, and output bites. Right? You should then convert it into some energy curves, some statistical models, and get a value, which is, yeah, which is your energy curve.

Asim Hussain: So what, let me ask you another question. So Srini, you and Navveen are just, there's a rare set of people in this world who've actually tried to calculate the emissions of software. I think you, there's probably, I used to say there's only two or three people who've ever attempted it, and now it's probably more like 10, 12.

But there's a small like limited skillset. So, and I'll start with you Navveen. I'll ask Srini as well. So what are your key takeaways from implementing something like the SCI specification?

Navveen Balani: Yeah, so that's a great question. So I would say we had three key takeaways from this case study one was around energy calculation, and as I mentioned, there were various approaches for calculating energy curves and there was no single source of truth. And the intent here was to use some method for future calculations to compare against the baseline.

And we documented the various approaches as part of the SCI Guide project, which I worked with Srini. And this can be a starting point for all software practitioners to calculate energy of this software system, whether it's using surgical methods, coefficient, or API-based technique. The second takeaway was around embodied emissions.

So the type of underlying hardware for running virtual machines and it's embodied emissions are currently abstracted by the cloud vendors. And we had to rely on certain approximations based on VM-type to calculate the embody emissions. And we relied on sources from the cloud carbon footprint, which had an Excel, which talked about for a given virtual VM type, what is the scope three emissions.

And there's similar study from Vista also, which talks about the cal- scope three emissions for a given hardware type. And third was around the cloud managed services. Uh, so there were certain managed services and shared services like the API gateway, load balancer, where the type of hardware is not listed by the cloud vendor.

In certain scenarios, we had to rely on certain approximation like we did with API gateway. We just assumed that 5% of our total carbon emissions goes to API gateway. And for serverless components, we relied on, uh, using, uh, a timeshare approximation and the utilization to come up with a carbon emission score.

Asim Hussain: So you just, for serverless, you just assumed it was like a normal workload, but the just your time shared it just cuz underneath it's a normal workload. It's a normal server that's been using it underneath. There is a server

Navveen Balani: that's right. So we wanted to attribute some value to it because we thought, okay, maybe a 5%, uh, based on our overall load might make sense to at least give some emission value to the API gateway.

Asim Hussain: Yeah. Good takeaways. Srini as somebody else who's been calculating SCI scores for a while, do you have any other takeaways from your experience implementing SCI scores that you think might be useful for other people to know?

Srini Rakhunathan: Definitely. Thanks for the question. I think I completely echo what Naveen said about learning how to do embodied emission calculations as well as for serverless. Other couple of things that comes to my mind, the first takeaway I would say is from a, the case study that I picked up was very similar to what Navveen was talking about.

Mine was a managed services set of application. I managed set of services combining to form a website. It's as simple as that. We had a Web application, you had a middle. I didn't even have an api. All that I had was a direct database, but to make things simple for me I chose managed services from a deployment standpoint, but from a calculation standpoint, it's the most hardest thing because a lot of abstraction is happening under the hood in terms of the, like Navveen has said, the database server, the front end server, and also if you talk about real production application.

You have replication. You have active configurations, active passive configurations. So the first takeaway that for me was from an infrastructure standpoint, the complexities around the deployment will we take into account the development machines that help build that infrastructure in the place? What should be included?

What should be excluded? So for me, the first and foremost thing is to define or be very clear. Around your software boundary, and that's where you need to make it iterative. Try not to boil the ocean in the first place. Say I've included everything. No, that's not how it works. Right? You need to iteratively keep adding and your calculation also ends up becoming iterative.

Second takeaway I would say is the networking piece, which was, I think we have talked about it enough in many of our group calls. It may sound the easiest to measure, but the most difficult to calculate. I can just go in and say, this is my data in, this is my data out in bytes. Hey, gimme the number. There's no reference number at all.

There's no reference multiplication factor that you can use. So I believe there are lots of studies going on and this is an area where we could do a lot, we could, we could invest in a lot of academic research to figure out what would be the best way to do this. And third, I think for me a revelation was I needed to brush up on my high school physics.

I needed to understand what's energy, what's power, what's joules. It's very important. It's just not software development anymore. If you talk about sustainable software, you need to know your basics, and so you should study.

Asim Hussain: I can tell you somebody's moved from like software development at Microsoft to Intel. I ask questions, which I think are just, I'm expecting an answer oh code it this way. Half my answers involve the word voltage. It's somewhere in the answer, and I'm like, whew. It's almost like a lost art. And I wonder if the secret to all of our world is this.

There's like a lost art to programming back in the day, like what you described, and I talked to a lot of people at Intel who, who write software at the very low level. And the understanding of what you're saying is inherent, it's natural. We've become so abstracted away and the abstractions are useful because make for faster development, but there is a nature of understanding how the discipline programmer in silicon has become quite distant.

Whereas if that reduced, that'll be something I think that interestingly might help us in our space. I've been thinking about this. I don't know if I've spoken to both of you about this yet, but this idea that I've had around, like you mentioned, software boundary Srini, and that's something we've spoken about a lot.

I know Navveen, you've been driving a project for a while now. The SCIentology, which I, I think we should change that name. Navveen. It's just the suggestion cuz it's SCIentology. But anyway, you've been driving a project around like when someone saves my application. What do they mean? What do you include in an application and what do you not include in application?

Because one of the things I've seen historically when people publish carbon footprints for applications is they conveniently cut out bits that they are either just inconvenient to calculate, but they are part of your application boundary. Like I've been thinking a lot more about monitoring. The monitoring of an application, like when you have these larger applications, you're monitoring all these different components of your software product.

You're monitoring the database, you're monitoring the front end, you're monitoring this, you're monitoring that, and I've been having this idea that what if you defined your software boundaries, everything that you're, if you care enough about some software component to monitor it, is that philosophically part of your software application? Does that make sense what I'm saying here right now?

Srini Rakhunathan: Definitely. And I think if you wanna make it systemic, if you want to make your calculation systemic and repeatable, you would. That's how we do, right? When we, when we want to calculate the number of users who are actually using their application, we fall back to telemetry, we find, and that's, we use words like user journey and scenarios, et cetera, et cetera.

But then here, if you really wanna understand what are your operational emissions, you need that telemetry, you need that telemetry across the, all the software components in a system. So yeah, definitely it's, it's, it's a must have.

Navveen Balani: I also think maybe if there is a simple way to calculate operation emissions and carbons directly from the data which has already been available or gathered, the whole generation of SCI's score might be very simplified. Maybe do whatever data you have, if you are maybe proxy data, cost data and some efficient way of converting that to a score because one of the challenges I would say is going through the SCI requires a fixed set of steps, right? Defining your boundary calculations, coming up with the conversions, and so it has sequence of steps, it takes time, but any available proxy, if you are able to quickly generate a SCI score, at least some benchmark.

I think that would be an ideal scenario, and maybe it's integrated as part of your DevOps in future. So you don't need to do much, right? Just build. Get a score next release, get a score. Maybe you're not there yet, but hopefully that's the intent and direction.

Asim Hussain: It's come up a couple of times on the podcast about the GreenOps, DevSusOps, that aspect of it, and actually monitoring is part of that. That's what you think of when you think of DevOps and is it, I think it is. There's monitoring. I hope it is. Someone should be monitoring my applications. It's not me. Hope it's a DevOps people.

Maybe the DevOps people think it's me,

Navveen Balani: Now we are giving more additional responsibility to DevOps. I would say

Asim Hussain: yeah, , that's just, that's just . Oh my God. That's an interesting way of looking at what we're doing. Like we can't actually, like we're not doing the work ourselves just piling onto the DevOps people to fix all of our coding problems.

Navveen Balani: it's a, I call it as a journey from DevOps to GreenOps, so finops and so on. Yeah. So you integrate all of these and then get a highly sustainable software system.

Asim Hussain: As long, as, long as it's not MyOps. As long as it's somebody else's ops. It's not my problem, but I'm joking. So maybe let's bring it back down to an interesting, I think both of you also work on, I think I teased in a previous podcast episode this idea of something called the CarbonQL Project, which Srini you're leading, you both heavily involved in, maybe let's flip over into that and talk about this brand new project that we're launching in the foundation called CarbonQL.

Do you wanna give it a quick overview?

Srini Rakhunathan: Definitely. Thanks Asim. Again, this has been something that I'm really excited that there is going to, it's gonna change the way we look at carbon value or how are we systemically capturing monitoring data. The intent of the project is to be able to provide a value, a carbon emissions value, which you can use.

It's more for action for you to continuously iterate and figure out where are you at a particular milestone after you have taken some of the measures provided as part of the SCI specs, which is around making it more carbon aware, making it more efficient or energy efficient, or all of it. So you need a way to tell whether you have progressively made or passed your different milestones, whether you're continuously reducing or you're stagnant or you're increasing because it's always possible that you need to pull all these parameters to make sure, because we are not building applications just to make it sustainable, right? We are building applications to make money for your business.

Asim Hussain: Add value to the world, let's put it that way.

Srini Rakhunathan: Exactly, and so you need a way to easily calculate across your different hosting infrastructure, whether you do it on the cloud or on frame. You host your app on your laptop. The project aims to tap into the different data sets available. And abstract away the calculation algorithm and just provide you a value, most intelligent value.

That's what we would say when we were kickstarted it, and I think it's going good. We should probably have something really cool coming out of this. Navveen do you want to add anything?

Navveen Balani: So maybe I think CarbonQL will be a Chat GPT of Sustainable Tech. You ask questions, what is your emission for laptops? What is your emission for mobile phone? What is the emission for your CPU? What is the emission for my software boundary? You give us spec, right? So can we ask easy as you just give us spec or a software boundary or the artifacts from SCI anthology if you define something and just please provide me a emission so it can make lives easier for people who want to calculate carbon emissions and maybe provide a simplified way of getting carbon emissions be for developers. DevOps, if they're able to provide that vision and implementation. I think the calculation journey would be quite simplified and more adaptable, right to all the development community.

Asim Hussain: I think we've mentioned the chat GPT every episode for the last 10 episodes now. So it's come. It's come up again. Yeah. Just for the audience. It is not a flavor of chat GPT. You are not going to be typing in what is the carbon emissions of my three tier thing. I think what you're alluding to is there's quite a few different methodologies and data sets and models and statistical models you can use.

And are you on a laptop and maybe you want to use this model. Are you calculating from mobile phone? Maybe use this model a server, use this model or this API or this data set. And I think what I've always found is that the set of people like you, two or two, a very rare set of people who've actually sat down and calculated this stuff manually.

The knowledge that's in your head is rare. Actual, real experience. People can read like roughly theoretically how to calculate and sum it all together. But you've actually got like that real knowledge and what I see is CarbonQL is gonna try and codify that so that anybody else. Does not need to be an expert in this space.

They can just say, I'm using the greenhouse gas protocol methodology. I'm running on these types of machines. This is the telemetry I've managed together, but I don't really have all of it. And then you'll just figure out the rest and like use all of your expertise and all of your best judgment to combine all the data sets into one, which I think is, yeah, it's been the number one problem.

I've heard from everybody about the SCI since day one was data. It's really hard. There's lots of data sets out there. We don't even know if they can merge. I get this number from one dataset. I get a number from another dataset. Can I add them together? Is that possible? We don't know, so the set of people who know that is low and you're two of the sets of people who know that.

Navveen Balani: I think the data, once we have CarbonQL implemented, right? I think through the data we'll get various insights across all data sources. Maybe that might be a good way to look at the various data trends, right across all these, um, APIs and SCI, right? Whether you want to use average value or which works for a given scenario.

Asim Hussain: Yeah, I'm really excited. I'm really, really excited for this project. I think this is one of the projects we, we've, uh, it's been on the tip of our tongues for a while now. and it's just really exciting to get it and there's been a lot of interest from various organizations in the foundation from getting involved, so I'm, there's a lot of interest in this.

So really excited to see what you both and the rest of the team deliver on this front, and we'll get you on the podcast again as things get a bit more mature to talk about kind of where the progress is. Let move on now to few of the other bits of news. One other bit of news that's happened recently is a PowerShell module for Watttime emissions data.

So Watttime are one of the organizations that price carbon intensity for electricity. They are a member of the foundation. So Henry Richardson and the team at Watttime have released a PowerShell module that retrieves near real time emissions data from Watttime for a supplied Azure region during resource deployment, this is a really lightweight solution.

It utilizes the limited functionality available with one of the Watttime's free accounts. Because it is a pay, they do need to earn money, and it's not really like a real solution for reducing carbon deployments of software. I imagine to get that data, you probably should pay for the full licensed versions of the data and get the more accurate data.

But it does provide some nice real-time values to simulate the behavior of deployments and software based on emissions without the cost of a paid account. So all they need, they'll need an account on Watttime, and you'll need the AZ resources PowerShell module for Azure installed. I need to connect all to Azure account.

This is a really great free resource, which we're gonna link in the show notes. Are any of you PowerShell users? Are any of you PowerShell users? Cause I've got to admit, I am not.

Srini Rakhunathan: I have used it.

Asim Hussain: You have used it. Srini's like, I work at Microsoft. I should know I should be a PowerShell power user. But you are like how I used to be at Microsoft, which is, I don't really use SharePoint, but yeah.

Power PowerShell. Yes, I PowerShell is, it's like a command line tool from my understanding for Windows, isn't it? Yeah. So it's really exciting. Cause I know it's using a lot of DevOps scenarios. Pretty simple idea and we think so complicated about the world that we do. This is a really simple idea. Just bringing a lot of this functionality into the command line interface all by itself is, I think, quite valuable.

Navveen Balani: No, I just remember we had a similar innovation right in the hackathon that we did some something around shell trying to get carbon intensity.

Asim Hussain: Circa that was it! It was called Circa. Yeah, in the hackathon we had last year. That's a good, that's a good memory. We had a, yeah, I remember now. There was one of the submissions was Circa, it was one of my favorites and it was a command line where you typed circa. Then you type the command that you wanted to execute, and it would literally just do a sleep.

Such a simple idea, just a sleep on the command line until the electricity becomes better, and then it lets the command run. So all these simple ideas. Very, very good. So let's move to the next one. So sustainable design toolkit for UX designers. This is from Vitaly Friedman. I've met him. We, I've been, I've spoken at his Smashing Mag conference.

I've met him at quite, quite a few conferences. So Vitaly if I've got your frame name wrong. I'm really sorry. So he's released a really cool list of sustainable design toolkits for UX designers, all with practical guidelines, frameworks, and tools to focus on what matters and removes what doesn't. The list is really comprehensive and includes a UX checklist for sustainability. It looks really useful. Again the link's down the show notes. I took a quick look at it earlier on today. He posted it. If you know there's a big post on LinkedIn and there's a lot of great stuff, I get asked about this. I just got asked about it last week. I spoke at a conference and people are asking for like a UX framework.

Because a lot of what we talking about starts at the UX level. We do need to build some of this functionality. I remember what there was one, again, like coming back to the Carbon Aware Hackathon last year, there was one proposal, which I thought was really interesting, which is like carbon aware components you can put into websites.

So like a literal carbon aware, I think it was carbon video or something. You can type a HTML tag called Carbon Video. and it would play the video if it was electricity was clean and if it wasn't clean it was, I think it would just play the audio component of the video, save on energy or something like that.

So I think a lot of this stuff, it all starts really at that stage. Before you even speak to a developer, it's probably the best time to do a lot of your sustainability thinking. Did you guys have a chance to look at this? Don't worry about it. If you don't, UX is for the other folks.

Navveen Balani: I think given how the UX will evolve, right? Particularly new interfaces maybe in Metaverse, right? I don't know. We will have a lot of thinking to do to decarbonize the entire UX story. Not now. Maybe when you see a lot of emissions where everything is virtual. I don't know what kind of carbon emissions we would have, right?

Maybe five years, 10 years down the line.

Asim Hussain: Yeah. Yeah. Very interesting point. Or may, maybe the only interface with how five years down the line is a version of GPT that just work GPT or your laptop just opens up and it's a chat GPT, like command line at the bottom and that's it. There's nothing else left for us to do, do my work for the day.

You know, I joke about this, I had to hunt around for it, but I actually found. Four years ago I was giving a talk. I think I tricked an audience into thinking I'd built an AI that could make a website. From a textural description, I tricked the audience. And then just last week it happened, and now the chat GPT four can see a picture of a website and code it for you.

So I, I wonder how long we've got left. Probably not as long as we think, but while we're here, let's make sure we go out with the bang. Last thing, lemme talk very quickly about the really excited the Meetup program. So the Green Software Foundation has a meetup program, meetup program where we support meetup groups.

If you wanna launch a meetup group around green software, In your region, or actually we've got like 20 regions where we already have members of groups, where we are actually missing organizers and leaders. If you're interested in launching a meetup group or or taking over or co-organizing a meetup group or even speaking at meet group or even being involved in any which way in a meetup group at all, go to meetup.greensoftware.org.

I'm very excited cuz this week actually GSF the London Meetup, which the first one that we're launching now is having an event on Thursday, March 23rd at 6:00 PM GMT. It's at the MasterCard offices in Angel Lane in London. There's more links in the show notes. Chris Adams is going to be there, so he's visiting from Germany.

I'll be Liya Mathew and Sarah Hsu from Goldmans is gonna be there. Daniel Vaughan's gonna be there for MasterCard. And big thanks to MasterCard for helping us host this event. And gen- are offering the venue. So if you're in London and you're listening to this podcast, pop over to the meetup. Thursday at 6:00 PM and if you're interested in launching a meetup, please reach out to us and we'd love to, uh, get your help and we'll to help support you launching a meetup.

So that's it for our news and events roundup. As part of the new format of TWiGS, we have a short closing question that we'd like to ask our guests, and it's gonna be different every single week. So whilst I go and figure out my answers, question's going to be, I'm going to ask Srini, what is a go-to green website that you have booked in your, bookmarked in your browser.

Srini Rakhunathan: So we are part of a sustainability action group within Microsoft and a couple of months back during one of the community calls, there was a website which is founded interesting. On green jobs. Yeah. And I don't think it's a carrier website or a job search site, but I found it interesting that it collated all types of green jobs.

You know, it could be an electrical engineer, it could be as simple as a sustainability policy maker. So I think I have it in my bookmark. I keep looking at it and trying to understand what types of carriers are there in the sustainability space.

Asim Hussain: Wonderful. Great. Send it over our way. We'll put it in the, we'll put it in the show notes. Yeah. Navveen, what go-to Green website? Have you bookmarked in your browser?

Navveen Balani: Interesting question. I would say, I think the simplest answer would be go to google.com. . I think from a, I think from a technology perspective also, and from a sustainability perspective, also, they have very simple interface where you can get whatever information you want, right? So I really love the way they have for the last 10 years.

Made the search more. Yeah, minimalistic. I would say for the two decades, I would say I haven't seen it change much. And maybe through that I'll explore more websites, but. . That is what I think it comes to my mind. That's probably billions of people might be using.

Asim Hussain: Just remembering, like back in the days when like search engines were like appearing, like there were just big messes on the screen, but Google's was just like this empty page. And then this thing in the middle, and it was like a breath of fresh air rather than all these other sites you're going to. So for me, I've gotta be honest with you, like my head is fully in the foundation and kind of the, the, the, I was just checking, I was just literally checking my history, my bookmarks right now and like we are currently like building out a wiki, so like pretty much every other bookmark right now on my website is a wiki page where we're trying to flesh out like every single aspect of, I, I joke that I manage kind of the, the operating system for the foundation and we're like building out what the operating system is and how we work and how to get involved and lots of information there.

So that'll be coming up not right now, but in a future episode be released in the Wiki, but that's, that tends to be what I'm working on. The only other kind of, I think, website that I really love in this space is is Branch Magazine. , which is, uh, Chris Adams who's the other host who of this podcast talks about.

He's built it originally, but it's a wonderful like magazine, which has like lots of articles about green technology and the really cool thing about, it's also carbon aware, so, when it, the electricity's dirty, that doesn't show images and when electricity's clean. So actually it kind of implements what it's, Oh! I have another one as well actually! I just remembered another website, but I'd admit I don't have it bookmarked cuz I just remembered it. But there's a website and I just discovered it the other day. I think it's very cool. It's low lowwwcarbon.com and it's a showcase of low carbon websites, how they've done it, and the case studies, how they implemented it, some examples.

That's a pretty cool website I've got. So yeah, I didn't think I'd have any, but I've had, I've got a couple actually now just so it just goes to show. That's all for this episode of The Week in Green Software. All the resources for this episode and more about the Green, Software Foundation are in the show description below, or you can visit greensoftware.foundation. That's green software, one word foundation in your browser. If you enjoyed the show, please consider leaving a five star review on Spotify or Apple Podcasts or wherever you get your podcast. Your feedback is incredibly valuable and helps us reach a wider audience. Thanks again for listening, and we'll see you on the next episode.

Bye-bye.

Navveen Balani: Thank you for listening.

Asim Hussain: Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review if you like what we're doing. It helps other people discover the show. And of course, we want more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation Thanks again and see you in the next episode.

The Week in Green Software: Web 3.0 and Energy Standards for Software

15 mars 2023 | 41 min

This episode of TWiGS has Chris Adams at the helm again with guests Asim Hussain and Anne Currie. They talk about the impact of Web 3.0 and why the future of immutable blockchains needs to be open source and sustainable and perhaps isn’t the only solution out there. They also talk about recent news from the BBC, AWS and highlight some great resources for you to expand your knowledge in the world of sustainable software.

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Anne Currie: LinkedIn / Website
Asim Hussain: LinkedIn / Twitter

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

How slimmed-down websites can cut their carbon emissions: / BBC [3:34]
Amazon denies claims hiring freeze is slowing AWS sustainability work / Computer Weekly [13:55]
Nori launches Web3 Marketplace: / Nori [24:38]
Why the future of Web3 needs open source, sustainable blockchains: / Linux Foundation [33:35]

Events:

QCon London Software Conference (March 27-29, 2023): / QCon [35:14]

Resources:

Sustainability at AWS re:Invent 2022 / Adrian Cockcroft
FOSDEM 2023 Responsible Clouds and the Green Triangle / Chris Adams [12:42]
OVH’s 2020 announcement of carbon and energy metrics APIs for customers / OVH [21:46]
The Cloud Energy Project / Green Coding Berlin [24:02]
Asim’s Interview with Nori’s Founder on The Climate Fix [26:35]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcription Below:

Asim Hussain: In the foundation we talk about carbon efficiency, which is minimizing the amount of carbon per whatever unit per value that you're providing to the end user. And I think there's multiple ways you can think about that cuz you can actually think about fundamentally changing the nature of your application so that you can actually provide the same value without even needing the same functionality.

And I think that's kind of the way we need to really think about this stuff.

I'm your host, Chris Adams.

Welcome to this week's episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and in this episode we have some exciting announcements from the world of energy standards for software, and yes, even Web 3.0, we also bring you some exciting upcoming events. Before we dive in though, let me introduce our participants for this episode of this week in Green software. With us today, we have Anne. Hi Anne!

Anne Currie: Hello, Chris.

Chris Adams: And we have Asim Hussain. Hey, Asim.

Asim Hussain: Hi Chris!

Chris Adams: Okay, so if you don't know Asim and Anne, maybe we should, they should introduce themselves.

I'll hand over to you Anne first, if that's okay. Cuz it's alphabetically ahead of Asim. And then we'll know to you Asim. So Anne, for people who've never met you, how would you introduce yourself?

Anne Currie: So my name is Anne Currie. I've been in the tech industry for nearly 30 years, and I've been an engineer in various other things. And the past six or seven years I've been doing quite a lot on sustainable software. I work for a company or I work with a company called Container Solutions, and I am one of the co-chairs of the GSF Community Committee.

Chris Adams: Cool. Thank you Anne and, Asim over to you.

Asim Hussain: I really respect the going via alphabetical order. If you go to the Green Software Foundation website, you'll notice that all our companies are listed in alphabetical order. So as, uh, to be fair, So my name's Asim Hussain. I'm the executive director and chairperson of the Green Software Foundation. One of the ways I used to describe what I do there is I'm in charge of the GSF operating system.

I'm like Linus Torvalds but for the GSF operating system, that's who I am. But I'm here to help all the wonderful people like Anne and Chris and every else involved in the foundation build their amazing solutions and help them to execute what they do.

Chris Adams: Cool. Thank you Asim. Alright. My name is Chris. As I said before, I am the executive director of the Green Web Foundation. I work with the Green Software Foundation. As their policy chair for the policy working group. And I also help run a small community online called climateaction.tech, which is passed recently, 8,000 techies working on climate and tech in this particular intersection.

And, uh, Yeah, that's what I've, that's the thing I do. And that's the milestone that we've passed recently. But before we dive in, we should just stop and I'm just gonna make you all aware that anything we talk about today we'll share links. So if there's something that caught your eye, if you go to podcast.greensoftware.foundation, you'll see some links specifically to those stories.

Okay. So I guess for that, folks, should we start looking at some of the news and see what's showed up in the news this week?

Anne Currie: Yes, let's do that.

Chris Adams: Okay, so the first one we have here is how slimmed down websites can cut their carbon emissions. This is a story from the BBC and this is the first time it's actually one of the set. The BBC has some, is some form in looking at this stuff.

But this is a recent story that came out in the last week or two actually, and uh, I might hand over to either Anne or Asim who've got some records on this one because it's quite nice to see the BBC looking at this. But there's always more to this story than what you actually just see just here.

Anne Currie: Chris, it's, it's interesting you've handed over to us for the reckons on this, because you are literally the world's expert in this particular field, are you not as the chairperson of the Green Web Foundation, which is entirely devoted to this very subject of how you make the Web less carbon intensive.

Asim Hussain: I think Chris is being humble, so

Anne Currie: A humble aside, Chris, what do you think? What do you think about this?

Asim Hussain: Because you know what if I was ever asked? I will just forward the request and ask Chris what he thinks and then mirror that statement out to everybody else

Anne Currie: Indeed, so would I

Asim Hussain: Yeah.

Chris Adams: Well, the thing is I, the thing I can actually point to is that, that's nice of you to say, although I work at the Green Web Foundation, there are lots of other groups who are mentioned in this now. One of them is the Eco-Friendly Web Alliance, which I'm not so familiar with, but some of the other names, Wholegrain Digital and Eco Grader are both listed in here.

In this, they talk about some of the tools that you can use to basically, estimate the environmental impact of a website by looking at how much data is sent over the wire. Now, the thing that I should share with you is that this is one way that you can, uh, Get a rough idea of what the environmental impact for Web website might be.

But it's also useful to bear in mind that a phrase that all models are wrong, some models are useful, and we maintain a library called CO2.js, which helps get some of these kinds of conversion factors. So if you know how much you're using something, you might get an idea of what the environmental impact might be.

But there's a lot more to just looking at the data that's sent over the wire. And I think I might hand over to one of you two here because there's a few other things I can tell you about, but I'd wanna make some space for either of you, because I'm pretty sure you have some questions at this point.

Asim Hussain: So they talk in this particular article or the show about how the highlight of the story is how a UK knitwear designer was trying to do everything else with her business to be environmentally friendly, sourcing surplus yarn. Everything is made to order using hand powered knitting machine. But, Not being aware that her website emits con nine, was it 9.89 grams of carbon every time a user visits her homepage, which is 10 times the eco-friendly Web alliance's recommended limit of one gram. So that's an interesting aspect of it all, which is like you, you've got this person who's building this business. Interestingly, just as a slightly slide, my wife is also a UK based knitwear designer, and literally yesterday in the evening, she actually was spinning her own yarn using one of those oldie style like foot things.

That's what she does. So this is, this is, we'll definitely watch that show. But yeah, it's interesting. So, You're doing all this stuff in your life, but you don't realize that your website, the thing that you're selling, your product actually has carbon motions as well. I think so. Chris, you were mentioning something before we used to hit recording.

You mentioned something about how it was like people just realizing that , your digital aspect of your business is almost just completely ignored. You don't even think about it. And I remember like those, there was some, a video that came up to my radar the other day was just a bunch of young guys sitting around talking to each other and one of 'em going, did you know you can edit the internet?

Another one going, yeah, you know, you can go to Wikipedia and just edit a page. No way you can edit

Anne Currie: Good luck with getting that edit accepted!

Asim Hussain: but it was like, I feel like that's maybe the statement that's coming through from this show is, oh, did you know by the way, that websites also emit stuff they have, and there's things you can do to reduce that.

And it's that realization in people's minds that this is actually an important aspect of their total emissions footprint.

Chris Adams: This is true in that yes, the energy has to come from somewhere. That's definitely the case. Now, whether a website being 10 times the size of another website mean it means it has 10 times a carbon footprint, that's another matter. And I see you shaking your head. So I think, I reckon you've got something to share before I come in on this one as well.

Anne Currie: Yeah, there are loads of factors, operational factors to take into account when it comes down to that. Where is the power coming that is powering the servers that are hosting that machine, for example, is, are you hosting somewhere where everything's green, all the power's green, in which case, you know, fill your boots, it's fine.

Are you delivering content at a time when basically everything's green, so maybe only business content and therefore you're not really serving up at peak times? For example, in the evening when people are home, if you are a very popular website, then you might be using a cdn, you might be caching your data, and in fact, the more popular you are, the more likely the data is to be cached somewhere close to the Edge, so your website isn't even hit when the big images are pull pulled.

So yes, there are loads of factors that are not just about whether you've got a really heavyweight framework that's serving up your webpages, although there are bad reasons for that as well. There are accessibility issues associated with very heavyweight web platforms as well. Cause often they do not have the accessibility features.

That's basically an HTML based website would have. So there are lots of issues and lots of questions.

Asim Hussain: It also depends upon the value that your website is providing to the end user as well. Because there's 10 grams per sending some lifesaving medication to something. It depends on the value it's providing as well. I'm sure we can always like, sure all of our products are as efficient as possible, but that is the one way.

I really talk about it is in the foundation we talk about carbon efficiency, which is minimizing the amount of carbon per whatever unit per value that you're providing to the end user. And I think there's multiple ways you can think about that cuz you can actually think about fundamentally changing the nature of your application so that you can actually provide the same value without even needing the same functionality.

And I think that's kinda the way we need to really think about this stuff.

Chris Adams: Thanks Asim. Alright, so the thing I might share now is, yes, I agree that we'll do all these things. Accessibility is absolutely a factor that gets played into this because there's actually studies and there's evidence showing that, uh, Basically designing a digital service in a particular way can basically lead to induced updates.

So people being forced to have to upgrade from, say an old phone to another phone and something like that. And when you bear in mind and Yeah, exactly. There's actually a really good study that was shared at a conference called The Limits Conference, where people are looking at the devices that were being used in various distance learning tools, and you could see very quickly, At a certain point how just certain devices just stopped showing up in the analytics at all simply because they were being, essentially they, the site was no longer accessible to people.

And given that around 75% of the impact for most devices is actually from the manufacturing, not the use, this is one of the big leverage points actually. If you are interested in this, there is actually some work from the Green Software Foundation. There's actually a training course for this. In my organization we've recently published a handbook for community tech specialists who are basically taking their first steps into this. But the thing to bear in mind is that, It doesn't automatically follow that a website being 10 times the size is 10 times as bad for the environment. It really doesn't necessarily work like that.

It's useful to start, but there are tools to, for you to get a much better idea and really observe directly where the energy is being used in the system. But in order for you to do that, you do need to have a bit more transparency through this. And because the internet is made up of so many different companies, that can be quite difficult and the story of transparency and companies when you're trying to understand where the hotspots are in digital services is a perennial one, but this is actually one that is actually getting better in that there are companies like Microsoft and Amazon and Google who now expose some of these metrics to end users.

Asim Hussain: There is a very significant percentage believe it's safe to say like greater than 50% of all the energy. Like if you think of like using laptops, what is the application that's actually been used the most on a laptop? Like we think it's like the actual things we install. It's not, it's the browsers. So that's why websites are so much more significant than we may be give them credit for, because a significant portion of the energy from a desktop is be of my laptop.

Desktop is being used to just to browse websites. I think the mobile, I've heard a slightly different story, which is that mobile use, people tend to still install things, but then they probably still install. That's a technology. Now, Web, you're installing effectively a Web. , but it's an installed thing. So I think like the Web is, at least in terms of end user devices, probably the most significant consumer of electricity I'm guessing here right now.

And I'm also guessing that on the service side, that's not going to be the case. It's probably gonna be something more like a numerically computational like machine learning thing or something. I don't know.

Chris Adams: So here's one thing that might be of use. So we spoke about the Firefox browser having some of these metrics cuz they've basically the people who build the Firefox browser set a target to make it the greenest browser possible. In order for them to do that. Need to see where the actual impact is taking place.

And I dunno if any other browser that provides the same level of detail right now into this, and it'd be really lovely to see the other browsers doing this to expose those kind of metrics out there. But this is probably one way you can find out. So we could actually see which applications are doing this now in Germany at least there is some work going on to start providing a kind of labeling system for various processes.

And uh, one thing that was shared at FOSDEM, a recent kind of open source conference was essentially comparing an open source word processing tool and a well-known proprietary word processing tool. And basically as you got a, a document open, the actual blinking cursor was basically one of the things that end up causing a massive spike in CPU every single time again and again.

It might be fixed now. If you look up the project called SoftAware, and I'll share the link for that. There's a bit of information around here, and the folks at, oh God, KDE, they have spoken about this at length, and there's a really nice talk at FOSDEM about specifically this stuff. We'll share a link to it. Okay, so what's next on our list?

There's another one, Amazon Sustainability work. This is a story in computing weekly and the short version, the headline is basically Amazon denies claims hiring freeze slowing AWS sustainability work. And, uh, as I can see it, basically a number of high profile people left the sustainability team within Amazon, and that's led to a number of people who are downstream as consumers or customers of Amazon thinking, oh Christ, what's gonna happen with the actual metrics?

Is it gonna keep developing at the speed it was? And given the size of Amazon, this is probably quite a substantial story, basically.

Anne Currie: Yeah, it does really matter what the HyperCloud providers do, really does matter. And it sounds like aws, have they, they had a bit of. They moved forward, they did quite a lot, but now suddenly they seem through falling back behind Azure and Google again. They're still making good progress, apparently on sourcing green electricity to power data centers.

But in terms of that whole architecture, so it was two years ago now, I think they announced the sustainability. Architectural pillar, which was an indication that everyone needs to actually start thinking about how to we as Amazon users, we were responsible for the sustainability of our own systems within the cloud.

So AWS said that, take responsibility for the cloud, but we had to be sustainable within the cloud. And that really meant things like using serverless, using spot instances, that kind of thing. Just being more efficient, turning things off when you're not using them and all that kind of stuff, and they had quite a focus on it, but lots of people have left and now there seems to be much less of a focus on it.

And the thing is that with Amazon, the reason why I'm animated about it, because there's something we can do about it because. Amazon are very focused on customers and customers saying they want it. So here, if we want that to get focus back again, and it is important that it does have focus. We need to be telling our AWS reps that we care about this and we're annoyed that we're not getting the progress that we are seeing that we would've had we been on Azure or GCP.

So yeah, use your wallet to have your, say, tell your AWS reps that you care about this and it needs to go back up the priority list.

Asim Hussain: I feel capable of giving a balanced viewpoint now that I don't work for one of the hyperscalers anymore. But I also do, I do work with a partner who partners deep with hyperscalers, so I'm not gonna. Say anything wild. But what I will say is that like it was always like, cause I was at Microsoft at the time, like a couple of years ago, I remember there was an article by Wired very early days, just ranked the hyperscalers and gave them an A, B, C, D, F score.

And I remember Google got B and Microsoft got B plus and Amazon got. Hopefully we'll correct, we'll make sure, we'll make sure the right number is there. I think it was like a D or an E or something like that. But the, from those days to just recently they moved, they accelerated very fast. I remember it was not last year.

It must have been the year before. I remember that. So they announced a couple of things that, not the last reinvent, the previous re is it is reinvent every year. Is it every

Anne Currie: Once a year. Once a

Asim Hussain: last, not the last

Anne Currie: It wasn't last night. It was the previous one. Yeah.

Asim Hussain: they announced that the sustainability content, the WAF content, they're from the rightly, the well architected framework content.

Then almost immediately they tease their cloud carbon footprint calculator, and I tell you right now, I was there thinking, oh, they're teased it. Well wait a year and then, then they'll announced it and within months it had come out. So I was at the sidelines going, wow, this's. They're really going for Amazon.

Really impressed. This is excellent work. I chatted to people who were in there and I was very, I was honestly very, very impressed with Amazon's execution in this space. It was happening very rapidly, very fast. I, from what I heard internally, you know, this was actually being driven the way I would, I think it should be driven from within organization, which is top down.

You were getting measured at the leadership level for sustainability commitments, which is why it was happening. So it's really sad to me to see an article like this, and I think there was this follow on conversation with Adrian Cockcroft on LinkedIn, which is about, there's cuts happening across the board in tech companies.

And I remember I was quite naive at the start of the year. I was thinking, no one will touch sustainability, but we're on the chopping block like everything else.

Chris Adams: Well, I guess if you're only making 300 billion in revenue each year, I gotta, I dunno how you're possibly gonna be able to hire anyone, right? How many hundreds of billions do you need to be earning before you can bring a sustainability team in? For folks who actually using Amazon and feel a little bit defensive about this.

We've also shared a link to some of the sustainability theme sessions at the last AWS reinvent from Adrian Cockcroft's blog, where he's gone through the the list of all this and put together a number of really interesting and useful ones for this. But this is the thing that, as Anne says, if you were a customer of Amazon, this is probably one thing you could ask for to help a customer obsessed organization move a bit more quickly on something like this, because this is one thing where I think they're quite comfortably behind the other two right now. And given the size, they probably have disproportionate impact.

Anne Currie: Yeah, so actually I do as well to add something like, because this has reminded me of something I'd forgotten about, but it's very relevant to this podcast. One of the things I'm working on at the moment is a book that I can talk about, which is a, a reissue of a book I wrote, ooh, about six or seven years ago about Cloud native and it had a lot of case studies in about people who are doing interesting things and we're just about to reissue.

So we've re updated all the case studies. One of the case studies is a particularly interesting one from Skyscanner who are a company in the uk. And what they've been up to recently has been FinOps with a climate's twist. So they have been looking at ways to cut their hosting costs. And that's mostly been through, they have massively reduced their posting costs in a large part by using spots, by moving over to using spots.

But the tool that they said, you know, actually it's the AWS stuff is better than it used to be, this tooling, but none of it's enough for them to really do a FinOps flywheel and proof test and a tool that they were really, and this is, I've never used this tool, but they were raving about it with something called Cloud Zero.

And obviously I don't know them, I know nothing about them, but that appeared to be what they were using instead of good quality AWS stuff. Basically there are other tools out there that, and maybe that's why AWS and I really want AWS to keep the foot on the gas here, but there might be other tools coming out that do some of the job quite well.

Keep your eyes open for other tools.

Chris Adams: So I'm gonna come in here with a question, if that's okay to put to the two of you. And we have it written here. So let's say you're a customer of a cloud service provider, Amazon GCP, Azure, maybe OVH or ScaleWay, what do you actually need to have access to to effectively manage the environmental impact of digital services?

Like Asim I reckon, this was some of the stuff that we were struggling with the software carbon intensity for figuring out some of the inputs for this, and I'm sure. I reckon you might have some reckons for people to know what to ask for.

Asim Hussain: Yeah, this has been one of our first conversations on it Anne you were challenge. Yeah, the data. It's data. It's all the data problem, like speak, whether you calculate the standard that we are developing here, which is just for almost an internal metric for teams, a software co-, whether you're calculating a greenhouse gas footprint, whether you're doing an LCA, whatever it is, you need data.

And the big problem is there's very minimal data that's surfaced to customers. It's not granular enough the methodologies are challenging to understand how given the methodology of that calculation, how can it actually be merged and combined with other data sets. So just a lot more transparency around that.

And I also understand, I hear it from multiple sides, I understand it. It's sometimes quite challenges for organizations to reveal this data. I've even heard that at some point becomes such a contradiction where you could actually, some lawyers have said you're actually revealing materially non-public information about your company and that gain you trouble with the SCC.

There's, it's a very complicated space, but the truth through all that complexity is more transparent data is what we need.

Chris Adams: So there is one thing I will point you to and I'll add a link to this because one of the things I do where I work is we maintain a directory of kind of providers who share some of these numbers. OVH are probably one of the largest cloud providers. In 2020, they announced APIs for exposing both energy usage at an instant level, but also embedded carbon usage to figure these out numbers out.

So these are the ones that we've been looking for ages for the SCI and like some providers expose them as part of the service. This is like mind blowing for me when I found out about, I tried to chase it up cause I'm not a customer from 2020. This was, I found a.

Asim Hussain: they announced in 2020 or they

Chris Adams: It. It was announced in 2020. I'll share the link.

I'll share the post on LinkedIn where I was asking about it, because I'm not accustomed myself, so I don't know what the numbers look like. And honestly, this would be game-changing if that was actually something that was made available because it's just very difficult for people to actually have access to.

And at the moment, most of this is people making educated guesses or using various kinds of modeled approaches, which don't necessarily match reality. A lot of the time.

Asim Hussain: We will have to review that data and maybe we should come up with a. This of next podcast actually reviewing the data and then coming up with what we conclude rather than just being our guess guesses right now. But I would say one thing that comes up quite often is what we call measurement for reporting and measurement for action, and then large, again, one of the limitations that the hyperscalers have is because when they're actually giving that data, they're giving you from reporting perspective, which has different constraints and it's not so useful.

It's not useful almost at all. I'd say from a developer perspective or very limited, useful usefulness perspective. If OVH are giving that data, hopefully it's providing it at a level of granularity, which is more useful to software practitioners, but I would then question whether or not it's the kind of data which would be approved from a regulator perspective to using GHG calculations or something.

Anne Currie: So I think the SkyScanner experience is interesting in that the payoff of moving to spots is financially very impactful. So it's, yeah, in some ways, There's just tweaking and tuning a little bit, but if you make a radical change, sometimes that is quite apparent.

Asim Hussain: It's gonna make it very easy to compare and contrast versus OVH isn't it?

Chris Adams: So there's one thing I will just move on, just share one more link for us to maybe discuss the other time. There's a unimaginably titled Green Coding in Berlin, have a project called Cloud Energy, where they have basically put together some of these modeled ideas for this. Basically, look at this, the machine you're using, make some inferences about the provider, the number of calls and everything you have.

To give you some numbers for this and, uh, I think they're doing a bunch of really interesting stuff in this field and they're doing it with a very open license. So open that maybe companies might not wanna touch , they're going AGPL rather than GPL. And uh, some people were okay with that. Some people aren't, they treat it like it's radioactive.

Let's move on to the next story because Asim, we were talking a little bit about the fund that was Web 3.0, and this seems to be a thing that you might have some reckon on. Nori launches a Web three marketplace for offsets. So the story here is that Nori a US-based carbon removal marketplace. They've launched a Web 3.0 Marketplace specifically to essentially our people, what they call nori, carbon removal tons, so not NFTs, NRTs. And this seems to be something set up to allow you to purchase an offset and for want of a word, retire, make sure it can't be traded to someone else. That's the idea behind it, and I think this is partly in this has been shared because the Linux Foundation had a recent report on open source sustainable blockchains and, and I think you had some reckon about some of this that I might give some space to you or Asim can come in as the good cop to your immediate, your inevitable things to say here.

Anne Currie: Yeah, cuz I would say that blockchain is so horribly tainted with general evil and terrible behavior that I would not want to touch anything associated with it with a 10 foot pole. We were talking last week about greenwashing, and to be honest, this is an awful lot in the financial press about ESGs being effectively ethics washing.

This feels like an attempt to, ethicswash some aspect of blockchain, but I think there's the time. I think that the stable doors has truly is open on that one.

Chris Adams: Has been closed after it's been bolted. Yeah. Yeah.

Asim Hussain: Well, oh man. This whole space is so complicated because it also gets tainted with the whole blockchain emits lots of emissions aspect of as well, and it also gets tainted with, it's just a wild west of people getting money, getting stolen, left. I got lost in the rabbit hole of watching CoffeeZilla on YouTube.

He's this internet detector who just rips into all these NFT scams out there in the world. But one thing I did actually interview the founder of Nori for an old podcast I used to have called The Climate Fix like quite a few years ago, like before I even really knew what an NFT was, he was describing to me this really complex thing and I was like, oh, I, oh, he was explaining it to me, but it was linked to what you were saying, Chris, earlier on, he introduced me to this whole idea. I was actually unaware of the fact that these carbon offset credits can actually sit as a financial asset in your books and be traded, which, which was shocking to me.

So I could buy some credits this year, claim those credits, offset my carbon emissions this year. Next year sell the same credits to Anne, who could then claim those offsets for next year. But no one that goes historically back into my time and then says, well actually you don't have those credits for last year's.

And so what they were doing with the, I'm not not going into the Web 3.0, but the blockchain aspect of their work was the ability to verify that a credit has been retired, and therefore cannot be traded against. I would buy it and retire it. I then can't sell it. I can buy it, not retire it, sell it to Anne, but if I buy it and retire it, claim those credits against my emissions this year, I then can't sell it again.

It's my understanding of how it works.

Anne Currie: That seems like a perfectly reasonable idea, except that as is always the case with these blockchain things, there are other ways, other cheaper, and it doesn't have to be decentralized. Sometimes centralization is okay. If you had an authority that was saying yes, that is an okay carbon credit, which arguably we should have, we should have some kind of better standardization of carbon credit.

Chris Adams: When I read this, the thing I don't quite follow for some reason. So this is where the ultimate source of truth is for lots of these, because as I understand it, blockchains are good for making sure that if you've put something, you can be sure that it hasn't changed because you have this kind of chain of kind of custody.

You can see what the integrity of what might go all the way back, but in many cases, The actual kind of carbon removal parts are usually issued by a government somewhere. So you're not really trusting a blockchain, you're trusting a government. But what you're doing is, so in many cases you, there are a number of schemes where this has actually been issued.

You have, there is some organization which is issuing these kinds of credits, and then they might be put into a blockchain to then be traded around. That means that I haven't actually got rid of the problem of trust. I'm still trusting an external organization, and I haven't really seen a really convincing way around that when I've seen any kind of blockchain related tool.

I have to be honest, when I've looked at this, I'm still not totally sure who is actually, I'm trusting at this point here, and that feels like the thing that's ultimately worth looking into. And if you are interested in this, there is a report. The report is actually that this came from, it's actually not a bad word, it's not a bad read to be honest.

Asim Hussain: Yeah.

Anne Currie: I think it's unnecessary in this case to do that. You don't need it to be done that way. But the classic example of effective use of immutable ledgers is certificate revocation. Now in there you have a trust or authority, which is the authority that issued the certificate that then sits on the immutable ledger.

And when it's revoked, everybody can see straight away that it's been revoked. And it means that nobody, for example, the people who no one trusts, which his governments in this case can't go in and arse around with that certificate without anybody knowing that, that that's happened. So you've got that transparency, you've got the immutability, but you still have the trust of the person.

You've still got trust in that system, which is the person who issued the certificate, but you got protection from the person you are, you don't trust, which is the government. So you're right, but in this case, Chris, yeah, the government's in the chain . So what's the point?

Asim Hussain: I would say one statement, which is when I did interview the founder of Nori a couple years ago, at the end of my interview, I felt more convinced that they were on the right side. I just can't remember all the points right now. And I would probably link it into the show notes here. They were doing this before there was even a language around this kind of stuff.

So they, and I do remember specifically, there was a time I was like, oh my God. Yeah, another climate startup that's using blockchain. Are you just using the keywords to get the funding? And I know that was a feeling for a while as well, and we were having lots of conversations with people and then you dig into, you're like, this could be a database. This could be a database. This doesn't need to be block-. But then there were other time, and I think when I had that interview them, I was like, I remember feeling afterwards, okay, I see the argument for why this is a, this particular use case was important. So I will put that caveat out there. But then I'd also say, you should be asking me too much about blockchain.

Chris Adams: In fact,. we can actually ask them. The folks at Nori maintain a really good, quite an informative podcast. If you are particularly interested in this field, then it's common has been going for a good few years and I actually was introduced to a bunch of quite interesting ideas when I found it. Like the carbon takeback policy, which was an idea related to oil and gas firms.

The idea being that, okay, if oil and gas firm is gonna get some fossil fuels out the ground, they should be responsible for putting it back in the ground. Like they're able to make some money in the meantime, but they need to take on the actual responsibility to put it back if they're gonna do that. And that's the only circumstances by which they're allowed to work.

Now you have an immediate reaction to this.

Anne Currie: Well, I, it is an interesting point to make because I was saying that's the classic only reasonable use for an immutable blockchain is to protect from state level actors like the American government arseing around,

Asim Hussain: That's the,

Anne Currie: it's basically there to protect, protect people from the Amer, from American government, asking around with the.

Chris Adams: I thought the American government was there to bail you out when your bank fails. Yeah, I thought that's what you're supposed to do when you're with your libertarian projects.

Anne Currie: I don't think it's controversial. I think everybody knows that immutable blockchain in certificate revocation exists to make sure that no one, no matter how powerful i.e. The US government can ask anyone to muck around with a certificate, because if they do, it'll be obvious. It's people tying their own hands against they're states, people who might be putting pressure on them. Now, obviously states are state level actors, but to a certain extent there are fossil fuel companies that have so much money that they are effectively state level actors, and in fact, some the biggest ones are almost. Indistinguishable from state level actors.

So to a certain extent I can see that argument that you might need something super tough to protect everybody from an immutable ledger of some form, not necessarily blockchain, which is not one of the better ones to protect against some fossil fuel companies who have a load of power.

Asim Hussain: because if it was a database somewhere, if you had enough power, you can get it changed.

Anne Currie: indeed you can. You've got money to bribe people, to threaten people to do all kinds of things. The value of immutable blockchain immutable ledger for certificate revocation is that you can do it, you can change it, but everyone will be able to see, and I can see potentially that there might be something there where if you're protecting against fossil fuel companies.

Asim Hussain: So I just wanna say one thing on the next related article to this, which is why the Future Web three needs open source sustainable blockchains, which is a report from Linux Foundation Research, which is my colleagues, and Tamara, who's actually joined the foundation, was involved in this as well. One of the specific conclusions in the last paragraph, I think is relates to what we're saying, which is there was a deep need for standards and regulation for Web 3.0, the sustainable blockchain space needs standard transparency and accountability, especially regarding claims about carbon offsets and other efforts to benefit marginalized communities. ESG regulations, making carbon measurement reporting more important, and standardization regulation will help build and shape the Web 3.0 Space.

I'm all for standardization and I'm all for regulations, so this is, this is a good conclusion.

Chris Adams: Last word on the report, which is really surprising to me, but a nice surprise. This is the first report we've ever seen talking about Web 3.0 That actually talks about climate justice and actually talks about the different kinds. Of climate justice in there for actually put bringing that in. She talks about distributive justice, as in who's actually getting the benefits and the downsides.

Procedural justice, who's actually getting to make the decisions, and then rec recognition, who gets to say their cultures are valid or not valid? So to see that in a report, Quite a techy report. Wow. This is some kind of STS level stuff, like actually in actual interesting humanity stuff being added into there.

So it's not just total tech solutionism in in this and it's probably worth having a quick look if you are new to this and you're looking around there. Alright, we're just coming up to 10 minutes, uh, left in this. So we might wanna talk about some of the coming events actually, if you folks are okay with that.

Anne Currie: Mm-hmm.

Chris Adams: All right, so I see there's two things actually. There is QCon London Software Conference, uh, at the end of the month.

Anne Currie: Yeah, so I've helped organize this track cuz I've been involved QCon for years and years and it'll be hosted by Sarah Hsu, who is one of the key members of the Green Software Foundation. And our opening speaker will be Sara Bergman, who is another one of our key GSF folk. We'll also have. Holly Cummins, who used to be at IBM and is currently working on Java, I think Efficiency.

Adrian Cockcroft will have a little bit from the financial sector. Goldman Sachs are gonna come and talk. Basically it should be a really very interesting track. So if anybody who's at QCon London come along to the track, which will be on the last day on the Wednesday, and say hi to me and Sarah and Sara, we'll be very pleased to see.

Chris Adams: Excellence. That's March the. To March the 29th taking place in London, and there is a link to the conference that's taking place if you are prepared to walk out your house and go to an event without a mask on. That's it for our news and events roundup for this week. As part of this format, we have a closing question now that we ask our guests.

It's gonna be different every week, but this week the question is, if you could only use one sustainable software tool for the rest of your life, what would it be? And why?

Asim Hussain: Is this a tool that exists or doesn't exist because most of them don't exist.

Chris Adams: Come up with the perfect, yeah, the dream. What's your dream tool Anne? Doesn't have to exist.

Anne Currie: my dream tool is easy. It's obvious. I would pick this. hands down every time, and that tool is a whiteboard that when it comes to being green and doing any kind of stuff, it's design that really makes a difference. Can you come up with a green design that will totally swamp any improvements you make in an code efficiency or anything like that?

Operationally, what's your design? Do your thinking upfront and share it and discuss it and come up with the design of your system that will be greener. Yeah. Whiteboard always my favorite tool.

Chris Adams: Okay, there we have it. Whiteboard is the number one sustainable software tool for Anne. Currie. How about you Asim?

Asim Hussain: Do know what I was imagining in my head as I was talking? I was imagining like a whiteboard and actually touring your architecture on it. The whiteboard goes, oh, you sure mate? It's gonna cost ya, kind of whiteboard. That'd be a great whiteboard

Anne Currie: That would be a great

Chris Adams: Asim software tool is a whiteboard with a cockney accent like a taxi driver, giving you commentary as you design your system.

Asim Hussain: Like a sustainable software GPT.

Anne Currie: Yeah. Sucks its teeth.

Asim Hussain: Surface plugin for Yeah. Like, ah, I wouldn't, ah, I wouldn't pick that choice. Uh, yeah, I don't know. I think we're still really early days in, in any tooling in this space. I know there are some great tooling and I, and I spoke about earlier on in this episode.

There's the Eco Grader, which came from MightyBytes, I think. Oh, there's ones the Green, Web, Foundation makes as well, and there's this tool in which you can just put in like the URL of a website and it gives you a score. And I think that would be, I've seen the impact of stuff like that. The almost immediate feedback you get from that thing.

And in fact that's what measuring, so we need much, much better tools for measuring, like you mentioned stuff again, Chris, you mentioned earlier on, there's the Firefox tooling as well. There's lots, I think of really great toolings available right now for the Web developer, the front end experience primarily of our website.

There's very limited tooling for everybody else, and I think that this is what we need to get to. I think the next stage in all of this is just much better energy measurement tooling that just are various solutions out there. I want there to be an out of the box solution that works for everybody everywhere.

Whether you're in a virtualized environment where you do your bare metal, it will work if, even if you're running on a hyperscaler, whatever it is, it just gives you the numbers. And we are working on something like that in the foundation called the Carbon QL Project, which is just getting kicked off. But I'm quite excited for where that might head to in the future, but

Chris Adams: Oh, you can't tease us with that. Just at the end of the show, Asim.

Asim Hussain: Well it's just, there's only been two meetings on the project. I can't re-announce too

Chris Adams: so, so what? That's called carbon ql, and that'd be something we discussed at a future date then? Yes.

Asim Hussain: Carbon QL Graph QL.

Chris Adams: And I think that's everything for this episode of this Week in Green Software or The Week. in Green, Software. Cause we're not sure at what point this week becomes next week or last week. All the resources for this episode and more about the Green, Software Foundation are in the show description below.

If you're visiting this on the Green Software Foundation website, the website for the Green Software Foundation. Green software, one word foundation. And if you wanna go to the podcast, you can go to podcast.greensoftware.foundation. If you've enjoyed the show, please consider leaving a five star review on Spotify or Apple Podcasts, wherever you get your podcast.

And as ever, if you didn't really enjoy it, please leave a five star review. But tell us why you didn't want, why you wanted to give us a lower score and respond, and we'll try harder next time. Your feedback once again is very valuable. So please do comment on this and uh, once again, thanks for listening and see you on the next episode.

Tira!

Asim Hussain: Awesome. Thanks all. Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation In any browser. Thanks again and see you in the next episode.

The Week in Green Software: Greenwashing

8 mars 2023 | 33 min

In this the latest episode of The Week in Green Software, Chris Adams is joined by first time Environment Variables guest Tammy McClellan and regulars Anne Currie and Asim Hussain. They discuss the concept of greenwashing; what it is and how companies can avoid it, and why green IT is no longer an option for the tech sector. They cover various statistics about the environmental impact of data centers and cloud computing, the importance of optimizing code and algorithms to reduce emissions, and how developers can’t just rely on hardware to reduce emissions. The hosts also touch on some valuable resources to further your knowledge in the world of Green Software - links below!

Learn more about our people:

Chris Adams: LinkedIn / GitHub / Website
Anne Currie: LinkedIn / Website
Asim Hussain: LinkedIn / Twitter
Tammy McLellan: LinkedIn

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Greenwashing: The Red Flag of Sustainability / Intel [4:01]
Green IT Is No Longer An Option For The Tech Sector / Forbes [13:46]
Everything is moving to the cloud. But how green is it, really? / ZDNET [22:01]
The Oversight Committee has been launched! / GSF [28:21]

Resources:

The SBTI [11:24]
ZeroTracker.Net [13:24]
Does not compute: Avoiding pitfalls in assessing the Internet’s energy and carbon impacts / John Koomey [17:21]
New WC3 Database of References / WC3 Sustainability Group [32:02]
The Transformation Impact of the Cloud (2016) / 451 Research [23:00]
There’s plenty of room at the Top: What will drive computer performance after Moore’s law? / Journal of Science [26:43]
Green Software Practitioner Course from Linux / GSF [31:13]
Green Software Foundation Software Carbon Intensity Specification Guide / GSF [31:31]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Asim Hussain: 80% of people don't trust corporate messaging is because I have a belief that members of the public trust nonprofits more than they trust for-profits, and that organizations like the GSF would gain more trust from people than like a for-profit company. And so sometimes I feel the problem is that organizations are trying to market their own thing instead of just aligning to like our commitments or SBTI approved tick!

Okay, everyone trusts you now, rather than, I'm going to try and explain my specific version of my climate target in the way that sells my products the best and shows me the most differentiators.

I'm your host, Chris Adams.

Welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I'm your host, Chris Adams, and today we're discussing the growing issue of greenwashing in tech and the importance of reporting and communicating corporate sustainability accurately.

Before we dive in, let me introduce my esteemed guests and colleagues with this episode of The Week in Green Software. Today we have Tammy. Hello, Tammy.

Tammy McClellan: Hi there.

Chris Adams: Asim

Asim Hussain: Hi.

Chris Adams: and Anne.

Anne Currie: Hello.

Chris Adams: And if you're not on first name terms, let's do proper introductions. First of all, I'll hand over to, um, Anne. You'll be the first person to introduce.

Anne Currie: So my name's Anne Currie. I've been in the tech industry for nearly three decades, which is quite depressing and good. And good, depressing, but but also good. And I am part of the Green Software Foundation along with everybody else here. So today I've been working on the introductory chapter of a new book on green software, which we can't yet talk about who the publisher is, but it's a good publisher.

Next time I'll be able to tell you who it is.

Chris Adams: thank you for that, Anne. All right. Next in the alphabetical order would be Asim Go for it.

Asim Hussain: Hi, I'm Asim Hussain. I am the executive director and chairperson of the Green Software Foundation. I'm also the director of Green Software at Intel. I'm excited to be here. I also do, I also, oh, I grow mushrooms, which is an active hobby at this time of the year, so

Tammy McClellan: No way!

Asim Hussain: Growing. Oh yeah. Have we not told you about that, Tammy? Yeah. Spring time's coming. So we're getting, as you are, I imagine, Tammy getting ready for a growing season. But anyway, I'll let you introduce yourself now.

Tammy McClellan: Sure. Hello everyone. I'm Tammy McClellan. I work for Microsoft. I'm a cloud solution architect, developer advocate. So along with Anne, I've been in the tech industry for a really long time. I think I probably just passed the 30 year mark. I'm also the co-chair of the community working group along with Anne here, and, uh, recently became the chair of the oversight committee.

So I'm super excited about that. And I also, as Asim alluded to, I have a small sustainable farm here in Chelsea, Michigan, where I grow lots of veggies and flowers. So happy to be here.

Chris Adams: Cool. Thank you.

Asim Hussain: What's the name of your farm? Tammy. What's the name of your farm?

Tammy McClellan: It's called Wonderful Tiny Farm or WTF for

Chris Adams: Sure

Asim Hussain: wtf

Chris Adams: That's very good. That's a, I think that's an appropriate lead in for the stuff we'll be talking about actually. Alright. All right. Before we start, I'm just gonna share a reminder for anyone who's listening, everything we do talk about will be linked in the show notes. So there'll be a link for every single story we do cover, plus some supplementary links for the things that do come up as we scrabble around to try to find them as we discuss them.

And with that, I think we'll start on the first story on our list, actually. So, What I have here is greenwashing, the red flag of sustainability. This is a piece from the Intel Podcast put together by Intel called the Intech Technology Podcast. They recently published an environmentally focused episode with I think Caryn Herder Fritz, one of the marketing initiatives sustainability initiative leads.

Speaking about how companies can avoid some of this greenwashing explaining what greenwash actually is. So, I suppose now is a good time as any, to answer the question, what is greenwashing here? Anyone want to go forward or volunteer something here?

Asim Hussain: So I actually work with Caryn. It was really exciting to hear her talk about this. It was interesting cuz I actually have a slightly different definition of greenwashing as well, personally, which is interesting. But she was describing greenwashing as a term to use to describe when companies make misleading or false claims about the environmental benefits of their products or activities in order to appeal to consumers who are concerned about sustainability.

So that's the definition that she gave about greenwashing. I'm actually interested to hear what, how do other people feel about that definition? Is that aligned with your thinking?

Anne Currie: That sounds pretty good, but Asim, you, you feel different, you say,

Chris Adams: What's your,

Asim Hussain: What a leading.

Anne Currie: so we who we're agreeing, disagreeing.

Chris Adams: what's your, are

Asim Hussain: or incorrect? According to my. My definition of greenwashing? No. I would also, cause I like, I don't know, I think about it a lot. I actually think about it a lot. I think about, I'm in corporate industry, so I think about it very carefully. I think about the messaging that I'm we're giving out.

What does it mean to be authentic? And one thing I haven't quite resolved in my head is the intention behind the work. And I think Anne, you've said something about this in the past, so if you did something without the intention of it being sustainable and then afterwards went, oh, actually, if we look at it through this lens, is it sustainable?

Let's talk about it from a sustainability perspective. I dunno where that lands

Anne Currie: I'm okay with that.

Asim Hussain: of

Anne Currie: I'm more of a consequentialist . I'm fine with people saying, Ooh, it was good. I'll talk about it being good.

Asim Hussain: Accident. Oh damn. I did something good by accident. Yeah.

Chris Adams: There are two things which are really interesting at the podcast that caught my eye, so first of all, one of them was, so there was a stat shared from some UK surveys, I forget this particular poll came from, but they basically said 81% of comm's messages about environmental measures from tech firms are not trusted by their audiences at present.

So this is one of the key things that she was saying was, This is a real problem that needs to be addressed right now. The other thing that came out of this, and I guess putting the question to Asim is quite an interesting one, is the genuine idea that using the word green itself is a bit of a red flag because it's so wooly and so kind of open.

I think Asim, I think you are about to speak to this particular as the director of the Green Software Foundation.

Asim Hussain: And what are you the director of again Chris?

Chris Adams: I'm a member of the Green Web Foundation, so I may also have things to say here as well, but after you first.

Asim Hussain: No, I think that was a really interesting insight. Now, the point she was saying was because greenwashing has the term green in it, and if you then just use the term green, it just reminds you instantly of the word greenwashing before you've said anything. So already frames. Like really pulls up that thing in their mind.

I was like, damn. Where were you Caryn? Like, where were you like maybe four years ago when we were like coming up with the name for this thing. But I would also argue a different point because I tell you that when we were coming up with the name of this foundation, and I've apologized to Chris Adams like so many times for this because I knew Chris and he had the Green Web Foundation, and the very first name that came to my head was the Green Software Foundation.

I always intended it to be like a pinhole name, and then we'd come back to it later and we never came back to it and they just ended up being the Green, Software, Foundation. But the other thing was, there was actually another organization called Sustainable Software or Sustainable Software Foundation. There's actually another organization out there, but they're much more focused around, can you, as a developer, can you, on a human level, sustain.

Chris Adams: Yeah, the Software Sustainability Institute, can you keep things going? So it's nothing to do with climate at all, it's just am I able to keep working without code collapsing under its own weight from like bugs and issues and usability problems and stuff like that.

Asim Hussain: So that's where the idea, it was like you either pick sustainable or you pick green. In my mind, I personally felt green was a bit more targeted than sustainable. That's why we went for the green in the first place. So that was my thinking on it.

Chris Adams: Okay. That's fair.

Anne Currie: Well, it's interesting, so the, the book that I'm currently working on, I originally pitched. Sustainable Software, but the publishers chose the name Building Green software and presumably that's because they're much better at marketing than I am . They obviously feel that green is the word that people want to be using or be interested in.

So that is to a certain extent, then that becomes greenwashing. Cause everybody likes the phrase, but everything that's more specific, I really like 24 7 carbon free electricity, for example, but it's very boring and it's quite specific, and it does not really get to appeal to folk in the same way. So

Chris Adams: Actually, I'm kind of glad you mentioned this Anne, cuz this was the other thing that Karen was saying. So she said there's green washing and there's gray washing where you go so far away from emotional and evocative images. The unit was something which is accurate, but basically impossible to get anyone to remember or respond to in any kind of meaningful fashion.

So I think that's actually, I've never heard it come across the term graywashing before or anything like that, but that's caught my eye. The thing that it might be worth actually talking about in this context is that, and Asim, you touched on this idea of is it intentional or unintentional? You can see parallels right now with the basically misinformation and disinfo discussion online right now.

Because one of the big problems about the internet, which is not necessarily being fixed by things like generative AI search engines, is that you have a real problem with it being very difficult to find reliable information online. All right, and in those circles, people call things misinformation, where you're unintentionally misleading people or disinformation if you are intentionally misleading people.

And like the kind of mental model that I've been using for this is, it's a bit like murder and manslaughter. You know, manslaughters, I don't intend to cause harm, but it's happened. Whereas murder is very much like a degree of intentionality. This is actually part of it. You might wanna think about where in your organization,

this kind of comms function might actually be alright if they're in finance compared to marketing, you're gonna have different drivers, but there's plenty we can refer to there. And I suspect the thing that might be worth looking at is that there are various kind of non-profit organizations who do try to keep track of all this stuff.

And one of the things that, if we could talk a little about, say some of the things that companies have, but, so I work for a nonprofit called the Green Web Foundation. We did a whole thing about net zero targets and uh, you can. There are some ways to tell if you have a good net zero target or a bad net zero target based on the kind of organizational changes you might need to see happen.

So if someone has a very far off net zero target, for example, where there's no meaningful action, the hat needs to happen in the next five, five years

Asim Hussain: We'll get it done by 2050. We'll, 2050.

Chris Adams: If you have that, then it suggests that maybe you're not prioritizing it. And the reports that from groups like say the corporate climate responsibility monitor and stuff like that, they basically say you need to have a net zero target by 2030, and sorry, you need to have something with interim actions in the next five years for this to seem meaningful.

Asim Hussain: Did any of you read? I don't think, we didn't appear in the last news that I, it might been one of you that posted it on, on, so I can't remember, but it was something somebody shared about, oh, it was cdp. It was, it was a report they had done about, if you are an organization that has a climate target that is a, an SBTI what's the term?

Chris Adams: Science based target? Is that what you're referring to here?

Asim Hussain: Is it is if your climate targets have been vetted by the SBTI? I think that's a term you are far more likely to over overdeliver on your climate achievements than if you haven't. And I think I wondered, cause one of, one of the foundation was starting like one eighty percent of people don't trust corporate messaging is because I have a belief that members of the public trust nonprofits more than they trust for-profits and that organizations like the GSF would gain more trust from people than like a for-profit company. And so sometimes I feel the problem is that organizations are trying to market their own thing instead of just aligning to like our commitments or SBTI approved. Tick. Okay, everyone trusts you now.

rather than, I'm going to try and explain my specific version of my climate target in the way that sells my products the best and shows me the most differentiators.

Chris Adams: Brief sidebar for folks who might not be familiar with SBTI. The SBTI stands for the Science-Based Targets Institute. They're a group of peer-reviewed scientists and and experts who look at various sectors to figure out what kind of changes and reductions in carbon emissions you'd actually need to see on a year by year basis in order to actually be responding in line with the climate.

They do work for various sectors, but specifically in 2020, they released information about the tech sector. So they basically said, you need to be hitting these targets for your actions to be considered credible. That's all. Sidebar over. That might be useful for folks who might not know what the SBTI is, cause we should have actually come in with that one.

Okay. There's a bit more here we could talk about. And there's a link here to zerotracker.net, which does track some of these targets and some of these actions by organizations. And there's even one pointed to specific companies. So you can see are they recording and are they reporting against these kind of figures that you've listed here.

But I suspect we might need to move on to some of the other stories we have if we wanna go on from here. So what's next on this list?

Anne Currie: So this is an article from Forbes. Green IT is no longer an option for the tech sector, although I think I would've called it no longer optional for the tech sector, otherwise it feels a little bit like it's saying the very opposite of what I think it's trying to say. But anyway, yeah. So a few weeks ago we did talk about, uh, some interesting and quite terrifying statistics that the cloud considered over 7.2 million data centers across the world, which actually is, that suggests that there's about a data center per thousand people in the world, which seems like the, the hardware utilization on data centers must be really bad for that. That's quite scary, although not totally implausible. If you think there's maybe about 7 million businesses across the world and each one has a, at least a couple of servers in there,

Chris Adams: I feel like you need a loose definition of data center here for that to be plausible, right?

Anne Currie: Yeah. I think it would have to be. I think it'd have to be, but even with that, actually, that really suggests cuz a thousand users. And these are not simultaneous users, but as Asim pointed out last time, these days everybody's pretty much connected all the time. So fundamentally far, then talking about a thousand simultaneously connected users.

But anyway, so we've got a cloud of 7.2 million data centers, one data center per thousand people, loads of energy and water and all that embodied carbon. Or embedded carbon, depending on how you like to say it's involved in the hardware and that cloud computing is responsible for broadly a percent of global greenhouse gas emissions, which now says it is about twice as much as the whole aviation sector potentially.

But broadly speaking, I always like to say it's about an aviation sector from the tech industry, which gives us some feeling and likely to surge, but might not surge because we get better at it. But fundamentally, we use computers for more and more. So those are the numbers that we're talking about.

Asim Hussain: So I think this article in particular, I liked it cuz it also touched on software. It didn't just say, data center's bad, go fix it, data center people. It was like, no, no. The one statement that I really liked was organizations of risk developing software that will run hot unnecessarily for many years if they do not improve the sustainability of their software today.

And for me, when I think about building inefficient applications, like I forget, like it's just gonna run like that forever. 10 years. Some of this stuff just runs forever, unnecessarily, and then just dies. I remember the, like, I won't name it the company I used to work at prior to working at Microsoft. It was a very small startup and the technology was so unbelievably inefficient.

We needed to buy one server for every 10 users.

Tammy McClellan: Oh my goodness,

Asim Hussain: but well. It was high net worth lending to high net worth, and it was very high profit business. They didn't care. But the, the software was built 20 years ago, and so for 20 years they've been running this unbelievably unnecessarily complex software.

And as far as I know, it's still running today. I haven't checked in on 'em for a while. I think that's really interesting, like when you write software. It will run unnecessarily hot for a long period of time, whereas a data center in harder will refresh. That was a really interesting insight for me. Yeah.

Tammy McClellan: Oh, good point.

Chris Adams: This provides a link to the next story, which is about moving things to the cloud. And there's points and counterpoints for that. Specifically. There's one thing I might share with everyone before we move on from this though. When you see numbers like say 1.5% moving to 15%, there's a really good paper by John Koomey, a well-known professor who basically has, there's a paper called Does Not Compute, avoiding Pitfalls in Assessing the Internet's energy and carbon impacts.

This is the guy who's been studying this for 30 years and generally speaking, if you look at our sector, People saying, oh, it's 2% now, but it'd be 15% by this time in the future. Aviation says that shipping says that every single sector, which is 2%, says they're gonna be 15% in the future. This is a recurring thing and it's really worth reading that to be able to interrogate some of these claims, cuz they can't all be 15% for this to be happening.

Anne Currie: There's a tacit assumption there that nobody else grows, as she said. But that's not a crazy tacit assumption that you are making your point that the point is gonna get bigger. It's already big and it's gonna get bigger. The fact that everybody else is getting bigger doesn't make your problem lesser.

And Dr. Koomey there, he has picked me up on this stuff in the past and said, oh, you can't say that this is going to continue. But I think there is a point here, to be made that, that we don't want these things to go up. The relative count doesn't really matter. It's, and, and it makes it, it makes a good point that we get more efficient as time goes, but we don't actually always get more efficient as time goes on.

Data centers get more efficient. Cloud data centers get more efficient. We use more of them. He has a platform which is all about, oh yeah, everything gets better and it'll all stay about the same, but it, you tends to use it to shut down people saying we should do better is my opinion.

Chris Adams: And maybe we should come up with a law for you, whereas Koomey's Law, maybe we need a Currie's Law as a counterpoint for this actually.

Asim Hussain: Everything will get worse all the time until the heat death of the universe.

Anne Currie: Actually, the thing is we do have a tendency to, over my entire career, 30 year career that Tammy and I have had. Hardware's got tons better, but utilization has been sacrificed to developer productivity. So machine productivity, no one cares developer productivity. Everybody focuses entirely on that, so we tend to move in the wrong direction, which then takes you down back to Asim's point that you end up with very inefficient software that could be a lot more efficient.

Asim Hussain: That's a really good point. Cause you're right and I have been using. I, I feel very much corrected cuz I have been using that percentage relativity. But that whole statement about relative increases is pointless because if everybody is saying it's going to increase by 15%, that also isn't a good thing.

It just means that. Maybe we should be talking in absolute terms of increases. Maybe that's the kind of statement. But I would also state that I think a lot of these things assume, like currently we are pressing 10% down on the accelerator pedal, and the statement is, if you keep pressing on the accelerator pedal, 10% you'll end up at this point in the future.

Doesn't mean you can take your foot off the pedal. If you take the foot off the pedal, then it just all just goes crazy. So I think it's important to note that doesn't just mean, even if, even with that statement of it's not going to be as bad in the future, I think that statement should be, it's not gonna be as bad in the future if we continue to put the pressure on that we're putting on.

It may not be that bad in the future, but if we just sit back and relax and say, apparently somebody says it's all gonna be fire in the future, then it won't be fine. I think it's about forces. We have to make sure we keep the force pressed to make sure that good actions happen. That's just one of our points.

Anne Currie: And so I remember when I see these statistics about the aviation industry, I remember about a talk I gave at a conference HashiCorp Europe in 2016, and I was the first talk I gave on green software. And I thought, oh, I need a really good statistic. I need a good statistic. So I had a look. And I would say it looked like the aviation industry.

We used about twice as much as energy. Maybe a bit more, but I thought, oh, I'm just gonna say it's the same, cuz then that's fine. I'm sure it'll get there eventually. But now everybody's talking about it being twice as much as the aviation industry. Things do. THe tech industry has got worse over the past five years in terms of carbon emissions.

No matter what Professor Koomey says.

Tammy McClellan: I'm just surprised you remembered what you were doing in 2016, so I'm just impressed by that.

Chris Adams: [laughing]

Anne Currie: So I I always felt a bit guilty cause it was a bit of a lie at the time, but I thought it's a bit of a lie, but it's going to happen, I'm pretty sure. So, you know,

Asim Hussain: At some point in the future or past it's true.

Chris Adams: I guess the good news we can talk about is that the technology sector is probably easier to decarbonize than aviation because servers don't need to fly through the sky all that often, and that's a nice link to the next story we have. Everything is moving to the cloud, but how green is it really?

Anne Currie: Yes, there were a couple of papers came out last week. One from Adrian Cockcroft, who is the member of the GSF, and there's a bit of an insider on this as well as a slightly more outside perspective saying that, and they made the interesting point, although the cloud is getting better from a very low bar, for some of them they have not been as good at helping their customers to become green as they said that they would be, or at least Amazon hasn't been. Google has been doing pretty well. Azure been doing okay, but AWS has really fallen behind and that is a, an opportunity for everybody because AWS. Amazon care what customers ask for.

So if you ask for it, you might well get it. And if they're not doing it, that might be a sign that people aren't asking for it. We need all need to make sure that we ask our AWS reps all the time for cloud carbon footprint measurement.

Chris Adams: So that was one of the papers. You mentioned there was another paper and so there's a piece in computing.com that I've shed a link to that pointed to basically some really detailed stats talking about the environmental impact of refresh rates on servers and stuff like that. This is actually worth being aware of because a significant amount of the environmental impact comes from actually making the servers in the first place and.

Asim as the guy at Intel, you probably have some insight on this one now, like there's a significant amount to making them and that part isn't particularly easy to decarbonize compared to the actual running of those. And this is the first paper I've seen, which basically challenges some of this narrative cuz this the reports that you do see that talk about the cloud.

Generally, like one of them is by 451 Research, which was commissioned by Amazon. So unsurprisingly, they say that Amazon's super efficient, but you'll see this are coming up quite a few times. It's quite hard to get some independently confirmed information from this, but this one seems to be more about where energy is coming from and how it's being sourced actually. That was my takeaway when I read through this actually.

Asim Hussain: Speaking to the point you just mentioned. Chris, it does surface an interesting stat. We're just gonna be like, maybe this podcast will just turn into one of us mentioning a stat and everybody else disagreeing with it. a, a well-researched stat from a very famous researcher and we just like that doesn't sound right, but the research seems that, that they say, the research indicates the energy consumption from data centers grew just 6% between 2010 and 2018.

well, its computing output increased 550%. And that I think speaks to what you were, you and Koomey was probably talking about, which is the computing industry is the efficiency has been increasing dramatically.

Anne Currie: I agree with you, and there's absolutely tons of efficiency improvements to be wrought from software. The difficulty there is that this quite difficult, takes a lot of time to do it, and this is my total stab in the dark guess. I think the next massive efficiency improvement will come from the, ironically, the same efficiency improvement that delivered to the industrial revolution, which is a move from generalists to specialists.

Has a tendency to deliver a thousand fold according to Adam Smith in the Industrial Revolution in his Wealth of Nations of 1775. moving to specialists. Is a thousand fold increase in productivity and performance, and I think that's where I have a dual hope for the cloud cuz that is specialists, that's people putting their homegrown software and saying, no, look, I'm not gonna write this.

I'm just gonna use the cloud stuff where they were specialists doing it for me. And also open source where they say, I'm not gonna write this library. I'm gonna use an open source library where specialists will have tuned this for me. I'm hoping that will to a certain extent, offset the end of hardware in terms of improvement.

But that might be my dream and an unrealistic one.

Chris Adams: Not necessarily. I feel there's a really nice paper called Plenty of Room at the Top, which basically it's a paper from the Journal of Science talking about specifically where the next generation of improvements are gonna be coming from. It basically puts this argument that, yeah, Moore's Law has slowed down over the last decade, so you need to find other ways.

And it says, yeah, the thing you need to be looking for is things like, Domain specific programming, matching the compute jobs or can matching the workloads too. The hardware much better. You can see examples of this right now. We'll share a link to some analysis. For example, Google using very specialized ASICS like application specific integrated circuits for video encoding or tools like that.

These are the things that are being used in production, a number of places which can provide these hundred or sometimes thousand improvements that there is an issue though about where you do if you wanna do something else. Cause we've seen exactly the same thing happening with cryptocurrencies, right? Where you get to an ASIC designed specifically for the Char 256 protocol.

So if you don't wanna do stuff for cryptocurrencies, I guess you might be to use it for your passwords on a website. But there aren't that many other things you can use it for. So there's a discussion there to be had. But no, you're absolutely right. We'll share a link to that cause it's a really interesting paper and it basically makes you an argument that you've described Anne but in lots of detail with those are really nice examples.

So we've spoken about those different tools and there's different ways that you can achieve some of these savings, both in software design and hardware design. And the Green software Foundation is maintaining and running a couple of projects like the Carbon Awareness SDK, the Patterns Project, and so on.

But recently, there've been some developments to how these project projects can be funded, or similar projects can be supported so people can organize and work on them a bit more. Tammy, I think this is something that you've. Been involved in, and I believe, yeah, congratulations in order for actually becoming part of the groups leading this now, maybe you could share a little bit more about the oversight committee for the Green Software Foundation.

Tammy McClellan: Sure would love to. I am super excited about this opportunity when talking to Asim and he basically came up with the idea of this oversight committee and the ability for G S F to scale as we get new members and organizations that are joining and it became pretty apparent, I think, that everything was funneling through Asim.

So this gives us an opportunity to scale at Asim's pace. And so I'm excited about that and being able to look at some of these technical experiments that we're doing and helping to drive some adoption in certain areas. But well, I'm just really excited about the possibility of advocating more in this space cuz I really feel that there's loads of opportunity here for us to make an impact in our overall carbon goals that we have.

And it's emerging tech, so we're flushing it all out. We're figuring it all out, and we're having some great success. So the Oversight Committee will also provide recommendations to improve the foundation's charter to set community norms and workflows and deliver budget recommendations to the steering committee, so folks like myself and Chris Lloyd-Jones will serve as the chair and the vice chair of the Oversight Committee, and hopefully we can get more OC members on this podcast to introduce themselves to our wonderful listeners here.

Chris Adams: Cool. Thank you for that, Tammy.

Asim Hussain: Yeah. Congratulations Tammy.

Chris Adams: Okay. Yeah, that's it for our news wrap up for now. But before we leave, there's a new format for this final part of the show where we are gonna ask a closing question to our guests. It's gonna be different every week, but this one we're gonna try to see what kind of resources or recommendations you might actually point other people to.

So the question is, if you could point one of our listeners to one resource about sustainability and green software. What would it be and why? Anne I'll start with you first if that's okay. Cuz it looks like a sea is holding a child now.

Anne Currie: Indeed. Yes. Oh yeah. One resource. One resource. Actually. So I'm gonna cheat and easily provide the one that everyone will do, which is, that's on the GSF websites available. Now from the Linux Foundation, you can get for free a certification in being a green software engineer, and that's two hours to do.

It's quite an easy read, and at the end you get a certificate of completion.

Chris Adams: Cool. Thanks Anne. Anyone else got one?

Tammy McClellan: I have one, although I'm cheating here as well, but it's the one I find myself going to all the time when I'm having internal discussions or discussions with customers. It's off the Green Software Foundation website, but it's the SCI guide. But it has just loads of guidance around applying the SCI some use cases, and it's just uh packed full of really good information.

Chris Adams: Okay. And the same yourself.

Asim Hussain: So yes, I had, I think I shared it on socials recently as well, so, Recently had a really great conversation with Lucas from the, from various things, but he is also involved in the W3C Sustainability group, and he shared with me just an amazing set of resources that they've been collecting over. I don't believe he was a particular person collecting, but he introduced me to it and will put them in the show notes as well.

There's all sorts of things in there from books to magazines, courses, events, media websites, even stuff from the Green GSF is in there as well. So just a great source of material.

Chris Adams: Cool. Thank you for that, Asim. So for folks who are following along the W3C have a Sustainable Web Group and they maintain a Wiki now, and that's where all that stuff is listed. It's a really good resource and we only found out about it in the last two weeks, but it's a really useful thing to look at.

So that's all for this episode are for The Week in Green Software. All the resources for this episode and more about the Green Software Foundation are in the show description below, and you can visit greensoftware.foundation. That's green software. One word. DOT foundation in your browser. If you enjoyed the show, please consider leaving a review on Spotify, Apple Podcasts or wherever you get your podcast, leaving us a five star review.

If you didn't enjoy the show and you hated it, please consider leaving a five star review and tell us why.

Asim Hussain: We only listen, we only listen to five star reviews. That's the only ones we listen to.

Chris Adams: Yeah, that's the way that, that, that's the . Your feedback is incredibly valuable and helps us reach a wider audience. Thanks again for listening and we'll see you on the next episode. Cheers, folks. Bye Cheers.

Cheer.

Anne Currie: Cheers. Bye-bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's Green Software Foundation in any browser. Thanks again and see you in the next episode.

The Week in Green Software: How Green is Your Cloud?

1 mars 2023 | 51 min

TWiGS returns this week with host Asim Hussain being joined by (now guest) Chris Adams. They talk about the environmental impact of the cloud and while some of the big cloud providers, including Amazon Web Services, Microsoft’s Azure, and Google Cloud, have introduced initiatives designed to increase the sustainability of individual data centres and reduce their overall carbon footprints, will it be enough to help reduce carbon emissions produced by cloud computing? They also cover Microsoft’s Surface Emissions Estimator and a recent paper surveying the factors that influence the emissions of machine learning.

Learn more about our people:

Asim Hussain: LinkedIn / Twitter
Chris Adams: LinkedIn / GitHub / Website

Find out more about the GSF:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

News:

Microsoft launches its Surface Emissions Estimator / Microsoft [2:42]
GreenOps Carbon Footprint Treads Closer To Cloud Developer Efficiency / Forbes [14:35]
How green is your cloud? / TechMonitor [22:11]
Counting Carbon: A Survey of Factors Influencing the Emissions of Machine

Learning / Alexandra Sasha Luccioni & Alex Hernandez-Garcia [36:21]

Ongoing Opportunities to Scale Green Software:

GSF Meetup Opportunities [46:24]
GSF Speakers Bureau

Other resources mentioned:

Ola Fagerström’s announcement about Microsoft Surface Emissions Estimator [13:01]
Cycloids.io [14:35]
Net Zero Tracker [27:55]
Chris Adams & Max Schulze talk Cloud Economics with a spreadsheet [31:02]
WattTime.org

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:
GSF_Ep17_TWiGS5_Transcript

Chris Adams: The flip side of this, maybe the thing they're going for is saying if cloud carbon footprint is just ubiquitous like hoovering is, maybe that's the thing you would just say, well, you're just gonna ccf it, or cloud carbon footprint it.

Asim Hussain: Yeah.

Hello and welcome to Environment Variables, brought to you by the Green Software Foundation. In each episode, we discussed the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Asim Hussain. Welcome to another episode of The Week in Green Software, where we bring you the latest news and updates from the world of sustainable software development. I am your host, Asim Hussain. In this episode, we have some exciting announcements from various organizations, and we're going to cover some interesting articles about the environmental impact of cloud and machine learning.

And finally, we're gonna share some opportunities for development for the world of Green software. But before we dive in, let me introduce my esteemed guest and colleague for this episode of Twigs. With us today, we have Chris Adams.

Chris Adams: Hi Asim. My name is Chris. For folks who've never heard of me before, I work as the one of the chairs of the policy working group at the Green Software Foundation. And I also work as the executive director of the Green Web Foundation, a small, non but fierce non-profit focused on reaching an entirely fossil free internet by 2030.

And I also work with the online community climateaction.tech, which is where I met Asim.

Asim Hussain: Oh, the climateaction.tech days.

Chris Adams: Yeah, that and OMG Climate as well actually.

Asim Hussain: Oh yeah. OMG Climate. I remember the OMG climate. You should start those again. Oh, have you started those again? Actually, I have interest. For the listeners, OMG Climate was a A A A conference, an on a conference that Chris was running.

Chris Adams: Yes, I can talk about this. We can segue into that gracefully in the podcast if you'd like. After we've covered some of the news. But this is something that we would like to do a bit more of, and I’ve got a fun story about why it's called OMG Climate and where some of that came from. So it's also open source.

So if you like the idea of onConferences around climate, then maybe we should leave a bit of time for that actually seen, because there are some kind of green software and digital sustainability related events in the coming months that are probably worth pointing people to. So that was me introduced

Asim Hussain: And before we dive in, just a reminder that everything we talk about will be linked in the show notes below this episode. So I think Chris to kick off uh, news from the world of Green Software, Microsoft Company I used to work for. Microsoft has launched a new tool called the Surface Emissions Estimator that helps customers understand the carbon footprint of the devices purchased.

It is the surface device range of the purchases that are purchased. It uses a lifecycle assessment model to provide accurate estimates of carbon emissions, it also, I believe, helps you adjust those emissions based upon where you are, what you bought, where it's been delivered, and things like that.

Chris Adams: I thought this was pretty cool actually. See, have you had a, have you had a go with it yet at all? Because I don't own a surface thing myself, but I'm happy to talk about it cuz I think this is actually a field that probably has some quite far reaching implications on how people like work within, with gadgets and things at all.

So first of all, you, you shared this link before, so we've, you've seen how it. Basically, when you talk about talks about surface, it shows you like maybe a surface. Are they called laptops or tablets or is there another word for this kind of removable thing? They have?

Asim Hussain: just called surfaces. I don't know. I just think they're Microsoft Devices

Chris Adams: that's confusing.

Asim Hussain: Yeah, but I believe the whole range. Everything is, the tablet and the laptop, they're all called surfaces. Funnily enough, I do have a Windows machine. Now I'm at Intel, but while I was at Microsoft, I had a MacBook my whole journey.

So I, I missed out, I missed out on the whole, the . Yeah, I was almost called a heathen on my first day of, of, of my Microsoft training. But anyway, no, I actually haven't. Cause while I was there, Ola, who's the person behind? Ola Fagerstrom. Hope I pronounce it a second name. . I spoke to him about it and I think I saw early preview design sketches or what it looked like, but I actually haven't tried out the latest tool myself.

Chris Adams: Okay. I could talk a little bit about it cuz I gave it a go before this. And there are some previous, there's like prior art that is actually quite useful to know about. So one of the things it does basically is, let's assume if you were to buy a laptop or a gadget or if you're gonna use the Microsoft parlance a surface, for one of these, it basically shows you.

Where the emissions lie in the actual creation of this. Cause a lot of the time people might think about what bought a machine, but I now to be, need to be really careful about how I use it, for example, because I might assume that all the emissions are from this kind of use phase rather than the making phase.

And the key thing that you learn from it when you were to choose something. See whereabouts you are in the world, how many you might use. If you are like, say, a medium to small, actually, if you basically have, you're working in an organization that's purchased a number of these, it'll tell you what the likely environmental footprint of those is over a particular time.

And then if you held onto them for, say, six years, for example, or three years or four years, it'll show you. What the environmental footprint of that might be to own that and to have it for this time here. And this is quite useful because people didn't really think too much about the embedded emissions in electronics for quite a long time.

It's only in the last year and a half that it's really become much more of a kind of thing that people focus on. People have typically been looking at, say, the energy more than the actual purchasing parts. And, and it's a French company called BOA Vista, which has created, has been collating lots and lots of data from lots of companies about this.

It's a nonprofit. Yeah, and the thing that's interesting there is if you don't have a Microsoft Surface, but you think this is cool, they have similar tools so that you can basically. Pick maybe a Dell laptop or an Apple MacBook or something like that, and then it'll give you some numbers. But the thing that's been a problem has been that some companies have been somewhat reticent about sharing these numbers.

So as a result, people have been either they have to like make guesses or they are do not have particularly useful guidance. So if there's a company sharing this stuff upfront in a kind of structured data fashion, which is necessary for this. And I assume they are sharing it as open data for everyone else to be using, surely Right.

Then that's, it's a good sign for you to see this and it helps you understand how useful it is to just hold onto a device for maybe a little bit longer and see how that might fit into some of your plans to basically reduce the emissions associated with the making part of running any kind of digital services.

Asim Hussain: I love the work of Bovis on there, and the latest version of the API, I was taking a look at. A day they've done wonders, BA based off of the, as you say, like the incredibly limited amount of data there is out there, but there's a lot of extrapolation

Chris Adams: Hmm.

Asim Hussain: the day, like some of the LCA work that they're using is like 10 years old.

Chris Adams: Yes. L C A here stands for lifecycle analysis. Manasi mentioned these things like ISO 14040 and 14044. There's basically complicated methodologies which people use to talk about what the environmental impacts associated with any tools, with building and operating something over its lifecycle. Over its lifecycle.

And LCA is the kind of short term for this.

Asim Hussain: And actually I hadn't seen the parallels myself, but it seems really obvious right now what Microsoft Surface. Estimator Emissions Estimator is, is basically hopefully a more accurate version of what Bar Vista is giving you because it's providing you with that data. Remember the conversations around this originally, becuase it was based around the idea.

If you're an organization with like 10,000 employees and you bought each of 'em a surface laptop, that's a lot of things to keep a track of. And also like those employers are gonna be based in different parts of the world. The laptop's got shipped over from different locations. Did you dive into the tool surface? Whether or not it took that kind of regional variability into account, I dunno where services, let's assume services all get shipped from the US. Would your US employees have less emissions than your European because you're just, the travel is less. Do does it take things of that into account?

Chris Adams: So when I was looking at this, what you could see is you could actually, it does show some information about the carbon intensity of different grids, and it does talk about the end of life part of it, but it doesn't mess it by. I didn't see, I, I didn't see so much specifically about shipping. So if I'm in one part of the world, is there an environmental impact of getting it sent over here?

For example, proportionally that's relatively small, meaning it's not being flown around, which in some cases it actually, unfortunately can be basically. So it doesn't talk about that, but it does tell you what the environmental impact is from the grid itself. So if you are running something in. I don’t know, let's say Pennsylvania, where there's load of coal, it's gonna say that proportionally the use face is gonna be heavier than, say, France or Montreal.

Where like more than 99%, this is Montreal, for example, or like, uh, Quebec. Most of the powers coming from the hydro or nukes. So therefore it's gonna be very relatively low carbon electricity. So it doesn't, does seem, does take that part into account.

Asim Hussain: So there seems to be a very unusual correlation between. , low carbon electricity and speaking French. There must be some research there about

Chris Adams: Honestly, I, this is gonna sound a bit weird, but like a significant part of it is in my view, having a real interest in there being a very strong state. So the entire. Thing about, say France being full of nukes on France like using, is because historically they had massive investment in the seventies and eighties in into nuclear power through the state owned systems, which is not really what you saw in other parts of the world.

And uh, also you've gotta remember that France didn't really have much of a kind of fossil fuel, didn't have much in the reserves, so they chose to have that as their way. Achieving some degree of energy independence. But the thing that when you see lots of people talk about nuclear these days is like they say, oh, we should be more like France.

But that means you have an entirely state owned system where you have a very different structure to how any of this stuff works. And people who tend to be talking about that tend to be the people who prefer to have a smaller state for this stuff. So it's like, okay, do you really want that? Because everything else you're suggesting suggests you probably don't think the government should be involved in all this stuff.

Asim Hussain: I think one thing I will state though is that I think it's interesting the way , because I think I will actually, I won't. I won't. Even though I'm very out. My debt, I'm going to carry in a little bit because, I dunno if you know this Chris, but the French energy firm actually owns, I believe half of British gas, which is really fascinating because the.

Chris Adams: You're talking Centrica, right? So they own…

Asim Hussain: Centrica? Yeah, it's a British castle, the energy, but they're in a significant part of the UK energy market, which is fascinating because you know, Britain privatized the energy market which was sold predominantly to a state owned energy firm in, in France. So now with the energy challenges that happening across Europe, France is somewhat protected.

Whereas anyway, we're all paying like double, triple our energy prices.

Chris Adams: Actually, so France over in the last year, there was a big thing about the cost of power going super high in France because while there was historically lots and lots of investment in the previous nuclear stations and what you might refer to as thermal energy, where you basically heat water up to make steam, to turn a turbine, to make, to generate power, what you found was.

The, you had all these kind of issues with corrosion and stuff, but also because you had all these heat waves reducing the amount of water available, that meant that it was really hard to keep things cool, which meant things were coming offline. So you end up losing lots and lots of what you would refer to as firm nuclear generation, which put the cost of power really high in France, except that.

Yeah. Nuclear thermal. Yeah, basically it wasn't just nuclear, it was any form of thermal energy had this problem because it's all relying on water to keep things running. If you're gonna turn water into steam, the water has to come from somewhere, doesn't it?

Asim Hussain: Yeah. Yeah. Oh, so it didn't increase the prices by nuclear going down and having to burn more coal. More coal and gas. It was just everything increased.

Chris Adams: There was just like, there was just a shortfall of power and as a result, the cost of electricity went through the roof really high in France as well. And you saw that manifesting and yeah, this is one thing that you saw a lot of, basically, so what you ended up happening was the French government ended up essentially bailing out the large countries, doing a massive investment at that point, which is somewhat different to how we did it over here, but.

Now we can move on. Cause I think we've just gone on off on one actually.

Asim Hussain: There is, I do wanna say one thing. I think more about the, not about what Dallas, let's leave politics for, uh, for. No, we can talk politics, but the one thing I wanted to say that I think is quite interesting what the Microsoft Service Emissions Estimator project, and I think this is parks back also to what we spoke about last week, which is Will Buchanan and his work at Microsoft as well.

Ola, I remember meeting Ola initially from this, you know what's called a green team. So this employee led grassroots, sustainably focused individuals inside Microsoft and that's. All I was and from my, where my memory serves. And Ola, please message out and reach. Now you're a correction. The Future podcast said if I'm wrong about this, but you, Ola wasn't in sustainability at the start.

This was a personal project. Something that he personally felt was important. Push and push. Years later, it's now being released. And this is what one of my things, and I think actually one of the things that we were really talking about in climateaction.tech, which was employee driven. Work inside organizations is so important.

I think probably a lot of what is eventually advertised as like a big corporate endeavor really starts off with a couple of passionate people or one passionate person inside an organization who pushes and pushes until the lever finally moves. And so

Chris Adams: This is true. The drawdown, you, if you've ever heard the term drawdown, they've got the nice kind of shiny coffee table books. One of the people leading Drawdown labs, I forget her name, damn it. But she, there's a really interesting interview with her on my climate journey and she basically talks about, yeah, I think that employees are one of the kind of untapped, unrealized groups that we need to rely on more to actually see achieve some of the.

Asim Hussain: Yep. So next up, we have just a really interesting article that came to our attention on Forbes actually reporting on a. A company called Cycloid, Green Ops. So Green Ops, we have the term Green Ops mentioned this week. Last week we had dev suss up. So the uh, decision is still not yet made as to which term will win out, but Cycloid Green Ops tool, and I find this Chris, a little bit fresh on, it's called Cloud Carbon Footprint.

Chris Adams: Hmm.

Asim Hussain: Which, so basically Cycloid have released a tool called the Cloud Carbon Footprint, which measures the cloud carbon footprint of cloud computing. Interestingly, there's a whole other open source project called the Cloud Carbon Footprint, which was exactly the same thing, and it's from ThoughtWorks. So there's a little bit of confusion there.

I initially, when I saw that was like, what's going on? Have they bought an open source product? But they're just named it the same as an open source

Chris Adams: Was this the case? I couldn't tell because when I looked at this, I thought, oh, they. I was confused by this as well. Cause I thought, hang on, those, these numbers look somewhat similar. And when I look at the, when I look at this, the screenshots don't look exactly like cloud carbon footprint. But yeah, cloud carbon footprint is, is a term that.

Is associated with a relatively well-known and probably like thee most well-known open source tool for this. So I am, I was surprised by this actually. And I'm actually meant to ask the ThoughtWorks I and say, hi, is this you guys? Or has someone actually just rebadged it and provided a hosted service?

Because it may well be that, in fact, cuz I, I. We know that's the thing that ends up being used in lots of places. And there are various other providers, like one company is called Green Pixie. They use some of the underlying parts of cloud carbon footprint in the same, and I suspect that this might actually be a.

it could plausibly be a kind of view on the existing version of this because if you don't want to run an some infrastructure to work out the footprint of your infrastructure, then I can see why you might wanna have someone else manage that. Because the cloud carbon footprint tool from ThoughtWorks is, it's got some stuff like how to set up with Terraform and stuff and how to run things in type script.

And if your team isn't comfortable using type scripts or this stuff here, then maybe it does make sense to use. Uh, hosted service for this. So that's my guess , basically.

Asim Hussain: While you were talking, I was double checking the blurb and that's, they actually specifically mentioned that it is based on that very same project, the cloud carbon footprint open source project, which makes me feel good. I was, I was confused a little bit, but that's, Very clearly mentioned in their marketing material.

This is based upon the open source project that I was talking about, which is really exciting cuz you're right, it is quite complicated to to set up cloud carbon footprint. It's not for everybody. It is a cloud-based tool that you need it hosted somewhere in order to work. And I believe how it works.

Remember how it works, Chris? I believe it works predominantly through billing data, at least the AWS component of it. I remember correctly.

Chris Adams: Yeah, there are two ways it would use the information. So the first one was that you could query the billing APIs provided. Large cloud provi providers and that, and based on that, they would say, you spent in the last week, you spent maybe a hundred a thousand euros on Amazon EC two, or Microsoft's equivalent or Google's version of that.

And then it would provide a conversion factor to say, for this many hours, it would likely be this. Based on the size of your machine and how long it's been running, and I'm not sure what time, but they might do that depending on if you have that kind of access, basically. So it'll give you some figures like that.

And that is the main way that it used to work. I think there is actually an alternative way that you can get the data from the, the, for example, you can also use utilization based approach. So they would read from say, Amazon, CloudWatch, Google's something, stack Driver, all this stuff. Yeah. Whatever.

Asim Hussain: The equivalent. The equivalent for Google and that,

Chris Adams: Yeah, and I, I actually dived into this cause I opened a Pi, a pr, a pool request on the project because I wanna look through it.

It's not actually that complicated to get this information for things which are not just the big three. So as long as you have an idea of how long something has been running and what the kind of utilization is like, how much of the CPU you're using for any of these things, which is exposed by lots of providers.

Then you could do this. So Hetzer could do this, Scaleway could do this, Digital ocean could do this quite easily. It's just a case of people not doing that yet. But no, it's open. It'll be really cool to see that. And ThoughtWorks provides some on-prem a service where they'll basically plug this stuff in so that you can have numbers specific from non-cloud infrastructure to have a kind of consolidated view of your emissions from all this stuff.

Asim Hussain: Yeah. Yeah, yeah. And it's good. This is, it's making the technology available to a lot of other people, which is very useful. I think. It's interesting, and I can, I see there's another, there's another announcement in the newsletter. I think it's interesting because Cycloid is a devops organization, I'm, this is more and more in the cloud space.

This is, these are the organizations that we're hearing more and more about if you are, or the, not organizations I should say, like providers and providers from DevOps. As Anne mentioned last week and she was quoting Adrian, it's a very much aligned, like carbon emissions reduction is aligned somewhat to the DevOps, to what's the cost reduction and all these other aspects that DevOps

Chris Adams: It's also easy to measure, dude, it's also easy to. Numbers from here, apply a number, apply a conversion factor, and come up with another number that you're being told that you need to report against by the C-Suite or the CSR or investors and anything like that. So in many ways you can think of these as a kind of pain reliever, whereas before they were considered like a kind of vitamin, oh, isn't this great?

Now you're like, oh. Can this help that person go away so I can focus on my existing work? That's how some of this is actually being presented to people because there are like regulatory drivers for this, for increasingly. One thing that's confusing with this esteem, cuz we were confused by cloud carbon footprint being the name of a well-known project and a commercial service from a totally separate, unrelated organization, and this kind of makes me glad.

There is a trademark on some of the green software stuff because I feel that if I'm confused by this, then I suspect other folks would also be confused by this. And I think when you look at other projects like say Firefox for example, or Jango or WordPress, people are a little bit careful about how the name can be used because it might not be obvious to what you use.

On the flip side with this, maybe the thing they're going for is saying if cloud carbon footprint is just ubiquitous, like hoovering is, maybe that's the thing you would just say, well, you're just gonna ccf it. Or cloud carbon footprint it. I dunno, but it's a, it let me realize that this conf confusion is only gonna happen more and more as people start thinking about this or have to be mindful of this.

Asim Hussain: yeah. Thinking more generously, which is not a usual trait of mine, but just to give the opposite viewpoint. So it also could possibly be seen as a sign of respect. You're a commercial organization, you wanna use a product, and you just name it the same as the open source, so you're not. The fact that they mention it very explicitly in their marketing material also is they're not trying to hide it.

But, uh, but yeah, I see your point. Cause this is the Kubernetes, I believe Kubernetes is trademarked, so you can fork it and call it Kubernetes, you can't call it Kubernetes. So Yeah, that's really, yeah. So that's, yeah, I think that's something pretty cool.

Chris Adams: a minor segue into IP law. Spoke about the actual project itself. Spoke about the fact that it's open source and can be extended and tagged in various directions.

Asim Hussain: Yeah. Yeah,

Chris Adams: Yeah, I'm happy.

Asim Hussain: I'm happy with that, how we covered that. Yeah.

Chris Adams: Yeah. What's next? How green is your cloud,

Asim Hussain: Ah, yes. How green is your cloud? So this is an article on Tech Monitor, and so as you might guess, this article is about the environmental impact of the cloud and just highlights some interesting stats. I always wonder, whenever I read any of these articles, I wonder what stat they're gonna quote.

the cloud's impact, cuz there's, it's a wide, the band of, as you could quote, is significantly wide, but they quote, the cloud computing contributes between 2.5 and 3.7% of global carbon emissions. And they are quoting a 2019 study from the shift project. I've seen other stats I've seen as low as 1%. 3.7 I think is one of the highest in terms of current stats.

The stat I actually find quite interesting is the one from Eriksson, which is I think is. Interesting because it talks about the growth of orange. So if we do nothing else, I believe in what we're doing right now. By 2040 we'll be, I believe it's 14% of global emissions, which I think is a really interesting way of looking at it.

Cuz it,

Chris Adams: 14% of global emissions. That's like steel,

Asim Hussain: Yeah. That's like all of, almost all of transport,

Chris Adams: That's,

Asim Hussain: it? Yeah.

Chris Adams: I'm struggling with that,

Asim Hussain: Are you struggling with that? Oh

Chris Adams: struggling with that being 14%.

Asim Hussain: and 14%. That's what we know. We, we are almost certain.

Chris Adams: Yeah, that's a, so for context, like shipping, all of shipping, that's like between one and 2%. And so I think agriculture's right, 20 to 30% or so. It's like a, these numbers, they are, we're not very good at like measuring, like keeping an eye on this stuff. But 14% seems incredibly high for us, part of the existing technology sector, right?

Not everything is gonna be.

Asim Hussain: Just so we're clear, they're not saying it's 14% now it's, it'll be 14% by 2022.

Chris Adams: Even then, and that would mean that cloud computing would have to overtake the manufacturing of steel or the manufacturing of concrete as a key emitter. And like you could possibly make the argument that in 40 years, like between now and 2040, that steel will become so clean. and people are gonna shift away from using all this kind of coke and stuff to make steel to go there.

And likewise, it's the same with cement, even though cement's been like the significant driver. Yeah, you maybe have that, but I think between now and 2040, like technology is probably one of the easier of the sectors to decarbonize. This seems like someone's taking some numbers and just like basically pointing it.

Yeah. Rather than actually thinking what's gonna.

Asim Hussain: I can see your point. There's probably multiple Variables at. They're all moving independently, but if you froze some of those Variables and extrapolated out, there's probably an argument to say there's 14%, like for instance, like in the incredibly complicated and lab that exists in ASMs mind that just run an experiment with his thoughts like, I can't imagine the manufacturer of chips is going to.

The major part of what that 14% is, it has to be energy consumption. That has to be that in in terms of what that model, it has to be the energy consumption. I cannot see 10% of all global emissions in the world making chips by 2040. And then if you maybe assume the current grid mix and all things out to 2040, and then maybe you can get that argument.

If you then have something a bit more complicated, then assumes the grid mix is going to get cleaner by the time it gets 2040. And then things may be balancing themselves out and probably that. That stat probably comes, well anyway, multiple levels of guessing even on my side

Chris Adams: All right. I, while you were saying that Asim, I looked up. And

Asim Hussain: looked it up.

You looked it

Chris Adams: looked to a, yeah, I wanted to bring some light rather than just heat into this discussion. This is our Weldon data, which is generally pretty good. Iron and steel is around 7% of global emissions. What we, what we have right now, all right, so it wasn't 15 and agriculture is probably around 18 ish percent, so like I was at the wrong end of the 20 thing.

So this still feels. So all of cloud computing being double the footprint of all the iron and all the steel being made, that seems very.

Asim Hussain: Chris last week. This is quite interesting cuz. Last week you quoted a stat which Anne found challenging to accept, which is a 7.3 million data centers in the world. And now a stat's been quote. I think what's interesting is there's stats here that boggle the mind because the scale of what we're talking about is really hard for human beings to imagine.

I've had colleagues of mine, one of my colleagues gave a, a presentation, which I thought was really fascinating. She took a picture of a rack, and then the, the picture of the room, a rack is a silver rack, a picture of the room. A picture of the building, a picture of the campus, A picture of campus is part of like multiple campuses, and you are already an enormous space, and that's just one of those 7.3 million data centers that exist in the world.

So I think that could form part of the resistance that we're finding in our minds as to the scale of where we're in.

Chris Adams: One thing that I saw from there, there's like another highlight in this piece that says, despite sustainability now appearing in the top 10 business priorities, only 9% of companies are allocating resources towards sustainability goals. And like I Canditt, thinking you can't have both. You can't say it's a

Asim Hussain: course you can.

Chris Adams: and then say you're not gonna be.

Can you imagine if we said revenue is one of our top 10 business priorities, so we're not gonna allocate anyone's time to chasing revenue inside this organization. Do you see how it sounds? A little bit unconvincing here. Or like a, and then if you look at the companies, let's say Google, Amazon, Microsoft, large companies, then we say, okay, if it's a priority, then why are the emissions continuing to grow every single year?

Right? At least between 15 to 20% each of these companies year on year. That suggests it's not as much of a priority as you might be thinking. If we know the science is saying we need to be reducing these year on. . So I've struggled with that part, but,

Asim Hussain: I think it's, there's, my colleague of mine recently did some analysis on, you know, this website, net zero tracker. Have you seen Net zero tracker?

Chris Adams: uh, I think so.

Asim Hussain: They analyze not just net zero, but like other, the various kind of sustainability commitments of organizations around the world. They like score the commitment on a kind of red, green, blue basis and they then score them on, this is your commitment that you've made.

Let's look at the plans that you've published for actually how you're gonna meet those commitments. And what's amazing is looking at it, it looks 58% of Fortune 500 companies have set very like green. targets and yet almost all of them, like any form of detail plan as to how to actually meet those targets.

So I think like setting targets is, is like a very easy thing for an organization. And in fact I, no, just to, I feel like I might be the one, cause I work in enterprise organizations, so I feel like I have a little bit more, I just have an insight that might not be available outside. And I think that. At some level in an organization, the leadership has got to set the direction of a, of an, of a, of an organization.

One of the ways, one, I think the very important first step for an organization is for the leadership to come out very publicly, not privately in an email, which then get ignored, but very publicly say, this is important to us. This is the commitment that we're going to make. So I think that is an important step.

That next stuff. I think the money, I don't think people fully understand how money shapes everything. Absolutely everything, and it doesn't even have to be intentional. It's just this is how our company makes money, a, B, and C earns us money. The whole organization is just absolutely geared towards maximizing.

That's what a for-profit company is these days. It's an engine to make money, and so all these promises are off to the side of. Rather than the primary thing, and I think this is why regulation is so important, some advice is given to me a while ago, which is that people that all they're focusing in is solving their pains, their pain points, and unless you're causing pain, you're not really going to be solve it.

So regulation is a pain for an organization. So they, if there's regulation on sustainability, they will put effort into resourcing it.

Chris Adams: Peyton really was a vitamin. Absolutely.

Asim Hussain: forces. is a pain. So if your customers are demanding more sustainable features and there's a competitive nature to this or another organization saying we will do it, that's another pain that you, you do it.

And I also argue that employee, internal employee forces are also pain cuz it's becoming increasingly the sustainability credential and organization is becoming increasingly important as one of the metrics talent is using to choose to whether or not to work in an organization.

Chris Adams: Yeah, absolutely.

Asim Hussain: say.

Yeah. Which counteracts the profit motive. Yeah.

Chris Adams: Yeah, actually, so here's how, here's one thing I see when, cuz there's an implication here that perhaps near liberal shareholder capitalism might not be the mis mechanism for us to actually get here. Right? And I won't di go down that particular rabbit

Asim Hussain: keep on wanting to go into politics, like

Chris Adams: But then, no, the reason I was saying this, cuz there's a good point.

I'm assure you. So, for example, we spoke about in technology firms, there is a whole thing about being 24 7 clean energy by 2030, right? This sounds really like a big thing. Microsoft has this, what they call it, 10, 10, 100, I think, or 10, is it? What? Do you know what it is? Is it.

Asim Hussain: yeah. It is. I think it is. It's a hundred hundred 24, a hundred percent renewable a hundred percent of the time. I don't, I can't remember.

Chris Adams: Alright. I've actually, I did a talk about this, so I'm embarrassed that I don't actually have the particular thing at hand.

Asim Hussain: don't think it's, I don't think it should be embarrassed. I think it's should be embarra. Like why is every organization choosing a different brand name for exactly the same?

Chris Adams: I don't know. Google mentioned, Google spoke about 24 7 being the key thing that they have and like the key, let, if we just step away from the N, the words people use, it's basically every hour of every day. Being matched with renewable generation is the key idea. And Google was one of the leaders for this saying, yeah, we are gonna do it by 2030.

We think it's hard, but we're just gonna manage it. All right. Then Microsoft came in. We are a trillion dollar company. It's gonna be hard. I think we're gonna get there. And then you look, and then earlier on this year, a small energy firm called Peninsula Clean Energy. Based in California, they were like, oh yeah, we're at 99% clean energy matched already, and we're on target to hit a hundred percent by 2025.

And here's the model we've used to figure out how to procure this. So this makes me think that, okay, if one organization is able to move, literally doing half the time of these large companies, then it suggests that it could be more of a priority and they could move just as quickly as this other organization, which has far fewer.

And I feel like this is why I, this is why like you said about the governing is so important. If it's a priority, you'll actually hit, you'll absolutely talk about this. And just like you said about like the pain thing, I'm really glad that there is now a really good example of a small, not particularly well resourced energy firm going so much faster than these trillion dollar companies, cuz I'm hoping it's gonna accelerate them to do this as well.

To an extent, to be fair, some of the funding and some of the work is somewhat funded by some of these organizations, but it does show that if you make it a priority, then you will actually move that quickly, and we totally can do this. It's just a decision that people are choosing not to move as fast as they really need to right now.

Asim Hussain: I, I, I agree. I, I've got two points to say. Here I was. One of them is I used to have a statement, which is, if you're working in sustainability inside a large organization, as as quickly as possible, you want to make sure that your work is not being supported through what I call grace in favor. So some executive leader, this is a priority for them, and they're pushing back the tide and pressure of all these other things saying, this is important to me.

I'm making sure that Chris Adams has got the resources to focus on sustainability. That's great. And most things start off with somebody doing that. , but you won't get the significant investment unless you kind of align with the rest of the business of the organization. And if that exec leaves, your whole division has just gone.

So I always say great. And I think that we should talk about it next week. I think. I wonder if some of the things are happening in Amazon are. Related to that kind of activity. But the thing I was saying, this is maybe like a call to the people who, who work in startups that are listening to this podcast, cuz quite a common piece of feedback I get when, and I talk to a lot of people who are in startups and I understand the pressure of a startup.

You're in survival mode. I mean, this isn't, you're not just sitting back, you're like, you're wondering whether you've got enough money for the next six months or next year. And it's sustainability a priority for you. But I think, Chris, you had a really good point, which is, A smaller entity is far more capable of reaching these targets and these goals such as the 24 7 ALI matching target than the larger organizations.

If your cloud businesses several hundred billion, it's much harder to reach like some sort of energy just because the market isn't there. You can't even just buy your weight out out of the solution. But imagine a very small cloud operator. It's much easier for them to achieve those kinds of targets. And I would just encourage you to explore that space a bit more because I believe if you were to achieve those targets, there is the market pressure there.

There is the customers there who would then choose you over the larger organizations. I think that's a missed opportunity for a lot of startups. I see.

Chris Adams: Yeah. I also, proportionally it's not that much, so if you, so I did a talk or I, me and Max Schultzer, he's another one of the members of the Green Software Foundation. We did a really nerdy recorded YouTube video. China basically deconstructing the cloud model to figure out, okay, how much profit is left over If you really were to step on the accelerator to try to actually achieve 24 7 by 2030 and like Amazon and Microsoft is 30% net profit for most of this stuff, there's plenty of cash left over to actually then like redeploy into this stuff and there's so much policy support both in.

America now with the infrastructure, the I, the ira, the Inflation Reduction Act, and in the uk uh uh, uh, in Europe as well. Loads of this, like I really feel that this is something that you could, that people could move on and people who aren't, those companies could quite easily actually compete on this, in my view.

Anyway, we're going way into something else cause we've got one story left.

Asim Hussain: Yeah, yeah, yeah. Finally in in the news, I think there's a really interesting paper. It's called Counting Carbon, A Survey of Factors influencing the Emissions of machine learning, which is one of our favorite topics of this podcast.

Chris Adams: This is cool. This paper, I'm really glad you mentioned it actually. Asim. Yeah. Okay. So Counting Carbon, the idea was there to this one woman, Alexandra, Sasha Luci. She works at Hugging Face in Montreal, Canada, and Alex Hernandez at the University of Montreal. They. Basically looked at something in the region of 500 papers where people were talking about the different machine learning models they've been creating.

And they basically, they sent an email and contacted every single one of the 500 authors of all these papers and said, hi, can you share some information about where this was run, what you used and what and how long it was running for? And they basically came up with some figures saying, these are the key things which will affect the environmental impact of machine learning.

And they talk about things like say the, the source of energy, the amount of training time that's being used, and a couple of other ones. It's a really nice piece and there are some really surprising things that came out of it. Basically, they broke this into five different tasks, kind of buckets of tasks that these machine learning models would do.

So like image classification, so, which is object detection, so that's like picking out a face in a, in something like that. Or machine translation, which is. What is you imagined, like Google and all these tools use question answering and named entity recognition? I'll be honest, I don't work in AI and ml, so I don't, oh, hey.

Your named entity recognition is basically through text, pulling out ideas and concepts, so that's what they were doing.

Asim Hussain: Adams is the speaker of this podcast. Chris

Chris Adams: So that's what they did. And the, there was one thing that surprised me was that of all of these ones, the only thing they saw was of all these ones here, only the image recognition one was the one that there was a strong correlation between the energy use and uh, the accuracy and effectiveness of the models, which was mind blowing for me.

Asim Hussain: so what is a difference between those models? Logically, those models should function effectively. Similarly, they're just nodes that you pump numbers in and these weights and all this other stuff. It's just matrix multiplication. What's the difference in the matrix?

Chris Adams: I don't know enough about it to really talk about it, but the quote, based on the comparison between carbon emissions and performance, we can observe that the only task in which better performance accuracy has systemically yielded more CO2 emissions was image classification. Really. So that was one of the key things that kind of blew my mind because you might naively assume that in order for you to have better models, you would need to just burn through huge amounts of energy.

And it turns out no, that's not actually the case. It's much more about the actual design and how people have actually been putting some of this stuff together.

Asim Hussain: Or it could just be those other types of models they could just plate. Plateau. Whereas image processing is like such a complicated thing. What's in an image is probably a lot harder to understand than what's in a body of text, which is a bit more structured. So what does that mean for chat G P T in large language learning models?

Because those are more.

Chris Adams: This is the thing that was surprising for me because you hear about chat, GBT four, chat, GBT three. You hear like, oh, it's used this much more compute time. Like it's now in the, it's now like maybe a hundred times. There's an implication. It's a hundred times more effective and like this paper is basically saying now that's probably not the case.

It may be more effective, but the link isn't as strong as you might think. It's not like a one-to-one thing where. Doubling the amount of computing you throw at something, you increase it by twice as much.

Asim Hussain: Oh, sorry. Is your argument is the argument. Models are not going to get much better the more we compute them. Right.

Chris Adams: Yeah, the argument is that yes, throwing computing at something can increase it, but it doesn't hold true that it's a kind of one-to-one correlation, and that by doubling the amount of machines you throw at a problem, you double the effectiveness of it. In fact, that's actually the thing they say. That's probably not the case a lot of the time.

Asim Hussain: Okay. But the one thing I did catch from reading the paper was they did discuss how, and I think this is interesting for this space as well. The energy source is a really big, the grid mix effectively is a really big cause of the emissions that they, they measured, which ODE 12 for the future.

Chris Adams: Yeah, it's it. What this kind of implies is that for something like machine learning where you don't necessarily need, where you're not, it's not like you're not waiting on the other end, waiting for the stuff to come through. You're training something for a long period of time, and it's kind of. It's something that probably is more interruptible than other fields.

Right. But that was one of the key things that led to the carbon footprint of the extremely heavy ones, is because not only were there was there lots of computing, but the actual fuel intensity, the carbon intensity of the fuel was actually a significant one as well. And weirdly, for like a significant number, like 12 of the papers, oil was listed as the primary source of power like burning oil, which is just, that blew my mind.

I didn't know that was actually. I didn't, I, I don't know where in the world uses oil as their primary fuel for generating power for the grid, basically. But for these folks who are in Montreal, in Quebec where they have 99% plus renewable energy, that's basically a really good place to be running things and.

For folks who might be using Amazon stuff, for example, Amazon have a, have a Montreal data center. So one of the most effective things you can do, probably run it somewhere where the energy is super clean. Even if you're not able to say, Hey boss, we don't necessarily need to be running loads and loads of machines.

You can say if you're gonna run machines, then running them somewhere where the electricity is very clean. It's probably one of the most effective ways to reduce the environmental impact of this.

Asim Hussain: That just made me realize cuz you know, a lot of a, some cloud provider. Do give you information about how much a renewable energy or whatever it is, different data centers use? I believe all of them, yes, I do believe. I think it's only Google, actually. Google provides that data with their market based measures included into it.

I'd be very interested to get a list of all the cloud regions around the world with actual grid mixes or average grid mixes. Because to answer the questions like that, because I think one of the things we talk about in the SN and the things is you should be picking, preferably picking just based on the nature of the grid mix, not based on the nature of the offsets that you purchased to, to,

Chris Adams: Yeah.

Asim Hussain: that'd be just a really simple thing.

Chris Adams: Yeah. Okay, so last year we announced where I work an IP to CO2 intensity api specifically to do some of this stuff. Now the thing is the information that is available for free. As in as open data go works at the country level. And for someone like Canada, this is actually quite an interesting one because let's say that you were looking, I'm gonna run everything in Canada.

So right next to Quebec is, it's the place where the tar sands, I've totally forgot the name of the province of Canada, basically two provinces right next to each other with radically different carbon intensities of power. So if you just say Canada, you could be running something in. Versus Quebec. So Alberta tar sands, super dirty, super carbon, uh, electricity.

Yeah, that doesn't have very green does it. And right next to is Quebec, which is hydro and nukes, which is problematic in its own ways, but very low carbon.

Asim Hussain: but are you saying, are you, because like surely what time and other providers, like they provide the carbon intensity data by grid level, not by country. When you mentioned that, what were you

Chris Adams: So for this one here, I actually think that what time does provide the figures at the kind of grid level. So I think the term is like either a balancing organization or a BA or balancing authority. So they will provide some of these numbers. I think those are the marginal numbers you would actually see, but.

As far as I'm aware, I dunno, of any open data source that provides a higher resolution in that.

Asim Hussain: source, right.

Chris Adams: And that's the thing, like you could be using what time stuff for like e either to experiment with, but as soon as you wanna put them into production, you have to pay for, there's a fee for that. And I don't know where that data is at a kind of free level like that right now.

And I think that's a thing that's really missing. Cuz in many cases you gotta ask yourself, how many times do you have to pay for this information? You pay once through the, your use of the energy bills, right? You're paying once there. In many case, you're paying through like taxation, so it to be generated and then to be like repackaged again so you can use it.

It feels like surely this should just be a kind of universally open thing that people are able to use, especially if it's like the stakes are this high and it has this much of an impact.

Asim Hussain: Yeah, something I think access to this and I, and I love what I think they do. These are great organizations and they have to keep the lights on one way or another, so I, I, I do understand it, but it would definitely benefit the world if a lot of this data was more readily available.

Chris Adams: It might actually be with, with Canada, to be honest. I mean, I'm probably just being lazy. Yeah. They provide a usual useful service and there's an API and stuff for it, but this feels like stuff which I. I feel like every single government everywhere in the world should be publishing this stuff automatically as open data because it does, it makes it, cuz you can still provide value added services on top of that.

You can still do stuff like that. But for it to be something which is so difficult and so not, not particularly open in lots of parts of the world, is a real problem for the policy discussion, cuz this is actually one thing that, going back to the paper, that the paper mentioned, this paper said, okay, all our models, all the kind of large learning models or machine learning models, There was zero representation from South America or Africa.

All right, so that's lit. So all the models published, all the, that were shared in papers were from universities or from institutions in either North America E, either what you might refer to as, say, Western Europe or China, or, or. The kind of North America, more North American continent, and it's not like there's no one living in Africa and there's no one living in South America and they don't have opinions and they're not doing this kind of research.

It just means that there's a whole sponge of things that we're missing out on because access to this stuff is not available.

Asim Hussain: Okay. Thank you Chris. I think there's one other thing that I just wanna mention before we finish our podcast, which is the meetup program. That we're launching in the Green, Software, Foundation. So one of the things that we would really like for there to be is a global network of people who are just.

Shared similar interests. We're meeting up on green software, different places all over the world, and we actually have a Meetup program, which means that we pay for, if you know what Meetup is, a meetup's, a platform which enables people to, to meet up and we pay for the costs of running a group on Meetup and we actually have about 25, 26, 27 meets up groups there.

A bunch of 'em are looking for organizers. A bunch of those groups have now become the actively looking for organizers for them. We're actually willing to also, Launch a meetup group in your area and if this is something you are interested in and if you're interested in organizing a meetup group, if you're interested in helping out with a meetup group, if you are even just interested in joining a meetup

Chris Adams: Speaking for one of the meetup groups.

Asim Hussain: for one of the meetup groups.

Really anything. I personally have built and grown multiple meetup groups in London and it's incredibly rewarding. Meeting up with people with a shared similar interest

Chris Adams: realizing they have legs. Yeah. Mostly,

Asim Hussain: Realizing they have legs. And especially in this space, I'd say cuz we're in a very challenging space and it can at times be quite hard to stay motivated even sometimes.

But I think I find that meeting people with similar interests is a very empowering thing. So if this is something you're interested in, please visit meetup dot Green, Software, Foundation, and you'll find like a bunch of resource information about how to get involved. So that's just the call to action here.

If you want to get involved in a meetup program, visit meetup.green software found.

Chris Adams: Oh yeah. Cool. And I suppose. Realized with the meter thing you're doing, if you were to choose to run an event somewhere, you've probably got a list of people you could ask already with the Speakers Bureau. So that would make it a bit easier to find.

Asim Hussain: Yeah, exactly. That's why we launched the Speakers Bureau because to help the Meetup program, that was one of the primary reasons cuz we, we have a speakers mentioned before, the Speakers Bureau works very closely with the meetup program.

Chris Adams: And just as I understand it, the Speakers Bureau, you don't need to be a member of the Green Software Foundation to be part of it, do you? You can be, as long as you've been doing research or you are able to talk about this and confident talking about this field, you can get yourself listed.

Asim Hussain: That's exactly right. And the same goes actually for the Meetup program. You do not need to be member of the GSF to be an organizer of a meetup group.

Chris Adams: Oh, cool. That's handy. Okay. That's so nice. Everything to end this. So nice I up for that. And I, I think you're listed and I think I'm listed. I can't remember if I am.

Asim Hussain: in the speaker's bureau. Yeah, you're listed. Yeah.

Chris Adams: Oh, okay. In that case, I guess that's one way to ask if you want me to speak at one of the events you or seem to speak at one of the events or even Anne and.

Oh, this is being recorded in February. If there's something happening in March, I might be around to actually be doing a talk in London about that stuff as well.

Asim Hussain: So that's all for this episode of The Week in Green Software. All the resources for this episode and more about the Green Software Foundation are in the show description below, or you can visit Green Software Foundation. That's green software. One word. Dot the Symbol Foundation in your browser.

If you enjoyed the show, please consider leaving a review on Spotify or Apple Podcasts,

Chris Adams: Five stars. Five

Asim Hussain: star , leaving a five star review on Spotify or Apple Podcast or wherever you get your podcast. Your feedback is incredibly valuable and helps us reach a wider audience. Thanks again for listening, and we'll see you on the next episode.

Bye-bye.

Chris Adams: right. Take care everyone. Bye.

To find out more about the Green Software Foundation, please visit Green Software Foundation. Thanks again and see you in the next.

Community Clouds and Energy Islands with Dawn Nafus and Laura Watts

5 december 2022 | 44 min

We’ve moved from a more decentralised internet running on centralised power, to a more centralised internet running on more decentralised power. Is this the only computing model of the future? What would a decentralised internet running on decentralised power look like? We see hints of what this looks like at the edge of the internet, but also the edge of the grid, and this is an area our two guests Dawn Nafus of Intel and Laura Watts of the University of Edinburgh have spent quite a lot of time researching. They join host Chris Adams in this episode of Environment Variables as they explore community clouds, datacentres, energy regulation, projects on Islands of Orkney and the book that they’re working on together!

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Dawn Nafus: Website / LinkedIn
Laura Watts: Website / LinkedIn

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
The SDIA

People:

Papers:

Paper: Carbon-Responsive Computing: Changing the Nexus between Energy and Computing by Dawn Nafus, Eve M. Schooler and Karly Ann Burch.

Projects:

Microsoft Underwater Data Centers in Orkney
Solar Protocol

Podcasts:

Environment Variables Episode on Carbon Aware Computing
Script Notes Podcast by John August

Books:

The Ministry for the Future: A Novel by Kim Stanley Robinson
Laura’s Book: Energy at the End of the World: An Orkney Islands Saga
Dawn’s Book: Ethnography for a data-saturated world with Hannah Knox

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript below:
Dawn Nafus: The implication of green software is not just that it's efficient in the immediate savings, but that you're opening the door to this much bigger infrastructure change that is enormously important.

I'm your host, Chris Adams.

Hello and welcome to Environment Variables. On this episode, I am joined by Dawn Nafus of Intel and Laura Watts of University of Edinburgh, and we'll be discussing community clouds and energy islands. We're shifting from an internet powered predominantly by burning fossil fuels 24 7 in large, centralized generation plants to one powered by a wider mix of decentralized forms of energy, generation.

And yet over the last 10 years, we've seen a shift from a more distributed internet to one where computing is concentrated correspondingly into large centralized Hyperscale data centers running 24 7, much like the centralized power plants of before. So we've moved from a decentralized internet running on centralized power to a more centralized internet running on more decentralized power.

Is this the only computing model of the future though? What would a decentralized internet running on decentralized power look like, though? We see hints of what this looks like at the edge of the internet, but also at the edge of the grid. And this is an area two of our guests have spent quite a lot of time researching to share their insights, and that's what we're gonna dive into today.

But before we dive in, let's do a quick round of intros in alphabetical order. My name's Chris Adams. I am the host of Environment Variables. I am the chair of the policy working group and the executive director of the Green Web foundation. I also help manage the community called climateaction.tech, and I'll hand over to the next person particularly, which I think is you Dawn, Nafus.

Dawn Nafus: Yes, I am Dawn Nafus. I am an anthropologist over at Intel where I focus on AI governance and responsible ai, specifically with an emphasis on AI's role climate change. I'm also an editor with Hannah Knox of for a Data Saturated World, which looks at the surprising ways that ethnography and data science intersection.

So I'll hand it over to Laura.

Laura Watts: Chris, thanks so much for inviting us to be here. This is fantastic. I'm Laura Watts. I'm a consultant and an ethnographer of futures, which basically means that I collaborate with organizations, companies, and communities to explore there innovations and how they, and together we might make the future otherwise.

And I have a background both in tech cuz a long time ago I used to work in the telecoms industry, particularly mobile telecoms. And I'm also a professor of Energy and Society at the University of Edinburgh. And as part of all that work, I've written a book called Energy at the End of the World an Orkney Island Saga, which is published by MIT press.

And that's based on the work that I've been doing for, oh, over a decade in the islands of Orkney, which is islands off the far northeast coast of Scotland. And that's actually also where I live. And I've been talking a bit and working with them on their energy futures, which we'll be hearing more about soon, I think.

Chris Adams: Cool. All right. Okay, so for the uninitiated, we've used some words like the edge of the internet and the edge of the grid. Before we go any further, I just figured it might be worth just putting that out to see what that might mean in this context to basically anyone who's up for answering that.

Dawn Nafus: The cloud is somebody else's computer. It raises interesting questions about what we mean by the edge. So if you think about this common distinction we make right now between, often when we say the word cloud, what we really mean is. The big Hyperscale, large infrastructural sort of entities where you can rent space, compute power, all the rest of it.

But what we're seeing now actually is also some serious growth at, in what my colleagues and computer scientists call the edge, meaning computers that live outside of those large spaces. And there there's a continuum really between the large data center. Large-ish by normal people. Standards that might be at a hospital or a bank, right, all the way through to servers that might actually be in mobile phone towers all the way through to smaller and smaller servers to something smaller that might be in somebody's basement. But all of those things right now we can think about as an edge in computing.

Chris Adams: Okay, cool. That's really helpful. And Laura,you mentioned about the edge of the grid.

Laura Watts: Yes, so that's actually where I live. It's a geography, it's a place. So we talk about the edge of the grid, we talk about it, or at least I talk about it in two ways. The first way is the fact that at the edge of the grid, we're in places where the infrastructure, the electricity infrastructure, the grid is more precarious by then when you have storms coming in, or the cables say between islands, so undersea cables might break because they have a lot of tension on them.

So places at the edge of bridge, you got precarious infrastructure. The lights go out occasionally. But unlike places in central areas, there's not this media panic or kind of social media meltdown. People just shrug and get on with. So what that means is you've got places where energy and electricity are visible, and people know where their energy and electricity comes from because they're either fixing it or they're maintaining it or they're thinking about it.

Imagine if you're living in a very stormy environment. When I mean by storm, it's hard to stand upright because that's the kind of level of energy there is in the air. You tend to think about energy. It's something that is part and parcel of your everyday life. You feel it on your body. So when I think about places at the edge of the grid, I mean it in these sites and locations where energy is part of your everyday thinking.

So of course, therefore you're gonna be thinking about. I've got visible infrastructure. I know where it is, but I'm also gonna think about how to generate it. So in places like Orkney, like I said, off the northeast coast, we've got 22,000 Islanders. Think a lot about energy. Very stormy. Long way away from London, closer to the Arctic Circle.

So there they've been generating huge amounts of wind energy, cuz I talked about storms, but also we have wave and tide power and we also have been doing hydro storage to think about ways to store this enormous amounts of energy we're generating in the islands. So that's one effect of being at the edge. You think about the energy, you generate the energy.

The second aspect of being at the edge is that the actual renewable energy resources. So environmental resources, how it's often talked about is at these geographic locations. So if you look on a map, where would you go for where it's windy? Where would you go for where the tidal resources or the wave resources you are going to the edge of the map.

So there is a correlation between being at the edge of the infrastructure and the grid and where the environmental resources, where the waves are big, where the tides are strong, where the wind is very powerful or where there is a large amount of sunlight. So that's the other aspect of the grid. And those two things are what's really changing the shape of the grid as we go forward.

Cuz you're shifting from this. As you talked about, Chris, we went from this centralized fossil fuel power structure. We've got big power station outside cities, and suddenly we're going, huh? Our grid is now the wrong shape because all our power's being generated a long way away from our cities, right? It's being generated in these wind turbines offshore.

It's been generated by other locations, and that has huge implications because the grid is no longer the right shape because we're having to change it as we go into the future.

Chris Adams: Interestingly, one thing you mentioned all the technology there, actually Laura was for the kind of like green software nerds; that's all energy tech that all that people are putting in data centers right now actually. And if I understand correctly, I think you had something you might want to come in on that actually, cuz I could see you nodding away there just then.

Dawn Nafus: Yes. Yeah, no, I do wanna build on that in the sense that if you think about, there's two things to think about here. One is, if you look at the geography of Hyperscale data centers, you can start to see something of a movement, right? You can see, for example, data centers going in the Pacific Northwest where I live, which is really in part about hydropower.

Some wind to a certain extent. You can start to see things starting to move, but it's also connected to. Other kinds of infrastructure, other kinds of considerations like latency, right? So we're not yet at the edges in the way that Laura describes it in, in her work on energy edges. But the other thing that also comes to mind is often when we're doing edge computing, it's true that energy becomes more about salient consideration, right? So you might actually be on battery and have to do a ton of tricks to get your compute down to something you can actually manage, right? You might be running a camera for whatever reason, and you might wanna actually do the computer vision at the camera and not move a bunch of data that you don't need to be moving around specifically either because there's an energy consideration or there's just, it just takes a ton of resources to be able to do that.

So we are baby stepping into this new world where energy is distributed in ways that are different from what we're used to. But we're in no way there yet, right? We're not in those kinds of places where the wind really is serious, or where in the middle of the Australian desert where the sun is, no joke, right?

So it's too slow for my liking in other words.

Chris Adams: Okay, and if I understand it correctly, some of the work in the Orkney Islands is actually seeing how some of the communities are using some of this technology and seeing how they relate to some of this. Is that the case, Laura? This is what some of your research was were you were doing before with the Orkney Cloud. Is that about right? Or maybe you could come in here at this point there.

Laura Watts: Yeah, though I think that picking up from what Dawn said, and this answer your question, we're in this really exciting and important moment where we've gone from cloud computation being about making infrastructure invisible. So if you're a developer, you don't have to think about it. But as we move to adjust energy transition, we're moving to thinking through, okay, we've gotta get, think about where our power comes from. Then suddenly we've gotta understand that question. Where does our power come from? Where does our energy come from? You know, the cloud can no longer be something that's untethered from the energy infrastructure beneath it. It's part of the protocol stack in some ways. So one thing we've been doing in Orkney, which has been quite fun, is asking this question of, and it goes back to your, what you were saying, Chris, but the data center industry is already thinking about where he gets its power from.

That's absolutely central PUE is everything. So what's happening is this reflection on moving away from just power purchase agreements, which is, as I'm sure your listeners know, how data centers try and often power their data centers from renewable sources as they cut a deal to buy renewable energy or invest in renewable energy.

But what's been going on in Orkney is to, rather than basically handing the problem off to a problem of the market. They've been thinking about different kinds of business models and that's where it's things, it starts getting really exciting because instead of it being about, okay, we're just gonna cut a deal and then some, again, distant unknown geography generating the electricity forest or an offshore or onshore wind farm or through wave or tide power, that's something I'm quite interested in - marine energy. But maybe there's things about where we could have a local community and cut a deal direct with them, and that money goes to a charity in the islands that support in the island community. So you can start thinking about fun ideas like fair trade energy or buying direct from wind farmers, literally.

So all these ideas, which we know that we have, but suddenly they become relevant to the energy and data industry. So some of the things we've been doing at Orkney through various projects, We've run an Orkney Cloud project, which was a collaboration with Mozilla, which was great. We've also had a project called Reflex Orkney, which is a government funder project to demonstrate a flexible energy system, and you'll have to ask me to explain terms.

If I get to geeky, it happens. But what we've been doing is saying, normally you manage a grid, so these power purchase agreements might go to renewable energy generation. And grid management at a grid scale. So you're looking at grid scale batteries or you're looking at something like switching turbines on and off what you might do through a smart grid.

But in Orkney, we're going 22,000 people. We're looking much smaller. So we're thinking about managing things like home batteries or electric vehicle charges or micro wind turbines, much smaller community scale things, hydrogen electrolysis, which we have connected to some of our community wind turbines, managing all these things.

To balance the grid. So it's a different kind of management and it enables the communities in Orkney to think about how they can take control of the grid and also to generate energy renewably. So Orkney generates 120% of its electricity needs from renewables. So it's got this amazing resource. And the first stage is doing that, doing like energy as a service type ideas, or think about community asset management.

But then the question is, what if some of those assets weren't just energy assets like a home battery. Maybe it was a server or maybe it was a data center container sitting on a beach plugged into the community windturbine, and you're managing that as an energy asset. Suddenly things become really interesting. We haven't done it yet, but it's an exciting idea.

Chris Adams: Okay, so as I understand it, there were some work with, say, Microsoft having some submerged data centers in various parts of Ireland. This is adjacent to some of that, perhaps. Is that the case?

Laura Watts: Yeah, what you're talking about is something, again, listeners might have heard about. They might have heard about Orkney, because Microsoft who ran the project, Natic said that Orkney had just become one of the most exciting places in tech. And so it was an underwater data center filled with nitrogen, and it was plugged into the European Marine Energy Center test sites.

So it's on. Planted on grid, running off the green electricity in Orkney, and it was underwater for, I'm not sure how many months, maybe a year. You know, listeners can look up project online. And that's about demonstrating the feasibility of underwater data centers for reliability purposes and obviously cooling.

Cuz data centers get hot. That's why talking about them. Cuz it's all about energy and it's all about cooling in Orkney obviously that partly inspired us, but we were also thinking about you could just simply have a server or you could just again know data centers come in containers. What if we plugged them in to our grids and had a conversation about how we think about data processing and data storage in a much more tighter relationship with renewable energy generation. Because as we know, renewable energy only gets generated when the wind blows. We have to either store it in hydrogen or other forms of storage or we have to change the way and when we process, cuz renewable energy needs to be shifted in time and space.

It needs to be moved on the grid over space and it needs to be moved in time to when you need it. And that's maybe something which you can start linking up with the data processing and storage.

Dawn Nafus: One of the reasons why I wanted to think about this topic with Laura is I had separately started doing some work in carbon aware computing, which I understand you had a whole podcast on, on this very topic. And encourage folks to listen cause it really was wonderful. But if you haven't, the short version is that it's about finding techniques to run your workload when and where the renewable energy is available, right? So when the wind is blowing, where the sun is shining, all the rest of it. We've been exploring in inside our labs how to do this with AI training, which is a good thing to do in a carbon aware way because you can wait, right?

The data scientist might be able to wait an hour, might be able to wait a couple days, cuz these things take sometimes weeks to train up. One of the things we quickly learned is carbon awareness is, yes, it's a scheduling problem, right? So there's some software that folks can build, folks have built about when and how to place your workload and all the rest of it.

But it quickly becomes not just a scheduling problem, right? So all the sudden you start to see, oh geez, the grid in California, in fact does look really different from the grid in Oregon. Looks really different. From the grid in New Mexico. So all of a sudden you start to have a relationship to place that you wouldn't otherwise.

Right? And you start to think in these terms that don't think folks tend to think in reading Laura's work, you can then ask the next question, right? Which is, maybe it's not just about the locality and the grid. There actually might be opportunities. To start to think more deeply about who's benefiting and who's actually running what and where workloads are actually going.

Right? And we can start to make choices about that. So as another example, one of the things that's really been heating up right now on social media, you might imagine is with the recent changes, shall we say to Twitter, there are a lot of folks like myself who have moved over to Mastodon. And on Mastodon, we've been having a rip roaring conversation about what would it take to actually stand up a Mastodon server in a place like Orkney where stuff is in fact community run and where there actually is community benefit to how the energy actually works and how it's organized. And there are a million challenges to that that we can talk about, but that's that next step. Once, once you get beyond scheduling, right, you can start to think about all these other social implications that are far deeper than just, you know, writing some scheduling code.

Chris Adams: So if I understand that correctly, Dawn, you're talking about like once you've solved some of the kinda scheduling problems, I suppose there's a chance to then layer over kind of higher level services, like some of the things you might associate with having to have in a more traditional kind of data center.

Like for example, we were using Twitter and that's considered like one way that people actually use to communicate and coordinate with each other. And you're saying like once you've got that, It's plausible to think maybe there's other ways you could create other kind of, for want of a better word, that I'm borrowing from say, some work by Rachael Coldicutt, like community technology, like provision of other kind of services that you might otherwise be getting from very large companies, like say Google Drive and stuff like that.

That's what you're alluding to there?

Dawn Nafus: Yeah, absolutely. It doesn't have to be that way all the time, right? Certainly when my colleagues are training their ai, they're gonna very much want it on their own servers for very good sound reasons, right? But the world is not all that. And so we can start to think about, okay, where does stuff go? To whom does it go, and what are you actually doing on those servers?

And that's when you can start to think about localities, communities, and ultimately who's benefit.

Laura Watts: I thought it might be really helpful for listeners to understand that how this kind of like connects together and how the community might benefit from A to B. So if we're thinking about doing some kind of data processing on a server or in a data center, basically that requires energy, it requires electricity, right?

It requires energy in terms of cooling. But it also requires electricity in terms of just powering the kits. So that's the first bit. If you're basically then putting a load on the electricity grid, so that's what your data process and does, it generates a load on the grid. That means a local community can sell electricity, either direct to the data center or server, or is gonna sell electricity to that distributed part of the grid.

So the other thing to realize is grids are getting more decentralized in the way they're managed, but by creating a load on the datacentre, By doing your processing on a very particular data sense, you potentially allow a local community with a community owned or locally owned energy generator. As I said, like wind is obvious, but it might be solar, it might be other things, or it might be something that's more complex like a, a flexible managed system to sell electricity and to gain revenue.

Because when a community or an individual sells electricity at a small scale. They generate money from what's called the feed in tariff. So you get money by selling electricity and therefore you can generate profit and revenue from the sale of electricity, and that money can be used to reinvest in local communities to support public services or to support local initiatives.

There's lots of things that can do, so I just wanted to paint it out end to end so people can understand why there's this direct relationship between. Not community benefit in some random cash sense, but real organized, governed. Cuz often these organizations at a community level have very clear governance about what decisions they're gonna make, about what they're gonna invest in.

So these are the things that make it a very powerful potential kind of business model. And Dawn talked about the Mastodon example because that's a really nice level to think about because most of the time, certainly in energy, it's about households in tech, it's often about users. Things are thought about as individuals.

Or it's the Hyperscale, huge data centers, big stuff. But when we start thinking at community at, you know, the fediverse level, we're thinking about groups of people. And that's a really interesting place to think about.

Chris Adams: I see. So what you've prescribed, Laura, there's actually another one of the Green Software Foundation members called the S D I A. They're the Sustainable Digital Infrastructure Alliance. They've got this notion, which they refer to as digital power, which is, if you imagine there being like a feed in tariff for electricity, they basically conceive some of the kind of primitives of computing, say storage or compute or networkers, there's

another kind of building block that they refer to as digital power that you put together to build other kind of applications on top of it. And what it sounds like you described there is if organizations are able to control if they're, instead of just actually like, say, generating their own power, it's plausible that they might be to do something like generate their own kind of digital power if you want, and provide something which is maybe slightly higher value that they could then use as a kind of basis for building services or things to meet some of their own community technology needs.

Laura Watts: It's a really interesting idea, so I'm gonna look into that. Chris, that's awesome. I think one of the things is really interesting, something that Dawn and I have talked about is also the challenge of expertise. Cuz the place where I work in ney has this huge amount of electricity and energy grid expertise.

What it doesn't have so much of is basically expertise in data processing infrastructures, you know, InfoSec. It just doesn't have as much expertise in that because Orkney has a history of North Sea oil and gas, so it's got its long energy history. So a lot of the things you are talking about I think are incredibly exciting, but the piece that's almost underneath that is how do we get the expertise or bring the expertise or bring training and understanding of those things to the right location?

So that's one of the reasons that Dawn and I have had in having this conversation. It's a, okay, so we have all expertise in the energy industry. It's cultures of keeping the lights. And then there's other sets of expertise in tech, which is thinking through issues of carbon aware computing, but is often at a very different game as different kinds of expertise.

So how do we bring these?

Chris Adams: You shared this enticing term, a digital blacksmith with us before, is this for some of the references or something else? Cause I heard that before and I, or just this idea of it. Addressing some of these skill gaps of people who might maybe know their way around energy but not computing or vice versa. We still have this kind of gap to create some of this.

Dawn Nafus: Yeah, Yeah and I think not everybody needs to be a software person necessarily to bring it back to the Macon example. And we might be able in a one off sort of a way, actually find the right people, right? Our networks are pretty big. Your listeners might in fact want to jump in and say, Hey, I'll stand something up over on Orkney.

But then it raises this question of, okay, what about these other corners? Other places where people might have onsite renewable energy of a kind and say, yep, I'm very happy to have some sort of server equipment in here. But what next? And so they're having some sort of abstraction layer or something where in a sense you're abstracting enough away so that folks need to know just enough to get done what they want done.

So you don't wanna abstract away the location in the way that Hyperscale currently does, right? You want some of that visible. But you know enough, we are actually getting into somebody's server in a safe and secure way, and doing that kind of orchestration with the energy that's important, and that's something that is easy and available off the shelf.

Chris Adams: So there's two projects which spring to mind when you talk about it. Actually. One of them is the solar protocol, which I'm getting lots of nods here on the podcast. It's a project which is a kind of collaborative project of various Raspberry Pi's with batteries and solar panels distributed all around the globe with the idea being that the website that has the most sun and the highest amount of charge and a battery will be the website that serves whatever website is actually hosted with the solar protocol. And for a while, that's literally just been one website, but they had a hack day a few months back where they actually started talking about the underlying technology. And basically the underlying technology is a raspberry pi with an Apache server, really reassuringly boring technology that you can run WordPress on.

And they now have an open project to start, hosting new projects on this distributed like a, you can, like super green CDN, I suppose is what you might refer to the solar cdn. That's one, one project they have. There's another thing that I've come across in the uk, which may be of interest. There's a company called Green Cloud, and I've taken advantage of the fact that, yeah, most computers have quite a lot of excess power now, excess computing power.

So the idea would be that if you have a machine with spare computing capacity and you have say like rooftop solar or stuff like that, you can add your computer as a node to a, basically a set of serverless style tools. So the, you have something which feels a lot like the kind of serverless tools that you do have, but you know for sure that you're running it on entirely with green power or basically a kind of mini rooftop, solar powered data center inside someone's home.

So this idea of distributed computing, there's a few interesting examples in the uk, but I would love to hear from more, because that's solar protocol is actually more like American and globally, but these are two ones which I immediately start thinking about when you mentioned actually Dawn.

Dawn Nafus: Yeah, I think they speak to two things. One is, you know when the necessary thing just becomes so evident, right? You, you tend to see it cropping up. Here and there and everywhere. But I think it also speaks to, you mentioned boring servers, . But I think it, it speaks to the need and the importance of doing the boring parts of green software development to making sure that your code is what it needs to be, right. That it's not collecting excess data that you actually don't really need, or that you've architected it in a way that you know makes sense for the infrastructure that you've got, because that infrastructure is changing, right? And so the software that's gonna run really well on a solar protocol or the other infrastructures that you mentioned, right?

Those things have to be developed for and kept in mind as you're building stuff. So the, you know, the implication of green software is not just that, it's efficient in the immediate savings. But that you're opening the door to this much bigger infrastructure change that is enormously important.

Chris Adams: So it's not just efficiency. There's a piece of about resiliency related to that as well.

Dawn Nafus: Absolutely. Absolutely. You're making it possible to be resilient, right?

Chris Adams: Okay. This brings us to a nice question about, okay, what this actually means for a natural developer and how this might be different to developing for, say, the cloud like you might have had before. I know that there is one paper that was released recently. Uh, Oh, God, I totally forgot the name. It's, I think it's, Nolan Bashir.

I'll need to check. It's this notion of an ecovisor. Basically, we have hypervisors, which basically take a resource, like a large computer with a certain amount of hard drives and stuff like that, and split that up into a set of virtualized resources. The concept behind an ecovisor is to do the same thing, but with power.

So rather than just having a steady supply of power that you come in that you just don't really have any real understanding of. It's instead, you have power split into kind of three kinds of forms of power. You'd have grid power where the carbon intensity might change over time. There's a renewable onsite power, which is very low carbon, but it's somewhat variable.

And then there's this notion of like battery power, which is coming from something local that you might be able to design with. And this idea is that if you have an idea of what capacity and what quantity you have of each of these going forward. Then you can design with this in mind to make the best use, reduce the carbon intensity of the software that you're using by either using a certain amount of battery over a certain amount of time, or if there's the cases where you have an abundance of power, you might say.

Just shell out like a set of extra subprocesses to use that kind of resources. That's the only example I've seen so far. But I would love to know if there are any other ones that you folks have come across, cuz that was a really cool idea and I've only found out about it from one of the previous guests, Philipp Wiesner actually, when he, he shared a link like Chris, check out this cool paper. It's awesome.

Laura Watts: It sounds really cool. I think for me what it highlights is this kind of coming together are two very different sets of expertise and development of different systems. So you've got the kind of like the eco advisor concept, and I'd be really interested, I haven't read the paper to see how much that's in conversation with people who are doing grid transmission and things like active network management systems, and flexible management systems. So people are working inside DSOs and DNOs, you know, on grid who are struggling at the moment to figure out these really important questions of how you manage your grid assets cuz you've gotta switch stuff on and off and you've gotta be able to balance it and you've gotta know as much as you possibly can about your wind turbine and how it's operating and about your kind of various different, I'm talk, I talk about it as assets, but essentially load or whether it's batteries and you talk to a really nice categorization there.

So the question for me is like, how do we bring these two together? Because there is enormous expertise that goes back decades that's trying to address some of these issues. And then if you look at, say, the UK government's white paper, it reads on energy grid transmission and the kind of like energy future.

It's talking a lot about data. It was like, we need to think about open data. There's the open data task force that's trying to think about this because all the things that are raised in that paper only work if you can get the data from the asset, you need to know about the battery, somebody else, they need to manage that.

And some of that exists. I mean, we've been trying to do this in Orkney with the Reflex Orkney project. Because we're trying to take the data from the home batteries, the micro wind turbines, all these things have apps. They will have APIs. We can get that data, and then we need to basically be able to manage those things on the grid.

So the first piece is, can you get the data you need? And that's not straightforward or trivial. And the second thing is also regulation. And Dawn, you just talked about really dull things like the dull side of green computation. I wanna talk about regulations. So before your listeners feel like I'm about to give a snoozefest, regulation is one of the biggest challenges to what we're talking about for energy and data and thinking about how we do things like the ecovisor, and that is manage assets because you need to be able to have permission from the regulator to basically switch these things on and off or be able to have any impact on the grid, because keeping the lights on is an absolute commitment. So if you're going to start changing the load, if data's gonna get into this space and thinking, how do we write code for using different amounts of energy sources, that's gonna change the load on the grid. And that starts getting into regulatory issues.

And it seems like a dull thing, but actually it's a really important space to start talking about because we can have huge impacts on what the grid looks like. What does a data electricity grid combined look like in the future? That's a regulatory and governance question as much as it is a technical; how do we shove the data about and change what the software looks like?

Dawn Nafus: Yeah, and just to highlight here, in a sense, data figures at two levels. One is. The data that you need to pull this off at all. Right? The instrumented assets that allow you to put stuff where it needs to go and what granularity that has, how that's governed, right? All the rest of it. But then also there's the data that you're shipping around and a straightforward example of how somebody developing software might develop differently if we did have something like, uh, This eco advisor thing or whatever it is, just in the face of changing energy situations.

One of the things we're starting to learn with our carbon aware AI training project is even just how you develop your training experiment. It makes certain amount of assumptions, right? If you assume that energy is just on and you just run the thing, you build it one way, right? You don't really care too much about pausing, right?

For example, which you might, if your energy is intermittent, right? As another example, you might not care too much about where the data is in relation to the compute, unless you're really pressed for time, in which case you really would. But if you're architecting the training code such that you actually know where the data is in relation to where the best place for compute is, well, You've gotta take this into account in your model, right?

Like how it's gonna train, right? So once you all those options open up, then there are implications for how you're developing machine learning models, right? And those are just the simple ones, right? And we're learning what those are as we go along, but I suspect it'll get more interesting as time goes on.

Chris Adams: That's really interesting what you just said because that's the first time I've heard people talking about fighting data gravity. The idea that everything is up, just being in one gigantic big box out of town data center. Like if you design it differently, then you need and have that, there's a way you can basically design away from that. The gravity issue essentially, right?

Dawn Nafus: Right. Exactly. Yeah. And there might be dynamic choices, right? There might be a time and a place where you look, the data gravity is just so big. It's not worth me messing around with the grids greener over there. Right. It's just not right. You know? You know you're gonna have to do. Training 10 times over , right?

You might want to move stuff wholesale. You might wanna know actually what the networking costs are in between in case you have to shuttle, right? There might be a moment where the grids are running renewable energy in such a way that actually it's worth your while to shuttle. Again, these are things that we're just exploring, but yeah, it, it might not be the same type of training, you know, everywhere for everyone.

Chris Adams: This is really interesting when you talk about this cuz this, this keeps coming back to some work by Lorenzo Kristov. About basically taking the lessons learned about computer science, like Law of Demeter and not having to see every single thing or basically applying lots of the ideas from computer science to grid design.

This sounds like really interesting. I wanna just check, uh, Laura and Dawn, you folks have been working on a book. Okay. You've been thinking about this for a while and I understand that you've been putting together a book or something around this. Is there a chance you might be to talk a little bit about that?

Then maybe we'll wrap up for the last few things cuz there's a few questions about what, what you're listening to and looking at right now. And I'm curious about this book now that sounds like there's some work going into this.

Laura Watts: Yeah, definitely. I think the first thing to say is that all the things we're talking about are the things that we've been thinking through because there is this extraordinary. I wanna say most like it. It's not so much a collision, but there's this entanglement of really important between energy and data.

Particularly important when we go through to this change to renewable energy. It's having an impact on data, and we can think about what the future looks like for this combined energy and data industry. Dawn, you were just talking about ai. One of the things that you and I have also talked about is an author respect for computing as a whole.

You know, data is, So I formally have a physics degree way back when, and that, and it leads me with a permanent understanding of the fact that when you talk about data, it's basically just an energy, it's an energy difference between a one and a zero, right? That's all it is. So, you know, whate, whatever you're storing any data, whenever you are doing any processing, you are always using energy and that has huge implications for whether you decide to make an AI at all, whether you decide to store data at all that we have, terms like data lake and all these kinds of things, right?

That seems to give the implication that data is some kind of inert object or it's just sitting there and of course, no. Every data that you store costs energy, and that has an implication to the environment. We can sit here and we can say that's bad, but the conversation we're having today is exciting because we go, actually, we've got lots of really smart ideas about what we might do about that.

We don't have to just shrug. That's really where the book came from, this sense of we've got this interesting combination of industries. That are coming together because there was a shared issue about the fact we know we need to use energy much more smartly and that requires data and energy together.

And so we've got these kind of closed loop issue. And that's really where we started off having the conversation and it led us through things like you heard at the beginning we're ethnographers, which means that we think about issues like culture because that informs the way you think you can imagine a future and build a future based on your experiences of where you come from.

Well, originally you might think there's a big difference between tech and energy because energy is about keeping the lights on. It's very conservative, it's very risk averse. And you know, the traditional idea of the tech industry, it's move fast and break things. It's alpha, beta, ship that really doesn't work in the energy industry.

And then Dawn, you and I started talking about that. They're actually much close than that. Yeah, I just Dawn, you were talking about data centers in particular, being much closer to the energy.

Dawn Nafus: Yeah. As computing has become more and more an important infrastructure in our everyday life, in the same way that energy is right, things have started to go in that direction, right? We really care if you're running a data center, you really care about keeping uptime, keeping your service level agreements, all the rest of it.

If you're in networking, which is computing, right? Same principle holds, right? You don't want packets dropped and things going down. So things are running a lot more like an infrastructure and not just bits of code here and there. And that's important. So there is a sudden point of convergence and that's also interesting.

So again, what we're trying to do in this book is to really think about, okay, what are the best of both worlds that are really gonna help us get through this energy transition?

Laura Watts: Just an example of things that the data industry knows about data privacy. We've had years of worth of thinking about this. We just talked about ai. We knows about this stuff that's really new to the energy industry. Energy industry's got smart meters, it's got home batteries. It's working with personal data, but there's not a lot of experience of thinking about these issues.

We can start realizing actually there's a lot of benefit to bring them together and sharing the expertise across them. And then in the energy industry, certainly in Europe, there's a commitment to making sure you keep the lights onto vulnerable customers and knowing who the most vulnerable people are and making sure they aren't without electricity.

So that ethical attention. Is something we can bring to potentially the data industry, for example. And that's just the tip of the iceberg of some of the things that you might be able to do. And all the things we talked about in this podcast were these different kind of business models, different ways of thinking about meso levels of scale ways of thinking through energy and data together, and the flexible management of assets and data.

These are all the kind of things we're thinking through and the kinds of scenarios for, for what a future might look like. That isn't where renew. Can run the grid and can run data and are an energy mix. I've talked about lots of different energy sources already, but there's a huge number of different ways you can generate energy around the world and that you know is enormous potential that data can really thrive from.

Chris Adams: Cool. Thank you. Before we go, let's get the name of that book. Before we go so we know what, what to be looking for. Is there a working title or is it Or, or something to look out for?

Laura Watts: We do not have a working title at the This is the most exciting thing because of the fact that the, your listeners are getting literally hot off the press ideas, but what you can do is actually you can get the precursors. So I have a book, which I mentioned, energy at the End of the World, an Orkney Island Saga, which is published by MIT Press, and that is available, it's published in 2019 won various awards.

It's written for a broad audience. It's intended to explain what's been happening so innovatively in Orkney, how and why, and gets into some of these issues. And Dawn, you also have a book out as well.

Dawn Nafus: I do, I have a couple, so there's one. On the relationship between ethnography and data science called ethnography for a Data Saturated World with Hannah Knox. There's also some earlier work on the quantified self movement and self tracking and all things to do with keeping track of both your body and the environment that it lives in.

This will be a new adventure for me, certainly, but I'm looking forward.

Chris Adams: Cool. Thank you, Dawn. So I'm just gonna wrap up now. So I'm gonna ask you with one question. Is there a book or a podcast that you're listening to or reading right now that you'd like to share with people?

Dawn Nafus: I'm certainly hugely inspired by Kim Stanley Robinson's Ministry for the Future, which I think really does a beautiful job of articulating what it might look like to take climate seriously and all of the social repercussions of doing so.

Laura Watts: And I think that for me, one of the podcasts I've been listening to recently is actually about the kind of practicalities of writing. So it's called Script Notes. It's a podcast which is for screenwriters and, and things that are interesting to screenwriters. So you seem might seem quite distant. It's a podcast by John August who's a film script writer and also Craig Mason, who is known for Chernobyl, who was the showrunner and and script writer for Chernobyl.

But the thing that, the reason why I find that so helpful and inspiring as myself as an author and writer is that many of us are actually writers. So all academics are professional writers. Many people are writing reports and words have enormous power. I mean, obviously words take power to transmit them.

Literally they cost energy to transmit, but also the style of writing we choose changes. How effective what we say can be. And I think a lot as an author about choosing the right words, about making sure the words are most effective they can be. Because whenever we're choosing to write, even if you're writing kind of a technical manual, it sounds an odd thing to say, but I think our words can really change the world.

Even when we're writing something quite simple, we, we transmit a lot of knowledge to the things that we say, whether it's here through podcasting, but also the actual written word itself. So I find that as script notes is yeah, really helpful for kind of reminding people for paying attention to the editorial process and thinking about the power of words.

Chris Adams: Wise words to end the podcast with then actually, Laura, thank you. Okay, we've come up to the hour and folks, I've really enjoyed this. So thank you very much for beaming in from the various parts of the world, from the Orkneys to California, and folks, I'll probably see you again on a future Green Software Environment Variables podcast.

Thanks folks. Take care ourselves. Bye.

Dawn Nafus: Thanks for having us on. Bye-bye.

Laura Watts: Appreciate it. Bye.

To find out more about the Green Software Foundation, please visit greensoftware.foundation. That's greensoftware.foundation In any browser. Thanks again and see you in the next episode.

Fact Check: Sara Bergman & Software Carbon Intensity

18 november 2022 | 54 min

Welcome to Fact Check on the Environment Variables podcast! Fact Check is a new segment where we take a deeper dive into the bigger questions in a one-to-one discussion with a special guest. Host Ismael Velasco, is joined by EV regular Sara Bergman, a senior software engineer at Microsoft, and an individual contributor to the Green Software Foundation’s Software Carbon Intensity project. They discuss Sara’s green software journey, the Software Carbon Intensity ISO standard, why it excludes carbon offsets, and fact checking what that tells us about offset-based green software claims, from Google to Blockchain.

Learn more about our people:

Sara Bergman: LinkedIn / Twitter
Ismael Velasco: LinkedIn / Twitter

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
The GSF’s Software Carbon Intensity Specification
Microsoft’s Green Design Principles
Microsoft’s Carbon Aware Windows (Windows 11)
Microsoft’s Sleeping Tabs for Edge
Browser Plugin: Tab Suspender
Tim Frick’s MightyBytes

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:

Sara Bergman: It gives us the software practitioner a way of evaluating if a future change will be good or bad. If an older implemented change was good or bad, it allows us to find the biggest culprit. Sets us up for success in terms of being able to change our software. And I think that's what makes it important because it, it gives us leverage.

It gives us opportunity for room to move, I guess, to do something, to not just stand there and be like, okay, now what?

Ismael Velasco: Hello and welcome to the first episode of Fact, check a new segment in Environment Variables. I am your host Ismael Velasco, and I am thrilled and inspired to be welcoming Sara Bergman to our home.

Sara Bergman: Thank you so much. I am thrilled to be here and I'm very honored to be on this this first episode. So thank you so much for having me.

Ismael Velasco: It's great. And for our listeners, the format that we've had in the past for Environment Variables is where you have a host and several panelists, and you've been a regular. But this will be the first time that you're interviewed one on one as part of fact check. The goal of fact check is to do that, to dive more deeply into a subject with a wonderful expert and just have a good conversation and see where it takes us.

So I am excited to, to start, and before starting, I just wanted to ask a bit. About yourself. I understand that you sort of got into green software in university and just kind of never let it go, but I wanted to know why that happened. You must have chosen to study it. So how did you end up in, in software engineering and how didn't you end up in green software engineering?

Sara Bergman: Yeah, absolutely. And I'm really impressed by, by the, by the homework you've made here , so thanks for that. Yeah. So I. Usually say I slipped into software on like a banana peel in the school system in Sweden is such that after 10 years of school you get to select program for your upper, upper secondary diploma.

So you have three years where you can. Go a bit deeper into something. So you have the university preparational programs where you have the more geared towards profession such as carpent tea or hairdressing or, or things like that. And I wanted to do something that was preparatory for university, that was close to science and technology, and that was a special program that they had.

Like a third was computer science. And at the time I knew how to start a computer and I could play the. . That was about the extent of my knowledge. But I thought, and to this day, I can't explain this reasoning, but I was like, yeah, computers would probably be important. At some point I better learn. So that was my sole reason for choosing that program.

And then in the first class, I had to sort of raise my hand and ask, sorry, but what's a CPU? Because I've never heard a term and. Of course were a lot of men in this class, they all turned around like, who is this chick? What she like, does she know what she's done? I did not know what, what I had done at all.

But then this was a great program. The teachers was really great, and after that it was sort of love at first sight once I got into programming and started taking more of those courses and I, I really loved the mindset of how you think about building software and. It's so creative. People don't think it's creative.

It's like the best kept secret of our industry is the creativity. So then after that I chose a, an engineering programming in university, which was geared towards engineering. And yeah, that's where, where, where it led to green software engineering, because we actually. In order to be allowed to graduate with this specific title, you had to take at least one course, which involved sustainable practices.

So I was forced to do something at least. And then I wanted, I took a course called Green itm.

Ismael Velasco: And did you have a previous connection to nature? I heard in one of your talks when you were giving examples and you were making a really great point, you were saying how we make all these efforts to green our domestic lives? And then we get to work and kind of lose our sense of control and forget all about it.

And you said, you know, we'll walk instead of take the car and I can't remember what other thing you mentioned, but then you said, you know, we will reuse the skis we give children instead of using new skis. And I just thought that is a Swedish vignette. I certainly didn't have to make those choices in Mexico growing up, and I just wondered whether that from the sounds.

You were exposed to nature you did grow up with presumably Skiing would be my guess.

Sara Bergman: Yeah, yeah, absolutely. No, I was very much raised into sort of nature close living. My parents competed in orienteering, which is where you run in Themeforest with a map and you find checkpoints. My dad was really successful when I was born, so I think I was at my first orienteering competition at three months of age.

So I've been

Ismael Velasco: Did you find your parents at the end?

Sara Bergman: I'm told my mom held onto me, but . But yes, no, I think that's fairly common for people both in, in Sweden where I'm from, and Norway where I live now, it's integrated into our culture in many ways. Same with skiing. You know, I live in Oslo now. I take my cross country skis on the subway, take the subway directly to the tracks, like with a bunch of other people, which.

When you tell people outside of nowhere, they are like amazed. But to me it's just very, very normal. And aside from that, my mom or we have a family farm. It's a very small scale, mostly forest farming. So yeah, it's, it's always felt like a responsibility to care for our planet and the legacy that we live behind to our children.

Cuz I am glad for the legacy that my great, great parents left me, which is divorced, which is now a big part of my heart.

Ismael Velasco: And I guess sort of that meant that you were predisposed that when you did that compulsory class. It, it made sense to you in not just in an intellectual way, but in a kind of spiritual, existential way. And the other thing that struck me is that I think. You mentioned how the creativity is the great secret of software engineering, and I was thinking when you were talking about your parents doing orienter, that actually software engineering and seeking green software paths is like orienteering, isn't it?

You have a few set of signposts here. You get recognize the landscape, but it's not a direct route. You don't have, you know, you know where you want to get. Roughly, it might not look anything like what you had on the map to start with. So, and you have to sort of be very agile and observant, et cetera. So I can see how growing up in all of those mindsets would have meant that when you landed in engineering, in software engineering and in green self, it particular, you would have resonated

Sara Bergman: And so that's a great comparison. I haven't thought about it. I will definitely steal it. That was great. Yeah. But I think I was also very naive, right? I, I imagine that if you wanna work in sustainability or make an effort, you need to like go into something that was very hands on, like you needed to be working directly with animal conservation.

Like that was my idea. So when I was. Taking this course, it was like, oh, I don't have to go outside my area or my expertise to do this. I can fight climate change in my own arena. And that was like eye opening and something that I'm very happy that I'm able to do because software is something I'm very passionate about.

So I'm glad I can combine those two passions.

Ismael Velasco: And at the same time, I heard your word fight for climate change in your arena and your career, and I think that's really interesting as well because it, it makes me think of you. A young girl asking what is a CPU? And not only had you chosen to go there, which is you know, pretty reckless to start with, but in an old male environment where asking a que a technical question as a girl positions you in a very gendered way, you were fearless and just went, sorry, what's a CPU?

And I get the feeling that then you got a job and you ended up working Microsoft and. You put your hand up again and it was like, yes, but what about the environment? How did that happen? Did you arrive sort of into your professional career and people were already talking about it and that space existed or did you arrive with that kind of passion, awaken you and thought it's nowhere here.

What do I do?

Sara Bergman: I think it was a combination for sure. One of the reasons I chose Microsoft is because they had at the time, and even more so now, a very clear position when it comes to climate. It's always been a pillar of Microsoft to first Carbon neutral and now this increased focus of being carbon negative, so, so that was definitely one of the reasons.

But I think also coming directly from university into a big corporation, there is a certain amount of time needed to find your feet navigating. A professional, what does it mean to go to work? You know, all those things. So it took a while before I figured out how I could marry the two. I used hackathons a lot in the start, which is something, uh, we have internally, which was great to sort of investigate this passion.

I've been so, so lucky to have the support of my managers and the mentors and other peers around me on this journey. And then I think it really took off after I saw Asim speak at March, build in 2020. It was like super early in the pandemic sitting at home and I was like, whoa, I wanna be like him. So again, I guess pretty fiercely.

I set up a one-to-one and was like, teach me your ways And uh, he was kind enough to accept and yeah, opened a lot of doors for me. So I do believe in maybe being a bit troublesome in like a good way. Make good trouble you. . I think that's, uh, that's I.

Ismael Velasco: Two interesting points there for me. One is the role of the hackathons and obviously the Green Silver Foundation just finished. Fantastically successful hackathon. We had such brilliant solutions that came up and 400 technologies mobilized into sometimes for the first time, thinking about carbon aware computing, which we will dive into very soon.

And I've often thought that part of the win, as it were, are the winners, right? You win your money, hopefully you take off with your idea a bit more. And then there's the next layer of winning, which is all of the projects that don't win the money but actually advance in some way. Their idea and some of those ideas are likely to go much further than the winners.

And vice versa. You just, you Canditt tell. And then there's a next layer, which is the judges and the companies and the observers being able to go, ha, I hadn't thought of that. We could do something like that, or, that's a good approach, or we should, instead of investigate this further. But finally, I. It's the live altering trajectories that there are people there who will have come to the hackathon.

You know, if you move forward 15 years, it might be a different life just because they went there. And so the fact that you had the initiative to go and the support to be able to do it as part of your work seems like a really powerful factor in your trajectory.

Sara Bergman: Yeah, indeed. They were truly a cataclyst and I also just wanna echo the sentiments you said about the, the hackathon, the gsf just closed. If people listening in haven't watched the recording or didn't attend the event. Please do the, the winners, like all the projects that were presented were, or the hacks that were presented were amazing.

I was blown away. So yeah, just do yourself a favor and go listen, go listening, cuz it's definitely a learning opportunity.

Ismael Velasco: Absolutely. Then the second element was this being troublesome in a good way at work and in. My last podcast actually, I reflected that people often think about big corporations as these monolithic things. They're good, they're bad, they're doing this, they're doing that. But really they're microsystems.

They're so big that they have all of us, and they have all of our motivations. And I will be very green in somewhere else of my life and then not green in others. And my weaknesses and my strengths will vary, as it were. And there's those set of competing interests inside myself that learn in the planet, that I learn in those organizations.

Microsoft recently released this sustainable design manifesto that I also highlighted in the last episode, and in it they speak about it only takes 3% of people to protest in order to achieve systemic change. And likewise, they say that it only takes a few troublemakers, as it were in tech, in your in a company to create systemic change.

And I wanted to ask you about those dynamics. So on the one hand, you have these positive tendencies in m. That are moving towards sustainability. And on the other hand you have, you know, tendencies that are perhaps not so, sort of mutually supportive of that and you're in the middle of, of this sort of navigating and you're one of that 3%.

How do you feel those two forces, as you were, the forces that drive sustainability and the forces that drive non-sustainable patterns of profit? How do they interact? As a whole and in your job as it were for yourself as a, as a technologist.

Sara Bergman: Yeah, absolutely, and and I agree with what you say. I firmly believe in those grassroot movements. If you have grassroot movement and leadership support, I think everything is possible. So if you have both that that is great. I think also what's been important to me is really try to understand people's motivation.

Like obviously I am very interested in sustainability to me, , that's gonna be the most important thing. It is not for everyone. Some people, they will care more about cost. Other people will care more about PR or security, or they have something else that's their cornerstone or corner value, which they will prioritize higher.

That doesn't make them bad people. So you have to think if you're one of those people who. Are emotionally invested in sustainability. For example, if you're gonna argue with everyone to disagree with you, your, your energy's gonna run out really quickly. So my personal mindset is to try and think that everyone is on a different place in their journey, and what I can do is give information based on which level they are right now.

Maybe they're not at all open to this idea. Maybe I can sell them on energy efficiency instead because they are also interested.

Ismael Velasco: Performance.

Sara Bergman: electricity bill. Exactly. Performance. Or I can talk about hardware efficiency because they work in asset management and they don't wanna throw out working hardware either, because that's their core motivation.

So I think that's a cornerstone. Trying to not think of people as your opponent, but you are not yet allies and you know, put on the charm or whatever it is you need to, even if they're never gonna align fully on. your vision. Maybe you can find areas where you overlap and you can be analyzed there and, and that can be super valuable.

So that's something that I try to, to work with a lot. I think also being here in Norway, we are have a more, it's not a flat organizational structure. It's still very. Hierarchical, but socially it's flatter. It's the threshold for interacting with someone who's my managers, manager, managers, whatever, is pretty low for me.

Whereas in other countries and other cultures, that can be quite a big step. I also think I have a benefit just based on the culture I live in. I'm working.

Ismael Velasco: That's really, really interesting and, and I guess one of the reflections I've had in the past is that change scales. When you can align altruism and self-interest, if, if you can put those things together, you can make them coincide. If something that you are motivated by also makes a difference to the world, then the chances of adoption and behavior change are much greater and you've.

Some of those patterns, for example. And in that sense, I've been in sustainability since I was a young man, so for, you know, 30 years, I'm now, you know, 20 years old. So time is, doesn't op operate the way it should, obviously. But having experienced this sort of, The work on sustainability in very different domains.

I find that in software, we are in a very unique place, in a unique position in almost every other arena. Advocating for sustainability is. Implies painful change. It's advocating against self-interest in some area or other, whereas in technology, mostly what makes software sustainable and green also makes it sellable and and cheap.

It's not universal, but it is mostly the case and if you create a piece of software that is sustainable, it's gonna be a piece of software that is fast, that is very well designed for usability that doesn't have a lot of data transmissions, that that is cheaper to run. So in terms of that alignment that you were saying, it is much easier in our industry than it is in most other industries.

And the other element is that that makes our, our industry unique in our ability to make a difference, is that everything we do is designed for scale. It is very, very, very rare that you. Somebody will hire you to build a piece of software that five people will use, right. It has to be at least a few hundred, generally a few thousand, and in your case, billions every month. our ability to have an outsize impact with a small tweak is, is gigantic. And I wanted to ask you about your experience of that trajectory. Where have you. What interventions have you made in your software that you are proud of, that you feel this, you know, took out a 10th of a gram of CO2 every hour for a billion people, or, you know, anything like that?

Sara Bergman: Yeah, absolutely. And then here I have to thread carefully, uh, because of NDAs, but, but something that I'm also passionate about is performance. So whatever I can align those things, whatever I can find an efficiency improvement that doesn't impact user experience maybe at all, or maybe even to the better.

That's gonna dry down CPU and also emissions. Like that's where I feel truly, truly happy and having those opportunities are, depending on where you are in your software life cycle, they will be more rare or there will be more frequent. If you have a piece of software that's very new, you have a lot of opportunity to to decide how things do.

And if you come in with this mindset from the. You can already set up your, your products for a more sustainable way of operating. Now, on the other hand, if you have a very mature software or legacy software, then you get to think about the problem in a whole different type of way. And I had the opportunity to work on both.

And I think they both come with interesting challenges. And for the more mature software, then you might also have to start working with partners and look outside the scope of your own little bubble and see how do. Interact with these piece of software, how can we change that? How can we, maybe there's a software that we know is going to go away, how can we speed that up or, or work with those kind of changes?

So yeah, there's definitely been opportunities, I'm afraid, I can't say specific things I've done.

Ismael Velasco: I think those design principles as it were, Sharing are are very important and they remind me of a specific example that Microsoft has shared widely very recently, which was that it made PC upgrades. Carbon aware and that's amazing. They, and it was announced. I love the fact, it's like I have mixed feelings about it, but at the same time, overall I love it.

I think the fact that the announcement was not tagged sustainability, it was mainstream. It was, this is what we've done in Windows 11, this is our new functionalities. And then at the end they said, and now whenever you charge your. Computer and where it goes to sleep. We've made some tweaks that will make it more carbon efficient and we will now schedule the upgrades for the times where your electricity is green if we've got information that we need from your device.

And that means is an example I imagine of those kinds of very mature technologies that involved a huge amount of partnership discussion and then that landed subtly, but hugely.

Sara Bergman: Exactly. Yeah, I think that's a great example. Another favorite of mine is the sleeping tabs and Edge, where I guess I'm like mo most, um, people in software. I keep way too many tabs. I don't know why. I am a hoarder. I collect tabs. So I'm very happy that the ones I don't use are put to sleep, so they don't have to waste CPU.

And I can still, you know, feel good about my 80 tabs

Ismael Velasco: Yes, I, that mean it has been a game changer to me. I, I use an extension called Tap Suspender because it's sort of more granular, but the idea, my CPU, the fact that my computer runs when I have literally. 500 tabs open. is remarkable. And I know that I'm not killing the planet and I still have my 500 tabs open.

So yeah. And also there's energy saver mode as well. I think that's come out recently in edge. So there's some really good work. And this leads me to another question. So what I see is that we are. That Microsoft is implementing the kind of carbon aware approaches that the Hackathon was all about, and it is seeking to improve efficiency but also diminish its CO2 emissions.

And I suppose a question for me is, where do these metrics live? How could we mainstream, I suppose those metrics. At a granular level where all of the software that we develop comes with a CO2 measure, and I know that in this context you have been really doing pioneering work with the Green Software Foundation around the software carbon intensity specification.

So I wanted to ask you a bit more about how that came about and how you came to be involved in.

Sara Bergman: Yeah, absolutely. So the SCI specification is by the. Working group in the gsf. We started meeting really early after the foundation was started. I'm not sure it's the first, but it was, it was up and running those weekly meetings really early since it aligned with things that I had been doing in my work.

I was asked if I wanted to join, and of course I did, and that's was sort of the start of it. Those weekly meetings have. So much fun, like getting to learn from people who are also software practitioner, but at other companies have other types of experiences. They have other length of experiences come from different backgrounds.

It's been, it's been so great to have those perspective and to learn from those people and to be able to hopefully give something back to this group as well. And I am so incredibly. Of the work that we did both with the, with the Alpha version, which I was very much involved in the now the Viv one version where I had to step away for little bit cuz of other work things.

But that we released now before COP 27. So yeah, I think, I think that was sort of how I ended up.

Ismael Velasco: And what is it? What is this specification and why does it?

Sara Bergman: The SCI is a methodology for calculating the rate of carbon emissions for a software system, and the goal is that you as a software practitioner, no matter, your role, should be able to take informed choices to improve your software so that you reduce or avoid the creation of emissions. So it's a score.

It's not really like a total, it's more a score where a lower number is better and a higher number is worse. But reaching zero is impossible, and it is important because it is biased towards action. It gives us a software practitioner, a way of evaluating if a future change will be good or bad. If an over implemented change was good or.

It allows us to find the biggest culprits. It sets us up for success in terms of being able to change our software, and I think that's what makes it important because it, it gives us leverage. It gives us opportunity for room to move, I guess, to do something, to not just stand there and like, okay, now what?

Ismael Velasco: And do I understand correctly that the way the silver carbon intensity specification measures or gets that. Is three kind of main components. One being how much electricity your software is consuming, so how many kilowatts per hour, and then how that electricity translates into carbon emissions. So that could vary if the grid is dirty or clean at a particular time.

But in general, it would be if one kilo. Our sort of produces 10 grams of co2, then your score would be that plus an estimate for the kind of the carbon that goes from your machine, the embodied carbon. And that would be everything from when it was manufactured to when it's been used to, when it's being disposed of or other, just when it's been manufactured and when it's been disposed of.

Right. Not necessarily the usage time.

Sara Bergman: Unless the usage time includes hardware updates, I suppose then, then you could consider including those as well. But yes. Yeah, you got it completely right. Those are the the three main components that we depend on, and I think when people hear this, they're like, oh, I thought of. It be much harder. , and people expect this to be like a magic formula, which will solve our problem, but really, it, it's quite logical when you start thinking about it.

And of course there are more nuances to it if you, if you read a full specification, you can get into all those details. But yeah, in essence, it's those three core components.

Ismael Velasco: So I want to get into some more of those details, partly because my thought is the opposite. Like, that sounds really complicated. it like I agree. The very, the concept is very simple and that's powerful, right? It's you use electricity, the electricity produces co2. And you have information on the, on what it goes to create the device you're using.

So you add it all up. There you go. So I have a number of questions around that. There are kind of two approaches that I can see that I've seen being used. One is a kind of proxy approach, life cycle inventory. So you may not be able to know exactly, I don't know right now how much electricity. My computer is consuming.

I generally don't know because I don't have a smart meter telling me. So I would probably guess by looking at how much a model of computer like mine consumes electricity using software like this for the hour that we're talking. And that would be my estimate for the electricity. Is that how you see it being used, or do you see it being used by an actual measurement of electricity?

Sara Bergman: So the specification can be useful in two broad. We have to remember that the. Is meant to be used for all software. I think sometimes we can tend to think that software is only something that runs in the cloud in a cloud provider, but that's really not all the software that we have in the world, and we want the s e to be for any software, no matter if you run on an FPGA in like an embedded system or you run on an end user device or you.

are the load balancer inside the cloud provider. We want this to be able to be used for, for every scenario, and thus it's gonna be slightly different and it's gonna be very context specific. So maybe that's where the, the slightly complicated part comes in. But for the two different ways you can, you can either measure it directly and that's very accessible to some, especially if you have a cloud provider that measures it's for you.

Or if you have your own data center and you run everything on premise. you, then it can be fairly easy. The other example is if you calculate it and you use kind of a benchmark, if you, for example, have an application that's, you have a very small service side, the majority of everything is done client side.

It's not really realistic for you to send back energy metrics for every single app on every single device at all time, just so that you can know. Will likely be a lot less green than just doing some experimentation on your side than try to find some kind of average behavior or usual or typical behavior and measure that.

So you can either measure directly or you can do a calculation and both are fine. It, it comes down to what kind of actions you wanna take and where your biggest culprits are. The most important thing is to look system-wide though, and not try to optimize a micro component that potentially have negative.

System wide consequences.

Ismael Velasco: In a sense that brings us to the why. There are two kinds of measurement and one. You measure for, as it were, the truth of the thing. So if you're an academic, you want to get as precise, accurate measurements because that's what you actually care of. And then there is measurement that you use for evaluation that you use for decision making guidance.

And the SCI is primarily focused on the later. And so the idea is not so much Did you catch the exact microgram? Correct. And is, if I run the exercise again, will the curve be identical? But it's more a question, if I understand you correctly of is the curve going up or is it going down? And if it's going up, is it coming from this area of my system or that area of my system?

And if it's coming from that area of my system has take making this decision, brought the curve. That's what really matters rather than, yes, but did it br bring it down by two or by five grams, for example? Would that be fair?

Sara Bergman: Yeah. Yeah, I think that's fair. I think once again, it's those, those actions that we really want to help enable, of course. And get down to the absolute gram that's good for you. But in the end, the important thing is that you find this to be a tool that is useful for your scenario. And of course if you compare, if you compares to evaluate some kind of change, it's of course important to keep the baseline the same drug just for otherwise it's not very scientifically stable, I guess you could say.

Ismael Velasco: One more question around this is how flexible, I guess, is the framework? So this makes me think of some other metrics that are out there and they actually follow the same approach. So when I think of in the Web space, the. Great Tim Frick multi bytes sort of created together with other really key players.

The Green, Web Foundation does whole grain digital and methodology for measuring the carbon impact of websites in particular. And this has gained traction and it's been used by libraries and bys, sort of browser extensions. And they too work through electricity, co2, and embodied carbon. They create a kind of proxy for their use case where they use data as the proxy for electricity.

So they go, if we've had this many gigabytes of data downloaded, this probably equates to this much electricity. And that's how we get that first part of the se. And then the co2, and this was the question, I guess the calculation of how a set number of electricity translates into CO2 or how, for example, what percentage to add to calculate for embodied emissions does the s e prescribe specific calculations?

Or is it flexible enough for you to say, as long as you are doing this three steps in an evidence based way, we are not too bothered as to whether you are using gigabytes or uh, gigabytes or not, or whether you're using one inventory or another for your emissions.

Sara Bergman: It is quite flexible in that sense. It comes down to the reporting. And how you do that. You may also choose not to report on it and solely use it for your own sake as a, as a tool in your own team or your own business group. I think what's important to know is that if you just take data transfer, for example, it only gives you one lever.

That means the only thing I can do. Is to decrease the amount of data sent over the network. That is not a bad thing, that is a good thing, but it doesn't really give you a full detailed picture. So if you also have the exact grid data, so the exact carbon intensity consumed at the exact point. When you wanna use your software, which you, there are several APIs which provide this data.

If you have this data, then you can also consider other levers such as, maybe I can time shift, potentially I can location shift. I can take other carbon aware actions that not only make my software greener, but through the use of renewable and green energy sources Over time, I'm also helping to shift the energy providers to incentivize greener and renewable energy producing.

That gives me two levers. Two levers is objectively better than one, I would suppose. And same with the embedded carbon. You can use a static number, which I understand that this methodology does more or less, or like a, a percentage based, which is a simplification, right? It gives you one less lever that you can.

Use. So the c I really want to encourage granular data use because it gives you more option, it gives you more insights, it paints the full picture. So really the more granular data you can get, the better. But sometimes you do have to make scientific guesses or reasonable estimations, and yeah, then maybe that's what you have to do in this situation.

But really the more granular you can get, the better.

Ismael Velasco: Would I be correct that all of those approaches would be still consistent with the SCI language? So to be fair, by the way, to the, to the Mighty Methodology, they do add different formulas. For specific emissions. So they do sort of take into account carbon intensity and they say, when you don't know it, use this.

But if you know it, actually use the precise one. But that's a good example. You might not know it. So in a situation where you don't have all that information, would you still be able to call your metric software Carbon intensity? So in the case of someone using. Not the great data they're using. This is the average global grid.

This is a time, this is theater. So you've got all the same components as INS E, but you have a thin surface layer with very few levers because of necessity. In that situation, would the final score be ans e score, or does it require a minimum amount of layers before it can be called?

Sara Bergman: It's a good question. We want this to be easy to use and to inspire action if it does. We are very happy. If you read the full specification, there are some data considerations, some fallbacks, for example, I do believe we recommend hourly or minute granularity on the carbon intensity, but if you can't get that, we suggest annual at the annual basis instead.

Most countries, it doesn't change super fast. Of course, with energy crisis in Europe here, it's been changing very rapidly in the past few months. That isn't the normal scenario though, so, so yes, I do think. To the full specification to get maybe some tips and tricks, how you can do this in the best way, but again, more in any insight is better than no insights.

Ismael Velasco: So that's fantastic because it seems to me from hearing you that really what the software Carbon Intensity Specification is doing is creating a common language. It's a statement of faith. It's like, if we do this, we can talk about it together. Let's talk about it together. The ISO is a kind of stamp of approval that says, when we talk about this, let's talk about it this way.

Is that correct? Can you tell us a bit more about the, the why you sought iso that and, and sort of where we are in that, in that process, and why does it.

Sara Bergman: Yeah, sure. So if we remember the, the mission of the Green Software Foundation, it is to build a trusted ecosystem of people, standards, tooling, and best practices for creating and building green software. And I think trust is such a key word here, especially when you think about standards and certifications and iso.

It is a very trusted way across many industries for sharing methodologies, for sharing standards, and I think it's a quite natural step that if trust is something that is a core value to us, then not only do we want people to not trust us, but also use ISO as a vehicle for. Conveying this trust and for, for conveying this to a broader set of the industry.

But really to my knowledge, we are just in the, the beginning of the ISO journey and we have every ambition to, to do it. We are not there just yet. So yeah, I'm not sure how much else I can say on this topic that would be interesting and useful to people, but personally I'm very excited for it. I think it would really lend that broad.

Industry wide awareness, hopefully at least to, to the standard.

Ismael Velasco: And one of the things you mentioned was the aspiration to make. A trusted standard and how that is, that word trust is so key in the Green, Software, Foundation mission, isn't it? It all, particularly in this arena, and particularly when it comes to metrics, the sci is very clear in that it is focusing on elimination of emissions, not offsetting of emissions.

And obviously this really matters because if you're eliminating them, you are actually greening the planet. There is no ambiguity, right? It's gone if you are offsetting them. There are, there may be more questions, so I wanted to ask you, first of all, what is the difference between elimination and offset, and why did the s e standard choose to go with.

Sara Bergman: Yes. So in an ideal world, elimination and offsetting would be the same. We are not in an ideal world wherein we don't have perfect technology, so they are not the same. If you eliminate something, it means you never emitted it. It stays in the ground, doesn't. Go up in the atmosphere. However, if you offset, there are several different ways of doing that.

The ones most talked about is forestation. So you plant trees, which in itself is great. It's good for biodiversity as well. It's good for oxygen that we breed, but are a number of problems around this. There has been reports of projects where treats were planted, for example, and then later they were deforested anyway.

There are also other studies that should suggest that there isn't enough space to plant the amount of trees we would need to offset all of the emissions considering the rate emissions are growing at. So there are. Bunch of questions there. Same with something that more talked about recently as the carbon captures.

It's like a giant vacuum that sort of sucks carbon directly outta the atmosphere. There are very few functioning examples of this and they are extremely costly and it is betting our future on a technology which isn't really mature enough to hold up to this promise. I mean, hopefully it will be, but I think it's a dangerous bet.

So why is it, why have we excluded it in the sci? Well, firstly, because of there are some controversy. See, and secondly, if you include offset, it doesn't help you or inspire you to take actions to reduce your impact. It just tells you how much you should pay, and that's not really what we are trying to do.

We're trying to help you reduce your impact and then offset to sort of complicates the pictures. It doesn't help you to change your software.

Ismael Velasco: So that's exciting because it basically adds trust, right? It makes me trust the E more because I know that it is doing what it says it's doing, and it's measuring what it says it's measuring. One of the other problems with offsetting is that where the offsetting happens is not where the pollution happens most often.

Sometimes it is, but generally speaking, you may be greening in Kenya and polluting in tanza. And that's great for Kenya and it's great for the planet, but it's not great for Tanzania. And whatever impacts that you're having are are not just global, but they're also local, et cetera. And of course there's a lot of criticism around greenwashing and around claiming of sets that aren't happening or verification and quality.

Sometimes if you plant the wrong trees, you are actually damaging rather. Healing the atmosphere, et cetera. But I suppose that the question that arises, so that's the bit that excites me about it. The bit that arises for me is that the way the planet is currently measuring its progress toward environmental survival, at least, is net zero.

And the concept of net versus growth is precisely the difference, right? Between direct emissions and emissions that you're offsetting. So the growth emissions would be how much pollution my software is creating, and the net ones would be how much emissions my software is creating after I paid other people to green some part of the world.

And if I've paid enough people to green enough that it's the same as what I've produced. Then I've arrived at net zero, even if my growth is a lot bigger. So I suppose my question here is how visible will s c I be if what people are looking at is the net zero effect? If people are not considering gross emissions particularly, you're able to go, my company is now net serious carbon neutral.

Yes. What incentive? I said, where will there be, or how do you see a measure of gross emissions like the se, making it onto the commitments, making it on the frameworks, making it onto the goals and measurements, for example, in Microsoft as an example, but the same for everyone else.

Sara Bergman: This is a very deep issue, I think, and it points to something that's. Way deeper than just the Green, Software, Foundation and us. And I think there's a, a lack of what really Net zero means. It means reducing as much as possible and only offset the rest. But some say the rest is like 10%. Others say it's like as close to zero as possible.

It's not really. A clear global consensus on the exact amount of CO2 we would not have to admit. So that is, I think, a global problem. Same that my understanding is that carbon neutral isn't necessarily the same. You could essentially pay to be carbon neutral because you could then offset everything where it's necessary.

It's really about reducing the majority in almost all and only offsetting the rest. The s e can really be a good tool here because we're all about reductions, so if you wanna reduce, which you will need to reach net zero, then the s e I is hopefully a great tool that will help you get there, or at least get you a lot closer.

Ismael Velasco: And that's a really powerful point. It reminds me of all of your discussion about working in Microsoft and my discussion of Microsoft being a microcosm of the world, that actually there is no technical solution to climate change. That it begins with values that people have to actually care. That if people don't care, it doesn't matter how you paint it, what you measure, what you call it, things are just gonna keep getting worse until people care.

And that in that sense, the sci is one more tool in that conversation, one more tool in sort of helping the people who care have an argument and have a vehicle. For expressing that institutionally. So you may have executives who care, but they don't have the vehicle to put that metric there. And if they say, okay, I am going to, I really do care.

And many and many more people are caring because the world is getting harder and harder. So in that respect, it sounds to me like the scis both, it's a tool for change in the present and potentially it's betting on a future that. He's betting in a future that says, what we actually want to do is these submissions, and now we're ready for this.

So if people are principled about this and not just expedient, it's not just about getting as far as you can to hit your targets and move on, then the SE puts on the map something potentially really impactful at that level. And I suppose along those lines, how do you see the. Of the SE standard happening?

Has it started? Are people using it? I've seen a paper on academia, on machine learning and se, I've seen a few bits and bolts, but what is the state of play now? I know it's only just been formally launched, like at the Carbonized software. One incredible event. Just that, I don't know, was it a week ago? But yes.

Where do you see adoption going? How do you see this taking?

Sara Bergman: It is definitely people using it, people implementing it. We have seen several case studies in the standards working group, so yeah, it is definitely starting and, and I think it will be sort of an avalanche situation. I think, again, going back to something that we talked about earlier is a. Think we're likely to see a grassroot adoption and leadership adoption, and then they will sort of meet in the middle.

I think there are teams who, for or, or business groups or startups or whatever, it's like small groups who can use this and are probably using it right now. Maybe they're not reporting on it, maybe they're not publishing work about it, but they use it as a tool for them internally because it helps them, they don't feel the need to, to write research about it or to publish anything.

they just needed us another tool. And then I hope that we will see more companies being really proudly using it than really using it as a PR metric, I guess, and to, to sort of marry the grassroot movement with the, with the leadership movement. So I don't know when that would happen. I am hoping. It will be soon-ish, but I think it's also, it's gonna be very different for different types of companies.

I think cloud providers, Microsoft and other cloud providers are in a, in a bit of a special situation because our business is software, but other also run software in our data centers and that makes it a bit special from other types of software companies. So I'm not sure cloud providers, if that be Azure, if that be Google will be the first.

I think they were the first, like Microsoft. Closely followed by Google and Amazon or AWS to have really bold strategies that they communicate broadly in terms of who will first publish their sci. I'm not sure the cloud writers will be first. I think they might be smaller. I think there's a huge movement in the open source community.

Maybe they will be first to sort of really go big on this. I don't know. I'm, I'm excited to see regard.

Ismael Velasco: And do you see the E being something that is usable? At the national and global level. So what I'm thinking is that people are measuring how much CO2 emissions is Norway meeting in Sweden and Mexico and Britain, and at the moment they just count electricity, I guess, and I C T, and they don't count any anything else.

They don't have the embodied carbon or stuff like that. Generally speaking, could you see regulators, governments being able to. Use the SCI in environmental indicators at scale in that sense. Is that something that you could see being possible or useful or not really? It's more at the business level, at the operational.

Sara Bergman: It is a very interesting question and it also points to the fact that software is kind of a global citizen. We shift our workloads, well, there are some regulations of course, but otherwise people shift workloads kind of freely, and I think if, I think that's also the crux of it then like why the European Union are.

Typically a front runner in these kinds of questions because they have a broad set of countries behind them where I think if one country said we're gonna start reporting on any software that's run in our company or in our country, would that then incentivize people to shift workloads to their country or incentivize them to shift out that country and would that mean something positive for opportunities for labor, et cetera, et cetera.

I don't know. I don't think it's impossible, but I'm not sure it's the primary use case. At least in the near future.

Ismael Velasco: That makes sense and it's a very great insight how it has to happen at sort of in. Rather than individual. I suspect it's gonna be the same with the bigger companies. It has to. Once you've got a few, then there is an incentive. If you are the first, it might not be, et cetera. You need the pioneers, don't you, at all of the

Sara Bergman: It's always scary to be the first. Yeah.

Ismael Velasco: and thinking along those lines.

I want to finish by asking you to imagine Sara Bergman in 2042. She's such a fantastic engineer that she has cracked time travel. You've been working on all of this now for all of these years. You are your wiser self. You've seen it all. You've been there, and now you've cracked carbon efficient time travel, but it only lasts one or two minutes, and then you transported back and it's anchored to you.

You can't just go anywhere, so you basically get a one minute chat with yourself. So you come back from 2042 and you talk to yourself, what did you think you'd be saying to yourself?

Sara Bergman: Well, an excellent question. I really hope I will be saying you did the right thing at the right time. I'm really hoping I won't say it was too little too. That would just break my heart. So I hope that, I hope there is some encouraging words. Yeah, I hope that's it. And just probably there will be some comment about work life balance up in there if I know my future self, you know?

But I don't think, I'm not the kind of person who would give spoilers away. I wouldn't say like, go do this. No. It's the journey to figuring out that's important thing. So just probably something like, go in this direction, talk to these people. You're on the right path, I think.

Ismael Velasco: Honestly, I find it so beautiful. It just hit my heart. I said, I hope. I hope I would say to myself, you did the right thing at the right time and not it was too little, too late. That is profound and beautiful, and I think it's a question we can leave our listeners with. Are you doing the right thing at the right time or too little, too late?

And what a way to live your life. So thank you so much, Sara, for a beautiful, insightful, deep and fun conversation. I wanted to ask if people want to follow you, and they will after hearing this, where should they find you?

Sara Bergman: You can, at least for now, still find me on Twitter. Not sure how that would be possible. I dunno if that's some dark foreshadowing, but so far I'm still on Twitter. Come find me.

Ismael Velasco: Thank you so much for helping us launch our first episode of Fact Check and helping us fact checking software, carbon intensity, helping us fact check offsets versus real emission cuts, and helping us fact check the future. So thank you very, very much and good luck in your adventure.

Sara Bergman: Thank you so much for having me. It's been a joy.

To find out more about the Green Software Foundation, please visit Green Software Foundation. Thanks again and see you in the next episode.

The Week in Green Software: Disintegration vs Integration

10 november 2022 | 51 min

👉 State of Green Software Survey - click this link to access! 👈

This Week in Green Software Episode, host Ismael Velasco takes you through the recent key events and happenings in the world of green software. He outlines a range of reports coming out from The WMO Greenhouse Gas Bulletin, IPCC, UNEP, UNFCCC, IEA and many other acronyms to boot! He also highlights huge positive (and negative) changes in Big Tech and how you can be part of the crest of the wave of change in green software.

Timings of mentions of links below are listed next to them for your reference!

Learn more about our people:

Ismael Velasco: LinkedIn / Twitter

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Microsoft’s Green Design Principles [25:52]
Microsoft’s Carbon Aware Windows (Windows 11) [28:52]
AWS: Well Architected Updates [31:20]
Google Cloud: Cloud Aware Computing Dashboard [33:56]
Meta: AI Powered Audio Compression [38:09]

Talks & Events:

Event: Decarbonize Software by The Green Software Foundation [39:13]

Reports & Articles:

Bulletin: The WMO Greenhouse Gas Bulletin [01:55]
Report: IPCC Report: Climate Change 2022: Impacts, Adaptation and Vulnerability [04:27]
Report: UNEP Emissions Gap Report 2022 [14:02]
Report: UNFCCC Reports [15:00]
Report: IEA World Energy Outlook 2022 Report [16:25]
Article: Beyond Catastrophe: A New Climate Reality is Coming into View by David Wallace-Wells in The New York Times [19:24]
Article: Carbon-aware laptop charging by Andy Stanford-Clark in Hackernoon [23:00]

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:

Ismael Velasco: Hello, and welcome to The Week in Green Software part of Environment Variables, brought to you by the Green Software Foundation. In each episode, we give you the most up-to-date news and events surrounding green software, a bite-sized smorgasbord of resources that will help you discover how to get involved in the world of software focused climate action.

I'm your host, Ismael Velasco.

I'm Ismael Velasco, and this is The Week in Green Software. In this episode, I will look at the raft of worrying reports released one after another by all the top international climate agencies. So recently I will explain why the news is dire, yet find the nevertheless significant milestones of hope buried in their reports.

Whereas last episode, I focused on the power of individual initiatives. This episode, I look at the complex role of big tech and highlight hot off the press green software developments in Microsoft, aws, Google, and Meta. Finally, I take the opportunity to trail the milestone represented by the forthcoming the Carbonized software event on November the 10th.

Check the episode notes for all the. So this week the planet received difficult news from a series of devastating reports, The Intergovernmental Panel on Climate Change, the I P C C, the World Meteorological Organization, the wmo, the United Nations Environment Program, UNEP, the United Nation's Framework, Conventional Climate Change, the U N F C C C, and the International Energy Agency.

The I. All released one after another. A series of scary reports. The World Meteorological Organization released its greenhouse gas bulletin, which found the highest ever CO2 emissions showing that we are back to pre covid levels and beyond. It also saw the biggest year on year jump in methane concentrations in 2021.

Since systematic measurements begun nearly 40 years ago, what does this. We hear a lot of the term CO2 emissions, carbon emissions, greenhouse gases, greenhouse effect, but I think it's become so prevalent that many of us don't really understand what it actually refers to. We talk about carbon. And CO2 emissions as a shorthand for all greenhouse gases, because other gases like methane stay in the atmosphere less so you're able to kind of do more about it.

Whereas CO2 stays for a very, very long time and removing it is a huge challenge. People sometimes think also that if we manage to remove all the excess co. Life returns to where it was before industrial times, but that is not the case. Whatever happens, the atmosphere has changed, the weather patterns have changed.

So although we desperately need to reduce and remove that excess co2, no one really knows quite what the world will look like even after we've cleared it all back to preindustrial. All we know is that it will be better, but not necessarily where it would have been at the beginning, and in fact, definitely not where it would have been at the beginning.

Now, why does this matter? Well, As I said, this growing heat in the atmosphere creates extreme weather events, and, and recently in a, in an earlier report, the wmo calculated that under current policies and trajectories by 2050, every single child in the world will be exposed to severe heat event. My son will be in his thirties, possibly about to start a family every single child in the planet, which also means every adult really will be exposed to severe heat events by 2050 at this rate.

What about the other reports? The I P C, UNEP, U N F C C, iea, they all release reports that coincide on certain finding. So back in 2015 there was a massive conference, COP 21. I was there and it was a real battle to arrive at unity of thought and. What we agreed as a trade-off between the economic needs of society, the speed of change, and the demands of survival, was that we would aim to control global warming so that it would not rise above one and a half degrees centi.

And certainly keep below the absolute scary maximum of two degrees. So again, these are figures we hear a lot. What does that mean? Well, Dozens of researchers who looked at heat deaths in 732 cities around the globe from 1991 to 2018, calculated that 37% of all the deaths from heat in that period were directly attributed to human cost warming.

Which means that our global emissions were directly responsible for nearly 10,000 people a year dying in just those cities. Of course, the numbers are dramatically higher because there are a lot more people in the planet than just in those 732 cities. So every day our choices. Responsible, not just our own, but also historic choices are responsible for the deaths of over 10,000 people a year, and that was in 2018.

The number will surely be significantly higher now. Now if you think about that, an extreme heat event, That occurred once per decade in a climate without human influence, would happen four times per decade at 1.5 degrees of warming, but it would happen 5.6 times per decade at two degrees. The difference between a massive deadly heat event happening four times and happening six times.

In terms of the damage, the deaths, the illnesses, the weather events is huge. If global warming spirals to four degrees, then such events could occur nearly 10 times per decade once a year. So again, if you think about that trend, that by 2050, every child will be exposed to a heat. If you think about the fact that 10,000 people have been dying at much lower temperatures from these heat events each year, not in total, but simply proportional to our human pollution, then the idea of this multiplying and multiplying is clearly.

So what do these reports that have just come out? Tell us about where we are in relation to those goals of 1.5 degrees and two degrees, and this broader situation of the consequences of global warming and climate change. Well, what all of these reports agree is that global surface temperature will continue to increase under all emissions scenarios considered.

They consider best case scenarios, worst case scenarios, the status quo, a whole range of scenarios, and under all scenarios, surface temperature will continue to increase. That also means that under all scenario, There will be increases in the frequency and intensity of hot extremes. So these heat waves will happen not just more often, but they will be more intense.

Likewise, more often and more intense will be the marine heat waves, the heavy rains, precipitation and floods, and in some regions agricultural and ecological drought. And there will be also an increase in the proportion of intense tropical cyclones that are devastating continents and reductions in the Arctic, cis snow cover and permafrost.

All of the reports agree that the updated national pledges since Cop 26 make a negligible difference. To predict the 2030 implementation of the current pledges at the moment, those pledges, if we were to implement implement them, we would have a 66% chance of reducing where we're headed to 2.4 to 2.6 degrees.

So in the best case scenario of implementing our best commitments, That were not currently implemented, just our pledges. If we were to implement everything we've pledged, we would overshoot the maximum that we agreed in COP 21 and hit not just 1.5 degrees, not just two degrees, but 2.4 to 2.6 degrees.

This is if we actually do what we say we. If we do what we are actually doing with our current policies in place and don't change them, we are on track for reaching 2.8 degrees in temperature rise by the end of the century to stay within the 1.5 that we consider vaguely safe, which means still harmful but not extreme by comparison.

We would need to cut four and a half times. 450% are emissions every year more than we are doing now, and three times more emissions per year to just keep to the two degree maximum that we'd agree and reduce global catastrophe. And at the moment there is no clear path to. What they all agree is that if we are going to reduce this growing severity, massive societal transformation is required at speed.

If China is excluded, the amount being invested in clean energy in emerging and developing economies has remained flat since the Paris Agreement 2015, The cost of capital for a solar plant in 2021 in key emerging economies was between two and three times higher. Advanced economies and China. So the countries that most suffer and the countries that have the most solar energy available have to pay two to three times more to create a solar plant.

That's crazy. And a prime example of climate injustice. And if clean energy investment does not accelerate, then there would. A need for higher investment in oil and gas to avoid further fuel price volatility. Clearly, when you take stock of these findings, we are in the midst of a disintegrating world.

Our decision making systems, our leadership, our lifestyle choices are leading us toward a break. At so many levels. However, this narrative of this integration, which is how the UK Guardian, for example, in an article that gained a lot of traction, chose to cover these reports is justified, but. There is another narrative that is also true.

There is also a process of integration going on in the planet. There is an integration of people thinking together, visioning together, agreeing on the reading of reality. We have agreed that climate change is real. We are putting in place policies, we are changing our lifestyles, we are changing our consumer choices, and we are building new institutions, new models, new technologies.

So actually inside these reports that were so terrifying, there were also some incredibly positive milestones. So in the unit report, we find that compared to our policy trajectory in 2010, we have cut our gap in emissions toward two degrees by a third. This doesn't mean that if we improve as much in the next two decades as we have done in the last decade, we catch up, we would need to change by 300% to do so.

Our change is dismally behind the. But it's not nothing and it counts. Whereas the worst apocalyptic scenarios looked likely even at the start of the Paris CO 21 Summit in 2015. If we look at the data in the UN report, we are currently in the intermediate scenario, and this is outta five or six I think it is.

So again, it is harmful. But it is dramatic progress. Another milestone, the UNF CCCs report identified that last year's analysis showed projected emissions would continue to increase beyond 2030. This year's analysis shows that emissions are, for the first time ever, no longer increasing after 2030. Again, the decrease is not fast enough to keep us on target.

But it is hugely significant. Think about this current long-term strategies representing 62 parties to the Paris Agreement account for 83% of the world's gdp, 47% of global population in 2019, and around 70% of total energy consumption in 2019. This is to say that most of the world is trying. Half of the population, but 83% of the economic engine and 69% of the energy vacuum is moving in a progressive direction.

This is a strong signal that the world is starting to aim for net zero emissions. Way too slow, but it ain't. Finally, the most exciting news in these reports comes from the International Energy Agency's World Energy Outlook report. For the first time, a scenario based on the status quo on current prevailing policy settings has global demand for each of the fossil fuel.

Exhibiting a peak or plateau, cold use falls back within the next few years. Natural gas demand reaches a plateau by the end of the decade. And rising sales of electric vehicles mean that oil demand levels off in the mid 2030s before ebbing slightly to mid-century. Total demand for fossil fuels declines steadily from the mid 2020s by around two exes per year on average to 2050, an annual reduction, roughly equivalent to the lifetime output of a large oil field.

Again, this is massive. Our demand for fossil fuels is the single biggest driver of climate change, and for the first time in. We can see it peak or plateau, starting this very decade. Similarly, they predict that annual clean energy investment rises to more than 2 trillion US dollars by 2030, more than 50% from today, clean energy becomes a huge opportunity for growth and jobs and a major arena for international economic competition by 2030, thanks in large part of the US inflation reduction.

And so learned when capacity additions in the United States grow two and a half times over today's levels, while electric car sales are seven times greater. New targets continue to spur the massive build out of clean energy in China, meaning that its coal and oil consumption both peak before the end of this decade.

That would have been unthinkable at CO 20. Faster deployment of renewables and efficiency improvements in the European Union, bring down EU natural gas and all demand by 20% this decade and coal demand by 50%. Japan's green transformation program provides a major funding boost for technologies including nuclear low emissions, hydrogen, and am.

And India makes further progress toward its domestic renewable capacity target of 500 kts in 2030, and renewables meet nearly two thirds of the country's rapidly rising demand for electricity. When you really take stock of this, you realize that these two processes of this integration and integration are equally powerful and they are recipro.

Integration drives disintegration. Disintegration drives integration, but there is motion in both directions. Perhaps the most balanced treatment I have come across this week of our current moment is this recent article by the New York Times. In it, there is a quote from Kate Marvel, and as a scientist and lead chapter author of an I P C C assessment.

She said, We live in a terrible world and we live in a wonderful world. It is a terrible world that's more than a degree Celsius warmer, but also a wonderful world in which we have so many ways to generate electricity that are cheaper and more cost effective and easier to deploy. And I would've ever imagined.

People are writing incredible papers in scientific journals. Making the case at switching rapidly to renewable energy isn't a net cost. It will be a net financial benefit, she says, with a head shake of near disbelief. If you had told me five years ago that that would be the case, I would've thought, Wow, that's a miracle.

So back to. We can be part of the integrative or the disintegrative process. Our industry is part of both trends. A really great example is aws. In their recent report and sustainability, they disclosed that they had achieved extraordinary. Advances in energy efficiency, that means that every process that they do consumes dramatically, exponentially less energy than it did some years ago.

At the same time, their net emissions keep growing because their business keeps growing. So even though they are generating less emissions, In every activity, they are having to generate a lot more activity every year because they keep growing. And this is really not particular to aws. We are all part of this.

Every technologist is helping technology grow. It's helping products grow and. Data consumption and our electricity consumption are going up even as other sectors are going down, so we can be part of that disintegration process. We can just keep adding to the electricity drain, to the carbon emissions, to the greenhouse gas effects, by the way, we build our software or we can be part of.

Integrative process and focus on patterns, methods, values, choices and designs that make us green,

and as we speak of greening our software and following where energy leads, then clearly carbon aware computing is a core part of our future as a. We discussed the CarbonHack 22 last week and the fantastic individual initiatives that it sparked around carbon aware computing. And one of the really exciting developments this week for me was that the Green Software Foundation published an article in Hacker None, and that article is called Our Code is Harming the Planet.

We Need Carbon Aware Design Patent. And the article trended number one in Hacker, which is an incredible thing. 50,000 people, nearly 50,000 people read that article and counting. It was such a privilege to be part of writing that article in a collaborative manner with a team because the resulting article, I think is the best introduction I've seen to carbon over computing.

Whether you are. More or less experienced in this field or completely new to it. So highly recommended as a guide. And along those lines, the CarbonHack has now resulted in. Over 70 teams and over 50 completed project submissions, and the ideas are incredible. They cover the full span from hardware to software to vr, to domestic appliances, to infrastructure as code.

I encourage you to visit simply to get a sense of what's possible and what's out there, because I think this is truly the beginning of something exciting, but really all of these individual initiatives that I discussed last week and that I'm mentioning now, are a very critical part of the roadmap to change.

But that roadmap cannot avoid big tech. And we discuss this massive companies, Microsoft, Googles, Apple Meta as if they were monoliths, just this one body thinking together. But the reality is that they are a microcosm of all of our society and just. Society has constructive and destructive forces integrity and disintegrity process.

So do these behemoths with tens and tens of thousands of people exist within each of these companies. And so we find that each of these companies has areas where it is contributing fantastic, groundbreaking, good things to the planet. And also has things that are massive blockages in the road to positive transformation, and both of these things are happening at once.

So I'd like to begin by highlighting Microsoft's recently released green design principles document. Now this is coming from their sustainable design. And this is an example of the part of Microsoft that is working to harness the power of this massive global corporation for the good of the planet.

Their initial principles, or manifest to, as it were for their getting started backpack on digital sustainability, is that. Climate crisis doesn't happen in a vacuum, that it is not just about carbon, but it is about ethics and relationships and inclusion and social justice, and all of these things come together.

That big change starts small. I'll come back to that in a second. That talking about climate can be hard and that digital is physical. These are the organizing principles of their green design principles manifesto, and they. Work on the idea that you should think bigger before you start, which means to challenge the status quo and to put care first, and that you should build better by default, meaning optimized, transparent, and adaptable.

Without going into detail into all these things, I do want to highlight that their principle of big change starts small, says. Start seeing yourself as a change maker. It takes only 3.5% of the population actively participating in nonviolent protests to ensure serious political change. Similarly, in companies and the tech industry, forming groups that can push change, can have lasting.

When they say, talking about climate can be hard, they say develop a willingness to meet coworkers, clients, stakeholders, or customers where they're at. Grow your ability to frame a conversation in a way that focuses on benefits, incentives, other than sustainability, just being something that is the right thing to.

It can be helpful to learn about climate commitments your workplace already has. You can use this to highlight how your sustainability work fits into these goals. So again, I find it very interesting that their design manifesto begins with designing for organizational change and effective communication.

And one of the recent examples of this in m. Is that they recently announced their latest version of Windows 11, and if you go to the announcement, you will see all kinds of features and they are all functional features, wonderful, good functional features, and then towards the end, without any sustainability heading whatsoever, but as part of the main core announcement, they also announce.

They have now made. Uh, change to their windows update experience. Windows update they say is now carbon aware, making it easier for your devices to reduce carbon emissions. When devices are plugged in, turned on, connected to the internet and regional carbon intensity data is available. Windows update will schedule installations at specific times of the day.

When doing so may result in lower carbon emissions because the higher proportion of electricity is coming from lower carbon sources on the electricity grid. And this is under the heading, Delivering Continuous Windows innovation and value. We've also made some changes they say to the default power setting for sleep and screen off to help reduce carbon emissions when PCs are idle.

And it's a wonderful example of those conversations. But it is also an example of the power of individual initiative outside of Big Tech because the director of sustainability for Microsoft made a post on LinkedIn about this and mentioned how they had built on the work of electricity maps. Electricity maps is not remotely big tech rather, it is one of a few innovations.

What time being the other big one of. Teams of individuals who came together to create an API that tells you what the carbon emissions are or other, what is the carbon intensity for the grid in any given time and place where that information is easily available, and their initiative, this API that they created has been used now to change the entire PC l.

Another development recently in Big Tech is that AWS announced updates to the AWS well-architected framework. And it's very interesting when you see, again, the trend, the movement, the I integrative process. The well-architected framework was born in 2012, and it had four pillars in 2015 when it was fully published for the first time released out.

But sustainability, again, has made it into the mainstream of their standards for good cloud computing. So what's new? The updates have done a couple of things. They have updated the prescriptive guidance of on best practices. To reliability, performance efficiency, and operational excellence pillars.

Performance efficiency in particular is close to sustainable design and greening emissions, but in addition, they have consolidated, including for the sixth pillar, for sustainability, they have consolidated. White papers and their best practice descriptions, which means that this is now the best tool you could look for if you wanted to understand how to build on the cloud in a sustainable way.

And to give you a sense of what this looks like, the sustainability pillar has six design principles. Understand your. Establish sustainability goals, maximize utilization, anticipate and adopt new, more efficient hardware and software offerings. Use managed services and reduce the downstream impact of your cloud workloads.

And there are six best practice areas. For sustainability in the cloud, namely region selection, how you select a region is clearly very significant to your sustainability practices, user patterns, software and architecture patterns, data patterns, hardware patterns, and development and deployment process.

So this is a collection of patterns, very practical. Guides that apply to all clouds, not just to aws, and then that drill down into AWS tooling as well. What about Google? Well, they're the smallest of the players in the cloud space and very early on they seem to have chosen to differentiate themselves.

By really committing to green software to sustainable computing. And so I think it is fair to say that out of the big three, Google is the one that has advanced the most in terms of its own footprint and also in terms of its transparency and tooling, and it's moving forward with it at pace. Quite recently, they rolled out a ui, a dashboard that allows users of Google Cloud to see which zones or which regions are currently running on clean electricity at any one time.

So, They tell you, for example, right now Europe, North Europe, Southwest and Europe West are on low CO2 with a nice little logo and a highlight that is clickable and takes you to the actual region cardboard footprint. In detail, it tells you that Europe West 1, 2, 3, 4 are not running on low co2, but West Six is and so on for the entire.

World and they tell you what their methodology is for calculating that, et cetera. So they moved from what was an internal benefit to an external benefit. And then again, this year they. Identified another tool. So they did a really interesting research project. They analyzed the aggregate data from all customers across across Google Cloud and found over 600 tons of CO2 emissions in seemingly idle projects that could be cleaned up or reclaimed.

These are your proof of concepts, your experiments, your tutorials. Moth bolt projects, the projects that were created by somebody 10 years ago or five years ago, and that have just, nobody's touching them either way, because nobody knows what they do. If those could be cleaned up, it would have a similar impact to planting almost 10,000 trees.

So Google's response was to create an active assist recommender that alerts. To your seemingly idle projects and tells you how many emissions they are consuming, how much you are producing waste in the world. This is exciting because the amount of dead data in the planet is huge. And finally this week they announced that they were launching their carbon footprint.

Product. A product that provides customers with across carbon emissions, associated with a Google platform usage. There are a number of dimensions here. They collaborated with atos, Etsy, hsbc, L'Oreal, Salesforce, ThoughtWorks, and Twitter. And if you know what's been going on behind the scenes, you recognize that ThoughtWorks have been the pioneers in the cloud carbon footprint.org tool, which pioneered the measurement.

Emissions across all the clouds. Instead of competing with them, they've learned from them. Likewise, you will know that Etsy was one of the pioneers in measuring CO2 emissions. Google have also announced that very soon they will be displaying for users of Google Workspace. We will be getting the carbon emissions of our usage.

So this is all extremely powerful and, and hugely scalable. As a very quick parenthesis, I think I want to make an honorable mention of an innovation that matter. AI has come up with recently that they announced the 25th of October and. It's not yet out. Out as in available to consumers. But using ai, they have managed to achieve a dramatic improvement in audio compression.

They have managed to make the equivalent of a CD quality mp3. Be stored in a 10th of the data. So where an MP3 compresses at 64 kilobytes per second. Their new process compresses at six kilobytes per second, and they are now hoping to take this approach and the techniques they developed with their AI algorithms and see if they can achieve similar gains for video compress.

Lastly, I want to finish by previewing the upcoming Green Software Foundation Cop 27 Showcase the event, the carbonized software in November the 10th. And this really brings together all the threads that I have been discussing because it involves many of the big tech actors. And it involves constituencies within those big tech companies that are moving to drive the progressive integrity process within the massive companies.

But it also includes some of the thought leaders, some of the key innovators that have already nudged and shaped and made possible. Some of the changes in big tech that I've. And finally, it also brings together and celebrates and highlights the individual innovators who are drawing the new frontiers of green software action.

And it brings those three constituencies right at the top, showcasing a raft of initiatives that will be crucial for all of us interested in this. At this showcase, there will be a number of high level panels discussing the state of the art in thinking about green software. In addition, there will be a launch of a new ISO standard, the software carbon intensity specification, a standardized way accepted by ISO to measure.

The carbon footprint of all software, and this is a standard that integrates both direct emissions and embodied emissions to take account of the manufacturing and disposal of the devices that consume the software. It will also be. Launching the training program I highlighted last week. Fantastic training program, which is now going to be owned by the Linux Foundation, so it will be a Linux Foundation course in green software practice.

It will also be launching a Patterns catalog. I mentioned the Well-Architected framework as a collection of patterns. The Green Software Foundation Patterns Catalog is an ambitious. Project to bring together. The green software patterns in Web, in cloud, and in ai, and this is not just a collection, but a vetting and curation.

Every pattern goes through a rigorous vetting process before. Being published and mainstream. So this will become a fantastic reference point and entry point to good practices. And we will have a preview of the State of green software report that I am working on for the Green, Software Foundation, and it will share.

The state of progress. What we will be doing with the survey, with the high level interviews and with the desk research and how we will be covering the, the level of awareness and adoption, the enablers and the points of friction in the expansion of the green software ecosystem. It will further highlight the launch of a green software speaker bureau where.

Anyone organizing a tech conference will be able to source an authoritative voice in green software. And finally, and very dear to my heart, it will be showcasing the most impressive carbon aware proof of concepts. It will be announcing the winners of the Carbon 22 Hackathon out of the Fantastic 50 something.

Solutions that have been completed and submitted and that you can go and visit and see and watch their videos and see their code and potentially work with them. We will be highlighting the absolute best of the best. So these are the initiatives that might be quoted. In one, two years time by big tech, by government, by regulators when they are scaling up new paradigms of green computing.

To finish today, I want to reflect on the nature of change. I believe that the movement toward a green energy sustainable just society. Is not only possible, but inevitable because the alternative is self destruction and human history is chaotic. It's violent, it is destructive, but ultimately it tells us two things.

The two things that human beings have. No matter what, and through all the pain and all the disruption and all the chaos, one is survive. Whenever we've been confronted with a choice of changing or surviving, we have changed. Not easily, not smoothly, not pretty, but we have survived by adapting. And the second thing that human history tells us is that we have.

Grown in our capacity for complex cohesion where we were able to collaborate only at the level of families or clans. After lots of bloodshed, we learned to collaborate at the levels of cities after even more. Trauma and destruction. We learn to collaborate at the level of nations, and right now, in the midst of all the chaos and Confucian, we are learning to collaborate at the level of the entire planet.

And we're not being driven there by Kumbaya songs. We are driven there by pain, and so I. Whether we are going to reach a sustainable, peaceful, just society by an act of collective will and consultation, or only after unimaginable horrors precipitated by our stubborn, clinging to old patterns of behavior that is the choice before all on earth.

That is our choice. And at the moment, let's face it, we are going for the slightly more painful option. And by slightly I mean. But significantly, as I said, the data suggests we are no longer going for the worst case option. We are in the middle in the really painful option as opposed to the catastrophic touch and go survival option.

We are adapting because we are wanting to survive, and I think this process of change is non-linear. It's like the waves of the sea and the rising of the tide in the. The tide does not rise in one go. The water doesn't just keep going up. It goes forward and it retreats. It goes forward and it retreat.

And when it retreats, it leaves dry land and foam and dirt, and then it comes back up in waves. And I think this is what this process is like. We have these retreats and we have these advances, but overall, the nature of survival means that we will have to advance and I believe we will. So we are part of these waves of change.

And I think as individual technologists, we can be in the back of the sea, completely unaware of the waves themselves right at the front, and be working as though the world was not burning. more likely. More and more and more of us, and very soon, perhaps already, most of us are somewhere in the wave of change.

We are at the base, we are at the middle. We are at the top of that wave and at the base. All that may mean is that we're hosted in the cloud and the cloud is already doing something to become greener. And so we're part of that change simply by being in the cloud, But we're not consciously making any decision.

Or we can be in the middle of the cloud. We are aware that this matters. We want to make a difference, and as far as we can fit it into our schedule, we do. But mostly our lives, our schedules are driven by motivations of a very different nature. Or we can be at the crest of the wave. And the crest of the wave is the thinnest part of the.

We are not with the majority of people, and it's the most turbulent part of the wave. It is full of foam and dirt and lack of clarity in that spot. It's the first bit that meets the sand and the resistance and the wind, but it also has a unique privilege. It has advantage point. It's the only point on the wave from which you can.

The other side where you can see the shore and what lies beyond it. And so I'd like to invite you, and I'm sure if you are listening, you are already near that crest. If you are, keep swimming, swim up, get to the cutting edge. Of that wave and it won't be comfortable and it won't be easy, and it might affect your career choices and might lead to awkward conversations with your colleagues and your friends and your family.

But it will give you the incredible gift of seeing past the despair and past the complacence and putting you in a position to be. The first to touch the New Shores. Thank you for joining me on another episode. And good luck with your adventures. Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

To find out more about the Green Software Foundation, visit Green Software Foundation, and please, if you liked what you heard, do leave a rating and review. It helps other people discover the show and join in the conversation. The more of us are exploring these issues at home, at work, in our free time and in our projects, the greater chances of taking effective action and making a difference in our own corner of the world.

Good luck in your green software journeys and see you in the next episode.

The Week in Green Software: An Introduction

24 oktober 2022 | 50 min

Welcome to our first episode of The Week in Green Software hosted by Ismael Velasco. This new segment on The Environment Variables Podcast will be a bite size smorgasbord of news, events, resources and tools that will help you discover how to get involved in the world of software-focused climate action. Check out the links below for all the resources mentioned in this podcast!

Learn more about our people:

Ismael Velasco: LinkedIn / Twitter

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Course: Green Software Practitioner Training Course by the Green Software Foundation
Company: Greenspector
Website: climateaction.tech

Talks & Events:

Event: CarbonHack22 by The Green Software Foundation

Papers, Books, Articles & Blogs:

Paper: The Sustainability Chapter of the Web Almanac by HTTP Archive
Book: World Wide Waste by Gerry McGovern
Book: Designing for Sustainability by Tim Frick
Article: Greta Thunberg on the Climate Delusion in The Guardian

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcription below:

Ismael Velasco: Hello, and welcome to The Week in Green. Software part of Environment Variables, brought to you by the Green Software Foundation. In each episode, we give you the most UpToDate news and events surrounding green software, a bite-sized smorgasbord of resources that will help you discover how to get involved in the world of software focus, climate action.

I'm your host, Ismael Velasco.

Hello. I'm Ismael Velasco, and this is The Week in Green Software. Each episode I will visit the cutting edge of software focused climate action and give you a whistlestop tool of the latest research and solutions standards and tools. A Bitesize smorgasbord of News and Resources. If you're just starting your journey as a responsible technologist, The Week in Green Software will give you the signpost you need to navigate this fast changing landscape.

And the first signpost I'll give you is the Green Software Practitioner Training course created by the Green Software Foundation. If you have found your way to this podcast, you probably already know that. Digital creations of our hands as we type on keyboards and build software have very real material impacts on the environment that aggregate into a massive driver of climate change.

And you want to do something about it, and this is why you're here, but perhaps a lot of the vocabulary terminology techniques. Are still fuzzy to you. We speak about CO2 emissions of software. What does that mean? We speak about carbon efficiency versus energy efficiency, and we also speak of carbon awareness and we talk about hardware efficiency and how that interacts with software.

When we talk about all kinds of metrics around all this, and finally we discuss all of these things. Very frequently in the context of climate commitments, whether it be the planetary commitment toward net zero or specific targets in particular companies or nations, and just like a junior developer, junior green software developer.

Can also feel somewhat lost in the massive vocabulary and of new concepts that are not part of the day to day of our practice to this day. So this course creates a very. Easy to consume, clear and evidence-based structure that breaks down the concepts of carbon efficiency, of energy, efficiency, of carbon awareness, hardware efficiency measurement, and of climate commitments, including each core component, each core terminology, and in very brief units.

Allows you to understand the landscape and gives you healthfully some good multiple choice questions at the end to consolidate your learning. By the time you finish this very brief course, you will be equipped to. Dive into the nitty gritty of making your software green, not just at the level of technique, but at the level of understanding why those techniques are having an impact and how they fit as a whole with the various other elements of creating green software.

In short, the Green Software Foundation training course will prepare you to consume. This podcast and others from a position of awareness. But the weaken green software is not only aimed at those starting their journey, if you are a pioneer already greening the material impact of our digital lives, this podcast will keep you updated on the latest advances and how they might enhance your own efforts.

And speaking of pioneers, Second resource I'd like to bring to your attention is the sustainability chapter of a hot of the Press Web Almanac Report by HTTP Archive. HTTP Archive are the creators of the Wayback Machine, and they. Are the foremost collection of copies of snapshots of the worldwide Web in the world, and that results in a massive data set that is indicative of the practices of Web developers.

This chapter was authored by real pioneers, Lauren De Verne Gerry McGovern and Tim Frick, who have helped define and continue to shape discourse and practice in green software on a global scale. Many of the green software practitioners that I know. Began their journey and trace it back to the incredible book, Indispensable book, Worldwide Waste by Gerry McGovern, and many of the people who started trying to respond to the moral imperative and call to action of that powerful occur, that powerful voice from the heart turned to Tim freak's groundbreaking textbook.

Sustainable Web design and he was one of the first people really together. A lot of the practices that today we are working on across the field. And Lo de Verne works with a company called Greenspector who have been really. Forefront of new techniques and new approaches to understanding in a very rigorous and actionable way, the environmental impact not just of websites, but of Web applications and of software in general.

And this chapter that they have written as a dream team together, I would say is probably the single best resource on green website development practices available right now in the world. I would recommend it as both an introductory guide once you've done that first training course to get you up to speed with vocabulary, but also as a truly powerful expert.

It summarizes the state of the art in Web related environmental metrics, methodologists and design patterns, and uniquely, in many ways, it takes an intersectional approach that engages not just with. What is called sometimes carbon tunnel vision, CO2 emissions only, and not even more broadly with environmental degradation around water consumption or other chemicals or energy.

But it also engages with the issues of climate justice and accessibility and says that actually building. Sustainable software is about more than just reducing CO2 emissions. It's also about engaging with the whole planet in a sustainable way. In addition to this already groundbreaking contribution of getting us all up to speed and summarizing in a very accessible but truly, Learn it and evidence based way, the cutting edge of thinking in this area.

It also breaks new ground in that it brings together a really vast. Data set that is open source and accessible to others with very user friendly and powerful diagrams that allow you to just go through that report and get a very granular sense of how the Web ecosystem is actually constituted today from an environmental perspective instead of an average.

Or a projection, which is what we mostly use when identifying the carbon footprint of the worldwide Web. We now have, uh, an empirically backed distribution. The authors tested 8 million websites and mapped in every single one of those 8 million websites, the presence or absence. Of all the good practices that we know make for green Web design, and they now allow us as a result to benchmark all of our websites in a multidimensional and rigorous way.

Your site, for example, might be in the 25th percentile for total size. It's beautifully small. It might be in the 90th percentile for number of requests. Every time that somebody lands on your tiny page, a hundred requests go off to trackers, to analytics, to libraries, to logs, and actually the net result is your very otherwise optimized website is actually a net polluter.

Not only does it give us this kind of data for our own individual practice, but for those of us who are active in. Shaping or advocating for policy agendas or in education for sustainable development in reaching our peers and fellow technologists. The data set provided also allows us to prioritize policy agendas.

For example, one. Impactful takeaway for me from this report was that only 10% of sites in the 8 million sample were hosted on green hosting providers. And that is both surprising in some ways and inexcusable in the sense that. There are so many options today, so many options for green hosting from going directly to the big three, which are probably the greenest of options to AWS or Azure or Google Cloud in reverse order of greenness.

So Google is the greenest followed by a zero, followed by aws, but also a lot of the even hosted providers. At the back of the trail are actually hosted in CloudFlare or in other providers that are green, but 90% of sites are not taking advantage of this, and this is significant because. Servers where our information is hosted and from which it is transmitted are the probably the biggest direct contributors of emissions in the software life cycle.

So the manufacturer of the devices that. Use is much bigger. I believe that the aggregated usage of all of the devices is probably greater, but in terms of the direct footprint of your piece of software, where you host it is hugely significant and it's one of the greatest. Biggest and easiest low hanging fruit.

If you were to shift your. Website from a bare metal or a dirty host and put it into a nice green cloud, you would dramatically improve the footprint, the optimization, the water usage, the electricity consumption of your website. So this kind of data point is a useful. Strategic guide for those of us who have to prioritize between all the various good patterns and messages.

Clearly the battle is not even begun, let alone won. We have to get the message out there that we need to be on green hosts, and perhaps even more significantly, we should be working with regulators, with producers, with industry bodies. To require green standards from all hosting providers so that we're not just having to leave it to the consumer to discover which host is green or not.

Although it's very easy and there are fantastic extensions to it, or you can go to eco greater. Put Globe Mellow on your browser or simply go to the Green Web Foundation, which has an eye wateringly large data set of billions of hosting providers of whom, significant proportion that they have identified our Green Web hosts.

It doesn't mean they're actually green, it just means that they are much greener than any alternative and so worth Priorit. Which brings me to a question that has come up again and again in my conversations this week. Why bother? And I mean, you're listening to this podcast so clearly you care, but in the secret recesses of your soul, or in the late night conversations, do you not ask yourself the question, Can I as an individual technologist, Really have a meaningful impact on this train wreck that is accelerating climate change.

This week, Greta Thunberg wrote of the climate crisis. It will take many things for us to start facing this emergency, but above all, it will take honesty, integrity, and courage. So honestly speaking, do we really believe that by turning off autoplay on a video on our webpage, we can actually change the world?

Is the focus on green software not a distraction? Placing the burden of change on individual technologists while the only people who can make a real difference operate on a much grander. What's the point of debating which bit of dust to clean in the middle of a mudslide? It is easy to be discouraged when faced with systemic societal problems, not just environmental, but across the board.

But it is a dangerous fallacy, I think, to think that the only form of power is authority resources. And coercion. Yes, governments could have a massive impact by regulating and enforcing greener ways of living. And yes, multinationals are in the privileged position to make changes that affect the entire planet in one go, but there are other forms of power.

The power to envision, the power to choose. The power to collaborate and evolve, perhaps the power to become. It is a tiny infinitesimal proportion of humanity who have envisioned tools, systems, and approaches that have allowed entire nations to change their energy con. The crisis of our times, I would argue is not least a crisis of hope and imagination.

For instance, if the extent of my hope is that I will be able to move in this room in which I am recording this podcast, but I have no. Ever opening the door that leads to the rest of my home. Then all of my planning, all of my solutions, all of my initiatives will be confined to these four walls. The. If my hope stretches further, if I believe that it is actually possible for me to roam the confines of my entire house, well that will be very good from the point of view of going to the toilet and having a shower, it will be fantastic from the point of view of feeding myself in a healthy way.

Above all, it will open up a huge range of options for interacting with my space. I can sit in the sofa, I can put some music, I can watch tv, et cetera, etc. I can even perhaps order food brought home and then my possibilities really expand. But if my level of hope only stretches as far as this house, I will never again place my bare feet on the grass.

I will not lie down at night and count the stars. I will not go and visit my children or my friends because all of my planning, all of my choices, all of my actions will be confined to this house. The extent of our ability to act is closely inseparable, in fact, from our ability to hope and to imagine, because if I can imagine not just leaving the house and I cannot just imagine.

But actually hope it, I can believe in its possibility even if at this particular moment I don't really know how to unlock that door because I lost the keys. But if I believe that I can actually get out one way or another, my initiatives, my problem solving will not be focused on which room I will sleep or eat in.

It will be focused in how to open the window or the door or find a way through to the vast landscape and the entire planet that awaits me. Similarly, those few people who have been able to see further, who have been able to imagine to conceive of possibility have allowed us all to. Beyond the confines of our immediate situation between the four enclosing walls, and gradually step into that space because in contrast to the very small number of visionaries creating collective planetary possibilities, it is in fact the aggregated choices.

Of millions of consumers to give their belief, their hope, their commitment, their effort to those ideas, which has been the engine of massive societal transformations. It is individual, often on coordinated, but values driven choices that have led companies and governments to adopt net zero target. And they may be moving too slowly and there may be a lot of green washing.

But the fact is that as a planet, we have agreed on a direction of travel and we are moving in that direction. It is those myriad of tiny individual choices that have aggregated to make esg. A corporate requirement, environmental, social, and governance indicators that have changed regulations, supply chains, manufacturing standards, and have made renewable energies power a growing proportion on an industrial scale at a time when, at the beginning, the very idea that solar could be.

Practicable, let alone scalable was elusive. But those few people there to hope and were able to imagine, and they gave us all the possibility to choose between those visionaries and that growing mass ethical consumers, the visionaries with the power to imagine and us with the power to. It's a dense network of networks of powerful collaborations, mobilizing to educate one another, to persuade one another, to track and implement and verify.

The civil society sector has been growing and growing and growing, not just in numbers, but in influence, in sophistication, in capacity to mobilize. To account for change and obstacles and to make that vision that the innovators and the dreamers and the spiritual and ethical leaders in the real sense of the world have seen for us and make it, reach us, make it, reach the.

And what all of these things have in common is that they allow us to be, not just to run an auto pilot, but to be true to our own best selves, our highest values, and thus not just exist, but really live in the end. The power to hope and to dream, the power to mobilize and collaborate and the power to. All adopt the power to become who we can fully be individually and ultimately as societies and.

The whole human race, which brings me to the Green Software Foundation CarbonHack 22, which is another thing that's coming to my radar this week, which I think is significant and of wider interest even for those listening who will not participate. I have been around the hackathon scene for a few years as a software engineer and as an I.

And this event is, in my view, a truly exceptional opportunity to envision, to collaborate, to choose, and to become. It is an inspiring confirmation of these other forms of less formal, but no less significant forms of power On the one hand. You have participating in this hackathon, the massive corporations with the power to have large scale impacts very quickly.

Intel, Microsoft, Goldman Sachs, Accenture Globe, and UBS ThoughtWorks. Together, they have made this a hundred thousand dollars challenge, which let me tell you is a lot more than is usual in Hackathons. With a $40,000 first price, this is a great motivator that amounts to more than just encouragement. It becomes a form of seed funding.

Any idea that really, really captures the imagination of the judges from these massive planetary organizations will receive. Money to be able to liberate time or bring in talent or buy equipment that allows their proof of concept, their dream, their vision, that hope beyond the borders of the four walls of our present to embody itself in action.

But perhaps more significant than that price, than that seed funding is the fact that these judges, that these big companies will be genuinely paying attention. I have had people reach out to me from some of these companies. Just wanting to participate as engineers in the hackathon, not for the money because they would not be eligible as sponsors, but simply from the enthusiasm to make a difference and others who are really paying attention, who want to see.

New approaches and solutions that they might support, incorporate scale, that would allow us as a planet and their mass companies to be more responsible, impactful citizens. A single idea that grabs their imagination could be the beginning of massively impactful change. And you also have the collaborative networks of change makers that can socialize, not just the tools that come.

This visioning process, this imagination, this hope in action. We have Code for All, which is the largest civic tech network in the world who are helping disseminate this happening across their amazing network of technologists trying to improve the world beyond just the environment. We have the Green, Web, Foundation, who are.

One of the greatest hubs. Proof of concept innovation. They themselves, just, if you go to the Green Foundation, GitHub re repository, you will see a whole range of things that all amount to windows into the future, to exits into hope threads that you can pull to advance closer toward a greener. And of course, the Green, Software Foundation itself, which is mobilizing to help reach those individuals across all three populations in whose power lies to imagine, to initiate, to sow seeds.

And then you have the innovators, hundreds of volunteers from both the global north and the global south with decades of experience or just starting. Finding one another and exploring what would be relatively simple ideas with the potential to really change the world or one corner of it in a significant and environmentally healthier way.

And already there are ideas, which, when or not I can see have huge potential. The purpose of this hackathon is to promote carbon aware computing. What does that? Well, what we really, really need in the planet is to drastically reduce our energy consumption and our consumption in general, which is. Really driving our planet to multiple cliffs, not just around climate change, but around the exhaustion of valuable resources and the degradation of societies, the exploitation of people, all of the very unhealthy patterns that create a system that is no longer functional for us, if it ever.

But to get to that point where we have achieved equilibrium in our consumption and production patterns is not gonna be like pressing a button. Even if we wanted to, even if we convinced absolutely everyone, there would still be a period of transition of experimentation, of learning, and in this transitional period.

We are all children of the half light with one foot in the sun and the wind and the water, and one foot in the shadow of a coal mine. Indeed, our energy mix is and will remain for a good while mixed. Some of it will be clean. Brought to us from by nature's hand, and some of it will remain dirty in the legacies of a dying world, so we cannot control.

When the sun shines, we cannot say to the wind, to Blauer right now or make the tides rise at will to power our washing machines and our computers and whatever you are using to consume this podcast, we have to follow nature. We are partnering with nature. We are not controlling it. Therefore, carbon awareness means that we ensure that the electricity that we will consume no matter what is sensitive to the flows and ebbs of renewable energy in the electricity grid.

So, for example, you could make sure that that washing machine only runs. When the electricity grid in your area is being powered by the sun or by the wind, And stops when it is powered by burning fossil fuels. Or you might have a massive computing job, which is happening in the cloud. And if you know a little about cloud computing, you know that there are multiple geographic zones where that job could take place.

So carbon where computing would allow you to say, I will run this computing job, not here where I am at this moment when the electricity is dirty. Instead, I'm gonna run this computing job in Sao Paolo or in Sweden where the electricity happens to be green at this particular. So this is carbon aware computing is the idea that we still use energy.

We haven't yet reduced our energy footprint in that sense, but we ensure that the energy we use is as clean as possible and that we avoid as far as possible the peak moments where the electricity is most polluting. The Green Software Foundation created this fantastic API that allows you to know when and where the electricity is being powered by renewable energy and when and where in the planet the electricity is.

Highly polluting and this hackathon is all about hooking to this API and coming up with solutions that allow you to apply this carbon aware computing. Now the solutions that people are working on are super exciting. Some of them are really quite small. For example, a plugin for the serverless framework.

So this is an infrastructure as code framework where you basically write a few lines and say, I want you to create, to provision a computing environment for me in. AWS or in Azure or in Google Cloud particularly, at least I am aware of it being used in AWS for what are called Lambda functions, serverless functions that only are activated when activities happening and then go back to sleep.

So they're very environmentally friendly and very efficient and powerful and composable. The research shows that something like 80 or 90% of all serverless functions, Lambda functions are deployed via serverless. So just a little plugin in this very, very niche area that allows you to ensure that those Lambda functions, those serverless functions only run in the greenest electricity zones around the world.

Could have a massive, massive impact because a lot of computing is being run through lambdas. Other ideas, for example, are around completely different part of the computing stack. So someone is creating green mold UI components. So the idea is that you create websites by composing various pattern. So you use a library like Tailwind or like Bootstrap or like Material ui, which already has pre-built styled components like forms, like buttons, like headers, et, cetera etc.

And those can be very attractive and very beautiful, but also potentially a little heavy. If you could have versions of the very same components of the forms or of the buttons or the nav bars that became simpler and lighter when the electricity was dirty, you would shave a tiny bit of electricity and of CO2 emissions.

And we're back to that idea and why do we care? Why would taking one gram of CO2 out of a webpage make a difference? Well, if that webpage is visited a hundred thousand times each day, then over a year, that is three and a half million tons of CO2 that your one gram reduction has achieved. There are so many other ideas.

There are some very, very simple ideas which are nevertheless really powerful. So simple schedulers, simple alarm belts as it were, that tell you when the electricity is optimal or so suboptimal. So for instance, in my example, you could get a scheduler that allows you to choose when. Put on the washing machine, but also when to game, and when to watch videos and when to engage in the most fun yet most intensive activities in your digital life.

So you can be a responsible gamer or for electrical equipment or for crypto mining if you're going to be mining crypto coins. Consuming a lot of energy and that's not a good thing to do. At the best of times, but if you are gonna do it and people are doing it at scale, wouldn't it be amazing if you could ensure that that mining happens at a time when the grid locally is running on renewables or stops.

You stop your mining when the grid is particularly dirty or when your emissions budget has gone past a certain level. All of these are little glimpses of hope. There are little windows of imagination. There are possibilities where, where a million collective individual choices might eventually lead us.

If those of us now engaging you, listening, the organizations that are participating, mobilized to let, to empower people to be able to make those choices. Maybe one of these ideas really takes off in a huge way. Maybe 10, maybe five of these ideas take off in a small. But still aggregate, not just in making a direct impact and difference, but in stretching our capacity to envision and to dream.

And the result is that a new solution comes on the back of these proof of concepts that actually scales. I want to finish by telling you a story and it's a real story. It took place in Geneva in the year 2000. I was there for a United Nations summit, and this was the Geneva 2000 Millennium Summit, which was five year son for the very first social summit that looked at social develop.

And it was taking forward this, and there were so many thousands of non-governmental organizations. I was representing the International Association for Community Development, and it was a fantastic space to meet incredible change makers. And there's. That stays with me. Out of everyone, there are hundreds of people I met there.

There is one person that I saw that I didn't even speak to properly. I had, you know, 10 seconds of conversation with. Mostly I just saw her walk against the flow of people to get to a meeting. She was from Somalia, and Somalia in 2000 was on fire. There was a terrible, bloody civil war and a mass immigration of war refugees.

This woman, too, had left Somalia, but she hadn't left Somalia to escape. She had left Soma. To launch an incredible vision. Think if you can, if you are old enough or if you've watched on series and movies where the world was technologically in the year 2000. Our phones, I think were barely Nokia, and the most that you could play was snake.

You probably could do a little more by 2000, but you were a far cry from smartphones and they were all in rich countries. What this Somali woman entrepreneur. Came to say in that conference, and I don't even know her name was, invest in mobile technology in Africa, come to Somalia and put in the infrastructure and bring in mobile phones.

At that point, that was sheer fantasy. Imagine starting a business in our war. Imagine bringing mobile phones to a continent that mostly struggled to have electricity and running water, and certainly no fixed phone, which in the year 2000 was still by far the primary way of phone communication. So people said, You haven't even got to the stage of fixed landlines.

You're convincing us that there is a business to be made installing mobile infrastructure and selling mobile phones to the poorest areas of the world. It really seemed ridiculous, and I don't know who listened. The vast majority did not. I was hugely inspired by her, but I have to confess, I was also hugely skeptical.

Fast forward 2022, and Africa is the pioneer in the application of mobile technology to social innovation. They are the. In mobile wallets with 60% of all mobile wallets in Africa, the incredible innovations, the interventions, the way mobile phones were used to save. The entire populations from starvation by, um, a version of universal basic income by sending to all these subsistence traders who were hawking wars and who suddenly could not survive cuz they could not sell and they could not go out because a pandemic was killing people.

They used phones to put money into all of those pockets and lifted an enormous number, enormous proportion of the country. I believe it was in Togo, from starvation and from poverty. Mobile phones are everywhere. I lived for three months in a very remote Maasai village in Tanzania. There were. The only technology that existed in that village was one very old cassette tape running on batteries.

A car that visited once a week or so, maybe three or four. Collective electric torches, battery run, and a few mobile phones. Everything else. Was built from nature and you follow the stars and you follow the seasons and you cope with a drought and you lived entirely by the flows of the natural world, except for mobile phones.

Those Maasai tribes were so entrepreneurial and globally aware and savvy, and they harnessed that technology in incredibly powerful ways that Somali woman was a visionary. If you listening to this podcast, if you got to this moment in this podcast, I know one thing about you. You care about our. You are not just passively despairing, but actively taking action, big or small to make a difference.

And I know you're not alone. That woman was a single undefeatable voice in a conference that wasn't listening. You are part of an incredible community of change makers, whether you've already found each other or you haven't met a single person who is thinking about digital. But you are not alone. Every single person listening to this podcast is another voice in the planetary conference room advocating with you, experimenting beside you, adding their impact to yours.

So I will lend by highlighting the most wonderful space I found to connect with others who are walking the very same path in very different shoes. It is called climateaction.tech, and it's Epicenter is a community on Slack comprising thousands of technologies from. Programmers to product owners, to designers, to investors, students innovators, exploring how to green existing tech across any and all domains.

This includes people who have just begun to think about these issues and also world authorities in this space. In an atmosphere of mutual encouragement, support, humility, and dynamism. That amazing Somali woman accurately envisioned the future. She was alone, but you could tell she knew she wasn't. We are not either.

And armed with this knowledge, we can share in her certainty of impact. How much more when we have so many more channels to find and collaborate with one. Not least The. Week in Green. Software spread the word. Hey everyone, thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts.

To find out more about the Green Software Foundation, visit Green Software Foundation, and please, if you liked what you. Do leave a rating and review. It helps other people discover the show and joining the conversation. The more of us are exploring these issues at home, at work, in our free time and in our projects, the greater chances of taking effective action and making a difference in our own corner of the world.

Good luck in your green software journeys and see you in the next episode.

Green Networks

12 september 2022 | 42 min

👉 State of Green Software Survey - click this link to access! 👈

Environment Variables is back! Chris Adams hosts our Green Networks focused episode and he is joined by Eve Schooler, Principal Engineer and Director of Emerging IoT Networks at Intel and Romain Jacob of ETH Zurich. They discuss how can we reduce the energy produced by networks? How could we leverage current research to make the internet more energy efficient?

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Romain Jacob: Website / Twitter
Eve Schooler: Website / LinkedIn

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Talks & Events:

Talk: The Internet of tomorrow must sleep more and grow old by Romain Jacob and Laurent Vanbever.
Event: Internet Engineering Task Force (IETF) Hackathon and Codesprint London (IETF 115)

Papers:

Paper: The Internet of tomorrow must sleep more and grow old for Hot Carbon by Romain Jacob and Laurent Vanbever.
Paper: Internet Backbones in Space by Giacomo Giullari, Tobias Klenze, Markus Legner, David Basin, Adrian Perrig and Ankit Singla.
Paper: ITU GHG Recommendations for the ICT Sector
Paper: Toward Carbon-Aware Networking by Noa Zilberman, Eve M. Schooler, Uri Cummings, Rajit Manohar, Dawn Nafus, Robert Soulé and Rick Taylor.
Paper: Greening of the Internet by Maruti Gupta and Suresh Singh.
Paper: Approaches to calculating network / website energy and carbon by David Mytton.

Open Source Projects:

The Watttime API
Electricity Map API
The Green Software Foundation Carbon Aware SDK - the SDK offered by the Green Software Foundation.
Grid Intensity Go - a Golang CLI and library, for use in projects to extend them for carbon awareness.
CO2.js
The Green Web Foundation IP-2-carbon-intensity API

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!

Connect with us on Twitter, Github and LinkedIn!

Transcript Below:
Romain Jacob: In many internet service provider networks. So kind of the edge of the internet, where we have strong, seasonal patterns into traffic, they are allowing fruits. There are many of those small networks. So the benefit you can get there actually add up pretty quickly. And if they, if they don't seem.

Interesting. If you look at a single network, if you apply those principles everywhere, you can achieve very large effect. And that's something, every network operator should, should have a look at if only to reduce their energy.

Chris Adams: Hello, and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode, we discussed the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Hello, and welcome to Environment Variables. The podcast about green software. I'm Chris Adams, your host today, I'm filling in for Asim Hussain. And on this episode, I am joined by Eve schooler of Intel. Hi

Eve Schooler: Hi.

Chris Adams: Eve and Hama, Jacob of et Zurich in Switzerland.

Romain Jacob: Hello, everybody.

Chris Adams: And today we're gonna discuss the levers available to us for greener networking.

Now, if we go by the figures from the international energy agency, data networks used around 250 terror hours of electricity in 2019. And while we don't have the figures yet for 2022. This same agency is projecting a estimate of around 270 terawatts to be used by the end of this year, which for context is more than the entire electricity usage of Germany, the fourth largest economy in the world, this results in a significant environmental impact.

And thankfully, we'll be talking with some people who've been spending a lot of time thinking about where the biggest levers are to do something about this in the context of the climate crisis. And they'll be sharing their research on how we can end up with greener more sustainable networking. But before, before we dive into the specifics, let's do a quick round of intros.

So maybe you can introduce yourself and your work at Intel.

Eve Schooler: Hi, thank you for inviting me. I am a principal engineer and I'm a director of emerging IOT networks at Intel. And my current work focuses on evolving the internet toward a sustainable edge to cloud infrastructure. My background in expertise is primarily in networking and distributed systems. And although I've spent much of my career in industrial research, I currently straddle a business unit at Intel called the network and edge business unit.

And as well as the corporate strategy office, where I'm responsible for sustainability, innovation and standard. And I'd say that something that's really colored my experiences that I've spent much of my career heavily involved in internet standards and standards bodies, such as N and the I E TF, which is the internet engineering task force.

Where earlier in my career, I developed control protocols for internet telephony and multimedia teleconferencing, but at present I'm heavily involved in and leaving a working group focused on deterministic networks and their extension to operate in. Networks. I'm also actively involved in the open group's open footprint forum, and that aims to standardize the carbon footprint data model.

So you can hear sort of my internet hat as well as the sort of sustainability hat, both coming to the, for.

Chris Adams: Cool. Thank you Eve. And thank you once again for getting up at the crack of Dawn to join us from California today. Okay. I know it's somewhat more sociable time, so maybe I'll just give you a chance to introduce yourself and then we'll dive into some of this.

Romain Jacob: Sure. So. I've been studying at et Switzerland since seven years for now, the first five I spent during my PhD on lo wireless communications for embedded systems. So the, the question of how to save the energy was, was kind of core to everything I was doing there. And after I graduated in 2019, I moved on to more internet.

And most recently I've been interested into how can we reduce the, the energy consumed by networks in general, with a focus on wired networks. And I'm trying to see to which extent the concept of low power wireless networking could be translated into wire networks and how we could leverage that to make the internet more energy efficient.

Chris Adams: Cool. Thank you. And for context, I found about raw math through the conference called hot carbon, which is a kind of energy and carbon nerd conference online where his paper. The internet of the future, that should, what was it? The internet of future will grow old and sleep more. Was, was that something like that?

That tickled me. I should introduce myself. My name is Chris Adams. I am the, uh, I am the, the head of the policy group at the Green Software Foundation. And I'm also the executive director for the. Green Web Foundation, an NGO based in the Netherlands campaigning for a fossil free internet by 2030. And now, you know, all our names.

Maybe we should jump into the actual topic of greener networkings. So Eve I, I first came across your work with, uh, one of the, with a recent paper called towards carbon away networking. And in that paper, There was actually some useful information about setting the scene in terms of how much of the internet, or how much of the tech industry, what, what, how much network makes up of this compared to say data, data centers, and computing.

I wondered if you might be able to just expand on some of that, because this is a useful piece of context. And previously we've spoken primarily about data centers rather than the kind of aggregate impact.

Eve Schooler: Absolutely. I mean, in the press, we hear a lot about how data centers are consuming the world in terms of their energy usage. And it's, it's interesting because there's studies that suggest that networking is as large or larger. Than the data center in, in terms of its consumption. And there, even when you look at, when you dig into the numbers, a little further networks have been estimated to consume as much as one and a half times as much as data centers and, and even within data centers, networks already account for between 10 and 20% of the energy there.

So those numbers set the context, which is why it feels like networking deserves some further investigation. And solutions.

Chris Adams: So, this is one thing that I might, I might ask you a romantic come into, come into on this, because previously we've heard while there are tools, like say CO2 JS or websites like website carbon, which will give you an idea of the environmental impact from say, What looking at a website or you'll see stories about things like say the environmental impact of watching say Netflix, for example.

But as far as I'm aware, the actual energy usage used by networks tends to be relatively, uh, has historically been run something which has relatively stable compared to the other usages from that. And I wondered if you might be able to expand a little bit on that part there, like, is there scope for a change or is it just a static figure that we have, no matter how much we use the internet.

Romain Jacob: Yeah. So there is two, two points to this. What, what, one thing, which is very true, that relates to what you were saying is that the, the energy consumed by networks at any point in time. Say in a time span of a year or months then to be fairly constant. So there have been a number of studies that shows this, that the energy consumed by the network is essentially independent of the load.

So if you are using 10% of the capacity or a hundred percent, essentially the same thing, the, the reason for this, so. The stable number, um, has increased over time as we've scaled up the networking infrastructure, but for the, for giving infrastructure, the energy you consume. So the power you draw at any point in time tends to be fairly constant.

That's, that's kind of worrying because we are typically operating very far from the hundred percent point. So we tend to over provision our networks, meaning we want to make sure there. Much they are capable of much more than what we typically ask, which means that we essentially use a lot of energy all the time.

Whereas we are using the infrastructure fairly little.

Chris Adams: So if I follow what you're saying, this is a little bit like maybe 10 or 20 years ago before kind of pre-cloud where you might have. A big fat, chunky server that you have and you plan for the maximum capacity. And as a result, it may be that if you look at say the usage you have there, because you can't really scale that server down, you've got that same.

You've got that kind of relatively core grained amount of energy usage. Is that the model that is actually helpful to think about when we are looking at the net network usage.

Romain Jacob: Kind of, yes. So when you talk about compute and you kind about what servers are doing a lot of, depending on the workload you're, you're actually working with, but. You can scale up and down the power, depending on how much computer you're producing. Whereas when you're looking at networking, it doesn't really work like this because you have very little compute that actually happens in the network.

What, what the network consumes energy for is to powering the memory on which you read the, the routing information, for example, in the optics. So reading packets in and out and all those things. Have are essentially dominated by idle power, which is the power you draw, just to turn things on.

Eve Schooler: I wanted to make a big distinction, which is that much of the core network has this property that whether or not you've got high usage, you know, lots of packets flowing across it or not. It is gonna have this constant amount of, uh, draw energy draw. But the wireless network. Inherently has was taught from the beginnings of its design to be fairly adaptive.

So I think that's the distinction. One of the distinctions being made here is that wireless and wired networks behave quite differently in the face of congestion or, or even just traffic on the network.

Romain Jacob: Yes, it's very true. Very true.

Chris Adams: Okay. So if I, if I'm were to apply some kind of mental model for this, you might think about kind of like backbone networks as almost like kind of current Deion, current, constant the entire time. And then as the closer you get to the surface or to. End users. You might have a bit more kind of spiciness going up and down.

And that's like a way to think about where some of the levers for reducing. Impact might be. So if we're speaking about consumption and that gives us some way to think about the energy used, there is another kind of source of leverage, which is the carbon intensity of the energy itself. And as I understand it, Eve, this was some of the work that was presented at hot carbon.

And some of the work that the paper that you've been con contributing to towards carbon away, networking, maybe you might expand on some of that because there are some really fascinating ideas I found in that.

Eve Schooler: Sure. I mean, as you alluded to earlier, there's been at least in the data center community, an awareness of what is the quality, if you will, of the. That is being drawn from the socket. Uh, and what I mean by that is what is the carbon intensity? How low a carbon intensity can we get towards using clean energy or renewable energy?

So the lower, the number, the better and data centers in recent years have begun to experiment with. And now are operationalize the idea of time and space shifting workload. To align with the availability of clean energy. That's interesting for a, a bunch of reasons, the most important of which is that as Roma was saying earlier, the, the footprint for data centers and I C T you know, information, communication technology continues to grow.

And especially in the face of all the increased amount of data that we're sending across networks. And so pairing a data center with renewable energy. Enables us to reduce the carbon footprint of those data centers as they consume more energy. But similarly, in the electrical grid domain, we also have more and more integration of renewables and in places like California, which is where I'm based and in Germany and other parts of the world where that integration is happening quite rapidly.

There are parts of the day where there's way more renewable energy than we can possibly consume. And so it just gets dropped on the floor. It gets wasted. And so there's been this lovely pairing of, you know, we've got an entity that's consuming a lot of energy going to renewables and the renewables, creating excess and looking to somewhere to consume that.

So if you can think of compute as load balancing or as being virtual batteries for the data centers and it begs the question that if network. Are using one and a half times as much electricity. Why aren't we using those same techniques in networks? And so there is this growing awareness of where are the places where we can put renewables in order that networks are consuming cleaner energy as well as can we, and is it worthwhile to time and space shift the transmission of our network loads in order that they have a smaller carbon.

Chris Adams: Okay. So this is actually quite interesting to me for a number of reasons, because if I understand the most common tools we might use for. Hacks of data around the world. We don't have that much control ourselves directly. So I might send something to the next hop, but there's something like the border gateway protocol that decides where the next hop is and so on and so on and so on.

So I might have some indirect control there. And there are say, clean slate attempts to redesign parts of the network, or even introduce a notion of kind of path awareness from connecting. Say something you have now here to maybe a website. So you could take a kind of greener route like you have here would either of you have anything you might could so share there because.

As I'm aware things like the border gateway protocol, BD BGP has maybe one main criteria that you have here. And it sounds like we might want to be able to use multiple criteria. Like I care about latency, but I also want to balance that with carbon intensity, for example, or even cost I'll open it up to see if anyone has anything they might wanna share here that might.

Romain Jacob: Yeah. So, as you mentioned, the GP is kind of like the glue that, that connects the internet together, and it is been suited how to extend it, improve it and change it. Over the past, I don't know, 30 years or so various directions for various objectives. Usually security is the main concern that people have with BGP.

But most recently there have been some different additive different approach to go away from BGP. One example of that is the Zion network or networking principle that is also coming for me. That that is trying to let the end. Pick which route the, the traffic should go through in the internet. So it's an idea that is generally known as source routing.

So like the, the source of the traffic should say, I want my traffic to go through this network. Then this network, this network in until I reach my end point. And once you have this tool, so this is good for security purposes, but if you have this, you can also use it for carbon awareness. You could also say, I prefer to go through California because they have a lot more renewable energy rather.

I dunno, some other state in the us, they may not have as much,

Eve Schooler: Like Virginia.

Romain Jacob: for example.

Eve Schooler: West Virginia.

Chris Adams: Yeah.

Romain Jacob: it empowers you to do this. If you want to do that, it is possible.

Eve Schooler: Another way to have a mental model about this is network performance has often been categorized or has attributes like latency, like packet loss, like jitter the variance in the latency. As metrics for the success of transmissions across the network. And so the idea is how to teach things, teach protocols, whether it's BGP or other parts of the network fabric, and even other parts of the network stock about carbon intensity.

So that there's this so that it is carbon intensity is another metric. That is a first class metric. In the selection of these routes, whether it's routes or whether it's software usage or whether it's scheduling. And so in some ways we need to teach many of the protocols that we know and love in the internet about these additional options so that we can do joint optimizations, or we can create source routes as a Rama was suggesting.

But it, it is really a very simpatico with this idea. Deterministic networks in the small, some of the work that is being done around time sensitive networks, for example, is all around selecting paths or subnets that have the lowest latency and even creating multiple paths in order to ensure that packets get delivered in time.

But what if the constraint that we really wanted to optimize for in certain circumstances was the carbon intensity and it really. Also leads us to ask, you know, how do we educate all of this software? That's out there to not only carbon intensity information about the loca, you know, carbon intensity is very location, time and space specific, but how do we also enable our applications to say how time elastic they are in order to be shifted around or delayed?

So they're both issues to.

Chris Adams: All right. So if so, what it's, what this seems to be speaking to is this idea of moving from maybe just one set of criteria to why this set. So for example, If I cared about latency more than cost, I might care about things like say, if I'm doing a video call in Australia, I might be prepared to care more about latency than the cost of say for carbon.

And if I cared about say, making, doing a download of Netflix for, for a video, I might, if I'm not gonna watch it right now, I might say, well, I care more about the cost and the throughput, not, not latency and making it go through a kind of green route. So I would rather have some. Low carbon internet trick shot bouncing through the greenest possible places to end up on my computer for when I come home tonight, for example, or something like that, that seems to be some of the directions and this might be heading towards.

Okay. Wow. That's quite exciting actually. So this also speaks to these ideas of. Maybe changing how we might design software in the first place and having different tolerances. Maybe if you might be to speak to some speak to this bit, cuz you, you mentioned this phrase, I haven't come across before time sensitive networks.

So maybe you could expand on some of that and some of the delay on, on the flip side of that, which presumably will be delay tolerant networks for this.

Eve Schooler: Yes, time sensitive networking community is for example, some of the work that I've been involved in in recent years comes out of the internet of things group at Intel and in particular, the industrial internet of things. Context where you've got control systems that have very, very low latencies, less than a millisecond, for example.

And so you're talking about subnets, very small networks, but some of the work in the IATF is about, well, how do you across factory floors? How do you enable them to be time sensitive across subnets, which may have different underlying technologies. And so all of them need to be taught sort of how to do this now in the time sensitive networking world, often one of the strategies, in addition to, as I mentioned earlier, multi-path.

Having multiple paths by which packets can go between sources and destinations. So there's redundancy for reliability, but there's also the reservation of resources along the path. And for that, it starts to look a little bit like what we were talking about when Roman was referencing the cion work. What if you knew how much time it took you along each.

And you had a certain budget. Well, you could send out a query message between a source and a destination to understand along the way, cumulatively, how much latency am I going to encounter? Reserve enough buffers in those queue. And eliminate congestion along that path. And that's sort of what time sensitive networking in the small has been doing.

Now, the kinship that it has with delay tolerant networks is that we wanna expand these time sensitive networks. We wanna teach them about energy usage and energy awareness, carbon awareness, but these delay tolerant networks. Back to the data center analogy, data centers are shifting their workloads to align with when the sun's shining or the wind is blowing.

And, and so they're holding onto their workloads. There's been a longstanding project in the networking community around delay tolerant networks that have been designed primarily for deep space. And because, because you, you know, routers come and go because planets align in certain ways or satellites align in certain ways and they're not always there.

And so that's why they have to be delayed tolerate. They have to there's this. Dynamic of the availability of the resources. And so the question is, could we be using delay tolerant, networking in context for more than just satellites, uh, in this context where we wanna align with, with the availability of clean energy.

Romain Jacob: Yeah, no, I. I totally agree with what Evo was saying and this, this idea of data or more de tolerance, more tolerance in general in networking is necessary to progress towards more energy efficiency or carbon efficiency. This is essentially what wireless networks have ever was saying have been doing forever in wireless.

How do you save energy while you keep things off for as long as possible, right? You, you just, you make, try to make sure that when you turn on your radio, it's. Achieve something useful and then packets, well go through and as efficiently as possible. And if you think of it, it's extremely easy to push the energy efficiency, right?

What do you do? What you, you turn off for 90% of the time, and then you schedule very tightly the time where you stay on the problem is that that induces delay, right. And your application is to be able to tolerate that delay. And embedded systems IOT, all this work, this, this field has been working in different tradeoffs to play with this so that the, the application performance does not take great too much due to thes by the networking part.

And that's like the story of what I've been doing during my PhD. The, the problem is that the internet networks, the wire networks, they've been building a different paradigm. It was. All about reliability. It's, it's, it's been designed to be as reliable as possible. Like if we have a nuclear war, the internet should still work.

Like that was the initial idea, right? So we need to make sure to provide all level of reliability possible to sustain anything, but we need to get away from this now because the cost of this is that we over provision everything. We have a lot of redundancy and we use very little of that. So some of the things that I'm, I'm thinking about together with several colleagues now, is that okay?

What if we were to redesign indeed those wired networks, so that reliability. It's not something we, we get rid of, but we modulate the requirements we set there and say require reliability is just one objective. How much performance degradation are we willing to tolerate in order to save an energy, to give a very concrete and simple example, most traffic on the internet is driven by.

Human activity. Right. And human activity has a very clear seasonal pattern. We, we use the networks more a certain time of the days and not at others. It's very easy to, to think that we could turn off part of this networks for certain part of the day, because we don't need that much bandwidth. And if we do, we might be able to tolerate a bit more delay than, than at peak hours.

It's very similar to turning off the public lights on the streets, you know, at night when nobody's driving, right. It's the same principle.

Eve Schooler: Or even in your home, right? The analogy of one's parents growing up, don't forget to turn off the lights. It's exactly the same analogy.

Romain Jacob: Yeah, it it's, it's the same idea. Right? And there is no reason this gonna be done. I we know we can do it. The question is, how far can we push it? And, and one, one limit limitations factor. One blocking factor at the moment is how quickly we can turn things on and off, uh, because turning, switching on a router or switch.

Takes as of today in the orders of several minutes. Right? So it's not something that you can just do multiple times per, per, per hours or so, because essentially your network will be completely unor. It can be changed if we were to change the hardware. If, when to change the operating system, we run on those machines.

We could improve on that. How far can we go? This is kind of an open research question at the.

Chris Adams: That's really, really helpful. And thank you for explaining it in that way. I presume this is the, the, the internet must sleep more part of your paper where the internet must sleep more and grow old. Right? What you're talking about here is actually the idea of things. Not necessarily being away there all the time, or the idea of liability, moving to different parts of the system is actually quite an interesting one.

And one we've seen with the cloud.

Eve Schooler: And actually there are a couple of seminal pieces of work that I think we can look back on as really setting us down this path. So for example, you know, Mati Gupta's work in 2009 from SICOM was all about, you know, energy efficiency and, and beginning to examine. How much could we save if we began to turn things off, another interesting paper that's been influential is Dina, Papa Gino's work, access points.

And whether parts of the wireless edge network could be, you know, in internal to buildings be turned on, on demand. So it's sort of the opposite idea of like turning things off, but it's like you turn things on, on demand. And as romance suggested, we understand the patterns. Usability of those things. We know when people come into buildings, whether they're in their homes or whether they're in their offices.

And so an on demand infrastructure at the very edges of the network makes a lot of sense. So those are two pieces of work that certainly have influenced my views on, um, teaching devices, how to sleep.

Chris Adams: Okay. Cool. And this idea of matching demand to supply Eve, you mentioned about kind of speaking in California. I mean, just this week we saw a really good example. Demand response where you saw like a Kaiso the grid operators basically say, Hey folks, we're about to kind of hit a blackout. Can everyone please just turn things down a little bit.

And if I understand it correctly, we were able to see basically one of the largest grids not fall over. And this is like an example. Yeah. The, you don't only just have to think about supply by the sounds of things.

Eve Schooler: And it raises an interesting question because how much of the network involves the user? How actively attentive are users when they route across the network. Right now there's very little engagement. So one of the issues that we probably need to solve is creating as you were referring to them, levers at different points in the architectural software stack, and even in the hardware.

That allows different levels of involve. For users that have different capabilities or interests in enacting those levers all week, we have been receiving warnings about the, the coming temperature. You know, it was 109 unheard of here. It broke all records the other day, but that was through constant messaging that we were asked to please be considerate during the.

In particular between four and 9:00 PM, I guess that's when people begin to come home and turn on their air conditioners, the network doesn't do that. it. Doesn't ask you to be thoughtful, but maybe our software and our software development practices need to incorporate this.

Chris Adams: Thank you for that. So for other listeners, we did cover some of this in episode nine, where we speak about carbon aware computing and the idea of annotating say jobs for Kubernetes or other schedulers to basically say, yeah, I can wait a little while I, you know, I'm important, but not urgent, but it sounds like a well there's.

This reminds me of a blog post by a guy called Ismail Philco he's in the climate action dot slack. And he's been speaking about the idea of, is there a chance to extend some existing protocols, like say we have open API for describing how APIs work on the Web. And there is an ay API, which is another way to say that, which is, as far as I'm aware, is used for lots of kind of programming tools these days, as a way of saying.

To do the same thing, very synchronously. And there he's, there's some work there to basically extend this with this notion of delay tolerance or location tolerance, so that you can basically say this thing is important, but it's not so urgent in the same way that with an operating system, with apples, for example, you can annotate particular tasks to either be returned very quickly for high latency when there's users or something, which might be better suited to a low, a low power core in a computer.

So maybe this is actually thing I wanna ask is right now we spoke about some kind of cool future things. If I could bring this to some of the stuff that's happening right now, these days, if people are listening to this and they want to do something or start playing around with some of these ideas, where should people be looking?

What kind of software is out there? What kind of tools, uh, exist for people to kind of experiment with some of these ideas to play it with their own time, or even possibly build some cool new services on top of, for example,

Eve Schooler: A couple of thoughts, at least one is that at Intel, there is a power, a dynamic power management. Solution that exists called speed select technology. And it does allow you to dynamically adjust the frequency of cores. And there's some interesting description of that technology at the most recent I C N.

Conference in 2022 in a joint paper with British telecom, there was a paper on NFV and energy efficiency describing that service. But for developers, I would say some of the most interesting APIs I've come across are from there. There are quite a few offerings to get carbon intensity information from the electrical grid, but use it in computing systems.

And some of the interesting, uh, APIs are from Watttime. And electricity map. And so I would say you could play with those to see, you know, whether you want to incorporate your both carbon intensity and understand what the patterns are of the carbon intensity, where you reside or where you want your workloads to reside.

I also wanted to put in a plug for a workshop that's coming up. That's being hosted by the internet engineering task force on environmental impact of internet, applications, and systems and the deadline for putting in publications, romance.

Romain Jacob: I'm aware.

Eve Schooler: It's the end of October and with the conference, with the workshop happening in December.

And then finally, one of the things that's been, I wouldn't say bothering me so much as frustrating me, is the long lag time between. Our assessment of the overall internet footprint and the time that goes to publication, there's a small group of people who diligently publish these assessments, but it's really backbreaking work to understand where are the pain points in the infrastructure and topology.

So I would provide a call to action if you will, for networking researchers involved in the internet to help speed. Accurate and timely assessment of the networking and, and overall ICT energy usage by participating in and contributing to these it documents. It's called L dot four seven, but it it's its name.

Human readable name is the G HG for greenhouse gas. Trajectories for the I C T sector. So if you have some insights into the pain points of the Internet's energy usage, where we could be more efficient, turn things off, age things longer, be aware of carbon intensity. We'd like to hear from you.

Chris Adams: When you spoke about that, that does remind me that. So this is the Green Software Foundation podcast, and it's worth, I would be remiss to not mention that there is a green software carbon aware SDK, specifically that. Apps, some of these APIs that you're able to use. I think it's primarily written in.net, but I believe there might be some a go build of somewhere this as well, independently of this, the organization I work for the Green Web Foundation, we've built a CO2 JS, which basically has, uh, a lot of the kind of carbon intensity figures inside it now.

And also grid intends to go, which is a Golan library specifically designed to allow you to essentially, again, wrap these APIs and use them in scheduling tools. The other work that might be worth being aware of is that there's some work with ripe, which is the people who issue IP addresses in Europe.

They've been funding us. Our NGO to basically annotate every single public IP address on earth with carbon intensity information. So if there's a chance to build some of the green routing stuff, yeah. You, we have an API which is API to carbon intensity from our organization that will give you some figures for free, but these are annual.

These are not gonna be fluctuating or updating the way that, what time and some of the other providers do. There's also some work from singularity. Who've also, who've recently. Started sharing some information and hourly resolution all across America for people to be looking at this.

Romain Jacob: One point I wanted to make is that I do agree that this, this question of carbon intensity and carbon awareness is important. For sure. We need to be able to improve on that metric, but we should not forget that at the end of the day, the best energy is the one we do not consume. And so we should also keep investing efforts into being more energy efficient.

While keeping in mind that consuming less energy, if it's energy has to be more carbon heavy is not necessarily the best tradeoff still. We should. We should not. We, we should look at the low hanging fruits in, in reducing the energy we consume for the current service the networks are providing. And I mentioned earlier already today, this, this studio of seasonality, the analyzing the level of redundancy that are existing in network.

I think in many internet service provider networks. So kind of the edge of the internet, where we have strong, seasonal patterns into traffic, they are doing fruits and as F paper of hot carbon was mentioning, there are many of those small networks. So the benefit you, you, you can get there actually add up pretty quickly.

And if they, if they don't seem. Interesting. If you look at a single network, if you apply those principles everywhere, you can achieve very large effect and that's something, every network operator should, should have a look at if only to reduce their energy bill.

Eve Schooler: I think you're absolutely right. Roman, I think there are three things to consider. Actually, when we think about green networks, one is first and foremost, this energy efficiency use less. If you're gonna use energy, ensure that it's decarbonized, but then there's this third facet, which we haven't even touched on, which.

The other environmental impacts, whether that's water or toxicity or air pollution, whatever it is that also need to be somehow captured in metrics as well. And ultimately comprehended.

Chris Adams: We're running short on time. So I'm gonna ask one question just because it's very rare. People who understand the network to help answer this question. So for people who might know people who feel bad about say watching Netflix or feeling guilty about being on video calls would either you have something to say to people who might be struggling with this to maybe put their mind at ease or help kind of come up with a mental model.

Like, should they be feeling bad about the environmental impact or the video calls they make, or the videos they're watching after.

Romain Jacob: Uh, I don't think, uh, trying to make feel people guilty will change anything. People don't have the levers to change anything like most, most individuals like you and me. I mean, you're you open your laptop, you have a service provider and you don't have any control. I mean, you can choose provider a or provider B, and they may sell you some broad characteristic of the internet connection they provide you with.

But I mean, you. Monitor this, if you're an internet geek and you care about these sorts of things, but you cannot truly influence where it goes. It's not exactly true, but in practice, the individual has very limited control. The network providers do that's that at this level, then that things need to be, need to be changed.

Now that being said, you can still do so. Right. You can, for example, just be considerate before consistently streaming and uploading to clouds, gigabytes, and terabytes of, of pictures and multiple the providers, because you care that if Google goes down, then Facebook is not and you get to access your things, you know?

Yes. But how many people are actually doing this? I don't think so. I don't think so many.

Chris Adams: Okay. So there we have from someone who's on the far stages of their PhD and Eve, is there anything you might add for people who are wrestling with this particular issue themselves? When they open up zoom to speak to a loved one or anything like that themselves?

Eve Schooler: I think it's like anything else in our lives, we need to be ACC acculturated to thinking about this as an issue. I don't think we should have that much guilt about it, but we should be thoughtful. And so if it doesn't make a difference in the, in the communication to have the video, when you're just a participant versus a speaker, or if you can do low Def versus high Def, those are really easy decisions.

And I think there will come a time when. People will be asked to fit within budgets of carbon footprints and companies and so forth. And so we'll have to do our part. So we should be getting in the habit of at least thinking about these things. But I, as others have said, we don't have that many choices except on or off, it's sort of a bullying choice.

Maybe, you know, one, one resolution or not. And, and something about teleconferences is we save a tremendous. Using teleconferencing technology over air travel and, and other forms of travel. So it's, you know, incrementally, we're getting more and more efficient as to all those Netflix that we're watching.

that's, that's another concern again, maybe we're, we'll be given a budget in time.

Chris Adams: All right. Okay. Thank you for that. From what I'm, from what I'm hearing, it might not be the case case that streaming is indeed in you flying. So that's one thing that we could take into account. All right. We're just gonna wrap up now for people who have enjoyed this and want to learn more, where can people find you online or where should they be going to learn more about the works that we've discussed here?

Romain Jacob: Yeah, so you can find, find me on my website, homo.net. This is where you will find most work related updates. Otherwise, with my name will be easy to find on Twitter. I will not read my handle because it's unreadable , but uh, my name works fine. I tried it before.

Chris Adams: Okay. Excellent. All right. We'll be adding your, the links in there. And Eve, if people have been interested in some of the stuff you've been talking about, where should they be looking?

Eve Schooler: Similarly, you can find [email protected] and linked LinkedIn as.

Chris Adams: Okay. I'm really glad folks. I've really enjoyed this session. And I think there we've covered a lot of really quite helpful ground for other people who are wrestling with this. And I'm just like curious about this for the listeners. Thank you very much for listening to Environment Variables. All the resources for the podcast will be available at podcast dot Green Software Foundation.

Along with copious show notes, with all our links for this. If you did enjoy this, please do write a review on wherever you've hear in your podcast. It really does help us find new audiences and yeah, that's us. Cheers folks.

Romain Jacob: Bye. Bye.

Eve Schooler: Bye

Chris Adams: Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on apple podcasts, Spotify, Google podcasts, or wherever you get your podcasts. And please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we'd love to have more listeners.

To find out more about the Green Software Foundation, please visit greensoftware.foundation that's greensoftware.foundation in any browser. Thanks again, and see you in the next episode.

From Carbon Aware to Carbon Intelligent

8 augusti 2022 | 49 min

In this episode of Environment Variables Chris Adams is joined by Colleen Josephson of VMWare, Philipp Wiesner of TU Berlin and Sara Bergman of Microsoft as they discuss the opportunities with making first carbon aware and then carbon intelligent computing. Variability, curtailment, disaggregation, 5G, 6G (!), delay-tolerant networks, intermittent computing, IoT and even a short segue about Raspberry Pi’s all make an appearance in this action-packed episode!

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Sara Bergman: LinkedIn / Twitter
Colleen Josephson: Twitter / Website
Philipp Wiesner: Twitter

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Talks & Events:

Talk: Breaking the Barriers of Stranded Energy through Multi-cloud and Federated Data Centers by Colleen Josephson, Zhelong Pan, Ben Pfaff and Victor Firoiu.
Talk: Towards a Fossil Free Internet by Chris Adams.
Talk: The Internet of tomorrow must sleep more and grow old by Romain Jacob and Laurent Vanbever.
Event: Hot Carbon at UC San Diego 2022.
Event: Solar Protocol Hackathon - August 14 2022.

Papers:

Paper: The Sky is not the limit: untapped opportunities for Green Computing for Hot Carbon by Colleen Josephson, Zhelong Pan, Ben Pfaff and Victor Firoiu.
Paper: Architecture of a Junkyard Datacenter by Jennifer Switzer, Ryan Kastner and Pat Pannuto.
Paper: Cucumber: Renewable-Aware Admission Control for Delay-Tolerant Cloud and Edge Workloads by Philipp Wiesner, Dominik Scheinert, Thorsten Wittkopp, Lauritz Thamsen and Odej Kao.
Paper: Let's Wait Awhile: How Temporal Workload Shifting Can Reduce Carbon Emissions in the Cloud by Philipp Wiesner, Ilja Behnke, Dominik Scheinert, Kordian Gontarska and Lauritz Thamsen.
Paper: The Internet of tomorrow must sleep more and grow old for Hot Carbon by Romain Jacob and Laurent Vanbever.
Paper: Internet Backbones in Space by Giacomo Giullari, Tobias Klenze, Markus Legner, David Basin, Adrian Perrig and Ankit Singla.

Open Source Projects:

The Green Software Foundation Carbon Aware SDK - the SDK offered by the Green Software Foundation.
Grid Intensity Go - a Golang CLI and library, for use in projects to extend them for carbon awareness.
The carbon aware branch of Nomad - Nomad, the popular scheduler by Hashicorp has an experimental carbon aware branch.
Carbon aware autoscaler in KEDA for Kubernetes - The issue mentioned in the podcast for extending Kubernetes.

Blog Posts & Articles:

Blog: Carbon aware autoscaler in KEDA for Kubernetes - The issue mentioned in the podcast for extending Kubernetes by Bill Johnson (Azure SRE).
Blog: Annotating jobs with AsyncAPI to be more amenable to carbon intelligent architecture patterns by Ismael Velasco.
Article: The Dirty Carbon Secret Behind Solid State Memory Drives | Discover Magazine
Article: Intermittent Computing in Satellites: Satellite Designed at CMU Launches into Low-Earth Orbit by SpaceRef magazine.
Article: A carbon aware internet by Chris Adams for Branch Magazine.
Blog: Nasa’s work on Delay Tolerant Networks
Blog: SCION and carbon aware networking the World Economic Forum.
Blog: Sustainable 6G Efforts by Colleen Josephson.
Blog: VMware sustainable digital infrastructure collaborations by VMWare.

Transcript Below:
Philipp Wiesner: So I think we're only touching on the actual potential of how much flexibility there is in many workloads. And I think this is also one of the biggest challenges in the entire field, how to identify opportunities for flexibility, and then especially how to make schedulers aware of these opportunities.

Chris Adams: Hello, and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode, we discussed the latest news and events surrounding green. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host, Chris Adams.

Okay. Welcome to Environment Variables. On this episode, I'm joined by Colleen Josephson of VMware, Sara Bergman of Microsoft and Philipp, Wiesner of TU Berlin. Today we'll be discussing the opportunities and the challenges associated with making software first carbon aware and then carbon intelligence.

So before we dive in, though, let's have a quick round of introductions.

Colleen Josephson: My name is Colleen Josephson, and I am a research scientist at VMware. And a lot of what I've been doing at VMware has been focusing on sustainability for the past couple of years. The core of it has been in the telecommunication space, actually, which a lot of people don't know that VMware has a business unit dedicated to that, but I switched teams in the past couple of years.

So teams have been moving around and now we're underneath the office of the CTO. So I've been doing much more broad and general sustainability work throughout the company. And as part of that, I'm actually the org representative for. VMware in the Green Software Foundation, we are members.

Philipp Wiesner: Yeah. Hey, I'm Philipp. Thanks for having me here. I'm currently doing my PhD in computer science at the technical university of Berlin. And my research is on ware and renewable Web computing in the cloud, but especially also in novel computing environment, such as fork and edge computing.

Sara Bergman: Hello. My name is Sara Bama. I am a software engineer at Microsoft where I work with Microsoft 365. I am also the chair of writers project in the Green Software Foundation.

Chris Adams: Okay. So every episode we talk about green software and today we're talking about carbon aware and carbon intelligence software, and, well, we all know that electricity has to come from somewhere, but not everyone, not. We don't always think about how it's generated. So I'm just gonna open this up to, uh, for someone who's done a bit of work in this for the uninitiated, what is carbon awareness in the context of computing?

Philipp Wiesner: So the idea behind carbon Web computing is not to save energy, but to use the, the right energy. So that's green or renewable energy. And this has to do with that. The carbon intensity, which is the, how basically how dirty the, the energy in the public grid is this varies over time. And of course it.

Different and different. And in ware computing, we're basically trying to exploit that. So we are trying to shift computational load towards times and towards places where we have green energy or at least where we expect green energy.

Chris Adams: And if I understand it, you've done some work in this field specifically. And there was a, there's a paper. I think that I saw from you. Let's wait a while. Maybe you could just briefly touch on this and then we can open the floor up to some of the other people here.

Philipp Wiesner: yeah. Sure. So in let's wait a while, this is basically an analysis of the potential of temporal workload shifting. So I just touched on it usually. Does come where computing there's two dimensions to it. You can either shift workloads on geo distributed data, data centers, like on a location scale, and you can also defer workloads on a time scale.

And this paper addresses the time scale. So we're basically looking at single data center and within the data center, we have certain workloads that are maybe not urgent. So it doesn't really matter if we compute 'em right now or in four hours or maybe in 10. And on top of that, there might also be scheduled jobs that are always scheduled.

For example, nightly, like nightly builds database backups, and so on. There's plenty of jobs that are scheduled nightly basically to not disturb anyone. But it doesn't really matter if we compute these jobs at one in the morning, two in the morning, five in the morning, as long as they're outside of business hours, but per scheduler for carbon Web scheduler, this makes a big difference because if it has 12 hours of flexibility, Using forecasts for renewable energy or for carbon intensity, it can decide on when to run this workload.

And that can make a very, very big difference. And in this paper, we basically analyze the carbon intensity for different regions and try to identify time windows where shifting is very promising.

Chris Adams: Okay, cool. And Colleen, the reason I, I came across your work initially, when I saw the, I think it was ACM energy. You did a talk specifically about this in the, in the context of your work with VMware as well. Is. Case.

Colleen Josephson: Yes. Yeah, it is Victor and I, Victor is over in the VMware research arm. We. That talk together. It's been a kind of a growing area of interest for VMware. When I arrived, it was just after we had partnered with the NSF, the national science foundation in the United States. And we actually specifically set aside some funds for a call on sustainable digital infrastructure.

And we funded, I believe. Three different projects from that area. And we've been working with those academics on the different aspects of overlapping a lot, actually of what Philipp was just talking about. How do you take data centers and make them more carbon aware through, you know, one aspect we discussed a lot is shifting through time and space and we we've done some, you know, early investigations and.

Crunched some numbers. And we've also been working with in particular, Andrew Chen at the university of Chicago. He has been working with Lena rad at the university of Wisconsin, and she has great expertise in kind of the power distribution. Side of things, which has been a really interesting consideration as we've talked with them.

Cuz when you start to think about, you know, beyond a single data center, you know, multi-cloud large customers, people who have workloads and data centers across countries or regions, when you're shifting things and in space, you can actually have a significant impact on the grid. The, these big data center providers, you know, customers that use VMware software M.

Amazon so on. If they are shifting their workloads around, you have the carbon intensity benefit. But one thing that, you know, our academic collaborators made aware to us is this very interesting fact that we will in turn be modifying kind of the, the economic dynamics of the grid have. You know, potentially the actual operating capacity of, you know, how power is distributed.

So that's been a really interesting angle to think about from kind of a major provider point of view. And how do we take these really interesting and promising ideas and begin to scale them up from, you know, a grid provider scale point and another thing that customers have brought up to us. Data borders, you know, if moving things from within, you know, one corner of the United States to the other corner of the United States is typically not a problem, at least legally.

And you might have latency trade offs, which need to be part of the equation. But when we went and kind of pitched our ideas, To people over in the EU, a lot of people were, you know, pretty concerned about the fact that the, the borders are a lot closer. So , you have customer data and it can get a little bit trickier moving it from one region to another.

Sara Bergman: As someone working for a major cloud provider and based in Europe, this is of course something that, that touches in my interest or peaks my interest because I think this is really interesting, at least for Microsoft, we have this Microsoft runs on. We have the cute running t-shirts and everything. Like it's a core part of our business.

And I know that is the case for a lot of customers, only Microsoft. So this is an important part of the equation, um, to make this. not only theoretically possible, but actually possible at scales for where we can solve this for the large business customers. But I think I watched one of your talks as well.

Colleen, I think it was, we can, you also talked about the, the difference throughout the day, because even if one country, which is in Europe quite small, doesn't have a lot of different energy providers. Not only can the. Type of energy that's generated change throughout the day, but also what is competing for that energy will change throughout the day.

Like when we're all standing up in the morning and everyone's taking a shower before work and, you know, it's, I live in Norway where everyone charges their electric car. Of course there's gonna be higher load on the grid versus in the night, there is a lot less, for example. So I think that's an interesting aspect as.

Philipp Wiesner: Yeah, I think directly to add on that this variability can be quite dramatic. So in France, for example, they have clean, not clean maybe, but like, oh, carbon energy, because of all the nuclear power they are deploying throughout the day. So there you have barely any potential. But then there's regions like Germany, for example, which are very interesting because they're super variable, like Germany employs comparably, much wind power as well as solar power.

So at many times of the day, they manage to have large fractions of the grid provided by green energy. But if neither sand nor wind available, we burn brown cold, which is pretty much the dirtiest of all fossil fuels. And this is why variability is really crazy. Like within a normal day, you can expect twice or like 50% fluctuations.

Like could be that one kilowatt hour that you consume now is twice as dirty. If you consume the same kilowatt hour, few hours later. And within a few days, you can even see like the difference between the min and the max can be factor four or something. So one kilowatt hour can really vary from 100 grams co.

Up to 400 grams or more 500 grams.

Colleen Josephson: Yeah, this variability, you know, particularly with wind and solar and renewables like that, that touches on something that we covered in our talk, which is the, the idea of curtailment, which is a really interesting opportunity. I, before I started in this area as an energy consumer, I had no idea about this concept, but.

Basically it's this, it gets into this supply and demand relationship that I was talking about where energy providers, they. Want to match the energy available to the energy being consumed. And if the demand is low, uh, as Sara was talking about there's times where people that you wake up, you take a shower, you know, or you get home and you, you start cooking or watching TV.

There's an ebb and flow of power, power. Is consumed. And that admin flow does not necessarily match with how nature is behaving. So if people aren't using a lot of power, the grid can't accept it. We, we not yet. Anyway. So what we have to do is actually burn it off, which I thought that was pretty shocking.

When I learned about it is we have all these renewables and we're just letting it go to waste at the moment. So there's this. Untapped potential as we start thinking about these workloads that can be deferred or moved in time or space. Well, when these curtailment conditions occur, can we pull from our back pocket, some of these compute workloads and start basically increased demand.

So we don't let this clean energy go to waste.

Chris Adams: Okay. If I'm bring spring it down for a second, just to make sure I understand the idea that you have here is that you, without making any changes to code itself, you're not changing the code. Really. You're just deciding to run the same code either at different places. In a kind of sci-fi move it through space or through time.

And this is going to have this, this is basically a way that you can essentially reduce the environmental impact of something without necessarily having to redesign it or from, from scratch. That's what it sounds like you're saying in this case. Yeah.

Colleen Josephson: Yeah, that that's, that's one way to put it. And of course, you know, the very process of taking a workload and moving it in itself has kind of, maybe you can think of it as meta code and it can be enormously complex to get all of that. Correct. But let's say maybe, you know, the core of what you wanna do is some training for a artificial intelligence model.

That same exact training is happening. You're just picking it up and plopping it in a new place or a new time.

Chris Adams: And when you spoke about curtailment, just then there was this idea that essentially because you can't get the energy, that's generated to someone who's able to consume it, it is essentially wasted and you're not able to make some kind of. Effective or economically useful use of it. That's the, that, that, that I'm sharing this largely for folks who might not have heard some of these terms, because if you're a software engineer, you might not have heard of curtailment or any of this stuff here, actually, we've spoken bit about carbon aware.

So we have this notion of you can be carbon aware in that, knowing that there are natural cycles on the grid that you can respond to. And Colleen, I think you mentioned this idea of, is it carbon inte. Was in one of the papers that you spoke about recently? I think I, I, I'm not sure. I think it was in your most recent paper.

It was actually kind of exciting for me, cause I haven't, I hadn't heard this term before and it feels like one of the next steps from this notion of carbon oil, perhaps it was among, I heard at hot carbon, which is basically a conference specifically for folks doing this kind of work.

Colleen Josephson: I think that actually was not our paper looking through it. Maybe. Carbon my colleague that one of the other authors spoke about it and that could be something he said, but we definitely talk about carbon aware computing in our paper. But I think that that in itself is kind of an interesting point to bring up is these terms are.

New. And they're very new because the conversations and I mean, there are some people who have been working in this area for a long time, but in the past couple of years, I think that. Interest in work in this space is really accelerated in pick up and I'm seeing more venues than ever in which to present this work.

So I think carbon aware versus carbon intelligence is, you know, kind of an, an artifact of the fact that we're still pretty early on in the. The conversations that we're having and standardizing on our terminology , which is, you know, Green Software Foundation itself has a standards group. So I guess this is a case in point why standardization is so important.

Philipp Wiesner: I think Google is using the term car intelligence across their papers, but I think it's just terminology. I think they all mean the same.

Chris Adams: Okay. Alright, we'll go with carbon aware for the time being in that case until we hear otherwise. So essentially when someone says carbon intelligent and carbon aware, it's more or less the same idea of essentially. Making your work or the workloads are making them kind of sympathetic to the natural cycles that I guess surround us really.

Colleen Josephson: Yeah. And I think to add to kind of this hierarchy, what we think about in VMware is kind of a few different levels. One of the most important things that a lot of companies are still struggling with right now, Is just how much carbon are we consuming through our operations and all the different scopes.

It's really hard to answer that question right now. How much does this application does this container? What, what, however I choose to slice it. What, what is the consumption of, you know, carbon and other re resources and then. From there, you know, that that's the classic. You have to know, you know, be able to measure the problem in order to solve it.

And then the next step is once you have this really good visibility into what you're consuming, you can then start to optimize for it. And that's where we get into kind of this awareness, uh, intelligence managing your carbon.

Sara Bergman: Yes. And, and I think that. Visibility is so important. And I think it's important to get the right visibility to the right people. I think scheduling of, of some sort isn't really new. I mean, we've had the supercomputers scheduled the work for them for quite a long time, but this type of scheduling is of, of course new.

But if you, if you get these concepts to people who have domain knowledge, then that's a concept they know, and they can say, oh, you know, This job can wait a while because they have the domain knowledge to make use of this technology. And, and that's where it becomes really powerful when we have the, the visibility to allow everyone to participate in it.

Because yeah, also, like we said, I think more people want to contribute and want to do something. And if you make it really easy, well then all the better.

Chris Adams: Okay. So I have to ask, are there any, so let's say you can tune things to basically. Take advantage of electricity when it's cheap and green and, and abundant. Are there any kind of measurable savings that are actually published or out there? Because this sounds great, but surely it does. Are there any early results to at least give us some idea of what kind of savings could be gleaned from this before you think you start thinking about redesigning?

Like, let's say you right to do you, you've got an app and you are trying to reduce the impact. Is there any kind of measurable reduction in carbon on this kind of work? For example, what, what kind of figure. Have been sh have been coming up so far from.

Colleen Josephson: I think what Philipp was saying earlier, really lines up what we've been finding these swings of 50%, depending on your region. You know, that can be that the high end, we saw a carbon savings of 50%. If you're in a highly volatile region and then even in places where it's less volatile, there's still.

This cycle and, you know, that tends to be more like 10%. And I think it's really hard to say right now, because a lot of the experiments as we talked about are not yet taking into consideration, which workloads are good candidates to move, you know, which workloads, you know, these data boards, you know, part of how we evaluated is we were just like, okay, let's take it from Germany to somewhere else and not think about these, you know, Policy issues that might make it so that we can only go from Germany to Poland, for example.

Philipp Wiesner: Yeah, I, I fully agree on that. I think the thing is that currently many results are still simulation based also our results. So you can easily craft scenarios where you can under certain conditions get 50% carbon savings by moving workloads from the night to the day or something. The only research that I'm aware of where like this was actually deployed somewhere, it was a Google paper that was publish.

last year or this year where they really have something running on their infrastructure. And then they report, like, I think it's 1% or something that they actually cut out. But this is then not only about jobs that were actually shifted. It's about like their entire workload of the entire data center.

And that's then already quite impressive because I think we're only touching on the actual potential of how much flexibility there is in many workloads. And I think this is also one of the biggest challenges in the entire field and how to identify how. Yeah, identified as flexibility opportunities for flexibility, and then especially how to make schedulers aware of these opportunities.

So most workloads are still black boxes in many, in many regards, you maybe have a deadline, but that's it. But there's a lot more information about workloads that would be good to have for cloud providers if they want to schedule them. For example, whether workloads are interruptable and what's the cost, the overhead of interrupting and resuming a work.

So for example, many machine learning trainings can take days. It's absolutely not uncommon. So if you know that interrupting these jobs is cheap and it often is because they do already do checkpointing. They write immediate results to the disc all the time. So if you can interrupt and resume these workloads, then you can really exploit it.

Short-term fluctuations in, in the grid. And.

Colleen Josephson: Yeah. I just wanna say that Philipp. I absolutely agree everything. You're saying really lines up with what we find. Really critical part of going from simulation to reality is this concept of how do you identify these candidate workloads? How do you do this? You know, figuring out what is good to move and, you know, maybe start off doing that by hand.

But in the long term, we need this to be automatic, you know? No. Human in the loop. So our systems need to be adapted so that we create a workload and it has some sort of metric for, you know, whether or not it's moveable at that time, how flexible it is. And so on.

Chris Adams: Okay. Now, if I, if we could maybe unpack some of this for the case of like these, these examples that you folks have just described, Colleen, you are, you're working at VMware. So. Whereabouts is this happening in the stack? Is it a product from VMware that is doing this stuff or is it somewhere higher up?

Would like, say Kubernetes, for example, is this happening at a hypervisor level or is there somewhere else in this? Cuz I know there are various parts of the stack that you could make an intervention. And I know there are examples of things like a carbon away, Kubernetes scheduler and where we work.

We've been doing some, we've got some P press open to Keta, which is an autoscaler specifically for Kubernetes to use some of this information so far. But I would love to hear cuz yeah, I didn't really know that much about VMware and it seems like there's a whole fascinating paper on this actually.

Colleen Josephson: Yeah. So I. We can't. I can't say this is like home to a specific product. I think one of the things that's things that's actually really nice about VMware is we have something called our 2030 agenda where we have 30 goals that we want to achieve before 20, 30 and sustainability. It's an ESG driven agenda and we have a whole bunch of goals related to sustainability.

So we've really taken the past couple of years since we announced the agenda to make sustainability kind of. In the core of what we do. So rather than having a sustainability office at the top, we wanna embed it and empower every single individual. Engineer. So we have these types of projects for moving workloads, measuring things they're kind of going on in a few different places.

So, and all I can really say for sure. You know, every company is a little hesitant to promise. Features for a specific project. we are very actively working on, we have engineers and project managers and researchers it's to be determined, kind of where it will emerge. But, you know, we've been talking with partners and stakeholders, and I think that it's been a very active space in a number of places in the company.

Chris Adams: Cool. Cool. Thank you for that. Okay, Sara did, there was something you would've come in on.

Sara Bergman: No, I would just say noting vigorously on, on a few things. So yeah, no, no. Additional to ask, add. Yeah.

Chris Adams: all right. So one of the things that came out of that was this idea of being able to, when you have a piece of work to be computed, it's either providing some kind of annotation or some, um, way of expressing that. Yes. It's okay. To pause me for example, or, yes, my I'm not I'm I'm not so urgent, but I'm, but as long as I'm done by this time, for example, So maybe have a few folks come across any kind of patterns that have actually that we might be seeing us wear, cuz based on what you folks tell me, it makes me think of the, the fact that I know when some of say Apple's work because they're now switching to a different kind of architecture that there is this notion of like annotating particular work that needs to be interactive to a user.

Something that might be a background thread. Do we have anything in the region of like a convention fr annotating stuff so that it's easy to. Especially. Yeah, go Colleen.

Colleen Josephson: I can't name any specific patents, but what this is reminding me of is telecommunications and I telecommunications as a deep. Degree of work in prioritizing data streams. And that's been a really active area for a couple of decades, you know, whether it's cellular or more traditional, just internet, there is this idea of delay in tolerant traffic versus more delayed tolerant traffic.

And there's a really rich body of research that we can look at and maybe borrow.

Philipp Wiesner: Yeah. Just one thing to add on that. I think what's special in particular about this scheduling problem in comparison to like scheduling problems that we had before is the, the time scale as well. Because like, if you talk about delay tolerance and telecommunications, And if we have like scheduling on a CPU level, that's really milliseconds or less or whatsoever, while in, if we talk about carbon awareness, like actually optimizing for the carbon grid, it rarely goes below 15 minutes that we have as a forecast and frequency, basically.

And if you optimize for, let's say your own solar panels, then you can maybe use satellite. in like the five, 10 minute scale, you can use like weather data for like a few hours or like, but if you want to go below five minutes, then you already need sky cameras. You already need like video information of where clouds move and stuff.

So when we talk about scheduling and delay tolerance here, then we do not really mean that we start a job and 30 seconds later we resume it. It's like bigger jobs that run for at least 15 minutes. And then we stop them for a few hours and then we resume them. So this makes it a bit different to, to what we have.

So.

Sara Bergman: Yes. I, I think this is super interesting because it's, if you start optimizing for perfect scheduling, it's easy to fall into the trap or you're investing more or spending more into actually. The savings. It, it, you need to think of it as economics. So yeah, you can maybe schedule it down to every five minutes, which may be great for your application.

But if you then, like you said, Philipp need to buy a sky camera, need to invest in like your own satellite network. I'm exaggerating now, but you understand the, the scenario. Well then the actual car saving is at the end of it might be like net positive, uh, in which is a negative in this disregard. So any, any sort of carbon project that would.

Whatever, wherever it is, need to take in the totality, because we are just one, one planet. And I see the same similar discourse in machine learning a lot where people are very eager to use machine learning for solving climate problem, which I think is great. And I'm not saying we should stop doing that, but sometimes we're spending massive resources training those models to then save the world where we actually polluted the world more while doing.

Colleen Josephson: Yeah, that we actually made that exact point in a recent white paper. One of the other organizations that VMware is involved with is the next G Alliance, which focuses on telecommunications in north America is looking at, we're talking a lot about 5g. We've already got our site set on 60 and Microsoft is actually also a member with.

And I co-lead something called the green G working group, looking at how, how we can make our next generation telecommunications networks intrinsically sustainable. And there, there is a lot of excitement, like you said, Sara, about. You applying machine learning, but you have to remember this caveat that right now, training these models is really carbon intensive.

So you have to remember the resources you consume to, to get the job done. and kind of the same thing comes in with upgrades. So if you look at upgrading hardware and data centers or telecommunications hardware, so 5g, I think got some bad press for how much power the base stations consumed, but what's actually true about it is that the power consumed per bit transmitted has gone.

Significantly. So there's a good advantage to upgrading your hardware, but then, you know, what about this hardware? You're getting rid of everything that we produce has this concept of embodied emissions. It takes resources and carbon to produce this hardware. So you have. Really carefully look at that sort of trade off.

It turns out that keeping our devices, especially smaller devices in use for as long as possible is one of the greenest things that we can do.

Sara Bergman: Exactly because it's very tempting to look at only your share and try to slim it down as much as possible. But if that means you're just overflowing into other carbon budgets, you, well, the net effect. What you want anyway. And that might be a leap from how a lot of us are used to thinking about software, but I also think it, it like triggers that natural engineering curiosity in us all.

So it's not necessarily a bad thing.

Chris Adams: So we're talking about the embodied emissions for this, and I'm, I'm glad you actually spoke about the network part, cuz this is one thing that. , I don't have that much access to experts when looking at this. But, uh, as I understand it, for example, with 5g, there was an, there's a significant amount of embodied energy in making each of these towers, but would it be the case that you would have more towers which are more efficient, but have higher embodied emissions compared to what you had before?

Like maybe it'd be really cur I'm curious about what the kind of trade offs you actually might have to make there because as I understood it, 5g. Tends to have a lower range. Is that the case compared to say 4g, for example, or is it able to fill in some of those.

Colleen Josephson: There's different types of cellular infrastructure for different types of transmissions. So I think what you're touching on is this idea of these micro cells, which are, you know, you have millimeter wave, they have short range they're deployed in dense, urban environ. But that doesn't mean that cellular providers have stopped having these longer range communications it's, you know, kinda like how your cell phone has different types of communication for different scopes.

You have Bluetooth for short range, you have NFC for ultra short range, and then you have your cellular and wifi for longer range. The same is true of cellular networks for really dense urban environments. You're gonna have these smaller. Micro cells trying to give high speed coverage in these urban areas, but you still are gonna have these larger cells deployed across the United States so that you still have, you know, a good range in coverage everywhere.

But when we talk about savings in telecommunications, I think one of the really big opportunities that's getting, you know, off the ground rate now has to do. Software defined networking and virtualization. So historically in telecommunications, everything you needed was kind of in the tower and there's this massive movement for disaggregation going on.

So you can begin to pick and choose providers and move different parts of the cellular network around. So, you know, one of the things that we've talked about is, you know, there's the telecommunications industry and then there's I C T and. The line between the two is beginning to blur because a lot of what used to maybe happen only on a tower or in specialized hardware can now be done in general data centers.

So the two industries are really kind of solving the same sorts of problems , which is really interest.

Chris Adams: So, if I understand that correctly, you are saying that some of the. Hardware that was all bundled or some of the functionality that might have been bundled into a single piece of hardware is somewhat being kind of unbundled. So that maybe a cloud computing thing might be happening. And this is things like, would this be stuff like open ran or like the open radio network kind of stuff?

Colleen Josephson: Yeah, that's definitely getting into that area, openness so that you can have different modules communicating with each other. And there's a lot of really interesting opportunities there. So one fact for example, is incumbent like really kind of older and existing cellular networks. You don't have the ability to easily turn off a base station when it's not in use.

So these base stations are what are consuming the overwhelming amount. Of power in the network and kind of this really low hanging fruit is, well, when there's nobody around turn down the volume and the ability to have it more software defined means that we can try out these algorithms to dynamically do that.

Right now it's a lot of, if people have to turn things down by hand, do a lot of really onerous changes to implement this, but as we start to move to a more nimble infrastructure for 5g, And beyond, we really have this interesting opportunity to rapidly prototype these power savings algorithms and see what we can do.

Philipp Wiesner: So I see that in networking, we have plenty of opportunities for energy saving. But do you see any opportunities for carbon awareness? Because from my experience, there's not so much, you either have like edge infrastructure that is wireless and really energy intensive, but it's by design critical it's by design has to be fast.

Otherwise it wouldn't be at the edge like, well, on the other hand, you have the big data centers that have like a lot of patch jobs that are like very flexible and defer. But they are connected via fiber, which is super energy efficient and there's barely any, any consumption on that. So do you see any opportunities in that regard or is it mainly computing that we can make car aware?

Colleen Josephson: Huge opportunities. And that's one of the reasons I'm glad that we're thinking about six G right now is we have. This major to reuse the word opportunity to design this next generation network to be carbon aware from the beginning. One thing we noticed when we started this work is right now, we're historically figuring out how much power 3g, 4g 5g network consumes has been very much a cyclical thing.

You build the network and then you look back in time and then you make measurements. So you. When we were working on our first white paper, we actually couldn't even answer the question. How much energy does 5g consume there? There's just the work on that is ongoing right now where, you know, kind of, we realized as we were doing this, it's like, why.

And we, we have infrastructure to measure uptime, to measure latency. We already measure these things across basically every facet of computing. So why are we not measuring energy consumption and figuring out the carbon footprint? So we really need to design in this ability to measure how much power.

Part of the network is consuming and report it with really fine grain in a really fine grained way in a real time way so that we know as we're starting to prototype six G exactly how much power it's consuming. And as we deploy different pieces of hardware or different algorithms, did this change actually work?

Chris Adams: All right. So this is actually something that we're probably gonna cover on our future episode. In more detail, there is some really fascinating work going on from an organization, working on a protocol called the cion. Just like a clean slate implementation of their staff, where we work. We've been doing a bit, bit of work with them and we have an ongoing project.

So basically annotate every single public IP address on earth with carbon intensity information such that you can start creating some of these paths. But one of the problems you have with the existing internet is that. You BGP doesn't necessarily account for these different criteria that you might want to have.

So if you go back to this notion of saying, I have a job where I need to move something, you only have one dimension right now with BGP, whereas you can't really talk about saying, I care about latency more than I do about cost. For example, if I'm on a really important video call, whereas if I'm not doing a big backup or shitting a bunch of data, I might care about cost more than latency.

So there is some work had to taking place in here and. The only example I've come across so far and it is, it's totally worth a look. I look up, I think actually, cuz it's kind of interesting, but I hadn't actually thought about this in the context of six G actually to be before this phone call. I hadn't even heard of six G so I guess there's a whole, whole little bit to actually add.

So if I may, I'm just gonna touch on this notion of. delay tolerant networks. Cause we spoke about the idea of having different criteria for jobs and stuff. Now, my understanding of, of this is that there is a decent body of work already. You used for interplanetary networks. That's where some of this initially came from is that I think cuz if you, when I did, when I was doing some research before this, for this podcast, I found out that the, the NASA actually has a whole bunch of really fascinating research on delay tolerant networks and it looks.

The actual timeframes for what they look at in terms of delay, aren't that different from the time scales we've been talking about, like in terms of 15 minutes to an hour, for example, I'd be curious if any of you folks have come across any kind of overlapping research here, because the idea of using technology from space sounds kind of cool and it kind of worked for Velcro.

So I figured maybe, maybe there's some stuff that we could actually take use, make use of.

Colleen Josephson: Well, I don't know a whole lot about delay tolerant, networking in specific, but this is reminding me of another. Area called intermittent computing, which has also ended up in space. Uh, and this is basically taking inspiration from the really small and embedded side of things, which is, I think a theme in general of green software embedded devices.

Historically they have a much, much lower power budget than something that's in a data center. And as we look at these. Networks of, you know, kind of internet of things. We wanna put them in more, in more inhospitable places, outer space, farm fields, et cetera. And these places have no fixed power or communication in infrastructure.

So it becomes really challenging to figure out how do you budget? When, you know, you're, you're not sure how long you might have to run. So there's researchers, I think at the university of Massachusetts and Carnegie Mellon who have done some really interesting work in how you. Have this architecture that deals with the fact that you could suddenly, and gracelessly lose your operating power.

So how do you, you know, checkpoint things so that when for example, the sun comes back out, you can pick up where you left off and make progress and continue to do compute or fire out a packet. and some of this intermittent computing has ended up in outer space. I think there's a, a satellite that was created that takes advantage of some of these concepts.

Chris Adams: Okay, this is something that is totally new to me. I'd never heard. If someone did wanna find out about how to apply space technology to these kind of problems, where should they be looking? This sounds really, really fascinating. I did not expect us to go in down this direction, but it, it sounds lots of fun to.

Colleen Josephson: Well, the, the researcher who did these space, toasters is Brandon Lucia. I think also, I mean, looking straight at NASA and , they, they do a lot of research in this area and they have funding and I was talking to somebody who works there and they actually have a lot of work. That's not directly tied to space.

So that they do stuff on, you know, underwater networking, for example, is one thing I was surprised to hear that they work on. So I wouldn't be too surprised if there are researchers already there starting to think about some of these issues and how you can apply what they've learned from delay tolerant networks or intermittent computing to some of these challenge.

Chris Adams: I've just realized that if anyone who is interested in the context on the concept of intermittent computing, there is really fascinating art project called the solar protocol. And it's a really run wacky project, which is essentially it's a website, which is. Really a cluster of Raspberry, Pi's all around the world in different parts of the world.

And basically there is a DNS server, which basically roots requests to whichever Raspberry Pi has battery. And when the batteries are run out, the Raspberry Pi stops serving websites, but because it's always sunny somewhere, there's always the steady supply of this stuff. And the thing that's really interesting is that they have a hack day coming up on the 15th of August.

So if anyone does wanna play with this stuff, you can actually do it. It's entirely open. There is a whole set of really fascinating stuff. And at a recent conference called limits 2022 computing within limits. I believe they've actually, there's a paper for this as well. It's a really fun project. And we've written about it on a magazine that we publish called branch.

And if anyone has a cur has some curiosity in intermittent computing, this is probably the kind of most iconic and wacky idea I've seen so far. And it should, I imagine it might be a lot of fun for the people who listen to this kind of podcast.

Sara Bergman: I love it when it feels like my work is becoming sci-fi, but that's the best

Chris Adams: Well, this is like part of the idea, the idea that you can move work through time and space. When you actually talk about that, that sounds extremely science fictionally. And this is partly why we wanted to speak about this in addition to just the conversations about efficiency, because in many ways it does feel like it's sympathetic to a lot of the kind of patterns we might normally have.

You can think of. Say seasonal food is a bit like kind of seasonal electricity or something like that. Just on a much, much more compressed time scale.

Colleen Josephson: Yeah, this is starting to remind. Of some recent work that a friend and colleague did pat Canuto at the university of California, San Diego, he, and one of his students, they started, they started working on this concept of something called a junkyard data center where they're using old phones, nexus four and nexus five phones to kind of serve this Raspberry Pi role that you're, you're talking about.

And they found that they were able to kind of match and occasionally exceed, you know, modern cloud compute offerings. And this is kind of like, I thought this was neat because it's the intersection of what we were talking about earlier. If you've got this kind of low, low, high power computing, but then you also introduce at the, the junkyard part is getting to this, what we were talking about earlier with embodied emissions, these phones people don't use nexus four and.

Phones very much anymore because they're a number of years old, but they still can do really useful and powerful computing. So, you know, this is the reuse part of reduced reuse, recycle, where we still have these very good, relatively speaking sources of compute power. So how can we extend their useful life?

So that is a pretty cool piece of ongoing work also.

Chris Adams: Colleen, the stuff you mentioned there reminds me. A service from a company called Lanum, who basically use XX Hyperscale data center. And they put them into shipping containers right next to all kinds of solar farms and wind turbine places to essentially do this kind of interruptable, low carbon computing.

And in many cases, when the cost of electricity is say negative, for example, or when you are paid to kind of scale back your power, they've essentially got another way to. These kind of services. I think this is really fascinating when you're able to DEC decouple a bunch of these open source pro uh, ideas from necessarily a gigantic data center.

You, they, you don't necessarily need to have a massive out of town, big box Walmart data center to take advantage of these techniques.

Colleen Josephson: Yeah, this idea of kind of sustainable. Computing has really also gotten a lot of attention. VMware has its own project where in that's reminding me a lot of this in collaboration with vapor IO, where we kind of have a container data center and we're looking at these sorts of savings. So I think there's a few instances of this sort of work starting to happen.

Chris Adams: all right. Sounds like lots of stuff for us to add to the show notes that we have here. If there's any kind of links or podcasts or projects that you'd like to draw people's attention to. Yeah. What's caught your eye recently that you'd say in the context of the podcast, that you'd suggest people look.

Philipp Wiesner: What I personally found quite interesting is recent work from. Monica vital from the poly technical to Milano in Italy, who looks at this entire topic of sustainable computing very much from an application side. So basically looking at, we have a certain business process, for example, where different components in our microservice architecture have a certain purpose.

And for example, like Sheena papers talks about flight booking process, and then maybe certain aspects like certain components of this pipeline may not really be necessary. They might add revenue. To the operator, they might add quality of service or quality of experience, but under certain times or conditions, we could trade this quality of experience or quality of service.

to consume less energy, or maybe we have different implementations of certain aspects of a system. I mean, I think this is really, really interesting work to think that maybe we can, if we like go deeper in the applications of how applications should be designed. So actually changing the code of software, like to refer back to what is at, in the beginning, we can actually trade some qualities of software for energy reductions during certain times to be to better align with the actual availability of renewable energy.

And I think that's really interesting.

Chris Adams: Cool. Thank you, Philipp. I definitely need to get that as a, as a link for the show notes. Sara. You've got something here to add as well. Right?

Sara Bergman: Yes. I've been reading a blog post called the dirty carbon secret behind solid state memory drives, which is a very enticing title, but it's about it ties into this episode because it talks. The trade offs between the embodied carbon and the lifetime emissions from using energy. So I'll link it. I, I thought it was insight.

Chris Adams: Okay. Colleen, what's showing up on your radar these days.

Colleen Josephson: I think I already mentioned it, the junkyard data center. So I'll find a link for that, cuz I thought that was pretty neat work. And then I'll also try to find the piece on the, the intermittent satellites.

Chris Adams: Cool. All right. Okay. In that case, I'm gonna share a couple, couple of mind, and then I think we're gonna wrap up the thing that I'm really interested right now in this particular context is a carbon aware branch of nomad, which is the alternative scheduler from Hashi called. It's very, very similar to Kubernetes, but somewhat simpler.

There is now a carbon away branch that actually does include some of this for its scheduling decisions. And that's something that I'm really excited about at the moment. And there's also a bunch of jokes around low carbon etes. Instead of Kubernetes, these days, we've been doing a bit work to build a specific CLI go based CLI to plug into tools like this so that we can.

A carbon aware, any version, any, any carbon aware cluster, any, any cluster you run should be able to be doing this kind of stuff. That's the thing I'm, I'm gonna be adding to the links here. All right, folks, I've really enjoyed this. This has been super nerdy, but that's basically why people sign up and listen to this podcast.

And I really appreciate you sharing your time with us. So folks, thank you very much for this. Just before I go. If people wanna hear more about your research or your work, where would they be looking.

Philipp Wiesner: Probably just Twitter. So just first name, last name without any just Philipp, Wiesner without any thoughts or anything.

Chris Adams: Cool. And Sara

Sara Bergman: Same, you can find me on Twitter. It's my name with an E in the middle.

Chris Adams: and Colleen

Colleen Josephson: Yeah. I post a lot on my website, see ColleenJosephson dot net. And I also share some on Twitter, which is see Josephson full because see Josephson was taken

Chris Adams: I could definitely identify with that. And my name is Chris Adams. Chris Adams was taken. So I am Mr. Chris Adams, which is @mrchrisadams. All right, folks. Thank you very much for talking to us about green software and carbon aware and carbon intelligence software, and hopefully we'll see you on future episodes.

Thanks folks. Bye.

Philipp Wiesner: Thanks a lot.

To find out more about the Green Software Foundation, please visit Green Software Foundation that's Green Software Foundation in any browser. Thanks again, and see you in the next episode.

How do we Teach Green Software?

25 juli 2022 | 43 min

👉 State of Green Software Survey - click this link to access! 👈

The future of Green Software lies in education. In this episode host Chris Adams is joined by Luis Cruz of TU Delft and Sara Bergman of Microsoft as they discuss how to teach and learn to make Green Software. With Luis’ unique insight as a professor, he is able to bring us up to date with pedagogical approaches that are laying the groundwork for future software engineers.

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Sara Bergman: LinkedIn / Twitter
Luis Cruz: LinkedIn / Twitter / Website (with course info).

Episode resources:

People to follow doing relevant work in this field:

Person: Simin Nadjm-Tehrani - IEEE Xplore Author Profile
Person: Hannah Smith: Educator - Hannah Smith (opcan.co.uk)
Person: Sabine Canditt and her book Small steps. Large... by Sabine Canditt
Person: Patricia Lago: Full Professor at VUAmsterdam.
Person: Elina Eriksson: Associate Professor KTH.
Person: Mikael Asplund: Linköping University.
Person: Sandra Pallier - ClimateAction.tech organiser and designer at Microsoft.
Person: Colleen Josephson - VM Ware. Collen’s recent talk and paper at ACM Energy.
Person: Bilge Acun - Facebook/Meta - See Carbon Explorer on github.

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Watch our videos on The Green Software Foundation YouTube Channel!
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:
Luis Cruz: If you want to change the generation of software engineers, you need to change the mindset of their leaders and you need to start earlier. And the obvious answer for that is to include green software as part of our university courses.

My name's Chris Adams. And on this episode, I am joined by Sara Bergman of Microsoft and Luis Cruz of TU Delft. That's technical university of Delft. We'll be talking about how people teach and learn to make green software. Before we dive in though, maybe it's time for a quick round of introductions.

My name is Chris Adams. I am the chair of the Green Software Foundation policy group, and I'm also the executive director of the Green Web Foundation, where we work towards an entirely fossil for internet by 2030, Luis. Would you be so kind as to introduce yourself, then we'll give Sara a chance to explain, introduce herself and see what she, and what she does.

Luis Cruz: Sure thing. Hello everyone. My name is Luis Cruz, and I'm an assistant professor at the software engineering group at TU Delft. And there we do research and educate the next generation of software developers. and our main topics, we've overrun, software testing, software deducting, and of course also about green software, how to build energy efficient software.

So most of my time is devoted on doing research on green software, green AI, sustainable software engineering, and of course also about educating and finally about managing because when we have a lot of people, we also have to manage. People and students are also part of these managing and management load that we have at the universities, like to you, Chris.

Chris Adams: Luis. And, uh, Sara, one of our recurring hosts. Why don't you introduce yourself for some of the folks who may not have heard the previous episodes?

Sara Bergman: Great. Thanks. Always great to be back on the podcast. My name is Sara Bergman. I am a software engineer at Microsoft where I work in M 365 or with M 365 products. And in the Green Software Foundation, I'm the chair of the writer's project, where we deal with some of our written materials, like the articles we produce and the newsletter that you're all hopefully subscribed to.

Chris Adams: Okay, thank you, Sara. And just for the record, while you brought that up, if someone did want to subscribe to the newsletter and find out about this through regular and. Updates, where would they be going to find out such a thing?

Sara Bergman: you go to our website, Green Software Foundation and it should be like right in your face.

Chris Adams: Okay. Awesome. I think even I could follow that, actually. Okay. So today we're talking about basically teaching, teaching green software, because if we're gonna be. Helping people figure out how to make more planet friendly applications. Then it probably helps to know what's going on in the world of teaching software and see where, where people are learning this stuff.

And this is partly why Luis is kind of, I'm glad to have you on for this because we actually met through things like the climate action.tech slack group. But one thing that was, I found really exciting from some of the work you did was the fact that you were working on. Entirely open syllabus for sustainable software engineering and obviously lots of us like open source, an open, an open culture, but I figured it might be worth just maybe you telling us a little bit about why the core, why you made the course and why you chose to make it open.

Luis Cruz: Yeah, that's a very good question, Chris. And that's a, maybe let me get back a little bit and tell you. How did all this started? Yeah, so it started by the fact that I was doing research in green software way before teaching sustainable software engineering and the always the main motivation point was the fact that.

Software engineers are eager to build more energy efficient software, but they often lack the resources or they often don't know exactly how to do it, but they, they are highly motivated to do it. So they keep asking about what are the best practices, how can we test energy efficiency, these sort of things.

So, so that was my main motivation to start this line of research, but then of course, I started realizing, okay. I should not only be doing research in my. Office, right. I should also try to communicate these back to developers. And that was when I realized that, in fact, it's not only about building or creating writing papers, but also about communicate them back to the community when, and when you Google about it, or when you look at the Web for resources about this, you almost don't see anything about it.

And then if you look at the conversations that we have. Climate action. How to change our software industry into a more sustainable industry. It always revolve around new policies, right? How to make sure that the industry is complying. with sustainable practices, how to make sure that we can ask our suppliers for energy efficient certificates, these sort of things.

But sometimes we forget that if you want to change our society. The strongest weapon we have is education. Right? If you want to change the generation of software engineers of the software industry, you need to change the mindset of their leaders and you need to start earlier and. To me, the obvious answer for that is to include green software as part of our university courses, because that's how you change the mindset of the future leaders in the next generation of software engineering.

But again, when I started, when I realized this, I started looking for content at, in my university, okay. Who is teaching these kind of topics. And I realized. , there was a lot of content around sustainability at UD Delft, but they kind of miss the, the fact that software also has an impact on sustainability.

So if you build a software system, this software system will probably have a carbon impact. And somehow I feel that the other fields kind of miss this point because they tend to. Target sustainable problems by creating a new AI model, creating a new software system that will automate something and help them improve the sustainability of their problems.

And they kind of forget that if you run something in a server, that server will probably be spending energy. So going back to the, to these pursuit of sustainability topics at the university, I realized that the computer science courses had nothing to do about. . And the funny thing is that if you think about green software, you will see that it, it, it covers, or it touches upon different topics of computer science testing, but programming languages, even in software engineering, the way you, the software development life cycle.

If you look at all the stages, every single stage. Should have a perspective on sustainability if you really want to make sustainable software. So in a way, the fact that none of these courses in computer science had a single chapter that would cover energy efficiency or sustainability was somehow a trigger for me and more motivation for me to think, okay, we really need to start creating.

Content. And at the Delft, they were very open. When I started pitching about a new course idea, they thought, okay, this is really something that aligns with our values. So they gave me all the freedom to start designing the course and creating the course. and I'm really happy that this first edition we had around 20, 20 something students that were really enthusiastic about the topic and they knew nothing about it.

Because as I said, there was nothing around this and there were already. Fourth year master students. So after four years of education, they, they were not exposed to this to me, the fact that it was so challenging to find content, to help me design the course. I think it was clear to me that I would, if I wanted to change our community, I would have to deliver this content and everyone can use it and reuse it and learn from it because that's the way that we can make.

A second addition of this course, or someone else can make an improved addition of a sustainable software engineering course. And that's the main goal. So this, this is not about getting a recognition. This is about changing society. And, and I think the best thing we can do to change society is to make this content open.

Chris Adams: Cool. So I have a couple of questions, but before I come in on that, I just want to see Sarah. Is there anything that you want to kind of come in on there before I come in with my questions on this?

Sara Bergman: This is where I wish the podcast was like a video recording. Cause you would've seen me nodding vigorously throughout entire thing. Yeah. There's so much of this that resonates very deeply with me. And, and I was lucky enough to be one like not, not one of your students, but when I was a student a few years ago in, in Sweden, we did have a course called green.

It led by an amazing professor called Simin who basically. she like her, the whole course was based on reading research papers. So it's like she has been a researcher in this field for a long time about energy efficiency. And she was basically like, don't just take my word for it. Take all of these other researchers word for it.

And it really opened my eyes a lot. And I know it changed other people's mindset to took the course. And that was the one course that made me wanna pursue this more. It's such a powerful tool and yeah, just a big plus one, I guess.

Chris Adams: Cool. Right. Well, Louis, when students are actually taking this course, were there some parts of introducing a sustainability element that seem to work particularly easily? Or were there some parts that they particularly struggled with? Because what I've learned is that some things can be a little bit counterintuitive and yeah, I'm really curious, cuz I've never heard from someone who's actually teaching, you know, undergrads home or even postgraduates for them to kind of pick up some of these skills that are obviously gonna be in demand in the future.

Luis Cruz: Yeah, that is a very, very good question. So one important part of the course is about measuring energy. Option of a software application. Yeah. And I don't want to be boring or anything, but of course, when you have to measure, you need a metric, right. You need to measure a particular metric and this metric needs to be, uh, Clear enough for the person that is analyzing the data.

Yeah. So to make it more precise, when you are measuring energy efficiency, probably you're gonna look into energy consumption, but you also have more metrics. You also have the power consumption, right? And, and power can be the average power over the execution of a particular software, but can also be the instant power that is being.

Uh, used by the, by the CPUs, by the memory, by the whole laptop system. So this is the most issue. So what should I be analyzing energy or power. And depending on the use case, you should be looking at one or another, but then of course you need to make sense out of it. So if your software is wasting 20 jewels after two minutes, I mean, what does it mean?

Is it energy efficient or is it like energy inefficient? Should I do something about it? Should I just assume that it's fine. And this kind of thing requires quite some discussion most of the time, because every single problem is different. Every, every software, every program language will have different thresholds and I won't have a clear answer.

Like I, I always don't have all the answers for every single. Result we get, but one of the cool things about having this course is that then we use that as an opportunity for discussion during the class. And this works quite well. When you have around 20, 30 people in the same room. Because if it's more than that, it's a bit difficult.

Sometimes it happens to have really large classes, but when you have a small group, they will come up with the most interesting observations about this. So that's really exciting when you bring people together just to critically analyze these, these, these bullet. And another thing that I'd like to bring up here is that, so my main passion goes around energy efficiency.

So the most of the course is about energy efficiency and software, but we also focus on other aspects of sustainability. Individual sustainability as in how to make sure that your software organization is actually helping developers or all the stakeholders be more productive and satisfied with the work environment.

So we call it individual sustainability. and the social sustainability, how your software is affecting democracy or how is it affecting the wellbeing of individuals in the soci in the society? And these kind of things are also. Also bring very interesting questions that not always have clear or not, not always everyone is on the same page.

For example, I can bring up the concept of inclusive programming language, but sometimes it's very difficult for some students to understand some, some of the guidelines that we have to make sure that this particular software is inclusive.

Chris Adams: so there's one thing that you spoke about was this, as I understand it, there's an issue about units. Like, what am I using? Cause you mentioned was like power and energy, for example. So that was one thing that you mentioned that was something that students struggle with. And then this thing about like inclusive programming is that like how easy it might be to learn in terms of how many concepts you need to have in your head.

Something like Python or by comparison to like say C plus plus, which has a lot more features than a language, for example, is it that kind of.

Luis Cruz: Now, when we, when we talk about inclusiveness, it's more about some keywords that you might have in your code that might have a meaning in our society. That. That might be not so inclusive. Yeah. For example, there are some guidelines around inclusive language that recommend you not use dummy Variables.

For example, like, uh, dummy variable is something that every code learns, uh, to use, uh, since the first stages, because sometimes you need to have a done variable.

Chris Adams: this is like food bar, Baz, that kind of stuff. When you say dummy. Okay.

Luis Cruz: And, and, and, and these dummy, some guidelines say, Hmm, you might be, you might want to use a different. Word here that, that doesn't denote that someone is silly or this kind of things, or that someone doesn't is not mentally well.

So maybe you want to use just like, I, I don't remember the replacement, but there is a dictionary with a few, uh, new replacements for all these Variables. And usually this is, this is so new for everyone that if you look at the code, no one is following these practices, but every single. Part of this let's call a dictionary has a meaning behind it that it's not immediate for us because we are kind of privileged and that we, we don't need to worry about this.

Or we, we never suffer from, from the users of these names in society, but they will actually affect someone. And, and that's why many organizations are starting to bring this up.

Sara Bergman: I think this is really interesting, cuz I've I remember being to the NDC conference, the conference here in Oslo, it was remote cuz it was 2020 I believe. And it was a keynote about the same concept and I had. thought about a lot of this concept otherwise, but things. Why is it called a whitelist and a blacklist?

When we really mean an allow list and a block list, those words are actually more descriptive, but it's true tradition. We've used other words, for example, if you mean a primary and a secondary data source, why would you use terminology like monster slave, cuz really what you mean is primary and secondary.

They, they better explain the scenario and they're also more inclusive, but I think that's a learning curve for, for a lot of people. No matter your, your age or experience level in the.

Chris Adams: Yeah, I agree. This is something that you you'll see in loads of code bases, kind of quote unquote in the wild as well. And like a, you do see some organiza, some large projects actually starting to adopt this kind of stuff, but on the subject of. Particular kind of like terminology, you said one or two really interesting things there.

Luis, you spoke about the idea of, okay. There's efficiency, but there's also like the individual sustainability and also the impact that you might have on the outside world. And I suppose internally we have. I mean when recently I, I was on a panel recently, and one discussion was basically about this distinction between say green it and green software.

And it for sustainability, like this idea that these are two separate concepts, which may be related, but they're different from each other. And a lot of the time it's very easy to see these ideas being conflated a lot. And, uh, I wonder if this is something that you've seen you've seen on the course, or if you've, or this is something students have been able to kind of pick up quite intuitively, for example,

Luis Cruz: Well, to be honest, I do know the answer. Case on the first class, I made sure that we all have a common definition, but I could imagine that some of my students were expecting something else when they joined the course. Yeah. Because indeed there are so many different names around sustainability about around the green software.

And when I mention green AI, people immediately think about AI for energy efficiency or AI for social good. And. when I talk about green AI, and I'm not saying that I have the right definition. because there are so many definitions that it's difficult to keep up with the right ones. But when I talk about green AI immediately think about building AI systems that are energy efficient, and they have minimal footprint foot painting our.

So, but I'm not claiming that I have the right definition because this is so new. I mean, the first paper on green AI is like three or four years old. And I don't think we are already at the stage where we can say that our, our field is mature, that we can already. Settle on these definitions because maybe next year a new problem will bring will, will be brought into the picture and we need another definition.

And then suddenly we need to rethink about this. And indeed, as you were saying, this, this can be quite, not only encounter intuitive, but confusing. And, and the simple fact that when we talk about sustainable software engineering, Uh, different people might have different definitions. And I'm gonna challenge that definition a little bit.

So some definitions, when we talk about sustainable software engineering also include economical, sustainability. and technical sustainability. And of course we, we don't care about economical, sustainability. Everyone is taking care of that. Saudi the world runs and we don't care about technical sustainability because software is a Newfield is already.

Quite mature for 60 years, we have been studying technical sustainability for software, but just to give you an idea that this doesn't make our life easier, but the good thing is that once we start getting our hands dirty, once we start measuring energy consumption and. Extracting knowledge out of these data and experiments, definitions are just a way of communicating it.

But what we need is groundwork people like-minded that are willing to create new tools, do new experiments, things like. What is the most energy efficient video call platform? Is it zoom? Is it Microsoft teams? No one use, no, no one knows, but if I knew I'll start doing all my calls with a particular system, because I'll know that I'll know there is an impact.

They are actually worrying about it and they are actually trying to, to, to do something about it. So they, they deserve having more users.

Sara Bergman: Yeah, and I think that's an interesting point as well. If you think about students and, and I can only speak from my own experience as I'm gonna self proclaim as a young person, as a millennial or a zillennial, it's a new word I've learned, which is the people between millennial and gen Z. I don't know.

Maybe that's where I fall in. Anyway, I think my generation grew up more with climate awareness. There was never a grace period in our life where we were not aware of climate change or where we did not feel the. Oh, climate change. I cannot remember a time in my life where I was told because I've always known sort of cuz I was, I, I, I suppose I was told at some point that I was too young to remember that now.

So there is an increased desire, I think, to, to lean in and to be part of the solution. I think an easier way for people is, is to use you say you become a software engineer, cuz that's your passion or, or a like researcher in the field, but you still wanna do something to help. And. Applying technology to a climate problem seems like the easiest thing you can do.

Like, yes, I love AI. I wanna use it for good. And that's great. Like we're not saying stop doing that, but I think when you flip it on the end and say, okay, but if the way you're doing this is actually unsustainable, it's kind of maybe not. counterintuitive, but not really meeting what you're trying to do.

And I've been speaking a lot externally, specifically about machine learning and AI. And almost every time I get people after almost frustrated, like, but I wanna know what I can do. And it's like, yeah, but I totally, you, a bunch of things you can do. No, but I don't wanna, I just wanna do my normal stuff, like code away and then like contribute to something.

And I think as a mindset, mind shift that need to happen and you can definitely do both, but I think we also need a bit of marketing for like this, what you said Luis, about building these libraries, doing these case studies, doing the groundwork to figure out, cause it's still very new, but that doesn't mean there are no opportunities and no chances for people to contribute there on the opposite.

There are tons of chances for people to contribute and, and make it a big, significant contribution and difference to, to this.

Chris Adams: So, this is actually maybe a good time to plug for people. Who've never heard this who haven't, haven't listened to previous episodes. Some of the work we did when we looked at AI green, eye green, AI and AI for sustainability. Of issues ago in that we had both, um, Dr. Lynn Kaack, sorry professor Lynn Kaack talking about basic AI for sustainability and some of the specific applications you could actually see in, and also where some of the blind spots might be when it comes to regulators.

But the nice thing was that there was a paper that, that, that she'd mentioned, but also one of the other people on the other guests will, I totally forgot Will's name, but his paper was actually published specifically, like iterating through all these techniques you can actually use. So, if you look through the previous episodes and the show notes, you'll see the paper, which literally says these are some of the techniques you're using.

So if you're using technique a here's how to, here's how to use technique B for almost the same impact, or these are the trade offs you might actually have. And I, Sara, I, I kind of wanna ask you a little bit about some of the study, some of the work that you studied on this. Cuz as I understand it, you were doing a bit of work.

Consensus mechanisms. I don't wanna, let's not dive too deeply into the world of, of cryptocurrencies and, and greening cryptocurrencies, cuz to an extent it feels like the price of the falling price is sorting some of the out for anyway. But there is a whole thing. Like you, you did some work comparing, I think some tools around this, like a like Hyperledger.

Sara Bergman: Exactly. Yeah. Yeah. So back this was my master thesis actually, which was later turned into a paper. At the time blockchain technology, I'm not, I'm gonna refer to it as blockchain technologies sort of remove the economics from it because you can view it as an economic structure, or you can view it as a data structure for this talk.

We're gonna look at it as a data structure, cuz that allows us more freedom in some sense. So at the time blockchain technology was really new. There were very, very few studies and to the best of our knowledge, the study I did was the first study where we actually compared. Blockchain technology to distribute the databases cause especially for permission, blockchains, so where you, where you have traffic, they can achieve a lot of the same use cases.

And it can be interesting for you as a practitioner to know when you should select one or the other. Because in theory, in my mind, a blockchain is kind of like an append only data structure. and that sounded interesting to, to investigate more, cuz it's very different from a traditional distributed database, even though they can do some things very similarly.

So I looked specifically at Hyperledger fabric, which was still kind of new at the time and at Cassandra who, who was a bit older at the time and did a performance comparison, cuz performances is one aspect you can like sustainability is very multifaceted. and when you have a master thesis, you kind of have limited times you can't delve into all the things.

Uh, but yeah, that was, that was the work that I did.

Chris Adams: So Cassandra, a distributed kind of key value store essentially, and a Hyperledger, which I, I, I can't even begin to understand how, uh, or, or talk confidently about hype, about her, the internals of Hyperledger. But I do know, like, since that's been published, There's actually been quite a lot of interesting work moving on from just looking at kind of key value stores, like, okay.

Cassandra, for example, I know that Cassandra's mostly written in Java, is it, is it not? It's primarily on the G JVM, right?

Sara Bergman: I think so. Yes.

Chris Adams: Yeah. I know that there is, there's essentially one, a kind of technique that I've seen relatively new, which I'm. I think I'm borrowing a term that IBM coined about a decade ago called scale in which I think is really interesting because it's essentially creating a API compatible implementation of an existing piece of software that is much, much more efficient in many cases.

So the canonical example would be say, ScyllaDB it's written in a very particular kind of C++ uses like a shared nothing kind of architecture. That basically means that it's. API compatible, but quite a bit faster. And, and as a result, you're able to kind of run the same workloads on a, on a fraction of the kind of resources.

And this is one thing that I've seen. And I'd be really curious about seeing some of that. Well, seeing that study rerun again, because you see that this is a pattern that I've. Seen once. And, uh, I think there are actually other examples. If you do stuff with SQL light these days, there's one called duck DB, which essentially lets you do the kind of large scale or medium data analytics these days, but onto something which is running in a single process.

There's a bunch of really interesting stuff in this field right now, actually.

Sara Bergman: Yeah. And, and I think what, what you said there in the middle was the sort of key, I think performance might not be an immediate. Into sustainability and green software, but the, the packed tighter is, is the keyboard here. Cuz if you use fewer CPU cycles, while you have more free CPU cycles to use for other things, that means you can use less resources.

So that's why we write efficiency is very important. That's also, when we look at machine learning, for example, the training. Time. So the latency you have for that is a very interesting factor for sustainability. And there's been papers published on this to support. It's just not my opinion, other smarter people's opinions as well.

Chris Adams: All right. I'm glad you mentioned the papers thing because Luis, this is actually one thing that I was gonna, I'll just ask you about the course. We spoke mostly about energy efficiency, so far as one or resource efficiency, but like, you know, hardware usage, for example, do. Have have students have had had much luck understanding one of the other pillars, the idea of like carbon efficiency or being able to change the intensity of the resources you might use.

Cuz this is one thing that is relatively new and it's something we've spoken about and where you can use some tools for this. Like what time, for example, to see. How green energy might be. And I wanna name shook a couple of papers, which have been doing some really, really cool stuff in this, but I figured it might be ask worth asking you how students respond to this part here.

Or if that even shows up in the course.

Luis Cruz: Yeah. So I, I was not very long on the, when I talked about units, but we have a. A single class dedicated to units on energy efficiency. And we start with the energy consumption, of course, power consumption, but then we go all the way to, to the data center level in which we no longer think only about energy efficiency, but we also need to think about the carbon footprint carbon.

I. The carbon consumption. What does it mean to be carbon efficient? I mean, what is exactly carbon, right? It's not only about CO2. It's also about all the other CO2 equivalent cases. And, and this, this kind of thing is not clear to everyone, even to me, sometimes I need to go back to the, my documentation and check whether I get the definitions.

Right. So. one class is fully to it, but one cool thing is after, after this class, I think most students got it right? If not everyone. So, so one part of the course is you need to come up with a project that revolves around the tool that will help software engineers build energy efficient software. Or sustainable software, I'd say.

And one of the projects that was developed by one of the groups was a scheduler of software tasks that will take into account the carbon footprint or the carbon intensity of the grid. So that means that when you are scheduling a task, you can say this task takes two hours, but I don't mind waiting 10 hours to get the result.

And then based on this, the scheduler would check what would be the best time to run this task. And this is a very simple idea that is based on a paper from Google. If I'm not mistaken where they have carbon aware data centers in it, they basically decide when a particular task can be run or not, but they have a very complex setup and these students thought, okay, that's a really cool idea.

It works for large organizations, but why not give. The power of these idea tools, just the ordinary software developer, that, that also wants to do something around these lines. So they created the very basic tool. I can share the link to it. It's a prototype of course, because they only had four weeks to develop it.

But you, you basically just. Define, what is the common you wanna execute? What is your time constraints? And there you go. We have a carbon efficient, uh, execution, and I find it quite cool. Uh, another thing, if you, if you allow me, Chris, and since we mentioned blockchain and energy efficiency in the same paragraph, and usually get a lot of people angry about it, regardless of the, uh, The idea or opinion we have about it.

I just wanted to say that, although I'm not against blockchain, I, I think we, we need blockchain, but that also means that we need energy efficiency in blockchain. So we, we need energy efficiency and blockchain and the same paragraph more often, and we need to have constructive conversations. So we need everyone, even the ones that.

Getting rich with Bitcoins. We need all these people to, to, to think about how to solve this problem, because this is a real problem. And, and I'm gonna stop here. I don't want angry people around there.

Chris Adams: Okay. Well, there's one thing that you've actually pointed to. One thing you've mentioned. I mean, if with your position, as a professor, I'm being able to like, have some influence on new developers and new new engineers, I'll happily let you give you that platform to like, say that. Yeah. You should probably be caring about that because.

Yeah. I mean, this is actually one thing we were talking about before Sara, you mentioned one of the reasons you chose to study what you studied on your masters was because there was some influence from one of your professors and they kind of like gave you some of the space to even think about this stuff.

And maybe you might would like explain that, cuz that was a really nice story. I thought.

Sara Bergman: Yeah, absolutely. And I touched briefly on it in, in earlier as well, but this green, it course that I was able to take in my fourth year. Actually, it, it really opened my eyes also to this concept of the duality of it. Like we've talked about in this podcast, not only that you can use software for good purposes, but the software itself actually comes from somewhere and, and data centers are huge when.

If you've never been to a data center, it's kind of hard to fathom. Like they're very fussy in like concept, at least to me, like if you come from a kind of normal non-data science background as a kid, which most kids do field trips to data centers, aren't that common yet, then it can be hard to understand just how big this impact can be.

Around the time I took this course, it was at the time relatively new paper. And I was like older paper that compared the carbon footprint of the ICT sector and placed the dust at 2%, which was the same as aviation. And that's a number I still see quoted over and over and over again because it's so impactful.

Cuz we talk a lot about aviation, but we talk less, but not more about. So, yeah. Simin who was the professor of that course and Mikael Asplund, who is on the big influence from Linköping university in Sweden. They were my supervisor and oh, I dunno the English word for it. Whoever approves your master thesis person.

Luis Cruz: Maybe promoter.

Sara Bergman: yeah, possibly that's the word I'll I'll I'll address you. Yeah. To give me the space to, to explore this deep end and they have done more research on energy EF. later as well. And I think it, it is so important to have that influence when you are sort of mallable I guess you never stop being mallable, but you're truly mallable at university.

So I am forever grateful to having had that opportunity so early. And actually my university enforces you, everyone has to take one course that has a sustainability topic. You cannot graduate until you've done that. Which also think is a good way. Cuz there are like, for some reason, some political. Tied into this, which I personally, I think is unnecessary, but there are for some people, but this way is a good way to make sure everyone at least gets a baseline of knowledge,

Chris Adams: All right. So we're coming up to last, like say a few minutes of the episode. So I figured while we're talking about some of the people whose work that's. who you'd recommend or point people to. And what add is it Simin? TU? I need you to pronounce it one time. Cause I'm not very good at that name. I'm sorry.

Sara Bergman: Simin Nadjm-Tehrani

Chris Adams: okay. Well make sure that's definitely in the show notes because that's, there's definitely worth looking at well we're here. I'm Luis. Is there any, are there any people you'd recommend or you think is whose work. Suggest people look up because there's a couple of names that I found really interesting of late, and I figure this might be a nice way to kind of wrap up with a top topic of pedagogy and learning from existing state of the art.

For example.

Luis Cruz: Yeah, I, I, I can, on top of my head, I can think of two rock stars. Elina Eriksson. She's an associate professor KTH. She has been awarded a number of times for her work on a education of sustainable software. Sustainable I CT in the universities and of course, Patricia Lago from view Amsterdam. She's one of the first researchers in green software in our software engineering C.

So, if you want to have an interesting landscape of, or, or, or a role model for research and education on sustainable software engineering, they are definitely, uh, worth, uh, follow.

Chris Adams: All right. I'll add a couple because just recently. There was a series of conferences. The ACM had a conference specifically about energy and there was one paper, which was co-authored. I think Colleen Josephson was the, uh, one of the lead authors. She did some really interesting stuff about this whole idea about like your student mentioned about moving.

Uh, basically, uh, time shifting compute in various places. That was one thing that I found really fascinating, and that was actually actually talking about, okay, she works at VM where these are the numbers we've sat, we we've seen and been able to use. And here's how we're actually where to productionize some of this stuff.

And the other work that I think is really interesting, I'm probably gonna, uh, mispronounce her name as well. And I'm so sorry, uh, for the author, I think it's Bilge, Bilge Acun or. Um, there was a work called carbon Explorer, which is both online on GitHub and is also a really fascinating paper that was also presented at that conference, which I think is consider as like one of the state-of-the-art things here.

Actually, I found it really, really worth looking through, because once again, this was someone who's worked doing some work at Facebook, looking at the ways that they're able to achieve the targets. Carbon reduction year on year going through this. And they both looked at the load shifting part. They looked at using mixing in renewables and they actually put together a server to, for each, each, every single data center.

Talk about what the optimal mix was for each of these. I've never seen that work before and it's really, really worth looking. And I've when I looked up at work, there's like just so much there to look at. It's so nice seeing that stuff in the open. Sara, I wanna give you kind of the last, the last word on this, cuz I'm sure there's some other people who've, who've actually found interesting or whose work.

I know fan of, or you direct people's eyes to, I suppose

Sara Bergman: Absolutely Sabine Canditt she recently wrote a book. It's in German. So sadly I can't read it, but I've had the synopsis explain to me. And she, she talked to me about the book I was interviewed also for it. Hannah Smith is another great educator. Who's involved in climate action, tech and green tech Southwest, I believe is the community.

And also Sandra Pallinger from, from Microsoft. Who's also involved in climate action tech, who I really.

Chris Adams: I didn't know that you were gonna mention Hannah, but, uh, the thing I probably need to say is that Hannah recently joined, uh, the organization, a work app called the Green Web Foundation. And she's now heading up all the training. So I'm, I really glad, happy to hear that. You've mentioned that because yeah, she's working on a bunch of that stuff.

So if this is interesting, 21, who's listening then. Well, I guess there's a plug Luis. I feel a bit embarrassed about mention talking about our own organization here. So I'm probably gonna. Try and leave the last word with you about this, because you've mentioned this course. If people do wanna find out more about this work that you're doing, or some of the other output that you are, you'd like to kind of people to look at, is there a particular website or is there a space you'd share?

Cuz we'll share the link to the, the, the engineering course, but there may be other things that you wanna draw people's attention to for.

Luis Cruz: Yeah. I mean, definitely. If you wanna have a quick start on sustainable software engineering, please do share the link at the end of the podcast podcast. I'm not gonna say it out loudly, but please do check it. I also write frequently. So if you check my personal website, I have a blog where I try to write the same content of the course, but in a way that is more friendly for the internet reader.

so I try to write it in the form of blog posts. So if, if you wanna reach out, of course, you can also follow me on Twitter, please. I'm Cruz, and I'm more than happy to interact to, to hear any thoughts or any feedback about the course. If there are more ideas. That are more content that I should add, please let me know.

And I'm, I'm making this request not only to salary increase, but to anyone out there, because I think this is how we can evolve and how we can change our software industry.

Chris Adams: Cool. Thank you, Luis.. Well, I think that's taken us up to the time we have left available. Sara. Luis. Thank you so much for joining. Yeah. Have a lovely evening or morning wherever you're in the world. Okay. Take care. Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on apple podcasts, Spotify, Google podcasts, or wherever you get your podcasts.

And please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we'd love to have more listeners. To find out more about the Green Software Foundation, please visit greensoftware.foundation that's greensoftware.foundation in any browser.

Thanks again, and see you in the next episode.

Accessibility and Green Software

11 juli 2022 | 31 min

This episode is taken from a recent panel discussion from the Green Software Foundation’s Global Summit of 2022. EV regulars, Chris Adams and Asim Hussain are joined by Anne Currie, Chris Lloyd-Jones and Elise Zelechowski as they discuss Accessibility and Green Software. What’s driving interest in Green Software? What ESG principles are being adopted by companies and what is needed to drive accountability and accessibility in this sphere? Where does the main driver for Green Software come from within an organisation?

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Anne Currie: LinkedIn / Website
Asim Hussain: LinkedIn / Twitter
Chris Lloyd-Jones: LinkedIn / Twitter
Elise Zelechowski: LinkedIn

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Green Software Foundation Summit
The Green Software Foundation YouTube Channel!
Policy: Blaue Engel / Blue Angel

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Connect with us on Twitter, Github and LinkedIn!

Transcription below:

Elise Zelechowski: I am excited about the Green Software Foundation for the very reason that I think this is about a movement of technologists who are really sort of coming up with innovative creative ways to address this problem. And so the way we need to take this forward is get CIOs to the table and say, now, how do we sort of come together and think about mutual benefit?

How do we sort of make this a systems approach within the organization to tackle all the different parts that we need to tackle?

Asim Hussain: Hello and welcome to Environment Variables. Brought to you by the Green Software Foundation. In each episode, we discussed the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software and your host Asim Hussain.

Chris Skipper: Hey everyone. Chris, the producer of Environment Variables here with a quick note to say that the following episode of the podcast was taken from a panel discussion from the recent Green Software Foundation global summit. If you would like to hear and see an unfiltered raw version of this discussion, head on over to our YouTube channel and look for the video entitled GSF global summit closing ceremony, 2022.

In fact, I'll do you one better and add a link below in the show notes of the podcast. So head there, click the link and you can check out not only that discussion, but all the other episodes of the Environment Variables podcast too. Now on with the show.

Asim Hussain: Welcome, my name Is Asim Hussain and I'm excited today to host a panel with guests from the Green Software Foundation.

Chris Lloyd-Jones: I'm Chris Lloyd-Jones.

Chris Adams: Hi there. My name's Chris Adams. Oh, sorry, Anne.

Anne Currie: hi, I'm Anne Currie.

Elise Zelechowski: Hi, Elise Zelechowski.

Asim Hussain: Awesome. I've got some wonderful, uh, interesting questions, hopefully for all of you. And I'd love to hear your opinions. Very interesting topics around green software. My first question is, you know, what's driving, I personally have experienced a lot more interest in this space in kind of the last year, at the very least.

What do you, think's driving interest in green software in, across the industry, Chris Lloyd-Jones. Why don't you go ahead with that one.

Chris Lloyd-Jones: So for me, I think it's the growing understanding from individuals that our way of life needs to be greener, to be sustainable and be resilient. And now there's a lot of talk from governments and around setting targets and policies, and those are very important, but green software lets every individual make an impact and make a difference, which I think is a differentiator from all the other initiatives that are going.

Asim Hussain: Mm. Yeah, really good point. I, I, I feel there's a ground spell movement across the world as well. Yes. Elise, what's your, what's your thoughts on that? On that question?

Elise Zelechowski: I would, I would echo that and just say, I mean, I think that technologists today are, are, are growing up with more and more real sort of impacts from climate change. They're seeing them manifested in ways that are really impacting how they experience the world around them and how they think about their future.

And organizations like the Green Software Foundation that really provide a forum to take action and to feel like you have agency to really make a change and sort of really not just make a change broadly, but make a change in what you're building every day really gives people a sense of the power to change.

And I think that that is one of the, the most exciting things about the, the Green Software Foundation. I also just think that, you know, Organizations are increasingly adopting ESG frameworks. I think there's more and more accountability expectations around that. And of course, organizations that, that have an outsized emissions footprint from their it right, are gonna be looking at green software as one of the big opportunities to make an impact.

Asim Hussain: Wonderful that, that that's. Yeah, that's definitely aligned to my thinking as well. So given the fact that organizations. You know, they're interested in green software, but they need to adopt, you know, processes, principles, patterns of, of engineering in order to greenify their, their software and reduce the emissions of software, what what's needed in your mind for organizations to, to adopt these patterns and principles, Chris Adams.

What's your thoughts in that area?

Chris Adams: So what really, our big, big one, right. Is capacity, like organizations having the skills in house to do this. And you know how, when like Ruby rails might have come out and then people are saying, I want a Ruby rails developer with 10 years experience. When the frameworks only a year old, we are kind of seeing something a little bit like that right now, where various organizations are trying to hire and ask for this stuff.

And they're not quite sure what they're after and we have this kind of shortage right now. So if anything, there's a real. Need to have some kind of way to provide, to kind of get this level of, of, of skill and competency up quite high, actually. And, and, well, I think this is something you've spoken about quite a lot, actually, as.

Anne Currie: Yes. Yeah. Cause we have to make sure that being green doesn't conflict with developer productivity in anybody's mind, because in the end developer productivity will always win. There are hardly any developers that compared to how many we need in the industry, we need more. And so everybody wants to use them to deliver on business goals, not kind.

Climate goals really. So one thing that I think will really help and is definitely helping at the moment is where you get open source or cloud providers who are offering services that are both good for developer productivity and also greener. The more we can, we can raise the profile of that and push it.

The more likely we are to get, take up, because what we cannot ask people to do, because they will not do it. I've asked them so many times over the years, we can't get people to rewrite their applications in sea or rust or making things more efficient is too expensive for people in terms of time. So we need to find services that will do it for them, make it the, the default, make it no or effort for them, or in fact less effort for them to choose those solutions.

Asim Hussain: Know, I do. I do think that one of the challenges is that we're either talking about software that hasn't been written. and the vast majority of software, which needs to be written has not been written yet. We are very, very early stages in this whole industry we're in right now, but there is also this legacy software that exists that we just have to figure out how do we manage that?

And rewriting that in a new language is not an option that, you know, that flies these days, but there is new development's happening, or maybe some of that stuff will. Will fly. Who made those choices early on? One of the things Chris, I just thought, I D dunno if you have any opinions on this or whether you've seen anything else on the horizon.

I, I was very, very delighted to see, I think just yesterday, the day before there were, there are several, you know, job roles out there. I saw some for Amazon. Of a sustainable solutions architect and I'm starting to see a more and more and more. And I know some of your organizations have roles, you know, with the title, agreeing the title.

I mean, what's, what do you see in terms of the job market regarding kind of our space? Do you see a lot of growth in this area?

Chris Adams: is definitely interest in this often at a kind of executive level and often at a kind of developer on the, kind of in the trenches level. But. The part in the middle people, aren't quite sure how to kind of prioritize it or even ask for, or even specify it right now. So like people asking, well, how do I buy a green, a greener version of any kind of service right now?

There's a real gap there right now. And I think even at the architectural level, we don't really have the language yet. Or it's not that common to really kind of talk about the trade offs you might be doing. So like, I could do one thing here, which takes advantage of how the energy markets have totally changed over the last 10 years to, to change the economics.

Right. But I dunno how to talk about the, the tradeoffs that might be, that I might make in order to unlock that stuff. And I think you kind of need this capacity or these abilities at the design level right now, before you can kind of get to the implementation part. That's what I feel is a real kind of.

Gap right now until we have like an iron triangle for things like greener software, for example, or this stuff here. I think we're gonna struggle to have like, informed decisions about how you're gonna spend a developer's time budget or an actual cash money budget on things. That's probably one of the big gaps I see in the next six to 12 months, that needs to be kind of plugged, I suppose.

Chris Lloyd-Jones: I just wanna weigh in on the chasm there. One thing that I've noticed is that some organizations are starting to see this as an either or that you either focus on sustainability in your business model and secular economy, or you do green software. Whereas actually the personas and the people doing these things are very different.

I think you can have sustainability in your technology and sustainability with technology. And I think organizations don't need to take that kind of zero sum approach to focusing on where they need to change.

Asim Hussain: great point, great point. And I think that's that's. The interesting question is, is. There is a gap. There's a gap between different types of roles, but there's also a gap between, as you mentioned, Chris CIO to, to, to in the trenches, they're in the middle, there's kind of this, this knowledge gap. And once that gets filled, then the rest of the roles and the understanding will spread throughout an organization.

But then that leads to one of my favorite questions, which is where will change come from? I think there's been lots. Attention thought strategy attempts in the past, going kind of top down, CIO CEO, down to the, down to the further down organization. And there's been other attempts going from maybe more. I call it in the trenches on the front lines, let's say, oh God.

So all these awful war metaphors, we need to stop using them up, up, up, up the chain. Where do you think the kind of main driver for adoption of green software come from? Which direction I'm gonna ask at least this question first.

Elise Zelechowski: Yeah. I mean, I think that's, that's right. Sort of one of the unique features of, and, and, and, and there is theory of change right at the Green Software Foundation is that we can create kind of a groundswell movement here of technologists who are really driving, driving change. But of course there needs to be support structures in place, right.

And there needs to be enablement from CIOs and other members of the team. And to the point that was just made. You know, there needs to be kind of a systems approach. Think about things as integrated, right? This is not, you know, an either or scenario. We need to kind of look at strategies that embed sustainability approaches throughout the entire space within the company.

But I think that CIOs are increasingly understanding that they play a role in an organization's sustainability ambitions. I think, as I mentioned earlier, I think ESG is, is part of that regulation sort of driving. Expectations around environmental performance. But I think that I am excited about the Green Software Foundation for the very reason that I think this is about a movement of technologists who are, are really sort of coming up with innovative creative ways to address this problem.

And so. The next sort of step or the, the way we need to take this forward is, is get CIOs to the table and say, now how do we sort of come together and think about mutual benefit? How do we sort of make this a systems approach within the organization to tackle all the different parts that we need to tackle?

Asim Hussain: Yeah, a holistic approach. We need everybody at the table, but yeah, they're definitely there. The, I know you have certain opinions.

Anne Currie: do I, I'm a big, I'm a big believer in the middle. Managers as being the, kind of the secret key to this, because 10 years ago in the financial services industry, when they wanted to try and stop everybody being Cowboys, and there'd just been a huge financial crash, whatever, and they wanted everybody's behave a bit better.

They realized that actually the key to that was the middle managers because. People at the bottom have, might have enthusiasm, but they didn't really have the power to make any changes. And they were all together. They didn't how to do it. People at the top would quite often get on board and say, yeah, we wanna do this.

But everybody assumed they were just lying in the middle and would just stamp on all the projects as soon as they, they go anywhere. So we it's, it's kind of senior architects, middle manager, other, otherwise they have a tendency to just kill everything. For good reason, because they know that loads of projects get kicked off that are actually banned for the business.

And their job is usually to keep the business alive and running and, and operating and not easily. No, not because somebody's NA the CEO's neighbors once said that they thought this was a good idea. So everybody has to drop everything and do it. They're there to protect from that and, and kill those projects.

So we need to persuade them. Otherwise everything's killed is stone dead. So yeah, we need it's. As Elise said, it's holistic. Everybody has to be convinced and we can't leave them to the end or they'll stop.

Asim Hussain: to Anne, we should fire all the CIOs and middle management jobs to say no. Got it.

Um, Chris Adams, were you about to put your hand up?

Chris Adams: Yeah was indeed. Yeah, like a little school, boy. I think there's actually. I think it's worth looking at the role that policy makers have been playing in this in there. And we can also look to other places where people started introducing things like what you might refer to as well. Not really non-functional requirements, but like other kind of key kind of indicators of quality.

And you might think about things like, say accessibility, if you look how accessibility was something, which is now become relatively mainstream in terms of like working. Yeah. You had a push. At a kind of regulatory level where people would say every single website built with public sector money now has to meet a minimum level, but you also needed something like ways to convert those ideas into something that people can act upon.

Like, if you have like a Web content, accessibility guidelines, you got poor, which is like, you know, perceivable. Understandable robust. You kind of need something like that for green software right now to make it a bit easier for people to manage success in this. And so like basically a product manager or someone is able to kind of accept something and say, yeah, this is actually meeting these criteria here.

And right now, We don't quite have that, but it'll be really lovely to have something like that to make it a bit easier for the people who are at that mid-level to essentially guide people along the way, or at least tell when something is actually hitting the targets, they think it's supposed to be hitting, for example.

Asim Hussain: We need a C level to say, yes, we need the middle middle management say yes, but also have the knowledge and the tools and the information to be able to guide and drive and, and, and drive work in, in, in the right direction. I also think it's interesting about the accessibility aspect of it. A lot of people assume that, you know, a lot of accessibility arguments.

Don't have financial benefits. There's huge financial benefits to accessibility arguments mean it's like on some circumstances, it's, you know, significant percentage of your customer base is someone who is, you know, differently. And because it can also be temporal as well, like cuz you can, you know, Maybe have a baby.

So if you don't have a one arm for like six months. Yeah. Well, this is the thing actually seemed like, I mean, so Microsoft did have some stuff like this, which is quite helpful when they spoke about things like situational positional kind of disabilities that is actually really useful in having the vocabulary to actually realize that there are benefits in lots of other places.

We don't quite have that language in. Technology right now, but it does exist. People do talk about co-benefits all the time. You can talk about greening electricity, for example, and saving literally millions of lives each year, that would otherwise be cut short with like particulate matter. There's all these things you can talk about.

And there's arguments you can make to say, well, maybe your staff might wanna hang around more. If they feel like they're part of the good side. For with using advanced humane things, rather than this really, really weird 20th century kind of stuff from before. There's all this stuff that we could be talking about.

And I don't know about you, but I think most developers would rather build star treks than build mad max and.

Chris Lloyd-Jones: I'd say that even with the code benefits you're talking about, I mean, that's kind of the values driven. It is the right thing to do. I think the thing around green software, there's almost a pragmatic business case without kind of wanting to relate everything back to money. If you reduce electricity in a data center, you reduce spend, right?

If you reduce consumption in the cloud, you reduce, spend as much as I think we should do this. Cuz it's the right thing to do. If it does come down to a business case for the bottom line for the pound, the dollar, the Euro. There's also, it can be a very strong, almost linear correlation between reduced carbon and reduced money.

And you can invest that in training, I think to then help those middle managers stop squatting all these projects. Yeah.

Elise Zelechowski: Yeah, no, that's right. The optimization is, is key. And I think that gets back to that point where, you know, companies with sort of outside impact from their, it are moving more quickly to kind of look at where those opportunities are. Right. There's sort of environmental benefits. Right. But there's also that, that big cost saving opportunity when you take this on.

Yeah, absolutely.

Chris Adams: this is the thing, though. Again, we can look from other sectors that do things like this. Like if you value. Then you can literally value it. You can price it. Like you see organizations doing this all the time. Like I'm gonna point to Microsoft because they have done things like this. They have like an internal carbon price.

So you've got a kind of carbon war chest for this cool green stuff. But they've also got things to say, well, maybe there's something which is the. Actually we're struggling to kind of bring down the emissions, right? For example, like an aviation, that's got a different internal carbon price. So you can price things to actually do this.

And like, this is literally what Boeing have been doing for like 20 to 30 years to achieve weight savings. They basically automatically give any engineer a kind of. Budget that they're, they're able to spend to buy weight savings. And I think you can do these kind of, if there's all these management patterns that are, that are like used lots of places, which would work really well in this field.

And now I'm gonna hand to Anne, cuz I think she's got something to say here as

Anne Currie: Yeah, because I, I totally disagree with all of you. So, so this is a good thing on the panel. I totally disagree with you. I don't think, I don't think cost savings sell to enterprises because they don't because for. Them the cost of developers exceeds cost of operating data. They, that's why they move to the cloud, cuz they're willing to pay more in order to give their developers less work to do so.

I don't think that's an, an argument that lad, how having said that, I think it is an argument that that can be made and it is convincing if you operate a data center because there electricity cost is a really key part of your business costs. So aviation. Aviation fuel is a really key part of your business cost.

You will go outta business if aviation fuel, if you spend too much on aviation fuel, but it's not, you will not go out business in the tech industry because you spend too much money on your data center. You go out business because you key business would a business, whatever it is, isn't generating the income you want it to.

It's not your number one priority.

Chris Lloyd-Jones: I think it depends on who you're speaking to though the C-suite level, for sure. That's not gonna make an impact, but again, I'm gonna use your analogy of middle managers. If you are looking at your spend and you look to what the spend was before, and you are seeing your use of function. An Azure or Orlando function, spiking.

And you see your app services. I think even if it might not materially impact the business, those people panic. So having a way to see that they can keep their cost in check. Then also, I guess, sell upwards, manage upwards that they're being green, doing the right things. I'm not actually saying it's a key business driver that helps those people make decisions, which might be a bit Machavellian.

Maybe

Asim Hussain: think, I think it's a co-benefit. And I think that's one of the things there has to be. There's multiple vectors in making decisions. I think I do. I do agree with Anne to a certain degree that I think if it's the only yard when you're putting forward, it won't, it won't land. But if you're saying it's, it's got this benefit, this benefit, this benefit, it is greener.

It's faster. It's more reliable. All those things added together. I do think, I do think it adds it's not nothing.

Chris Adams: When I talk about pricing something, I'm not necessarily saying the cost of a computer, for example, you can cost things in all these different. And there's a whole set of management theory, all around cost of delay, which is like, what is the cost of, of us not shipping this product, for example, the day before, okay, let's choose a bad example like black Monday or something like that.

Right. That is a clear cost of doing that. Right. And that's what I'm talking about. We price things accordingly. And if you're able to talk about this and, and then, and if you, if you accept that in many cases, organizations are driven by numbers. If you can translate these into that, Incredibly reductive single one dimension that helps you get, make it argue for something.

Then I think that's actually useful and we have patterns for doing this kind of stuff. So it's not necessarily the cost of compute. It's the cost of the opportunity cost of what you could be doing. Otherwise, for example, you can express these in numerical form and people make lots and lots of money building models and designing that and managing this way anyway.

So I think that you could apply this to carbon because there are absolutely. Consequences while the science spells out, there are consequences for us not pricing in carbon into how we work.

Asim Hussain: If you're working for a for-profit organization, it's it's money it's, you know, that, that's what that's the purpose of the organization is to, is to make a profit it's a for-profit organization, but there's other aspects of the world. You know, we have these free markets, but then we, hopefully we have regulations and regulatory authorities that work that temper and control the access is perhaps of, of a free market economy.

So we talked a lot about, you know, if, if there was just the money argument, Just the market driven money argument, you know, what, what, what could happen? What are some regulatory policies that if it, if they came out would help accelerate the adoption of not just green software, but you know, more sustainable tech solutions.

Anne Currie: Cast our mind back that we talked earlier about accessibility and, and you know, that that that's useful for generating additional customers and things. But actually, I, I, I remember at the time I was head of it for a company and the main argument for why we should make the site accessible. Was that the SEO was better.

Google used to give you an SEO boost if you were accessible. And that was one of the key performance indicators that they had, which was where are you in Google? That was a top line item for every C-suite business discussion. So you could easily say, well, we have to be accessible because it'll move us.

Through Google. And that was the reason why, you know, it was clearly measurable thing. And I think people really cared about, and I also used to work for companies like Microsoft and for them, the reason to go for accessibility was that there were government mandates on accessibility standards that a lot of applications had to pass.

So that was, well, you just cannot sell into the us government, which is a key customer that you want, if you don't meet these levels. So if we had. Something like that again, that would be effective, um, in the past that was effective and I think it would be effective again. So how do we get those kind of like both business?

So Google provided the push through the SEO and government provided the push through. You have to have these meet these standards in order to sell to us. And that was effective.

Chris Adams: like this is currently happening in Germany and in Hong Kong we're already. So I think last month, the first ever eco certified software was granted. Blauer angle or Blauer Engel certificate. And one of the reasons that people were doing that was because in government, which had many cases have legally binding targets to reduce emissions, they need to have a thing to ask for.

So now you see that and there is now a German project called software. Which is like soft aware, but with a kind of German accent, what that does, that's basically something a bit like continuous integration checking every single time. Are, are you making, are you moving forward or moving backwards in terms of the actual missions associated with any of the things you're doing?

So there are loads of efforts like this, but it's still early days. There's Germany and there's Hong Kong, but no one else has new certifications yet. But there's like a loose network of like 28 different countries who are trying to figure out what to be asking for so that when they put a big fat procurement tender in, they can say, I need this to be, I don't know, instead of perceivable, operable understandable and oh, Christ robust, maybe green, open, lean, decentralized gold, something like that.

We need to have things like that for it really.

Elise Zelechowski: And I'll just add, I think that's, I mean, that's right. The role of the Green Software Foundation, right. Is, is we have this forum to help drive and sort of bring kind of. It right to this question of like, what does good look like and what should we be really aiming for as a community? And our, I think our approach and our processes, you know, are based on transparent principles, right?

And, and really about let's get everyone at the table and really talk about this in an open forum. And so I think that regulation is key, but we need to sort of get to a place where we start to, to drive that alignment on, Hey, this is kind of a good stand and keep building. As you say, Chris, more sort of visibility awareness in the.

Asim Hussain: Yeah. Wonderful. So it's even when it comes to government influence, it's still, still going back to money, cuz it's like, where, where is the government gonna spend its money? But I think that will evolve to something hopefully a bit closer to policy regulation. And I would also say that, you know, Google, the Google SEO example, that was a Google policy.

It was a decision that they made not, I dunno, we probably had some, I don't know the background of that decision, so I'm just guessing, but it was this decision that they made, which drove a lot of action from a lot of organizations. And that's why these things are such great levers to, to pull because it's a small change can have a big impact as we're kind of reaching to the end of the, of the, of the panel.

We might have gone over a little bit. But I just wanted to ask each of you, you know, what advice do you have for someone looking to start having conversations regarding green software in inner roles, or even just in their communities? Let's start with you, Chris. Lloyd-Jones. Do you have any advice for people?

Chris Lloyd-Jones: me. It be about identifying that business case to support green software, but also understanding your culture. I mean, focusing again, gonna repeat it on that management example. I do think you need to know what your organization's motivated by and be quite honest about that. If you are a pragmatic organization cost or one of the other drivers, Chris mentioned will be important for some organizations, it being the right thing to do will be enough to start these conversations.

We need that organizational support to have that driver support adoption. You also need a sponsor. Cause without that, buy-in, it will be difficult to encourage teams to actually adopt these new practices. Go out for training. Like, like Anne was saying, if you're taking away from developer productivity, there has to be a really strong reason for that.

So that's the top down. I think the other thing I would say to start having conversations would be grassroots advocacy, meet the like-minded individuals. Those would be my two main.

Anne Currie: Great piece of advice, Chris Anne, what's your thoughts? Well, my thought would be, remember you have consumer power, almost everyone in the tech industry has consumer power to make change. Yes. So when you are buying, when you are buying stuff or. When you're thinking about buying staff, talk to your suppliers and say, I care about this, and this will, this is something that will make the difference between whether I buy from you or I buy from somebody else, because if you can then get them to go and change their products either immediately, or just because it's the feedback they're getting from their customers that will have outside impacts without you having to do anything at all, or get any internal sign off.

Just, just ask for it.

Asim Hussain: Right. Absolutely. Use your wallet, Chris Adams. What's your thoughts?

Chris Adams: I'm actually gonna agree with a lot of what Andrew said. There's a whole phenomena called the values perception gap. That's common in psychology, where everyone basically assumes that everyone. Doesn't care about anything until they ask them and realize that they do care about things. And like, until you do that and are explicit about that, or explicitly give a team permission to do this, or explicitly talk to someone, they'll basically assume that you do not care.

And if you don't do that, then. Well, we are kind of here because everyone is assuming that no one else cares and no one is prepared to ask, like, can you please do to our provider or Hey team, do you reckon people might wanna join us? If like we were, what some, we look more like the good guys, rather than the bad guys, there's all this stuff that we could be doing.

And I feel like a lot of time is basically be very explicit about this, about what you're after and what your values actually are lead with your values in many cases. But back 'em up with some numbers that you can come up to justify when someone is asking when someone is challenging, you, you later.

Asim Hussain: Yeah. And I will, I will just add to the, I do work for one of those large cloud providers and I will, I will. That the sales org, they do. It is flagged. It is put in a database when a customer reaches out and complains about anything. But like specifically there's a there's, you know, if, if the request or suggestion is around sustainability, it is flagged then is go.

It does go in a database, which is brought up in discussions, at least with you, like, what are your, what's your, what's your guidance?

Elise Zelechowski: Yeah, I, I would say all, all great points and all part of the kind of systems thinking that needs to be applied here of sort of activating right. Sort of different stakeholders and levers within an organization. You know, I can speak, you know, you know, specifically to, to ThoughtWorks, you know, one of the things that I have found, you know, we are a very decentralized culture, wherever agile culture, and it's.

So important for us to find champions in different parts of the business and then do the education and training and bring them along. And, and we've learned a lot from our participation already in the, in the foundation, but just getting enough people who are starting to spread the word, you know, sort of in a frankly, in a grassroots, right.

Sort of like pollinating different parts of the business. So you have. Suddenly you have this sort of like, oh, I'm hearing about this over here. Oh. And I'm hearing about this over there. And then as you're sort of building that business case that Chris is talking about, you know, and putting this together, you've already sort of primed the organization culturally and kind of gotten enough momentum building that, that it, I feel like you can just move then more quickly and get gain more momentum.

But definitely I, I feel like that grass. Piece has been very important at ThoughtWorks, even though we're a very purpose led company, it's just the, the way our organization works. You need to kind of do that pollination.

Asim Hussain: I would, I would absolutely agree with you. I think that we, one of the things people don't realize about Microsoft is it has one of the largest kind of green teams of any organization in the world. I think we're almost 6,000 people now in the organization kind of actively sit and participate and talk and, and, and make it very, very clear that sustainability is a value to them.

And. I think that's a strong signal to everybody else in an organization. So the middle management and the CIOs CTOs and the CIOs that, you know, they to make a decision around sustainability would be, it would be supported. So, um, yeah, absolutely. I agree with all of you, all of your opinions. Thank you.

Thank you. All of you. Thank you all for being part of this panel and thank you for being part of the foundation, your guidance and support. And your knowledge is instrumental in us achieving our mission and our goals. Thank you. Wonderful. Thanks so much. Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on apple podcasts, Spotify, Google podcasts, or wherever you get your podcasts. Please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we want more listeners to find out more about the Green Software Foundation.

Please visit greensoftware.foundation. Thanks again, and see you in the next episode.

The Price of Oil and Gas and Green Software

27 juni 2022 | 40 min

This week Asim Hussain and Chris Adams are joined by Sara Bergman of Microsoft and Henry Richardson of Wattime.org to discuss the current global surge in oil and gas prices. Why is the price of oil and gas so damn high at the moment? How does intermittency influence the price surge? Are high gas prices actually good for society as a whole? How does Green Software come into play? All this including a wrap-up of the Green Software Foundation Summit.

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Sara Bergman: LinkedIn / Twitter
Asim Hussain: LinkedIn / Twitter
Henry Richardson: LinkedIn / Website

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Green Software Foundation Summit
Event: RIPE 84
Organisation: The Rocky Mountain Institute
Article: From Deep Crisis, Profound Change by Kingsmill Bond, Amory Lovins, Oleksiy Tatarenko, Jules Kortenhorst & Sam Butler-Sloss.
Article: Understanding Energy Market Trends At The Layer Below the Internet Stack by the Green Software Foundation.
Podcast: The Climate Fix: Transitioning heating to hydrogen w/ Professor Averil Macdonald OBE
Resource: Carbon Aware SDK by the Green Software Foundation.
Book: Apocalypse Never by Michael Schellenberger

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Connect with us on Twitter, Github and LinkedIn!

Transcription below:
Chris Adams: We kind of do incentivize publicly traded companies like legally fiduciary responsibility wise to act a little bit like sociopaths. So we need to like hack that and do something about that. And like, this feels like maybe one way that you do need to resort to, but probably not the only one it's also useful to maybe talk about values.

Because, well, I know I respond to values more than most people, and there's lots of research, which says lead with values, but follow up with numbers, if you're gonna try and get any of these changes to actually be implemented or to see any change within an organization.

Asim Hussain: Hello, and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software.

I'm your host. Asim Hussain.

Chris Adams: So my name's Chris Adams, I'm the executive director of the green web foundation.

Asim Hussain: My name's Asim Hussain. I'm the executive director of the green software foundation and the green cloud advocacy lead at Microsoft.

Sara Bergman: My name is Sara Bergman. I am a software engineer at Microsoft. I'm the chair of the writer's project at the green software found.

Henry Richardson: I'm Henry Richardson, an analyst at WattTime and also the chair of the specifications working group at the green software foundation.

Chris Adams: Okay. So I suppose Asim, I figured we would spend the time today doing a brief recap from the GSF summit here. And I figured that would be a relatively easy thing to talk about because there have been like how many hours of programming now from various people over the last month has seen.

Asim Hussain: I haven't got the stats yet, but there's quite a lot. There's quite a lot. I mean, it's 27, no, actually 20, 27 regions and 28 events cuz one region had two events over two days. So yeah, a lot of, a lot of people have been connecting and sitting and chatting and talking about green software. , which is just really incredible.

Actually, I, you know, I, everybody was just working so, so hard to get the summit together, that to actually just sit there for a second and think that, wow, you know, people all around the world are now sitting down and having conversations about this topic. I don't know. It was really inspiring. One of my hopes for this podcast has always been like, just getting a bunch of people on and just talking about something that's happened recently that's of interest.

And I think all of us kind of can have opinions on, on this topic. So I don't know. I mean, other than the foundation, other than the summit, because I think that Henry only took part in some of it and, and Sarah. Got hers canceled and I had a fraternity, so I kind of just checked out for most of it as well.

so I didn't really like doing much. Is there another topic, like something happening in the news recently that, that

Chris Adams: shit. There's loads. So in Europe you've seen a whole shift. This you've seen a kinda landmark rolling on things like right to. Hardware efficiency in that basically the Europe has decided to standardize on USBC for any kind of devices or devices for charging now. Right. So that totally affects that part of the S uh, of the se.

All right, that's gonna have a meaningful impact impact. There also, you've seen a. Some similar model legislation come through on the U on the us level. Now that's been put forward to actually say, well, if they can do it, why can't we do it? So you may end up with the Brussels effect that was being referred to in the previous podcast about, well, okay, you set a decent default and then this ends up being something that ends up having a meaningful effect on the hardware.

Part of a bunch of the tools that we do in the end user device part. That's actually one thing that I think is worth actually exploring.

Asim Hussain: Yeah, I think, and I think GDPR has proven that, uh, something can happen from Europe and just, you know, affect the whole, the whole world, you know, like regulation from there. Cuz everybody has, is forced. Everybody's forced to kind of follow. Europe's such a huge market in the world, but if something happens in Europe, everybody else, everybody else has to follow suit.

so, is that the only is that the only one is that the, I remember that the USBC came along and I'm, and I'm quite happy cuz I'm an Android user. So everything I, and at Android, Macs, everything I have is, is the USBC dramatically works.

Chris Adams: works. Yeah.

Sara Bergman: Also, but it might be too political, but we could talk about cutting off Russian gas and how that leads an energy shift in Europe.

Asim Hussain: That's interesting.

Chris Adams: Yeah, there's absolutely some fascinating stuff along that because, well, this is what I did. I did a talk at ripe RIPE 84, which is basically a event it's ripe is the European entity that issues IP addresses to people. And one thing that we were speaking about was, okay, well, what are the drivers that you might want to use to kind of.

Be thinking about, be thoughtful about your, your source of energy. And if you look up over the last 10 years, basically the cost of renewables has fallen more or less tenfold. And particularly in Europe, I'm not sure in America quite so much, but because it's a global market and global commodity, you are seeing this like oil and gas is like 14 times more expensive now than it was 10 years ago.

Right. So you basically have this scenario where something is either a 10th of the price or 14 times the price. And if you've been able to. Negotiate say green supply here you are exposed. You're not exposed to like 14 fold increase in your energy costs. So there's like one thing which you're totally seeing.

And there is some fascinating research that's come out of the RMI, the Rocky mountain Institute, where they've spoken about how this war in the Ukraine has essentially sped up. A bunch of the existing predictions about how quickly you'll see various parts of the grid, decarbonize are on a global basis.

So yeah, there's a bunch of stuff there, which will absolutely have a knock on effect to any large consumers of electricity, like data centers, which are in terms of energy usage per square foot. They're more dense than aluminums melt. And this is before we switch to kind of liquid calling, which is gonna increase the amount of energy per, per rack to be even even higher, because you can get away with it.

When, whereas previously the limiting factor has been, is it possible to kind of cool this stuff down? If you have things like liquid calling, which could be maybe 10 to a hundred times more E. In terms of moving heat around, then that's gonna have a knock on effect in the energy density inside a given data center, for example, and therefore the marginal stuff, which

Asim Hussain: But going back to the point of the wall that's happening right now. So, so the Rocky Mountain Institute is, is this, cause I think there's, there's lots of mixed information out there. Cause I've heard, I've heard there's other, other things that have come along. So the RMI is saying that this is gonna help accelerate the energy transition, but I'm also reading things that, you know, in the hunt for energy independence for, for, for entities that they're going.

On previous commitments to roll out renewables. And they're just, they're just rolling out gas and, and coal faster.

Chris Adams: We might wanna speak to the expert here. Yeah.

Henry Richardson: There were two, two things. I, the union of concerned scientists just came out with a report saying that natural gas, power plants are running at a loss, a financial loss kind of like coal was four or five years ago. Um, and so there it's following, following the same patterns where. It's uneconomic to run coal or natural gas at power plants at various times.

So I think we're starting to see the same fossil trends emerge. It of course, leads to self dispatch and other. Out of market AER. And then we're also seeing, I think Biden instigated the, uh, defense production act for heat pump, hot water heaters. So I think there is a movement towards like accelerating some of the electrification fuel switching trends that everyone's been talking about for a long time, but we'll see if there's actual action on.

Asim Hussain: if, if gas prices are increasing and that's making gas, power plants, less economical. I mean, I'm, I'm seeing my, my, my electricity prices increase. Isn't just gonna, just kind of, aren't just gonna pass them on to me and therefore we're just gonna pay for more gas and, and that's it.

Henry Richardson: so renewable energy generators. There's a whole bunch of problems with supply chain issues, which are starting to get ironed out with the tariff reductions in the us at least. they are so financially viable because as the gas price of gas rises, the price of wholesale electricity also rises making it more attractive to build more wind farms and solar farms.

So there's like, anecdotally, we've heard that traditionally, a lot of renewables in Europe and the us are purchased through power purchase agreements where there's a guaranteed off taker through a corporate partner. Anecdotally, we're hearing that a lot of the developers want to delay the start of their PPA.

Because they're making so much money in the wholesale market in electricity right now, because they're able to produce power and make a significant profit, and they don't need the guaranteed price to put steel in the ground. So there's a lot of like, because it's pushing the price of power up that makes other technologies much more attractive.

We don't know how protractive this will be. There are a whole bunch of caveats there. Of course.

Chris Adams: is basically one of the thesis put forward by the report from the RMI it's. From deep crisis, profound change by the accidently named Kings mill bond. And AIE Levins along with Alex Tako, Jules Khost and Sam Butler, SL it's a surprisingly short, but interesting read in this, in this discussion, but I'm not sure.

Henry, I'm glad you're on the call. Cause it might be worth actually asking to what extent does this end up affecting things like margin intensity? Because this is one thing that I haven't been able to understand quite so much. Cuz what I have seen recently in our world is I've read a, a blog post recently about from electricity map, talking about the choice of module versus average and the fact that there's a whole ongoing bunfight about that stuff here.

Henry Richardson: The price elevations can do a lot of weird things

to the dispatch order. So for example, natural. Might have been roughly the same cost as coal. So they might have been intermixed in the dispatch stack the order in which they're dispatched a significant rise in price may make coal cheaper than natural gas, so that you'll run coal before you run natural gas.

So it could affect the total emissions changes, or even if coal's lower in the stack, low cost periods could be dirtier than high cost wholesale periods because you're bringing on. Coal will respond to low cost changes, and then you bring on peaking natural gas. And if you increase demand, then they'll just burn more natural gas and not more coal.

So some weird dynamics like that, you can all see, see inversions. So the price effects are very, can be kind of dynamic and update in real time.

Sara Bergman: I think this is super interesting and I'm just using this to like, learn more, but coming from a, or currently living in a nation that sort of made our wealth from oil and gas, uh, because way for, and we have vast amount of resources that the government luckily decided to not. Dig to because they are destroying coastal areas, cetera, but what's to stop countries from just, oh, gas is lucrative.

Let's just bring up more.

Chris Adams: So it might be useful for context, for people who knew were new to this. Cuz we have a few kind of energy walks on this. There's a helpful blog post on the green software foundation website called understanding energy trends at the layer below the tech stack, where we talk a little bit about this. And if you are, you've never heard of what a dispatch stack is or any of this is there's a little bit there which might elucidate and provide a bit more background on this. So we can, if, if, if you're totally new to this.

Asim Hussain: Because this whole world is so complicated. We've got on one side there's right. There's rising energy prices. And on one level that is going to help accelerate the transition towards renewables because then, well, if you, if you. If you've got a renewables, your crane, electric renewables, you just can make more money from selling your renew energy.

But on the other hand, they're just, you're gotta make a hell of a lot more money by drilling oil and gas. So what I'm trying to get in my head is, is it better? You are saying the RMI saying they've done the maths. And they said, actually raising rising prices is better, period. Or are they saying that rising prices will just help the transition?

Chris Adams: Okay. So here's one of the issues is that over the last, say 10 years in Henry, you can probably back me up on this in a bit more detail. And what you have seen in America in particular is you've seen this kind of glut of kind of cheap gas coming onto the market. Right. But a lot of that has happened.

That a loss. So you had all these people who are like drilling, drilling, drilling, drilling, drilling, but they weren't necessarily turning a profit. They were just like going for like a little bit, like how, at our level, we incentivized to go for growth and worry about revenue or profit later. You've had the same thing happen for a long time.

But in the last few years you've seen in particularly the last like year and a half, I suppose you've seen a, you've basically seen all the people who actually do have been doing this. They said, oh, we need to kind of make a return to our investors. So rather than actually. Bringing more gas on stream, for example, to bring down the cost.

Every single large provider has basically said, no, we're not gonna do anything about supply. We're just gonna make sure we are gonna give a return to our shareholders. We're gonna exercise kind of shareholder. Discipline is the term that people have been using a lot. And this basically is why we have loads and loads of why you have such high prices right now.

So yes, you've had demand in lots of places go up that you haven't seen the corresponding increase in supply, even though there's lots. People there's lots of basically operators who have been able to do this and do have permits for drilling. They've said, well, we're gonna keep the price high and actually make sure that we can get return to our shoulders rather than increase the, uh, supply and therefore the cost, which would probably be good for like people who want to.

Buy cheap electricity or cheap, cheap fuel. Probably not so good for the climate, but definitely not so good for shareholders. Who've been quite annoyed about them not being able to make a return over the last say, say five to 10 years while there's been this mania to just expand, expand, expand, and drill, drill, drill.

But Henry, I'm happy to defer to you because you are in Houston and I'm not, and I'm in Berlin.

Henry Richardson: This is not something that I have a ton of expertise in, but if you think about you are seeing price increases across the board in, in like your natural gas delivered to your house for consumption, electricity costs, fuel costs. So like all prices are rising. So it's. Pressure on all forces. So for an electricity transition, going from natural gas, heating to electricity, heating there's, the forces are kind of equaling out.

That's like neither one is any more attractive than it was before, but we're seeing renewables as an enormously valuable hedge against the cost of power increases. So the fuel costs for renewables have not increased the fuel costs for everything else have. So there's still a huge benefit. To, to power systems that have already decarbonized to some degree.

Chris Adams: Also come from fuel and you don't have to pay for fuel with renewables. Then you kind of got quite a pretty good advantage, right? Like this is the, this is the key thing. And as long as you have a way to address the intermittency issues or the variability, and there are ways around that, then you can do something to basically do better financially than, than you otherwise would be.

Henry Richardson: a really interesting point about the intermittency and it's something, I guess the green software foundation actually talks about pretty frequently. Awesome. You coined the like carbon. Air, which means both time shifting and location shifting. And while we, we don't really talk about it very often. So we talking about software itself and like where it runs and when it runs.

But the things that software controls like heat pump, hot water heaters, EVs. Smart thermostats. And you have to build really sophisticated algorithms to take advantage of this. So it's still very much so software, those will be key in managing intermittency down the road. So if we have a ton of EVs, that's an enormous amount of storage that can be scheduled and managed.

If you think of a heat pump, hot water heater tank as kind of a thermal battery, if you either oversize that tank. Or include a mixing valve and overheat that tank. You can actually consider it a, a thermal storage. And I think those are two of the largest energy consumers in a household that have flexibility, but even generally two of the largest.

Sara Bergman: cuz I think that's a key difference from like when we think of data centers or software. Where that software requires. We need like electricity. That's what we need. But if you take the homeowner homeowner perspective, you don't like electricity, you need, but you also need heating, especially in, in cold places.

And if you're in more places, you need cooling. So you need not only electricity, but other things and, and oil and gas have this benefit, right. That you can use them for both. You can use electricity for heating. It's actually quite common where I live, but it. Much more expensive with these prices right now, and much less efficient use of it as well.

So that duality, I think, is quite interesting.

Asim Hussain: This might be a bit of a tangent. One of the, in I, in a previous podcast I used to have, I interviewed, I do. I'm gonna put in the show notes. I can't remember her name, but she was talking interview someone about the use of hydrogen as a replacement. For, for, for natural gas, for heating and actually most countries around the world, that's actually not possible.

The UK just happened to be one of the countries where it is possible because we have like essentially a grid for natural gas. Whereas a lot of other countries don't have a grid for natural, for natural gas. And apparently it just requires, you know, a 10 minute adjustment to a, a, a boiler to get it, to burn hydrogen instead of sort of natural.

I remember the other fun, fun fact. She told me, which was that hydrogen burns colorless. So they actually have to add color to it so that it, so, and I was like, well, you can have, we can have whatever color you want. You can have pink flames on your, on your gas, boiling your boiling, your kettle instead of, instead of blue ones.

So I think the question I'm asking myself in my head is, is, are rising prices for. energy. Is it better? And that's kind of like, I think one of the, the questions that's coming in, is it really better? And one of the,

Chris Adams: Is it better?

Asim Hussain: yeah, well, that's the thing. Is it better for what, what is it better for? I mean, it's obviously it's really, really hard for a lot of people in the world.

I mean, right now, I mean, right now in the UK, it's quite good. It's summer, you know, we're not gonna be hit with big heating bills when it comes to winter, it's gonna be really, really challenging. But the same level when the price of something is high, it means you're focused on efficiency. Is much, much more important things like intermittency is much, much more important.

If electricity was 10 times more expensive than what it is right now, how would that change our landscape and the conversations that we're having with people? How would data centers work? What would be the, the focus on green software is the problem that we have right now that electricity is too cheap and we're just wasting.

Chris Adams: This is my free idea and the thing I wish I had time to add a pool request for cloud carbon footprint, right? Cost of energy is absolutely a thing for like equity and people having access to like, you know, warmth and heat and, and cool. If you want, when you're in someone like say Texas, cause I suspect Henry you probably care about being cold rather than too hot right now.

Yeah. Like if, you know, if we might be okay in a, in a relatively separate place, but. It's like 54 degrees in some parts of the world right now, like Iran, for example. Right? So the dream feature for something like cloud carbon footprint or any of these tools would be allowing organizations to price this at the level of their organization really would price it.

So right now, if you were able to do something like, well, we value the cost of carbon emissions. That may be a hundred tons. And then you factor that into the price. When you saw a dashboard, this is how you can end up having the same ways to track this and measure this stuff within, inside a team. Because if you're gonna have like any kind of tools for track for in the same way that you might track say compute bills on a weekly basis to see, have you seen in a weird spikes, if you have some way to say, show me the full, the actual true cost that's who that's.

Push that, that that's, that's actually transferred to everyone else outside of the binder of, of our organization. I think there are ways that you can do this because every other organization or every other sector, or most other sectors are currently experimenting with things like internal carbon pricing shipping does this now aviation there's stuff around this and various organizations like say, well, Microsoft does it, you know, 15, 15 bucks per ton, a hundred bucks for trans for, for aviation, but like groups, like say Lloyds of London.

They have figures of like 250 pounds per ton for this stuff. And when you look at it like this, when you realize that it's actually the CO2 emissions, rather than the energy, that's the thing that we are really trying to reduce in the context of climate change. It makes sense to price it accordingly. So you have the right signals inside teams who are able to act upon it.

Really. I feel like that's a better way to think about this and like there's loads of precedence of us pricing, other things we value. So why would, why wouldn't we do it with carbon and, and expose it to our own metrics inside our teams to act.

Sara Bergman: Yeah, that really strikes a chord with me. I think. Actionability on. This is key, cuz when you are in a software team, whatever role you have, if you can't measure it, if you can't track it, if no one knows about it, it's incredibly hard to put it on the agenda to say that, yes, we should care the most about this.

And people are like, okay, great. How will we know we're doing better? And you say it's a gut feeling. It's not gonna fly with stakeholders. It's not gonna fly with management. It's not gonna maybe fly with the rest of your teammates either. So having this actionability, and we've spoken a lot about the, the energy grid now, and it's like super fascinated for people like us who are deeply interested in it.

Not everyone needs to be this deeply interested in it to act on climate, right? You don't need it if you want to. Great, but we wanna make it simpler and more actionable for people with like hands on desks in code rather than. Just the geeks, I guess I'm proudly a geek. So I'm saying that with all, all my love.

Henry Richardson: And I think we're seeing a lot of efforts towards that at the green software foundation, especially the SDK will hopefully make it much more accessible. So you. Have to think about the, the carbon intensity of the grid. You don't have to think about like, what tools do I need or the carbon intensity of different things.

It'll just kind of be built into a, an accessible tool. Awesome. Do you have an update on that? You were about to mention something.

Asim Hussain: You just said the SDK, I just gave the full name carbon or SDK. And I do know the team is, is working on an release in July. So we was keep an eye out for that. Cause I remember there was a report or a paper out a couple of years ago about, was it called the social cost of carbon? Cause you mentioned $15 for Microsoft.

It used to be seven. They just doubled it one day. Because you can, because it's so arbitrary, you can just double it and whatever, but like the social cost of carbon was that calculation, which is you actually put a dollar term on it. What is the impact on the environment, but also what the impact on my children and, and, and down the line, I think it was like $300 a ton.

If I remember rightly it was around that, that, that price

Chris Adams: So it's changed depending on who's been in power. All right. So with the Barack Obama administration, it was $50 a ton under Drumpf. It went down to $1 per to.

Asim Hussain: Oh,

Chris Adams: It's now gone higher up. I forget. I think it's actually close to 50, again, we're under the Biden administration, but what you might be thinking about assume is the UK.

I know in the UK they have a figure. They don't actually have, they, they use something called a carbon value, which is another price for this. And their figure is one that climbs over time. And basically they use it as a kind of way of checking the impact of the policies. So in 2022, the cost that they use is.

Yeah, they have a, they have ranged from 124 to 373 pounds, which is about what 500 bucks, 500 us dollars roundabout per ton, but that ratchets up. So every year it climbs up to the point where in 2050, you're looking at a hundred between 189 and 560 odd pound, which is close to eight or 900 us dollars. I think.

I mean, I'll let someone else kind of cut. Help me with the actual specifics, but yeah, they use that to basically make decisions about policy. So they don't charge people the way that say Microsoft does. So it's not like a heavy price in that sense, but is definitely used for framings. And this is actually some work that we're doing.

And we shared inside the group, the GSF slack about ways to think about carbon pricing. There's there's like high, you can think in terms of the height, like how, how high is the number? How broad is it as in. How much of your organization organizations operations, does it, does it impact? So a broad one might be 15 across the board, whereas a narrow one might be, say a hundred dollars, but only for aviation, like the Microsoft example.

And then you might think in terms of like weight, which is, does this. Impact me and my team or my organization right now, like with Microsoft, where the money is taken out of someone's budget and put into like this kind of carbon war chest for savings compared to a kind of shadow price that you might see, let's say shell uses or other oil companies where basically say we are gonna use this to factor in an investment decision, or maybe with the UK government where they use it as saying.

They're not taking money out of any purses, but they're using it to decide whether they should or should not go ahead with a particular policy decision because they've decided, and they're using this number to basically account for all the kind of other impacts that basically carbon mission can have because they have long reaching impacts in terms of health and well, the economy and everything like that.

Asim Hussain: can you buy anything in this world? When can you buy a ton of anything in this. for a couple of hundred dollars. That's why I'm just a tons, a large amount of something. So is that price? That price just, just feels too low. I mean, I, I can't, it doesn't feel right. I mean, does it really give you the incentivization to make those kinds of changes?

I'm not too sure.

Henry Richardson: This, this might go back to Sarah's point around like there's a, there's a, there's a significant difference between not considering carbon at all. And thinking about carbon in the optimization, that is a much larger leap than deciding what price to put on carbon, because it becomes a decision making factor versus one, not at all.

And I think that is a while we're talking about like, should it be 300? Should it be 50? I think the idea of just including carbon at all is a much more radical concept than, than picking the correct cost or the correct impact.

Asim Hussain: It's called a Maserati problem. I think it's what you usually called the Maserati problem. It's the startup space. I think it a long time ago in the startup space, like it's like a Maserati problem is, is a, is a problem you'll only have once you're driving Maseratis, cuz your startups made so much money.

there's not really much point worrying about it right now, but yeah. Hear your point. Yeah. So it's yeah. Hmm.

Sara Bergman: And that that's the flip side of it, right? Because if you are a small organization, then this number might matter a lot. If you're really, if you're a government, like this number becomes arbitrary very quickly, even when you price it super high, it's. Dropping the ocean compared to, to other things. And you maybe don't want it to be in direct competition with other things like basic healthcare, et cetera.

Chris Adams: Is basically I'm, I'm mindful that we getting way away from green software with this kind of tangent, but this is the idea, the reason the purpose of the carbon value, one of the UK is designed to kind of capture some of this stuff. There's a guy called, uh, John Cooey who talks about this. And he basically says the.

Absent other things, pricing stuff is one way to show what you're valuing things. And that doesn't mean that you wanna have this as your only kind of somewhat reductive lever for this. But if you, I think you can make an assumption that if you price something at $1, a ton versus $50 a ton, there is like a, an implicit value judgment there about, we think that the impacts don't really matter.

Versus we do think that the impacts actually matter here. So I think it's, it's, it's. I mean absent any other tools. This is one thing that may be more useful, and this is kind of what, Hey, I've seen people repeatedly use as a way as a mechanism to essentially help inform a decision. Cause if you're in, if, if you are in an environment where this is one of the only levers you actually have, and if you're say inside a organization, which is primarily driven around numbers, sometimes you'll need to speak a somewhat reductive language, even though it doesn't encompass the full gamut of human experience, just because.

Like we kind of do incentivize publicly traded companies, like legally fiduciary responsibility wise to act a little bit like sociopaths. So we need to like hack that and do something about that. And like, this feels like maybe one way that you do need to resort to, but probably not the only one it's also useful to maybe talk about values because well, I know, I respond to values more than most people, and there's lots of research, which says lead with values, but follow up with numbers.

If you're gonna try and get any of these changes to actually be implemented or to see any change within an organization.

Sara Bergman: And I think that's interesting if we think about the, like a global market where we actually price carbon really high and what that means for software, because we know data centers are big, big consumers of energy, which produces a. Of carbon. So if that means we're gonna see a rush of veiled and expansion in countries with hatch, a more renewable energy reducement yeah.

Like for example, what does that mean to the local grid here? Because typically they are kind of scaled for the usage that you have now. Cause the grid always needs to be balanced. How fast will that shift go and will the grids keep up or will there be a need to import other electricity? And how will that affect the price of both building new data centers, but also the people who live in that region.

And I think it's one, it can be a really good thing, right? Cause it syntheses the increased use of, uh, renew low carbon energy sources. But if this happens too fast, it can have the opposite effect for those specific regions.

Henry Richardson: Sure. We definitely see that where like a large increase in demand is tends to be backfilled with fossil because that's what's available and it is dispatchable. And then only over time do renewables start to get built to fill that gap. So I think that's a key point is that's that kind of gets back to the marginal question of when you make a change, what.

What fills that change and it tends to be fossil, unless there's already an oversupply of like, for example, in California, there's a huge oversupply of solar most of the year in the middle of the day. But the rest of the time you add demand and it'll be met with a fossil resource. I don't know exactly what the answer for Norway is.

You might know though.

Sara Bergman: We have only hydro basically, because we are a very mountainous, rainy country. So hydro is perfect. But with the cold winters that we had that would likely continue to have with climate change. A lot of that water is frozen and it's hard to make use of it. Also, it rains less or snows less in the late fall, which means there is less water.

So we're seeing more and more like local incentive wise people adding solar to the roofs, et cetera, because energy prices are going up and, and with climate change, actually, they may continue to do so, even though the energy source itself is renew.

Henry Richardson: And I think Norway exports a lot of its power to, I don't know if you call it mainland Europe, but so the more power that is consumed in Norway means there is less hydro to export to the rest of Europe. And that means that power that is not exported to Germany, for example, will now be fed by coal. So there's like, there are.

Global implications of local energy consumption or data center development. So understanding how those dynamics can be super difficult and.

Asim Hussain: That conversation's quite interesting. Cause that's making me think, well, the transition has to happen slow. Are we saying the transition has, has to happen slowly?

Chris Adams: No, it doesn't, it's not that it has has to happen slowly, but in the short term, if you don't have something to fill it in, then you may end up with something filling in that gap as Henry suggests, simply because even if you have loads of batteries and things ready right there, it's gonna take a bit of time to get planning, permission, getting all their stuff, getting them actually deployed into space until you have that people might be resorting to something else in the meantime.

And that means what one thing you need to have now is. Ways to either streamline that or make it ideally streamline that in a way, which does support some of the issues around equity so that you don't end up having a backlash against this or against people looking to do things. So, so for example, in Europe, right now, we have seen.

Like this new plan repower EU, where there's basically a plan in the next two and a half years to deliver three Germany's worth of solar in two and a half years. Now, even if we could make all that, that's like incredible amount of solar, but it's nowhere near as ambitious as China is doing. China's going way, way, way faster than this.

Right. But the thing that, that. Yes. Like there's a whole thing about some money being set aside for that. I think it's in the region of 300 billion euros has been kind of earmarked for this, but the thing that's probably gonna have, the bigger, biggest impact is changing of the laws. So it becomes faster and easy to get this stuff deployed.

Cuz a lot of the time you have this backlog of things of projects being able to get set up, usually because it takes a, a good few years and if you've set yourself two and a half years, then you'll need to do something to make those changes. The worry is if you don't, if you just kind of make it easy to steam, roll this stuff through, then you'll end it with a backlash a little bit later on, which will make it hard to get the rest of the way.

So this is one of the, I guess, challenges that we currently have at the moment, but. There are kind of ways around this. You can, you know, share the rewards with people in a more equitable fashion. And when you have seen that happen, then people are more prepared to deal with to, to accept say wind turbines and solar farms and stuff like that.

If they get to, if, if the, if the relationship isn't quite so extractive with the existing local communities, for example,

Sara Bergman: I, I think it's interesting here as well. Uh, Europe has kind of a complicated history with nuclear, but nuclear is a low carbon, uh, renewable energy source and, and that's also something. I think if I get to predict something, it will be a, a mitigation approach until we have that renewable capacity that we need, because we already have the power plants.

And I know Chris, you can probably speak more about this, but I know in Germany that there's been talks for a long time to shut down all the nuclear plant power plants, but, but the current energy crisis, those are not really going forward as planned anymore because you can't spare that energy that they can produce, which I think is interesting.

Cuz a lot of time we conflate. Environment and climate, they are not necessarily the same thing, at least in, in my language that I speak there, they don't mean the same thing because nuclear is not great for your local environment, cuz you need a final storage, but for climate, they are great. It's very low carbon.

Chris Adams: So there's some stuff I can share for this specifically in the German context. So we've had this whole energy vendor for a very long time and you are absolutely, you can absolutely make the argument that yes, having very useful, very, very low carbon or, you know, Basically carbon free energy available for a very long time is a really good idea.

The thing that it's worth bearing in mind is that in Germany specifically, it's not necessarily the energy, that's the issue. It's the heating, that's the issue, which is why you've seen there's such dependency on like fossil gas and stuff like that for heating. So heating is the problem more than electricity right now, or has been for most recently?

So it's not just as easy just saying, well, if only they'd switched on the new, it kept the nukes on it. It'd be fine. I. They absolutely are really, really helpful for that, but that's, it's not as simple as just saying, oh, if we had that, then that would, that would've solved all the problems. They still wouldn't really have solved some of the heat issues that we do have because Germany can be a cold temp, cold country at times.

Asim Hussain: The advice and the plan for the whole world is to electrify. So I think, you know, I hear your point, you know, and same in the UK, like I've actually got gas heating at home. It would cost me an insane amount of money to heat my house to electricity right now. But I kind of stayed away from it for a while, but I finally read the book from Michael Shellenberger - The Apocalypse Never. Yeah. As Chris, Chris. Yeah. As the eye rolls, I've been going through like a process recently of, of kind of reading stuff that people told me never to read, but he, he made a really solid argument for nuclear. I find it very, very hard to kind of question a lot of that. Because, you know, the amount of energy that we need in the world is, is phenomenal.

And, and especially the new, the new nuclear power plants, the new, the new stuff's being designed and deployed are, are, you know, it's not the same as the old ones. And just to kind of answer a point that was made that I, I, it really resonated in my mind with nuclear, was that the, the, all the nuclear waste from the entirety of the United States from history to.

Could could fill an Olympic size swimming pool and it's easy to manage and you just keep it there. Whereas coal and gas, it's just all goes in the atmosphere. So there there's this waste, there's this, this concept of waste and nuclear, and you really hate it. But then, and it's horrible waste, but at least it's self-contained whereas this other stuff is all out there.

So that was an argument I thought was really, really compelling.

Henry Richardson: I think everyone has their pet technologies and they're like, we need to build nuclear. We need to build geothermal. And like,

Chris Adams: Yeah,

Henry Richardson: yes, we need all of it. Like, we're gonna need nuclear. We're gonna need hydro. We're gonna need geothermal. Like we just have an energy shortage or a non fossil energy shortage.

Is that a better way of phrasing that?

Asim Hussain: non-fossil energy yeah.

Henry Richardson: deficit gap? However you wanna describe that. So I think. I, if, if nuclear is successful affordable, it would probably one of the solutions.

Chris Adams: If it arrives in time. That's the thing, assuming it arrives in time. Cause the first predicted thing to go on at a common stream in America, 2030, right? So that's, this is one of the issues, so much needs to happen this decade, that when we only talk about this thing, which at earliest could happen there, this basically we can fall into a discussion about, oh, it's just delay.

Let's just wait. Let's let's not make too many changes or be too fast. Like we really don't have that time right now. Like we're really, really late in the day, but we can also not forget the first fuel efficiency.

Asim Hussain: true. And, and by 23, we'll have cold fusion reactors. So don't worry about. I’m joking. That's a joke.

Should we do a wrap up?

Chris Adams: Okay. So what we've covered in this discussion is dived into all the trends happening at one there below. Most of us tend to work in the stack and there are all kinds of geopolitical implications of what we have of what what's going on. And they do affect not only people who are. Dealing with it day to day, but these also ripple out across the rest of the world.

And as a responsible software engineer, knowing about this makes it easier for you to prepare for turbulent times ahead.

Henry Richardson: Another way, phrase it, the war is happening and it's having wild effects on energy prices. And we're like, there's some, there's some thought out there that these increased prices could accelerate decarbonization of different sectors. Which might help mitigate climate change effects - full stop.

Asim Hussain: That works.

Sara Bergman: That was good!

Asim Hussain: That was good.

Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we want more listeners to find out more about the Green Software Foundation. Please visit greensoftware.foundation . Thanks again, and see you in the next show.

How does AI and ML Impact Climate Change?

13 juni 2022 | 40 min

This week Chris Adams takes over the reins from Asim Hussain to discuss how does artificial intelligence and machine learning impact climate change. He is joined by Will Buchanan of Azure ML (Microsoft), Abhishek Gupta; the chair of the Standards Working Group for the Green Software Foundation and Lynn Kaack; assistant professor at the Hertie School in Berlin. They discuss boundaries, Jevon’s paradox, the EU AI Act, inferencing and supply us with a plethora of materials regarding ML and AI and the climate!

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Will Buchanan: LinkedIn
Abhishek Gupta: LinkedIn
Lynn Kaack: LinkedIn / Latest Paper

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Green Software Foundation Summit
Project: Pachama
Project: This Climate Does Not Exist
Legislation: The EU AI Act
Paper: The Green AI Movement
GitHub: Dynamic Batch Inferencing - Taylor Prewitt & Ji Hoon Kang of UW
GitHub: NVIDIA Triton server on AzureML & Model Analyzer

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Connect with us on Twitter, Github and LinkedIn!

Transcript Below:

Abhishek Gupta: We're not just doing all of this accounting to produce reports and to, you know, spill ink, but it's to concretely drive change in behavior. And this was coming from folks who are a part of the standards working group, including will and myself who are practitioners who are itching to get something that helps this change.

Our behavior change our team's behaviors when it comes to building greener software.

I'm your host. Assim Hussain.

Chris Adams: Hello there and welcome to the Environment Variables podcast. The podcast about green software. I'm Chris Adams filling in for Asim Hussain, the regular host while he's on paternity leave, with a brand new baby! I met Asim on climateaction.tech, an online community for climate aware, techies. And I work for the green web foundation where we work towards a fossil free internet by 2030, as well as working as the co-chair for the green software foundations policy group.

Today, we're talking about climate change, AI and green software, and I'm joined by Lynn, Will and Abhishek.

Will Buchanan: Thanks for having me. My name is will. I'm a product manager on the Azure machine learning team. I'm also a member of the green software foundation standards and innovation working group within Microsoft. I foster the green AI community, which now has a few hundred members and I'm also a climate activist.

That's focused on pragmatic solutions to complex environmental issues. Recently I shipped energy consumption metrics within Azure machine learning, and we are about to publish a paper titled measuring carbon intensity of AI in cloud instances, which I think we'll touch on today.

Abhishek Gupta: Well, thanks for having me. I'm Abhishek Gupta, I'm the founder and principal researcher at the Montreal ethics Institute. I also work as a senior responsible AI leader and expert at the Boston consulting group, BCG and I serve as the chair for the standards working group at the Green Software Foundation. So I've got a few hats on there.

Most of my work. As it relates to what we're gonna talk about. Runs at the intersection of responsible AI and green software in particular what's offer interest to me is looking at how the intersections of social responsibility, the environmental impacts of software systems in particular AI systems.

Can be thought about when we're looking to make a positive impact on the world while using technology in a responsible fashion. I also, as a part of the green software foundation, help through the standards working group, come up with the software carbon intensity specification, where we are trying to create an actionable way for developers and consumers or software systems to better assess and, and mitigate the environmental impacts of their work.

Chris Adams: Okay. And Lynn last but not least joining us from Berlin. Thank you very much for joining us.

Lynn Kaack: Yeah, thank you so much. I am an assistant professor at a policy school, public policy school called Hertie School in Berlin. And I am also a co-founder and a chair of an organization called climate change. AI. And with climate change AI, we facilitate work at the intersection of machine learning and different kinds of climate domains, focusing on climate change, mitigation and adaptation.

And in my work, in my research, I am looking at how we can use machine learning as a tool to address different problems related to energy and climate policy. And I'm also interested in the policy of AI and climate. And today, actually, since we're talking about papers, I have a paper coming up that is called aligning artificial intelligence with climate change mitigation, where we look at the different impacts from machine learning and how they affect greenhouse gas emissions.

Chris Adams: Awesome. So we actually have some decent deep domain expertise and I'll try to keep this quite. But we might drop into a little bit of like data science nerd around here, but the podcast has done that previously and it turns out to be something that we've got some decent feedback from because there aren't that many podcasts covering.

Okay. So let's, uh, get into this topic of green AI and climate change. As we know, it is a significant driver of emissions in its own, right? When we think about other climate crisis this year, the IPCC, which is the intergovernmental panel on climate change in their big reports, which synthesized literally thousands of papers explicitly called out digital as a thing we, we should be talking about and thinking about, and if you're a responsible technologist, it seems like a thing that we should be taking into account here.

Now I found it helpful to think about it, uh, a little bit like how we think about the shipping industry, partly because there is similar in terms of emissions, which is around between one and 3%, depending what you look at it. But also in that both of these acts like kind of connective tissue for society.

We also think of it as a kind of force multiplier for the existing forms of activity. So if you use it, which is in line with the recommendations of the science, that's a good thing. But if you use. To do something which is kind of rejecting some of the sites. It might not be such a good thing. And within technology, AI and machine learning particular is one of the fastest growing sectors and often seen as one of the biggest levers of all.

So we're gonna highlight some interesting projects we'll start off with, and outta that, we'll probably dive into some specifics about that or some of the things you might wanna take into account. If you're a technologist wanting to. Incorporate an awareness of climate into how you work and build greener software.

Then finally we'll hopefully leave you with some actionable tips and techniques or projects that you may contribute to or use in your daily practice. There's another term that we might be touching on here when you're making AI greener and that's specifically green AI. Is that the case? Well,

Will Buchanan: Correct. And that actually was coined by, uh, researchers a few years ago. Uh, Roy Schwartz, Jesse Dodge, uh, and it's really focused on making the development of the AI system itself more sustainable and it's to be dis Abid for. On the term using AI for sustainability.

Chris Adams: Okay. So that's something we'll touch on both today. We'll talk about some of the external impacts and some of the internal impacts. We're gonna start with something quite easy first, because, well, why not? I'm just gonna ask each of the people here to. Kind of point to maybe one project that they've seen, that's using ML in quite an interesting fashion to ideally come up with some kind of measurable win.

Well, if there was one project you'd actually look to that you think is kind of embodying these ideas of like green AI, AI or something, which is really helping us essentially face some of the challenges. Maybe you could tell us about what's catching your right at the moment. What you'd look.

Will Buchanan: been thinking a lot about natural capital recently. And I learned about a startup called Pachama, which combines remote sensing data with machine learning to help measure. Monitor the carbon stored in a forest. I think it's really, really valuable because they're providing verification and insurance of carbon credits at scale.

And they've protected about a million hectares of forest. I think that's really when you have IT and remote sensing and machine learning combining to help nature restore itself.

Chris Adams: Okay. Cool. So if I understand that they're using satellites to basically track forests and sit and track deforestation, is that the idea that they're doing?

Will Buchanan: Yes. And also to verify the amount of carbon that a forest can sequester.

Chris Adams: Okay. Cool. All right. I know there's a few other projects related to this. If I just hand over to Abhishek, can you let us know. What's caught your eyes recently and then we'll see what other projects come out of this.

Abhishek Gupta: absolutely. I think one of the projects, and I don't know. I mean, if it, what the. Impact has been so far. In fact, it's. It's something that's come out of MILA, which is, or, you know, called the Montreal Institute for learning algorithms, which is Dr. Benji's lab in, in Montreal. In fact, one of the people who led that project as a part of climate change, AI as well, who I'm sure Lynn can talk more about too, which is SASA.

And she's done this project called this climate does not exist, which I think was a fascinating use of machine learning to visualize the impact climate change will have on. You know, places around you in, in a very arresting and, and visually capturing fashion, which I think when we think about what impact climate change is going to have around us, sometimes it, it feels quite distant because it's a, it's a slow rolling thing that's coming our way.

And this. Puts it in, in, in a way that's quite immediate, quite visually arresting. And I think stores people to action. I, as I said, I'm, I'm not sure what the measurable impact of that has been yet, but I, I certainly feel that those are the kinds of creative users of AI. When we want to galvanize people into action around climate change.

Lynn Kaack: happy to also talk about an application, which is also kind of difficult in terms of measuring impact, but I think it's. Another interesting component of what AI can do. And this is something that at the Austin Institute of Technology do on a project called Infrared and they use machine learning to help design new districts and cities.

And especially at the moment in many countries, a lot of. New urban districts are being built and how we build these has a huge impact on energy consumption in cities, both in terms of transportation, but also how buildings are heated or cooled. And by the use of machine learning, they can drastically improve design choices because now they can approximate their very computationally heavy models and run them much faster, which means that they can also have more runs and can try out more design configurations.

So this is. Rather indirect application, but it has huge implications also on emissions for, for many decades to.

Chris Adams: Using kind of housing policy as climate policy there, because there's just a huge amount of emissions built into how people live and whether they need to drive everywhere in a car and stuff like that. That's, that's some of the stuff that it's doing and making that part.

Lynn Kaack: So it's not really looking at housing policy, but it's looking at how districts are designed. So they take. The group of, of houses, like if the, a new district is to be built and then they simulate the wind flow going through these cities, which are very expensive simulation models. And then they take the outputs of their model and approximate it with the machine learning model, which makes it much, much faster.

So from hours or days, you go to milliseconds or below seconds for one run, and then you can try out different design configurations and understand better how. The build infrastructure affects natural cooling. For example, in cities or walkability energy impacts generally of the micro climate on, on the build environment.

Chris Adams: I had no idea that. It was actually possible. That's really, really cool.

Will Buchanan: That's very cool. That's similar to generative design.

Chris Adams: Generative design. This is the phrase I haven't heard actually will. Maybe you could elucidate or share something there actually.

Will Buchanan: It's similar to some software that Autodesk has built, where you can try out many different iterations of a design and come up with optimum solutions. I think what's really cool that you're just consolidating it and running these models more efficiently.

Chris Adams: Okay, cool. And that's a bit like following, say a fitness function saying I've have a chair or, you know, I wanna have something works like a chair and needs four legs and a seating pattern. And then it essentially comes up with some of the designs or iterates through some of the PO possibilities, something like that.

Will Buchanan: Exactly.

Chris Adams: Oh, wow. Okay. That's cool. All right then. So we've spoken about AI and there's a few exciting, interesting ones that we can add into the show notes and list from, and for people to look into and see how that might relate to what they do. I suppose I wanted to ask a little bit about measuring impact from these projects because.

There's quite a few different ways that you can actually measure impact here. And in many times it can be quite a difficult thing to kind of pin down. And this is continually thing that's come up. When I know that people have tried to come up with specs like the software carbon intensity, and I'm Sureek, you've had some experiences here will, you've mentioned a little bit about.

Actually measuring impact internally. And it sounds like you've just had to do a bunch of this work on the ML team right now and exposes some of these numbers to people consuming these services in the first place. Could you talk about some of that part a bit, perhaps?

Will Buchanan: Certainly. And so, as I mentioned, we have shipped energy consumption metrics for both training and inference within Azure machine learning. And that's really complex when you think of the infrastructure required to just report that. That doesn't necessarily account for the additional power that's consumed in the data center, such as the idle power for devices or for the utilization of your servers.

So there's so many different factors there. So you really, you could encounter scope creep when you come to your measurement methodology. So it's really necessary to put boundaries around that.

Chris Adams: Okay. And when you use the term boundaries here, you are saying I'm gonna measure the environmental impact of the servers, but not the environment impact of building the building to put the servers in. Is that the idea of when you're referring to a boundary?

Will Buchanan: Yes, that's great.

Chris Adams: Okay. Alright. I think this is actually something we've come across quite a few times in other places as well, actually, because maybe it's worth asking about this kind of boundary issue that we have had here, because automatically that sounds complicated here.

And I know that Abhishek you've had some issues at your end as well with defining this staff for deciding, deciding what's in or out, because I think this is one thing that we've had to explicitly do for the software carbon intensity spec. Right?

Abhishek Gupta: Exactly. And, and I think when we talk about boundaries, it's, it's, it's trying to get a sense for what are the actual pieces that are consumed, right? From an operational standpoint, from an embodied ion standpoint and how you make those, you know, allocations across. You know, what, what your system is consuming.

Your pieces of software that are sharing tendency across the pieces of hardware and how different consumers are perhaps sharing that piece of hardware with you and, and thinking about whether you've booked the resource ahead of time or not, whether it's hot or cold in terms of its availability and what implications that has.

I mean, there are so many different facets to it. And each of those decisions, what I wanna highlight here is. That it comes with a trade off, right? So we also don't have any standards in terms of how we should go about measuring that and what should be included, what should be included. And so the way people report out these numbers today also doesn't really make it actionable for folks who are consuming or who want to consume these reports, these metrics in, in taking decisions as to, you know, whether something is green or not.

And I think that's one of the places that the software carbon intensity specification is trying to help folks. Is to help standardize it first and foremost, but also to make it actionable so that if you are someone who's environmentally conscious, you can make the right choice by being informed about what the actual impacts are.

Chris Adams: This is a question that I'm curious about here. Cause so far we've only been speaking internally about, okay, what is the environmental impact of it itself? Like its direct emissions, but the assumption that I have here is that there are ways we might talk. About the impact that it has on the outside world, in terms of what activity we're speeding up or accelerating or supporting there, is that the only issue that we need to think about?

Or are there any other things to take into account about like this system boundary part that we've just been talking about?

Lynn Kaack: Yeah. So these system effects are really important to look at and to consider, maybe just to give an example, like if you use machine learning in, let's say the oil and gas sector to make. Small parts of the operations, more energy efficient that, and the first site looks like something that could be considered sustainable and green, but you also have to realize that often then you are reducing costs as well.

And that might change the way that oil and gas in this particular example is competitive for the particular company is competitive. And that actually might shift also how much oil and gas we are able to use in the short run, how the prices change. So. Indirect effects can actually then have much larger impacts than the immediate effects of such an application.

So drawing boundaries is really important and also opening this up to, to have the broader system level view, and really try to understand how does the technology also change then than to larger consumption and, and production patterns. It's important.

Chris Adams: if I understand that correctly, that's talking almost like the consequences of an intervention that we might make here. So even though we might have reduced the emissions of. The drilling part by putting a wind turbine on an oil rig, for example, that might change the economics and make people more likely to use oil.

In which many cases they might burn, for example, or stuff like that, is that basically what you're saying?

Lynn Kaack: Yeah, essentially what I'm saying is that efficiency improvements in particular, and often they can be done with data science or with machine learning or AI systems. They often come with cost reductions and then those cost reductions do something and change something. And often this is also considered under rebound effects, but it's not only rebound effects.

So it's systemic. The system level impacts that come from more small scale applications that need to be considered.

Will Buchanan: That's such a good point. And I think I've also heard it called J's paradox.

Chris Adams: Yes, J's paradox. This is stuff from like the 1800s with steam engines, right? Like my understanding of the J's paradox was back when people had steam engines and they made steam. More efficient. This led to people basically burning more coal because it suddenly became more accessible to more people.

And you ended up using an integrated number of factories. So there's a kind of rebound, I think that we need to take into account. This is something I think has been quite difficult to actually capture with existing ways of tracking the environmental impact of particular projects. We have like an idea of say an attribution based approach and a consequence based approach.

And maybe it's worth actually talking about here about how. Some of the complexities we might need to wrestle with when you're designing a system here. I mean, Abhishek, I think this was one of the things that was an early decision with the software carbon intensity part to not try to have an attribution approach versus a marginal approach.

And if we're not diving too deeply into jargon here, maybe you might be able to kind of share a bit more information on that part there, because it sounds like it's worth expanding or explaining to people to the audience a bit better.

Abhishek Gupta: Indeed. You know, the reason for making that choice was, again, our emphasis on being action oriented. Right? So as we had started to develop the software carbon intensity specification, One of the early debates that we had to wrestle with and, and, you know, will, and will was of course a crucial part of that as well as were the folks who were a part of the standards working group was figuring out how, for example, the G G way of going about doing that, you know, accounting doesn't really translate all that well for software systems and how perhaps adopting a, a slightly different approach would lead to more.

More actionable outcomes for the folks who want to use this ultimately to change behavior because. You know, without getting into specifics of, you know, what marginal is and what consequential approaches are. And, and if we want I'm, I'm sure, you know, we would, would be happy to dive into all of those details as would I.

But the thing that we were seeing was that we we're doing all of this great work around, you know, talking about scope 1, 2, 3 emissions, et cetera, but it's not really helping to drive behavior change. And that's really it. The crux of all of this, right? Is that we're not just doing all of this accounting to produce reports and to, you know, spill in, but it's to concretely drive.

Change in behavior. And that's where we found that adopting a consequential adopting marginal approach actually helped make it more actionable. And this was coming from folks who are a part of the standards working group, including Will and myself who are practitioners, who, who are itching to get something that helps us change our behavior, change our team's behaviors when it comes to building greener software broadly speaking.

Chris Adams: So that helps with explaining the difference between a consequential approach and a marginal approach. As in the consequences of me building this thing will mean that this is more likely to happen. And if I understand it, the GSG protocol that you mentioned, which is the greenhouse gas protocol and this scoped emissions approach, this is the kind of standard way that an organization might report.

It's kind of climate responsibility as it were when, and when you say scoped emissions, that's like scope one, which is burning. Say that's emissions from fossil fuels, burned on site or in your car. For example, scope two is electricity and scope three is your supply chain. If I understand what you're saying, there's like a kind of gap there that doesn't account for.

The impacts of this, perhaps. I mean, as some people who've referred to this as scope zero or scope four, which might be, what are the impacts an organization is happening to essentially that we mentioned before, do something around this systemic change. Or as Lynn mentioned, like this is changing the price of a particular commodity to make it more likely to be used or less likely to be used.

And this is what I understand. The S St is actually trying to do, it's trying to address some of this consequential approach because the current approach doesn't capture all of the. Impacts an organization might actually have at the moment. Right.

Will Buchanan: That's a good summary. One challenge that I have noticed is that until it's required in reporting structures, like the greenhouse gas protocol, then organizations don't have an incentive to really take. Action that they need to avoid climate disaster. It's something I encounter on a daily basis. And I think broadly, we need to bring this into the public discourse.

Chris Adams: I think you're right. I think it's worth it actually, Lynn, I think that when I've seen some of the work that you've done, you've done previously, this is something that's come into. Some of the briefings that I think that you've shared previously with climate change, I eight work and some of the policy briefings for governments as well.

Is there something that you might better add on here?

Lynn Kaack: Yeah. So something that comes to mind is for example, like a concrete legislation that's currently being developed is the EU AI act. And that's a place where for the first time AI systems are being regulated. also that scale and climate change almost didn't play a role for that regulation in the first draft.

So here it's also really evident that if we don't write in climate change now as a criterion for evaluating AI systems, it will probably be ignored for the next few years to come. So the way that legislation works is by classifying certain AI systems as high risk, and also just outright banning some other systems, but as high risk systems, Could as the original legislation stood, weren't really explicitly classified as high risk, even if they had a huge environmental or climate change impact.

And that's something that I talked about a lot with policy makers and trying to encourage them to more explicitly make environmental factors in climate change effective for evaluating my. So that'd be a very concrete case where making climate change more explicit in the AI context is important also in terms of legislation.

Abhishek Gupta: There's, there's a lot. Said about the EU AI act. Right. And, and, and a ton of in has been spelled everywhere. I think as, as you know, it's, it's, it's, it's called the Brussels effect for a reason, right. Where the I, whatever happens in the EU is taken as gospel and, and, and sort of. Spread across the world, which I think has already, Lynn has pointed out there.

It's not, it's not perfect. Right? I think one of the things that I've seen being particularly problematic is the rigid categorization of what, you know, high risk use cases are. And, and whether the EEO AI act, as we'll see, hopefully with some, you know, revisions that are coming down the pipe is whether.

We'll have the ability to add new categories and, and, and not just update subcategories within the existing, existing identified high risk categories. And I think that's where things like considerations for environmental impacts and really tying that to this. You know, societal impacts of AI, where we're talking about bias privacy and all the other areas is going to be particularly important because we need multiple levers to, to try to account for or to push on getting people, to consider the environmental impacts.

And given that there is such a great momentum already in terms of privacy considerations, bias considerations, I think now is the time where we really push hard. To make environmental considerations, an equally first class citizen, when it comes to, you know, thinking about the societal impacts of AI.

Will Buchanan: This is something I'm incredibly passionate about. I. It needs to encompass the full scope of harms that are caused by an AI system. So that could be the hidden environmental impacts of either the development or the application. The application could vastly outweigh the good that you're doing. Even just expanding oil and gas production by a certain percentage amount.

I think it just must account for all of the harms for both the ecosystems and people.

Chris Adams: thing. Does this category. Actually include this stuff right now. What counts as like a high risk use case? For example, when, when mentioned.

Lynn Kaack: I haven't seen the latest iteration. I think there has been some update on, there's been a lot of feedback on the version that was published. In April last year, I haven't seen the latest iteration. I think a lot of things have changed in yeah. In the first version, there was high risk systems where, when, uh, those that affect personal safety, like human rights in a sense of, of personal wellbeing, but the completely overlooked environmental protection aspects of human rights.

Chris Adams: Wow. That's quite a large one, especially when you take into account the human rights. Okay. We've spoken about the external impact, but I am led to believe there is also an internal impact from this as well. Like the AI has, has some direct impact. That we might wanna talk about as well as I understand it, we spoke about two to 3% of emissions here, but if we know there's an external impact, why would we care about any of the internal impacts of AI at all, really here, what we might be doing or why we might wanna care about the internal impacts of AI as well, example like the direct emissions.

Will Buchanan: So by direct emissions. you're talking about, let's say the scope, two of the operational costs of the model.

Chris Adams: Yeah, there'll be things that we have, there's an external impact. Or there is a, we use this phrase scope four, for example, to talk about all the other things that induce in, in, in the world. But there is a kind of stuff which happens inside the system boundary that we've spoken about. And presumably that's something we should be caring about as well.

Right. So there'll be steps that we can take to make the, the use of AI, particularly like say the, the model is more efficient and more effective and, or all these parts here. This is something that we should be looking at as well, presumably. Right.

Will Buchanan: And so in our paper, which Is

going to be published, I think on Monday, we've calculated the emissions of several different models. And one of them was a 6 billion parameter transformer model and the operational carbon footprint was equivalent to about a railcar of coal. And that's just for training . So it's really imperative that we address this and provide transparency around this.

Lynn Kaack: that for developing a model or for training at once? I mean, is that with grid search architecture, search.

Will Buchanan: For a single training run. So it does not account for sweeps or deployment.

Chris Adams: right. So there's a, there's some language that we haven't heard for here, so, but maybe it might be worth it. maybe will, could you maybe talk about just briefly, you said a rail car full of coal. I don't actually know what that is. I mean, in metric terms, what does that look like? Okay.

Will Buchanan: A hundred million grams. I don't have the conversion handy, but we took the US EPA greenhouse cast equivalencies, and I should add the methodology that we applied was the green software Foundation's SCI. So we calculated the energy consumed by the model and multiplied it, multiplied it by the carbon intensity of the grid that powers that data center.

Chris Adams: Cool. And that was per training run? So that wasn't the, in the, the creation of the entire model, is that correct?

Will Buchanan: correct.

Abhishek Gupta: That's the other interesting part as well, right? When you're thinking about the life cycle is or life cycle of the model, I should say, because life cycle has multiple meanings here, which is that once that model's out there, what are the inference costs? Right. And are we, are we, if, if, if this is something that's gonna be used.

You know, hundreds, thousands, tens of thousands of times, if it's something, you know, if it's, if it's a large model that's, you know, now being used as a pre-train model and is going to be fine tuned on by, by other folks downstream. Are we able to then, you know, talk about amortization of that cost across all of those use cases.

And again, Again, I think what becomes interesting and, and is, is how do we account for that stuff as well? Right? Because we don't have complete visibility on that as well. And, and I know Lynn's nodding here because her paper that's, I think coming out, getting released in an hour and a half, actually the embargo gets lifted on our paper, actually talks about some of those system level impacts.

And maybe, maybe learn you wanna chime in and talk a little bit about that as well.

Lynn Kaack: Yeah, thank you so much. Exactly. So I think what's a crucial number that we're currently still missing is not what is emitted from a single model in a well known setting. But what is emitted overall from applying machine learning? So what are the usage patterns and practices like how often do people develop models from scratch?

How often do they train or retrain them? People? I mean, of course organizations and typically larger organizations and companies. And how do they perform inference on how much data, how frequently. There are some numbers out there from Facebook and Google and in their large scale applications actually inference outweighs their training and development costs in terms of greenhouse gas emissions.

So inference might become a bigger share depending on the application. So we really need to understand better how machine learning is being used in practice. Also to understand the direct emissions that come from.

Chris Adams: An inference is a use of a model once it's in the wild. Is that what an inference is in this case? So there's an environment. So you could think of the making part and then there is the usage part from the inference, right? So is that how that part works?

Lynn Kaack: exactly. So if you use a model on a data point, we call that inference. So you've fed the data and given you a result. Then training means you. Sort of train a single configuration of the model once on your training data set. And then development is what I refer to as if you search over different configurations of the model.

So there are lots of hyper parameters that you can use. Adjust to achieve better performance. And if new models are being developed, then there's an extensive search over those hyper parameters and architecture configurations. That then of course gets really energy intensive because we are training the model thousands of times, essentially.

Will Buchanan: really. Me, I think Nvidia posted on their blog that referencing accounts for about 80 to 90% of the carbon costs of a model. And I think Lynn, in one of your papers, it was Amazon had also claimed around 90%. So these are really non-trivial costs and I'm not aware of any framework to measure this.

Lynn Kaack: That Amazon number just to be clear is costs. So monetary costs that came from a talk, but there are numbers now published. Google and Facebook, but they look at some applications of theirs where inference outweighs training in terms of energy consumption. They're not exact numbers. It's not entirely clear which applications those are, but there is some data, at least that shows that.

And I think it just highly depends on the application that you're looking at. And sometimes, you know, you build a model and then you do inference once and you have the data set that you, and then in other types, you build a model and then you apply it a billion times. so of course that can then add up to a lot more energy consumption.

Chris Adams: Wow. I didn't realize that was actually an issue cuz most of the numbers I've seen have been focusing on the training part. So, well I think this is something we spoke about before that training. There's there's, there's a kind of trend in the use in. Use from, from training already. Is this something, cuz I've seen figures from open AI and, but my assumption was that basically computers are journey getting more efficient about twice as efficient every two years or so with like Moore's law or kumis law or things like that.

But if you are seeing an uptick in usage here, is, does that mean that they're staying about the same or is there, is there, is there a trend that we should be taking into account?

Will Buchanan: So I think the computational costs of training have been doubling every 3.4 months or so. And so I think the trend is only accelerating. The models are just getting larger and larger and you've got, I think GT three is one of the largest. Ones around at this point, I think we might challenge Moore's law.

Chris Adams: Okay. So if Moore's law is doubling, once every two years, I mean, what is the impact of doubling every 3.4 months? I mean, over a few years, what does that work out to be? Because I don't think I could do the exponential numbers, the exponential math, but it sounds like it's, it sounds like a pretty big number, basically dub if something is doubling on a, every three or four months, right.

Will Buchanan: I also don't have the math handy, but I think it's important to note here and Abak was talking about this earlier. These models are very flexible, So,

you can train them once and then provide some fine tuning or transfer learning approach on top of them, and then repurpose these models for a number of different applications.

And then you can even compress them. Let's say, using OnX runtime, so you can be very efficient and you can really amortize the cost of that.

Abhishek Gupta: Yeah, just building on Will’s point there's a lot of work on quantizing the weights of a trained network, applying distillation approaches using. And model approaches that actually helps to shrink down the model quite a bit, especially with the whole push for tiny ML, trying to shrink down models so that they can be deployed on the on edge devices has been something that's helped to manage to, to a great extent the, the, the computational impacts. One of the other things that I wanted to highlight as, as you know, will was talking about Mo models getting larger is there's this almost fetish. In the world today to continuously scale and keep pushing forever larger models in, in, in chasing soda as, as they would say.

So chasing state of the art, you know, which is, is great for academic publications, where you get to show, Hey, I improve state of the art performance on this benchmark data set by 0.5% of whatever. Right. And in performance, I think what what's being ignored is that. That has a tremendous, tremendous computational cost.

In fact, one of the hidden costs that I think doesn't get talked about enough is there's this statistic out there that, you know, 90% of the models don't make it into production. And that kind of relates to things like, you know, neural architecture search and, you know, hyper parameter tuning, where you're constantly trying to refine a model to achieve better performance.

A lot of that actually goes to waste. Because that stuff doesn't make it into production. So it's actually not even used. And so there's a whole bunch of computational expenditure that is done that actually never sees the light of day and never becomes useful. That obviously has environmental impacts, right?

Because of the operational and embodied carbon impacts. But none of that actually gets talked about, reported, documented anywhere because, well, who wants to know that, Hey, I trained, you know, 73. Different, you know, combinations to get to where I'm at. You just talk about the final results.

Chris Adams: Okay. Let's say that if you don't wanna go down one of those rabbit holes, what should you be using? Or where would you start if you wanted to start applying some of these ideas about greener AI in your work on a daily basis, do not have anything that they would lead with. For example,

Will Buchanan: is not always better. Sometimes you really should choose the right tool for the job. We've had some really great graduate student projects. University of Washington's information school and they built some case studies and samples around green AI. As an example, a project led by Daniel Chen was comparing a sparse or a dense model for an anomaly detection setting.

And they found that using sparse meaning less trees and a shallow being smaller depth per tree, random forest would save a massive amount of carbon and provide the equivalent accuracy. So I think it saved about 98%. In terms of the monetary cost and energy.

Chris Adams: Okay. Uh, wow. That's bigger than I was expecting. What would you say to people if they're in production, they're trying to do something.

Lynn Kaack: I think. The big goal should be to not only develop more energy efficient machine learning models, but then also ensure that those are actually being used. And surprisingly, even sometimes within the same company, certain model developments are not being passed onto other parts of the company. So really trying to develop stand up models that are now also being used and practiced is important.

So interoperability of energy, efficient machine learning models. So,

Chris Adams: Someone does wanna look at their stuff and they do want to apply some of these ideas. You spoke a little bit about using some other models. Where would you suggest people look, if they wanted to operationalize some of the kinds of wins or some of the better ways to make green software greener? For example, I realize you've got a paper coming out and you work on this day to day.

So yeah. What would you point us to?

Lynn Kaack: I mean, as I understand, there's a lot of ongoing research in the machine learning community for energy efficient machine learning. So. I don't have any names on top of my head in terms of workshops or community resources, where one can see what are the most energy efficient model types for specific application.

I know that there are some very comprehensive papers also that summarize all the different research approaches that are being taken. But I would encourage you if you are looking for using like a, a deep. Learning models of some kind, just inform yourself quickly if there's also a leaner version of it. So many of them like widely used models, like bird, for example, smaller versions that can almost do the same thing.

And, maybe your performance doesn't suffer much. If you're using a much lighter model.

Chris Adams: Okay, so lighter models and looking around what we have there. And will, is there a paper or a source you might point to

Will Buchanan: It's actually gonna talk about the carbon aware paper that we're about to publish, but I think that's a slightly different track.

Chris Adams: That's up next week, right? So that will be the 13th or 14th of June. That's when that'll be visible. Correct.

Will Buchanan: Exactly.

Chris Adams: Okay. Cool. All right, then there's a load more that we could dive into. We've got copious, copious, copious show notes here. So what I'm gonna do is I'm gonna say thank you everyone for coming in and, and sharing your wisdom and your experiences with us.

And hopefully we'll have more conversations about green software in future. Thank you folks.

Asim Hussain: Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we want more listeners to find out more about the Green Software Foundation. Please visit greensoftware.foundation . Thanks again, and see you in the next show.

How can Open Source Help Reduce Software Emissions?

30 maj 2022 | 47 min

In this episode Asim Hussain is joined by guest Chris Lloyd-Jones; Head of Open Technologies at Avanade and co-chair of the Open Source Working Group at the Green Software Foundation, and Dan Lewis-Toakley; Green Cloud Lead at ThoughtWorks and co-chair of the Open Source Working Group at the Green Software Foundation. They discuss the benefits of open source versus closed source, what tools are already out there and how open source can help reduce software emissions.

Learn more about our guests:

Dan Lewis-Toakley: LinkedIn / Twitter / GitHub
Chris Lloyd-Jones: LinkedIn / Twitter
Asim Hussain: LinkedIn / Twitter

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Green Software Foundation Summit

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Connect with us on Twitter, Github and LinkedIn!

Transcription below:

Chris Lloyd-Jones: If everyone has their data out in the open that they don't feel precious, like they're being compared in a negative way, this incentives to make things better, or you also let all of these open source tools consume from a curated, trusted data source. So open source is almost like that trusted gate for good.

Asim Hussain: So welcome to this podcast episode. My name is Asim Hussain I'm the executive director and chairperson of the Green Software Foundation. And I'm also the green cloud advocacy lead at Microsoft.

Chris Lloyd-Jones: Hello. I am Chris Lloyd-Jones. I am one of the co-chairs of the open source working group, that Green Software Foundation. And I'm also head of open technologies at Avanade.

Dan Lewis-Toakley: Hi folks. I'm Dan Lewis-Toakley. I'm based in Brooklyn, New York. And I'm also a co-chair of the open-source working group at the Green Software Foundation. And I'm the green cloud lead at ThoughtWorks in North America.

Asim Hussain: it's actually just so amazing to have so many people. Uh, Mike focused on grid software and you know, in these environments. So Chris CLJ Chris Lloyd-Jones. So you are at a conference today and you're talking to people about open source software and green open source software.

Chris Lloyd-Jones: Yeah, that's correct. I am at bill today or specifically I'm at the UK chapter in Reading and London. And it's a pretty timely topic based one because I co presented this morning on green software. And there's a really big discussion started with the Green Software Foundation. Then we quickly moved on to CNCF the DPIA and tons of other open source communities.

What was cool is everyone's first thought was how can we work with each other? And also all I see is green software.

Asim Hussain: Yeah. So out of interest, what are people, what is the general impression of people as to what is green software? Cause this is a question which is just, I think, going to constantly be redefined year and year.

Chris Lloyd-Jones: It's interesting actually, because I've seen three broad definitions. The first one was sustainability with technology. So anything from sustainability, like sustainability cloud things like how you track people around your building to make them more efficient. So using technology to make things greener, then there was software.

From my quantity, intrinsically green. So making code greener and less emissions. And I think that the final piece is kind of some combination of the two. So things like using green software for carbon accounting, which was interesting. So I don't really consider that cause I tend to assume that a lot of the accounting protocols that blockchain are inherently not green, which I might be a bit out of date on now, but yeah, three definitions there.

Asim Hussain: When both of you talk about kind of green because you both lead the open-source working group in the foundation. What do you think of when you say, if someone says, what is green open source software, whereas open-source green.

Dan Lewis-Toakley: I can, I can have a stab, uh, same, I suppose, maybe two starting points. One is open source information. I suppose it doesn't necessarily need to be code, right. It could be data or research, but open and readily available tools. Frameworks of research that help enable. Practitioners in the technology space to build software that is green on what we mean by that is software that consumes less energy, particularly dirty energy, and that emits less carbon.

And so we've got quite a wide gambit within the Arkansas working group to sort of explore and discover and help drive and accelerate projects, open source projects that help really enable that space. I think. Been predominantly our focus over the last year, but I think another component is trying to find and understand and accelerate the software itself that is doing things in a green a way.

And so maybe leveraging some of those tools or frameworks that we've discovered, but to help create examples of, you know, software that's written in a Greenaway as sort of a ways for others to learn through that process. So some of the ways that I'd like to think.

Chris Lloyd-Jones: Yeah, I guess I probably agree with that. So I liked the way you started. Cause I think about three opens, open data, open software and open hardware. So kind of sharing. Sharing those different approaches to kind of collaboration and making a lot of important information available outside of large corporates for the common.

Good. And then for me, green software, I do think about carbon and reducing electricity, but I also like to think about the planetary boundaries theory, which are pretty, always banging on. But was thinking about in future, how might we think about e-waste pollution and even vague things like land usage, reducing chemicals, which are much harder to, to affect in any way as software engineers, but I still think we could have something.

Asim Hussain: I think, I think that's one of the space that's kind of what the challenges that we've had is. You're trying to get really focused in on this space. I think required a lot of, you know, narrowing down, not on down the targets. And a lot of us really talk about carbon emissions and things like that. But, you know, you're talking a Palm tree bound.

There are other things, there are other problems. Believe it or not. There's other problems in this, in this world other than just carbon emissions. But, and I think you're right. I think we will over time kind of open that, open that gate open. I don't quite feel that way. We're getting there, but I don't feel quite feel like the carbon, the carbon problem has been solved just yet.

Although I was kind of worried, cause I know we'd be trying to Bovis, which is another non and they released actually they released their open source API, the carbon API, and one of the things it does. I forgot Chris, what the different, it, it, it tells you that carbon emissions is it's an API where you can provide it, you know, adjacent.

Description of this is the, these are the machines I'm working on. This is how many, how much CPU or use et cetera, and returns you, not just your carbon emissions estimate, but also think it was primary energy. And I forgot what the other one that we're returning a biotic,

Chris Lloyd-Jones: Yeah, that was a whole set of manufacturing. So they broke it down and put the different scopes and they scope 1, 2, 3, and broke it down in that way, which, which was interesting.

Asim Hussain: but, but also beyond, beyond carbon, it was also.

Chris Lloyd-Jones: All the minerals

Asim Hussain: Mineral. Yeah. Yeah.

Chris Lloyd-Jones: and reclamation.

Asim Hussain: yeah. Yeah. And, and, and that stuff as well, which I think is interesting there, they're also like looking to broaden out the scope of green to beyond just carbon it's other things.

Chris Lloyd-Jones: I like that because my concern is where is the tipping point? Because so many organizations have net zero carbon targets, you know, either in the next few years or decades in the future. Do you think that now is the time to stay narrow, but eventually we may have to widen out because otherwise if you leave it too late, Who's addressing the other problems, but yeah, I think now we're still in a good space at the moment.

Asim Hussain: Yeah.

Dan Lewis-Toakley: Which, which isn't it really interesting. Right. Because as what building these tools and frameworks, we want to architect things in a way that delivers value in the short term, in terms of, you know, quantifying and measuring carbon emissions. But we don't want to build in a way that it's going to be really difficult in the future to consider other factors.

And so we want to think about our interfaces and design tooling in the space that can be extensible to like future future use cases, not beyond carbon, which is like really interesting problem to think about.

Asim Hussain: Well beyond the carbon, that's a, that's a good title for something Matisha wall.

Dan Lewis-Toakley: Maybe a future podcast

Asim Hussain: It's a podcast. Yeah. Podcast. And I think, I think, I think you, you hit on something that as well, Christmas is like, you know, it's about the targets. So like organizations that, so if an organization sets a net zero target, that's a target relates to carbon and therefore all of the underlying infrastructure and tools and everything tools up to, to, to solve the carbon price.

But then as Oregon, I don't know what an equivalent, I have no idea. Actually one equivalent target would be for some of these other things. Like I know Microsoft, they also look at, I want us to just want to say the word water. I want to say potable water. Like, you know, water scarcity is, is, is one of the key things that are looking at and he works and they have set targets for that as well.

It's just not as high as the carbon target. Like if you want to get something done, you talk about the conference.

Chris Lloyd-Jones: In a way, I feel like it's because carbon isn't easy, but it's graspable now in a way that it wasn't the. And I think the UN SDGs are at that same level of the greenhouse gas protocol, the way that that was in that high-level government global target. But we now need to do that same work of translating it into impactful actions that we as individuals can have.

There's a great book called mission economy, which looks at things like the NASA, the NASA moonshot, and how you can identify these bold challenges to make the targets achievable. Like what you're talking about, like NetSuite.

Asim Hussain: I like that a lot. Yeah. Yeah. So what are those opiates? I would say in general was just kind of general around run, run, run, run, run, focus a little bit. But what are some of, what are some kind of open source projects that you could point to? Some, some, maybe some of the stuff working in the foundation, but also stuff outside of it related to, you know, green, what does exposed some green open-source

Chris Lloyd-Jones: I'll probably refer to when a dance actually, first of all, cause it's one that I actually use. And

Asim Hussain: and he's, and he's probably not going to refer to it. Cause he's so he's so polite if like.

Chris Lloyd-Jones: that's why I was

Dan Lewis-Toakley: I didn't want, I didn't want it. Yeah. Thanks Chris. I appreciate that.

Chris Lloyd-Jones: the cloud carbon footprint calculator. One of the initiatives that I know you're a part of is which I used a day in, day out tracking kind of. Carbon footprint from many different clouds and helping people to identify where they might be able to shrink that. I think that's great because it's an organization just making something that others needed and just sharing it for the common good.

And it does seem to have active pull requests and issues, and people are very welcoming. So that's why I picked.

Asim Hussain: Do you know what let's dig into that? Cause that's actually quite interesting because I think that that touches on some interesting areas because. There are some closed source versions of cloud carbon footprint. So that's an interesting angle to talk about that as well. I mean, why, what are some of the advantages of having something like cloud?

So just for the benefit of everybody else, cloud carbon footprint, I said, why don't you give us an overview of what cloud carbon footprint.

Dan Lewis-Toakley: Yeah. So cloud carbon footprint is an open-source tool, mostly developed by ThoughtWorks where I work, but also we have many contributors in the Arkansas community in some of the members of the Greenstone fire foundation. And the way that the software works as it connects to a cloud provider, API APIs, and it supports AWS Google cloud and Microsoft Azure connects to the usage and billing APIs.

And then with a custom methodology based on, you know, best practice and academic research, we convert that usage data in terms of compute, storage, networking, and memory. Firstly into a quantified energy consumption or an estimated energy consumption. And then based on publicly available data sets of emissions factors, based on where that cloud usage exists, what data center, the cloud provider, we then convert that into estimate at carbon emissions.

And it provides the data in a front end dashboard where you can view it in some, you know, data visualization. You can also just consume it via API or CLI. We wanted to provide a way to understand and explore that data in variety of different ways, based on the context. Recently added support for on prem where you sort of provide your data in a CSV format.

And we estimate on, on premise, and also, like you mentioned, with the Bellavista API as SIM, we added support for embodied emissions as well, based on the carbon intensity standard. But yeah, it's, it's a tool we've sort of put out there about a year and a half ago, and it's sort of grown in popularity and usage in that time.

So something where hopefully a little bit. And we hope to see people use one thing I will add you notice. I said the word estimated energy consumption. It's not the same as the actual energy data that cloud providers would have of their, their data centers and major cloud providers all sort of provide different tooling uses.

More potentially more accurate and measured approaches for the usage. But as you say, it's sort of closed box in some it's to varying degrees. It's a little bit of a black box in terms of where that data is sourced. And so there's interesting trend. Like we went for a trade-off of easier access to the data, better usability, better overall support with multi clouds because we don't have access to the underlying data.

So there's that trade-off with accuracy versus sort of usability in various.

Chris Lloyd-Jones: I liked your mention of reference to close source tooling as well, because one. Your project made me think of a number of things that Green Software Foundation doing. Make me think of it. We launched as a foundation of the software, carbon intensity standard last year and the offer version. And I'm glad that it's done almost early in the development of this software, because there was already starting to be a lot of different organizations measuring their carbon intensity in many different ways.

So by having this standard earlier, I feel like it's going to help a lot of these open source and closed source companies, at least settle on a way that they aren't kind of comparable versus incentive to drive carbon down.

Dan Lewis-Toakley: This is more being out of our longer-term planning, but we've talked a lot about. Uses of these various tools or software could optionally opt into sharing some of the statistics or data about maybe their company size number of employees, industry, and, and some of the outputs of these various tools into some sort of centralized way to understand comparatively across injury industries.

What does good look like? What does not good look like? So not just you internally using the sci to measure against yourself, but how could you compare against competitors? Other comparable organizations in some way. I think we might be a little bit a way off from achieving that, but it'd be really interesting to think about.

Asim Hussain: Yeah, that's a really good, good point. I think it might just only happen because it's cause you're talking about using the cloud carbon footprint as an SEI measurement is what you described.

Dan Lewis-Toakley: I believe we haven't implemented the sci yet within cloud carbon footprint, but it's definitely on the.

Asim Hussain: Yeah, just, just in case you haven't mentioned it for the audience, because we get getting into that phase of, of, of, of our field where we just, we just kind of dive straight into our own kind of terminologies. The sea is a, is a, is a measured methodology has been developed in the foundation. For essentially scoring a software application for carbon emissions.

So just, just going back to the cloud common for it. Cause I think it's an interesting thing that there's a debate out there, an old debate, you know, open source versus closed source. You know, what are the advantages disadvantages of each? I think this is again, just a great example because every single cloud vendor had a podcast episode about talking about all the various cloud and, and, and there are different, everyone will have their own, their own measurement tool.

And with cloud carbon footprint is an open source one. And I think one of the interesting things about that is it's actually a lot easier to engage with. Because of the openness, you know, when you don't remember Dan, when you reached out to me, kind of in very my, my Microsoft hat to, to engage, it was actually quite challenging to get people to kind of start having those conversations with you because everybody's in there thinking, well, what can, what can we say?

Like, I don't want to reveal anything. This has been NDA signed and all this other stuff. Whereas as soon as something becomes open source, boom becomes easy. It's easy. It enables that collaboration and everybody can just see the work that you're doing. They can verify it. There's not, there's no hoops to jump over and that's kind of one of the Clare kind of advantages of open source I would see is.

Is or whatever. I don't even know what word to use described, but I just said like collaboration. Well, I suppose it's the whole, the fundamental nature of open-source is collaboration. Anyway, I see. No, it's not, I'm rambling. I'm rambling now, but it's it's to be open, but it enables collaboration in a way which closed source solutions.

Don't.

Chris Lloyd-Jones: It cuts through legal. And governance technical barriers too, because I mean, I don't think it's just a green thing. I mean, if you look mid pandemic, there are lots of people struggling with solutions for vaccine passports, and you had cities and towns in the U S collaborating with universities across Europe, collaborating with small villages from Australia.

And can you imagine those sort of. Very geographically and size different organizations collaborating without open-source it wouldn't have happened. So it provides visibility, I guess, on a shared platform.

Asim Hussain: Yeah. With closed source. You can't like your methodology is fully is out there. It's completely out there. Every single number and people can review it and back it up. And it builds that trust. In, in UCA, essentially, you're creating something which provides data, which makes decisions, but everything is, is, is out there and open to be reviewed, which is a real advantage.

Chris Lloyd-Jones: One of the things, which yeah. Yeah. I mean, I love open source. It's what I do day in, day out. I want to be real. There are challenges which organizations and people need help overcome. I mean, first of all, you get a lot of people working on that side of their desk around the day job. And I guess in the foundation, we probably even seen it.

That means that people might take their name to project, but not how to time commit or things get spun up and they go stale. I think open source is sometimes seen as a panacea for let's just get something cool, done without thinking about all the hard work and support that needs to go to make that work.

Dan Lewis-Toakley: Yeah, I was going to say something similar to Christian that open source doesn't mean that you don't need things like leadership and project manage. And product development and thinking, right? Like there are still some of those, those core roles and responsibilities that would be required with any in-house software development team is still sort of required.

It might be a little bit more piecemeal in that it's provided by different community members and sort of globally at different times zones around the world. Some of those key aspects to successful software development is still a still really, really important. And oftentimes I've seen organizations just, oh, there's this, like, let's use this open-source tool, bring it in.

And they, you know, don't actually invest some of the time to understand how it's working or, or invest time in investing back into the tool itself. So there's still, it's still somewhat some of that work involved. One other note I'll make is that you mentioned the. Sort of things, Chris, I think we thought long and hard about the license that we were going to use for cloud carbon footprint.

And for us, we settled on the Apache two, which has a similar in terms of permissiveness to MIT. And it's, we still do is really important to have a license that allowed anyone, any organization or individual in the world to use the software in any way they wanted. If, if that meant they were going to take cloud carbon footprint code and build a page closed source service using that.

That was okay for us, because for us it wasn't, it's about engaging the community and growing adoption of this tool and similar tools rather than necessarily holding it tight to our chest and trying to have the secret sauce that you know, that we want to provide to our clients or to partners. So that was a, that's sort of a key ask to get different license types, more restrictive ones can often be a deal breaker for some companies and organizations to adopt.

Chris Lloyd-Jones: Yeah, I don't know about you, but sometimes it can also feel like, you know, when you got. Holding a sweet and you have to kind of prize their fingers off it at times for an organization that can be quite hard to make that decision. And I say that because we we've contributed some code as Avanav to the Green Software Foundation is to start with that CICB pipeline tooling.

But one of the reasons why I pushed for that is because I also think. That secret source, close to your heart. You run the risk of people. First of all, jumping over you because they just want to get things done. They want to get things done quickly. What you do becoming out of date and stale when you've invested a whole load of time and something which isn't compatible with what people are now using.

You can also just look like a bad actor, particularly in fields like this, where we're not, I don't know, making a search tool in comparison to that whole open search elastic search for Roy. We're trying to actually solve a problem here. We're fundamentally, if the world doesn't go near zero, we might all die in a massive heat.

Death has good reasons to do it.

Asim Hussain: I think that's also where this is one of the things that I recognized early on the foundation, because Hey, look, let me be honest. Both of you work for organizations that. Our competitors, let's just be out front your competitors. Right. But the, yeah, but, but your, but what's wonderful. I realized is that open is, is a way you can collaborate with each other because there is an underlying layer.

Every single time you talk to a customer and you're going to be implementing something as an underlying layer of trying to solve a problem, which is common. And that's kind of really what the advantage of open source. And then I love the fact that you use that, that permissive licensing. I always say, even at the foundation we've used, we've used MIT for some of the, uh, for since before content is, is creative commons.

I forgot the one way you can just make a derivative and just rename and sell it the most. Cause that's cause it's all about how do you kickstart this ecosystem than, than anything else? The open-source is again, collaboration between. Method of collaboration between organizations that would typically be competitors.

I think.

Chris Lloyd-Jones: Yeah. And it's nice to be able to set that aside, but also you avoid all the antitrust concerns operating out in the open in a way. It also removes a lot of the, kind of the way whom I did this year, the best cause people could see what he worked on. They can see each one of these meetings, which means we recognize the contributions of those that don't.

Which is an important piece. It's often overlooked because consensus meetings, design discussions is important. It's everything else, but you can also see the code and the minutes, which is nice.

Dan Lewis-Toakley: I don't know about you both, but I love feedback. Like I love getting feedback and there's nothing better than, you know, me pushing out some code to a project and then getting a pull request. Correcting me and say that I was like wrong about this. And like, sometimes that can be a challenge, but open source provides so much more scope for have a feedback mechanism from experts and community members.

And it allows projects like cloud carbon footprint, but many others to sort of better cost cost, correct. Towards things that are valuable because you have more feedback and more users, you know, giving you yeah. Improvements.

Chris Lloyd-Jones: I think open-source is a bit of a mindset as well. Cause I, I'm not going to be at as nice. I sometimes like feedback by half to be in the right mindset to accept it other times I'm like, oh, I just want to get on and do things. But source does get you into that mindset of go actually, no, hang on. They've got a point I'm wrong.

I can acknowledge it and we can move on when you're in that head space.

Asim Hussain: I think it gets a lot into kind of a psychology and where you are in your life as a human being at the current moment assignments to whether you can, uh, whether you can access. But as soon as I've seen, I've, I've seen feedback where I'm like, that is, that is not okay. Or that's, but then again, even that feedback is public.

So you know, that, that conversation that happening in private, so you can then, you know, it's happening and then you can correct them. And, you know, do you think you, you reminded me of something? One of my, I don't know if it's related, I'll say it anyway. What am I kind of like happiest moments of the foundation was one of my first poll requests got rejected or one of my, one of the first time, one of my poll requests got rejected because I was like, Um, thankfully there is like a community of people and they're confident enough to kind of feed back to each other and kind of reject things and accept things.

And I was like, excellent. This is going to the stick, the culture's there. It's going to work. They're rejecting being okay with having poor requests. Retracted must be, it must be a really important thing in any kind of open source community.

Chris Lloyd-Jones: Yeah. I like the visibility that gives them because it's a bit of learning experience. And as long as it's public of those, then see what the feedback was that they gave to you. That'd be the only thing I struggle with sometimes is. Sure people can give that negative or critical feedback in a way that you can learn from it because you don't want to put off few, two contributors.

I feel like if it's your first one, you can also feel a little bit burnt too. Particularly for junior devs. I see them contribute. Once they get strong feedback, I'm like, oh, I'm scared of this now.

Dan Lewis-Toakley: I think you mentioned that what culture statement. I think that's really important, like different communities or that are a foundation or a particular pencils projects have different cultures. And I think. Being deliberate about the type of culture that you want to foster is really important. I know that the foundation, early days, we worked on some principles like principles, cultural principles to underpin how we engage within the foundation.

I think similar thought process is really useful for openings. Projects having a really grounded code of conduct. Ideally you don't ever need to use the code of conduct, but it's important that it's there in case, you know, people sway away from the intended culture. And Chris, I love your example of like the, that junior devs me and then getting negative feedback.

I've seen it happen. I think something that I like to think about is before providing. Direct or constructive feedback, like ask, ask a question instead. Like why ask questions to understand before delivering any sort of judgment or value in terms of people's contribution?

Chris Lloyd-Jones: Yeah. Question is, is, is a great one, particularly being out in the open and chatting because. I know internally and in the past, this is certainly something I'd like to think I've learned from. I can see something and I take a negative intention from it. And it goes back to that psychology piece. When you ask often you find that are really good reasons for say why someone did something the way that they did.

And it also brings you closer together to work together and future, sorry, we're making open source sound very touchy feely. Now I'm good, but good things get done with open source.

Asim Hussain: Now there is it is, it is, it should, it should be is. I think there's a, there's a human component to everything, which I think really does need to be addressed. And I think it's actually, you might, you might need that. Cause that, that was one of the, I'll have to find it right now. Can't find it by. We we wrote, we wrote, I think we wrote in our manifesto assume good intent, which I think is one of the heart we have that shop that Microsoft has all like assume good intent and like ask it enough.

And that's the hardest advice. Great advice to give as hard as vice to remember, to use your sound. So, so ways here, but I'm sure there's many times that we kind of fail to do that ourselves. Yeah. Asking that question. Why? Oh, what was your reasoning behind this instead of this? What are your reasoning behind this?

And then you realize, oh, that's a really good reason. Oh, that's kind of how it goes. Yeah. So talk a little bit more as great as for us to talk about generally kind of source kind of challenges and solutions. I was wondering, do we want to, do you want to kind of touch on some more of the, the kind of green open source projects that are in that happening?

I know Chris, you just mentioned the, as the sci CD tool, I forgot what CICT stands for all of a sudden.

Chris Lloyd-Jones: Continuous integration, continuous deployment.

Asim Hussain: all right. Okay. There you go.

Chris Lloyd-Jones: Yeah, I know that was a contentious naming. And then there's also the carbon where a software development kit. Those are two of my favorite projects that we have because they come at a very similar problem at our different parts of the life cycle. And I'm excited with the fact that the green stuff, I foundations a lot of different roles and people in it.

So we put developers, data scientists, we have business folks. And what the carpenter. Well, I start with the CIC D tool. What that does is that lets developers at the moment and just that infrastructure is code, I guess, almost the design patterns for the code that you want to end up in the cloud, that the services which you want deployed in Azure, Amazon, Google, and other cloud providers and forecast what the likely carbon emissions are going to be.

We would like to be static code analysis. It's a harder problem. It doesn't do that. So you forecast the carbon west software development kit is almost the next part of the chain. Once you have your applications up in the cloud and you have your infrastructure, this that you instrument them to figure out what times of day should this software run?

When is the energy clean? Where should this run? I'm unsure workloads. Then you have tools at the other end that can look at that forecast of both the infrastructure and when software should run, I'll tell you the actual sales. So I feel that we're getting a really whole end to end view of here's the forecast.

Here's what actually happened. I was slowly plugging the gaps that every part of what a developer would do is covered. And that makes me really excited.

Asim Hussain: I actually hadn't joined the dots between the CACD static analysis and the carbon until just now, because with the CIC, the tool will do is I can point it to my GitHub repository of my entire backend application. And it will just by looking at my Docker files, try and figure out this is this, these are all the instances you're going to probably create.

This is probably what the utilization is going to be. And therefore this probably probably what the carbon emissions will be on the carbon or SDK. Is it that, what that does is it essentially gives you your advice? I don't know how to describe it regarding like when is the cleanest time of the day? To run stuff and they never really joined those two together.

So we could actually figure the actual carpet emissions and potential carbon emissions are. Oh, wow.

Chris Lloyd-Jones: So forward looking and then what Dan's got like the cloud comfort for nasal my step backwards looking viewpoint. So eventually one day we'd get that Delta between this, what we forecast, this is what happened. How can we slowly close that gap and figure out what bits can we improve more? And in an ideal world, every part of everyone's job could just be greened in that way, which would just be.

Asim Hussain: One of the thoughts I've had for the longest time is, is, you know, I like the way you phrased it in terms of the software development life cycle.

Chris Lloyd-Jones: Yeah.

Asim Hussain: that's the way I kind of think about what, what we need to, to get to is kind of one of, wouldn't it be great. If, if, as you're developing code, like on your laptop, you had information there and then that would inform you somewhat about, well, you've just made this.

I'm not even gonna call it coaching, but architectural change some sort of change to your software, by the way, it might have, you know, this, you might, it might improve this, this change, my increase, your carbon emissions somewhere else. And then I, and I, then I was imaginable. What's the next stage. And everybody, they even develop as journey is to check something and then push it to some sort of a get lab, get hub, whatever it is, code code repository.

And then that's what you're describing. Chris is kind of like that, that started code analysis there. At that point, you know, do some actions, then you can see like, well, my push might coach. Merged in with the actual code base estimates, this, this kind of impact. And then once you actually are deploying the into production, that's when tools like the calc carbon footprint.

So yeah, I liked the way that, so I think that's, that's a really good way of thinking about it, kind of all that chain through, um, all of that really has to be open source. I just, I mean, maybe part from the deploying to the cloud providers could come up with their own proprietary way, but everything else in that journey has probably has to be open source.

Chris Lloyd-Jones: And even they're collaborating. Cause I mean, if you look at Google homes in Nashville, it's in their interest to make it as easy as possible to get on their cloud. People don't want to be locked in. So in a way, people compete in the quality of their servers.

Asim Hussain: That's a contentious point. So it's three dig into that a little bit,

Chris Lloyd-Jones: Go for it. Go for it.

Asim Hussain: I don't know. Cause I've had that because I think there is this, there is, and I know Chris Adams as well, has this idea as well. If you have open source. If you, if you're building something on some kind of non vendor Lockton opensource system, let's just say, Kubernetes, for instance, you then have the freedom to move, but you can therefore use that power.

If provider a is an adherent to some green standard or something you can easily move to provide a B. Is that kind of what your, the open source

Chris Lloyd-Jones: It is. I can see where you're going with that is that when you start to use vendor specific extensions, you then get cruft and little bits of luck and little hooks in your code, like tracking a fish out of the water that make it really hard to move from one to the other.

Asim Hussain: But, but also I would all get it. I I've heard. And I, and I, I'm not an engineer in those platforms. I do not, I don't know, first hand noise. This is second hand information, but the more vendor looking you get, the more efficient the actual platform is, does that make sense?

Chris Lloyd-Jones: I think that was the case. And I'm going to strongly push back on that one. I will repeat the fact that I am at Microsoft build a Microsoft conference. And later today, I'm going to be talking about as you container apps, but that is built on dapper. Dapper is a non-green project, that distributed application platform, runtime.

What that does is it takes all these vendor lock in pieces.

Asim Hussain: It's opensource.

Chris Lloyd-Jones: totally open source. It was contributed to the cloud native computing foundation. What it does is it hides all these proprietary pieces behind abstractions. So you're not using an Azure SQL database or RDS in Amazon. You're just storing data.

You aren't needing to know necessarily how even Kubernetes works. It has different building blocks, calling services, saving data. Publishing observing and secrets. And then it has little adapts. It's that which gives you that developer productivity of moving fast. So you're allowing you to stay efficient, allowing apps to stay small so they can be carbon efficient, Watson not locking you in.

And, uh, that to me is the ultimate expression of freedom from one place to another. So it really serves as again, as we can.

Dan Lewis-Toakley: Another example I might add is some Spotify. And source backstage tool that provides an interface for developers to deploy resources across all the cloud providers, but many other infrastructure and services as, as sort of like a central dashboard for them to do it. And the reason I mentioned it is we actually recently worked with Spotify to publish a plugin for cloud companies.

So, if you're using backstage, you can now sort of install the cloud carbon footprint, plugin connected to your cloud providers and get that same data and data visualization within, within that platform. And I think that's a really neat example of different open source communities collaborating in a way, and adding building blocks together to build better solutions, which I think was really.

Chris Lloyd-Jones: Yeah. Sage is another great example because that does a similar thing of dividing up tools into categories, infrastructure monitoring discovery, and let him develop his peak was right for them through that agnostic portal. Yes.

Asim Hussain: I see now. So I'm, I'm, I'm standing corrected. Well, what or.

Dan Lewis-Toakley: As him, but to your original question, like that efficiency gains or cost gain, I'm going to give you the classic consultant answer. It depends. It depends. It depends on what your goals are, right? If your goal is to get the most efficient bang for your buck, in terms of like dollars spent, then. Deep partnership with a single cloud provider where you can negotiate every single cent for all the instances.

Like maybe that produces the best gains because you pay a lot of money for that. But is that your goal is your goal to scale most effectively is your goal developer effectiveness, developer efficiency. It really, it really, it really depends. The tradeoffs. I like to think of software architecture or software in general is trying to pick the least worst decision.

Chris Lloyd-Jones: Yeah.

Asim Hussain: Right.

Dan Lewis-Toakley: There's always trade offs. There's always,

Asim Hussain: always always trade off.

Chris Lloyd-Jones: People often think of it as, as Lego, but I think as marble run, cause you've got this whole runs, you've got to get the marbles to fit through as quickly as possible. I like that because it shows like the Rube Goldberg desk, like maze of parts and pieces.

Asim Hussain: Yeah, that actually fits with me because I play with my sons, but I do play with his novel run a lot. Both of you, you mentioned that Chris, you mentioned dapper and Dan, you mentioned backstage and they essentially sounds like FAFSA. To has mean like bringing up my design patterns from, I did use the code, but facade for essentially kind of the wide range of, you know, not in this example, car providers and kind of their interfaces.

Um, so maybe it's like a layer thing. Like, you know, as long as it's a thin layer on. Well, you could have like all the, all the optimizations underneath, but like the interface layer itself is the same and that's kind of where open source comes in. And I was going to say, Dan, like, that's one of the big things about cloud carbon footprint is it's multi-cloud and the only way that could be is if it was the only way a multi-cloud solution could exist.

Well, I, I take that back. If there was a startup criticizing or they could do, but th the only realistic way it's going to create as if it's open source, because. Like I can, I think I can say this, but like Microsoft, I think it might be obvious that Microsoft customers like Cisco is car platform, but they also Shakara use other cloud platforms as well.

And then there are some questions while you've got your wonderful tool. Like, can you, can you, Microsoft, can you make your carbon measurement tool calculate my Azure and my Google workloads. Unlike Microsoft would never decide legally that's just system minefield. You wouldn't even go in that direction from a closed source tool, but like an open source would actually allow that.

So that's kind of like for creating a facade amongst a lots of different cloud providers, lots of different APIs, Chris, I feel has different opinion or

Chris Lloyd-Jones: No, no, I don't think my would never do it. I mean, look at Azure up. That's exactly what that's trying to do,

Asim Hussain: yeah.

Chris Lloyd-Jones: but actually a lot of the underlying tech from Azure, there are other cloud providers out there to these open source.

Asim Hussain: Yeah. So that's maybe like the mechanism that would allow that kind of cross from the feeling I had is like, if, if, if you can't publish while you're saying your Amazon number is X, then there's then the one, if you could publish that in some open source, if the methodology for why you're calculating numbers and numbers of certain ways public, then that's going to stop you from an illegal.

Chris Lloyd-Jones: And does that take us back to the sci data project? Like another open source project, the foundation. We're trying to get providers and vendors and manufacturers to share their data publicly, to prevent these legal concerns. If everyone has their data out in the open that they don't feel precious, like they're being compared in a negative way, there's incentives to make things better.

Or you also let all of these open source tools consumed from a curated, trusted data source. So open source is almost like that trusted gates, a good data.

Asim Hussain: Yeah, trusted as a thing as the big term, big word there. And I think by the fact that, you know, well, the opensource working group is like a. Objection of people sitting there curating it, experts in the field. That's what gives it the trust because anybody can, I can create a source data in my, my, my, my, I have got several open source projects that no one should

Chris Lloyd-Jones: Heavy burden, but Dan and I and yourself were trusted in this one day. We're going to have our equivalent of elastic search and open search. I'm sure. But like it's, it's cool to be in. And to see where it's all developing at speed.

Dan Lewis-Toakley: Chris, you've done a much better job than me at going over the projects in the, your working group. So I might I've barely listed any, I might list one, one more to add alongside the ones you made. Is the, the sci open ontology project, which looks at a different part of the problem domain of the software carbon intensity standard.

So when access to the trusted data, which you mentioned, press is totally a problem that we're trying, we're really trying to solve. But another problem let's say you do have the data. Another problem that comes up when trying to utilize the STI standard is where do I draw the software boundary? Let's say you have a, some software running on an instance in the cloud and you have a database and then you have a large number of users maybe accessing that API or something.

Do you include all of the end user devices that are being used to access? API. Do you include the networking over the internet to access it, you know, or do you just include the software code that's running, like making those decisions is something that is up to users of the sci. And so the, the open ontology project is about defining a standardized way of making those decisions about how and where you draw the software boundaries.

So it can be consistent and a lot easier for people to sort of make those.

Chris Lloyd-Jones: Yeah, which is important, I guess it helps the training as well, because I mean, you made an awesome workshop, which again, you shared with us around how to calculate sci. I tried running it, um, our organization. And it's incredible actually, when you get people using tools like the SEI and you see, you think you've written something really clearly and have people adding up all these different kill Afric is averaging them, doing them together, picking different or figures from monks.

API calls are as a part of the sci calculation. So the SEO told your project, you mentioned is almost like a training tool to help standardize the way people do that. Yeah, and I will share more details of that workshop in a way, because it was a car crash in a good way. I learned a lot.

Dan Lewis-Toakley: Awesome. I'm running meant to be running in a couple of weeks, so yeah.

Asim Hussain: my, my, my advice and works, I used to do a little bit shops is like, just make it as simple as you can and then make it 10 times simpler. And you still there still be people struggling. So, yeah, so. We're reaching kind of the end, the end of the hour that we have. I just thought, maybe give, give I'd love to kind of ask you a broad question, actually, each of you to, to, to see what's going on in the world.

So, you know, what else in the world of green software has kind of got your attention? Recently? I asked that to Chris first.

Chris Lloyd-Jones: Ooh, I feel really put on the spot now, can I take off piece and just pick something which I think is a really cool in open source generally. So that is that the United nations, the us federal government and the European. Have a setup open-source project offices and a collaborating. And a lot of them are looking at digital public good, which includes screen software.

So the fact that you have not just your traditional corporate community and other organizations, but also now governments taking apart, that's incredible. As long as their heft and size, doesn't put off other people and having a chilling effect, this is going to be a really great scientific collaboration of the.

Dan Lewis-Toakley: I'm going to give you two things that I'm thinking about. One is you sort of touched on this a bit as seam in that the. software delivery life cycle and How we think about that. Something that. With our clients about and partners is there are many ways to solve a problem with software architecture, many different ways, maybe unlimited ways.

Imagine if you could model the carbon and energy impact of different software architectures that solve a given domain problem prior to writing any code, um, and make and factor that in as a cross functional requirement with alongside cost and performance and security. Right at the earliest stages before any lines of code is written, anything's pushed to a CICB pipeline.

You can at least put some guard rails around the architectural direction that you had in. I haven't seen that done successfully yet. Maybe I missed it, but I think that that would be really cool to think about and something that is sort of top of mind. Secondly, I just want to give a call out to the Green Software Foundation summit coming up next month, you know, dozens of in-person and virtual events all around the world.

I'm super excited to see some people in person. In some cases for the first time in a, you know, two years and that yeah. Go to the Green Software Foundation websites, check it out. That it's something on the, on the calendar that I'm really looking forward to.

Chris Lloyd-Jones: How could I miss that?

Asim Hussain: that Chris, Chris is involved in that. I actually think Chris, I think, I don't think we've ever met in person. Have we? So I'll be meeting you in person at the event. I think the London event. Yeah, yeah, yeah.

Chris Lloyd-Jones: There's so many folks, the foundation I need to meet around the place.

Asim Hussain: Are you, are you told, I don't know. You don't know, you don't know, you know, people are quite surprised and quite a tall person anyway. Um,

Dan Lewis-Toakley: was literally about to say height is the most surprising

Asim Hussain: it's the most upright.

Dan Lewis-Toakley: but I'm like, you're so tall or in some cases it's not as though it's like, I have no idea how

Asim Hussain: That's what people, yeah. Some people say too.

Chris Lloyd-Jones: And I say a lot of the current chairs, many of the projects. So from surrendered project approach, pretty tool and understand average. Someone's going to be there by giants.

Asim Hussain: yeah, not in terms of my answer, kind of like one thing that it's just been a number of changes, number of essentially jobs going up recently, one thing I've noticed, you know, in the world of green softwares, a number of jobs with the words, kind of green or relates to green self. I mean, honestly, two years ago, if you didn't claw your role yourself inside your organization, there were no like jobs out there kind of publicly.

You know, posted, and it was deeply unlikely. Now I'm seeing, you know, you've both got titles that kind of relate to the degree of software I saw. So Amazon posted up, they've got roles for a sustainable solution. Architects. There's a new role for agreeing that advocacy that Microsoft and I'm seeing more and more, not often, but more and more kind of roles appear in this space.

And I think that's one of the most important indicators of the validity of what we're doing, you know? And the moon behind this, there is an ecosystem, the business behind it. And that's really what will drive this, this field, I think, in the, in the future. So can you let a confidence?

Chris Lloyd-Jones: Well, I want to say on that is, I think also your lead share from the Green Software Foundation has been a big part of that because I'm going to say I was watching bill near a year ago today when the GSF was announced and I text, I was on holiday. I texted my boss via people. I mean, we're conditioning.

Why aren't we in this what's happening. We sh we love green. What we're doing as in that.

Asim Hussain: Oh, wow. Is that, was that, was that the, was that the inception story for other? Not

Chris Lloyd-Jones: Yeah, we were so happy to join.

Asim Hussain: yeah. It's really good to you. Yeah. Very good to have it really, really glad you're here. Maybe this is before I wrap up any other information you want to give any, any, any, tell people what they can find you socials or anything.

Chris Lloyd-Jones: Sure. So you've called me a few times. That's my initial. So you can find me @sealjay the two animals undescore, the letters CLJ because I, 70 organizations have Chris's so that's always me. Yeah. I don't want to sweater.

Dan Lewis-Toakley: Also also on Twitter and get hub the handle DToakley as in DTOAKLEY.

Asim Hussain: Wonderful. Thank you so much. Both of you for being our guest today and also being such active participants and chairs of the foundation, your, your leadership. This is just been instrumental in us getting to where we are today. So thank you.

Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we want more listeners to find out more about the Green Software Foundation. Please visit greensoftware.foundation. Thanks again, and see you in the next show.

How do we make Green Changes in Organisations?

16 maj 2022 | 49 min

In this episode Asim Hussain is joined by guest Anne Currie; Tech Ethicist at Container Solutions and Lecturer in Tech Ethics at the University of Hertfordshire. What are the real factors that drive organisations choices around increasing efficiency within their organisation? What needs to happen for senior leaders to make sacrifices for sustainability? Can regulation push for real change inside organisations? They discuss the role of middle managers, developers and their love for ops people!

Learn more about our guests:

Anne Currie: LinkedIn / Website (Sci-fi Novels included!)
Asim Hussain: LinkedIn / Twitter

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Green Software Foundation Summit

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Connect with us on Twitter, Github and LinkedIn!

Transcript below:
Environment Variables_Ep3_Transcript

Anne Currie: We need to be shifting it so that people think how efficient is this and demanding it? I think our power is developers is not to do stuff. It's to demand stuff of the people. We are buying things.

I'm your host. Assamese sane. So welcome to Environment Variables. My name is Asim Hussain. I am the executive director and chairperson of the Green software foundation.

Anne Currie: And I am Anne Currie. I'm a tech ethicist. That's a long time veteran technologist, I'm Tech Ethicist for Container Solutions, which is a member of the Green Software Foundation. I'm a lecturer of Tech Ethics at the University of Hertfordshire. And I am a serial startup founder.

Asim Hussain: So I'm a serial startup, a failure, which is, I think I will. This is, I think maybe this is an interesting direction to go into the, into the topic because the topic I want to talk about today, it was, how do you make change inside and organizational more specifically? How do you. Make changes relates to green software, the greening of our field in how'd you have those conversations inside organizations.

And actually one of the things I have because I've had some success, not a lot, not a lot, not a lot of success, but some success inside the organization, main organization I've worked in while I've kind of gone to space, which is Microsoft. And I often wonder whether the success I've had is linked to all of my startup failures or the failures I've had. And what is a startup, other than trying to convince somebody of something, you know, on the, on the most literally buying into your idea is, is what a startup is. Is that, how is that? How you see it? Do you see similarities?

Anne Currie: Absolutely. Yes. Yeah. And it's not just slightly buying in really buying in.

Asim Hussain: Yeah.

Anne Currie: Yeah.

Asim Hussain: I always say like, you, you need to, like, you really need to come out with going off on a tangent, but I always say to be successful in a startup, you need to give a, a poop about $5. And that is because you need to, you need to get somebody to give you $5 and the, you need to care so much about the first, relatively small amount you're going to get.

You to put so much energy into it that you need to, which is why, if, if, if you're reasonably successful already, it can be quite difficult to get started because you're like, I don't really care about the first $5. Um, I want the billion, but yeah, cause I, I remember maybe describe like, what what'd you describe, you had some, some successes inside your organizations you worked in, in, in, or even, not even in the organization, but generally regarding spreading the word of Greenville.

Anne Currie: Well, actually for me, green software is a relatively recent thing. I've been, I've been pushing it for in my career. I've been pushing efficient software for a long time. And I got, I started down this route. Based on pushing software efficiency. And I found that it really doesn't sell at all. It's I really thought, well, if I can sell people on cutting costs that who doesn't want cutting costs is always on people's lists of things they want to do.

But what I realized was unless it's the number one thing on my list. Like, there'll be, they'll take the meetings with you. They'll talk to you. But fundamentally what they'll actually want you to try to do is solve the problem, which is at the top of their list, which for development organizations is always developer products.

Asim Hussain: Yes.

Anne Currie: That's and the machine productivity will all add cost is always secondary to solving their developer productivity. So you go in trying to sell them on efficiency. And what they actually want you to do is to increase the fee is increasing the developer productivity of their teams. And I had to change the message.

And, and I'll talk to you a little bit. I have to talk about how I changed the green message to align it with. So I had, I had a bit of an epiphany. I was cause I used all of talking about developer efficiency, efficiency of our containers, Docker containers and that kind of stuff. And I was giving a talk at a conference HashCorp EU, about that.

And the person who was immediately before me was Mitchell Hashimoto, who was the, uh, who's the founder of, of HashCorp. I've watched a couple done, fantastic work around, no match, no match scheduler about increasing the efficiency of hosting applications. And my talk was all about increasing efficiency. His talk was all about.

No one could sell increasing efficiency because all anybody cared about was developer efficiency and they didn't care how much it costs to make the lives for their developers simpler. So they were, they were having, they were basing everything off that. And I thought, yeah, I really need to change my message here because he was quite right.

That it is, it is all that anybody cares.

Asim Hussain: Is it, I think, and I just inserted it's up to you, but I think it's because we cost too much. I talk about this. We're very, if you to look at the costs of the cloud workload infrastructure, and it can be quite significant, I will admit, but the cost to developers and also the opportunity cost of your developers.

I think you, you, you, it's not just the cost, the developers with like, if they weren't there and they weren't doing what they're doing, your, your competitors are doing something and that might so. Fearful thing that you just got to deliver as fast as possible. You've got to be in the market first and that, that drives a lot of the investment.

Anne Currie: Yeah, absolutely. It does enable developers to move on. And, you know, I've, I've run development teams in the past. And hiring developers is really, really hard. It's hard to get them. And it's hard to get ones that are not going to be a disaster when they appear. So nobody wants anybody to leave. No one wants, no one wants to, everyone's desperate to hire people, but, and they want to just get the most out of them when they're in.

They do not want to spend any time on anything else unless they have.

Asim Hussain: Yeah. So how so, how have you changed? How have you managed to change the narrative somewhat to support kind of greening of software systems? Like what arguments.

Anne Currie: Well, I, cause I started, I'll tell you what didn't work for me. So I started describing efficiency improvements and things like, because I'm quite old. And I used to work on backend servers in the nineties and everything was written in C and exchange and things like. Which, which are a hundred times more efficient than Mo, than modern equivalents often, usually in fact, so, you know, it was, it, it there's so much efficiency to be gained by just write, using a more lightweight language.

You see, you could use rust or whatever, but that's landed appallingly people, developers went, oh yeah, that's true. But then never be able to get signed off on it because it's too expensive and slows what slows them down. Those languages are very inefficient for develop. Unless they're absolutely necessary.

So unless you're writing something that's super high performance and therefore has to run at crazy speeds. Everybody's just going to go look, let it run more so late. Let me pay for more hosting that burned more carbon in the atmosphere and, and I'll have it in Python. And thank you very much.

Asim Hussain: But I'd also argue that. I mean, it's not always the language that dictates how efficient or inefficient the code is. Oftentimes. I mean, you can write extremely inefficient code in C and you can write extremely shockingly efficient code in JavaScript. Make fun of JavaScript. You can write incredibly if you, if you, if he understand the language really well, you need to be good at what you do.

But what I was surprised me was that it doesn't really matter because the end of the day, all that matters. Yeah. People are happy to ship inefficient, pour quality code that delivers on the functionality to whatever agreement that is required to meet the business goals. Yeah.

Anne Currie: I agree. That's all that people care about is shipping quickly and that it basically works. That's. So I changed my pitch a little bit to, well, how can I, how can we go to something more systemic? So rather than have developers tune their code and make it all efficient and super amazing. Can we put pressure on somebody else who is making that case?

It's a big hit, super efficient, so specialists so that there are two, two parts to play. As there you've got the pipe cloud providers who are providing services, can you make their hosting efficient? And can you make their services efficient? And can you make them offer more services at the tar green and efficient?

And then the second thing is, can you make open source projects start to. Really targets, carbon efficiency and offer those carbon aware features and things like that. Now, I haven't really looked at it. I, at the time I went full hell for leather on cloud, but we also, I think, need to start working on, on the open source side of things.

Asim Hussain: I mean, I I'm constantly reminded that how much open source is actually used in every single project. I mean, the GIP team tell me that it's about 90% of an enterprise stack. Open source. And actually some of the things that they've mentioned to me in the past also is hotspots. Like there are, I mean, if you're in the organization, I think actually if you can, you can navigate the public database of open source projects.

And there are like a few key libraries that are used every. You know, if you look they're there. So there, those are the hotspots and any improvements in code efficiency, which is not something I'd necessarily like to talk about, but that, that those are the places where you probably should put effort into code efficiency because the impact is magnified so many times from an open source perspective.

Anne Currie: I mean, things like service meshes. If you're, if you're running Kubernetes, the service mesh runs all the time and it is it's on demand and it's often horrendous. It's like, you know, mining Bitcoin on your applications. Absolutely redness. So there are some coming on the market that are more efficient, but at the most.

It's not necessarily the key selling points of those things. I think we need to be shifting it so that people think how efficient is this and demanding it big. I think our powers developers is not to do stuff it's to demand stuff of the people. We are buying things

Asim Hussain: Ah, yes. Yeah.

yeah. Using our purchasing power to, for the good, but isn't, isn't it? Cause that, that has been my experience kind of throughout, throughout this whole. You know that this whole process, you might, you might get some buy-in. If you're talking to your leadership, you're talking to different organizations, you might get some buy-in, but I always describe as grace and favor is through the grace and favor of a leader who sacrifices some sort of form of some sort of metric that they are being heavily measured upon to just, well, let's give sustainability.

This is important to me. The number of senior leaders who've mentioned to me that, that children. Saying things to them at the dinner table, as well as driven them. I mean, I think that is, I I've heard of that statement before, but I, I really believe it now is kind of as there's an aspect of that, but it's through grace and favor and you might get, I mean, it's not the worst thing in the world.

You might get a project kick-started, you might be able to do some research or something along those lines, but it will not scale. It just will not scale to any, to any large. And until you can align what your offering to what they're getting measured upon. And I think that's really what they are getting measured upon is the challenge.

And no one is currently significantly getting measured against sustainability. If it was, I think sort of the things that some of the conversations we have, it will be very, very different.

Anne Currie: um, my suspicion is that, I mean, Isn't it. It's not like I love the cloud providers. You'll worry about

Asim Hussain: I should be more in that.

Anne Currie: your threat. I don't love the cloud providers, but I think they are soft targets. And that if you, you know, it kinda, it's aligned with people. If you say it will make your life more easy. If you use cloud services and cloud services agreed, then you kind of like you're selling your life is more easy and also you get green for free, but I'm, I'm thinking one of the sales in the future would be, I know full well people hate to measuring things and having to report that's terrible.

Everybody hates doing that. So I suspect. Being able to use cloud services where you can just say, oh, give me your report. Here's the report I didn't have to do anything is actually another USP for cloud.

Asim Hussain: It is a USP, but I think interestingly, it becomes cause then these cloud provider has to offer a different value proposition of that reporting. And then you get into, like, I just probably went to get in trouble saying this, but we really have to be really convenient. If every single cloud.

provider just gave the same report,

Anne Currie: Yeah, the net. We're going to do that

Asim Hussain: but because there's no, there's no capability then saying, well, our reports are better than the next person's reports.

Right? You don't. Cause it's all constantly between the cloud providers. It's all about, I would say is jokingly saying that there's so much more that than. You can say this argument for all of life, to be honest with you, there's so much more that those is commonality between his and his difference. Like the different, like you've got a piece of code, it needs to run it on someone else's computer is pretty much the same between the Google, the Amazon, the Microsoft app.

It's all pretty much the same. We have, but we focus in on the, on the differences. Like what are the key difference at differentiators between them? And I think this has just been one of the differentiators, like reporting is just another thing to be differentiating. It's very few areas where there's a lot of focus on, on standardization.

Anne Currie: I can also see why they did that. I mean, so I don't have any inside knowledge of any of these, but I remember hearing a stat that I found quite musing about Microsoft that might stop now hires more lawyers than engineers.

Asim Hussain: I don't know. I honestly couldn't tell you if that was true. I should probably also acknowledge it right at the start. My other affiliation is towards Microsoft. I don't think, I think I would probably rerecord the start maybe in the kind of acknowledged that also. But yeah, my I'm most of the green cloud because Microsoft, so I do have some insight into that.

Anne Currie: But I can see why, if you were a lawyer, you would want to steer clear of standardization of reporting on the green side, because then you, you introduce a liability. Then if you just say, oh, it's hand-wavy about this, then you want to avoid someone coming up and saying, well, hang on a minute. I made all of these legally binding reports based on your, your thing.

And actually I've just compared it to get to Google and they say, you're wrong.

Asim Hussain: But if Yeah.

but if you always different as if what you're providing is always different to everybody else, you can always put an argument is there's always a spin on it where standardizations forces you to the same level playing for where you have to then compete on the same rules. And you really do discover who's who's better than the other.

I think it's interesting. I think I do. I do. I mean, obviously there's many people inside an organization and everybody has. You know, motivations and it's not all the same and it's not just one voice all the time, but no.

one says they want standardization until the standardization happens. And then they really want the standardization because then they, but then cause then they just finally on the same level playing field and then they can compete fairly, but they have to be dragged in. You know what I mean? Which is why I was away for me. Cause I, I spoke this similar stuff. I talking on so many levels of so many people inside organizations about green software. I did have successes, I call and grace and favor successes. You know, you know, people who are willing to just put in investment to, to, to something to see, see how.

To the scale investments? No, I'd never really had any successes there. The first time I got the hint that this might not be the right direction or there might be a different direction. Was when we started talking about regulation when there starts to be hints of regulation on the horizon. And one thing I realized, I mean, just the conversations, just the threat of regulation opens more doors.

Than anything else. They, for instance, one of the things I've learned early in my years at Microsoft was that you really do have to find customers. You can't just like be waving around going, Hey, I'm really passionate about technology. And I know my area and, and you know, like if we were to build this feature, you know, I, trust me, trust me a lot of people with.

I said, no, one's paying attention. You've got to come in with like, I've got five customers. They all want this feature. This is how much money they want to spend. If they get this feature, we should prioritize this. Let's get this prioritize and because, okay, look, let's, let's do this. That regulation surpasses. Would open the doors to passing that fear. I w I used to work in investment banking and one of the lessons I learned leaving investment banking was there's only two things people really care about. And that's fear, fear, and greed fear, I think is greater than fair regulation was the grit and the grit of, of, of money.

So I think that's the direction that would really help us out a lot is more regulation. This.

Anne Currie: It would, it really would. I totally agree. Yeah, without it, life has been a lot more difficult than aftermath. I would say another technique, a tic technique that I used to use that to some effect, not, not to massive effect to someone like him with that. We didn't have any regulation. Apart from the threat, the threat of regulation is to find someone who is accidentally doing something really good.

That right. And talk to them and say, do you know you're doing this thing? It's really, really good. Let me write it up is an amazing thing you were doing. And then they would go, oh yeah, we're doing this thing deliberately. And then they would let you, and then you would, they will talk at conferences about it.

And they will say to that to everybody that this is something really care about that because they're not, they don't do any, there's no effort. But so once they're already doing it, they'll talk up what they're doing and then say, oh, this is something we would care about massively. And that's why we're doing it, but getting them to do it, it's impossible.

Asim Hussain: Yeah. Yeah. So it's like someone who's already doing some work around or something that could, that is more efficient or it does reduce carbon. And I do. Thoughts as well. But then I like, for instance, for me, I get into the whole realms and it gets stuck in my head regarding these kinds of moral and ethical tech ethics.

So this is the perfect person to talk to, but I'm like, this is going to be a, this will be quite a tangential conversation actually, because then where is greenwashing? Right? Greenwashing is a really challenging concept for me to wrap my head around because I work in a large enterprise organization. You know, it's kind of everything that you do has multiple vectors inside it, right?

When someone's making a sustainability announcement. Yes. There are people involved who cap passionately about this space. There's also marketing people who, you know, they're getting measured on how many clicks and all this other stuff. So everybody's, it's this mixture of things that go into kind of announcements.

Some of it is also just gonna be like, well, we can make money out of this as well. Like there's whole that aspect coming into that. But. Always upset of core people who care passionately about this and then figure out what, in order for this to be successful, we need to get marketing people interested in it.

We need to get business people. Who's get the financial people in, in, in, in, in, on our side. And so what I always say is that there has to be that the initial intention. That's I think what Greenwood greenwashing is is, is when there's no initial intention to do good in the first place for me, that's, that's what I define as greenwashing.

But then again, other people have told me, I may ask him, there are companies that really agree mushing and what you're on the other end of the whole spectrum to what we really mean is greenwashing. But anyway, yeah.

Anne Currie: I think that's very, it's very hard to be active in this without constantly thinking about is what's the right thing to do when it wasn't. The other thing I do is I write a series of science fiction novels. I've just published book number seven. And the entire thing is, is like the question of. What's the right thing to do, you know, are you, is it can't, is it it's you have to be truthful all the time or is it Bentham?

Is it utilitarian? If it, if it comes out on to on top, is it good? Even if the intentions were wrong or if the intentions were right, but some of the things that happened along the way until the line in order to make it happen, but it has a good effect. Is that right? Or is it wrong? It's very hard.

Asim Hussain: It is. It's really hard. I bet we have enough.

Anne Currie: Nobody has done.

Asim Hussain: Yeah, I suppose they would have to have their internal answers. So for me, myself, I'm like, I, as long as the intention is, as long as my intention is good, I'm happy to speak to whoever I need to speak to and put whatever argument I need to spin on it to, to make it happen is, is, is the way I kind of view it.

But Yeah.

Anne Currie: So you, you are a follower of Jeremy Bentham,

Asim Hussain: Is that a bit? Is that it's I utilitarianism,

Anne Currie: Yeah, you

Asim Hussain: the sound.

Anne Currie: Emmanuel Kant truth at all costs.

Asim Hussain: No. I'm not like an idealist like that. No, no, no, no, no failing whilst having a good feeling about myself. Isn't isn't for me is not, you know, so what that's just, that's actually ego. That's like your own ego, you know, I, I would argue your ego shouldn't matter in any of this. Yeah.

Anne Currie: Yeah. It's, it's interesting. It is. It's not. Aye. Aye. Aye, aye. Also utilitarian, cause I want to Tappan, you know, and I, I, but at the same time I can see the argument that. Humans should be given all the information and allowed to make a free choice. But, but the reality is that they don't have all the information and they can't make a free choice.

So, you know, it's what world do we live in? We live in a world where you're going to have to make an argument to your boss and, you know, cap Emmanuel Kant says, so is not a sufficiently convincing argument.

Asim Hussain: I think that's the end of the day is, is, is it's all about convincing people, that's it? And that's all that your life has ever been about. It's about convincing people. Yeah. Anyway, get back back to topic, back to topic, back to topic. Yeah. So one of the things I found was, was, was, was, was regulation was one of the things that I, that really seem to open the doors.

It didn't, I mean, until it until actually. I mean, some of the conversation I thought was the thought was interesting was that there is, there is a strong desire to, to be ahead of regulation, not to just respond to it so that there that is there. But I do think the world we're just not quite there just doesn't seem to be any regulation.

That's just on the verge of like pushing for real change inside an organization. Yeah. I mean, what, I don't even know what some of that regulation would look like. I mean, a carbon tax would obviously be the main thing.

Anne Currie: What makes Spain, Spain ready have variable electricity pricing. So. There's a lot of S that the sun is shining and the wind's blowing electricity is one hell of a lot cheaper than at times when it isn't. So that's, I think that's quite likely to come at some point, but will it be enough to change?

Asim Hussain: Well, if it, I think going back to the original point of developer velocity is if that's all there is, I mean, how much of, I mean, one of the key challenges that we have you talk about, I talk about it is, you know, server utilization. I mean, if there's one, if there's one thing that you should work on inside an organization, which would have the biggest bang for your book, it's increasing your server utilization because most workloads for more.

Organizations are running a very low levels of utilization, but when you and I, and I initially thought that just because I start having conversations with customers and people who have low utilizations, and I initially thought it was something along the lines of, oh, they just don't know. Um, oh, if I just, if I just explained to them, this is the situation, did you know, have you heard of auto scaling?

Have you heard of auto scaling? Oh, this is wonderful thing. A lots of skeletons. But no, like the reasons for that there's many reasons for why they have chosen. They've chosen that path, but not, they're not fooled. Nobody knows what's going on. They've said, well, we're willing to live with the added expense of having quite a lot of our machines idle because of X, Y, and Z and X, Y, and Z have almost always got to do with money.

Yes. We could go auto scaling. We could use as all the other methodologies, but you know, we've done it in the past and we lost three days worth of trading and that's social. It's not really worth it for us. We'd rather like live with the risk.

Anne Currie: Yeah, yes. Which is why, again, I, I try to point people towards. Cloud managed services for that kind of stuff, because it's just too hard. You, you, you don't have the skills in your team. Almost everybody wants to, to go to zero ops these days. So the last thing they want to do is to give additional complexity to their ops teams.

Asim Hussain: this is maybe this is what cause you, you and I have kind of had quite a few conversations in the past about how to actually cause at this, I think we're at the crux of. The challenge that we have. And I think we both agreed, this is, this is it. This is this, this log that we're trying to move this, this rock that we're trying to, to roll one of the ways that I've spoken about this and I've thought, Well, this is going bottom up.

You know, uh, we need this to change. Maybe the solution is engaging with all of those, not just developers, the people who are in the business of building software, all their software teams. Making a priority for them so that it just kind of happens because some of my experience in the past and working in, in, in, in, in engineering organizations is you can always, if you want to try and get something done, you can usually go the official route, which is through the top and get it prioritized.

And like all the other stuff, or you can just sideline have a bunch of conversations with the actual people building. And they're like, yeah, right. I'll just implement that tomorrow. You know, something along those lines. There's a lot of stuff that actually gets done. I think in the bottom up approach, I think you have a, more of a, a slight different thinking on the topic then if you want to.

Anne Currie: Well, I would love to do it bottom up. And I, and I do agree with you that, well, there's been a lot of psychological thinking organizational thinking about how do you make changes in organizations? And because a lot of it came from changing the finance industry because it had to really change after that, after the big crash of 2008 and 2009, and they did loads and loads of psychological research and organizational psychological research.

And what they found was that really. Uh, less middle managers decided to do it. It all stopped, you know, they would stop it going up and they stop it going down, which is where I think the tech conferences are very good because they tend to be attended by people in the middle managers, senior architects, if you can get them on board.

Then that's all this really required. Top-down gets stopped by that bottom up, get stopped by them. So tech conferences are a good, a good place or, or, you know, magazine said th the register who is traditionally been a bit, a bit slow or pushing green and stuff. So they had no use, but that, that kind of level of folk are probably the people that we, that if we can convince we can change things.

But I mean, if you look at Kubernetes, that's. Bottom-up thing to do. And I think it crazy to implement Kubernetes is unbelievably large amounts of work.

Asim Hussain: so it's crazy to implement

Anne Currie: Yeah. I think it's

Asim Hussain: Oh, wow.

Anne Currie: amount of work. I would go to a manage service any day of the week before.

Asim Hussain: Oh, I see what you're saying. So like a man is Kubernetes is kind of a containerized solution. Yeah, yeah, yeah, yeah, yeah. It doesn't stop people. I've heard people like I've walked into projects and I'm like, you you've implemented what by yourself? Yeah.

Anne Currie: I know what I can see, what people used to and anybody over a certain age has written their own orchestrator in the

Asim Hussain: well, it's fun. Isn't it? It's like you start on a project. You're like, I'll just get.

Anne Currie: and necessary if he wants to have performance, you know, if you want something to stay up, you had to write some kind of orchestrates and to do that. But yeah. Can we as just a company.

Asim Hussain: but is that maybe it may, maybe that's what it's all about because maybe it's certain types of decisions need to go middle, middle. And certain types of decisions in this space needs to go bottom up, because now what we're doing here is, is, is wide. This there's a wide variety of things you can do.

Like choice of technology. People think happens higher up inside an organization. And I don't think it happens higher per second organization at all. I think it happens right in the trenches. I always see a leaders, rarely ever pick a technology, which that team is not already comfortable using in my experience. usually a lot of resistance if they're not, if you're picking a different technology, a lot of resistance. So maybe some of the, maybe it's like some of the leveraging, some of the open source technologies or some of the solutions, some of the more engineering type solutions should be bottom up.

People can just pick them. We make them make those choices easier for people. But what kind of, what kind of areas could middle management help out within their space? Just measuring prioritization of issues or where money goes.

Anne Currie: Well things like whether you're going to move to the cloud or you're going to be on.

Asim Hussain: Yeah. Okay.

Anne Currie: very much a middle management decision. That's not something that the young person is going to, that, that kind of level of change. Isn't isn't really going to happen.

Asim Hussain: that's true. That's true. I get a lot of feedback from people who are kind of at the lower, early stages in their career. And the feedback to me, oftentimes for a lot of advice has been, this is all great, but there's nothing I can do. There's nothing I can do. Whereas middle management have more capability.

Anne Currie: Yeah, but, but I think you have an interesting point, which is that there are, there are decisions. There are large-scale decisions. There are middle management. Which might be okay. Well, do you want to, actually, if you want to start measuring this stuff, do you want to move into the cloud? Do you want to adopt this new service?

Uh, at a large scale, large scale, but there are also small level decisions, like, well, given your already in AWS or Azure, or do you want to use this service or this service, or do you want to use this open source tool or this open source tool? Because that kind of thing you do have change. You do have country.

Asim Hussain: Just thinking about cloud. I think that is an easy one to think of, but yeah, one of them is like moving to the cloud, but I think it's not just moving to the cause. There's two aspects of moving to the cloud. There's one, which is the cloud is some, may be somewhat more efficient. Than where you are currently.

And I'd say maybe It may be somewhat more efficient than we

Anne Currie: It depends what services

Asim Hussain: depends on services use. Yeah. I think we're having some conversations about this previously, but I think the, the great thing about cloud is it allows you. To architecture solutions in a way which increases your utilization. I think that is kind of like the fundamental crux of moving to the cloud.

If you're not on the cloud, if you have your own private servers, you there's no point doing auto scaling because you're just, you have your own servers. What's the point you have to, you have to deal with your own headroom and the public cloud. Yes. It's more efficient. And B it gives you the, just the potential.

Of of, of increasing utilization, but then it's moving to the managed services and that's when the conversation gets very, very tricky, because then you starting talking about a vendor, lock-in

Anne Currie: Yeah.

Asim Hussain: kinds of things. And I've had conversations with, with people, engineering, people who, who are in engineering organizations in kind of cloud providers.

You like build those managed services and they're like, look, if somebody look, if you want to be more effective, You have to use vent in a vendor specific services because that they are, they are making it more like the, the, the Microsoft service is incredibly efficient because the Microsoft engineers are building it for Microsoft platform and the Microsoft system.

But that's a Rudy for the longest time. That was the hardest sell in our space. It's like we are we trying to avoid.

Anne Currie: I, I think I would embrace. I embrace lock-in. I think the only way you get efficiency is through locking. If you do lift and shift into the cloud, really, I can't see how you get any benefit from it whatsoever. I think I suspect much more expensive and not really any more efficient, marginally, more efficient.

You have to use the service.

Asim Hussain: Less efficient as well. Yeah. because you're, you're, you're moving from one paradigm to another paradigm and there are differences fundamentally in the cloud.

Anne Currie: Yeah. Yeah. But the only thing that does is that once you have lifted and shifted, as you say, people are more likely to start comping off staff and using the managed services. So it's kind of a gateway drug lift and shift to the gateway drug

Asim Hussain: get into the cloud wherever you can get into the cloud. And then, and then, but then like, but then why, but then we back to that same challenge that we have, which is why, like, what are the motivating factors to, to them? Implementing change. I mean, you know

Anne Currie: Well, it used to be the big lie of lifted shift would save you a load of money. The cloud was cheap.

Asim Hussain: But then, but then even once you're in that, but w we need people to change and use more efficient services costs. Isn't a motivating factor by itself.

Anne Currie: But I would say by far the most, the most effective argument for getting people to use Monett managed services is zero ops. No ops. That, that, if you can say you could get rid of all of those, you know, you know, all those people in ops that you really don't like.

Because I have to say, I mean, I say I love ops people. I love well, because I'm

Asim Hussain: I love it. This is let's just make sure

Anne Currie: I took, I love off street mall, but most of the rest of the business always found them quite challenging.

Asim Hussain: Yeah. I actually started off my whole career. It's actually, now that I think about it. Yeah. So they go to SIS admin, we call CIS

Anne Currie: Well, you were very tool guy, people who are very tall, tend to, I find that it's opposite. It's full of people who quite tall. I think probably because in the old days you could lift machines around,

Asim Hussain: Yeah.

Anne Currie: never found any girls in all, because we all just went, can't lift this

Asim Hussain: Yeah, that's hard. Jars are so heavy and was bringing them up from, from four inches off the floor to eight inch. Oh, it was so hard.

Anne Currie: That's why there were no women.

Asim Hussain: They're all women

Anne Currie: And the cloud enabled women to answer ops because there's nothing to lift

Asim Hussain: It was literally lifted and shifted in,

Anne Currie: and deed. It was literally lift and shift, but yeah.

Asim Hussain: I forgot what we're talking about.

Anne Currie: Oh, yeah. Yeah. Well, how do you persuade people to the managed services? I think it's really all about look, you know, cause you don't have to fire, so you might want to fire ops people, but you don't have to fire ups people. You can, you can use them for develop the dev ops.

You can use them in fear of. And, and whoever is at your managed services or AWS managed services or Google managed services will do some of their work for them. And, and, and that's really, and then at the same time, you need to push Google as your AWS to make those services more green. So that at the same time, you're pushing that developer productivity by a backdoor.

You're also upping their green side of things that was more kind of Fairfield.

Asim Hussain: When you, what you're really saying is you're upping the ops productivity by dividing by a fewer number of numbers until you've reached zero. Which mathematically doesn't work anyway. So you're basically saying, is this the same argument you're saying like, we're going to increase your ops productivity.

We're going to increase as you're increasing your developer velocity. Like if you use a managed service, you increase the developer velocity. So basically it's, it's, it's it's, that seems to be one argument from our conversation. And it's the only argument that is of any importance ever in any developer situation, which is a developer productivity.

Anne Currie: Yeah.

Asim Hussain: And I felt that my felt, because I felt, I remember in my early years of my career, I've been in this space for 20, 20 odd years now. Yes. I graduated in the.com boom. So I was very lucky to get a job, uh, for university. That's just like, you have a job. And I even at that point, if you valuable to do things. And I, and I remember, I know there's been certain points throughout my career. It wasn't because it happened so slowly. It wasn't like one day something over so slowly. I was like, it just doesn't seem to be that important that I write anything efficiently anymore. And then, you know, I started off in C plus plus, and it just got less and less and less supported.

And in fact, my earlier the career was doing. now, I don't know what to call it now is great. Regionally grid computing. So high-performance computing. So like being very, very like super efficient servers were still pretty expensive in those days. So there was like, it was worthwhile investing some time from an ask him to make your code more efficient because the actual server costs would, would like don't micro save on the servers would like be less than an asset.

But now for some reason, it doesn't, it doesn't that that equation doesn't doesn't relate anymore. Is it? I think a, you know, our costs have skyrocketed, you know, B S cloud costs have gone down, but as a Jevons paradox argument, isn't it Cod costs have gone down. But also I think that there's also this kind of opportunity cost thing, which is what is this it's it's triggering.

The fear is that we could spend, we could spend an asset making it more efficient, or we could spend an asset in building another feature to be our competition.

Anne Currie: Yeah.

And it's, it's not just that. I think it's the, those, those complex languages. They, they were quite, it was quite, quite slow to develop.

Asim Hussain: mm.

Anne Currie: it just, yeah, it, now everybody wants it, this small thing out, out, out, out, out to see what happens and to be more interesting. And so bull Paul is dead. Most of these two work quite well with, with CNC plus plus, and yeah, nobody wants to do it cause I, I, there, you know, the dirty secrets of agile is it is not more efficient than waterfall.

If you can, if you can get all to full rights, if you, if you really have a clear idea about what you're doing. You can do it with quite a small number of engineers, quite cheaply. It takes you a long time and you don't get to any iteration on that. So if you were right first time, then great. If you've got it wrong from the start you're screwed.

Asim Hussain: I think agile, like if with waterfall, if you, if you, if you know what you're doing, if you know what you want to build, it works. Agile is like, I don't really know.

Anne Currie: Hmm

Asim Hussain: We're just responding to like requests, like trying to agile when you don't have a bunch of customer requests coming in as is doesn't really work.

I don't think. Yeah. w we've been quite negative. I think someone listening to this podcast might just like, Table flip and give up. I don't really know. Look, they want to ask what are the successes that we've

Anne Currie: The huge success. And there was an you and I both know that we did not necessarily see this coming was for the big three cloud providers to all commit to carbon zero operations by 2030 in 2020. We, because we've been pushing, I'm sure we'd both been pushing for this for a long time. Not really expecting it to happen.

And then suddenly. And that sets, that sets the timeline for everybody that says, look 2030, I expect your, your app, your applications, to be able to run carbon zero, which means. It's something that's that, that, that we need to discuss. What does, what does that mean? Does that mean that basically 90% of your CPU load is on carbon-free electricity and you can 10% maybe, maybe no more than that is running at times when this get, it has to be on fossil fuels.

Asim Hussain: I think my definition is a lot stricter, which means that it's it's it's it's it. I don't know how they're going to do it. My definition is like super strict. And what I do know is like, I think you're absolutely right. Like they, they, they, that that's been a wonderful success. The fact that the major cloud providers have not only committed to that, but also they provided the vault now very, very new, shorter provided measurement tools, you know, which is very, very impressive.

What's interesting about that is that every to achieve that goal every single year, you need to make incremental improvements and you know, all the low-hanging fruits. So, like, I won't say like, people are like really thinking extremely hard, you know, next year, you know, we, we committed to this, what, what do we need to do now to make that happen?

And the wonderful things about public commitments is that they're public.

Anne Currie: Yeah.

Asim Hussain: like, you're like, I don't want to make it out. I don't want to, I don't want to say we failed is a big driving, driving factor for that. So, Yeah.

So that's been a big success.

Anne Currie: Th that there was, there was a really interesting paper from Goebel. I know that you've read this paper, the paper from. In June last year saying, okay, how will the hell are we going to do this? But they say, we CA they can know how to do this on their own workloads. They look, and they can do it on the word workloads by introducing more time shifting, but the public cloud, because it's all black boxes to them.

But then really that, what you're saying then is you need instance types, the main things that aren't, aren't that things aren't black boxes anymore. So.

Asim Hussain: Price signals.

Anne Currie: it, yes. I'd say sports instances, instances with some degree of SLA, which is not quite as, as on demand, as you know, let's get rid of those on demand service and the lift and shift and move everything over towards containers.

Whether look, you know, I'll, I'll run it within 10 minutes, but I'm not going to run it within 10 seconds.

Asim Hussain: Yeah, but they also have to like, cause I mean it's for a net zero target by 2050, they have to have eliminated 90% of the image. And that's in a world where we've probably quadrupled 10 times grown by 20, 20, 20 50. And so that kind of like, yeah. Yeah. So yeah, you're right. Setting. Those kinds of targets is, is really, really important.

And I, and I was going to say like, one of the things that I've seen successes internally is soon as soon as your organization sets a target that helps you and proposing prep pre. Why would they need this whole grace and favor? Once you've got a target, then you can say, well, Hey, you know, you're that target you set, but you're starting to lose a little bit of hair over because you're going to like have to, well, if we were to do X, it can maybe meet, not 0.2% of your target, whatever, you know, and if enough people do that, that's enough.

So I think that's kind of you're right. Those companies setting targets is one of the big things is one of the big things. And, and, and I think that's one of the first steps. If you, if you, if you work in an organization that hasn't set a target. Forget about trying to push any green software work. Your first job is to, is to advocate internally and do what you need to do to get your company to set target.

Anne Currie: Yeah. And, and set a target for a long enough distance in the future that all the executives think they'll have left by then

Asim Hussain: because you're right.

Anne Currie: managers know they might be around that's where middle managers who I'm good. So they tend to be here in 10 years. I'm going to have to meet that target. That's why middle managers stop things happening.

Asim Hussain: Oh, I love, I just, I just got your point because if they're going to be leaving, then they're willing to set a target as someone else has to. Yeah, yeah,

Anne Currie: the cuteness of setting the target without any of the pain of meeting the target.

Asim Hussain: a, It's like a musical Chaz, like whoever's, who's the CEO left of the table when they've got to get to get to zero. Yeah, absolutely.

Absolutely. So I think, I think, I think those, those bold aggressive targets and kind of any, any internal advocacy you can get to kind of push those targets, upwards is, is, is, is one of the big thing, big successes, I think like, well, I have seen a lot of, I've seen a big sea change since January last year, a big change in the number of.

People customers, people just, just asking about this question, which is like, Hey, I've got this software. Like, what do I do to make it greener? Like, just asking that question is, is, has been coming up a lot and that's going to that's one of the things we've been trying to do in the foundation is just try and get some answers together for that.

You know, what is the advice that people have for that? So I am incredibly infused over the last, because I've been in the space of craft for years as well in the first couple of years. What do I have to do? The last, the last year has been very, very exciting because of all the other, you know, it felt like vindicates and validated.

A lot of other people who care about this, I'm asking those right questions. There's been a lot of interest in that space. I just don't think we have a lot of great answers yet working

Anne Currie: It's an interesting gods. In some ways we dealt once, once today's questions. Cause we want people to ask those questions of their providers. So that they know that there is interest. We want people to be saying, how do I do this? Where is my green region? How, what's your plan for making the regional I'm already hosting in green.

Asim Hussain: but then they're all competent, but then everybody's joining the foundation and going, well now let's, now let's figure out the answers to these questions. So I think that's that, that, that, that's what Sonos, and I think kind of one of the things I, I found one of the most impactful things that I've personally done and I you've done it as well.

And a lot of people have done it is. Talking about this stuff is educating teaching, training, making people aware of it, making people. I remember I was speaking at a conference. I won't say where I was speaking at a conference last late last year. And it was, it wasn't like a tech conference was more general purpose sustainability conference.

And there were a bunch of people giving talks and I gave my talk and the feedback I got afterwards is people coming up to me going, wow, you know, you're the first person who's talking about solutions. Everybody else is talking about, you know, oh wait, where we're all death destruction. It's all doing the problems much worse than we think.

And I'm like, well, I don't have time for that anymore. It's this? I've heard it all before. Well, the solution is like, what are the things? And I think that's kind of where we are. We're in that even what you just described right now, like we're talking to cloud providers, that's one solution and we're now we're kind of talking about other ones, like finding other solutions and just talking about it. Yeah.

Anne Currie: Yeah. And decide. We'd love to know, have to decide to, to act and they have to act. So, you know, we, we. Making sure they know, but we also have to pitch the no in a way that they will decide to act. So if you say, Ooh, you need to know about this. It's going to cost you a fortune and it's going to reduce your developer productivity.

They will not decide to act. So you need to frame it in a way that they'll go, okay. Well that sounds doable. I could decide to act on that.

Asim Hussain: Yeah.

Anne Currie: So that's the next thing is. Um, how do we make it? So it doesn't acting doesn't immediately blow away all our, actually that thing, the metrics that they're actually already measured on.

Asim Hussain: Yeah. And I think that that might be how we, how we end this podcast, like with like a question, uh, relevant, but yeah, that's been a really refund chat, I should say. And I hope, I don't know if it's, I don't know how useful this, this conversation is to other people, but I found out a lot of fun.

Anne Currie: What was I was, I think that was good. I think it was good.

Asim Hussain: Okay.

Wonderful. So, thanks for listening to Environment Variables, all the resources for this podcast, including links to our guests and more about, well, the topics that we discussed as well as the Green Software Foundation are in the show description below. We hope you enjoyed the show and see you on the next one.

Hey everyone. Thanks for listening. Just a reminder to follow Environment Variables on Apple Podcasts, Spotify, Google Podcasts, or wherever you get your podcasts. And please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we want more listeners to find out more about the Green Software Foundation. Please visit greensoftware.foundation . Thanks again, and see you in the next show.

Carbon Aware Computing

2 maj 2022 | 40 min

In this episode Asim Hussain is joined by guests Scott Chamberlin formerly of Microsoft and Henry Richardson of Watttime as they discuss how time-shifting, location-shifting, curtailment and other terms are important to Carbon Aware Computing. How can we build sustainable software that reduces the impact on the environment and how these decisions may just lie in the hands of the developers instead of the CSR teams.

Learn more about our guests:

Scott Chamberlin: LinkedIn
Asim Hussain: LinkedIn / Twitter
Henry Richardson: LinkedIn / Website

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter
Green Software Foundation Summit
Magazine: Branch Magazine

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Connect with us on Twitter, Github and LinkedIn!

TRANSCRIPT BELOW:

Asim Hussain: And one way I think about carbon awareness is actually I'm building software, which responds to natural cycles of the earth. And it connects me with nature in an indirect way, but it's one of the few ways you can connect with nature. I think in software.

Hello and welcome to environment variables brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding green software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host, Asim Hussain, welcome to the Environment Variables Podcast.

We have an exciting episode today talking about carbon aware computing.

Scott Chamberlin: Hi, I'm Scott Chamberlain. I was previously at Microsoft leading some of the sustainability efforts in the windows organization. And the day, this podcast is airing. I'm actually starting a new role at Intel leading their software sustainability effort.

Henry Richardson: And I'm Henry Richardson. With Wattime we're a nonprofit really focused on making grid emissions, available to partners to achieve impact through load flexibility of sighting of renewables. And so we're really excited about the kind of expansion of capabilities in software to take advantage of flexibility in grid emissions.

Asim Hussain: Henry cause this, this, this is your, this is your bread and butter. This is your space. Do you think you can give a, go trying to explain, you know, carbon intensity in these concepts to the audience here.

Henry Richardson: Absolutely. One of the things that we spend a lot of time thinking about is how. Clean or dirty that the electric grid is. And what we mean by that is when you make a change on the electric grid by increasing or decreasing load, how does emissions change? So if you decide to schedule a compute load at a specific hour, a certain set of power plants will be responding to that change in load and they'll have an associated emissions.

And so you can see how by scheduling. Load updates or sorry, windows updates to specific times you could actually affect which power plants are operating. Ideally, we would be scheduling those two when there's excess solar, excess wind, which can happen pretty often in the great Plains. There's a lot of excess wind, a lot of excess solar in California, but you can also pick between coal and natural gas, if you can have that flexibility.

So. We measure the intensity, the electric grid, and then we make that information available the software and making it available to software as what we consider carbon awareness. So can the software take advantage of that? The time varying emissions, intensity of the electric grid and actually change when it trains machine learning loads, as Scott was saying updates, major pieces of software, can you run.

At different times because they're, they're run chronically or regular. So things like that. So we see lots of opportunities in software to be kind of carbon aware and take advantage of this, this flexible.

Asim Hussain: Yeah, I think there's lots of other examples. I mean, there's, there's other big, big examples of well as well from other organizations, but where they started to apply the ideas relate to, to a carbon Alanise I think broadly, it, it, if you have the kind of. Software workload, which can respond to a signal, not all software can, that's the chatter challenge.

Not all software can respond like this, but if you do have the kind of workload that can respond like this, and that's why the windows update is such a perfect type of workload, because it's something that, you know, you need to do at some point. But you can have a reasonable amount of flexibility over when, when that happens.

If you're, if you're, if I'm visiting a webpage, I don't have that. You know, I need that web. And in three seconds, the other kind of famous use case I've seen is a large, large scale implementation of this I've seen is, is the work that Google's done with their carbon way data centers, which I think is quite an interesting, like the work that Microsoft's done with the windows is, is, is, is on a device.

And then Scott you've told me before I keep on saying, I'm sure I get the number wrong. Is it 10 billion devices around the world? Use windows? Is it? What is it?

Scott Chamberlin: think the last public number is either 1.4, 1.5 billion devices use windmill point for 1.5 billion client devices use windows, and then there's a separate, separate staff for our servers and data.

Asim Hussain: Well, so there's a, quite a few people in the world who now think is 10 billion, because I just threw that stat out so many times in the past, one of the things you're involved with at your time at Microsoft Scott was well windows, but specifically, and then an announcement that windows made recently.

Do you want to give a quick summary as to what that was.

Scott Chamberlin: sure. Totally. And again, this is in partnership. Wattime and Henry's organization as us in partnership with electricity map and the tomorrow organization. One of the first things that we did in windows was figure out how do we bring carbon awareness into the. Operating system, right? The operating system is responsible for scheduling tasks.

One of the things that does, and the question we had was if we had a CO2 intensity signal, could we number one? Change the behavior of the operating system in a way that was beneficial for the environment and had minimal user impact. And number two, would that have a significant impact in the emissions associated with the energy used by PCs around the world?

It was a PC focused saying not necessarily a server or a data center focus thing. I'm still, yeah. Recently in preview, Microsoft released the. First implementation of carbon aware, scheduling for windows update. And so windows update is essentially how windows applies new features for users. And there's this whole set of criteria which go into when is the optimal time to apply an update for a couple reasons.

Number one, in a lot of cases. It causes a reboot to happen on your machine, not all cases. And in some other cases, it requires CPU cycles requires a bunch of things to happen, perhaps to close that kind of stuff. So we added to that list of criteria, carbon awareness. And so if we can, you know, with this within a certain timeframe, Find a period of the day where we think that the CO2 intensity and the grade is going to be lower.

We're going to try to do the update during that time, rather than at a time of day, which might be optimal from the other criteria, point of view, but might have higher suit to intensity. So that's. Feature, that's just been released to windows preview. And in an upcoming version of windows 11 is available to windows insiders, not available to the general windows population yet.

And they're doing evaluation and testing of, of that feature at this point, that's the ability to shift across the time of day, which we would call time shifting. And then there's the ability to just shift appropriate workloads. Yeah. To the place where it's being run and that's what we would call it.

Location shifting right. In, in the, and there's two, I think, critical challenges with each, right? In, in time shifting, you have to have some ability to be able to move the load to a different period of time. And that's where you were kind of referring out. Some is like, Hey, when I'm using a webpage, I can't really move the processing of that visit to a different time of day.

So. When you, and you have to be able to sometimes predict when that is, because you have to, a lot of times you're scheduling into, you're always scheduling into the future, but you don't always have. Long period of time to look forward and wait for a real time signal. So sometimes you have to prep your workload and predict when that is.

There's some interesting ML and AI stuff that while time invest in to predict when that period would be, so you can get ready, your load can get ready and do it at that time. And, and that's one thing that is really important for time shifting now, location shifting. It also has to be appropriate workload, but in a different sense, many compute workload.

Art require huge amounts of data to be able to read in and data shifting is really hard. It's actually probably a it's something that would block a really large scale location shifting implementation. If you had a huge large data dependency on that. So things like training machine learning algorithms are pretty hard to.

Location shift. If you're not already geo distributing your data to multiple data centers around the world. And many people are, you know, I'm sure Google in a lot of senses, geo distributing their data to many data centers around the world. And then they shift to, they could, I don't know what they're doing.

They could shift their. Processing to those, you know, regional locations, which had lower carbon intensity at the appropriate period of time and, you know, say follow the sun or follow the wind around the world. As long as the data was already there.

Henry Richardson: And that doesn't speak to the political challenges of shifting data, which is like, you might have different jurisdictions, like you have specific rules and won't let the data outside of the boundaries. So they're not only technical challenges, but also geopolitical challenges. I would say.

Scott Chamberlin: Totally agree Henry. Yeah.

Asim Hussain: Every time you try and have this conversation with anybody about location. Shifting that's the, the word data sovereignty just comes up almost immediately in the conversation and it's challenging. But then again, like within large countries, like the United States, there is still a lot of variability between east coast and west coast.

And the date, I believe the United States is one data sovereignty region and the same thing can, is it not? Um, she's shaking her head, Scott.

Scott Chamberlin: A lot of the data sovereignty laws, the privacy laws are being written by the States, today as though like say Illinois has a really strong one in California is really strong. it depends on the nature of the data, whether it actually falls within that data, sovereignty law, not all data is going to, there's a lot of data that's just generic and is not tied to individual privacy stuff.

And so that certainly wouldn't apply, but when you're doing, you know, Machine learning or algorithmics, or, you know, big data processing on things that are associated with users or have data privacy policies associated with their collection and use. yeah.

there's going to be even in the United States, a lot of times per state laws, you would have to comply by.

So again, it depends on the nature of the data about exactly. You have to consider when thinking about these kinds of things, there's a lot of things like in processing, you know, batch processing, a lot of these cloud concepts there start when you, when you think of a cloud native, you know, world, right.

There's a tons of cloud concepts that are really appropriate for. Both time shift and location shifting, you know, in NLS to me, you did a lot of work on, on batch processing. You know, there's the, the work that has been done both in Google and Microsoft on, on cargo or Kubernetes, like how do you build it into the infrastructure so that if you do have appropriate data, you can start to have the data center operating in a carbon aware way and that's, that's analogous, you know, like We in windows, On the client and data center, you have similar concepts, but are more operating on those cloud data workloads, which are very different than what the client workloads are like.

So.

Henry Richardson: We focused on a lot on the challenges. We were just surprised by seeing, by seeing how many people are actually figuring out how to navigate a lot of those. Like maybe they identified instead of, because data can be so. Maybe they identify two data centers that are in different regions and just have local copies of both of those so that they can pick when they train.

So they're not picking amongst the entire set of data centers, but a specific set or like the windows opportunity. I, would've never thought of updates as an opportunity for flexibility, but it's a huge, like you have up to a week of flexibility. Whereas a lot of the conversations we have are like, we need this job to be done by the morning.

We only have. 12 hours of flexibility, but so the more flexibility you have, the more savings potential you can achieve. So I think we talked to a lot of creative engineers who have identified opportunities within their very specific software to figure out how to make it.

Asim Hussain: And I think that's one of the exciting things about this space is that there's just a lack of knowledge. And this is kinda one of the things I believe for a while is if you pass on this knowledge to people, I mean, hopefully some people listen to this podcast now, or then have an idea regarding some aspects of their workload or something that they can maybe explore with, with carbon where computing one of the things I've always, there's always been a lot of interests.

You know, one of the things we do in the green software like movement is we, as you look at kind of various, you know, as you know, various touchpoints to reduce the emissions of software and carbon awareness is just one of them. There's always been a lot of interests. The interest comes from the fact that relative to the investment, the return is quite high.

It's not, this is not going to be the solution, the one solution and organization adopts to, to, to reduce all of their emissions. But relative to the investment you've got to put in, you know, the return is quite high. I think I've seen those even that there was a paper recently. I'm not too sure. Much adheres to what I've heard from people I know in this space, but it's talks about an upwards of up to about 30% emissions reductions from workloads.

Although I've heard kind of at a top start more about 10%. I don't know how kind of, what are your, have you, have you guys heard anything about this, about the potential improvements from, from the.

Scott Chamberlin: I think, well, I can't really reference the Microsoft saving specifically. I don't think they've released that information. I think what you can say is that it's highly dependent on the parameters of your problem. You're trying to solve like Kendra is referencing in terms of the amount of timeframe you have to be able to shift, or the amount of locations you have to choose from, for shifting and what the marginal emissions are in those locations.

Right? Like shifting from you know, coal to natural gas might have a certain percentage opportunities shifting from coal to. You know, a hundred percent renewable, like wind or solar is going to have a much different, and if you can completely shift or partially shift, you're going to have a bunch of different stuff.

I think almost every implementation is going to have a different upper bound at what the savings is and, you know, getting good at measuring that and identifying what that is, I think is if you were to like break down, if somebody's thinking. Building some time shifting or some locations shifting carbon awareness into their application.

Certainly in the number, one thing that any user, any developer would need to do is model their potential parameters that they're going to be con they're going to constrain their problem. You know, come up with some estimates. Like if I can S if I can move 5% of my workload within 24 hours in DC location.

I have this much potential savings. And then going back to ask them your previous point about the potential, like the cost versus benefit and the modeling, the work like there's, it costs this much development work to be able to do that versus this much savings and any developers are going to need to do.

Like modeling and that estimation before they go forward with an implementation, because I can certainly think of many problems, which might not benefit greatly, especially if they already are very small and the amount of emissions they're generating and it might not be worth the implementation for that.

You might focus on other things, but there's certainly ones that are, you know, generating a lot of missions. The, the attributes necessary, like the flexibility and time or location, the.

data dependency stuff we've already discussed. There's certainly problems that could. Great benefits in terms of implementation time shifting.

But again, the prereq for all of this is To model that out, understand what that potential is before implementing. And then if I can, I want to talk about one of the things you mentioned asked them in terms of the cost is a benefit. I think you're totally right. Like I, you know, in other places as some, I'm just, you haven't mentioned here, but you talk about software sustainability.

The first, if you were to like, create a. A classification of software sustainable, and you've done this previously and I've seen it. And the first branch and that is, you know, making carbon efficient applications. And then the other branch is making carbon aware applications. Right. And so those of us who are new to software sustainability might think of.

Efficient applications in the past, we might've talked about this performance engineering or improving the efficiency of your algorithms or stuff like that. Almost all of that is really hard. If it was easy, people have Buddha probably have done it already. Right. The nice thing about carbon awareness is that it's a different.

Way of thinking about your algorithms that are already running and it doesn't necessarily require you to reengineer your algorithms or to change the underlying. Implementation of your software. You're instead changing the scheduling about how that, how those underlying things work. And yes, I totally agree with you.

Like from a concert has been at Fort point of view in a lot of cases, the low-hanging fruit is in carbon awareness and software that I've seen.

Henry Richardson: To kind of build on that. Once you've identified a piece of software that could have flexibility, both spatially or temporally there kind of tend to be two big factors that drive the potential one is how variable is the, is the location that you're in. So is there a lot of variability in the admissions rate?

And can you take advantage of that with your flexibility and then comparing across regions? But then the second piece is how capable are you a forecasting that variability, because then, you know, can you take advantage of that variability by scheduling it? So do you have 24, 72 hour week long forecasts that you can begin to say, how well does that forecast match?

What's actually going on and can I take, can I use that forecast actually think about when to schedule? So the first step is really saying, what software do I have that can take advantage of flexibility? And then the next section is. Once I had that flexibility, is there an opportunity to actually reduce emissions with that flexibility?

So you need both of those pieces to really, to be successful.

Scott Chamberlin: Yeah. And Henry, so, you know Wattime provides a forecast, correct me if I'm wrong, it's up to about 24 hours. Right. And

Henry Richardson: We just extended it to 72, but yes.

Scott Chamberlin: great. And do you have any stats that say how they speak to the accuracy of the forecast over certain periods of time and. Time parents start to be really unpredictable and it is, it like correlated with whether it is correlated with a bunch of other stuff.

That's becomes more unpredictable. The further out you look, you look

Henry Richardson: Yeah, that's a, that's a really interesting question. We we've shifted away from an accuracy metric towards an efficacy. So, if you were to shift based on this signal or this forecast, the signal, how effective are you at reducing emissions?

Scott Chamberlin: okay.

Henry Richardson: And so if we get the magnitude a little bit wrong, but we get the rank order or the, the, the, the time, right.

That's much more important than the absolute magnitude, but that's just a training trick that we use on our backend, but it kind of can be represented as accuracy as well. But to answer your more, the deeper question of like, what characteristics do we see? We see that like solar dominated regions tend to.

Slightly easier to predict because solar is much more reliable when the sun's up. And if you have cloud cover, you'll be reasonably okay. When can be much more unpredictable. So wind dominated regions tend to have variability. That's hard to detect far ahead, but we might be say, we think that this hour is likely to have curtailment.

We might not know the exact five minute period when we're throwing away wind, but generally we can shift load to. The periods that are much more likely to see that high variability or low emissions period.

Asim Hussain: You just used the word curtailment and I feel like we need to educate people as to that's what that magic.

Henry Richardson: Yes. It's a very jargony, I apologize, but in the industry we often refer to when we throw away wind and solar, because there's an excess of it or where there's not enough capacity in the transmission system to, to move that wind or solar to other places as curtailment. And we're starting to see. Certain grids throw away wind and solar at kind of pretty prodigious rates.

California throws away quite a bit of it in the spring because there's an oversupply of solar because the sun is shining, but it's high temperatures haven't arrived yet. So we're not running air conditioning also in the great Plains. There can be a lot of wind at night, but low load periods. So there's an excess of wind, I believe even in the Pacific Northwest occasion.

In the spring that the same low load situation, when there's lots of wind and solar, they'll actually spill hydro over the dams and not generate with it because they have to release it. So you can see how, like, if we can take advantage of these opportunities through load flexibility with software, that's, that's an amazing opportunity.

We also talk about devices often, too. So smart devices, EVs, that type of thing can anything that has load flexibility. We're very focused on software in this conversation, but you can see how it could be other things.

Scott Chamberlin: Right. And again, I think I was thinking about that very concept Henry, in terms of, you know, you have to think about software in a very broad sense. When you talk about the total opportunity here for only talking about PC software, the total opportunity is, you know, going to be limited by the number of.

PC's in the world as, you know, windows devices, Mac devices, you know, and throw in, obviously, you know, the mobile devices in the world, which kind of sip power, but, you know, we all need to think about the broader definition of software it's software running, you know, in our thermostats, even though it's driving, you know, Both, you know, energy, if you're, uh, you know, electric heat or electric heat pumps and stuff like that, or natural gas use, which doesn't have the same benefit of the time shifting, but it's software, that's running like, you know, you, my robot vacuum cleaner, that's sitting right here.

It's software, that's running, you know, almost everything, you know, Future how homes and businesses are being controlled by software and have differing abilities to take advantage of the topics of carbon awareness that we're talking about. Right. And so, you know, the IOT space is huge relative to the PC space.

When we typically think about when we think about software or the cloud space, when we think about the Microsoft software, but those are all software developers and they all. It a lot of sense have connected, you know, internet connections and can take advantage of some of these signals. And, you know, I think another area that, you know, we, we talked about, I don't have the ability to talk about too much, but we need to think about what, how does this take advantage of disconnected environments?

Not every, not every phone is connected at all times to the internet. And can you still do carbon awareness when you're disconnected? I think there's a lot of work to be done there. A huge impact or not. I don't know, but some of the modeling and stuff like that, it has seasonal variability as well, which might be able to be built in as a baseline.

If you don't have, you know, rich live internet connections on at all times.

Henry Richardson: We've had some conversations with our partners. Like what about fall back schedules or. That's still to still do some scheduling that can maybe not perfectly identify that variability, but can still take advantage of some of the grid emissions, variability. And we've also had conversations around like how frequently should that device be connecting to the, to the kind of grid emission signal and making decisions based on that, because that has a, a carbon penalty to it as well.

Because every time you reach out, run the computation to decide when to schedule load, there's a, there's a carbon cost to that. And are you, you have to make it. The flexibility is achieving savings greater than the cost.

Scott Chamberlin: Yeah, we ran into and out of a feature, another feature, I won't name the feature, but we did investigate a machine learning, you know, approach to reducing, you know, matter of power, use a windows. It turned out in that case. The amount of processing and power used for the processing to run the algorithm was greater than the potential savings for it.

So, yeah, you're touching on a really great point. Then that goes back to the point I was trying to make in terms of, you got to model all of this, but yeah, Henry, you totally got a model. The. New stuff you're writing as well to make sure you're not, you know, stripping out all the potential savings by the new code.

You're going to start running here. And hopefully you're looking at loads that are large enough that, you know, the amount of algorithmics and. You know, connections and, And services you need to ride to do time awareness is probably going to be much, much smaller for an appropriate workload that you're looking at.

But again, that's where the modeling and the measurement is super important to be in with.

Henry Richardson: And we've seen people scale that level. Like if it's a small workload, they'll just pull the forecast once, make a decision and then not check it. Or they can even do it every like three hours instead of every 15 minutes or something like that. So there's lots of ways of like adjusting the workload to the, to the job.

Asim Hussain: One thing I wanted to cover. I think, I think it's quite interesting to also cover the future because one of the things I think Henry you've mentioned to me and I think is quite important, is that everything that we're doing today, like if you talk about modeling something today, That's today's impact yet.

The world is actually when we're moving towards a future where more and more of the energy is coming from renewables. And therefore the impacts. If you were to build something today where your carbon, where workload has an impact of 10% in five years’ time, it might have an impact of 20% because the world's becoming kind of a lot more variable.

I mean, do you have any estimates of how that's, how that's gonna go in the future?

Henry Richardson: We definitely are seeing an acceleration of renewable deployment, which is increasing the variability of grids. I mean, historically. The electric utility is balancing authorities. Grid operators have always matched generation to demand. And I think we're shifting into a paradigm where we're going to have to be matching more of the demand to the variable generation coming from wind and solar.

And so as that variability increases, we're just seeing kind of dramatic increases in curtailment. Renewable deployment that enabled just much greater savings. You're kind of shifting from a world where you're occasionally trying to pick up that excess renewables to a world where you're trying to avoid the peaking fossil plants, which is just a much greater opportunity from an emission savings span, where you just move a load as far away from the peaks, instead of trying to find those trucks.

Asim Hussain: Can we just dig into that for a second? Cause I think that's quite, that's quite interesting because that's almost the opposite of curtailment because you, are you talking about peaker plants? They're so.

Henry Richardson: Exactly. So you could have a hundred percent renewable all the time, except for occasional periods where they have to turn on those really dirty peaker plants, whether they're fossil oil, fossil oil, fossil gas, or fossil coal, as you just want to avoid those periods at all costs. Instead of right now, we're saying that seeing occasional periods.

Where we're throwing away renewables, and you want to move as much load into those. So it's kind of like this expansion of opportunity, which is really.

Asim Hussain: Because because those, my understanding those peaker plants is, you know, the grids need the capability. The energy very, very, very fast. And they tend to be natural gas don't or some sort of gas. Cause you can just burn that quickly interrupted workloads, I think is what it is. It's interrupting. So not running something for five minutes could be as valuable as shifting your workload to another hour because you're avoiding the worst emissions.

Henry Richardson: Absolutely. And of course you want longer periods, the more flexible. Great of that opportunity. So there could be a two hour period in the afternoon where they have to bring those peakers on. And if you can avoid that, that can be really good.

Scott Chamberlin: But, but talking to this variability, one question that comes up pretty frequently is the nature of different grids and the makeup. And especially when it comes to grids that are. Are more towards the grids. We're going to need say 30 years from now, which a lot of times, you know, has the peak and Henry you're more of an expert on this.

So, correct me if I'm wrong that the peak loads are going to be handled by in some of these. Nuclear. I think I take long-term ideally in some cases, and then the base load is going to be, you know, renewables for the most part. And so in those grids, you know, I, when I think of a couple regions today that we deal with Iceland, you know, typically we treat it mostly a hundred percent renewables.

France is another one because it has high investment nuclear. We kind of treat this hundred percent. Carp R zero carbon region. And again, Henry, you have to correct me. I'm wrong, those two regions, but it's because they have this nuclear renewable, you know, not the nuclear in Iceland, but in France, high Metro renewable.

And in that case, the carbon awareness is just kind of a flat clean signal. And so as the grid evolves to these things that are like zero carbon grids, like the techniques we're talking about, they don't have as much impact.

Henry Richardson: So, this is a really interesting disconnect that we're seeing right now. And especially in the near future load, flexibility will have a lot of emissions savings potential because it'll be able to shift out of those dirty periods into the curtailment periods. But once we eventually attain those a hundred percent or near a hundred percent clean grids, the flexibility won't be saving emissions directly, but it will be enabling a hundred percent clean grid because will be following wind and.

And if we didn't have that flexibility, we would have to be fossil resources. So like it's an essential piece of a clean future grid, but it's going to be harder to quantify the benefit of it in that view.

Scott Chamberlin: That's a great way of putting it Henry it's it's we get to clean grids faster. The more we have carbon awareness because carbon Alinea, or this allows us to maximize the use of our renewables. Whereas today we're already curtailing them. Right. I think that's, that's an excellent way of putting that.

Henry Richardson: Exactly. And so it's, it's like a critical piece of that future grid without it. We wouldn't be able to obtain it as quickly as efficiently as, as chief.

Asim Hussain: It makes me realize I had like a, an epiphany moment a year or so ago when I, I realized the way this is interesting in computing, but just generally the way we consume electricity. Is based upon the way the energy grid was created. Like a lot of other things we do in our life, we flex based off of what's going outside in the world.

Like, I, I don't try and grow plants in winter and my garden because it's cold. Right? So we, we normally have this thing where we flex and we change what we do based off of the natural cycles of the earth. But because we've just been, had this thing called coal, which you could just burn whenever we wanted, we've not had to have.

That pressure in the rest of our world. And what renewables is bringing into the world right now is like, well look you, you can actually do I find it? I actually find it quite beautiful because oftentimes we're way disconnected from nature. And one way I think about carbon awareness is actually I'm building software, which responds to natural cycles of the earth.

And it connects me with nature in a, in a, in a kind of abstract, an indirect way, but it's one of the few ways you can connect with nature. I think in software,

Henry Richardson: It's a really good, interesting point about the electric grid in that it's very unique in the sense that it needs to be balanced instantaneously at all times. There's no flexibility in terms of timing. So if there's a demand on the electric grid that has to be met immediately, you can grow rice and store it in a silent.

Or it's an, a grain elevator for a while and then release it. The electric grid has to be instantaneous and until we have a lot more storage or pumped hydro, we're not going to have that flexibility. And so demand has to kind of follow supply much more closely.

Scott Chamberlin: I think it was in California, right. Where they're starting to look at in terms of increasing storage by starting to enable all this, the growing amounts of EVs in the world. Connecting those batteries and utilizing them as a bunch of local storage. So we have this like future that's way more complex in the sense that, you know, we're, we're drastically increasing the amount of storage.

Sometimes it's centralized storage. Sometimes it's this municipal storage. That's like, you know, you're, you're , and it's just can be a local buffer, not only for your house, but also for your driving. And you could charge that in a carbon aware way. And then. You know, we're adding municipal, solar, ratting, you know, utility, solar, adding all this.

And I think, you know, the grid gets way more complex, but like you're saying, ask them, I think it all slowly, you know, you get these new patterns, these new natural patterns that start to arise out of it. Yeah. The technology is going to play a really key role in all of this, how to, how to implement.

Henry Richardson: I'll toss another bit of jargon in here, V2G vehicle degrades. So there's both smart charging of your vehicle, but there's also can that vehicle actually push power back to the grid at important periods? Whew. It's all software driven, even though it's living on hardware. One other question for you.

Awesome. Cause I, I think we're approaching the end here, which is. One of the things that we're excited about, this is that it puts the capabilities in the hands of developers to actually affect this. This is something that a developer can make a decision about actually make, may affect the software that they're working on and have a real emissions effect.

Is that something that you've explored with?

Asim Hussain: Oh, you talking about some of the projects that are working in the foundation, because we have one particular project, which is a carbon aware of soft SDK, the software development kit, which is a lot of what we're describing here. I mean, the logic is the same. Every single company wants to implement carbon awareness is pretty much just creating the same logic.

And so one of our projects is create the carbon where software development care, which is gonna enable people to enable developers to much more easily implement some of this logic and functionality. And I remember when I was having, we were having conversations with our team and kind of the same thing came up, which was I w w we started this with this idea that we're going to reduce the carbon emissions of software, but actually software drives everything.

So this SDK could be used. For, I have an E vehicle in my driveway. So it could, I could build something to leverage it and charge my Evie based off of, you know, some sort of signal or heating my house or something like that, because I think that's, that's one of the things we're seeing is if we can just make it easier for people to do this, then the more likely to, to, to implement a lot of these things.

And I'm seeing it, I'm seeing it implemented another place as well as lots of websites that, and I love the impact might not be so good. In terms of carbon savings, but impact's quite high in terms of making people aware of what the potential is. There's a really great magazine called Branch Magazine from one of our colleagues here, Chris, Chris Adams, and you know, the it's an online web magazine, which changes its behavior based off of electricity grids.

And so the images will disappear if it's, if it's high carbon intensity and replace it with text and words. And that's really good because everybody reading that is suddenly then aware of this entire concept, because people aren't aware of it in the.

Henry Richardson: One of the pieces that I think I really like about it too, is that a lot of this is being driven by the developers themselves. Not necessarily the sustainability to. Like the corporate sustainability team at these organizations, they're like the developers see an opportunity. They understand how the code works and can actually make a decision about how to drive emissions or that.

Scott Chamberlin: I a hundred percent agree with you. I mean, some of that is natural in the sense that the. Inability to step in. Most corporations are driven a lot of times to the supply chain organizations. And a lot of times that's because the measured CO2 impacts a lot of the majority of is sometimes coming from.

Supply chain, but I think the opportunity in terms of cost versus benefit on the software side, I think is it's, it's an area where we can change faster and have some initial impacts greater than some of the supply chain teams, which supply chain changes, which are our longer term kind of things. And to be clear, all of those, both sides of the thing are totally interleaved.

There's not a fine line between them.

Asim Hussain: So I think that's all we've got time for today. So it was really wonderful conversation that we want. And just a final thought for me. I just want to give a shout out to an event that's happening in the middle of June. So the foundation has a summit, a global summit, which has been run over 20 locations around the world.

If you want to meet other like-minded people, people like us kind of thinking and talking about these topics. You know, come find us a confined your local event at summit.greensoftware.foundation.

Scott Chamberlin: I think the final thought is that, you know, having gone and tried to build carbon aware software, it was as long as you're making sure that you have the ability to measure and that you are. Actually doing the engineering that is going to have an impact it's actually super motivating to look at. And it's actually the technologies.

It's a rich, rich area of technology. And it's may seem intimidating when we add these new kind of terms about to, about how the grid operates. You have to think about a, yet another thing is software, but it's, it's really, once you get in there, the concepts are pretty straightforward and adjusting yourself.

Do this kind of stuff is actually not too hard. So I'd encourage folks to, to try it out at least.

Henry Richardson: I think one of the things. About, as people are coming up with use cases that we never thought about. So scheduling windows updates, like we hadn't even considered that as a possibility. So people come up with very creative ideas, but with region shifting location, shifting that, that we would never have thought of.

And so we're always excited to see that kind of expanding possibilities for loaded flexing.

Asim Hussain: Thanks for listening to Environment Variables, all the resources for this podcast, including links to our guests and more about carbon where computing as well as the Green Software Foundation. The show description below. We hope you enjoyed the show and see you on the next one. Hey everyone. Thanks for listening.

Just a reminder to follow environment variables on apple podcasts, Spotify, Google podcasts, or wherever you get your podcasts. And please do leave a rating and review. If you like what we're doing, it helps other people discover the show. And of course we want more listeners to find out more about the Green Software Foundation, please visit green software.foundation.

Thanks again and see you in the next episode.
[END]

Amazon's Customer Carbon Footprint Tool

11 april 2022 | 45 min

In this episode Asim is joined by guests Chris Adams, Sara Bergman and Danielle Erickson and they discuss the impact that Amazon’s Customer Carbon Footprint Tool is having on the green software landscape. How do services like AWS affect climate change and what are the effects on the environment of these huge data centres? We also learn about how you can use heat from greenhouses to grow tomatoes!

Learn more about our guests:

Chris Adams: LinkedIn / GitHub / Website
Sara Bergman: LinkedIn / Twitter
Danielle Erickson: LinkedIn

Episode resources:

The Green Software Foundation Website
Sign up to the Green Software Foundation Newsletter

Person: Adrian Cockcroft
Blog: David Mytton
Blog: SRE Methods & Climate Change - Benoit Petit
Presentation: Green Cloud Triangle
Magazine: Branch Magazine

If you enjoyed this episode then please either:

Follow, rate, and review on Apple Podcasts
Follow and rate on Spotify
Connect with us on Twitter, Github and LinkedIn!

Transcript below:
[background music]

Danielle Erickson: We're looking at the AWS tool, the Google tool, and the Microsoft tool and understanding the broader strategy, so when you combine those two things, I think that's what we have to do right now to strategize in the best way to reduce our emissions.

Asim Hussain: Hello, and welcome to Environment Variables brought to you by the Green Software Foundation. In each episode, we discuss the latest news and events surrounding Green Software. On our show, you can expect candid conversations with top experts in their field who have a passion for how to reduce the greenhouse gas emissions of software. I'm your host, Asim Hussain. Welcome to Environment Variables, our new podcast. My name is Asim Hussain. I'm the executive director of the Green Software Foundation.

Chris Adams: Hi, there. My name is Chris Adams. I am one of the directors of the Green Web Foundation.

Asim: The Green Web Foundation, the Green Software Foundation, there's a story behind that.

Chris: There is indeed.

Danielle: Hi. I'm Danielle Erickson. I am the product manager of the Cleantech service line at Thoughtworks and the product manager of an open-source tool we created called Cloud Carbon Footprint.

Sara Bergman: Hi. My name is Sara Bergman. I am a software engineer at Microsoft and I'm also the chair of the writer's project in the Green Software Foundation.

Asim: All right. I think we're here today to at least start off talking about Amazon's Customer Carbon footprint tool that they announced recently. This is something I saw they announced in re:Invent which if I remember right, was it November or December in last year, in 2021? They announced it in November or December and it's a method of measuring the carbon emissions of customers' workloads on Amazon.

I don't know how all you all feel about it but I was really impressed with just the speed with which they made their second announcement, which was just, when was it? Weeks ago now or maybe February, they then announced it in preview. It was incredible breakneck speed. I was expecting from their re:Invent to wait a year for them to publish something. For me, that was a really, I felt like I had a whiplash. I was like, "Whoa." Has anybody else had a chance to look at their announcement?

Chris: I have. I think it's really cool. I'm really, really glad that something like this exists because if you consider yourself a responsible technologist, I figure the electricity has to come from somewhere, and being able to have this information allows you to optimize for carbon and given the information that is shared with us increasingly in the news, it's worthwhile actually referring to this. For example, the WPC, so the IPCC report explicitly mentions digital and the role we have to play now in that. Having the tools to instrument that is really, really handy. We actually used to work on something like this. A couple of years ago, we built a tool called Amazon Green Cost Explorer, which basically used some of the billing API to work out which regions were green and which ones were not green, so you can act on this. To actually have something much more fine and green is really, really cool, but it's not the only project going in this field, I suppose. It might be worth talking about that, yes, there are other ways that you can understand the environmental impact if you compute than just using this, for example, even though it's very, very useful and very, very welcome.

Sara: Yes. I think that's a really good point. I think, for example, if you think about performance, some people are very interested in getting into the nitty-gritty. I want to read the logs, I want to really get down to it but others are just I want to see my latency. I think the same goes for carbon, right? For some, this will be revolutionary and this will be a great way to get any kind of insight but there are others who are ahead of the pack where maybe the granularity isn't enough or isn't timely enough or whatever but this isn't a really good first step in my opinion.

Danielle: This is something that customers have been asking for for a while, so to see Amazon respond in this way and give some visibility that I believe is generally available, this tool. Anyone who uses AWS can see their carbon emissions over time is really incredible. It's a great step in the right direction and as Chris already said, there is a lot more regulatory pressure to be reporting on this. Everyone's going to need this and I think if Amazon can make this first step, we can hope that in the future, they'll continue to respond to this demand and this need the customers are having, so really excited about this.

Asim: This is my mistake as well. It's generally available, is it now, Danielle? I just assumed it was a preview.

Danielle: That was my understanding, but Sara, you may know better?

Sara: No, I heard the same. It sounded like it was for everyone who uses their services.

Asim: Yes, it's done.

Danielle: Yes.

Asim: Wow. That is very impressive to go. I mean, it means they were working on this for a long time, I think. You don't come up with a tool like this in two, three months. This has been something they've been working on for a while under the covers, I imagine.

Chris: I think you're right, Asim. I mean, if you've been following this, Amazon have been hiring for sustainability specialists for the last two or three years. Also, if you look at the VP of cloud at the moment, I think, or one of the VP of Sustainability, Adrian Cockcroft, he's been speaking about this for a very long time. If you followed him on Twitter, he's actually had a lot to say about this even in the pre-Amazon days, actually. It's really, really cool to actually see some leadership here on this. I'm quite impressed with this as well because between 2019, there was actually a talk by AWS specifically at a conference called, Map Camp where they were explicitly calling this stuff out.

They said, "Look, if you want to do this, you're going to need to tell us as customers because we're not seeing the customer demand for this." There's a slide of Mr. Cockcroft standing with a big thing behind saying "The thing you can do is move to the green region." Now the thing they've created now is something which provides a bit more visibility to that, so rather than just having that as your option, you've got ways to optimize the actual compute in place rather than having to take on what in many ways could be a risky or scary migration that you might have to weigh up against other things like feature development or the other things that product manager or a CEO might be asking for.

Asim: For me, it felt like the cherry on the top because Microsoft announced their, I think it's now called the Emissions Impact Dashboard a year ago. Then I think it was six months ago that Google announced their dashboard as well and so with Amazon coming out on it as well now, that's all three of the major clouds, all [unintelligible 00:06:39] use that major clouds now have a capability of customers, basically being able to answer questions along the lines of, 'Well, how much carbon emissions is all of my emissions?" This is where I think Thoughtworks would be leading the pack as well because I've forgotten what you call it. I keep on getting confused, Cloud Carbon Footprint. Is it?

Danielle: Yes, that's it. Yes. [laughs]

Asin: Okay. Yes. Cloud Carbon Footprint tool because you've actually been developing essentially an open-source version of this for a while. You must have some really deep insights into how do you actually go away and calculate some of these numbers?

Danielle: It's been really cool seeing the three different tools come out and right now, our team is going through an internal process of reviewing each of the different tools and understanding the variety of features that each of them has is what they have to offer. As much as we can understand how they're getting the data and the methodologies they use to calculate carbon emissions, we're trying to do so because each of them are going to give you really the best numbers you can get for each of those individual cloud providers, but one thing that they're unable to do at the moment is compare between each other.

For many organizations, the majority are multi-cloud users and if you're trying to look holistically at your sustainability strategy and your cloud emissions, you likely want to see them in one place and also using a similar methodology. If you're looking to compare, if you're looking to really optimize, take action, you'd want to compare them apples to apples, not oranges to apples. For our perspective, using all of these tools together is really the best strategy. Have a lot of tools in your pocket to understand what's going on and then begin to understand the areas you can start to make changes.

[music]

Asim: I think we've talked enough about all the glowing praise for all these three tools. We now dig into the issues with them. I think one of them, like you just touched on there is exactly how is Amazon calculating its numbers? How is Microsoft calculating its numbers? How is Google calculating its numbers? There's a lot of opaqueness because they're not revealing that. They're just revealing, 'Here's your total number."

Chris: This is one thing that I could share some light on, I suppose. There is some good news in that increasingly organizations are now talking about essentially, how they share which parts they do measure, which parts they do not measure inside this. We have established ways to track some of this stuff. There are things from the GHG protocol, which is an organization that pretty much sets some standards for this and they talk about things in terms of Scope 1, which is burning for carbon emissions on-site, Scope 2, which is electricity, and then Scope 3, which is stuff in your entire supply chain.

A lot of the time when you might look at some of this, you might have people talking about just Scope 1 and Scope 2, for example, without necessarily talking about the Scope 3 part. If you look at say, I know this is one thing that both Google and Amazon don't include in their numbers is basically the environmental impact from creating the servers in the first place. This is one thing that's probably worth talking about because well, they have to come from somewhere, and it's obviously an energy-intensive process to turn sand into silicon chips.

This is one thing that I've been quite impressed with because there actually are a few open issues on cloud carbon footprint to start piecing together some of these numbers because this is actually very much considered the next step, now that stuff is being done on the energy front. There's a really good blog post in the show notes by David Mitten, who's been writing about this. I'd really recommend his blog because he provides a really, really useful set of incisive analyses in this field.

Sara: Yes, I think that's an excellent point. Depending on the type of application that you have, the hardware emissions just from creating a server or whatever, the network devices, whatever you use can actually outweigh the pure energy cost of it. It depends, of course, on multiple factors, but it definitely can be the case. When we talk about engineering and engineering enablement, there are some pretty easy things that you can do to decrease the amount of hardware that you use, but if you're not getting measured on it, how will you be incentivized to do those actions?

If it's pure cost, well, we are very much relying on cloud providers being kind enough to give us a cost which is mapping to carbon, but that isn't necessarily true always. Right?

Asim: Yes. I think just essentially from my understanding, Microsoft's emissions dashboard gives you Scope 1, 2, and 3, so it tells you the carbon emissions of your workload, your energy consumption, just to break it down a simpler format, your energy consumption and your hardware. Google currently just gives you your energy consumption of your workload. I have actually assumed the Amazon one was all three but is it just energy again? It's just energy again, so Amazon is just energy?

Danielle: Yes, one and two.

Chris: I've read the 451 report. In the announcement, there is a report by 451, and they explain what's in the model and outside the model. They basically said, "We're not looking at embedded energy, and the actual machines themselves, and we're not looking at Scope 3 at present. We're not necessarily looking at Scope 1 because it's not quite so tight." This is primarily about the energy part and this is also why the numbers, as far as I'm aware, there is a lag, because they're looking to get the most accurate numbers, just like how Google do where they basically say, "We will move as fast as we can, but we are working with very, very large providers who might not bill on the same monthly basis. We wait until we have the information from energy providers, so we can give you an up-to-date number." This is what is actually shared to my understanding.

I've got to stress, I do not work at Amazon, so there may be much more detail that may be there that I'm not so aware of.

Danielle: I'm not exactly sure the full reason for that lag but my understanding is it's about three months, which if you're getting very accurate information, can be helpful to look back and understand over the course of the year. I do see a challenge to the consumer actually trying to make changes and use this data. How can I act on data today that's three months old? It becomes a little bit difficult to build into your workflow, to make decisions day to day based on three-month-old data. That's something to consider, I think, with this tool, and maybe something they can improve in the future.

Sara: Yes. What they're stating in their announcement is that it is the underlying billing cycle for the electrics utilities and I believe that Google is doing the same, but they are also quite late in showing. It really limits what you can use it for. It's still great for some type of comparisons. If I have two applications that are similar, which one do I continue with? Those sort of things, it's very good for, or comparing over time, but doesn't really tell you what I should do tomorrow. I think as more and more software companies move away from the waterfall and move into more and more aggressive agile- three months, no, is anyone going to be really happy with that? I don't know, maybe.

Chris: Maybe there's one thing that you can talk about here in terms of, there may be different uses for this data, for example. I know that when I've spoken to people who are looking to use things like cloud carbon footprint, they've told me that there's two main use cases that you tend to have. There might be engineers like yourself or me when we want something like an SLO for carbon, I want to be a green SRE and there's a really nice post by a guy called Benoit Petit, who is one of the lead contributors to a French project called [unintelligible 00:14:34] which basically provides per-process level energy usage information that could provide these numbers.

He talks about this stuff as an SRE saying, "Well, these are the numbers I need to basically optimize for, and I should have dashboards like this." There's some really nice work by the folks at Mapbox, who've been speaking about this for a while. They were some of the early contributors to the early green cross-explorer stuff for this where they were talking about, "Well, if we review our bills on a weekly basis, and we use that to shape our usage, it'd be awesome if we could do this for carbon because we're already good at optimizing for some kind of metrics, so it would be really nice to have something that."

Increasingly, we do actually see numbers like that now. There are schemes which do make this stuff possible. Just last week, for example, there was a new standard which has been proposed called granular certificates by a number of organizations. This gives you hourly settlement for this stuff, which is really, really, really impressive. This isn't that well known yet, but this is the kind of stuff that what the future looks like in my view. I look forward to the time where this is actually a thing that you can optimize for as an engineer, and you can see on a dashboard, for example, for your team.

Asim: I think really, the issue here is that we want to celebrate this work on the work that we're doing, but it's not quite there as a dashboard, that from an engineering perspective, teams can use to actually give them information to make decisions. That's basically the challenge that we've all got. As we're sitting there and we've got options between one, two, and three different architectural types or different choices. This doesn't quite give you that level of granular-- Regardless of even the methodological differences between the different platforms, even the granularity won't give you that.

I can't speak for how Amazon does it. I do have some experience for how the Microsoft dashboard works, and it is very averaged out data. Multiple servers will always report the same energy consumption regardless of what you do because that's just how it's been calculated. That works great the thing you're talking about what is it used for? They're designed for reporting purposes. They're designed so organizations can calculate and report their carbon emissions to CDP or perhaps have an understanding regarding what are the offsets or neutralization strategies we need to employ. That's just what they're designed for. It's not built for engineers with the caveat of-- I think Google is on an interesting track.

Sara: Yes. If you think like a person in an individual team with a small portfolio, then I completely agree, but maybe if your step up, so if you're someone who is responsible for a larger portfolio of services, then suddenly this means you're able to compare them. Sure the data is older, but I can then start to evaluate, "Okay, how much value is this service provided me compared to service B, and how much is their carbon footprint."

If one is vastly higher, but providing me less business value, then that's a decision on a leadership or a planning level that you can take that these dashboards enable that you would not have been able to reach without this. It really depends on what kind of decisions you're trying to make based on this dashboard.

Danielle: Yes, this is something that we thought about a lot when building the open-source Cloud Carbon Footprint tool. Our perspective has been trying to reach that engineer level, that day-to-day decision-making level with as much granularity as we can build in and as much real-time as we can try to make the tool, taking billing data immediately and turning it into carbon emission estimations. Not to repeat myself, but I think the benefit of having multiple options here is that you can combine them for these different uses that you have.

Your engineers can look at both tools, combine the data that they're seeing from the Cloud Carbon Footprint on a day-to-day basis, and then talk to their infrastructure leads who are looking at the AWS tool, the Google tool, and the Microsoft tool, and understanding the broader strategy. When you combine those two things, I think that's what we have to do right now to strategize in the best way to reduce our emissions.

Asim: Especially the fact that because Cloud Carbon Footprint is open source, not only is your methodology public, but your data and the underlying data assumptions are very low, granular level are public. I can see what is the energy-- If I'm using this particular server, this particular load that data is public. We're actually using that in the foundation in the software carbon intensity standard, where you're leveraging that data because it helps engineers calculate the carbon emissions of processes or estimate the carbon emissions of processes, so they can then make those kinds of decisions.

It's the openness of the data is, I think, also missing with these tools. I've also heard it's extremely difficult for Amazon and Google and Microsoft to make this data public. It's not only they're revealing competitive information, there might also be legal constraints. If you reveal some of this information, the SEC might come after you because you're revealing proprietary information. There's actually lots of complications around that, from what I've heard. I wonder if others have thoughts on that, on the openness of data.

Chris: I can actually weigh in a bit on this, which might be of use because you have a similar thing happening in the energy sector just the one layer below right now. One thing that we've seen pushback from people who run the energy grids in various places, they've typically said, "We are not able to share information about how congested various parts of the big transmission wires that move power around because we see that as a security risk." but this is actually a thing that we've heard in lots and lots of places. In many cases, a lot of the time you could see there's a trend towards opened for a bunch of this stuff.

I feel like a lot of the time, if you're not designed or if you aren't used to sharing things open by default then you can come up with a lot of- it's understandable that you might not want to share a bunch of this stuff. There will be cases where you might not want to share this for very valid reasons. For example, there are probably valid reasons for not listing where geographically every single data center might actually be. Even if this may be information that as a customer you might want know if you want to understand what climate risk is associated with all the machines running in a particular place.

Especially when we refer to examples like say- a most recent risk example might be the big Facebook data center, the big data center from Meta and Zeewolde in the Netherlands. That's eight meters below sea level, that gigantic data center. That's the thing you might want to know about in a world of rising sea levels. That's some of the stuff which is useful to know about, but going back to the original point. not everyone's ready to share information on a very, very open basis just yet, but I suspect that over time this will have to come up because well, environmental factors will increasingly push this and necessitate this kind of disclosure.

This is actually one of the things that's been driving a lot of this stuff right now. It's because investors are basically saying, "I need to understand the disclosure in my supply chain." or "I'm invested in you as a company. I need you to share this information so I can end up with a net-zero portfolio. If you don't have these numbers, it's going to be very, very hard for you to share that." In many cases, organizations will basically say, 'Well, I'm not going to invest in you. I'm going to invest in someone else. At least I know whether risk is there." We're not open yet, but the more often we do get, the easier it is to make data-informed decisions as we move forward into this changing climate world.

Sara: Now, I see the same security issues for hardware as well. Do you want to state exactly what type kind of servers are on your server [unintelligible 00:22:25] floors? Maybe not because there has been hardware security incidents in the past, I'm sure we'll see them in the future. Then you might not want to say exactly what you have, but there can also be an argument for finding what is a valid enough proxy that you don't state explicitly that this is this type of server, but exactly this hardware. I built it like this. I specify carbon cost or some other tangible number that gives you the information that you need without being a security risk.

That is, of course, a lot of work especially if we think across all cloud providers even if your company is your own cloud provider while being on Preem, you would want to be able to compare across the stack. The lining on that without being open is difficult. That we're going to guess what our competitors use. I don't think that's a good approach. It's quite exciting from an engineering perspective, just the complexity of some of those things.

Asim: That's a really good point that you mentioned about what data can you reveal? Because this is what we're talking about with the software carbon intensity specification in the foundation is what we want. We're talking at one level about give us all the data, but really, why do we want this data? We're actually trying to calculate our carbon emissions. Well, what really would be quite useful is just the carbon intensity. It's like this server, I don't necessarily care what the components are. I want to know how much carbon per CPU, per minute of this [unintelligible 00:23:55].

I want to have that data, and if I have that kind of data, that's actually probably all that I need from an engineering perspective. That's probably all that I need in order to make decisions. It'll be a wonderful world in the future where everybody is essentially giving you this data is what is the carbon intensity of my service? What is the carbon intensity of this streaming service we're using right now? What's the carbon per minute? That's all I really need.

Chris: It might be worth looking at some work that's happening in the web world that I've seen. There are tools like website carbon and increasingly there are tools that plug into analytics like Google Analytics to give you an idea of what the environmental footprint of some digital services over time might actually be. One thing that I've seen in the web world right now is this real push for having carbon budgets for websites. One company, Wholegrain Digital, they basically say, 'No website that we build will cost more than two grams of CO2 emissions per page load. That sounds really silly on a per-page load basis. Some websites get quite a lot of page loads, so over time this stuff adds up.

If we just zoom out for a second and think in the outside world, there is a huge amount of science saying, "Yes, we have a budget we need to stay inside." If you look at the energy sector, they themselves have a carbon budget that they have agreed to stay inside which is why you have massive compliance markets. It makes sense. Probably we would also need to have something like that as well if we want to stay inside, like I guess the dictates of science. We don't get to change the physics of climate change but we can at least change some of the economics around climate change.

We can at least do something around this so we can optimize for carbon as developers, so when we're building services, they tread more lightly or as lightly as possible given the various other requirements would be nice to meet as professionals, I suppose.

Asim: It's interesting that you mentioned the carbon budget as two grams per page load, because that's an intensity, not a total. I think that's the thing that I talk about a lot that total budgets are really challenging in our world because there's just- how can you set a one-ton budget for our website? You have no idea how many users are going to land on it, but an intensity is, "Oh, [unintelligible 00:26:10]. '"

Chris: I'm not so sure about that.

Asim: Oh, interesting. Let's go.

Chris: Here's the thing, Let's say you're going to go with this. You have a $2,000 budget or something like that because you know that you're probably going to get this many page views over a given time. This is the thing that you're seeing in procurement and contracting these days. They're basically saying, "Well, we have been given legally binding targets that we need to reduce our emissions by 5% a year, a year between now and 2030. That's it. We don't get to do not do this now it's in the law. If they have that, then they're going to have to say, 'Well, we're going to spend €100 million, £100 million on this contract for the next two or three years. We basically have an implicit budget that we need to stay inside."

You do have something like that now. It may be the case that okay, having just one number over three years isn't very useful. You might want to have some smaller timeframes or something like that. This is why it might be useful to have a rate for this because you can say, "Well, given that I have this, I now have something I can act upon. I can either change the size of a page for example or I can change the intensity of the electricity so that's going to allow me to stay inside it. It gives me more options." I think it's useful to have the total number because this is essentially what's driving things from a science and regulation point of view. As a developer you might not be able to use it on a daily basis.

If you have CI for example, you're going to want to have a unit because that's what you're used to using for your score from say, Google's web vitals. A web vital score is going to be a rate that you can refer to or something you can look at. It's not going to say over six months. It's a kind of volume basically. I think you need both, Asim, not just one, but it's very, very useful to have the ratio, absolutely.

Sara: I think tools like the tool we've been talking about today, Amazon's new tool, that can give you that from an OKR perspective because you can see, "Okay, what's my cost? What's my page views? I do the simple division and I do get these numbers, but once again, it's for reporting purposes. If you've never reported on it, this is better than not reporting on it for sure. Totals are also interesting because you can go to a rate assuming you have the other end of the fraction, but sometimes you want to go the other way. That can be a lot trickier without the really granular data.

Chris: Sarah said something really interesting here about going both ways and about if you've got a total number, you can go down from there. I've mentioned Wholegrain Digital before because I'm a really big fan of what they do and EcoPing is another group that do this stuff as well as Mightybytes who built a tool called Ecograder almost 10 years ago where they were tracking this kind of stuff. The model that is used, they call it the sustainable web design model that is basically based on a global figure for all the initial used by all the entire internet tech sector divided by all the data transfer that is facilitated by this.

This is a bit of a course figure, but at least gives you something that you can act upon and work with. This is actually one thing that I think it's going to be live next week, as being able to use some of the tools that you do use if you build websites and have things like that. It's useful to have those kind of stuff. In many cases, you need to understand what the model is actually representing to see what's going into that, for example. In the example of CO2-GS, for example, this is using network transfer as a proxy to talk about things like, say, usage at a device level or usage at a data center level.

Once again, without having access to the open models, it's very difficult to know where your interventions are going to make a meaningful difference. This is why I'm actually quite happy that things like Cloud Carbon Footprint are open enough and are accepting [unintelligible 00:30:07] requests. You can basically say, "Well, this is what I think is going on. This is what I'm trying to do in good faith to reduce the emissions of whatever service I'm building."

Sara: Yes. I think we should also maybe mention that the granularity only matters if you have an application or service that changes rapidly. Not every software does that right, that we have stable software that's in a maintenance mode or for whatever reason, isn't that interesting to change frequently. Then this is honestly good enough. You don't need hourly data for a service that you're going to update twice a year. That's not needed. I think it's interesting to compare to, for example, the transportation sector.

I was taking a train recently in Norway, and on the app, they showed me by taking this train, you've saved this much carbon. I got annoyed. What is this magic number? Then I clicked on it and they actually showed the entire methodology. I'm like, compared to flight, if ou are one person, da, da, da, this is how much. If you were to take one four-person car, you were one of those four people, this is how much you would save. I was pleasantly surprised. The reason they can do that is that is because the cost of fuel and the cost of building a car doesn't change that frequently. They can calculate this once and then use it for a really long time. For software that's tricky. You have rapidly changing software, so this is something to keep in mind.

Danielle: The thought that you brought up, Sarah, about stable applications made me think of a trend work seeing with usually more mature organizations to try to start understanding, if I have stable applications and I also have typical architecture practices that I continue to use, can I start to understand particular architecture choices that impact different carbon and energy use? How can I understand and learn from those architecture choices over time, and then maybe even automate that process? I learned that this particular architecture choice is less carbon-intensive. Can I just make a dashboard or facilitate provisioning certain services in that way?

Asim: Danielle, I think you're right. This is one thing that we haven't really got around to developing a language for yet about how you optimize for carbon at various places. in the show notes, I've shared a link to a thing called the Green Cloud Triangle, but we've spoken about this. There's a kind of iron triangle of compute cycles, response time, and cost that you might want to be doing trade-offs of. For example, there may be cases where you want to optimize for response time and cost. This is stuff where you might say, if you want something to be cheap spun quickly, you might go for say, static pre-build stuff.

For example, you're not doing too much dynamic stuff. This is stuff we know already a lot of the time. It may be that there are some cases where you don't need to have things happen immediately, right? You might be more interested in keeping the costs low, but making sure you've got lots of compute cycles. We might use this in terms of having queues or tools or things like that.

Then finally, this is like the default that a lot of us end up using when we're not thinking about this, which is basically optimizing for compute cycles and response time, not really thinking too much about the cost part or not really knowing that the cost can change in this way. This is like speaking to the fact in many cases and what I think we're going to see more of every time is that the cost of electricity changes depending on the time of day. This is not really exposed to us right now, but it's something that is definitely visible, that does definitely happen especially if you look at the markets such that sometimes the cost of electricity can go negative, so you can be paid to use Compute, for example.

I feel like there needs to be a set of tools or a way to describe this stuff so you can take advantage of these changes that have been happening one layer down in the stack, so that you can basically architect for better, more responsive things, but also in a way that's actually very. very planet-friendly as well as wallet-friendly. I think there's a couple of good posts on the Green Software Foundation blog specifically about this. I might have written one of them, but the other ones have also been written by other contributors.

[laughter]

Sara: Well, I can speak about electricity because I think that's interesting. Many of these cloud providers, they say that it takes a long time for electricity utilities to get to them. During this winter in Norway, we only have hydro here, so if it doesn't rain, electricity becomes expensive or if it's really, really cold, all our water is frozen, it becomes expensive. There were these newsletters reporting almost hourly on like, 'This is how much it costs to charge a smartphone right now. This is how much it costs to use the oven right now." Obviously, we know what the electricity costs right now. I wonder how hard it would be to propagate that.

It doesn't always correlate to carbon though. My carbon cost was the same throughout the day because it was hydro, but their cost changed.

Asim: Yes. If cost- [unintelligible 00:35:05] cost is a proxy for carbon, really at the end of the day, I think here.

Chris: All right, you're talking about cost as a proxy for carbon, and that's because a lot of the time when you have more electricity than you need on the grid, it's lunchtime and it's sunny as hell, or it's windy as hell, you've got more than you need. The problem with the grid is that the grid has to basically be balanced the entire time. Otherwise, basically, very, very bad things happen and very expensive hardware gets damaged that falls off-grid. You can end up with incentives to basically-- If you operate a grid, it's cheaper. It's easier for you to basically just set the cost of electricity to be lower than it is to ask someone who runs a big nuclear power station to please turn down the output to make sure stuff is balanced, for example.

This is the thing that you are often doing. This is why the cost will change over time, depending on how much demand there is compared to how much supply there is. Most of the time we're shielded from this, but it's actually quite fascinating. It's stuff that you can absolutely take advantage of because companies do this. Google make a really big thing about shifting loads to when energy is green, but the reason that they do that is it really saves a bunch of as well as just carbon, basically.

Asim: Yes. That's the secret to a lot of our spaces that there is cost savings, a lot of this stuff as well.

Sara: I don't necessarily think that's a bad thing. Capitalism runs large parts of this world. If we can get those forces to work with us, promote us to a greener future, we shouldn't necessarily be against it. I'm saying capitalism is always awesome, but you take the wins where you can find them.

Chris: This speaks to incentive design, basically and who's making good use of this. There's a really nice example of organizations. There's one organization in America called Lancium, and there's another one in Germany called XMesh. They take advantage of this. They basically take data centers which is basically a shipping container full of machines, which will otherwise be thrown out from hyper-scalers like Facebook. They put them on renewable energy parks. What they end up doing is they end up providing stuff for either-- okay, I'm not a huge fan of the cryptocurrency stuff, but you can use the same thing for machine learning models as well.

Anything, which is a plausible load that is quite compute-intensive is a really good fit for this use of an oversupply of renewable energy in many cases. This is what Lancium and XMesh both do now. By being able to be plugged straight into places where they have energy, that they otherwise would not be able to use, they basically end up being able to provide Compute for a much, much lower cost. You can get your machine learning models done at a fraction of the price from some of other larger providers by going with this because they're taking advantage of the economics and how they've changed over the last 10 years that in many cases say some other there haven't taken advantage of yet.

Asim: Why is that more economical? At the end of the day, those servers are servers that they got essentially maybe for free or very, very, very reduced cost. They must be four or five years old.

Chris: There are two reasons. Moore's law has slowed down over the last, say five years, for example, before you could just rely on [unintelligible 00:38:20] to do this work. As a result, servers which are maybe two or three years old, aren't actually that much slower than they were previously. If you've got something like a plausible load, because you are not trying to run it 24/7 all the time, if you have different requirements for keeping this stuff cool, for example. Unlike, if we're going to talk about keeping cool, for example, there are some really nice examples in the Netherlands where they basically have shipping containers full of servers.

Once again, these are servers which are end of life. They plug them into greenhouses with the idea being that the waste heat, rather than basically vented into the sky, or you spend loads and loads of money trying to get rid of it because you see it as a waste product, they use it to pump into greenhouses, so they end up with really nice juicy tomatoes. This is a really, really cool use of heat because the greenhouse folks, they were like, 'Well, we can either burn fossil fuels for heat or we could just use that heat from over there." This is an example of taking advantage of-- If you understand the underlying energy systems, then there are all these fascinating, new, pretty cool use cases. I don't know about you, but the idea of, I don't know, a greenhouse connected to a data center and juicy tomatoes, that sounds cool. I like that idea.

Asim: I do know, actually that heating greenhouses is one of the biggest costs for greenhouses. That's wonderful. I love that.

Chris: There are loads of examples here. When you look at the next challenges we are facing between now and the next say, five to 10 years, one of the big ones is heating things up. What we have right now is we have a massive data street full of data centers thinking, "I've got all this heat, how do I get rid of it?" It feels like maybe the people saying, "Well, if only we could find a way to get heat." and then we'll tell the people saying, "Oh if only I could go find it, we'll get a way of getting rid of this heat." If they could talk to each other, then maybe you could end up with a slightly more efficient system.

Now, this isn't going to happen all the time because if you put a gigantic hyperscale data center miles away from everyone else, it's going to be harder to integrate that into in an urban environment. Then maybe that speaks to the fact that our idea of what a data center needs to look like could change over time to end up with a different topology for the internet because the internet did use to be quite distributed. What we've seen right now over the last 10 years is that the energy sector has ended up looking a lot more like the internet and the internet is now looking a lot more like the energy sector was 10 years ago.

I feel like maybe there's scope for us to find some happy medium rather than just zipping past each other in mad decentralization or centralization mania that we have at present.

Danielle: I think, Chris, with all that you're saying there's so much opportunity. My question is, where does the responsibility lie to provide that information to consumers, and who is responsible to make these choices of shifting workloads, taking advantage of the energy at different times of day, that type of thing? How much can the cloud providers do and how much can the consumers do and what is that balance? How do we get there? I think it's going to be a really interesting problem that, hopefully, we get to solve in the next few years.

Sara: I'd love to see a carbon throttling thing that you can add to your services, whatever cloud provider you have. It's like, Yes, you can carbon social this application. That's fine for me."

Chris: There is loads of cool stuff happening in this field right now. Brunch Magazine does examples of this. If you go to brunch.climateaction.tech, it throttles based on the carbon intensity of the grid right now because this is exposed to it. There are also tools [unintelligible 00:41:49], for example, that let you do this stuff. This stuff exists and there are examples of it being built. I think it's a really exciting fun place to be looking at this but there's a whole policy piece that would map to what we're doing here.

[background music]

Asim: This has been a wonderful conversation. I love all the places we've been to. Maybe let's just end with just one quick thought or idea from each of us and think about the future and something from our conversation. I might start because if I don't, I'll forget. There's something you touched on, Chris, earlier on, I thought was fascinating. You talked about what of Meta's datacentres was going to 8 meters below sea level.

One of the things for one of our future podcast episodes I would love to explore is the SEC has just had a proposed I think ruling, I don't know if that's where you got the data from. The SEC has proposed ruling now that organizations have to disclose their climate risks. I'd love to have a conversation about what are the climate risks related to software and green software and sustainability and technology? That's a great example that you gave, and I just thought that's something I'd love to explore in the future.

Sara: Final thoughts, placing my tomato plants next to my laptop, number one. Number two, it will be interesting to talk in the future about how the pure economical aspects of where to place a data center will impact the grid. If you're only placing data centers where the grid is green, will that power a green shift in the energy markets?

Danielle: I'm having trouble wrapping up all these thoughts. There were so many different avenues. I think something that stuck with me that I'll continue to think about is the idea of carbon intensity and viewing that in conjunction with totals, using these variety of numbers to come up with a strategy. I thought that was really interesting.

Chris: I guess that's me left now, actually. Asim, I'll keep it short. I think this points to us having a carbon-aware internet. I think that's a really cool vision, personally. I'll leave it with you, philosophwith you, Mr. Hussein.

Asim: Thanks for listening to Environment Variables. All the resources for this podcast, including links to our guests and more about Amazon's customer carbon footprint tool, as well as the Green Software Foundation, and everything else we read discussed today is going to be available in the show description below. We hope you enjoyed the show. See you on the next one.

[background music]

Asim: Hey, everyone, thanks for listening. Just a reminder to follow on Apple podcasts, Spotify, Google podcasts, or wherever you get your podcasts. Please do leave a rating and review if you like what we're doing. It helps other people discover the show and, of course, we want more listeners. To find out more about the Green Software Foundation please visit greensoftware.foundation. Thanks again. See you in the next episode.

[music]

[00:44:52] [END OF AUDIO]

00:00 -00:00

Environment Variables

Each episode we discuss the latest news regarding how to reduce the emissions of software and how the industry is dealing with its own environmental impact.

Om podden

Avsnitt

Backstage: Green Software Patterns

The Week in Green Software: Sustainable AI Progress

AI Energy Measurement for Beginners

The Week in Green Software: Transparency in Emissions Reporting

How to Tell When Energy is Green with Killian Daly

Backstage: Impact Framework

Backstage: Carbon Aware SDK

Deep Green Technologies

Finding Signal Amongst the Noise in Carbon Aware Software

The Cloud and the Climate: Navigating AI-Powered Futures

Green Networking with Carlos Pignataro

The Quantum Entanglement of Software Sustainability with Wilco Burggraaf

The Week in Green Software: Powering AI on Atomic Energy

The Week in Green Software: Wooden Data Centers

Shaping Web Sustainability with the W3C

Backstage: Awesome Green Software

Remembering Abhishek Gupta: How does AI and ML Impact Climate Change?

Backstage: TOSS Project

The Week in Green Software: The Sustainable Data Paradox

Electricity Maps

The Week in Green Software: Obscuring AI's Real Carbon Footprint

Making Testbeds for Carbon Aware Computing

Academic Forefronts

Green Software with Gaël Amongst the Whales

A Greener Internet that Sleeps More

The Week in Green Software: Tackling the Energy Challenges of AI

Preview: CXO Bytes

The Week in Green Software: AI’s Power Problem

The Week in Green Software: FinOps, GreenOps and the Cloud

The Week in Green Software: Carbon Hack 24 Recap

The Week in Green Software: Building Green Software 2

Environment Variables Year Two Roundup

Greening Software Procurement

Community Publishing and Greening Software

Building Green Software

Greening Low Code

Greening Digital and the Rebound Effect

Nowcasting and Using Computers to Reduce Emissions

AI Legislation

Greening Serverless

Carbon Hack 2024 Announcement

The Week in Green Software: Embodied Carbon

The Week in Green Software: CNCF TAG Environmental Sustainability

BETA Impact Framework

Cloud Footprints with CCF

Driving Climate Change Solutions with AI

The Week in Green Software: Google, Grids & Green Software

Responsible AI

Decarbonize Software 2023: Recap

The Week in Green Software: Greening the Front End

Introducing the Impact Framework

The Week in Green Software: Modeling Carbon Aware Software

The Week in Green Software: Greening Web Standards at the W3C

The Week in Green Software: Mapping Green Software on the Grid

The Week in Green Software: New Research Horizons

The Week in Green Software: Net Zero Cloud

The Week in Green Software: Automating the Software Carbon Index

Decarbonize Software 2023 Preview with Namrata Narayan

The Week in Green Software: Cleaner Energy with Molly Webb

Sci-Fi Fantasies with Anne Currie and Jo-Lindsay Walton

Decarbonize Software 2023 Preview with Adam Jackson

The Week in Green Software: Complex Carbon Accounting with Gaël Duez

We Answer Your Questions Part 2

The Week in Green Software: Green Kernels

The Week in Green Software: Carbon Aware Spatial Shifting

The Week in Green Software: Open Data with Fershad Irani

The Week in Green Software: Data Center LCA with Stani Borisová

We Answer Your Questions!

Environment Variables One Year Round Up

The State of Green Software Survey with Tamara Kneese

The Week in Green Software: Code Green and Clean Power

The Week in Green Software: Open Source Innovations with Tom Greenwood

Fact Check: Colleen Josephson, Miguel Ponce de Leon & AI Optimization of the Environmental Impact of Software

The Week in Green Software: DIMPACT with Andri Johnston

The Week in Green Software: AWS & Scope 3 Emissions Data

The Week in Green Software: IaaS, PaaS, SaaS!