 Welcome to the CUBE's coverage of KubeCon EU 2024, live from Paris, France. Join hosts Savannah Peterson, Dustin Kirkland, and Rob Stratche, as they interview some of the brightest minds in cloud-native computing. Coverage of KubeCon cloud-native con is brought to you by Red Hat, CNCF, and its ecosystem partners. The CUBE's coverage of KubeCon EU 2024 begins right now. Good morning, cloud-native community, and welcome back to Paris, France. We're here at KubeCon cloud-native con. My name's Savannah Peterson, joined by my fabulous analysts, Rob Stratche and Dustin Kirkland. Gentlemen, day two, how you feeling? Awesome, fresh, fresh as the daisy. Ready to go, let's do this. Freshest Paris morning in the springtime. Absolutely, absolutely. No, I mean, I think it's been a very energizing morning, that the keynotes have just wrapped up and what we saw was some great news around sustainability. And funny enough, something I kind of hit on a little bit harder than I expected to yesterday afternoon, but a Kubevert, and running VMs on Kubernetes. We had Goldman Sachs on stage talking about that. With Red Hat, we had numerous sponsors, including Deutsche Bank, talking about sustainability. In fact, I think sustainability was the big topic and how to develop sustainably. I think one of the things, and Dustin and I were briefly talking earlier, that it's one of the things that seems to be really in flux is how do you deal with AI and sustainability? And I think that's one of the big ones. And I know you had some thoughts on that. Yeah, the sustainability thread was common across all the keynotes. I'm sure that was an orchestrated theme here. What's interesting there, though, is man, it's hard to talk about sustainability and the power consumption driven by GPU. I was just going to say, GPU and sustainability is not. And I think rightly so, blockchain and crypto has gotten a fair amount of flack about how much of the world's energy is spent mining coins, okay? Now, if you're not mining coins with GPUs, you're probably running an AI ML workload of some kind. And I think we got to think a little bit about whether, if that is worth, if the product that is being produced by those ML workloads can be sustainably performed. I don't know. Well, it's also about optimizing it. If it's sitting there idle and you're just sucking power, it's not exciting. Well, it sucks way more power when it's running, less when idle, but some when idle, for sure. So utilization, that's a piece of it. And then running as efficiently as possible. We heard a little bit from the ARM folks on the stage today about, that takes care of part of the problem, I think. But the CPU is not drawing the power. The GPUs are drawing the power. And I think also, and I think part of, yeah, I think that's, some of them had gone into the fact that storage was a big component of it, but it was also about, hey, how are you planning for high availability? And I think that it was really interesting how they were going through a number of these different high availability strategies and what they were still not seeing between how you use auto-scaling and how you use actually these different CPU planning tools to go out and do that. But then there was the back end storage, which moving to non-spinning disk definitely helps, but there's still a ton of spinning disk out there. So I think that's an interesting piece in a way that they were going about it and the string that they kept pulling on, so. Obviously a conversation around sustainability and I love that you brought up the blockchain piece was a part of that movement and it's true. Those are really our power-suck moments. So yeah, we're talking about it. How are companies actually going to do this? Is it possible or is this a bit of a myth? I mean, it's a great talking point for a keynote. I think you're right though. Right, I mean, love the lip service and love the planning. What are we going to do about this? But there are a number of projects. The ElephEnergy group is looking at this. So there are a number of projects going down. I think it's still really hard. And in fact, I thought that Deutsche Bank, who was a sponsor, who was up on stage showing some of their dashboards, I thought they hit on a very interesting point. And when I left Amazon some number of years ago, they literally came out with, at Reinvent, a carbon footprint tool. And my whole thing was that I thought it was a little bit of greenwashing. Well, Deutsche Bank, now we're two plus years or three years out from me leaving there. Deutsche Bank basically laid into all the cloud providers saying their carbon footprint tools are not sufficient. And that they've gone out and built their own and started to do that, but then they showed the architecture for it. And it's a lot of gear. It's a lot of containers. So I think there's a lot of different vendors that are kind of aiming at that and to try to provide that as a really how to get sustainable. There's got to be more done in the community. There's a lot of different approaches, right? There's the one piece that we have some can, I say we, the people who are part of the cloud native computing foundation community have some control over is writing code as efficiently as possible. You can write your code inefficiently. That consumes way more CPU, GPU, disk storage and whatnot, we need to do that as best as possible. With the silicon vendors do or choose to do or can do, that's going to be largely driven by their customers. What customers decide to pay for, decide to pay a premium for, what customers refuse to buy. And then on the backend, the cloud operators, energy credits, sustainable energy initiatives. I spent some time at Google with a number of data centers that are carbon neutral. And the way to get that is consuming energy from renewable sources, but also buying CO2 offset credits where possible. It's going to take multiple layers. It's messy. Yeah, we actually had a conversation at Supercomputing in Denver, about a company in Denmark that's using, they're actually heating Copenhagen with the power, like liquid cooling quite literally, and then taking that liquid and then heating it. And it's a nice way of regenerating that and using that power consumption for good and creating a commodity for the community. And they actually had one of the local providers who's here in France and has data centers here in Poland and somewhere else I can't remember, but, and they were talking actually, another thing that doesn't get talked about is water, and the water consumption and things of that nature. And I think, to your point about GPUs, the Iowa data center for Microsoft, where chat GBT runs, for every five queries or every five prompts is 16 ounces of water. Wow. Glasses of water, which is insane. So you start to look at it. There's a lot of measuring management better. Like you said, there's another Tigra who's another service provider up in the Nordics. They're actually building into mines, abandoned mines. That's where they're building their data centers to get ambient cooling. So I think it's going to, it's got to be in all of the above strategy. I tend to believe and talk to people that the credit stuff, what the hell's a credit? Where does it go? How does it really get applied? It doesn't take carbon out of the air. It's a nice thought, but I think when you start to look at it, and I think organizations, especially here in the EU are starting to hold people more credit. It transfers the responsibility. The credits programs transfer the responsibility. It helps others invest in sustainable initiatives. There are ways of taking CO2 out of the atmosphere, but it's very expensive to do so. And so that's not something that can be done on the cheaper free. Yeah. No, not at all. I mean, and I'm glad that maybe that'll be a theme of our day today. We've got some really interesting guests on. We've got Scott from Docker. We've got Microsoft. We've got more guests from Red Hat. It's going to be power packed. Absolutely. And I think they talked about Wasm was another one and talked about SpenQ. Yeah, we got a Wasm panel. And we'll have that, talking about how they're packing more in, using actually SpenQ on Kubernetes and packing in more smaller applications and being to your point about more efficient compute and more efficient development. They were looking at that. And that was a little bit of a message as well. So get a lot of fun. Well, so today we're doing our swag segment tonight. Have you seen any cool swag out there? I haven't. This is a pretty swaggy event. Yeah, absolutely. There's socks galore if you want to. Chronosphere made me a pocket square. It's going to be really hard to beat that. It's fun. Yeah. Yeah, I know I feel bad I don't have a jacket on today to show it off. But anyway, what do you think is the most surprising thing about the show? Hot take, huh? Yeah, hot take. What is your hot take? It's time for that. Starting day to super fresh. I think again, I'd just go back to what I pounded on yesterday, which was security, moving SecurityCon out of the main KubeCon and making it its own thing has kind of forced people to get in. We were talking about one of the people saw a huge line for a security, one of the few security tracks that was going on. Yeah, yeah, yeah. So I think the fact that security kind of got separated out may be a miss and maybe it should just be a co-located thing. I don't know if how that works out, but I think that to me is, that's my hot take for so far for this week is that security is missing. Yeah, I'd like to see a lot more security focus for sure. I, you know, if we look at the discussion around AI ML, I think it's surprising to me that it's still, that it's still so front and center. We spent all day yesterday talking about it. It's almost like apps going on the line in the late 90s. Like at some point being on the internet is no longer like a badge of pride. You just, you were on the internet at some point, right? We stopped talking about it and bragging about it and designing around it. I guess we're still in the early days of that, but at some point, I think there's AI ML everywhere and it's less of a story than it is right now. In fairness though, I think when we were, you know, ChatGPT came out after the call for papers for KubeCon Amsterdam. This is actually the first cloud native event in Europe since. In Europe, yeah, that's true. That's a good point. Since we saw this whole surge. That's a good point. So I would say in fairness. Both continents get to have their AI ML. Exactly, I would say, come on, come on, come on, come on. Okay. We got to give the community a second. I understand you're on the front lines, Dustin, but I do think it's fair. On that note, actually speaking about this, there are 233,000 contributors to the cloud native foundations projects. There are 183 graduated incubating and sandbox projects. And like we said, this is the largest KubeCon ever. So maybe that's why everyone's excited to talk about AI. Got another hot take for you. Just thought of one. The number of VCs and startups here outstanding. Actually really, really impressive to see the VC community in Europe and supporting, you know, hopefully some founders with some new ideas. I will say that the ecosystem is different here than it is in the United States. And it is a much more collaborative and inclusive. And maybe I'm obviously jaded because we're in Paris and it's lovely. But genuinely, it always is a different energy when we're here. Well, I'm excited for our day. We're going to have some great guests. Can't wait to do it with both of you. Dustin, Rob, thank you so much for being here. Thank you to our fabulous team over there for holding down the fort. And thank all of you for tuning in from home, from work, or from wherever you are. Here for our live coverage at KubeCon, CNCF's largest event here at CloudNativeCon in Paris, France. My name is Savannah Peterson. You're watching theCUBE, the leading source for enterprise tech news.