 Hi, this is Yoho Sapnibhatiya and today we have with us Webbrown, CEO and co-founder of KubeCost. It's great to have you on the show. Great to be here. Thank you so much for having me today. I was actually looking forward to this discussion because you might be surprised or not surprised that I was at KubeCon. We did a lot of interviews there. Your name will come up in almost every other discussion. So I'll tell people that, yeah, I'm going to talk to him in a couple of weeks. So I'm looking forward to that. So there are so many things to talk about. I was just working on a story and community's complexity and cost, it does become a big topic. But before we go there, you are not at KubeCon. We will see each other at KubeCon, but I'm pretty sure that you were keeping an eye on it remotely. What was your observation? Because when we look at KubeCost, the kind of discussions that are going on there, the kind of folks that are there, that gives you some indication where the market is heading. Yeah, so I wasn't able to attend this year. We talked a little bit about it. I actually had my first child right before, but looking forward to being at the next one. But we did have a team of 14 that was there and came back with like really lots of interesting observations. And it really feels like energy is kind of reforming in this ecosystem, if you will. It felt like with the pandemic, we all kind of were really focused in our own silos in a lot of ways. And now it feels like just this reemergence of this community. And it's really interesting that like a lot of digital transformation and cloud native adoption acceleration has happened in that period. So it feels like we've really moved the ball forward in terms of production kubernetes, production kubernetes at scale. And I see more and more teams starting to think about these day two like problems, right? So now really thinking about operability around security and cost and reliability, et cetera. Which is the phase after you've really got your hands around kind of your first set of microservices, if you will. Now let's talk about open cost, which is an open source project for kubernetes cost monitoring and management. What was the driver behind this project? Why you felt, it's actually quite obvious, but creating an open source project specifically for that. And if I'm not wrong, there is FinOps Foundation, which deals with a totally different problem but which is also in terms of cost. So talk about the origin of the project, why you created it. Yeah, I think it really goes back to the origin of Kube cost, which is born out of our experience at Google, where we'd spent about five years doing various like infrastructure monitoring projects, like mostly focus around performance costs and health. So my co-founder and I worked on like internal observability as well as external observability for Google Cloud and a DevTools team. And when the Kubernetes effort was born, we saw really even early adopters at scale, struggling with even just how to like understand and monitor and allocate costs. Because all of a sudden you don't have a dedicated VM per team or application, et cetera, you have these like really dynamic services and pods containers being deployed by the Kubernetes scheduler. So that was the kind of the genesis of Kube cost. Over the last about three years now, we've helped thousands of teams get monitoring and then optimization in place. And two things kind of emerged during that time. One was that there's actually a lot of teams when we start working with them that have like implemented their own kind of DIY monitoring, if you will. Sometimes it's a simple Grafana dashboard. Sometimes it's like, you know, doing something more advanced with like Prometheus. And what we saw is that like, basically all these teams were doing it different, right? Like there's just tons of ambiguity and no like agreed upon standard in terms of how to measure costs. And then secondly, all of these partners were coming to us saying, we're hearing this demand and are, you know, problem from our users and we also want to give some visibility. And so we came together with this amazing group of partners and launched the open cost effort just last week. And it's this brand new, you know, vendor agnostic standard, which comes with a detailed spec on how to implement it. And then also comes with a going implementation, which is the Kube cost core allocation engine, which we've contributed. Because it's an open source project. So when we talk about and people talk about ecosystem, you know, when people are community, there is no community, there are communities. It could be users, it could be vendors, it could be maintainers. So if you just look at open costs, talk a bit about what kind of communities are on that or what kind of community you're... I mean, you cannot build a community, community itself, they folks come together, they create a community. But I do understand what kind of ecosystem is around it, who are supporting it. Just give us a picture of the community. Yeah, so it's early days and we are very much actively welcoming other contributors, but we've got this amazing mix of cloud providers, ecosystem partners, and then end users. So like AWS and Google made really big contributions from the cloud provider side. Then we have others like Adobe and New Relic and D2IQ. Those were the lot of like deep Kubernetes experience. So just tons of input from around the table from that really broad array of ecosystem partners. And now we're just in the stages where we're welcoming kind of even broader community input now that we've launched open costs 0.1 and already getting tons of great input in the almost one week that we've launched it. So we are looking for and hoping to get just input from a broader community. That includes the FinOps community. That includes the CNCF or like cloud native Kubernetes community and others. We feel like this is just the beginning. And again, we're super focused on containers and Kubernetes now, but we also think this framework can apply and this implementation can apply to other areas. Talk about the scope of the project and how do you actually help organizations or developers, I mean, depending on who really is the target there to kind of keep an eye on where they're spending and going and of course, you know, optimize it as well. Yeah, lots of great points there. We view Kubernetes itself as this incredibly flexible and powerful platform, right? And that can be applied in a lot of different ways. And that our view is one of the like core reasons why it's been so successful was that enterprise needs are complex. And, you know, Kubernetes has developed as a complex platform to try to meet all of those needs. So if you look at the core scope of the open cost project today, we are targeting Kubernetes environments that are like where the Kubernetes API is available. So, you know, that could be, you know, managed service at any of the major three cloud providers. You know, that could be on-prem clusters. That could, and we're starting to discuss this as a team, but technically the implementation will support environments like K-Native that are, you know, abstractions on top of kind of, you know, underlying Kubernetes environments. So, you know, it's that core use case. Already we're discussing around, you know, containers being run in other places and already we're discussing around, you know, external cost or cloud services that are consumed by a core Kubernetes tenants. So those will be discussions that are, you know, unfolding now or, you know, will unfold in the coming weeks and just starting now. When we talk about how this really helps teams, this, the flexibility of the spec starts with understanding cost in real time at the container level. So once you have a cost at that very granular level, no matter how you're using Kubernetes, you can aggregate it in a way that's meaningful for you. So you can look at cost by microservice or by cluster or by namespace or by deployment or stable set or job or label or anything. You know, that is a lot of the core power is this open cost model is truly built up from a per container basis. So that allows it to really support, you know, the nearly infinite, you know, deployment, you know, configurations possible with different Kubernetes environments. Project is very, very new, but you folks have already planned to contribute it, you know, to CNCF. It's too early as you earlier mentioned, but just give us kind of, you know, what is your long-term goal with the project? Where do you see it, you know, evolve? Yeah, I mean, we'd love to contribute this to the CNCF, you know, it is in the queue, you know, for review now. And ultimately, this is about, you know, similar to Kubernetes, which is, we want to see a neutral governance model, you know, where the community can be very actively involved and ultimately help address a lot of this ambiguity around measuring cost. Because our view is again, once you have a real-time cost language that is, you know, agreed upon, then you can start doing really powerful things like real-time alerting and dynamic optimization in ways that just weren't possible because you didn't have that, you know, complete, accurate, fair, et cetera, language for measuring costs.