Latest / The MonkCast / Chopping Wood, Carrying Water: VMware by Broadcom's Open Source Story at KubeCon EU 2026
Transcript
- speaker-0: Hey, it's James from Redmonk. I'm here with Dilpreet. I'm here with Zach. And we're here at KubeCon Europe 2026 in Amsterdam. It's a beautiful day outside. It was raining this morning, but the sun has come out. I'm really happy to have you both here. Why don't you introduce yourselves, say a little bit about who you are and what you do at VMware. speaker-1: Well, Dilpreet Bhindra, I'm the Senior Director of Engineering for ⁓ a number of things, but also the Kubernetes stack that comes out of VMware. speaker-2: I'm Zach Shepard. I'm the distinguished engineer responsible for the vSphere Kubernetes. speaker-0: Awesome. Well, great to have you both here. Open source, obviously key topic. That's why we're here. The CNCF has done an amazing job of fostering an ecosystem, a diverse ecosystem. But yeah, open source, you're not necessarily the first name on people's lips when it comes to open source contributions and the open source ecosystem. So I wonder like, what's the situation there in terms of your commitment to open source? speaker-1: Yeah, I think it's really interesting that we, you know, it's probably the best kept secret ⁓ in the industry is that, you know, VMware and ⁓ Broadcom have been significant contributors to the CNCF ecosystem ⁓ starting ⁓ a number of years ago. And, you know, we, for the last decade or so, I think we're in the top three in overall contributions and of late, it's probably top five, right? And, you know, in terms of the contributions, they're very focused on sort of ⁓ the things that really help Kubernetes run. So I find it very interesting that we, and frankly, we don't do a very good job of talking about it either. speaker-0: Go ahead. speaker-2: I think one of the words he used that it's interesting is the commitment to open source. For me, one of the things that I feel shows our commitment is we've moved from our cluster lifecycle management API being a proprietary API wrapping open source technologies to exposing cluster API directly. That is the API for the vSphere Kubernetes service. You want to create a cluster, upgrade a cluster, you're using cluster API. That means we've committed to maintaining that API as a part of our product. We wouldn't bet on something like that if we weren't serious about the community. speaker-0: Okay, so talking about the seriousness of the community, think one of the couple of things here, A, brand lags. So, know, VMware storied history and it didn't grow up in that era of open source. So it's going to take a while to get that to change. But one of the things they talk about in the Cloud Native Computing Foundation sort of community in this ⁓ Kubernetes community is chopping wood and carrying water. That you've got to do some grunt work and support the community. ⁓ in order that you can be recognized. So tell me bit about sort of chopping wood and carrying water and some of the activities that VMware is involved in from that perspective. speaker-2: Yeah, so I think there's a, again, a couple of perspectives that we can look at that from. One is the CNCF has a Chop Wood Carry Water Award. And if you look back at the last, ⁓ say, five years, I think we've had ⁓ four employees who have received that award. So I think there is a recognition that we have employees who are chopping wood and carrying water from the community. I think a different way to look at it is... ⁓ When we see opportunities in the ecosystem, like we recently saw with etcd, we have an etcd maintainer who saw that there were opportunities to improve the ability, not just for us, but for anybody to diagnose and repair etcd clusters. And so we contributed those sorts of components back to the community, ⁓ not just for ourselves, but as open tools that anybody can use. speaker-0: I love that story for one reason, special reason. When Amazon first arrived on the scene in this community, they had some opinions about EdCDA and they had not yet done quite enough of the chopping wood and carrying water and that caused friction. So it's a great example and yeah, that's exactly what we're talking about. speaker-1: I, you know, that example is pretty cool because, ⁓ it takes, ⁓ the things that we are doing from a support perspective that are helping our customers. And then we're contributing it back to the overall community as value add for a number of different, ⁓ distributions to take advantage of. But beyond that is like one of our support folks was telling me recently that his brother. was that use this ⁓ tooling to be able to help another distro to be able to get out of trouble. And we were like, this is great. speaker-0: That's awesome. Yeah. Yeah. That's exactly how it should work. So I think, ⁓ when we look at this, this question is, okay, we are involved in the community. are, using open source. We are contributing to open source. ⁓ now one of the things I assume you want to do is, kind of lead some projects. ⁓ the chopping wood, carrying water can win you some influence. ⁓ you've got some. ⁓ Project Valero, think, is a good example. Now, on the other hand, so Valero is a project led by yourselves, but Zach, you had some opinions about how important ⁓ diversity of contribution is. So tell me about that, because I was kind of surprised. I'm like, aren't you just going to win some influence so you can ⁓ drive a set of technology in particular direction? speaker-2: Yeah, so that's a good reminder. feel like, know, Valero is a VMware-originated project, but it's, I wouldn't even call it a VMware-led project at this point. We've got maintainers from multiple companies, and I don't even think we have a majority of the maintainers. And kind of related to this evolution for Valero, it's part of why we've contributed it to the CNCF sandbox, to make sure that it's is viewed even externally as a community project rather than as a VMware. speaker-0: ⁓ by the way, what even is Valera? speaker-2: Valero is a Kubernetes native backup restore tool for backing up any Kubernetes cluster. speaker-0: So this is a. speaker-1: I know it's the fundamental API that you would use to drive any of your data protection on cloud-native environments. And so it's used by the industry. As Zach mentioned, there are many contributions across the industry. And I think it's pretty cool that we were able to not only ⁓ think about internally that we should ⁓ contribute this to CNCF, but... How resounding the feedback back was, think, we had a very fast turnaround from the time that we went to the CNCF community and said, hey, we'd like to contribute this to them coming back and saying, hell yes. speaker-0: And I think, well, there, mean, the hell yes. mean, look, you're VMware. If you don't know how to do backup and restore, then I don't know what we're living in. So that's a good area for you to be investing in. And it's great that, oh, who are some of the other companies you say there is some adoption outside or where are the maintainers? Who else is adopting this? speaker-2: Great question. So I believe we have maintainers from ⁓ Red Hat and Microsoft. I feel like I'm forgetting somebody, but it's a diverse ecosystem and it's hard to keep track. speaker-0: Yeah, no, 100%. And I think Red Hat and Microsoft and yourselves, that's already showing some diversity right there. speaker-1: I think just lingering on Valero for a second, like, you know, some would think that this contribution into the community may be viewed as, hey, us ⁓ contributing something that we don't believe is valuable. It's quite the opposite. And back to your point is this, it is a tool that we believe in our overall VKS story as well as our overall ISE story is going to have huge dividends that we're going to keep. investing in over time because we believe in it so much. And that's why we think the governance of CNCF is important. speaker-0: Okay, awesome. Now, slight gear change, but here we are, it's 2026. We can't get away without talking a little bit about AI. Open source and AI and Kubernetes, it's one of the big themes here at the moment in terms of that underlying infrastructure for all of these services that seem to be popping up like mushrooms every day. There's a new thing that we need to be thinking about how to run and so on. So what are the stories in terms of where we are around AI? the products you're building ⁓ and the wider ecosystem. speaker-2: Yeah, I so I feel like the place I'd start with this is that VKS is a certified Kubernetes AI platform, and that's a part of the CNCF conformance program for AI. Why does that program exist? It's really about ensuring that as different Kubernetes platforms become able to run a wide variety of AI workloads, users maintain that portability, the ability to ensure just like any traditional workload can run on any certified Kubernetes distribution, any AI workload should be able to run on any certified AI platform. So that portability I think is important to ensure that as the community evolves and as these new projects show up and as people start to take bets on those new technologies, they know that they're not locked in. speaker-0: And again, mean, to be frank, whether or not VMware's history was as an open source company, your history is definitely as a portability company. So that is built into, guess, the way that you've always thought that makes sense. And so we need to see that in terms of AI workloads. As I say, partly just because it's a mess out there. speaker-1: Yeah. And this particular function, and I love the leadership from the CNCF community on this compliance. ⁓ I think this is core to our belief around Kubernetes in general. any workload that runs on Kubernetes on a CNCF compliant Kubernetes distro should run on any other CNCF compliant Kubernetes distro. And that we want to make sure is true for VKS, but all across the board so that customers can choose the best platform to run on. speaker-0: Mm-hmm. So here we are. say, it's 2026. It's amazing. It's the biggest KubeCon ever. So Europe is dominating America at this point. 13,500 people plus, I think, is the number that we're talking about. A lot of buzz, some interesting announcements. What are you most excited about? This is the last question. Zach, what are you most excited about this year at KubeCon EU? speaker-2: There was a lot of great stuff in yesterday's keynote, but if I had to pick one thing, I think that would be LLMD being accepted to the CNCF as a sandbox project. I think it's really important that the CNCF is taking leadership in that space of running AI workloads, not just on a single node, but across a cluster of nodes. I think that's an important and exciting space to be in. speaker-0: Awesome. Yeah. speaker-1: I'm, I, I frankly, I, I am very, very excited about the comfort that conversations and the type of conversations we're having, you know, the traffic in our overall booth, the set of, ⁓ customers at differing stages of evolution on their journey for either modernizing their stack or, ⁓ you know, determining how to build applications better. And also in the AI journey and those conversations that we're having are really, really, really deep. and interesting at our booth and I get really excited about that. speaker-0: That's awesome. Well, so here we are. That's another episode of the BunkCast. If you enjoyed it, please do like, subscribe, smash all those buttons. Thanks for joining us. And yeah, thank you very much to both of you. speaker-2: Thank you.