Cloud Native Theater | KubeVirt Summit: Opening Remarks and Commu... Andrew Burden and Itamar Holder
About this talk
In this session, Andrew Burden and Luboslav Pivovarch from the KubeVirt community introduce the first in-person KubeVirt Summit, celebrating the transition from virtual to live events. They discuss the importance of community engagement and contributions to the KubeVirt project, including the release of version 1.8 and the introduction of Virtualization Enhancement Proposals (VEPs). The speakers explain how VEPs facilitate project evolution through user collaboration, enabling contributors to propose, design, and prioritize features for KubeVirt. They also highlight the structure of their development process, detailing phases from design to implementation and release. The talk concludes with encouragement for attendees to participate and contribute to the community.
Full transcript
So, we're going to get started just like a little bit early cuz we've got so many people here and we're going to be a bit of a rush. Just uh I can't see my slides, so this might be a little bit muddled, but let's do it. Woo! All right. Good day, everybody. Uh it's really exciting to have you all here. My name is Andrew Burden. This is
Luboslav Pivovarch. Uh we are with the KubeVirt community. Um and this is our very first KubeVirt Summit live event in person, seeing directly face-to-face. This is really exciting for me. We've been running a virtualized event for the last 5 years. We'll probably run another one. Um but this is a big deal. So, I just wanted to thank the CNCF for putting forward the Cloud Native theater. Um
we've got a great bunch of presentations and speakers, so I'd like to thank them for coming forward and bringing their use cases and what they're working on. It's really exciting and I'd like to thank you all for being here as well, for being interested in the project and engaged with the project. So, thank you all for being Um if you're not familiar, if you're using KubeVirt and
you're not aware of how you can be more involved in the project, we have an adopters file. that QR code will go to it. It's also got the web address above it. Um we would love for you to open up a PR, uh put your um organization there, a short description of how you're using KubeVirt. It really it's it means a lot to us as we kind
of uh move towards graduation. Um it's one way to help us. Also, the CNCF is interested in case studies. So, if you're using KubeVirt in an interesting way and other CNCF projects, the more the merrier. Um they would love to hear about it. Um the CNCF website has more details on that. They're also interested and I only learned this the other day, but cloud native reference architecture.
I don't entirely know what it means, but uh George Castro, who's around here, knows all about it. The CNCF webpage I think has details on it as well. Also, our meetings. We have a whole bunch of meetings. I got a slide further on with a link to it as well. You're welcome. I don't know if you know this, but you're welcome to come to any and all
of them. So, I'd be love to see you there. You don't have to say anything. You can just kind of like ghost a little bit. That's totally cool. Well, like I said, we've got a big schedule. Here it is quickly. Um, so we'll be running all the way to 1:30. and once again, thank you to our speakers here. And we have some cups with KubeVirt logo. They're
really cool, insulated. They might fit giant coffee or can of big beer. So, if you're going to ask a question as a kind of like thanks for being engaged with the talks, we'll offer you a cup as long as stocks survive. Um, directly after our last talk here, you've got 15 minutes to get to our maintainer talk, which is an EO3 EO3105. Did I say that right?
E103105. it'll be covering a little bit of what we're talking about here, but in much greater detail. So, if you walk in a brisk but orderly fashion, I'm sure you'll make it there for 1:40 1:45. So, before we look at the what's on the horizon of KubeVirt, what's kind of on the bleeding edge of KubeVirt, let's take a little bit look back. Just to yesterday, we released
our version 1.8 release. It's a really big release. This link will take you to our blog post about it. I've just got a high-level kind of a couple of features there. The blog has more details. It also links to the release notes and the release details itself. So, check that out. And also, in the last 6 to 12 months, just a little bit more of you. So,
VEPs, my man here is going to be talking about that in a little bit more detail. Um but they're very exciting and in short basically it's a it's a guideline, process, timeline, and template that basically creates a space for end users such as yourself who have like some big, complex, crazy ideas that maybe you want to see in KubeVirt you can help us understand it through the
VEP process. We'll work together and we'll try and get it into our release. Maybe not the next release, maybe it's a subsequent release, maybe it's in pieces, but there's a framework there for you to bring those ideas. Um and yeah, Lubo will be talking about that in a minute. We have new working groups. So, these are some big ideas that have come and they they're not going
to be solved in a couple of hours. So, every fortnight we have these meetings for confidential computing, DRA, AIE, which is our first meeting is today, the accelerated infrastructure enablement working group. And we also have a V a VEP open door. So, those are that's a place where you've got a VEP or an idea for a VEP, you want to talk to someone about it, you can
just come to that. It's on Tuesdays. Uh and that's all for me. It would be remiss of me not to mention our graduation. We submitted a ticket uh late last year. Um we've been in the queue and just a couple of weeks ago we were um um assigned a sponsor and so I believe we are now beginning our due diligence phase of that and so it's a
long way till December, so I'm an optimist. Maybe this is the year that KubeVirt graduates, which will be a big deal. Anyway, without any further ado, Luboslav, can you please tell us about the VEP process? First of all, let me welcome you here. It's great to see you all in person, not only on the screen. Uh I see also some contributors, some users, I think. So, happy
to talk about anything after the talks. so, what are the VEPs? Uh they are something like a cap, so the Kubernetes uh, enhancement proposals, but we call them virtualization enhancement proposals or Kubevirt proposals. And they are here to uh, enable us, you know, more uh, natural, but also major uh, process of uh, evolving the project. Uh, first key aspect is like a focus. The community is uh,
a group of people who are uh, contributing to the project and they don't work together in one company or they are not meeting that often. Uh, so one of the key aspect is to focus everybody on the same aspect and same features uh, to guide us into the future. Uh, in order to do that, uh, we created the VEPs which are basically designs design proposals uh, which
you start with creating an issue in an in our repository. And then uh, contributing the design proposal as a PR. Uh, once you do that, we have a phase where we look at the Uh, we try to uh, give you a feedback, uh, see if it's a fit or not a fit for the project. And after we, you know, collect some of them, uh, we can get
into the phase of prioritizing uh, what things we need to work on in the next After we uh, gathered all the VEPs, we know what we want to work on in the next release. Uh, these issues I mentioned that that's the first step of how you basically create a VEP. Uh, are a tracking issue uh, tracking tool uh, to see what's the maturity of the project, what's
the maturity of the feature, uh, in what phase we are, what problems we stumbled upon, what we implemented. And it should contain all the links to discussion some kind of talks as well. Uh after we have this uh of course implement the feature and after the graduation is still going to be still there for future contributors to get the them familiar with the to help them maintain
the project in the future as well. So, who owns the the VEPS? Uh those are owned mainly by the SIGs uh SIGs are the special groups that we have in our in our project like a compute, storage, and networking. Uh the VEP can also be uh owned by multiple uh SIGs if it, you know, uh goes uh through a lot of the area of the project. And
uh final thing I want to say is that all of this is kind of a like our roadmap that you can always have a look of uh what the next two or three releases are going to bring into the So, how this fits into the release? Uh basically, right now at this at this point we are 1 day after the release of 1.8, so we are we
are looking at 1.9. Uh so, in the beginning we have a phase of designs uh so, con- contributing the uh creating a bug tracker issue uh creating the proposal and design. And in this phase uh we try to focus the reviewers and approvers on on only on this repository. Of course, we are still doing bug fixes, uh security stuff. But uh any other feature should be uh
be kind of in the background in this phase. And of course, it's not only these five weeks where you can work on the feature and design. You can work on it anytime, but we are focusing our resources, which are mainly people, right? Which are contributing into the project in their free time, maybe as a work uh And once we hit this five weeks mark, we call it
an enhancement freeze. So everything stops and reviewers move to the main repository or other repositories which the contributions are coming to and the switch focus to developing these designs and contributing reviews. For this this this is a for implementation we have kind of like eight weeks. We have few checkpoints. We have a alpha release, beta release so users can consume them, give a feedback on the features
as they go. Of course, not always it's a not always a happy path so most of the features are coming in the last week or two unfortunately. We will try to make them behave a little bit better. But once the eight weeks mark hit, we have a code freeze. After the code freeze, we have a look if everything was implemented, if we are happy about the quality
or if we need to give a exceptional one week for to to finish the work and we switch to the stabilization where we have a few release candidates. If the current release candidates are okay, users are not reporting any bugs, we move to the release as we did last day. Yesterday. Yeah. And that's that's pretty much it. Yeah. We're out of time. Super jazzed for you all
to be here. Enjoy all the talks. Enjoy KubeCon Summit. Enjoy the rest of KubeCon. Thank you very much. >> Enjoy.
More from this event
See all 436 talks →
Best of KubeCon + CloudNativeCon Amsterdam 2026
2:17
The Quiet Work of Forever: Sustaining Open Source Communities - O. Hope Amaechi-Okorie, JSON Schema
26:24
Evolving KServe: The Unified Model Inference Platform for Both Predictive and... F. Spolti & J. Lee
32:40
Preventing S3 Cost Storms: Applying Cortex’s Efficiency Lessons to I/O-Heav... A. Fishman-Lichterman
5:32