 Hello, everyone. Welcome to the East Joe roadmap update. I am John Howard. I'm a member of the East Joe technical oversight committee who was responsible for kind of being the project roadmap and a bunch of other things. And I'm joined with Eric and we're both excited to tell you a bit about what the future of East Joe holds. So I'll hand it over to Eric. Thank you, John. My name is Eric Van Orman. I started with IBM nearly 34 years ago. I've probably been working on IBM cloud for the last nine or 10 years and been working on this deal for probably the last four. Currently, I am a member of the technical oversight committee. I'm also the test and release work group lead. And I'm also a maintainer on the documentation work group. In the past, I was release manager. Actually, I believe I was released manager with John. And I continue to help the current release managers as they ask questions. I was also a member of the product security work group. In my free time, I try to fly my airplanes and do some ham radio stuff. So with that, let's take a, let's see here. Let's take a look back, since this is a con 2022. Typically, we, we try to start these talks with a set of discussions and there's usually a slide and I didn't make one up. The talks about something about either Istio or service mesh as just sort of a, you know, kickoff slide. So I did look up a couple of stats to sort of kick this off. I did look at the 2022 CNCF service mesh survey. And there was a couple of interesting statistical sections in there. One was what percentage of the respondents were either using service mesh and production development or evaluating it. The results were 60% were using service mesh and production of 10% and development and 19% were evaluating. So basically 70% of the respondents were using some service mesh, either in production or development. I'm sure that number is probably grown. And some of the another section of the survey was what features were these adopters service mesh adopters actually looking for 79% were looking for security 78% for observability. So both very close to 80% 60% were looking for traffic management and just less than that 56% looking for stability. So just sort of a quick recap from, you know, the survey, which was probably done about a year ago on at least trends then. And again, I think service mesh will be a little bit more used than the survey results at that time. I talked about Istio Khan 2022 that was basically held the end of April and 2022. One of the key things that we talked about there for the future roadmap at the time was day two operations. We wanted to pay a bunch of attention to stability, security, extensibility and reach as well. And at that point we had, I think six fairly full slides we and I actually presented the roadmap at that point of things we wanted to do. The first things we wanted to do was, or one of the first things I guess it was announced at Istio Khan 2022 and I was mentioned in the welcome is Istio had applied to the CNCF to become an incubating project. And we basically continued to work with that and then September of 2022 CNCF accepted Istio as an incubating project and through a lot of diligent work with a bunch of people. We actually became a graduated project in July of 2023. So that too long ago. So that was one of the things we talked about. One of the other things that we didn't talk about actually. This was introduced in September was a new data plane mode ambient. I know John will talk a little bit about that in a little bit. But one of the key reasons that it was brought about was to bring a simplified operational experience to users as well as a bunch of other benefits. It was originally released as experimental and with the 1.18.0 release. We now have ambient mode as an alpha feature. Since then, we've been continuing to add features as we get to progress the beta. I know it was mentioned earlier, the state Z tunnel. We've done some stuff with service entry workload entry peer authentication and DNS proxy. Then we sort of get into a bunch of feature promotion. So one of the things we talked about was stability and trying to make Istio more stable. So what one of the things we're pushing on is to try to get a bunch of the features promoted. Some of those promotions that happened where external authorization got promoted to beta with one dot 16. The gateway API support. A gateway API went beta itself in July of 2022 and at that point we claimed data support as well. In terms of trying to increase our installation and upgrade support, we moved helm installation to beta as well as the canary upgrade revision takes going to beta and one dot 17. Earlier this year dual stack support. We introduced this as experimental. And then we talked a little bit about workflow group and getting that promoted as well. Talked a little bit about spilling. We also should talk a little bit about security. What happened with security as part of the CNCS. And to become graduate, we had another security formal audit done in 2022. And I believe the beginning of 2023. There's a blog where we talked about some of the results from that audit. We did we started a bunch of fuss testing way back probably in 2021. And we continued to expand on that. I think the last number I saw was we added another 50 tests and I'm sure we've added more since then. In terms of additional platform support. We did also release arm support in one dot 15 to go back a little bit on the security. And hopefully I'm not going to cause any issues going forward. Knock on wood back in 2021. We had eight security bulletins that we released for Istio. We stayed flat in 2022 and we've had eight. We had eight last year. And so far this year we've had three. So hopefully we continue the downward trend for this year. And with that, I think I'm going to turn it over to John with looking at where we're going. Great. Thanks, Eric. That's that was awesome. You know, I've been involved day to day on the project for quite a few years now. And I forgot about a lot of stuff we've done in the past year so it's really great to kind of look back and see all the progress we made. I'm also really excited to look forward at what we're continue to work on because even since last year. We've just grown quite a bit. A lot of that likely related to the CNCF donation as well as some new great community members getting involved. So we've seen a lot of increase in project velocity and development. So I'm excited to talk about what we're working on and going to be working on for the next year. If you've been to any East Joe cons in the last couple of years or East Joe day, you've probably seen actually slides that look very similar to that. And to this, sorry. And that's because what we've been working on hasn't changed so much on a on a year to year basis. The details of course and where we are along those those priorities has changed. But the general trends have remained the same. We started first started working on East Joe almost five years ago. The product's priority was to kind of build out the value of service mesh, convince users of the value service mesh, and then make sure once they were convinced and tried out East Joe that it didn't explode and cause production outages and whatnot. And those three things took a lot of effort and many years to build out. At this point, we've mostly solved those issues, right? I think there's general industry acknowledgement that service mesh provides a lot of great functionality. Right. We have this rich security layers we have great observability that you gain, and then the whole slew of, you know, bespoke API is to do things like traffic routing mirrors canaries circuit breaking all these things that you may want to add after your initial adoption phase. So the area that we're focusing on is making sure that we can expand that adoption. So service mesh and East Joe provides all this functionality, but it does come with the cost right as the complexity costs the operational costs the resource costs, etc. So for the kind of next stage of service mesh, it's really about reducing those costs, breaking barriers to take compatibility and entry and making it so that we can give those features to all users. So everything in the roadmap is really centered around that and there's a bunch of different aspects of what we're doing, but they all tie into that same goal. So some of the things we're going to be talking about is some new big projects that tie into that goal. The two main ones being the new ambient mode, which we've talked about a bit already, and we have some previous talk from Lynn a few minutes ago. And later on in the sessions we have some more talks about ambient. So I'll just give a more of a brief overview. And the gateway API, which is a new traffic management API from the Kubernetes group, which basically takes a lot of the existing functionality East Joe had in its own APIs and moves them into common core APIs that are shared between other vendors, and they have a lot of improvements over East Joe APIs. We're also going to be promoting a lot of much smaller features and stabilizing existing features or making small tweaks that help reduce their areas to entry. And finally, we want to have really strong integration with other projects and standards because we know that East Joe is not an island right East Joe fits into a much larger picture of in a platform. So if you have a certain certificate management or telemetry or observability back in, we want to make sure that East Joe seamlessly fits in with that so that you can easily adopt East Joe without, you know, having to deal with the incompatibilities between your different products and solutions that you're using. Good. Excellent, please. So I talked about ambient. I just want to dial in on this a bit more. It's been almost a year I think since we announced ambient in the initial experimental release. And since then we've spent a lot of time working on stabilizing it we've been making kind of refined API of it implementation and all these things so that's continuing to be one of the biggest investments from the project. So we're working to drive this towards more stable stability level and try to get to production readiness sometime in the next year. I think you have the next slide. Unlike other features, ambience huge. Right, if we introduce some new, I don't know, routing feature some cool load balancing functionality or something. You know, we have to implement some new API, we write a little bit of code we write an integration test, a document, and off we go. We're pretty much done right. Ambience kind of a whole different thing, right, it's kind of a whole re architecture of how service mesh works, and it's has all these different layers that we're working on. I won't go over all these but right we have new API is we have to rethink how multi cluster multi network work. There's more issues with compatibility with the different platforms. Justin talked about this a bit in his previous talk. You know, we want to make sure we handle all the different platforms that do networking in different ways. There's all sorts of new trust and security boundaries need to worry about. So there's all this stuff that we're working on. So it's a really big project that's kind of tying in every aspect of East DO. You know, there's no, there's no part of these two that's untouched by ambient. So we are hard at work on all these areas, driving it to, you know, more stable levels. One of the things that is most helpful for people that are listening here is, I mean, if you want to contribute to do directly through code or documentation, that's awesome. But even more critical thing we need is actually user feedback. So we want to make sure that all these decisions that we've made actually aligned with what you the users of East DO are interested in a service mesh. So if you haven't tried it out, we would strongly encourage you to go try out ambient in a non production environment and give us feedback on GitHub Slack, wherever. Not just like how I ran into this small bug here. But are you excited about this? Is this something that you would adopt in production? What would you do differently? What do you like? What do you just like, etc. So we're really looking to gather user feedback. So please try it out if you haven't already. Next up is the Gateway API and Gamma. So there's another talk on this later actually by myself and Keith tomorrow, I believe that goes into a deep dive. So I'll just be brief. But this is a new API that we've been working on in collaboration with the broader Kubernetes ecosystem, not just the core Kubernetes but a bunch of other vendors as well like engine X, HA proxy, AWS, etc. To kind of define a consistent API across the ecosystem that is implemented by East DO. And we're working on adding ingress support, which was the initial use case for Gateway. But we have also been driving support for mesh so that we can use the same API for ingress and mesh traffic just like we do with East DO. So this API has been progressing quite a bit. It's now at the beta level and it's rapidly approaching stable. So we are actually working on that as well as moving the mesh support out of experimental mesh supports trailing behind a little bit. So part of reference as cyber stabilization is to make sure that East DO users can use all the functionality that they used to date in East DO but with the new APIs. And I'll hand it back over to Eric. Alright, thanks, John. Yeah, I'll go back and sort of reiterate looking for feedback for for for the ambient mesh stuff. I know we want to progress it from alpha to beta to stable and and part of that is trying to solidify the API. Right, we don't want to we don't want to come up with some set of APIs and then get it out there and decide that, you know, they were slightly wrong. We should do a tweak here or there and then have to to sort of reflect that back into the into the product. It's easier for us to to take that feedback earlier on to help solidify that so that we can sort of sort of get it right the first time I guess I know back in the early years early for Istio days. I know it seemed that there was a lot of churn. And that wasn't necessarily a good thing for both developers and users alike so anything we can do to collect some of that earlier feedback would be extremely helpful. I know there's an ambient Q&A sort of thing coming up later. Or as John said, Slack, GitHub, anything like that. So, besides an earlier chart we said ambient and and the gateway APIs, we still want to continue with the stabilization and promotion of existing features. So part of that is to get back to do it to doing some of that originally a couple of years ago there was a there was some work to try to help do this. We created a repo at the time and and so some charts and sort of, you know, define, you know, what we thought the different levels would be, you know, what does it mean to be alpha? What does it mean to be at a beta level? And some of that is sort of sort of fell on the wayside. And there's some people now that have stepped up and we sort of started this issue enhancement subgroup and Whitney is leading that effort. So thanks to Whitney and all the people that are working on that. And the part of the idea of that group is to take a look at, you know, the existing features as, you know, they are verified that indeed they're at the level that we say they are. So if they're at alpha, make sure they're at alpha, if they say they're at beta, make sure they're at beta. We also want to make sure that we're going to continue the momentum. So if they've gone from alpha to beta, you know, we're still continuing the momentum to stable. Maybe we think they were beta and in reality, they were probably really still at alpha. Let's make sure that we either get it to be beta or, you know, we move back to alpha. And for those things that we're just not really using anymore, and we need to deprecate, let's make sure we're moving things in the right direction. The whole idea being make sure that you as users know where things are, understand what we're doing and try to keep things working on. Either up, down, or continue where they're at. We want to make sure the users understand what we're doing there. In terms of features, in terms of features that we want to continue to promote, the telemetry API, adding some things in there. I mentioned earlier the experimental dual stack support. We want to continue to promote that to alpha, as well as the IPv6. We want to make sure that things relate to was and we continue to move forward. And another feature that is being looked at by Costin and Keith, I believe is this Istio foundational safe mode where you can be assured that you're not ending up using stuff that's, you know, alpha and experimental, basically by default, those things to be disabled. And only those things that are stable would be available to you. So that's some stuff that we're trying to move forward with. And with continued integration, I'm going to turn it back to John. Yeah, so I briefly touched on this, but, you know, Istio provides a lot of functionality on its own, but its real power is its ability to kind of integrate with other offerings, right? If you install Istio today, you get a lot of things out of the box. We come with a certificate of authority. We can admit some metrics in the Prometheus format and a few other formats, but the real value comes when you start plugging things into all your existing infrastructure, right? You already have some enterprise-grade certificate authority that you want to use to now provision and mess certificates. We want to make sure that that integration is seamless and easy to do. Similarly, there's a lot of integration work on the telemetry and observability side, right, with things like Keali, Prometheus, and one of the big new ones is open telemetry, which is kind of an effort to consolidate the disparate telemetry, tracing, logging landscape under one project with the unified set of schemas and protocols and et cetera. So we've been investing a lot of effort into open telemetry and making sure that Istio can really easily fit into an existing platform that's utilizing open telemetry. Similarly, we have a lot of orchestration tools like Argo CD, Flux, and many others, and Helm kind of in the same area. We want to make sure that using those, Istio is easy to deploy using these tools, and that Istio can also augment these tools to make them more powerful, right? So there's kind of a double integration there. Like, Istio can be used to provide secondary routing features that can be used to safely roll out your own applications. And the same tools can also be used to safely roll out new versions of Istio or to enable Istio for workloads. Finally, there's a lot of existing standards that we're working on closely integrating with Wasm as a big one. There's been a lot of work in the Wasm area for supporting Istio as well as kind of the specifications around that. And open telemetry, again, I mentioned, open telemetry does a lot of things, so it's both a standard and kind of a protocol. We're looking at it from all angles. With that, we have a few more minutes left that were open for Q&A. Right now, it's just me and Eric here from the TSC. In about 30 minutes, we have a whole session with everyone on the TSC that's around for a just dedicated Q&A. So if you don't get your question answered now, we will bring all those questions to that session and answer them. And we'll have, you know, a lot more folks involved there as well. Yeah, so I also asked a few questions of my own on the chat about what were your favorite changes in the last year and what changes are you looking forward to most in the next year. If you're interested, we'd love to hear feedback of how users are using Istio. It's always great to hear from users. Otherwise, we'll end up getting back to that coffee break that got skipped due to the late start. Yep, yeah. If you have any questions, start thinking about them now and come back and join us at, I think, about 50 minutes from now for the Q&A talk. We happen to answer any questions you have on Istio. I think the talk is focused on ambient, but if we have time or, you know, we can definitely talk about any Istio topic. Alright, thanks everyone.