 from Austin, Texas. It's theCUBE, covering KubeCon and CloudNativeCon 2017. Brought to you by Red Hat, the Linux Foundation, and theCUBE's ecosystem partners. Welcome back to theCUBE, SiliconANGLE Media's two-day live production of KubeCon and CloudNativeCon. I'm Stu Miniman, my co-host for this segment is Matt Broberg. Happy to welcome back to the program, Clayton Coleman, who's the architect of containerized application infrastructure with Red Hat. Clayton, great to see you. It's great to see you too. All right, so first of all, 4,100 people. You impressed? I am hugely impressed every year this gets bigger and bigger. The community is out in force, people building on top of Kubernetes and in the CloudNative ecosystem around in force, and it's really phenomenal. Yeah, so, John, for your interview to last year at the Seattle show, I think it's what, triple the size, the number of projects have gone from four to 14, but at the core. I mean, it's Kubernetes, and you spend quite a lot of your time. Tell us what have you been working on the last year and what's important in your life? Well, I think the biggest things that we've really tried to focus on are making Kubernetes a good foundation for both the community as well as for a technology stack because Kubernetes is about empowering developers, it's about empowering operations teams, and we always anticipated there to be many levels and many ways of building on top of Kubernetes to make it an ecosystem so that people can build and deploy software, but other people besides us can succeed. And I think that's more than anything else in the last year, it's about ensuring that everyone besides the Kubernetes community is successful, not just Kubernetes itself. Yeah, it's interesting. We think back to, like, Linux. It's, you know, Red Hat did quite well with Linux. Also, you know, from the enterprise standpoint, from the company, we appreciate what Red Hat did to make sure that Linux could be used by everyone. Seems like a lot of similar themes that we see, but how could you kind of compare and contrast Linux versus Kubernetes today? It's interesting, everyone is a lot more conscious of open source and the idea of building a platform because of the example of Linux. And so, we've tried to actually be pretty conscious about that, which is we want there to be a strong community, we want there to be a technical respect among not just the core of the project, but also the different layers. And the Cloud, the Cloud Native Foundation has actually done a really good job of bringing together mutually complementary technologies, but also helping and support those communities. From a Red Hat perspective, a lot of the things we work on are stability, security, reliability. We also work on extension because extension to us allows us both to support customers, but also to help the open source ecosystems that we depend on that are- Sorry, just for audience, can you explain what extension is? Sure, extension is actually, it's a number of things. In Kubernetes, we really want it to be possible. If we're going to build in Kubernetes things that make running applications easier, we want everyone else to build their own tools that make it easier to run applications. And we don't want to be opinionated in kind of the same way as maybe some other ecosystems about who gets to build what. Instead, we want to open the doors for vendors, for partners, for deployers, for individual users to build their own extensions and points of contact with Kubernetes to really solve their own problems. We can't solve all those problems for them. That plays really nicely into it, right? The Cloud Native Foundation's gone from four projects to 14 in just a year, and you're talking about the extensions. What do you want people to take away from that proliferation of projects that are all being supported and seen as essential to the ecosystem of Kubernetes? It takes a spectrum. We want everyone to be able to use Kubernetes and to use the other projects, either independently on their own. But I think a lot of us in the Kubernetes community and the CN-CF communities believe that a lot of these tools work really well together and finding new opportunities to make it easy to work together. So Prometheus is a great example. It's exploded across the ecosystem. I think at the last Cloud NativeCon, Prometheus was really the talk of the show. And I think what I've seen is that a lot of people around the ecosystem, not just in that core community, on a very specific project, have taken the ideas that underlie that technology and tried to apply it to other things that they were doing. So you see people building integrations into Prometheus. You see InfluxDB working with Prometheus to share data across. A lot of really exciting cross-collaboration. And the end goal really is to make building and running applications easier, which is something we really believe in as well. All right, you use the word a spectrum. When you talk about users out there, there's lots of them that are kind of in the one-on-one phase. We know there's people doing things through production. What are you seeing, kind of the help us walk through some of the spectrum as to where customers are, what you're seeing, some of the big challenges that they're facing? At Spectrum really there's no other word because the range of people using Kubernetes in production and development is so incredibly diverse. I would say the two extremes are people who are today deploying microservices-based production applications on public cloud and they're bringing three or four or 10 or 100 applications. It might be a two or three developer team and they're really finding a lot of value in that because Kubernetes is taking a lot of the heavy lifting and they can rely on that to keep their applications running and to rapidly deploy. On the complete other end is giant corporations, people with decades of investment in IT, finding ways to use Kubernetes and OpenShift which is the product that Red Hat ships around Kubernetes to empower tens of thousands or hundreds of thousands of applications and in those models, Kubernetes is just one small part of a larger whole and this is where ecosystem really comes into play. In the middle I think we're starting to see a lot of really exciting things as people have the, they've got their one team working together and as they start bridging out and bringing other teams as companies grow, as they say, find more reasons to use Kubernetes, they start asking questions like, well how do I have all these teams working together without impacting the other teams? And that's where multi-tenancy and that's a real specialty for Red Hat and OpenShift is multi-tenancy and we're actually really excited to work with people in the community to build out these technologies at many different levels to have kind of that spectrum to spread from the middle as well. One of the things coming into this show, the last year or two it was like, okay, who's going to win kind of the orchestration battle? And it's like, okay, Kubernetes, here it is. Well, now there's like 42 different providers, OpenShift being one of them. Where does Red Hat look to add value to the customers? Is it just a piece of the platform? How does Red Hat look at it and how do customers, when do they come to you? When do they say, oh wait, I'm just going to go build all my own pieces and use some of the Red Hat pieces? If working with OpenSource and Linux has taught us anything, it's that one of the key components of a successful story is a distribution. The idea of curating, making a few choices, making it easy to bring things into that distribution and we've actually started to really apply the distribution mindset to Kubernetes. So if you look at OpenShift, it is a platform. It has tools that help you run tens of thousands of applications together with tens of thousands of users to bring operational control. But as hard, it's about taking the best technologies in the community and bringing them together. And so I would actually expect over the next year or two to start seeing the idea of the distribution emerge in Kubernetes in the cloud native ecosystem where it's not ever going to just be one company dominating OpenSource. That's not how OpenSource works. I would expect to see an effort in thinking about Kubernetes as the core, the kernel, if you will, and bringing together all of the successful technologies like the ones that we've seen at Cloud NativeCon here today and bringing even more of them, letting people mix and match to find the solutions that work for them. I really like that view of it too because you're saying that the OpenSource at its core is open and not opinionated while distributions aren't outlets who have opinionated and refined business problems. So how do you see that playing out a little bit? There's always going to be some trade-off when you make choices for people. And so I think the way that we look at it is we try to make choices that make sense when you're dealing at certain scales, when long-term support and lifecycle becomes really critical. If you can't afford a production outage because you have 10,000 applications running together, then it becomes really important to focus on those. But at the same time, we actually expect there to be different choices and trade-offs to be made and we'll want to actually encourage people to mix and match the different parts of the ecosystem. And what patterns are you seeing in enterprise readiness or any enterprise feature sets that are combining into what you hope to see out of a distribution? At the heart of its security tends to come up a lot. You know, everybody who's making the leap from, we made the leap from bare metal to virtualization and then a large number of management platforms grew to encompass it and virtualization brought its own changes. Containers were starting to mature and how we understand how the software lifecycle works with containers, how it works in large, multi-tenant environments. I think the next step will be, as we become more mature, that a lot of these patterns will be baked in and so you'll see standard solutions. We all kind of need to work together to make those standard solutions happen. We're actually seeing that in a number of the things. Even today, I'll talk about the CNCF conformance profile for Kubernetes. It's a new effort that intends to take the tests that we use in Kubernetes to make sure it's working correctly and use that to say, this is something that you can rely on every Kubernetes distribution also supporting. And just like any other mechanism that we use to make sure that we're delivering something that is stable and predictable across a wide number of spaces, I would expect in the future to see things like conformance for multi-tenancy, conformance for security specifically in CUBE and to see vendors bring their own approaches, partners, ecosystem, players, integrating their solutions and then new open-source solutions fitting into that as well. The keynote this morning, there were a lot of these projects, getting to the next rev. Kubernetes is going to be at 1.9, many of the underlying kind of supporting pieces are hitting kind of the 1.0 out there. Your top contributor for Kubernetes, what's that experience like today? Lots of new people still coming on. How's the balance of kind of the view that are heavily involved versus kind of the majority? When we started Kubernetes, it was a very, it was an interesting mix. It was a lot of engineers working on very concrete ideas, things that we'd want to try to bring to fruition together in the community. And it's been a very deliberate goal over the last two years to broaden that into a successful and healthy open-source ecosystem, which means a lot of mentoring, which means working to find the different ways that people can contribute in an ecosystem. Sarah Navani from Google often uses the chop wood carry water analogy. There's many different ways that people can work together and everyone has a spot. So we've spent a lot of time being very deliberate about being open, trying to organize ways for new contributors to get oriented and to bring their value. But at the same time, we actually want to mentor and grow the next level of technical leadership in Kubernetes. I won't be here forever, and I don't want to be here forever. I want people to replace me in the open-source community because that's a healthy community. Yeah, I think the stratification of contribution is one of the number one signs of success from my perspective. And I see a distinct different invitation for each type of user. So you have the user, you have the administrator, and then you also have a developer. Are there any things you've noticed changing in one of those patterns that really hits home for you? I think the developer pattern is the most interesting. There's a lot of focus on how do you use Kubernetes in many different ways, and a lot of developers want to get their hands on and dig in. And so there's actually been a lot of great community projects that are focused on making Kubernetes easy to consume at a small scale. All of that then ties back into, well, in Kubernetes, we want to be pretty opinionated. If we're going to be a kernel, there needs to be a space for things like the compiler and the programming language distributions. I'm actually hoping that we can keep that focus on making sure there's a good set of projects in the ecosystem that meet developers where they are so that they can start using Kubernetes. And then, I don't want to say trick, but trick them into becoming contributors and help us get that feedback about how we can make Kubernetes better. Helping to paint the fence is very fun. That's right. All right, Clayton, last question I have for you. You're doing two keynotes this week. Give our audience that won't be there in person. Give us a taste for that and especially want to hear kind of the outlook for the next 12 months through 2018. Sure, so my first keynote tomorrow is just a real quick one. I'm going to try and convince everybody that Kubernetes should be boring. And I'll leave it at that. Boring is good in very specific ways. Boring equals mature, right? I would certainly hope so. And on Friday, I'm going to talk about what's coming up in the Kubernetes ecosystem in 2018. A lot of people have finally jumped on board the Kubernetes bandwagon. And what I'd like to do is kind of help people find those exciting projects to get involved with. If we're going to have a vibrant ecosystem and community, helping people understand where they can get involved and to find the things that match their interest is going to be really important. Okay, anything specific that you're super excited about, looking forward to next year, any project or area? I've got to say, and this is not a company line, but Istio is incredibly exciting because one of our goals with Kubernetes was always about making it easier to run applications. And Istio and the idea of service mesh is taking that to the next level. And I actually hope to see even more projects like that over the next few years in the ecosystem that solve things like serverless and database as a service. And I think we're actually starting to really see that develop. Yeah, well, companies are all looking to move faster, get those applications up and running. Istio definitely one of the ones we heard buzzing before the show. Clayton Coleman, thanks so much for joining us again. Hope to catch up with you again soon. For Matt Broberg, I'm Stu Miniman. We'll be back with lots more coverage here from theCUBE's coverage of KubeCon and CloudNativeCon 2017 here in Austin, Texas. You're watching theCUBE.