 Live from Los Angeles, it's theCUBE, covering Open Source Summit North America 2017, brought to you by the Linux Foundation and Red Hat. Okay, welcome back everyone. We are live in Los Angeles, California for theCUBE's special coverage of the Open Source Summit North America with the Linux Foundation. I'm John Furrier, host with Stu Miniman, my co-host, our next guest is Chris Wright, Vice President Chief Technologist, Office of the CTO at Office of Technology at Red Hat. Welcome back, good to see you. Thank you. So this is your world, everything's going down, you got container madness, you got the Cloud Native Foundation on hot as could be. Everyone's joining Cloud Native, looks like it's looking in the middle of the fairway, as they say in golf. Everyone loves Cloud Native. So what's your take on this Open Source Summit? It's a big tent event, it's kind of a celebration from the roots of Red Hat, really the success of Open Source. Well it's really fun to be here, partly it's the people, you develop relationships for me it's been, I guess since the late 90s, early 2000 time frame, so that's a long time that I've known some of these people that I'm running into here and then balance that with kind of the influx of new folks, which is critical to the longevity of all these projects that we're working on together. It's a fun event and I think it's cool to see the, call it rebranding from, was LinuxCon and then LinuxCon Cloud Open and LinuxCon ContainerCon and now Open Source Summit really kind of brings it all together. It's a big tent event. It really puts a big tent around the trend, which is pretty obvious from Jim Zemlin's talk this morning. By 2026, it's literally from 64 million libraries to 400 million Open Source libraries, the code sandwich now where literally 90% of most IP is going to be open and 10% could be your individual work product, your unique IP that people will be doing the original idea. So this is a completely beginning of a growth market. Who would've thought? You think we're already winning, right? So like there's more to go. So you guys have been there from the beginning. How do you rationalize that? What's going on in your mind as you look at the future and say, okay, there's a lot of moving parts still of people, people business, it's still community-based? It's kind of daunting. What's your reaction to those numbers? Well, I mean it shows the power of Open Source first and foremost. I think one of the things that's interesting is it's really easier and easier to get involved in Open Source and especially creating new projects. And that in and of itself brings on a balance and even a tension of it's so easy to create something new, we also have to balance that with focus on what we've already created and evolve. So if everything is green field and created fresh, then we're not really leveraging everything that we've built so far. But if we're kind of trapped by what we've already built, then we can't do that fresh innovation. And I think that's an interesting tension for the industry. And the new developers that are coming on board, they're being attracted by the legends in Open Source, you guys and your particular other luminaries. But these guys don't have a lot of them institutional, I won't say baggage, but knowledge. Like, hey, I just want to write apps. Infrastructure, they gravitate towards containers, they gravitate towards Kubernetes. So of course the new blood coming in loves the container stuff, loves the Kubernetes. So that's a big factor in the popularity and one of the factors. Huge, developer ease of use is an ability to move quickly, is a big driving force behind that container space. And I kind of look at it as a spectrum of back in the day, we had our special handcrafted artisan servers where you're supplying the entire operating system, you're keeping it up to date with patches, you're including your application, plus it's runtime and runtime dependencies all together. You can just lift that and put it in a virtual machine and you can kind of move forward into a more cloud-based virtual machine where you get some stateless behavior. And as you're moving along, the developers are responsible for a little less of that underlying infrastructure. You get to a container and you're starting to focus more on the application as direct runtime and application dependencies. Focus being for the developer, writing code, and for their company that they're working for, creating value for them. A new artisan role, which is creating a great product. Yeah, yeah. And you could even roll that all the way forward to say serverless fits in that spectrum and there you're just focused on your business logic and letting the underlying platform really take the rest of the responsibility away from you as a developer. So on the one hand, we have DevOps bringing things together, on the other hand, we're really enabling specialization. So operations teams can build platforms, developers can write applications. Yeah, interesting, that interesting dichotomy. We talked about everything collapsing down. Red Hat has a software-defined version of every piece of infrastructure, yet really what you're doing is building distributed systems where everything is disaggregated. How do you really, as you build this technology, balance that? Where does Red Hat put its focus? How do you fit in with what the foundation's doing? Initially, we're customer-driven, so we're looking at what are the key requirements from our customers. In that space, I love that you brought up distributed systems because from my perspective, distributed systems are non-trivial, it's a challenging space to work in. One of the things that we're trying to do is make distributed systems accessible to the broad enterprise developer population. So one of the things you see in a platform like a container orchestration platform like Kubernetes is the ability to schedule your workloads around a cluster so you don't have to do the cluster management. The next generation will see bringing, as you kind of decompose your application to more and more services, the network becomes a core central part of your application. So recombining your applications to the network is a critical part of your application and bringing that network topology, network definition down into the platform so that it's not a key consideration of the application developers, again, making those distributed systems accessible to the app. Chris, I want you to take us inside the open container initiative, OCI, the 1.0 release is done. Red Hat's been working in containers about a decade now. Yeah. Some people look at it and be like, wait, it's just a container format. Is this a pretty granular level? Why is this important to be able to building the systems that you've been talking about? Well, we're working in major industry trends at this point and containers are a huge part of the industry and having some standardization so that we know that images are functional from a format perspective with runtimes supporting a specification for the runtime and the image format, independent of the orchestration platform really helps build that ubiquity that we're going for with containerization and containerized application workloads so that we can be confident when you're building, developing your container image that you know it will run in a standards compliant environment. So I think it's really important. It's also interesting. We've done a lot of work in open source to create what I would call de facto standardization where everybody uses the same platform or tool like Linux. And now we're looking at how do you create some sort of formalized standardization without going into full blown standards bodies and kind of killing the momentum of open source development. Yeah. You mentioned serverless. I have to ask you your opinion on that because today there's a lot of different options for serverless. There's a couple of open source versions but predominant one, we look at AWS Lambda, not open source, I can use Azure functions. Where do you think we need to go with serverless and how does the work that you're doing in containers lead into that? Well, container environment is a great environment to host the runtime that's supporting the serverless environment. So serverless obviously misnomer because there's a server involved just that you're not managing it as the application developer. So it makes a lot of sense to launch a container with the core runtime and load the application content into that runtime as part of a serverless environment. Creating some portability across different cloud platforms I think will be really important for a whole class of I know our customers. And so using something like AWS Lambda is going to really tie you into the AWS infrastructure. So if that works for you, great. If not, you need to look outside to an open source platform that can run independently of the underlying infrastructure. So for us, we've been spending a lot of time doing work in the open-wisk community, trying to bridge that into Kubernetes. So there's a good relationship. So you got to contain orchestration to launch the serverless environment managed by open-wisk. And then the next question is how do you get events triggered out of whether it's your own application stack or the external cloud that you're running on, events triggered to run the functions that you're writing in this functionless, function as a service environment? We're just red hat heading with all this. So if I'm going to boil it down and talk to a CXO and someone says, hey, I just ran into Chris on the Cube and what did he say? Well, boil it down for the CXO. What's that message to the C suite? Because, you know, they hear Kubernetes this, Pivotal Container Service that, VMware's got, it's like, okay, boil it down to me. I got red hat, I got RHEL, 15 years support, I've been using tier one open source, but I want to go to the serverless direction. I want DevOps. What does all this mean? How do you bottom line it for the CXO? Probably a couple core tenants, if you will. One, on the operations side, we're really trying to move towards simplifying operations so that operations runs more policy-defined. On the developer side, we're trying to simplify the developer's life so that what you're developing is directly translating to business value. You're really trying to build something for your company. To do that, we're building standardized platform. Kubernetes being a primary example. That platform of Kubernetes is actually orchestrating applications that are Linux applications running in a Linux environment. And from a red hat point of view, we've been invested in Linux for quite some time, so we're really confident in our ability to build up that application stack that was an application sitting on a single instance of Linux now distributed across the data center, managed with Kubernetes as the standard platform, but still looking at Linux as the real host for the actual application and the application runtime. So you say, basically, the paraphrase, I'm the sales guy, or I'm the consultant or architect. Look it, we've been doing Linux for decades. We're your guys for Kubernetes. We have a standard. We want to make sure it works for you. That's right. Containers are Linux. No one wants to have orchestration of stuff that needs to be managed. I mean, it's got to work. That's right. We've got to be debugged. Developers are going to break stuff, right? That's a whole, a really important part of just the distributed system world is insight into what's happening in the application. So if you- It's a black box. No one's going to buy into that. We've got to tap into that. So this is just one thing I said earlier. I want to get your thoughts on this. Is the Kubernetes container business then the better mousetrap will win? Because that seems to be what I keep hearing. It's like, yeah, I got this, but at the end of the day, ops guys just don't want stuff to break. That's right. So it's like, the better work, the better product in this case will win. And I think it's- You agree. What's supportable from the ops team? So they're looking at deploying something that needs to be manageable so there's simplicity of the management there. Also, who's going to stand behind that? So if you're working in an environment where you have a vendor, obviously we think that support and services are a really important part of keeping that up and running in mission-critical environments, that ops team is trying to keep a platform stable to run applications. The app team is really trying to build those applications as quickly and move them as quickly as possible to address market requirements or opportunities or threats. And that's where we're going. So Chris, want to circle back to OCI. 1.0 is out. Yes. How's Red Hat feel about what's in there and what's still to come? Well, it's a great starting point. So we've got core image format and runtime sort of specified. We had some initial code associated with those specifications as part of the OCI project and also expectations that as we solidified that specification we get further adoption from other parts of the ecosystem and we're seeing the evidence of that happening today. So it's a great milestone. It's definitely the beginning. You see in Kubernetes different ways to manage container images that are OCI compliant. It's really important. We're excited about the CRIO project that we're working on within Kubernetes which is bringing OCI compliant to the container runtime in Kubernetes. So great first step. You were going to see the next steps of caring about the distribution, the kind of signing and verification of signatures, really making this a robust system but we had to get started and it took a while to get here and we're pretty excited even though I know we wish we got there sooner but it takes time to get everybody on the same page. Well Chris, it's great to see you again. I really feel like I was bummed out that I couldn't make the Red Hat summit this year. Stu handled it before I was in Boston. I was in California. It's great. You guys are doing great. Great to see the continued success of Red Hat. Again, another generation of growth coming down. Congratulations. Thank you. Chris Wright, the VP and Chief Technologist at Red Hat. Inside theCUBE here for day one of open source summit, the new event at the big tent event for the open source community powered by the Linux foundations, theCUBE's exclusive coverage. I'm John Furrier, Stu Miniman back with more live coverage after this short break.