 Live from Seattle, Washington. It's theCUBE on the ground. Covering KubeCon 2016. Brought to you by the Linux Foundation and Red Hat. Here's your host, John Furrier. Okay, we are here in Seattle for KubeCon Cloud ContainerCon. We are here, Cloud NativeCon, I should say, in Seattle. It's theCUBE special on the ground. I'm John Furrier, getting all the action from the event. Our next guest is Mark Ochepe, who's the engineering manager at Canonical. Welcome to theCUBE. Thank you for having me. So you've been, before you joined the company, you've been a community leader. Ascubuntu has been the community you've been working with. I got to get your take on something. This show here, KubeCon, has now included Cloud NativeCon, which kind of gives it more range and its focus. But it's got a vibe that's much different than it was last year. Obviously the numbers are huge. What's your take? Because this seems to be kind of a coming out party for this explosion of energy from software guys. Absolutely, it's fantastic to see. I'm really happy to see just the breadth of people coming out, the ecosystem that's being formed around Kubernetes and this whole idea of the Cloud Native Foundation. It's really exciting to see the participation, not just from programmers and software engineers, but from a lot of different people. It's very exciting. So the other thing that we've been observing on theCUBE, all the events we've gone past few years, especially with the DockerCon and with Docker coming out with all the momentum, is containers kind of lifts the fog, if you will, and people can start to see a clear path with containers. And now with Kubernetes growth around the orchestration side of it and the management of microservices, it opens up the market. How has that changed the game for developers? And what is the impact to existing projects, people in the stack, like the past layer? What's your observation on that? Sure, so Docker for sure and Docker-like things where it's really easy for developers to package up their content and deliver it out into either a testing environment or ultimately into production as really sped up the cycles of developers that are producing software. It still incurs the problems that we see a lot in these communities, which is when you start actually dealing with software at scale, the operations is really the cost that people invest their time into. So for a lot of those things, Docker sure has sped up the consumption of software, but we're still left in a vacuum a bit about how do you operate that software? It's something we're really interested in solving for concerns. Yeah, that seems to be the number one conversation, operationalizing both within the community and in the enterprise. What are you guys doing with Kubernetes? What's the latest update, status, coolest thing you guys are doing? What are the key things you're working on right now? Absolutely, for sure. So Concle itself and the Ubuntu project, we've been basically born into containers, containers in our DNA from being one of the first platforms to support Docker all the way through to being the largest and most used distribution of Linux on top of any public cloud. Most private clouds were the most built on top of platform. We really understand the idea of clouds and ultimately how containers work on clouds. So for Kubernetes project, we have the ability now to distill down the operations of Kubernetes in a way that makes it really easy to reliably and repeatedly run Kubernetes. All of this work that we've done in open sourcing our operations of running Kubernetes over a year of working with our customers, working internally as well and running Kubernetes, we basically distill down into open source code that's consumable, it's upstream and tree, and it makes it really easy to operate Kubernetes so that you can get that ability to operate software like Kubernetes at scale but still get the speed that you need for developers to deliver their platforms, their pieces, their software quickly into that. It's interesting, the interesting pattern I'm seeing in all the conversations I'm having is exactly like what you're saying is that people have so much experience dealing with open source, they see this formation happening and they stand up and quickly align with it. So there's a lot of alignment with Kubernetes, still a lot of work to be done but this doesn't seem to be any friction, big friction in my mind around where Kubernetes fits with pre-existing open source projects. Your thoughts on that and impact to what needs to get done next. Well, so we see a pretty interesting trend within our customers and within the larger community where all new applications that developers are building, they're building it on top of Docker. However, there is left that legacy idea of infrastructure, how do I lift and shift my legacy, VMs, my legacy bare metal into machines that are either Docker or Docker-like. For those who are seeing a lot of uptake with things like Linux containers which are the same primitives you get from a Docker machine except instead of it being a single process, you get an entire machine modeled as a Linux container with lightweight, super fast, durable and 13 times the density than things like KVM. So between those two stories, we see a lot of organizations moving to containers where they're lifting and shifting their traditional applications onto Linux containers, machine containers. And then we're seeing new development being done with Docker and Docker-like platforms for delivering the next generation of software in a way that's really quick and reliable. It's interesting, you guys are so close to the action so I'll ask the question as if I'm a customer. Hey, you know, it's so confusing. I don't know what container version of OS this and that and Docker does this and CoreOS does that and people are forking this and there's a core, common core, what the hell do I do? I don't know what to do. So how do end-user customers get the data? I mean, how do they figure out what to do? Well, the best way that we can tell people what to do is really not to tell them exactly what to do but to support the experimentation and trial of each piece of software. Everything that's coming out, whether it's Docker, whether it's all these threats of Forks of Docker or different variations that fulfill the same needs, it's best to evaluate those pieces of platform with your needs and we work with our customers and the community at large to help describe each of them having their pros and cons and give them the platform tools to make it so that it's really quick for them to experiment trial and then ultimately choose a platform forward for that. So that's your approach. You guys take that approach to say, here's a first wave tranche of things to do and to experiment. So it's really more of getting that, the sea legs, if you will. Absolutely, and there is no one overall solution that wins for everyone. Everyone has unique challenges and requirements and most of the tools available, especially in this idea of Dockerized containerized worlds, there's pros and cons for each of those. Certainly the buzz of this show has been phenomenal. The hallway conversations have been some of the best I've been to in years in terms of real energy and real action, meaning technical, people are riffing on ideas and a lot of inventions, entrepreneurial activity going on here. What's needed? What's next? As this continues to grow, I'm sure we'll start getting into the growth side of this where we're seeing the job growth is up. What's next? What needs to get done? What are the key work areas that need to be focused on first in your opinion? For Kubernetes? Yeah, for Kubernetes. Well, in our experience around Kubernetes, it's a great platform for managing the containerization of those processes. The things that I see people moving forward next is how to get Kubernetes out in the hands of people quicker for evaluation. And then ultimately comes down to how do you make it a really robust and reliable platform like the platforms that've come before it? Diving into things like security, extensibility and making sure they harvest and really promote a healthy ecosystem around it. Without that, it'll get segmentation. And then ultimately, it's not very good for a community when you just have a few entrenched partners and players into the world in the ecosystem. Share some experiences that you guys have had with your customers with Kubernetes, cloud native. What are some of the experiences and changes and shifts that have been positive for customers? It's very positive when we go into a customer location or even internally for ourselves, standing up just a couple of machines, two or three, and seeing that grow very rapidly as customers engage and realize this is something that can really work for them. So watching these small engagements, butterfly and blossom into actually quite large deployments with multiple workers, large scale deployments where they're starting to gear up and getting ready into production, it's quite exciting to see that traction. What's your take on the drivers of the Kubernetes container, revolution, evolution, and some of the inhibitors that customers have? There's definitely some, there's definitely a lot of it's around education. It's a different way of thinking when you start entering a world of process containers and dockerization. So a lot of that is just making sure that we educate customers and users on what it means, the right way to do this, how workloads are. It's a different way of thinking from microservice architectures all the way through how you develop the code for it. The benefits are quite clear. The easy way of distributing binaries and applications, the quick run ramp from developer all the way through to the end system makes it really tractable in that sense. It speeds up that process by a lot. All right, Marko, thanks for coming on theCUBE. Really appreciate it. Thanks for sharing the insight. Any thing you want to share in terms of coordinates where people can get involved with some of the things you're doing with Kubernetes you would like to share? Absolutely. We're quite president in a lot of the SIGs here, but all of our source, again, is upstream in the cluster directory and then we have Ubuntu.com slash cloud slash Kubernetes for more information on what we're doing about distilling the operations of Kubernetes. Asky Ubuntu.com is your site, is that your site? Yes, Asky Ubuntu.com is a stack exchange powered site. So if you're familiar with Stack Overflow and all the developers out there, Asky Ubuntu is a stack exchange community run website. It's separate from Ubuntu and Canonical, but there is a great Q and A location there. So if you have any questions about Kubernetes, Ubuntu or the breadth of our projects, it's a great place to ask. I'm Marko Cepa here in theCUBE on the ground. I'm John Furrier. Thanks for watching.