 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. Hi everyone, I'm John Furrier. We are here on the ground with theCUBE at KubeCon, not to be confused with theCUBE here. It's Kubernetes conference all called KubeCon and CloudNativeCon. Our next guest is Robert Hertzfeld, he's the CEO and founder of Racken, entrepreneur, native industry leader in this area. Been through OpenStack wars. Now we're in the Kubernetes wars, CloudNative wars. What's going on here? Tell us what's that, is this OpenStack event here? It certainly feels like it. Last year there were very few big vendors and this year everybody showed up. All the operating system providers, not the hardware people haven't shown up yet, but Cloud, everybody, it's here. I want to get your thoughts on CloudNative versus DevOps and all the challenges that are going on. And certainly joking aside, OpenStack and this community kind of have overlap. But I want to get your plug-in for your startup, Racken. What are you guys, because you're doing some interesting stuff I want to share out there. What are you guys working on? What's the problem that you're solving? What's the company all about? So Racken is really about workload portability. So we specialize in hybrid automation starting from the bottom, the very bottom, so metal and cloud. What we're really trying to do is solve what we call underlay problems. So when we look at Cloud today, a lot of things that are really hard are hard for everybody, building a public key infrastructure, doing load balancing. Hold on, before you go under, what does underlay mean? First to find what's the line where it's under? What it's under and what's over? Is Kubernetes under or over the line? So when you mean underlay, be specific. That's a great question. So Kubernetes is over. So a lot of people want to use Kubernetes because it hides all of the mess of infrastructure and automation and provisioning. And they just go to Kubernetes and it takes care of 80% of their problems, which is great, 80% is a lot of their problems. But there's a whole bunch of things that use the messy stuff underneath, provisioning metal and dealing with networking, real networking and security and all that stuff that's really ops. DNS as well, we saw the hacks coming in there. DNS is a great example. You have to have that stuff working, it has to be right. OpenStack really struggled with that underlay where the neutron, the networking layer for OpenStack has a hard reputation, bad reputation. A lot of that reputation was earned because it's just hard to install overlay networks correctly. It has nothing to do with the APIs and the code, it's the actual operational underlay to make that stuff work. And when you do that back. I mean the day-to-day operational dynamic environment that exists called the internet. We're all trying to hide, but that's what people want with OpenStack or Amazon or Google or Kubernetes is they want to hide that mess and they can. That's what's awesome. But you can't, it doesn't go away. It's just under the rug. And at some point somebody's sweeping up under the rug and cleaning it up. So what are you guys doing in the underlay? What are you attacking? What's the core problem? So that's our specialty. What we did was we built an automation platform it's called Digital Rebar that actually orchestrates and automates all of these underlay tasks as a unified thing. So we go all the way from booting and configuring Raiden BIOS all the way through installing Kubernetes and then we can even push things into Kubernetes. So we're really building this sort of- Skeleton, if you will. And the concrete, Rebar, I love the name, very clever. So you're really trying to create more structure so that you have some stability. Correct. Hence Rebar, concrete. It's foundational. So the idea here is that we're not trying to replace Kubernetes or OpenStack. What we're really trying to do is create an environment where they can be successful. And that's really what we felt like when we started OpenStack work operationally that was what was hard. It was building the environment around OpenStack. OpenStack actually work, it still does. A lot of marketing range with the Rebar cracks in the foundation. No pun intended, meaning all these open source foundations. But foundationally, this is key because the building blocks of cloud really is all about sequencing layers of evolution and on top of it, each other. So having a crack on the foundation really can become problematic and things can come crashing down. We've been struggling for years to try and build OpenStack upgrades. And the problem with OpenStack upgrades is that you don't have a good foundation. You have a good foundation. It's much more coherent to do an upgrade. One of the things we just announced with the Rebar framework was doing a public key infrastructure root rotation. It's a long string of technical stuff. But basically, if you need to change your security infrastructure, you have to roll out a change across a whole bunch of servers in a very coordinated, precise way. Those are hard things to do. They're not Kubernetes tasks, they're underlay tasks. So all the hubbub in the industry outside of Kubernetes and Docker, we see that as a massive scale point because there's more developers in that area in that category. The DevOps is a very, kind of, I won't say niche, but very targeted set of skillset. The professions are specific. How do you separate those professions like, obviously, app developers or app developers and they just want containers, they want to orchestrate the hell out of it and compose it, it sounds great. That's what they want. Kubernetes. They're not network guys, they don't want to deal with the mess. They don't, and they really don't have to. In some ways, the whole containerized revolution is this developer saying, look, I don't want to deal with writing scripts to deploy stuff. I just want to package my stuff, ship it to you and let you run it. Kubernetes does that actually pretty well. So DevOps has to find a comfortable ground with Kubernetes and containerized orchestration and things like that. And I think companies are going to find that ops teams are still doing this heavy lifting that they have to do and releasing developers from it. DevOps, in a way, went too far because we asked developers to do a whole bunch of operational work that they didn't want to do. And Kubernetes is, in some ways, a developer revolt from having to write DevOps scripts. Yeah, we all want to compose an orchestrate. That's the key. But I want to get your thoughts on this because we were talking before we came on camera about DevOps and Cloud Native. Can you describe the difference between what Cloud Native is and what DevOps is? So Cloud Native is really an architecture that you put into your applications and build them in a certain way so that they will get the benefits of the platform, scaling and HA and automatic upgrades. All those things come from doing Cloud Native development architecture. It's not DevOps. DevOps is really, there's a lot of workflow and lean and pipeline stuff, but generally people look at it as how do I automate my infrastructure to do these routine underlay tasks? So there's elements of it where you- They work hand-in-hand, but they're different. They work hand-in-hand, but Cloud Native really is a development architecture. DevOps never told you how to build an application. It told you that you should automate it and manufacture it in a certain way, but not how to construct it. That's a great, great, thanks for sharing that definition, it's awesome. Back to the underlay problems, again, because this is where the action is. Certainly, the security exposure is deep on that area. A lot of security holes back there, if not, just a classic example of patching, right? Worlds, that world still exists. You have DNS you mentioned we talked about. So there's a lot of important work in the foundation that needs to get done. Thoughts on priorities from your standpoint, how you guys attacked that with the rebar approach? So we attack it really by focusing on being very flexible and heterogeneous and allowing companies to be hybrid. When we talk to companies, nobody's single infrastructure. Everybody's in multiple infrastructures, they're moving between infrastructures, they've got to replatform. It's very frothy, it's incredibly frothy, and they don't know what to pick, and even if they pick the right Kubernetes, but they might pick the wrong software to find net storage, and so there's all of this churn. There's a playbook problem, no one knows which playbook is related to which. And they won't, because everybody's a little bit different, there's a lot of good choices. And so what we look at is how do you say you can make multiple choices and not break the whole infrastructure? And that's a lot of work. How do people engage with your company? Do you guys provide services? Is it software? Is it cloud? So we sell software licenses. It's our primary, we have enterprise support, RACAN is our company, four digital rebar, and then we will do some services and consulting. But our real goal is to provide this platform and then help people use it and expand. So we look for partners who want to add things into an overall deployment. What's your thoughts on cloud native con and cube con here, Kubernetes con here in Seattle? It's election day, it's sunny in Seattle. It's a unique day, but there's a lot of action up there. A lot of people here, we've talked about five, but there's still a lot of people here. It's tremendous. So the enthusiasm, there's a lot of users. There's people really doing real work with Kubernetes. The barriers to get started in Kubernetes are much lower than say an open stack where you really needed to run it on servers. So Kubernetes is super easy to get started in a small way, tip of the spear, get into your organization. And from that perspective, I think you're seeing a lot of people sort of get running with it and it's following the Docker wave. So people said, hey, this Docker stuff is cool. And then they got to attend containers and they said, wow, I need to coordinate and schedule and manage. And so we're in this interesting wave of transformation. It's a developing ecosystem. It's got growth to it, it's got good vibe to it. But I think there's a real danger of square pegging Kubernetes. In the echo chamber of the conference, everybody's got their apps and they're running Kubernetes and everything's great. But when you step outside of the conference in Enterprise, they're not there yet, right? They're getting in the VMs. The enterprises aren't there yet. The enterprises aren't there yet. They need, their applications aren't architected correctly. There's a lot more to do. And so I think we actually- So there's a lot of rah-rah in the hallways here. And it's well-deserved rah-rah. Yeah, absolutely. But then the operationalizing it, this is back to operationalizing things. The enterprises either aren't set up for it or aren't on the maturity curve yet. They have to be given the patterns. I mean, it's exactly, it's the playbooks. It's exactly what you were describing. That's how we help enterprises move into it. And it's remarkable how many of them are doing it and getting into that process and then it's gonna grow and expand. So I think it's gonna follow a different path than OpenStack did, because OpenStack was really a switch from A to B rather than, you know, I can incrementally land and expand. What's the skill set if you had to kind of look at this world? What's the kind of profession of the people here in terms of developer category? I see you seeing more and more business-minded developers who understand business value, but from a geek standpoint, what are we seeing here from a profession standpoint? So it's actually pretty broad range. You know, some of the things I'm seeing are a lot of Java developers coming into the platform using how OpenShift has been really good for Java developers to jump in. There's a ton of go development, but it doesn't really take a lot of expertise to use these platforms. You just have to be willing to adopt the architecture, right? And that's where people really have to do some learning. They have to figure out how you manage an application differently and re-architect it. I mean, we went through that process ourselves, and it took us time to figure out the pattern, but I don't think it's outside of any developer. I don't think you have to be a rockstar to figure out your capabilities. I'm a software developer, basically, you know. Just software development. Application development is back. I mean, so awesome. Rob, thanks so much for sharing. Congratulations on your startup. You did some funding. Any news on, you know, big mega VC round? You guys self-funded. Give us a quick status of the startup. So we've been doing seed. We've been announcing some incremental seed and some more coming. We'll be looking for a series A in the spring, so if you're an investor and interested in hearing more, please contact us. We are always, always looking for- I love the digital rebar brand. Thank you. That's phenomenal. Is that open source project? Is that your code? It is an Apache 2 open source project. So we are looking for community who wants to participate. And is that your project? Did you guys start that project? We founded it, and yeah, we've been maintaining it. Don't want to crack on the foundation. Rebar, great analogy, and certainly very relevant. Thank you. Thanks for sharing. We're here at KubeCon. I'm John Furrier. We're with theCUBE. We're on the ground.