 Live from Austin, Texas, it's theCUBE. Covering DockerCon 2017, brought to you by Docker and support from its ecosystem partners. Hi, welcome back. I'm Stu Miniman, and this is SiliconANGLE Media's theCUBE, Worldwide Leader in Live Enterprise Tech Coverage here at DockerCon 2017, Austin, Texas. Happy to have on the program. Guess that's been on before. Mazda Marvasti, who's the CEO of AppLariat, and has brought along a managed service provider of his, also customer of his, Thomas Champty, who's the founder of K-Mocro Technologies. Gentlemen, thanks so much for joining me. Thank you for having me. All right, so you're both from Orange County, come down here to Austin. I have some local friends of mine that are like, God, all these tech conferences are coming down here. I'm like, this seems to be where open source kind of gathers. There's the OZCon coming a couple of weeks. I'll be back later this year for theCUBECon, come there. So, first time for both of you. Just quick, how's the show been for you so far? So, it's been really well received, I mean, in terms of our product announcements and our company, and not just open source, enterprise, right? Enterprise applications are all over the place, right? So, for us, doing a launch at DockerCon was very appropriate because our company is all about taking your existing applications, new applications, and taking them through the journey of not just into the cloud, but into the container era, running on a modern container environment. So, it's been well received so far. Yeah, Mazda, I think some really interesting stuff this week. And even the last few months when I've been hiding here in the cloud space, we've been talking about microservices and cloud native and everything like that. But, you know, customers have, you know, you talk of the enterprise, you know, they've got hundreds, if not thousands, of existing applications. Never, what about them? You know, Bengal have got on stage that it's not a bimodal world, we need to give them a platform to be able to move forward. So, I think that ties into the vision of your company, right? Oh, absolutely. I mean, the real problems and the real opportunities are with the existing applications. How do you get those applications, same kinds of capabilities, that you're trying to give brand new applications being done, you know, with microservices? And that's why Thomas with KMicro, their managed service provider, that they're working with customers along the same journey. Yeah, Thomas, before we get into the solution with that, Larry, just give us the thumbnail, you know, your background and your company. Well, basically, we, KMicro as a company, is a three-year-old company, we do managed services for mid-to-large-sized customers, obviously, and I myself have been doing IT for about 20 years, not the fresh guy on the block. But with these, you know, we were introduced to Apple area, when we were looking to do a little project for a client of ours, they had a lot of, they had a specific legacy application that they needed to modernize. So we were looking for a solution for them that would help them modernize the application at the same time, stay up-to-date with the latest technologies. And Docker seemed, you know, in addition to containerizing all their application was the solution, was the way to go for them. And that's when we were introduced to Apple area and worked together on that project, and we've been working with them since then, and we love working with these guys with it. Their solution seems to solve a huge issue for a lot of our clients, and we look forward to actually doing that again and again for more clients. Yeah, I mean, the challenge here is, how does KMicro repeatedly do that across customers? So you have consistency, which is the same thing when you look at an enterprise. Yeah, I can handcraft an application. So that it's containerized and I can deploy it, but how do I get repeatability and consistency across different teams across my applications? That's the scale issue, that you can't use muscle memory to move your applications into the cloud. Yeah, I'm actually totally right. I've watched the kind of app modernization environment, and when you go say it's like, oh, what's the killer app? What's the number one thing? It's custom, and that's not scalable. That's not repeatable. So maybe unpack for us a little bit about what you guys bring that journey along there and how you fit in that spectrum. Right, so again, the key is not to be reliant upon software artists to sit there and script away per application, how does this thing get containerized? What you want to bring is automation and uniformity across these applications, so that you have the same type of consistency for the applications, and then not only that, but when you deploy it, how does the orchestrator manage those deployed containers? Because if you're not going to run it on an orchestrator, it becomes really difficult to manage it. So we use Kubernetes from the back end to actually orchestrate these containers, but then you have to have a policy layer that manages that entire infrastructure. Kubernetes is great at knowing what to do. It just doesn't know when to do it. So we provide that when capability. Again, that's a solution that a managed service provider can essentially say, okay, now I can repeatedly do this across my customers and manage the environments in a similar manner, right? Versus having to do one-offs. And that's what we're trying to get away from. Thomas, could you maybe, if it's possible, walk us through a little bit that project with the customer, how long it took, how many people were involved, where some of the heavy lifting is there and anything that you've done this that say, oh, okay, here's what we've learned for kind of the next time. So one of the clients is a GPS tracking company and they had a DevOps team that were struggling, maintaining their code, and they had code all over the place. And Mazdo is more involved in on this project than I might have been, so he'll chime in on that. But what we walked out at the end with this client is that they had a platform that all, whether DevOps could go to and test their code against without having to spend a tremendous amount of time or efforts into putting all the pieces together. The application will actually do all of that for you from A to Z. They didn't have to worry about storage or processing or memory or any of that stuff. It was just there for them to use. In a second, or a few seconds, they could just spin up an environment, test their code against it, and all of a sudden, when they're done with it, it just shuts down, and it scales up as well. So if they needed to do some testing against an application or a code, they would do the test and then shut down the environment and they're ready to go and move on. Right, I mean the primary use case was that I'm spending way too much money on AWS for dev tests, right? Because developers would go spin up their VMs and use it to develop and do some unit testing and then the VMs would just live there, right? And so on an ongoing, continuous basis. So once we put the AWS under a Kubernetes cluster, we're then able to manage the cluster size based on usage and availability, et cetera, right? So not only that, but then the IT side of the house is able to govern that environment for the developers. The developers don't care to go provision machines or use IaaS or anything like that. They just want to deploy their applications, be able to test it and go back and modify code. So that's all they care about. Well, IT cares about where does that code go? Who looks at it? How does it get tested? What is a cost infrastructure for that? So by using our product to manage that AWS cluster, they were able to save 50% on their AWS costs by just managing that dev test environment. Now we're moving it into production. So those are the kinds of use cases that really you find containerization can bring, right? You talk about bringing new capabilities to the apps, right? But it also really goes to cost savings as well. And that was something that... 50% is probably conservative, even I would think. We had one of the keynote case studies this morning. I think it was Visa. It was like 90% of my environment is like being utilized less than 10%. I mean, we know that. And heck, I think back, we had server sprawl. We had VM sprawl. Now we have kind of cloud sprawl. Is it the whole API economy that allows your software to be able to plug into this, manage some of these environments? How much of it is just cloud? How much is it containers allowing us to do some of this? So every layer we talk about is another layer of abstraction. Cloud is an abstraction. Container is an abstraction. Orchestration is an abstraction. So every time you bring these abstractions, you're introducing inefficiency into the system, right? It brings efficiency in terms of how you develop, how much more secure your application is, how easily you can bring your applications up and down. It brings inefficiencies in terms of these things living on in terms of sprawl along the cloud boundaries that your applications are running. So we bring the efficiency in terms of policies that says who gets to have what, where, and when does it die, right? Because nothing should live forever. Unless it's your production app. It should not live forever. And when should it die? Expiring those applications and then reclaiming the environment to a smaller size so that your costs are lower or giving the resource to someone else who may need to use it. Mazda, Docker's made a real emphasis this week about the ecosystem. Can you talk about really partnering with them? How easy it is to work with them? You know, what you're seeing over the last year or so. Yeah, we're primarily using the open source Docker containerization mechanism because that's really prevalent in the marketplace. And we found the same thing in terms of using Kubernetes. Customers are looking for having a degree of control, right? Close source at a level where your application is running at this point is becoming really difficult, right? For customers to kind of be able to manage that ability to be able to say, okay, I don't have control over this part of my environment. That doesn't make any sense to me. So the open source community has really come along in terms of moving the enterprises into their direction. What we're doing is that we're leveraging those open source elements. So at the end of the day, we are not in the interconnect between the customer and their applications. They can always go back to those open source tools to manage the applications however they want. We're providing the orchestration layer that sits on top of those open source tools, really unifying them and bringing them together. Thomas, what is this whole containerization and your partnership with Applire that mean to your business? How do you expect it to kind of change what you guys are doing? Well, there's a couple of things. One of them is modernizing the applications that we talked about a minute ago. But another thing that we wanted to bring on the table and use the solution for was disaster recovery. No one had thought of real time disaster recovery without the efforts of going through a whole bunch of configuration and maintaining a whole bunch of things just to have disaster recovery for a company, for a company no matter what cloud you're in, no matter what infrastructure you use, it doesn't matter. And that's when Applire yet we think is a really good platform that we're going to build on to actually provide what we call real time disaster recovery services for our clients. And I think that'll take us, that'll be pretty good for our clients and it's going to help us grow our business as well. Yeah, that's another one of those areas where you see a lot of managed service providers providing disaster recovery services. But what that means is that they'll provide you a location in the cloud for you to have your application and then services along the lines of people in bringing those applications to run over there. And then when there is a disaster, have to go through a lot of manual effort to get that site up and running, right? So what we're talking about is that if you have your applications already containerized, you can snapshot it anywhere, right? And once it's snapshot it, it can come up and go down very quickly. So having that ability in terms of being able to provide disaster recovery services on a containerized application, it's a whole new set of capabilities that now I think it's viable for organizations. Let alone the cost savings as well, like you were talking about, the cost savings are huge. I only need to spin up the environment when needed. It doesn't have to sit idle sitting there costing us money or costing the client. And do you see passing that savings onto your customer? Absolutely, absolutely. I mean, that's one of the areas that was untapped in the past and with dockering containerizing applications, technologies, this is the next thing. I mean, this is the future of doing disaster recovery. We see it that way. All right, Mazda, oh, sorry. Yeah, Norris. Mazda, want to give you the final word. When people leave DockerCon 2017, what do you want them to know about AppLariat and the solutions you're providing? Yeah, what I want them to know is that you don't need to become a scripting ninja to be able to containerize our applications and bring cloud native capabilities to your existing applications. You don't need to become a scripting magician. You can take your existing applications through a very easy process, get them containerized, deploy it on Kubernetes without having to know anything about the underlying infrastructure that actually takes for those platforms to run. And then on an ongoing basis, as newer technologies comes along, will be the abstraction layer in front of the application for those customers so that they don't need to bother. Do I need to make a bet on something? Do I need to learn new technologies? Do I need to upgrade my people? Keep doing what you're doing. Your existing CI-CD process stays as is. Your existing developer work levels and skill set remains as is. Everything else should be abstracted and taken away from you. Awesome, Thomas. Thank you so much for joining us. We'll be back with lots more programming. Thanks for watching theCUBE.