 Hey, welcome back everybody. Jeff Frick here. We're in our Palo Alto studio for a cube conversation, getting deeper into virtualization and containers. And as we get ready for DockerCon that's coming just around the corner, it's fourth year. There's a lot of exciting things going on. And one of the exciting things is a new company that'll be launching there called Applariat. We're excited to be joined by Applariat CEO, Mazda Marvasti and former VMware CTO and Applariat advisor, Ben Fati. Gentlemen, welcome. Thank you. So Ben, why did you join these guys? What are you helping them with? What opportunity, you've been in the business a long, long time. Aren't all these problems solved yet? No, we create new ones. We create new ones, okay. Let's solve a few, create a few more. Job security. So Mazda and I worked together at VMware for several years and I initially managed the vSphere operating system and hypervisor. And Mazda was the CTO of the management business unit. He was responsible for the vision of how we're going to manage all these virtual machines that people are creating and the applications running on top of them. And then later on I became the CTO of the company and continued to work with Mazda. And one of the things that happened over the last year or so that we were at VMware together was the rise of containers and Docker and cloud native applications. And we actually worked together at VMware trying to integrate containers and Docker directly into vSphere to enable cloud native applications to run on vSphere environments. Anyway, make a long story short, so we both left and I became an advisor to several startups and Mazda started this company. So after about a year or so he came back and we had conversations about what he was trying to do. And meanwhile I'd gone off to a startup where I personally delivered software as a service using cloud native applications. So I built the operating system and was also trying to deliver applications in the cloud native format. So I felt the pain of what Mazda was trying to solve. And when he explained his approach of what he was trying to build, it was really, it resonated with me in that he was seeing the IT organizations and developers in existing enterprises struggling with this move to cloud native 12 factor applications and moving to the cloud. So the environment that built really makes it easy to containerize applications and move them to the cloud. And by doing so, they make it more portable. So you have the ability to go back and forth and get all the advantages that you would have a containerized modern cloud native application. So that resonated with me and I felt I could add some value and obviously he felt I could add some value. And after all, it's not often you get a chance to work with somebody who's named after a car company. So that was a big factor. Is he autonomous? That's what I want to know. I'm working on that. I'm working on that. But it begs the question, obviously a lot of activity around Docker and a lot of buzz around Docker and DockerCon and Kubernetes and this move around containerization. You know, what was the big challenge that still has yet to be solved that you guys are addressing? So from my perspective, the big challenge is how do you deploy these things at scale to existing applications, right? If you're developing a brand new application, right? You can just go hire yourself a software artist to handcraft one of these beautiful works of art that you can then deploy. And if that guy leaves, good luck, right? Which again, it's all over the place, right? If you go look at kind of the graveyard of applications sitting at the enterprises, right? So how do you take that environment and bring it into the modern environment of running it as cloud native? So that was a technological challenge for us. Now when Ben and I were talking while we were at VMware, we were kind of discussing all these issues and what we were seeing with customers and how we could kind of approach and solve it. But then we both ended up leaving and then after I had a chance to go think about it some more, kind of approached Ben and said, listen, if we do it this way, what do you think about that, right? And then you started kind of contributing back, well, have you thought about this? Have you thought about that? So that's how we kind of got into a mindset of, this is a solvable problem, if you just approach it from the correct angle. And I think that's where we kind of came together on that. Right, because it is very different, right? If you're building a native application versus now people want to get some of those cloud native benefits with the legacy stuff either a rewrite or a lift and shift. And so there's, the momentum is there, obviously cloud is a giant, we just got back from Google Next and IBM Interconnect and AWS show. So the clouds there, enterprises are all in, but there is now the legacy stuff where people want to get that stuff into the benefits of cloud architecture. And that's what you're helping them with. That's exactly it. I think there's going to be a mix. There's obviously going to be a long tail of legacy monolithic enterprise applications that are going to continue to run for the next 10, 20, 30 years on prem. And at the same time, a lot of those applications are getting re-architected, broken down into microservices, built cloud native, or sometimes built on prem because of the timeliness and the availability of resources and corporate requirements. But then you also want to be able to go back and forth to get advantages of the cloud and the on-demand nature of cloud that you don't necessarily get when you have an on-prem application. But it's early days. I think you're going to see a lot of improvements, not just in the approach to move the applications to the cloud, but also how to manage it once you get it there, how to do day two operations. And a lot of things that Mazda and the team have built into the App-Lariat environment is the ability to do load balancing, to do failover, just have an inventory of your entire application stack so the IT organization can also have more control over how they're running and how they're behaving after they've been on site. So that's interesting. Of all the times we've been to DockerCon, which are numerous, I've never heard the expression day two operations. There's a lot of talk about spinning stuff up and how quickly you can spin stuff up, but it's that pesky day two. There's still day one. There's still a day one. But then you got to run the stuff, which is not the exciting step necessarily. It's not the sexy part. It's not bringing these things to life. It's the day to day grind, if you will. Yeah, I think, and that's some of what you get from Mazda and myself is our experience and many years of working with enterprise IT environments that do have, it's not sexy, but they do have to worry about security and compliance and archival and management and failover and load balancing. All of those things are key parts of capabilities that we need to automate in the new world of the cloud native applications and that we're not there yet. Right, and especially in the enterprise requirements and compliance and we were at RSA and the security's got to be baked in everywhere and traditionally, not so much anymore, but the traditional cloud entry was test dev, test dev, test dev or cloud native, he didn't have these issues. I mean, even back when we were at VMware, we were seeing small startups come in and we were looking at acquisition or partnership. Frankly, a lot of what we saw were toys. People hadn't quite grasped the whole concept of what it means to create cloud native applications and containerized environments and operating environments and management of those applications. One of the reasons that I liked Mazda's approach was that it's an API driven mentality, which really resonates with the developers and it has a simplified UI on top. A lot of the applications and startups you see out there today are toy UI environments. They work fine for four or five apps, but you really can't use them in an enterprise that has thousands of applications. So having the capabilities of both the API driven model as well as a simple UI that App Lariat or their partners can go in and help an enterprise build those large catalog of applications, I think the two really resonated with me as something that's mature and will help. And was the market clamoring for this? Is it more of a band aid to a fix that people were having an issue with or is it more of a kind of a vaccine that you're kind of looking down the road a little bit further and trying to help them out? Well, so think about it. So containerization is something that everybody's looking into. Every enterprise you go to, there's some kind of a science experiment already going on, right? Because they got to jump on the bandwagon, right? Whether it's happening from the developer side, whether it's happening from the IT side, so they're already there, right? But like Ben said, nobody's really thinking about the day two operations forward, okay? And so how do you, so let's just say you do have your environment containerized and let's just say you're now about to deploy it into production. Great, after the party, what happens, right? You've got to manage this thing on an ongoing basis. How do you roll changes into the code into it, right? How do you scale it up? How do you scale it down? Those things are not trivial, right? So having a policy-based mechanism of managing those environments on a day two forward is really critical to get the whole solution working properly. It's not just about getting your things, getting your applications containerized. It's about ongoing basis. How do you manage and maintain it so that it's still useful to you from an IT perspective, not just from a developer perspective. Right, right. Because that's part of the whole promise of the containers, right? Is that you can bring these apps up and down and scale them big, small, depending on what the requirements are and how they change. But again, day two, I'm going to steal that line, the day two operations, that's a good one. So, so look. I think if I can just add a little bit to that. Absolutely. What Mazda is trying to explain is back with the old traditional enterprise applications, you had a thick operating system, let's say Red Hat or Windows. And on top of that, you had a fairly thick and monolithic stack that may have had a database, let's say an Oracle database. But there were relatively few knobs that an IT organization had to play with. As long as you knew you had Red Hat 7 with these five patches and Oracle or SQL with these 12 patches, you were okay. You could pretty much support any application that had been tested with that. What containers have done and what the cloud native service oriented architecture has done is they've completely blown that up. The OS is now a very thin layer, whether it's core OS or one of these container environments. And then on top of that, each application stack really declares its dependencies on other services and other libraries. What that means to the IT organization, and this is the day two issue that they haven't really grappled with yet, is you can't just have two or three knobs of a Red Hat and Oracle. You now have 27 knobs for each of those applications and services you're trying to run. And it becomes really difficult to keep track of it, both for security purposes. If I've got 10 apps running, each of them could pull in 20 different libraries, each with some security vulnerabilities. How is an IT organization going to deal with that? Those are the types of things that I think this environment really helps in terms of having a single inventory of your software stack and your application stack, so you can quickly go in and say, I know what version is running, I know what the security vulnerabilities are, what updates I need to do, and it brings sanity to that environment. That's the dirty little secret then of the API kind of based application. Because the other thing where I thought you were going is where before it was that infrastructure that determined the apps that you could build, now really it's driven by the application on top and then what microservices and infrastructure does it need. And then of course there's this thing called portability, as you move things from test dev to production, maybe you want to put part of it in a cloud, you don't want it in a cloud, it's on-prem based on data server and the issues and those types of things, so it's a much more complex environment with just like you said, a lot more knobs like that analogy, 27 knobs, just not just three. Every time you introduce a new abstraction layer, somebody gets helped, somebody gets hurt. Right, right. In this case, it's much easier for the developers to be able to develop the unit that they're working on, develop a reusable unit, be able to test it and then move it along, right? When it gets thrown over to the wall on the IT side, all those, every one of those little units becomes an independent variable that has to be managed by somebody on the other side. That's where the complexity is. That's why you can't really do it the way you traditionally did it. You have to do it via policy driven mechanism. Because the developers don't want to deal with it. That's the whole problem too, right? They don't have to deal with the infrastructure so much but like you said, it's the day two, they actually have to operate it. Excellent, so as you look forward, any particular opportunities in terms of verticals or application type that you see as kind of a little hanging fruit where you want to go first or where there's the kind of biggest ROI for your customer? Sure, so I mean, it's interesting that we've seen it across verticals, right? These are custom apps, right? That enterprises use to essentially deliver business, right? And so we haven't really come across one vertical that says, oh, this one is really aching for this problem, right? The problem is, you're not able to compete with your smaller competitors that goes right across verticals, right? So how do you bring these capabilities into your environment? So we really haven't come across, maybe you haven't been out there long enough to find a particular vertical but right now we're seeing it cuts across verticals because it's a business problem that cuts across the verticals. All right, so Ben, I'll give you the last word. Your next advisory meeting, board meeting, whatever you guys get together, coffee, Friday beverage, what's kind of top of mind awareness of really critical things that that player it's got to make sure we're taking care of? I think to harp on the same comment, I'll say day two stuff. And these guys have seen that every time they come to me, they've built a really smooth on ramp for taking the applications to the cloud but I keep harping on this, what about day two? Have you thought about providing a central inventory? Have you thought about easy search if I have a thousand applications that I'm trying to manage? Have you thought about security implications of, what if I can quickly find all the versions of this crypto library 5.6 so I can apply a pass to them? Those types of day two operational IT additions and improvements, especially at scale when you're dealing with lots and lots of applications, I think are what's going to turn this from a toy application to a real enterprise ready application that an enterprise can work with. So that plus working with partners because of the API driven nature of this, I think it's really important to build that team of partners who understand it are experts and can go in and help the enterprises who don't want to do it themselves. It's just like the buzzkill. Everyone's all happy. Like you said, they had the party, they go live, there was drink and champagne and then you're waiting around for the sun to come up the next day. Here we are guys. Here we go back to work, right? All right. Well, that's exciting, right? Big problems are big opportunities and thanks for stopping by and taking a few minutes and best of luck. We look forward to watching the story unfold. All right, Ben and Mazda, I'm Jeff. You're watching theCUBE. Thanks for joining us. We'll be back.