 Live from San Francisco, it's theCUBE, covering DevNet Create 2017, brought to you by Cisco. Hey, welcome back everyone. We are here live in San Francisco for Cisco's inaugural event. First time they're having DevNet Create, an extension of their classic DevNet program. I guess it's not so classic, Peter. It's been only three years. I'm John Furrier with theCUBE and here at my coast, Peter Burr is general manager of Wikibon.com. Our next guest is Jim LaGuardia, who is the founder and CEO of Nermata startup. Welcome to theCUBE. Thank you, John. So thanks for coming on. First question before we get started. What do you guys do? Take a minute to talk about what your company does and why are you here at Cisco DevNet Create. Right. Yes, Nermata is a SaaS for cloud application, delivery and management. So what we do is you can think of us as a logical layer above. You know, it's the big three cloud providers, as well as private clouds. And we provide a common set of application services for developers who are looking at multi-cloud use cases and even edge computing moving forward to provide a common layer. I was just covering SAP Sapphire last week. Again, on multi-cloud, again coming out. Multi-cloud is the hottest trend right now in terms of what people are seeing. And that makes a lot of sense. No one cloud's going to win it all. It's never been a winner-take-all. Jerry Chenick, Ray Locke said that many years ago. Turns out he's right. However, you get the big cloud guys lining up. The question is, multi-cloud, is it reality yet or is just hybrid IT, hybrid cloud, just the stepping stone to potentially a multi-cloud world? Your thoughts? Yeah, good point. And hybrid is certainly the stepping stone, but what we're seeing more and more is the application sort of being chosen to go on one cloud and on another. So it's not at a point where we're seeing the same application span multiple clouds, but based on the workload, based on the application type, enterprise is deciding whether to put them in private cloud, public cloud, or a choice of public clouds. So define multi-cloud real quick. Take a minute, so I'd like to get your definition of what is multi-cloud? Right, so to me it's a combination of being able to choose your infrastructure services primarily and being able to have a portable set of application components and constructs which can span either these public or private cloud deployments. And today, of course, there's a lot of momentum towards public cloud, but private cloud is also going to continue to grow and will continue to grow for various reasons. So having that choice of deployment is really what we're seeing as multi-cloud today. And of course, to put a plug in for Wikibon and Peter's research, they just put out a new true private cloud report and had it pegged at a market of what, 260 billion? For true private cloud, yeah, yeah. So you're right. It's going to be big. It's going to be big. And John, just another plug. We're actually doing a multi-cloud crowd chat tomorrow at 9 a.m. Pacific. So anybody that wants to participate in a crowd chat about multi-cloud, 9 a.m. Pacific tomorrow. Good plug. Check it out, Pacific, crowdchat.net, check it out to be right there on the front page. You should get on that, Jim. But I want to ask you to go to the next level. Multi-cloud, let's peel the onion a bit a little bit. Does that mean I can run workloads on any cloud or do I put a workload on one cloud and I put another workload on another cloud? Or can this workload, if capacity's bad, move over to another cloud? This just smells like a latency problem to me. This seems like un-gettable at this point. What's your definition of that multi-cloud? What is multi-cloud? Yeah, so what's happening in the developer space, of course, with the big adoption of containers and the push towards containerizing applications, now we have that ability to rapidly spin up services as needed on different cloud platforms. And really, a cloud becomes a place where you can have a container host and an endpoint for deployment. So you combine that with management services, application management services like Nirmata, and now you do have that choice of being able to set policies either based on demand and scale or usage, or based on recovery from faults in the infrastructure to span different clouds for the same workload. Okay, next question for you. It's great to have you on. Great subject matter expert here. Thanks for answering the questions, but this one is a little bit different. If I want to secure a cloud with the Amazon, I put my stuff there, you seem mostly test-deaf, and then Oracle CEO talks with us all the time. It's pretty much all test-deaf. Okay, that ship has sailed, pretty much no brainer. What percentage of the workloads now or what workloads specifically are going beyond test-and-deaf that you're seeing that are going into production? Because now with hybrid, it opens up more range of apps beyond test-and-deaf. So certainly test-deaf is happening. We get that, low hanging fruit. What's the next level? Yeah, so I think the one way to categorize it is systems of engagement and systems of record, of course. So we're seeing anything public facing, whether it's mobile, web app properties, web applications, more and more microservices-style SOA applications, those are the next wave that's going to cloud. Data residents tend to stay with private cloud for a longer term, but even that, over time, we're seeing with VPCs, with the right security constructs, being a viable, public cloud being a viable option there. One of the top questions we have in our CrowdChat community that comes up all the time around, Dev Ops, I'm going to give you thoughts on this. What advice would you give to operations practitioners who are afraid Dev Ops is going to automate away their jobs? Yeah, so yeah, great, great question. And that's very far away from the reality, right? So there's, what's happening with Dev Ops is now we're getting to a better definition of what Devs need to be concerned about and what Ops needs to be concerned about, right? And again, pointing to containers as one of the enablers, microservices as another, we're seeing where application developers want to operate their own applications. They want control of their destiny, but the furthest thing from their minds is to worry about IP addressing and security concerns and things like that. So there is, you know, and it's interesting because enterprise Dev Ops is very different than what you would find in a startup or in a cloud or internet giant, right? And there is no mythical enterprise developer who can do all of this themselves. You need a Dev and you need an Ops, and that's not changing. The mythical man month kind of goes out the window. We had CMO, EVP, earlier on. We had, it was, you know, Matt Howard, and he's an experienced guy, but he was saying, you know, 100 developers have 10 IT supports and one security person, he sees that completely flipping around. So if you take this whole notion of the jobs are going to go away, which I think is BS, certainly things have to be automated. Speaking learning is great for that, but you can see the shift happening certainly in more security guys, more operational IT guys not doing escalation, doing actual real IT. So I think there's going to be a shift of jobs. So you might be displaced functionally. You're a plumber, now you're a machinist. I get that. We're the hot jobs. So if that's the case, if you believe which I think you do, where are they going to shift to? What's the job profile look like? Yes, much like we're seeing even in software development itself, right? The level of abstraction and the amount of knowledge that has to be absorbed keeps increasing. So it's more similarly in operations, what we're seeing, like you mentioned, rather than doing something at a low level, now it's understanding what are the best policies for let's take security as an example in AWS, in Azure, in private cloud. How do you now make sure you have the right visibility and governance with things like containers, microservices where the applications are so dynamic across various environments? So it is a transformation in the type of role and skill set, and I think it's for the better because now you really have time to step back and look at this holistically and contribute back to the business. Here's a philosophical question for you. And maybe Peter, you could weigh into, what single misperception about DevOps would you like to see change out there? As people try to grok DevOps, we hear it's a movement, we hear it's a playbook it's this, it's an agile manifesto, grow organically, conways law, and all kinds of stuff we've been talking about. So bottom line, what is the most misunderstood or misperceived issue about DevOps? Now you'd like to see changed. Yeah, so to us, the one issue that we always emphasize is there will be a dev and there will be an ops, and any product that tries to minimize one role or another is not a good fit for enterprises. So what's needed is a transformation of that ops role to the role from just being the direct service provider the hands-on ops person to more of a governance curation in some ways an architect type of role, right? And that's what we're seeing is that ops role is not diminished, it's actually heightened and highlighted. Great point. I would, we've already talked about it in many respects the idea that we're going to go from application development to pushing a button and having it, having the business suddenly run differently is just silly. At the end of the day. You think people think that that's what DevOps is? There's a magical, rub the bottom, the genie pops out. I think there's a lot of people that think that DevOps is a step on the path to know ops, to having no people involved in operations at all. And that's just not going to happen. Do you believe that ops is still going to be relevant? I think ops is always going to be relevant. I think that dev is going to evolve to better understand and have greater data invisibility in what's going on in ops. Ops is going to have greater predictability in what's going to happen from a development standpoint. So I think we will see a combination of roles. We'll see the productivity of ops continue to grow. But the idea that this is going to be, there's magic in here and Gandalf is going to wave his dev ops. What would Trump say about dev ops? I don't work great at it. Trump would say, Trump would say, I've never been to Mexico. I'm going to make it amazing. We're going to build a wall on IT. I had to bring that in. Sorry, I know you were laughing about Trump earlier with the whole thing going on. Okay, good point. Some are saying in the community, not no ops, but new ops. It's a new kind of ops. Yeah, the way we see it is what we think of as dev ops is splitting more into functions like application operations, security operations and infrastructure. So really all three need to be accommodated and they need to work together. And that's sort of how we have built up Nirmata as our product. There's ops for all three of them. In fact, the last conversation we had, John, was going to test you on this, is that the inherent quality, the inherent distributed quality of a lot of the new applications that we're building absolutely dictates that we start to parse ops up differently. That it's no longer running it on a single machine or on a single database with a network out in a client server domain. It is just inherently distributed and therefore the tasks and the responsibilities and roles associated with the operations side of that are themselves going to be inherently distributed, which requires new ways of thinking, new conventions and new tools. Jay, I want to give you a final word. Give a plug about your company. Thanks for sharing your insight. By the way, I appreciate you answering the questions. What are you guys doing? What's up with the company? Talk about the status, the employees, how much funding you have, how much revenue you have, what's your goals, go lay it all out. Yeah, so myself, my other co-founders, our background is enterprise software and we come from a network management background where we build centralized management systems for complex networks, distributed devices, et cetera. What we saw happening is with cloud applications are starting to mimic that complexity, right? And as applications move from back office productivity functions to these hyper distributed mission critical real life functions that we use day to day, there's a need for this enterprise grade management. So that's the type of centralized management we're delivering as a service to our customers. You have to become network-ified so you have to have app management. I mean, that's pretty much what you're doing. You're bringing network management paradigm to apps. Exactly. Versus a monolithic app and some dashboard. Now it's all over the place. Multiple form factors, access methods. Right. It's a network in the app. It is. Yeah, and today the customers are left to cobble together about 12 to 14 different tools, correlate data across tools. And what we need to do is move beyond systems with just observe and report to being able to observe, react and learn and do things in real time. Exactly. So you guys are simplifying that process. Absolutely. And is it a single pane of glass? Is it a service? Is it a software product? It is a cloud service. So you can think of us as an overlay across any public or private cloud. And early on we kind of decided the best way to deliver infrastructure is as a service and we've learned that in real life. People who are doing that are winning. Absolutely. So Trump would say, winning. Yeah, yeah, yeah. He would say, I'm going to drain the data lake swamp. Who knows what he'd say. Well, I could say I couldn't get that in there. So we're built. Drain the swamp. You didn't get the data lake swamp. No, I got it. I didn't. You just didn't laugh. Okay, go ahead. So we're going to monitor completely as a cloud service because of that philosophy that we started with. And we want to give developers and DevOps teams the choice of any platform, right? And today it's all about cloud. I mean, the edge is also very real. We have industrial IoT customers. We're looking at containers. Yeah, so your world's getting, your TAM, your total, just what market's getting bigger and bigger. Absolutely. Every IoT device has data on it. Because data is an asset. It's part of the app. I mean. I want to bring that up. Just if we have just a second. Yeah, go ahead. Yeah. I'm curious, because one of the things that we believe is that increasingly the whole concept of digital business is how will data feature as an asset in your business? Especially for creating sustaining customers. Totally by end of the idea of the external view versus the internal view. For customers versus for employees. That for customer side, the engagement side is really driving a lot of this. But at some point in time, it makes me wonder if we're going to move from a DevOps orientation to a data ops orientation. Where at the end of the day, the physics of how things run is where is the data? What's the latency to get at it? How do you handle the state of it, et cetera? Do you foresee at least or an extension of the DevOps concept? So the data as an object is something that we act upon and we understand what role it plays in this whole bringing together a lot of piece parts to create distributed digital systems? I think so and the starting point of that that we're seeing is the split between data services and behavioral services. Look, I mean, in any form of programming, it's all about packaging behaviors and data, right? So whether it's in a programming language and with object oriented, it was about putting things together in object. Now with service oriented and microservices, it's these service boundaries. So having the data services and then having the behavioral services separated gives a lot of flexibility. And then being able to move the compute to the data versus the other way around, that is also very interesting. So we're working with some partners where we're looking at cross-cloud data. Can we, as even services and containers are spun up on one cloud, can we clone an entire environment into another cloud? Can we migrate some of the data efficiently? Challenges like that. Well, Jim, we're going to recruit you. I just made a note to ping you for tomorrow's crowd chat to see if you could make it, or one of your co-founders. I'd love to get your input at the community as part of sharing insight into this really fast growing, changing world of management with all this complexity. I mean, there's more tools out there than ever before. They're all different types, a lot of complexity. So we'll hope to bring you back in the studio or have you come in and be a Skype or a crowd chat. Okay, this is theCUBE's exclusive coverage. Cisco's inaugural event, DevNet Create. I'm John Furrier. Peter Burris, stay with us for more coverage after this short break. Hi, I'm April Mitchell, and I'm the Senior Director of Strategy and Planning for Cisco.