 I'm at EMC World, this is theCUBE, our flagship program. We go out to the events, extract the signal from the noise. I'm John Furrier, the founder of SiliconANGLE, and we're here at EMC World in Las Vegas. Our fourth year, consecutive with theCUBE, theCUBE was born in 2010 here at EMC World. And one of the great things about theCUBE is we get to go to the events, extract the signal from the noise, and talk to the executives of the companies at the event here, it's EMC, so we talk to all the head haunches. But more importantly, we love to go out to the crowd and find out the emerging players, the ones that aren't the big whales, the ones that are gaining ground. And we've got a great guest here, the chairman and CEO of ServiceMesh, Eric Fleer. Thanks for coming on theCUBE. It's great to be here, thanks. So, ServiceMesh is in the hybrid cloud management for applications, which is the cutting edge area right now because cloud mobile and social has become the buzzword, the pillars that everyone's building and transforming their infrastructure to literally go to, as what I call, modern infrastructure. And it's a huge inflection point. I once called it the combination of client server computing and the PC revolution boiled together, and that's kind of the action we're seeing in terms of dynamics and impact. And it's happening very, very fast. So, I want to ask you, you guys have an uncontested lead in this area, but you're not a huge company. I mean, you've got all these big companies could be doing it. Why are you guys gaining so much traction? Yes, well, we're not a huge company compared to the ones that I'm sure you're thinking of. IBM's the MC's of the world, but we are growing. And I think we're growing primarily because that we've taken an approach that's very customer centric. So in other words, when this IT transformation, as you put it, started to become ubiquitous, the CIOs of the world started to look for a problem, to solve a problem that was really focused on applications, getting them to market faster, and having a lower cost base to run them. So there was a lot of pressure on them that really had a lot to do with business metrics. And yet the field that had really come up through virtualization and infrastructure automation had a natural affinity to approach the same set of problems from their place of knowledge, which is infrastructure, right? So you start with infrastructure optimization, and you start to move up the stack. The problem is that this is really not, ultimately, an infrastructure problem. It's an infrastructure in search of an application. In other words, what good is infrastructure if it's not helping applications? So we really approached it from the app down, and I think that's changed our model. Yeah, Sean Douglas, a recent hire of yours, actually from EMC, I've gotten to know Sean, he was at EMC Ventures. You guys really had a stellar hire with Sean. He's now CTO of your company, and you're co-founder of Frank is now in charge of strategy, kind of looking at the next mountain climb for service mesh, which the market's growing like crazy. I got to ask you, I mean, we saw DevOps really change the game around web scale, companies saying, hey, I want to be more hyper-scale. So DevOps became a very popular, kind of pre-cloud developer community. That has now blown up huge into cloud operations, right? So cloud operations is the public cloud. We know Amazon, OpenStack, we had you on OpenStack, but more importantly, the enterprises aren't necessarily that eager to go to the public cloud for a lot of variety of reasons, compliance, security, data, et cetera, et cetera. They're going there. It's a destination for them. So the hybrid cloud is a nice fit for you guys. So I got to ask you about the competitive landscape. What do you see out there for companies that are taking the best of DevOps, what was proven as, hey, I can develop on cloud resources, launch new apps, and actually make this stuff run fast and deliver business value. So that's a mainstream requirement right now. Hard to get to. It is. So what is the competitive landscape? Okay, well, I think you've hit it on the head in terms of what the requirement is. How do you actually create an automated DevOps tool chain? Which is, how do you get a concept to production faster? And how do you do that with some of the existing tool sets that you have, because in an enterprise they're not going to throw it out everything and just buy your new fangled toy, right? So what we see in the marketplace are people attacking this from different angles. You have the guys, as we discussed just earlier, that are looking at it from infrastructure optimization or automation. So they take these kind of old fashioned run book automation tools and they try to apply it up the stack. Then you've got guys who are really focused on DevOps who have an ability to do what they might call continuous delivery or some kind of automated way of bringing you along a release management chain, but they're not well connected to the infrastructure. And at the core of all this is really a policy problem, right? From an enterprise point of view, you want your constituents, your business units, to go fast, but they have to stay safe. If you turn over the reins for IT as a service to the business and you don't have an effective policy or governance model, then everything falls apart. So what we've done at Service Mesh is we've really started there. What if we could build an extensible, incredibly powerful policy core and apply that up to the DevOps tool chain and down into the infrastructure so that you can essentially provision real time applications or application platforms and then treat code as a first class citizen and move it along a software delivery lifecycle? So this entire landscape of problems that one has to solve, if you don't have a policy core that has some fidelity across these different problems, then essentially you end up with point solutions and they become more trouble than there were. Yeah, I mean, I love the infrastructure as code concept because that's about making things programmable. And the buzz here is software-defined storage. And obviously software-defined networking kicked off the software-defined everything or as we say software-led infrastructure. But let's take a step back. I want to ask you one question around the operating system of the data center and how the data center is changing from a more like API model, like a lot of Amazon. But how did you guys get started? Talk about Service Mesh because you guys aren't a big well-known company and the vertical people know you're doing extremely well. How did you guys get here? Tell us, how did you guys start the company? You have a co-founder, met Frank at OpusTrack. You guys have teamed up on other deals to tell us the story. Well, Service Mesh is a company that's a long time in coming. So it seems like maybe we've done about half a decade of work and done a lot in half a decade. Five years of hard work building an application that we think could be fundamentally important to this space. But the truth is we spent a lot of time prior to that, which is really the story of Service Mesh. So the folks that have built Service Mesh have decades of experience in the enterprise. So we didn't approach this from a consumer perspective or even an infrastructure perspective. We approached it from an application-centric mentality from other ventures that we had done in the enterprise. So for instance, a previous venture that Frank and I had worked on together was called SOA Software. We spent quite a bit of time really focused on what does it take to have- As in service-oriented architecture. Service-oriented architecture. What does it take to scale service-oriented architecture from a governance and policy point of view? Well, that's a stepping stone to the same type of problems that we needed to solve here. So the way that this came about is that we've built, I think, a certain amount of credibility and relationships in the enterprise space, and we were able to call from those relationships the set of requirements that were converging on a similarity. The bet was that by the time we were able to build something that was scalable and referenceable that the market would arrive. So rather than take a normal approach, which would be early VC money, a lot of PR, a lot of marketing, everyone knows about you, and then you do a lot of pilots, but the market's not there. We went the opposite approach. We went really deep with select, highly complex, regulated, multinational organizations to really prove out that we had something. And you self-funded. We self-funded. And when we got to the point where the market started to arrive and we had referenceability where companies could stand up and say, we've saved $150 million. We've fundamentally transformed our productivity metrics and we were able to point to those companies and they would stand up proudly and speak to the same. Then we felt we were ready to scale. And so that brought us maybe to about 18 months ago where we took on our first capital from Ignition, which had been fantastic for us. Frank Artali. Frank Artali and John Connors and the rest. Cube alumni. Yeah, amazing. Praise the Cube alumni. Is that right? Yeah, he's been on multiple times. We love Frank. He's a friend of ours. He's great. And so we couldn't be happier with our partnership there and they've really supported this company, backed it, and made it what it is today in the sense that we took that backing and then built out the structures for scale and that's where we are now. So basically, and if I can summarize, you made a big bet. You made a big bet that web services and services in the arts, aka now APIs would be big. Yes. Now, if you look back a decade ago and even more than a decade ago, that was web services and that's now hitting. Yes. I mean, so that kind of went through this like, well, yeah, it wasn't really ready for prime time. The web services, but it's changed too with cloud. So you're in a good spot. So you essentially skated to where the puck, you went where the puck was coming. You were already there waiting for it. Oh, that's right, yeah. Okay, so take us now today. So you made a good bet. Both of the chops to do that. You got the multi now clients, now you're growing, you got the funding. How does this world that you're living now change? Because we had the CIO of EMC on earlier as consistent with other thought leaders, validated again that the operating system model of the data center really is what people should be thinking about. But the data center is now an operating system and operating systems have subsystems, layers of software. So things like configuration, ability, management, automation all become part of that fabric. So you agree with that? Absolutely. Okay, so that's cool. So we don't need to spend that. So with that, what is the mandate? I'm a CIO. Yeah. Okay, I say, okay, the future is obvious now. The fog has lifted. I got to get to a cloud like environment, have utilities and APIs, but still keep my on premise. What is the mandate for the CIO to move there? And what do they do? And how do you guys help them? It's a very good question because that actually is the question that we most often have to ask ourselves. Because ultimately you want CIO buy in. This is a really a moment in time that is, I don't think that we've seen since the late 90s. It's truly IT transformation. Nobody's business as usual. So the CIO has a set of common concerns right now of an enterprise. One is that the CFO has come to them and said you have to cut out anywhere from 150 million to a billion dollars of your annual spend. So you have to cut the run rate down to the extent that you can't just shave around the edges. There's an operating model ship that's necessary. The other thing is that the business now needs to compete with a different set of constituents. The business is no longer satisfied with the monopoly internally to supply them with what's necessary in order to do their job. So what happens is when the business gets a mandate from the CEO and the chief marketing officer, et cetera, to get into new markets, new emerging markets with new products faster, to compete with upstarts who don't have the same difficult quagmire of regulatory issues, et cetera, then what happens is if the IT organization can't respond fast enough and can't be a better partner, they go around them. And this phenomenon is known as shadow IT. Because the business units say, okay, in the past we would just be sluggish along with you. But now they're saying, we'll go to Amazon. We'll go to Salesforce. We'll go to Rackspace. We'll go around you and put in our credit card and we'll start doing our work because we have to get something done. But they're cobbling together, right? So what happens is that goes outside the operating system model. They're going, the shadow IT is, as we said, it's innovation opportunity. Yes. I mean, so yeah, it's rogue, but rogue sometimes is R&D and innovation. You justify it because it's more convenient and you have to get your job done. However, that brings us to where we are today. What the CIO wants is not that. CIO doesn't encourage shadow ITs. It's a regulated environment. Don't ask, don't tell policy. No, it has to be managed. It's governed. You have auditors, right? So what the CIO wants is to be able to offer the same level of convenience and agility that they can get externally, internally. Essentially, it's this. The CIO wants to transform the IT organization to become a cloud broker that can deliver IT as a service directly to the business. That's what every CIO is looking to do. And the way that you do that. And you've got to cut the middleman out. Well, you become the middleman, right? And as the middleman, you now have an ecosystem of service providers. And this is very interesting because now you say, okay, well, what does the CIO think about? Always thinks about pleasing the CEO and the COO and the CFO. Do they care about software-defined networking, software-defined storage? Only in so far as it supports applications, right? Because applications are where you make your money and it's where you spend your money. So what we like to show at ServiceMesh is that we can be that connective tissue, that glue. We are behind that software-defined data center vision and the practical application of that to what the enterprise actually needs to fulfill their IT transformation goals. So let's talk about the large companies out there because you mentioned them earlier that people might build a point solution and then it kind of breaks. And you also mentioned a comment that early on that you saved the company $150 million. Give an example of a use case. You don't have to name names, but where you've walked in and literally saved a company over $100 million. Okay, so I'll give two if I might. Is that accurate? You've done that? Yes, not me personally, but our team. Yeah, ServiceMesh. Yeah, you was the name. I wish I could do half of the things that our company heroes. Yeah, they don't let me touch anything these days. But what we've done at ServiceMesh is actually just that. We've been able to come into organizations that are looking to affect that model, to become an IT broker and give them the tools where they can do that, but still have the same manageability and auditability and security that they have now. So they can feel that comfort to shift to this model. So what we've done at one particular bank is we put in that model and we did a few things. The first was to look at existing applications and understand how you can migrate them onto more standardized platforms, right? Because a lot of the cost structures in the enterprise today come from all these insane Rube Goldberg machines that these applications live on top of. As you shift to a cloud model, you're also shifting to a model of more standardization, less variability. So you can have less people to run them, right? So that's where a lot of the cost comes is you can lower your FTE count around that particular function. So that's one thing. The next thing is how do you actually create those standardized platforms so that they can be automated from beginning to end? And what we mean by beginning to end is really no human intervention. The only place you want human intervention is where you want to go back later and maybe insert some for some reason that you say, okay, we want Joe to approve this or Sally to approve that. You don't want to be forced to any human intervention which means you have to replace every sticky point where there's a human today with a policy. So the central policy engine from service mesh allows you to, in essence, create those policies but manage them centrally, change them centrally, version them centrally, in essence scale. So you end up with this capability of having an end-to-end provisioning, okay? Once you do that, you now can provision complex topologies which are known as application platforms and put them in an app store. And that's an example of something that we've brought forward, that by reducing the manual steps and the number of people necessary and increasing the standardization, we've been able to save company significant capital in the range of what you've described. And economics is a great proof point too in terms of valuations. Especially if they're willing to stand up and verify those economics. Not a lot of people can say, have people stand up and saying, we've saved over a hundred million dollars. Well, let me ask you about speaking of large complex deployments. I mean, let's talk about VCE. We're going to have them on the cube here. So VCE is a flagship offering for around EMC, of EMWare, everyone knows this was involved. Big problem, a lot of reference ability, same kind of challenges. How do you guys play with VCE? And how do you compliment, say, VC? Because that's an environment that is not easy to turn key. So, or is it? We think it is easy to turn key. If just because it hasn't necessarily been a turn key to the extent that it will soon, it doesn't mean that it's not getting there. In fact, work with VCE at all? Yes, and we think it's a fantastic product. The VBlock is, it's on fire. I mean, it's a staple of the enterprise, or it's becoming that way, right? So we actually believe very strongly in that vision of converged infrastructure and the units of provision being a unit that you don't have to assemble various pieces and parts, right? So from the end user experience, what you really want is the ability to put a cloud in the box in the data center and instantly begin getting business value. And what you're describing is that there's still some friction between that arrival of a VBlock on site to when somebody can get business value. We think that we can add a lot of value to that. The automation side, management in general? Two things, one is obviously on the automation side, which is core to what we do, but even more importantly, to bridge that gap from a policy and governance directly into the software delivery lifecycle. Because why does someone want a VBlock? One is complexity reduction, right? You could assemble it yourself otherwise. And the other is business value. So the complexity reduction piece, we can add because you can install service mesh in a matter of hours as opposed to weeks that a lot of competing capabilities cause. The next thing is, well, now that you've got it in an hour, is what are you going to do with it? The idea is that you want to deliver IT as a service, which is the ability to spin up application platforms, not VMs, but true platforms that you can start writing code on and deliver that directly to the end user. Now you can't do that without the right policies. You can't do that without a notion of application lifecycle, because what are they doing? They're building applications in a lifecycle. We've spent five years doing that, and I think that's our value that we can bring to VCE. We're getting the hook here, so I'll give you the final word, talk to the camera and tell them what's next with service mesh, what's your year looked like, what's your goals? So our goal now is, it's a very exciting year for us. So our goal really is scale. We put the structures in for scale, we're beefing up on the support side. Our channel partnerships are becoming increasingly important. I think what you'll hear from us is a number of SIs and service providers, MSPs, and even ISVs that will be OAMing and reselling our product. So we're excited, we think this is the year where we'll become better known. Service mesh, up and coming, hot startup, scaling, and not startup anymore, they're growing like crazy, self-funded, great entrepreneurial story, congratulations Eric for the year. Chairman and co-founder and CEO of Service Mesh, company to watch here at EMC World. We'll be right back with our next guest after this short break.