 Hi, I'm Peter Burris and welcome to another CUBE Conversation. I'm here with Tom Joyce, CEO of Pensa, from our beautiful Palo Alto theCUBE studios and we're talking a bit about some of the trends and most importantly, some of the real business value reasons behind some of the new network virtualization technologies. But before we get there, Tom, tell us a little bit about yourself. How did you get here? Okay, thank you, Peter. Thanks for having me in today. I am CEO of Pensa. I've been there for about six months. Company's about three years old, so I joined them when a lot of the engineering work had already been done. And I've been around the tech industry mostly on the enterprise side for a long time. I worked with Hewlett Packard in a number of different roles. I worked at Dell, I worked at EMC and a number of startups. So I've been through a lot of different transitions in tech as you have over the years and I've got excited about this because I think we're on the cusp of a number of big transitions with some of the things that are coming down the road that make a company like Pensa really interesting and have a lot of potential. So it's been a tremendous amount of fun working in startup land again. So what does Pensa do? So Pensa is a software company and again we're based here in Mountain View. Most of our operations are here. We also have an engineering team over in India and they're all people that are focused on networking technology. They have a long history there and what we do is help primarily service providers. You can think of the classic telecommunications industry but also other modern service providers build modern networks. We're very focused on network functions virtualization technology or NFV which is about building network services that are highly flexible in software that you can deploy on industry standard server technologies, kind of cloud native network service development as opposed to what many folks have done with hardware based and siloed networking technologies in that industry for a really long time. So what we help them do is use intelligent automation to make it easy to build those things in incredible combinations with a lot of complexity but do it fast, do it correctly every time and deliver those network services in a way that they can actually transform their businesses and develop new apps a lot faster than they could do otherwise. So Tom, I got to tell you I'm an analyst. I've been around for a long time and every so often someone comes along and say, yeah, the telcos are finally going to break out of their malaise and do something different yet they always don't quite get there. What is it about this transition that makes it more likely that they succeed at becoming more than just a hauler of data to actual digital services providers? Yeah, I mean it's an excellent question. Frankly, it's one that I face all the time as you traffic around Silicon Valley, people are focused on certain hot topics and getting folks to understand that we are at a cusp point where this industry is going to fundamentally change. There's a huge amount of money that's actually being spent and a lot more coming. A lot of folks don't necessarily, we don't spend their time there every day realize what's happening in these communication service providers, which we used to call telcos because what's happened, and I think you're interested in your perspective on this, over time you see long periods in that industry of things don't change and then everything changes at once. We've seen that many, many times and the disruptions in that industry which were very public 15 years ago and then another 10 years before that, those were trigger points when the industry had to change and we strongly believe that we're at that point right now where if you look at the rest of like enterprise IT where I've spent most of my career we've gone through 15 years of going from hardware based proprietary siloed to software based industry standard servers, cloud and cloud native. And service based. And service based, right? And the formerly known as telco business is late to the party, you know? And so it's almost like that industry is the last domino to fall in this transition to new technology and right now they're under enormous pressure. They have been for a while. I mean, I think if you look at the industry it's a trillion dollar plus business that touches basically every business and every person in the world and every business and every person has gone to wireless and data from wire line and the old way of doing things. And these service providers have pretty much squeezed as much as they can possibly get out of the old technology model and doing a great job of adapting to wireless and delivering new services. But now there's a whole new wave of growth coming and there's new technologies coming that the old model won't adapt to. And so frankly the industry has been trying to figure this out for about five years through standards and cooperation and investment and open source stuff. And it's kind of only at the point now where a lot of these technologies work. But our job is to come in and figure out how do we make them work in production? How do we make it scalable? And so that's why we're focused there is because there's an enormous amount of money to get spent here. There are real problems. It's not crowded with startups. We have kind of a free shot on goal to actually do something big. And that's why I'm excited about being part of this company. Well, the network industry is always unlike the server and storage industry has always been a series of step functions. And it's largely because of exactly what you said that the telcos, which I'll still call them telcos but those networks of service providers historically have tied their services and their rates back to capital investments. And so they would have to wait and they'd wait and they'd wait before they pulled the trigger in that capital investment because there was no smooth way of doing it. And so as a consequence, you've got these horrible step functions and customers, enterprises, like a more smooth set of transitions. And so it's not surprisingly that more of the money's been going to the server and the storage guys and the traditional networking types of technologies. But this raises an interesting question. Does some of the technology that you're providing make it possible for the telco or the network service provider to say, you know what, I can use NFV as a way of smoothing out my investments and enter into markets faster with a little bit more agility so that I can make my customers happier by showing a smoother program forward, make my rates, you know, adjust my rates accordingly, but ultimately be more likely to be successful because I don't have to put two or three or $10 billion behind a new service. I can put just what's needed and use NFV to help me scale that. That's exactly right. I mean, we're really bringing software programmability and DevOps kinds of capabilities to this industry. Us and other folks that are involved in this, you know, this transition, which we think is enormous. I mean, it's probably one of the biggest transitions that's left to happen in tech. And the old model of set it and forget it, I put in my hardware based router, my switch, build out my, make a big investment, that step function you talked about and depreciated over a long period of time, doesn't work anymore because during that long period of time, new opportunities emerge. And these communication service providers haven't gotten all the growth because other people have jumped into those opportunities, the over the top people, the Netflix's probably increasingly cloud players and saying, we're going to take that growth. And so if you're one of these, you know, there's a few hundred large communication service providers throughout the world, this is an existential problem for them. They have to figure out how to adapt. So when the next thing comes along, they can reprogram that network. You know, if there's an opportunity to drop a server in a remote branch and offer a whole range of services on it, they want to be able to continually reprogram that, update those. And you know, we've seen the first signs of that with software to find advanced, right, as an example of that. But not just take a hardware approach to adding new services and improving the quality of the experience that their customers have. That's exactly right. They want to have software programmability, they want to behave like everybody else in the world now and take advantage, frankly, of a lot of things that have been proven to work in other spheres. So the fundamental value proposition that you guys are providing to them is, bring some of these new software disciplines to your traditional way of building out your infrastructure so that you can add more, add new services more smoothly. Grow them in a way that's natural and organic. Establish rates that don't require a 30 year visibility in what your capital expenses are. That's right. So one of our, you know, our flagship customers is Nokia. Nokia you can think about as kind of a a classic network equipment supplier to many of those service providers, but they also provide software based services through things like Nuage that they own and some things they got from Alcatel Lucent and they do system integration. And they've been kind of on the leading edge and using our technology to help with that of saying, look, let's deliver you industry standard, you know, Intel based servers running network functions and software. And what we help them do is actually design, validate, build those capabilities that they ship to their customers. And you know, without something like Pensa, somebody has to go in and code it up. Somebody has to really understand how to make these different parts work together. I've got a router from one place. I've got a virtual network function from someplace else. Interoperability is a challenge. We automate all of that. And we're using intelligence to do it. So you can kind of go much faster than you otherwise could. Which means that you're bringing value to them and at the same time essentially fitting their operating model of how they operate. So you're not forcing dramatic change in how they think about their assets. But there are some real serious changes on the horizon. 5G, net neutrality and what that means and whether or not these service providers are going to be able to enter into new markets. So it does seem like there's a triple witching hour here of the need for new capital investment because there's new services are going to have to be required and there's new competitors that are coming after them. We like to think that, or we think in many respects the companies that are really in AWS's crosshairs are the telcos. And you guys are trying to give them approach so that they can introduce new agility or be more agile, introduce some services and break that bond of rate based capital investment based innovation. Yeah, exactly right. And also frankly break the bond of having to buy everything from the same telco equipment provider they've done for the last 20 years in extraordinary margins. People want to have flexibility to combine things in different combinations as these changes hit. 5G you mentioned is probably the biggest one. You know and I'd say even a year ago it was clearly on the horizon but way out in the distance. And now almost every day you're seeing production deployments in certain areas. And it is going to fundamentally change how the relationship works between businesses and consumers and the service providers and the cloud people. All of a sudden you have the ability to slice up a network. You have the ability to program it remotely. You have the ability to deliver all kinds of new video based apps. There's a whole bunch of stuff we can't even conceive of. The key thing is you need to be able to program it in software and change it when change is required. And they don't have that. Without technology like this. That's right and 5G provides that density of services that can actually truly be provided in a wireless way. All right so but this raises an issue. Look we're talking about big problems here. These are big, big, big art problems. And no company let alone pens is unlimited resources. So where are you driving your engineers and your team to place their design and engineering bets? Yeah I mean look there's clearly a set of problems that need to be solved. And then there's some things that we do particularly well. We have some technology we think is actually unique in a couple of areas. Probably the heart of it is intelligently validating that the network you designed works. So let's say you're a person in a service provider or you're an SI providing a solution to a service provider. You make choices based on the requirements because you're a network engineer that I'm going to use this router I'm going to use a Palo Alto networks firewall I'm going to use Nginix and use Nuage. Whatever that combination is. I've got my network service. Very often they don't have a way to figure out that it's going to work when they deploy it. And we build effectively models for every single element and understand the relationships of how they work together. So we can pretty much on the fly validate that a new network service is going to work. The next thing we do is go match that to the hardware that's required. I mean servers they're not all the same and configurations matter. I mean we know that obviously from the enterprise space and we can make sure that what you're actually intending to deploy you have a server configuration or underlying network infrastructure that can support it. So our goal is to say we do everything frankly from import network services or onboard them from different vendors and test them from an interoperability standpoint to help you do the design. The real heart of what we do is in that validation area. I think the key design choice that we are making and frankly have had to make is to be integrable and interoperable. And what that means is these service providers are working with multiple different other vendors. They might have two different orchestration software platforms. They might have some old stuff they want to work with. What we're going to do is kind of be integrable with all of the major players out there. We're not going to come in and force our orchestrator down your throat. We're going to work with all of the major open source ones that are there and be integrable with them. We believe strongly in kind of an API economy where we've got to make our APIs available and be integrable because as you said it's a big problem, we're not going to solve it all ourselves. We got to work with the other choices that one of these customers makes. So we like to say at Wikibon that in many respects the goal of some of these technologies, the NFV software defined networking technologies needs to be to move away from the device being the primary citizen to truly the API being a primary citizen. People talk about the network economy without actually explaining what it means. Well in many respects what it really means is networks of APIs. Is that kind of the direction that you see your product going and how are you going to rely on the open source community or not to get there? Because there's a lot of ancillary activity going on in creating new inventive and innovative capabilities. Yeah, I think that's a really big question and to kind of tackle the key parts of it in my mind. Open source is extremely valuable and if you're a communication service provider you may want to use open source because it gives you the ability to innovate. You can have your programmers going and make changes and do something other folks might not do. But the other side of the coin for these service providers is they need it to be bulletproof. They can't have networks that go down and that's the value of validation and proving that it works. But they also need commercial software companies to be able to work with the major open source components and bring them together in a way that when they deploy it they know it's going to work. And so we've joined the Linux Foundation and we're one of the founding members of Linux Foundation Networking which now has open NFV and has ONAP and a number of other critical programs and we're working with them. We've also joined OSM which is part of the European Telecommunication Standards Institute which is the other big standards organization. I'm not aware of another company in our space or related to NFV that's working with both. And so we feel positively about open source but we think that there's a role for commercial software companies to help make it bulletproof for that buyer and if you are a very large service provider you want somebody that you can work with that will stand behind it and support it and that's what we intend to do. Well as you said the fundamental value proposition sounds like yeah you're doing network virtualization, you're adding the elements required for interoperability and integration but also you're adding that layer of operational affinity to how telcos or how service providers actually work. That is a tough computing model. I don't know that open source is going to do that. There's always going to be a need to try to ensure that all these technologies can fit into the way a business actually works and that's going to be a software, an enterprise software approach whoever the target customer is. It's a great partnership between the open source community, commercial software companies like us and the service providers. To build this thing and we've seen that happen in enterprise. DevOps was that kind of a phenomenon. You have winning commercial software providers, you have a lot of open source and you have the users themselves and we think a lot of those concepts are going into this service provider space and for us it's all about at the end of the day we want to have the ability to get people to do their job faster. If things change in the industry a service provider using Pensa or an SI using Pensa can design, validate, build and run that next thing and blow it out to their network faster than anybody else. It's time to value and certainty that it'll work. And in many respects at the end of the day we all want to be digital businesses but if you don't have a network that supports your digital business you don't have a digital business. That is correct. All right, so last question. Yes. Pensa two years from now, what does it look like? Yeah, I think our goal right now is to line up with some of the leading industry players here, folks that service those large service providers and help them build these solutions and do it faster. I think our goal over the next two years is to become a control point for service providers and again folks like SIs that work for them and sometimes help run their networks for them give them a control point to adapt to new opportunities and respond to new threats by being able to rapidly change and modify and roll out new network services for new opportunities. You know, the thing we learned in the whole mobile transition is you really can't conceive of what's next. What's next two years from now in this space? Who knows? If your model is buy a bunch of hardware and depreciate it over five years you won't be able to adapt. We want to be. You do know that. We know that. We want to be one of those control points that helps you do that quickly without having to go wade into the code. So our goal is to allow, our whole tagline is think faster which means use intelligent technology to drive your business faster and that's what we intend to be in two years. Excellent. Tom Joyce, CEO of Pensa. Thanks very much for being on theCUBE. Thank you very much. And for all of you, this is Peter Burris. Once again, another great CUBE conversation from our Palo Alto studios. Look forward to seeing you in another CUBE conversation.