 Welcome back everyone, we're here live in Silicon Valley for the open networking summit 2014, hashtag ONS 2014, this is theCUBE, our flagship program, we go out to the events extracted secretly from the noise, I'm John Furrier, the founder of SiliconANG, I'm Jordan Mike Coho, Scott Rainovitch from the Raino Report, analyst on theCUBE, Kelly Wanzer, CEO of Stateless Networks, welcome to theCUBE. Thank you. First time on theCUBE. You're awesome. Vanessa. Already I'm awesome. Thank you. Friend of Vanessa, it was Clodorati. Vanessa's awesome. Friend of ours and we met at Open Compute and then the OpenStack event in the Computer History Museum. We couldn't get you on theCUBE, but we got you on now. Thank you. So welcome. Tell us about the company. It's a startup. You guys are innovating in the space. We love talking about startups. You got the old incumbents. You got the new incumbents. Yeah. Trying to survive. And then you get the startups, which you're running one of them as the founder. Tell us about Stateless Networks and what you guys do. So it's a fantastic time to be a startup and networking. So you have this big transformation happening in the biggest, most critical part of the infrastructure. And you know, I come from a different space. So I came into networking from messaging infrastructure and security. Do I need to lean forward? No, you can leave that there. It's fine. So I came into the space from somewhere else and I came into networking originally looking at the challenges of big data infrastructure and sort of hit on the fact that there was a combination of a bottleneck in the network when you try to deploy big scale-out systems and this shift happening in networking technologies where a network was moving from hardware appliances to software and the software was disaggregating between the physical network operating system, which was becoming open and a virtual layer, which was going to lift the network up and make it really agile and flexible and elastic. And so seeing this happening and saying, well, what's needed to operate and manage a network like this? And so from my point of view, we came in and we started working on this problem of, well, you have this new type of physical network that's coming and you have this new type of virtual network that's coming and it envisions a software layer that operates it and uses intelligence coming from the network to operate it and it doesn't exist. And so we started working with the first vendors who were doing open network operating systems, Arista, and we put a software agent on the box and we started to take feedback from the network to use it to drive. Well, Jay Shree, she's been on theCUBE. She's an innovator. Yeah. She was doing this stuff when it was hard. She actually was telling us when she started Arista, you couldn't get any money for networking. Well, yeah, but to be fair to her and their team, I mean, you have Andy Beckelsheim, you have Candida, you know, you have some of the geniuses of networking. They had a lot of job. So they had a really great technical vision of where things are going. And now you see folks in, you know, because really where it's going is about this open Linux kernel-based networking underlay and then you have this virtual network all the way on top. And so the vision is I have a simpler network underneath and it's automated and it's operating very smoothly. And if something happens, you know, it works like an assembly line and everything's good. And on top of that, you have a lot of the brains, a lot of the richness, a lot of the flexibility that comes from letting people running applications say, look, I need some bandwidth. I want to bring these applications online and I can click to do that. And that's awesome. So are you guys targeting developers? So who's your target audience? Are you selling a box? Is it software? So we're going to say something a little bit different than you see in the cloud world. I'm going to say my target is the network operator. And what I want to do is bring to the network operator the visibility and the power and the control of these new model networks. I want to say to you, you know what? I know you guys in network operations and network engineering, you're there. You keep the lights on, right? And you've operated the network in one way. You've optimized against fairly static technology and you've done an awesome job. Now we're going to give you better technology. But we realize that you need to have high visibility and a high degree of control because you guys are, you are the, the, the bulwark of security, of compliance, of keeping the lights on. What's the psychology of the network operator right now? I just don't know. So I'm asking, is it, are they feeling like I've been battled down? I'm tired. I'm irrelevant. I'm being worked to the bone. I'm being consolidated. Do they, do they feel more? I'm not Amazon. You know, do they, do they feel like empowered, like the future's mine, the world's spinning in my doorstep? What's your take on these guys? Are you giving them an aspirin or are you giving them a new weapon while they above? So my experience is, it actually starts to separate out, that there are different kinds. And so you, you meet the guys who are energized and like, I'm going to learn Pearl. I'm going to learn Ruby. I'm going to, I'm going to become a network developer and it's exciting and there's never been a more exciting time to be a networking, maybe ever or maybe since the 80s, early 80s. And so you get some guys who are very energized and they've got, you know, one of the banks we work with, they've got a team of guys and they have coding books and data books and they're, and they're really energized. And then you have some guys who are really nervous and they've got these models that they're using to operate and they don't see how to translate them. And they're worried. They, you know, they carry pages, they know the requirements they need to meet and they don't necessarily see how it's going to work. And then you have the guys who are worried for their jobs. But Scott, I want to get Scott in here. Sorry, sorry. No, no, it's okay. He's, he's, he's to analyze and critique. What's your take on that? So a startup, you agree? It's the best time to be a startup. I'm curious, I'm curious how you got into this business because a lot of people spend their time trying to get out of, you know, telecom, but apparently I've never succeeded. But that's okay. Yeah. You, you are a serial entrepreneur, a serial entrepreneur. You've done several companies and then you came into this market. You hadn't been in it before. You got this big relationship with Arista, which that's not an easy thing to do. That's one of the premier names in this space. Tell us how. Well, and they're not our only relationship. Right. But tell us how you got that relationship, how you got involved in this market. So, so I actually, for my sins, you know, I really like enterprise infrastructure and I think it's a really cool area of technology, especially because you have, you have technologies that become widely distributed and once they're widely distributed, they can, they can become very static over time. So you get these big areas of kind of lack of innovation and then sort of sea change of innovation. And I was coming from email, which is one of the oldest technologies on the internet. And so, so enterprise infrastructure, I think is really cool. And I was looking, so I also have a background in data analytics and started my career in pharmaceuticals. And so I was looking at big data and so you read, okay, everything's moving to the cloud. It's going to be driven by data. And this idea of driven by data, I was, so I was looking at infrastructure for big data and saying, you know, there's going to be a lot of this world out. And so, when I got into looking at, I was actually looking at building like a green plum, network storage, compute, call the hardware, software management layer, and I had a couple of guys and we were working on the software management layer and we started with the network. And we were working with Arista and Coret and a couple of big Wall Street banks. And the other thing I like to do is very early collaborate with, you know, thought leading customers. If you look at financial institutions in particular, they're usually early adopters of technology. They identify the gaps and they try to find where people can innovate. So tell us what these early adopters are doing because you have an analytics platform and obviously there's all sorts of different applications and use cases for that. Tell us what some of the more cutting edge use cases are for that technology. For us and what makes us different is we're analytics plus, you know, analytics plus configuration or analytics plus change. And so we consider ourselves an automation system and in networking, in other areas of technology, this is where not coming from networking is interesting, in other areas of technology, systems, they have this concept of data driven systems, right? Data driven content, data driven ads, data driven. And in the network, it's a one area, it's not data driven at all. I've got monitoring over here and I've got people that pick up and manage what happens when I see something and I've got change management over here, I've got configuration over here. And so the idea of a data driven network where, you know, you're pulling, you're extracting real time data out of the network and if you want to make changes on the network, you want to do positive things like, you know, add new capacity or add new software or you want to respond to events like I see congestion, what should the network be doing. In order to do that in an automated way, you need to pull together the information about not only the network but what's attached to it, what's of the stack from it. And so that's the piece that's missing. Yeah, I just did a whole research report on this, very topic actually, and Vinod Kozlo was speaking this morning about automation and adaptive networks, pointing out that we're not really in an adaptive world. The analogy he used, which happened to be my same analogy, I don't know if he stole it from me, but airplanes, right? So if you look at, if you go into an airplane, you know, 95% of the flight is automated. The pilots are just drinking coffee, hopefully watching, you know, this instrumentation. They might land manually, but the plane is fully capable of landing itself, and this doesn't happen in networks today. It's still a manually driven world. So you are, your goal is to be the autopilot for the network. But tell me, do you have a customer use case you can talk about? So there are a number of them, right? And so I'll give you a simple use case and then I'll give you a more sophisticated use case. So take a simple use case today where I have a manually operated network, and 70% of the outages on my network are caused by mistakes or misjudgements in manual operation. So I'm not, they could be typos, they could be, I thought I needed this snippet and it was that snippet, or I didn't think this would affect that, but actually, so, so one of the very basic use cases that we can do before we do anything risky is come in and interact programmatically with the network. So I'm going to start getting real time data, which the traditional polling method doesn't give me. So the old way of talking to these network appliances, you know, polling the network, I've got 10 minutes old data. No, I need, I need data that's a couple of seconds old. So I get this real time data and I'm going to take it from the physical network. If I've got a logical network, if I've got a virtualized network, I'm going to take that data. I'm going to take data from everything that's attached and it's status and put all that, maybe, yeah, whoever's tweeting. I'm going to take all that data and then when somebody, then I'm going to, I'm going to allow operators to write policies or logic that say, how should the network be? How does each device need to be? What does the system need to look like? And then when somebody goes to implement a change, I'm going to run checks. And you know what? If that shouldn't be happening, I'm not going to let it happen. So that's a basic use case. We think of all this like guardrails. And you want to set up those guardrails that's this data driven logic. And then that's going to allow you to do more sophisticated things. So one of the biggest preventable causes of congestion on a network, and congestion is where you got too much traffic. It's just like traffic. One of the biggest preventable causes of congestion on the network is, in this new virtualized world that I'm able to create, I can move things around. And I can move an application from here to here because I think it's a good idea. And when I do that, I take a whole bunch of bandwidth, and I can make congestion. And I can actually interfere with applications that are operating. Sure, so what, how can we trust that stateless is going to do this better than, say, Cisco or Juniper? Because you would think that the bigger players have already are already working on stuff like this. What, when you go to a big customer like Verizon or a big bank, you said you work at the bank's JP Morgan, how do you convince them that a little startup is the way to go? Well, one of the things about this is it is it requires combining different ways of thinking. So you have to take a deep understanding of the way networks operate in their hardware and these new network models. And you're going to combine it with a software architecture and a data architecture that does look like Twitter. So our system actually looks more like Twitter than any traditional networking system. And so when you, when we come to customers, they realize, OK, this is the innovation. It's not just an innovation in technology. It's an innovation in mindset. And there's more recognition that that innovation in mindset is more likely to come from the startup. And we actually see the companies that we work with, banks and service providers, they're setting up little teeny mini startups in Palo Alto. You know, their own little startups to try to emulate that. Because you need this sort of, you know, what sort of looking for radical thinking or this sort of, you know, willingness to be a rebel and say, no, you can, you can use a Mongo database on a network management system or no, you can use a bunch of open snippets to do something. And so it's different. So I think there's a lot more acceptance now than there used to be of a startup doing that. It helps in our case that, you know, we're walking into a gap where we come in with use cases that are very pressing for what people want to do. Sorry, I keep talking. John, you got to come in and I can tell. No, we just got them just much. Keeping an eye on the clock. Make sure to have them left in the dark here. Sorry, I'm supposed to talk in sound bites, aren't I? No, no, don't be yourself. It's the cue, very, very friendly place. All right. I want to get your take on the show here, honestly. As an entrepreneur, you got to get a position in there, find a little beach head. The focus is critical. You got a nice focus. Share with the folks out there. What is this show about? Why is the opportunity and compare and contrast after the big megatrends of what's going on in cloud? You mean the show, the ONS show? What's going on here at ONS? Why is the show important and compare this networking future to what's going on in the cloud and what are the big things people should pay attention to? Well, I think the open networking summit and the open networking ideas in general are really important for people who care about cloud and for people who care about IT and business systems. They're really important because the network has been the bottleneck. And now you have a networking. This revolution that's coming. And it's technology transformation. And you have an industry that's been pent up. So the open movement, I think, is where things end up. And the network becomes as dynamic and as fluid and as collaborative as other areas in the cloud have been. So I think everyone in the cloud space should be paying attention to what's happening in networking. And it sounds like, you know, with Vinod's comments and others' comments that they are. There's radical change coming. Innovation. Kelly, great to have you on theCUBE. It's a pleasure to have you. You're awesome. Thought leader, great. And then the founder of a company, which is amazing, which is, you know, as founders, we all know how hard it is to build a company. So share with the folks really on the last word here. What's going on with the company? Your goals for the next year? What are you looking for? Are you going to hire? Are you looking for funding? Just give a quick update on the state of the company, of stateless networks. Give us the update on the state of the stateless networks. Yeah, state of stateless networks. So we're in growth mode. So we are definitely hiring. And we're especially looking for people who are very passionate and very innovative minded and in engineering. But for the company, we're really looking to establish ourselves as a dominant force here in providing this essential layer in the network and helping to drive people to think about how networks operate in terms of the feedback that they require and the new operating models, which say the networks are going to be comprised of software and they should be driven by software. So we're really excited to spread that message. We're really excited to be here with you guys in theCUBE. I'm sorry I missed you last time. This format is fantastic. It's really, really cool. We love having you on and again, sharing your information and what you're doing to the crowd. We love that's what theCUBE's all about. Big, big place for ideas to grow and have great smart tech athletes here. This is theCUBE where we'll be right back with our next guest. We are live in Silicon Valley for the ONS Summit, the future of networking, impacting cloud, software driving, software networks, code writing code, infrastructure as code. You call it what you want. It's a revolution. We're here at theCUBE. We'll be right back.