 Live from San Francisco, extracting the signal from the noise. It's theCUBE, covering Console Connect Live 2015. Sponsored by Console, there's your host, John Furrier. Okay, welcome back everyone. We are live in San Francisco for special CUBE coverage from SiliconANGLE medium. John Furrier, the founder of SiliconANGLE. My cohost, Jeff Frick, general manager of theCUBE. We are live at Console Connect 2015 and our next guest is Brittany Brockman. CEO of Edge Connects, fastest growing data center now, 23 and growing. Welcome to theCUBE. Thank you very much. Pleasure to be here. I love that stat, fastest growing data center. I thought data centers were supposed to be shrinking and getting built in Alaska, Arizona, Iowa, farmland. Couldn't name better spots. That's probably fastest growing data center plural. You hit the nail on the head. We've knocked out 23 data centers in the last 23 months. These are data centers we've built, not bought for the most part. And you hit on the very markets. You hit on the very thing we're doing, which is putting data centers in markets that typically have not been peering markets. What drove that was this explosion of over the top content. Explain this peering market dynamic. Was that center on the May, West, May, East, major kind of network overlays or autonomous networks? Correct, that's kind of what if you think about the original peering markets, May East or May East and May West, depending on which way you're sitting with the table. And then it kind of expanded to six to eight major peering markets. And then, lo and behold, along comes my kid's generation. And they start consuming. And these things. Exactly, and mobile phones. And they start consuming incredible amounts of over the top content, especially video content that just is tremendous amounts of bandwidth being consumed. And what we anticipated and saw was the opportunity to really bring that content to the eyeballs rather than the eyeballs going and chasing the content. And to do that, I think our dream was and vision was you needed to build data centers near where the broadband traffic is aggregated in pretty much every market in the U.S. ultimately. And bring the over the top content there, first and foremost, now cloud. And we'll see where it goes from there. This is a nice natural extension of expansion strategy. We saw that with head ends with cable. We saw it with cellular, with towers. You guys are doing the same thing where it's a new way where you're going through where the demand is. And that is not necessarily mobile. That's just all internet users. That's correct. It's very, you know, in real estate, it's location, location, location. That's actually how we viewed the data center. When we thought of this model, we actually thought of a tower model. And we said, this is going to be a search ring. Only the center of the search ring is actually going to be the primary aggregation router or routers in a market like Nashville or Detroit or Pittsburgh. And what we're going to do is put a data center near there. That may not be the cheapest real estate in all cases. So this isn't a land bank and we build a data center. We're actually going to put a data center where the traffic is being consumed. And then we'll connect directly into whoever the largest broadband provider is in that market. Make sure that that works well for them. And then make sure it works well for the content and now cloud ecosystems. And I was telling folks earlier, we built the first edge data center in July of 13. There's 23 now, which I think just, you know, the ecosystem spoke loud and clear. And it was one of these, once they got it, once they did it, they've been just a wonderful customer base. And our customers dictate where we go. It's kind of a great way to do it. Now we cover all the shows with theCUBE. We've done 70 shows last year. VMworld, we used that last week. IBM, HP, Google shows, Facebook events. And one thing that's very clear is that the workloads are driving the new architecture, meaning the apps are driving the infrastructure. So you mentioned over the top, that's fallen up and that's now it's going to the mobile. So the edge of the network, not only is the data center, it's location, real time location with devices now, internet of things, self-driving cars. Yeah, I'm going to put you on my fundraising. You've got it. And it's interesting when I- We take 20% of all exits. We can do it in every other bank. In every department too. No, it's, you hit the nail on the head. It's funny when I, when myself and the co-founder Edmund Wilson, when we thought of this, I tell you we were mobile centric. We had this grand vision that was all going to move to mobile and that's how it was going to play out. Turns out there was a whole lot of consumption on your basic broadband networks, but it's really this year, it's really this year where we started to see the transition where mobile's catching up and you're absolutely right. So we started with the over the top content, cloud followed, you hit the nail on the head, internet of things. Now you start to look at some interesting announcements like Verizon doing their TV over the top and this interesting marriage of AT&T and direct TV. And all of a sudden those whole worlds are blurred and it's truly content everywhere. Well another thing that I'll add to your fundraising deck would be the identity systems are merging, right? So you're seeing that hybrid cloud is showing that you can run infrastructure anywhere, or apps anywhere, but identity of the user could be driven by the app. So the identity login, you can say, hey I'm a diss subscriber but I'm here, I'm on HBO Go, I'm on the mobile, I'm in Nashville for a show, I live in Palo Alto. I mean this is complex tech that can be simplified by an edge, how do you talk about that? So we did, you know, I mentioned the over the top, I mentioned cloud, we've talked about some of the other stuff. Just in the last week or so we did a press release with Cloudflare and another great company, rapidly growing company. But it was interesting, the focus of our press release was actually around security. And the need for getting security as far out in the network as you can. Simple, most simple, probably look at a DDoS attack. You want to stop that as far out as you can. You want to put those measures in place. But you just hit the nail on the head, which is my identity is in my phone right now. Me is here, and the ability to kind of bring my version or my cloud with me is just the exact type of things we see happen. So we're pretty excited about it. Talk about some of the technologies in the hood. Flash, you're building data centers in real time, so you're looking at the OPEX, CAPEX piece of it. Can you share some best practices in terms of what you guys have done? Obviously, you got converge infrastructure, you have flash, you got the big data opportunity. And two, can you talk about the geography challenges on a state by state basis with respect to data? Okay. You don't have time for that? I think they're going to cut us off. So let me start with some of the best practices on what we did and what we learned from others, quite frankly. When we looked at the, you asked about the OPEX and CAPEX trade offs, it's interesting. When we looked at the edge markets, what we realized is this is something that needed to think about a different OPEX strategy. Because these data centers didn't necessarily need to be 10, 20, 50, 100 megawatt facilities. In fact, what we had to be able to do was create the same level of economics you might get in that type of a center, but in a two, three, four megawatt type of facility, because our customers are, they're actually really, really smart, right? The reality is the largest broadband providers and the largest content and cloud providers, they're loaded with smart people, smarter than we are in many ways. So what we saw was a need to really change the operational support systems in the data center industry. I came out of telecom 20 years ago, 30 years ago now, I hate to admit it. I started in the Bell system and it was remarkable when I looked at the data center space, how kind of yesterday, the operational support systems were. Nobody had actually looked at it and said, if I can run a global telecom network with no people, how come I can't run a network of 20, 50, 60 data centers that are critical nodes on a network in a highly automated, super secure fashion? And five, nine, so you need to have a high availability. Yeah, tier three, certainly tier three does our inspect and operate it. Yeah, that's, you don't get to, you don't get to, that's a day right, so it's mission, right? Right, so it's mission, and that's right. And so it was interesting, we looked at it and people just didn't think of that way. A lot of them thought it was good systems for an individual center or for individual pieces of a data center, the equipment and so on, but not a holistic view that went from security to data center information management to building management, trouble ticketing, the whole thing. I know we've done it right because our customers actually use our system to manage their own workforces inside our centers. That's the ultimate compliment. But we invested a lot in the OSS. So we can actually run these in an automated fashion. Now that takes some getting used to. You know, if you come from the old world, people are convinced you have to have people there, but we can guarantee response times. You pick the timeframe you need. Does your OSS traverse multi-faceted? It does single pane of glass. You as a user, if you were my customer, you'd have a single pane of glass. If you were in all 23 facilities, you'd see into all 23 facilities you could manage raising and lowering the loading bay information. You're looking at all the data centers as one system, one resource pool. Now all across security, across trouble ticketing, across, you actually see all your equipment, you see your temperature, your power consumption in real time, your humidity on every single rack you have in every single facility. Everything's tracked with the SLA. So we did that on the OpEx site. That changed the game a lot, right there. That led us on a five, we don't call it one to five, two to five megawatt facility, as though we're 20, 30, 40, 100 megawatt type facilities. Does that shift your focus to allow you to run faster on the build out and other things? Yeah, let's me do wholesale at that level. Let's me do wholesale economics. Let's me do wholesale economics to sophisticated customers and do a long-term... Sophisticated customers are used to paying full freight for... That's right. And they're telling you where to go next, right? And you know something? We're okay with our customers telling us where they use it. So you're a price advantage, because you're basically those lowest wholesale, but they're used to paying retail. That's right. That's right. And so it's kind of retail-sized deployment at wholesale economics. That's right. On the CapEx site, the rule of one, my friend. What we did was we iterated, we iterated, we iterated. My center one and two and three are three different designs. Then we stayed the same. But what we did was... So you expected that. You iterated... We expected it to get bigger, by the way. It turns out one, two, and three, we had to grow across the three. We learned that quickly. So one and two will be like the History Museum. Remember that one? Yeah, we... You got it. They're a little bigger now. Punch cards and... Now we're kidding. The uptake... The uptake, right? That's not that funny. You said you were from the South Coast 30 years ago. And I had to use punch cards. That's not that funny. The... But what we did was we actually came up with a modular design where we could literally kind of start off with a one or two megawatt pod and then expand it to, you know, call it three, four, five megawatts depending on what it is. And we set a hard design goal for us. What we wanted to be able to do was try to build at a KW rate of 50% what the Uptime Institute said you should be able to build that. You can't do that if everything's customized, right? So we really... We set that out as a design goal. I'm not there on some, not at all. You asked about the differences. There's actually a difference, right? There are certain states where labor costs more, where you have to use union labor or not use union labor. And, you know, those drive material cost differences in it. And let's not kid ourselves. Land in some of our markets like Boston and Miami, that's not cheap. Land in others of our markets, such as Detroit or Pittsburgh, are more affordable. And so there are definitely variables in there. The challenge for us was, could we largely provide a within, call it 10% consistent pricing mechanism across all that. So I want to ask you about, just in compare and contrast a little history lesson into where we're at today. So totally get the model. But let's go back to the web, internet. AOL was just a small little startup, UUNet, all these exchange points, they dial up with phone lines. And then those were concentrated infrastructure points that kind of dictated kind of how things were done. And it was not that very innovative. And then things started to expand. And then like Yahoo and the web came, and then Akamai created this overlay. Is this similar concept from a data center standpoint? You guys building kind of a, is it an overlay network? Is it pure network, peering network? How do you describe that? And what does that give you as an advantage from a customer standpoint? What do they get from you? That's the equivalent. And Akamai was clear. You can go to the edge and your images load fast based on the dial up. So a customer of ours is Akamai, right? It's an extremely valuable customer of ours. And what we've tried to do for Akamai is say, hey, let's try to give you a, I don't know if it's an overlay or an underlay, but let's try to provide a set of infrastructure in a whole bunch of markets that here too for didn't necessarily have that peering infrastructure. So we tried to build that, maybe it's a fill in, a densification is probably the best way to describe it. It's a densification of infrastructure across the entire US so that now your Akamai or Limelight or Fastly or Cloudflare or name your favorite or- They can't scale. They need you to scale. Or Google or we hope Google said- Densification. It's, and so what we are is providing a densification of that- Densification goes to API efficacian of the world. Densification, making it more dense and not necessarily to watch more movies. To, yeah, so you can watch more movies. So you can get it more cloud services, a lot of what we're talking about today. And do more Snapchats. And all this good stuff. Yeah, that's exactly what's going on. And so that's what we've done. We've tried to provide the infrastructure you might see in Ashburn, Virginia or in Santa Clara, California. We've tried to- So what's next for you guys? And how do you fit into console? Because obviously you're now becoming essentially a point of presence in these local markets. Yeah, we think console's a great idea. We think in many ways, I'll use my example in Detroit or in Nashville. If we can get our console and us as we continue to expand our relationship, they place a console node in our facility in Nashville. Now all your enterprises in Nashville have a wonderful on-ramp to the console ecosystem. Some of which is the same as ours. Some of which expands our ecosystem. So I now have a great on-ramp to go. I can be a Nashville business getting on the console on-ramp here in Nashville, going to Ashburn, picking up my cloud services on a very direct, secure, highest quality connection possible. Yeah, mission critical, just to extend out the connection to the edge. And now what we've done is kind of make an on-ramp in every market. That's awesome. Final word, I wanted to get you to your point of view. Share with the folks here watching live and on demand who didn't make the event. What's going on here at Console Connect Live? What's the vibe? Who's here? What are they missing? What's going on? Give a quick summary. I think the vibe is a little bit of wow. That next generation of software-defined connection is here. I don't think it's, I don't even know if folks had thought about it in some ways. I think we got infatuated with SDN and NFV and some of that, but there's an element of it here in software-defined connection. And I think there's a sense of it's real. I think that's the real vibe I get. I think the other big thing we're hearing today is the ecosystems that are starting to circle around and starting to really get excited about it. I'm seeing big key anchors to lots of ecosystems sitting at the table today. And that's exciting. So one of the things I want to just, Mark Andreessen just retweeted one of my comments I mentioned earlier about 10X developer, I was referencing to one of his comments. He said, in DevOps, in the development world, with cloud, one developer's like 10 developers in the old days. Network Engine, we coined the term here in the queue this morning, 10X network engineers. There's not enough guys out there to make your market. So what we see is this console connected, 10X network engineer, they're making your market for you. They're essentially saying, hey, you don't need to be a network total geek. You can be sufficient enough with pushing a button, knowing how networks and provisioning basic stuff. And they're up and running. You can be competent in IT and be a network engineer in many ways. And be able to do some of the things that before, here too, for you might have to hire an expert network engineer. And you're lucky with that too, right? I mean, just not enough people in the world. There are. I mean, you know. How do you know if it's good or bad? How many Bill Norton's are there out there? How many Paul games are there out there? Not many. There's only one Bill Norton. I just want to stop here and read it out loud. Thanks so much for sharing your insights. We've been a little long there, but great content. Congratulations, and how old is the company and what's next? Company was founded in the middle of 2010. And what's next? We're actually this year moving outside the US. We'll be moving internationally this year. So hopefully when we do this next year, we'll be talking about the first 10 or 11 markets outside the US. Well, we'd love to find out what adventure you're going to, get you on theCUBE, and hear about your global consumption contract with customers and how that's going, especially with your data in multiple countries and all these jurisdictions. That's a whole other CUBE segment in itself. Yes, it is. We're here inside theCUBE, sharing some insight with you. With sharing the data, we'll be right back with more live in San Francisco after this short break.