 From San Jose, in the heart of Silicon Valley, extracting the signal from the noise, it's theCUBE, covering OCP US Summit 2016, brought to you by OCP. Now your host, Jeff Frick and Stu Miniman. Hey, welcome back, everybody. Jeff Frick here with theCUBE. We are getting towards the end of day two of two days of wall-to-wall coverage at the Open Compute Project Summit in downtown San Jose, California, where it's all about the hardware, it's all about the infrastructure, it's all about what's happening that enables cloud, enables mobile, social, big data, all the things that we've come to love and really take for granted in a way, but this is really where it's happening. We're excited to have somebody who's really right in the thick of it, Ehab Terazi, the CTO of Equinix, welcome. Thank you very much. Yes, we've had Brian Lillian before, and I know you guys made a lot of news years ago on kind of the direct connect with AWS, and so you've been, you know, kind of unnecessarily maybe in everybody's radar, maybe just a step below Amazon and Azure, but you guys have been really active in the cloud space for a long time. Yeah, we have, over the last few years, we had hundreds of clouds come to our data centers, and Equinix is now known as the home of the cloud. We've had the network ecosystem for a very long time, but the last few years, major deployments from all the key cloud providers. We have over 500 cloud providers in our data centers now and almost 1200 cloud and IT service providers. Oh, good numbers. Yeah, so, you know, you guys understand the ecosystem required for cloud. You know, we've been talking about OCP, that's about 100 providers there. You guys are new to OCP, so what brought you guys here? You know, why is it important to Equinix and your ecosystem? Yep, thanks you. This is our first OCP meeting. We joined OCP a few weeks ago under something new called TIP, Telecom Infrastructure Project. That's the same thing that was announced at the Mobile World Congress, if you saw the keynote. And TIP brought in the big network providers, Verizon, AT&T, Deutsche Telecom, and we joined that as well. Those are our biggest customers. We provide network interconnection, and that's why we joined. And what we announced a couple of days ago is that we're expanding that partnership with Facebook and Mesosphere to do even more with OCP. All right, so I believe in the news there was a discussion about Equinix using the wedge switch. We've had a lot of discussion networking probably top discussion, so explain to us why you're doing that and what that really means is the whole cloud now going to be connected by Facebook switches? Yeah, I think the bigger premise is that our customers are adopting OCP and taking advantage of all the OCP standards, the efficiency, the innovation. And I'm sorry, when you say your customers it's the Telecom providers that you're talking about. The Telecoms, the cloud providers, the 100 providers that you have here, most of them are Equinix customers. So all these customers are bringing OCP hardware and standards into our data centers. And what we want to do is enable the next generation of software-based stack to support all this OCP hardware, things like SDN, IoT, et cetera. And with that, we want to leverage Facebook's amazing innovation engine and all the stuff that they've done. So that was part of the announcement. Okay, so you're saying lots of the cloud providers, the one that's not here that we've had a bunch of questions on is of course Amazon. So how does that play into your thinking? A big customer of ours as everybody probably knows and therefore we continue to support Amazon with whatever they want. Our model is based on building ecosystems. So if people want to use the equipment that they built in-house, or they want to use vendor equipment like Cisco, Juniper, or they want to use OCP hardware, we support all of that. And we have massive public exchanges. We also have cloud exchange that we built the last few years to provide direct connect to AWS Azure. But what we're saying now is also for the community of customers who want OCP hardware and they want to innovate there, we're going to be there for them as well. So we support all of it. So why the move? Why did you join Atlas Beats? Been around for a while, you guys hadn't been part of it before. What was kind of the catalyst to say, hey, this is something we want to get involved in? Yeah, this community have gotten to the point now where after innovating in servers and storage, they got to the point to build credible, high-scale networking equipment. The Wedge 100 is one of them. And the community got to the point where there's enough players now want to get together and build open source software to support all this dispersion of hardware. So that plays completely into our game, into our value proposition to build ecosystems. So with this movement, this was the time to join. And you can see all the big telcos announced joining at the same time. So we want to be at the middle of it. We want to solve the issues and we want to be building the next generation interconnection and be forward-looking on technology. So in one of the keynotes yesterday, they had a panel of your customers, those telecom guys. There was a question, AT&T, Verizon, Deutsche Telecom. These guys are competitors, but they're working together. Can you give us your insight? I mean, the telecom world has been a lot of fighting over the years. So are these guys really working together to help the users? Additionally, the telecom companies work very well together through standards bodies, IETF, IEEE, and that's how companies came together, as well as the vendors like Cisco, Juniper, to create standards and build the next generation equipment. I see this as another version of that. I see this as the cloud software world version of that. Where the biggest users and innovators, you know, Microsoft, Facebook, et cetera, Google are all coming together to say, here's what we're innovating. And the network's being part of that. So I see the model continuing. It's just in a new space now. And that new space is more open source driven than the model before where people built, you know, based on specs and standards. And the model here is much more faster, much more agile with thousands of players innovating every day compared to maybe longer cycles that took months to develop. Okay, so being an open source project, what does Equinex bring to the community? What does Equinex? What do you bring to the community? Will you be making contributions? So let me explain what the three of us are doing. Facebook is bringing the OCP skillset hardware and knowledge and innovation. Mesosphere is going to bring their knowledge of the e-source management, container management, which will allow us to build an exchange for this next generation. What Equinex brings is the ecosystem. We have 1,000 networks, 1,300 clouds. We have a very high quality and high performance operations model. We are able to create a physical place where people can come in and test and operate and also a place when we build this for people to deploy. One of the things that are missing today in this open source world is, where do you take your hardware components to deploy? Where do you put it? What kind of software? How do you make sure it can work with other components? So all the stuff that traditionally happened to standards bodies that haven't happened yet, sufficiently, there's a lot of innovation inside a single cloud like a Microsoft or Facebook, but now all of us want to come in and say, how do you build that for the multi-tenant environment and that's Equinex business model? And so we feel this is going to really spark the innovation and bring new players to our data center and allow a lot of people to start to innovate at a faster scale. Okay, so can you unpack for us a little bit the mezzo's piece of that? I think much of our audience are probably familiar with mezzo's sphere, but it's really grown quite a bit in the conversation recently. Mezzo's sphere is already a key provider for a lot of the networks and a lot of the cloud providers here. And what they do is they do basic resource management, plus they also do a lot of specialized services for them. What the piece that attracted us about it is, allows you to do a federated exchange model where customers can own their own containers, control the quality and policy on them and be able to pass them to each other in our data center at very high performance, probably very low millisecond or maybe microsecond latency. But every customer will be able to control the domain, control the containers and policy and feel comfortable that as the service gets split across multiple infrastructure and clouds, they can still feel comfortable, they can support the performance. It's also very flexible, allowing NFE, SDN, all kinds of containers that people are looking for a place to put to go on top of that platform. We're at the beginning of that journey, we see unbelievable opportunity for hybrid cloud. Today, hybrid cloud is a little clumsy for a lot of customers and we want to make it seamless. This is critical for enterprise customers coming in and we see a very good path to that. We also see the need for IoT ecosystems to connect somewhere and be able to move containers and this is going to also be helpful. Yeah, I'm a lot of good points there. So I want to touch on one from the IoT standpoint. You have great connections into the cloud providers, but one of the things that you definitely need for IoT is resources out at the edge. So how do you work with your customers on that? How do you bridge between the edge and the cloud? What is that interconnection? Yeah, IoT is a natural extension of our model. It's very exciting for us because all the wireless devices are mostly coming through the same networks. The same thousand networks we have are the ones controlling the wireless networks. And then the IoT resource management analytics and all the back office processing is done by the big clouds, but also our customers. And the new players, things like what Cisco is doing with Jasper, et cetera, all these guys are also part of our ecosystem and part of our customer base. So physically, most of the IoT ecosystem is in our data centers. What we need to do is enable that exchange of data at a different layer, at the application layer so people can take advantage of the analytics and be able to take advantage of the scale. We see the edge of IoT remaining the same edge of what Aquanax has today and the customers for the most part are the same with few more players in the ecosystem. It's just a matter of enabling a different connection at a different layer to energize and open up that model. Because, for example, some of the IoT ecosystems are done in open Linux. This is above the IP layer and we have to figure out how to enable that. The partnership we have with Mesosphere will enable us to do some of that stuff. It's good stuff. I mean, you guys are right in the middle of all the action. You got to be really excited. You're in a terrific position. I'm curious to get your take on the evolution of the data center itself. We've always talked about the mega data centers because of latency and you guys have Direct Connect AWS. You saw that and that's a big issue. But now with IoT, I think what's interesting is now we need a more compute and more store out to the edge. Where exactly and how close to those devices is still to be defined on an application-specific process. But you see the emergence of all these different footprint data centers based on whatever the requirement is as opposed to just kind of piping everything back to the mothership? Yeah, I think the picture has become clear in the last couple of years. And I agree with you, we're an unbelievable position. It's so exciting to be in the middle of all of that as Aquanax and us as a company. What happened over the last two years is we started with the backbone of all the networks in our data centers. And then the last few years, the backbone and the edges of the big cloud providers moved in. And now you have the biggest infrastructure companies, networks and clouds. What's happening now is the next set of backbone changes, the submarine cables, is starting to be done instead of stopping at the beach to go directly into Aquanax data centers. We have 25 submarine cable projects in the works and some of them were any one. 25 submarine cable projects. Yep, where people are going to come in, skip the beach and go with ducts directly under the beach all the way to our data centers because they're looking for clouds and networks. There's no need to stop somewhere out there when the destination of your traffic is at Aquanax. So that's going to redo the global backbone of submarine cables again against our data centers and Aquanax in the heart of it. And then we see IoT as a natural extension of what we have with the networks and clouds with massive volumes of data and also billions more devices coming in. And with it comes the complexity of 5G, new unlicensed spectrum, new Wi-Fi and the need for analytics cloud-based architecture. So all of that plus the fact that more and more clouds are showing up, more and more application providers. So it's unbelievably exciting. Our announcement here with OCP and what we're doing with Mesosphere is we want to allow our customers to come in and continue to go innovate unstopped. If they want to use OCP hardware, we're going to support it. If they want to use customized hardware, they can do it. If they want to go to the application layer or move containers with policy, we're going to also be first to market and innovate on that. So we feel we have the continuum for customers from public IP, cloud exchange, and now the software layer, OCP hardware. And I mean, I see it every day, that just the momentum is unstoppable. The last thing I want to add is the thing we're really focused on right now is enterprise customers coming in. And now all the pieces are there for them to build hybrid cloud. They can do multiple clouds, multiple networks. And I think you're going to see a lot of our innovation towards enabling them. This piece we announced with IoT and multi seamless multi cloud is specifically targeted to those enterprise customers. What a great close. I don't think I could do any better. I know I couldn't do any better. So yeah, thanks for coming on and sharing your insight. Really a tipping point for Equinix as well as Google to join OCP. And you said you did it because the ecosystem has now developed to a point that it's time. So thanks for taking a few minutes with us. Yeah, thank you very much. It's an exciting time. Absolutely. So I'm Jeff Freik with Stu Miniman. We're wrapping up day two here at Open Compute Project Summit 2016. Again, I want to quick shout out to Micron, Melanox, Pika8, our sponsors without sponsorship. We can't bring the guys, the crew, the gear to cover the events like we know you want us to do and we like to do. We got a really busy month, it's March. So that means March Madness. So keep an eye out for March Madness. We run a concurrent with the real tournament. Cube Madness, Jeff. Cube Madness with March Madness. March Madness and Cube Madness. Thanks Stu. So go on. You'll see, we're working on the brackets. Vote for your favorite Cube interviews. They'll work their way up the brackets. We got a busy month coming up. It's March. We got Open Networks coming up. The cloud march continues. We're going to be at the Google Next show in a couple of weeks. Oracle Cloud World in Washington DC. And then we wrap up the month of March with Big Data SV at the Fairmont Hotel. Our signature event that we do twice a year once in New York or once in San Jose. And that will really wrap up the month of March, but we're off into the races. Go to siliconangle.tv. Look at the upcoming to see where we're going to be. Go to wikibond.com. Get the latest research and what the guys are working on. And then of course, siliconangle.com for the latest news and tech. So thanks again for watching. I'm Jeff Frick, signing off here from Open Compute Project 2016. Thanks for watching.