 It's The Cube. Here is your host, Jeff Crick. Hi, Jeff Crick here with The Cube. We're on the ground in Santa Clara, California, at the Open Daylight Foundation event, their second event. Foundation's been around for, I think, three years. It's their second event. And we're excited for this next segment, Jay Etchings, Director of Research Computing from Arizona State University. Welcome. Certainly. Well, thank you. Absolutely. So, research computing. What is research computing? So, what research computing is at Arizona State University? It's an amalgamation of two sciences. One, high computational or high performance computing and data intensive computing, or big data put together under one offering. And how do you delineate the two? One is really more just computational intensive versus all the data? So, the traditional approach to high performance computing is bare metal machines that do high computation for life sciences, for weather forecasting and all the traditional physics domains. Big data or data intensive computing is something recently adopted in higher ed in the National Science Foundation, which is what you think it is. It's Hadoop. It's big data analysis. It's analytics that fits in with our business school. It fits in with life sciences. All right. We're live. Thanks, Matt. And several other things. So, these two were completely separate before the research computing initiative came along and put them together, put them both on the same fabric so jobs can go from one to the other. Right. Okay, excellent. And so, what are you doing here? What's your role here at Open Daylight? So, I've been with the university for about two years, but my appointment to director of research computing only came about in the past eight months. At that time, we had a multiplicity of controllers. So, we had Cumulus, we had some NEC stuff. Everybody had their own and we were looking for some way to unify them. And Open Daylight gave us the methodology to do that, to get everything together, get everybody developing on the same path. So, before we went on there, you were talking about Open Source and Open Stack and Open Networking and Open Computer Project and we cover a lot of these shows. Talk about the impact of Open and all these different initiatives on innovation and what you guys are able to do at the university. So, one of the foundations that President Crow is trying to reach for and one of his key goals was to extend higher ed to all Arizonans. And to do that, we have to be able to extend research computing to all Arizonans. Well, as a publicly funded university, we have to do a fair share of that on our own. So, the Open methodology allows us to interoperate with all sorts of different universities, some of different socioeconomic backgrounds without the high price of licensing and proprietary and interoperability with applications. So, everything that we can do open, we do and we try to build a supported model around it so then we can put it into production. Okay, and then what about in terms of being really active participants in these foundations? How much are you guys contributing code back? Are you really active or are you just kind of taking advantage of it? So, we're definitely taking advantage of it. I mean, that's the ultimate goal. But we're on the customer advisory board for Open Daylight. We contribute code back to the Viata project which is an Open Daylight controller that Brocade sponsors. And we work pretty closely with our computer science division to develop these projects and we also, quite honestly, are looking for funding opportunities and what I mean by that is, you know, NSF-funded network opportunities or Department of Energy-funded network opportunities where we can build an application and then pass it back and get some funding and support for those types of things. Right, so we talked a little bit about some of the things you're working on. So, you mentioned Flow Guard. What does Flow Guard all about? What are you guys doing? So, Flow Guard is an open flow based firewall that we were using for host protection or endpoint protection. The current initiative is around obviously distributing that firewall so all the devices that are in that flow path can understand when a state change occurs or when there's an indirect violation versus kind of the nomenclature around firewall methodology that only sees direct violations to policies. So, this can look for state field changes and know to adapt to them within the firewall architecture. So, a big component of that is really this NFV that's emerging out of this movement as well that allows us to have a firewall that sits on all these different appliances. Right, but really one of the growing kind of constructs is this perimeter of security, right? So, the concept of a firewall, where's the firewall when everyone's running around with their own devices, that nobody's got control over when they're accessing all these applications and then that's before we talk about Dropbox and all the kind of whole host of shadow IT things that aren't spinning up in Amazon instance but taking notes in the meeting on Evernote. How does that play? I mean, in terms of really thinking about security and putting in systems that can meet the standard that they need to meet. So, first you got to take a step back and look at some of the nomenclature and who's defined it and who's driving it because no matter what we come out with for every term that comes out, whether it be cloud or even back to voice over IP or even the things that happened around SIP vendors adopt these things and then they apply their own definitions to them and that's how they sell them in the marketplace, right? So, when we talk about firewalls we're still kind of talking about traditional perimeter protection but it's also hybridized with an IPS component or a secondary UTM component so if you're looking at that architecture and what it looks like, so there's perimeter defense but there's also some inside detection and then there's an action that happens tertiary to that and then that action is sent off to the expert who might be the UTM device or where secondary deep inspection could occur because we understand putting a firewall or putting some sort of IPS device on everything similar to how we tried with desktops years ago and it sucked up a whole bunch of resources that's not possible in network equipment so we're not trying to reinvent the wheel we're just trying to take the existing components and make them inter-operate more efficiently and then watch out, here comes Internet of Things, right? so there's going to be IP enabled devices connected all over the place, yep, just a whole new challenge University is filled with that, 82,000 students so biggest university in the US, right? 82,000 students Yes, ASU is a really big school so with the funding goal of, you know, 700 million by 2020 and being a research one university we're measured by who, not by who we exclude but by who we include so we have everybody come they all bring their three devices their iPad, their phone, their laptop and what we've been trying to do through our BI team is establish ways to allow them to inter-operate with all these environments but do it in a safe way and that challenge in itself opens up a whole bunch of new doors not only that, but we offer internet to access on the research education network as well so we have a science DMZ where there is no perimeter security, right? it's friction-free to 100 gig and we use that for our specific for those research groups and that's a place where the talk about how a perimeter firewall doesn't fit anymore brings up a great discussion Well, it sounds like you got your hands full We definitely do and this summer, I'm surprised I made it here today because this summer we're really heads down trying to knock out a lot of these tasks so the beginning of the semester we'll be ready for the students but I'm sure they'll throw up some stuff we don't expect 89,000, that's a lot of people a lot of things, people are things too, right? A lot of devices, that's a lot of things, certainly Absolutely Alright, Jay, well thanks for taking a few minutes to stop by Thank you Good luck in the semester Go Son Devils Go Son Devils Alright, Jay Etchings here I'm Jeff Frick, you're watching theCUBE We're at the Open Daylight Summit Thanks for watching