>> Announcer: Live from the Computer History Museum in Mountain View, California, it's The Cube! Covering DevNet Create 2018, brought to you by Cisco. >> Hello everyone, welcome back to The Cube's special coverage of Cisco DevNet Create here in Mountain View, California, the heart of Silicon Valley. We're at the Computer History Museum. I'm here with Lauren Cooney, who's co-hosted with me for the past two days' wall-to-wall coverage. We've been kind of getting down in the community with Cisco's DevNet Create, which is an extension to their main DevNet developer program, which is mostly network-centric, classic Cisco developers, guys configuring networks, the power players in the enterprise and all around the world as we know it. But now that the cloud native's taken off we're here exploring that DevOps equation. Our next guest is Rowell Dionicio, who's a network engineer at Stanford, welcome to The Cube, thanks for coming on! >> Thanks for having me. >> Love Stanford, very progressive, always having state of the art facilities, I mean, hell, the campus just gets better every year. It's like a cathedral of new buildings- >> Rowell: It's always under construction. >> Always under construction, football team's been decent for the last decade, which is good as a season ticket holder, but the network drives it all, the great facilities there. So now see you're here as a Cisco practitioner. Networks have been running the show for many, many years, now comes cloud, Stanford's got a lot of stuff going on on campus, obviously academic computing, business computing, is there a lot of cloud going on there? And is there a lot of DevOps happening? Give us a quick take on- >> There's a lot of cloud. I come from the infrastructure side, so this is my first time here at DevNet Create. I wanted to get a feel for what's coming. What do I need to learn in order to make that next step to help bring a better network, help students connect, help staff connect. >> Yeah, the network guys have all the power, always have been, but what's interesting is Susie Wee at Cisco, she's leading the team around DevNet Create, we talked at their last show in Barcelona about on top of that she was introducing, which I love, network ops. Which is essentially what DevOps is, but making the network truly programmable, at a level where it's a service. That's the nirvana scenario, that's the dream scenario. >> It is, yeah, and we actually have a lot of that already in place, but obviously there's still a lot of areas that we can improve, especially in maybe the wireless space, and that's why I'm here. What can I do on the wireless side to help drive that? Is there something that we can do better, more efficiently? >> I mean, we always do this ad hoc, unscientific surveys. We interviewed the guy who runs the stadium for the San Francisco Giants, the guy who runs the stadium for the Sharks, Levi Stadium. The number one complaint is wireless. And it's like in the Maslow's Hierarchy of Needs. >> It's a tough one to crack. I hear those complaints, I get 'em, and I try to fix them as quickly as I can. But it's one of those things where you can't see it, and I think wireless is just such a robust technology that it'll work even in the worst scenarios. >> That enables a lot of IOT, but also the consumer side with the students and the faculty. Is the strategy at Stanford just to blanket coverage of campus, you guys just throw the RF all over the place? >> We don't, we don't just put it everywhere and anywhere. We actually think about it and it's not just in terms of coverage, it's also capacity and how people want to use it. And so we try to design around those requirements, and also if we're bring in IOTs, how do those devices work with wireless? Am I going to deploy something that those devices actually work well with it? I don't know, and so we have to do a lot of testing, ask a lot of questions. What's the use case? Where do they want it? Is it even possible? >> The analytics are interesting, right? You look at the patterns, and they're humans, they're connecting, so you can see where the crowds are, probably, I imagine you look at the concentration? >> We're not even at that point yet. We're actually just looking at it. That's why I'm here, to see how can I do this on our network: is it possible, how do we deploy this and make it work with other schools on the campus? To see whether or not it's a great use-case for us. >> 'Cause the schools have their own kind of kingdom kind of thing, or how does it?- >> A little bit, yeah. >> So there's some job there maybe, yeah? Well, let me ask a question, as you're creatively looking at the solution, if you could have the magic wand, what are some of the things that you want to do, if you kind of think about some of the dream scenarios, the futuristic kind of view? >> Yeah, if it was just as easy as putting it up, and making it work, that would be fantastic. But we have to work with physics, radio frequency, so it's not that easy, not yet. >> So what are you thinking about when there has to be a lot of compatibility that you're looking at in terms of the different campuses, what will work with what, how can we make it more streamlined, mesh-like, etc., is that something you're considering? >> It's a lot of planning that's involved. So not so much mesh, we don't do too much of that, but a lot of it has to go around with the requirements of the building, for one. A lot of the buildings on campus are considered historic, so we can't really place access points the way we want them to be installed. So we have to work around that challenge. And then it's getting it to the areas where people want wireless, which is also another challenge. And then budget and infrastructure. Then people start throwing devices and then that we don't even know about, so they'll want IOT everything, whatever you can put wireless on, they want that. >> How are you mapping for security purposes? What are you doing for that? I mean that has to be something that you're looking at. >> We definitely have a network that's secure, which uses certificate-based authentication. We have our regular Stanford network, but we really secure the infrastructure side and allow students, staff, teachers to really try to innovate around that. So we don't put a lot of restrictions on the network. We do protect anything coming from the outside coming in, but going out to the internet, if they want to develop something, there's a lot of great stuff that comes out of Stanford, and we don't want to inhibit any of that process. >> As a Cisco kind of champion, you guys can look at Cisco, and honestly, certainly the network enables a lot. What are you learning here, what do you hope to walk out of here with, what sessions have you played around with, what did you gravitate to? >> I gravitated toward some of the beginner sessions, which would have to be with how to program using Python. I looked at some location-based stuff. Maybe there's location-based services that we want to roll out to the campus. That's a big topic amongst the industry right now. And then efficiency as well, how can I deploy faster if it's just me working on a certain project? Those kind of things, and even reporting, how can I get statistics, how do I know how many devices are on a section of the campus or an AP? Those kind of things, something that will be easier for me and maybe my co-workers as well to get the information we need and then be able to deliver the services and the infrastructure faster as well. >> How's the tooling for you guys over there? Obviously with DevNet Create you can almost see the dots connecting. Apps could be developed, either custom apps, and they're different, you can't really have an off-the-shelf app. You could have general purpose EDU apps for maybe networks, but you guys are a pretty unique environment there. Are there apps now that you use or are they coming? >> It's very unique. It's a big campus, so there are apps that just don't fit right out of the box, so there's a lot of custom apps. Some of the stuff I'm not part of, but I do use them and they are custom. It's very tailored to what exactly we need, what information are we trying to get, and they build tools around that. >> What the Stanford network like? Stanford as a school, top shelf, everything's great. They have a smokin' network? I mean, what's the bandwidth, give us some numbers! What's the upstream? We know from a live-streaming standpoint, we've been there- >> We have a good upstream, I'll tell you that. And there's multiple, for redundancy, so at least 10 gigs for some parts of the campus. And we do get a lot of devices on wireless. I think the last number I've seen was around 40,000 unique devices on wireless. So it's getting larger. >> Rogue devices, I mean obviously, we were talking before Cameron, just joking, there's a lot of power there, a lot of network, I can see kids bitcoin mining in their dorm rooms. I mean it's what I would probably try to do. >> I don't monitor, we don't monitor what they actually do on the network. We just deliver the pipes. >> You realize there's thousands of people rejoicing now over what you just said. (laughs) >> I'm sure there's entrepreneurs out there. >> I'm not on the security team, so maybe the security team does something, but as far as I know on the wireless, we just try to deliver connectivity. I don't want to do anything that inhibits somebody from doing a project that they're trying to do. 'Cause they always develop a lot of great applications, a lot of great products, I don't want to be that guy that says no you can't do that. >> But you got to also make sure, you don't want to restrict the creativity, because Stanford does have a lot of students who go out and start companies, Snapchat, you name it, they're all there. >> We'll see a lot of rogues, and I do go and get the bad ones, but there are some people who are trying to build a network to create a use-case around this application that they're building, and that I won't block because I know what they're doing. I tell them how you should go and approach it, so that way there's no security issues. If there's a potential security issue, I say, hey, you need to talk to the security team and get them on board. >> So you guys are lackadaisical, but you're actually encouraging them, but there's an honor system it sounds like, if they kind of come clean you guys give them some barriers to bounce around on? >> Yeah, we have the fences in place. I won't talk too much on the security side, because I'm not the guy who does the security. >> But you're not locking people down, it's not like a hard-core, chop your hands off- >> It's not like we're filtering a lot of content or anything. But if you're doing something bad, you'll be found. >> What else can you tell us about what's going on at Stanford that you think is well-positioned vis-a-vis the theme here, which is take the network, move up the stack, these things like kubernetes, this is bringing kind of a new concept. You guys are already progressive in the way you posture to the audience out there. >> A lot of the people on campus have the freedom, I would say enough freedom, to go out and try these kubernetes or maybe like Node-RED. And those are the kind of things I want to see if I could leverage those technologies as well, on our side. I think the campus is adopting the cloud, so a lot of people are moving to the cloud. I think there was some push-back there, but I think people are starting to see the full benefits of using it. >> Are there some bug bounties out there all, any incentives for students? >> Oh I don't know, maybe for the other guys. >> Rowell, thanks for coming by, I appreciate it. And good luck on your journey, appreciate it. Thanks for coming on The Cube. Okay, Stanford here, talking about network, It's hot, I've been there, I can tell you the bandwidth's strong at Standford, a great university. It's The Cube, bringing you all the action here in Silicon Valley in Mountain View, at Computer History Museum for Cisco's DevNet Create 2018. We'll be right back with more after this short break.