 from Orlando, Florida. It's theCUBE, covering Cisco Live 2018. Brought to you by Cisco, NetApp, and theCUBE's ecosystem partnership. Welcome back, I'm Stu Miniman, and this is theCUBE's exclusive coverage of Cisco Live 2018 in Orlando, Florida. Happy to welcome to the program first-time guest, David Arnett, who's a technical marketing engineer with NetApp. He's living in the heart of the show, which of course is the knock. Thanks so much for joining us. Thanks Stu, it's good to be here. All right, so Dave, first, before we get into it, give us a little bit about your background. Where are you based and what your role is? Well, I'm based in Research Triangle Park, which is NetApp's east coast headquarters, basically right across the street from Cisco's east coast headquarters. So we all know the RTP, those of us that work in tech. Right, so my role, I'm a technical marketing engineer on the FlexPod Solutions team, so I build systems in the lab and then validate them and then document them and publish them for our customers to use to build similar systems for themselves. Okay, we go with all the TLA's, you're the TME from RTP, working in the NOC. That's correct. All right, so you are not responsible for the network here. So we're not going to ask you why the network went down for a few minutes and people are troubleshooting everything right there. Thanks, I appreciate that. Why don't you explain, you know, NetApp and your role at the knock? Well, NetApp is a big sponsor of this conference and one of the ways we do that is that we supply storage systems for the network operations data center. There's a lot of systems that have to run to kind of, in addition to the actual network in order to make everything run and those systems require data center resources. So we bring servers and storage to run all of the services necessary. Okay, and luckily I'm sure everything went really smoothly, there's never any challenges and that's why it's just this great glass thing that we look at as we walk by. It just magically appears like that, right? All right, so give us a little bit of insight. You've done this at a few shows. What kind of things do you run in? What are the stresses and strains and how does the architecture hold up? Well, you know, we started Wednesday of last week putting the knock together and of course the room here was completely empty. There's nothing here. The rigging is laying all over the floor and we end up with delays just because things are in the way we can't put our equipment where it needs to be, then the power gets pulled in, we had the racks in place and then waited an hour or two for power to come in and then another couple hours for the network drop to arrive before we could get connected to the outside world. So it's always kind of a challenging, there's a lot of moving parts in order to get this thing off the ground. It reminds me so much if you talk about customer environments. All right, how does day zero go? Well, things like we interviewed one of your colleagues talking about FlexPod. Converged and hyper-converged infrastructures helped simplify that initial rollout and then it should also help once you're up and running. So once you've got the knock up and running, what's your team working on? Are there knobs you've got to adjust? Are there outside stresses that need to concern you? I'm happy to say that the storage and the data center infrastructure is one of the most reliable parts of the knock, right? Once we get it up and running, it's more about monitoring and management than anything else. I'm personally monitoring to make sure that we're not running out of storage capacity to make sure that everything is still online and make sure that basically everything is running as smoothly as it can be. Okay, are you doing any analytics on this? Do you have like hero numbers that come out after the show? We do, we participate in the session on Thursday, the knock round table session and so we're collecting all of the numbers, how much capacity we're actually using, kind of what the performance envelope of the system is and so on. It's interesting, when we talk about customers and the deployment, one of the biggest challenges is, okay, I'm going to deploy this, how long am I going to have it and when am I going to run out of storage? When am I going to need to grow? It's kind of a unique beast when we're here to show like this, because you've got some ideas, but what if something's really popular or stresses and strain challenge? How do you plan for that and is anything ever come up that has to worry about? Well, we've never actually hit the wall yet. We try to be very careful. We've actually provisioned a considerable amount more capacity than we need and the system that we've deployed is an all flash fast so it's got performance to spare. So we really try and avoid any of those problems up front. We have seen in the past issues where the cameras and recording and such generated more data than we expected, but it was not more than we could handle. We had planned ahead and made sure we had plenty of extra capacity. Oh, trust me, this is our ninth year doing theCUBE and some of the early days we were scrambling. Luckily, we actually worked with a lot of storage companies so sometimes there's spare drives or things that we can grab because yeah, it's more and more data. The devices get larger megapixels higher resolution. It's challenging to deal with things like video. Yeah, it is pretty challenging and of course every time we do this there are more and more cameras, not just the data they produce but the actual items that are producing it. So things really can grow very quickly. Yeah, if there are any interested, are they playing with like IoT here that brings data back? Not so much. I mean, the video recording is more just for general monitoring and security purposes. There's not a whole lot of kind of deeper level analysis going on like on the video or anything like that. As far as the rest of the systems, everything is constantly being monitored and for pretty much every system that we run at the end of the week we'll kind of produce some metrics around what we saw and where the challenges were. Okay, you said at the end of the show that there's a round table to talk about it. What are the users looking for? What kind of things do they learn going to a session like that? Well, they're really interested in how it's possible to bring such a huge scale network into existence especially in such a short period of time, right? So we talk a lot about the deployment and how the wireless access points get spread all over and how all the applications come up and come online and then take advantage of the networks that we put in place. They're really interested in the deployment details, right? Because they're doing the same things in their own shops and they're looking for guidance on how the experts do it. Yeah, it's interesting. I remember I read a book once. They're like, you can't teach a kid to ride a bike at a conference but there are some interesting lessons that we learn going through some of these deployments. Any interesting points over your time working with Anak? Well, you know, we build this system to be really highly reliable. That's key for the operation of the show, right? We just can't take any chances that it goes down. We've had some incidents where wrong circuit breaker gets switched and so we're left with a failed over situation but because the system was designed to withstand those kind of failures, it's really nothing but a thing. We flip the power back on and make sure the other half comes back alive and we're back in action. All right, yeah, definitely I'm sure people's running around a little bit trying to, you know, fix this. Oh yeah, as soon as lights go dark, man, there's chickens in the hen house, you know? All right, and it's something people can walk by and is there tours of it? Is it, you know, big plexiglass thing? Yeah, the knock is actually, all of our data center equipment is out on the floor for everybody to see. Rotala supplied an enclosure in the cooling equipment so we can run it right there in the middle of the floor and still keep it at proper data center temperatures. It's there all the time for everybody to come look at but then there are tours daily. Starting at noon, I think, every hour on the hour, there's a tour related to more specific technologies like the wireless and the routing and switching and then of course the data center. Okay, and the stuff you're not using, you're doing Bitcoin mining on that now? That's correct. No, actually we do run some, we do run some, oh, I can't think of the name of the application now. It's like not SETI, but something similar, folding at home or something like that to really just kind of drive the systems a little harder and run them at an operational pace more like what a customer would see. We actually, in terms of the entire infrastructure, we are way over provisioned in terms of compute capacity and stuff. So we have some room to do that and that's always an interesting number at the end to produce how many projects we closed in a folding at home scenario. Okay, so you've got like half a week to set it up. How much time did they give you to tear the whole thing down? About 24 hours. Yeah, it goes down really quickly. It's remarkable. Yeah, it is something. If you've ever, people, if you've been involved in these events, it takes such a long time to set things up but they usually are designed to break it down, get out of town. Yeah, all the equipment is in racks, pre-racked, it arrives in the rack and we just connect up the patch panels between the racks and when the time comes to go, we just power everything off and pull the cables and roll it back into the crate. Yeah, do you do similar things at other events? We don't do similar things at other events. Cisco Live is actually the only show I know of that actually runs the operations as a centerpiece of the show. It's really a remarkable thing. Yeah, well, it is, you know, the network is obviously pretty critical here. Yeah, the attendees expect a world-class experience, right, and so our job is really to make sure that happens. All right, Dave, I want to give you the final word, you know, key takeaways you have coming to events like this? Well, it's really kind of an honor and a privilege, right? I mean, NetApp is really proud to be a part of what Cisco has going on here. We've got a lot of synergies with our FlexPod program and so it's really great to be here and be a part of this show and really specifically to work on the knock team where I can say I had a hand in making it a success. Well, Dave Arnett, really appreciate you joining, giving us some insight into some of the inner workings that help everything going on here at the show. Thanks very much. All right, we'll be back with lots more coverage here at Cisco Live Orlando 2018. I'm Stu Miniman and thanks so much for watching theCUBE.