 Live from Las Vegas, it's theCUBE. Covering Discover 2016 Las Vegas. Brought to you by Hewlett Packard Enterprise. Now, here's your host, Dave Vellante. Hi everybody, we're back. This is theCUBE. We're live here, day three of HPE Discover 2016. It's my pleasure to welcome back Patrick Osborne. Patrick, you know, longtime, Cube, guest, Cube alum, saxophone player. We can't get enough. Let's talk about that a little bit. He's joined by Michael Lindsey, who's a system architect at RNDC out of Atlanta. Welcome to theCUBE, great to have you on. Yeah, you like Lenny Pickett? Love Lenny Pickett. He's like the original, right? Yeah, of course, Tower of Power fan. He's on Saturday night. He's the saxophone player on Saturday Night Live. Big inspiration, he's good. Awesome. Okay, well, let's start with Michael, because Patrick, we talk all the time, but so. So Michael, tell us a little bit about RNDC. What do you guys do? You're liquor distributor, but tell us more about your business. RNDC's second largest liquor distributor in the country. We're across 20, 22 states. I can't remember the exact number. We've grown over mergers and acquisitions. Started out as an East Coast primary organization and kind of spread out from there. We run the gamut of HP products. We started out as an old deck customer, went to Compact, followed Compact over to HP, and have been going strong. So they still sell billy beer? You guys are probably too young to remember what that is. I know what that is. Okay. Chibi Carter's brother used to have a beer for those young people in the audience that don't know what I'm talking about. So your role as a systems architect is basically to architect the entire infrastructure, or are you mostly focused on storage? We can talk about that a little bit. I architect the entire infrastructure, anything aside from networks. And wireless communications. So storage, servers, virtualized systems, hyper-converged systems, backup and recovery, BCDR, remote office, that's all under me. So paint a picture of what your operation looks like. Infrastructure. So we have a primary data center in Atlanta, Georgia. We have a secondary data center for disaster recovery in Houston, Texas. And then we have remote offices all over the place, sales offices as well as warehousing offices that run local systems, file services, domain services, some SQL services for the warehouse systems there. And the apps that you're running, maybe talk about those a little bit. Are they custom? Are they off the shelf packaged? You run Oracle database? A lot of stuff that we have is custom. One of our primary order to cash systems that we have currently is on OpenVMS, written in die ball, which is a very old language. It's a cold ball derivative. We actually own the source code for that product because the company has been out of business for so long. So we have quite a few developers that have learned that language. And right now we're bringing in younger people and teaching them that die ball code to continue to program on that. So people might think, wow, why are you doing that? That might be crazy because it's adding value to your business and to convert off of that would be insane. We are actually getting ready to ramp up conversion to get off of that system as well as get off three other order to cash systems that we have. We just recently signed a contract with Oracle and have brought in Superdome X into the environment. So we're going to be running everything off of Oracle EBS and SOA suite for an integration bus. And real time processing. So we have about a five to seven year project to do that. Okay, that could be another 20 minute conversation. So we'll just stay away from that. And the storage infrastructure that's supporting these apps is consistent or is it a bunch of stovepipes? How does it? Not to be pejorative, but that's how it works. For me it's consistent. It's HPE. Our tier one storage is XP P9500. Okay. Our tier two storage is three par 7400. So we just got into three par two years ago. We came from an EMA system years back when we were running Alpha processors. Moved to an EVA, outgrew the EVA and decided we needed something that was a little bit more stable and rock solid for replication and went to the P9500. Okay. So you hear a lot of talk about software defined. Maybe Patrick you could talk a little bit about that. You guys are, you know your big messaging this week is all flash data center and then the whole fabric, you know software defined piece. Give us the update on what you guys are talking about at the show, what the reaction has been with customers. So we have a number of customers like Michael who you have your core IT, right? And you're using technologies like all flash to take cost out, accelerate, you know the performance of those applications regardless of their sort of legacy apps or if you're moving to Oracle but that's a core IT that we're looking to accelerate with three par and all flash. For a lot of customers like R and DC they've grown through acquisition. You've got a ton of IT operations that are outside of your core data center. So we're really trying to help customers take cost out in those areas and also drive simplicity. So Michael started a journey with using our store virtual technology, right? So bring in store virtual which is a software defined storage layer putting that on servers for the purposes of putting a number of workloads on there that I wouldn't say they're tier one workloads but they're pretty important, right? So file services, print, some SQL, right? Taking that and it's very simple. It's, you know, you've got servers co-locating storage with your compute and then to further that journey since Michael doesn't have a ton of IT staff to manage this and build these things on their own moving that model. It's the same architecture but moving to an appliance model, right? So the hyperconverged appliances that we have. So HC250, HC380. So you get that same experience and the same architecture but in something that's very easy to deploy less than 15 minutes, right? You can scale that out. Very intuitive user interface. And then you don't have to really deal with it at a local office of patching servers, updating disk firmware, dealing with your SAN, right? It becomes all collapsed, very simple infrastructure. So that's kind of what we're doing right now with software defined. So Michael, you mentioned you were on a panel earlier this week. Panel is next right after this. Oh, okay. Patrick and I spoke about hyperconverged yesterday. Oh, okay. So what was that conversation like? Share with us what you guys discussed. Any questions you might have had? Kind of our journey, where we started five, six years ago. We needed to replace all of the physical servers in our remote locations. We don't have a lot of expertise in every remote location. We have some tier, some first level support in those locations. So we needed something that was a little bit more reliable than a physical server that could die or hard drive die and somebody has to go replace it. So we look for a high availability solution. One of our partners said, hey, have you ever looked at VSA? Said no, and I don't know anything about it. He's like, well, let me show you. So he said, take two physical servers, put this VSA software on it, and you've got a cluster that you can fail over, you can upgrade without any downtime. He said, okay, we like that. So let's start deploying that. So we started deploying that with Gen6, DL380s, kind of grew those, increased memory on those, distrives on those as the use case built out, and continued those through Gen9s. So we even deployed some brand new systems last year that were dual Gen9 DL380s. And all the time we're looking for the next best thing, that gives us more consolidated systems because we don't have a lot of room. Each one of these locations does not have data centers. They have a closet that they have a two-post rack, and that's really all the space. Some of them have a table. So we're putting servers on tables. So we needed something that was a little more compact, but also had the high availability resiliency. And one of our storage guys from HPE came to us and said, have you heard about this hyperconverged thing? No, tell me about it. He said, well, we've got this HC242 system. And we looked at that and thought, it's not quite a fit for us, because they didn't have a lot of the, I guess what you would call special sauce built in yet. And he said, well, hold off for a little bit. We've got this new thing in the works, HC250. So when that came out, they started talking to us about it. And we said, we like this. This is great. It gives us everything that we need in one package. We've got one software package to deal with. We've got one firmware package to update everything. It's all inclusive. Our guys in the remote locations don't really need to know anything about it. We can deploy it. It's invisible to them. It's invisible, right. And things just run. When you say it didn't have the secret sauce, you've been talking about the maturity of the stack, or is that- The maturity of the software. The one view instant on piece and the management capabilities. The 242, you couldn't suck into your own vCenter. And you couldn't manage in one location. You had to manage each individually. So we waited for centralized management. You didn't have the resources out there to do that. We have two people. One of those is me and one is one other guy that works with me. The administrative assistant managing the servers. How do you back all this stuff up? We back everything up with store once and HP Data Protector, or excuse me, HPE Data Protector now. The documentation still says HP, probably. Right. We haven't updated that yet. Okay, so- We started that journey. We had Barracuda appliances. And we've had those for five, six years. And things just didn't quite fit for us. Their trickle out to the cloud did not work well for us. We never could figure out why. So we wanted- What was the issue? Just the performance or the backup window? It was a performance. A lot of those appliances weren't right sized for the locations. So we ended up filling those up and having backup retention policies of two weeks or some of five days. So we wanted to move to something that we used in the data center, which was HP Data Protector. So we used that in our data center for all of our systems aside from OpenVMS and iSeries that we have in there. So we wanted one stop shop for the entire organization so we could manage it holistically. So you had mentioned earlier that you're not responsible for the networking piece. But when you talk about, when you hear a lot about the hyper-converged, converged infrastructure and storage networking and compute coming together, do you see a day when that networking piece is all under one roof, as well as the compute in the storage or small shops? We are investigating software defined everything. I would say we're in our infancy on that, but our networking team is really on board with that. They see the value in it and they're fine with it. We have virtual connect now, so my team still manages some of the networking for blades and things of that nature. What's the appeal of that software defined vision and what are your concerns about it? My concern is the size of our staff and all the back end work to get that up and running. The appeal to it is, I think it will greatly simplify our disaster recovery and our ability to transition to an agile deployment for systems. Right, okay, so it's the learning curve of getting there. And then what about Flash? Are you guys investing in Flash at all? It's an interesting story. Three years ago, my director purchased a pure storage device and a month ago I pulled it out of our infrastructure. Really what? I pulled it out because of the price. The price point for the performance that we needed for the application just, it didn't match. But it was overkill. It was overkill. But you acquired the system, so you're gonna use it somewhere else? No, we're not. Our support runs out early 2017, so we're gonna go ahead and retire it. It's already depreciated off of our books. Okay. But we are gonna introduce some SSD Flash into our three-par array and take over that piece. Okay, and see how that goes and then evolve. Patrick, what else are you hearing at the show from customers? Day three now, so you must have a three. Good read on things. You know, the HP split going well, the spin merge, the nice tailwind, the product portfolios, Crankin, Antonio had it with you last night, had a spring and a step. Things feel pretty good. Yeah, things feel great. I would say that we have a number of customers that are in transition to, Michael has your traditional IT apps, that you're always gonna have to keep up and running, you have to cost optimize them, and we're gonna help them do that with things like upgrading his three-part, all flash and things like that. But the vision around automation and making things simpler for customers is that's becoming a reality here. So everyone, software defined storage is in slideware right now. We have people that are doing it at small scale, they're doing it in remote locations, they're doing it at remote offices, smaller data centers. We have people moving to hyperconverged, so that's at node scale, and they are absolutely doing that in the core data center and then being able to take that same experience in the same software stack at rack scale on composable and synergy is, that is for me that's super exciting because not only do you solve a bunch of problems around simplicity and cost, so we have this model of disaggregating servers from storage appropriate for a lot of workloads today, like obviously that's where you're gonna run your Oracle on, a lot of other workloads that you can deploy on servers with this stack and co-locate compute and storage together at a different price point, but be able to automate that is, I don't see anybody else that's gonna be able to pull that off in the industry. We have the portfolio to do it and that's what excites me about the portfolio today. You hear a lot about digital transformation and every conference you go to, that meme is, they hit you with that, is, what does that mean to a business like yours? So I would say for the last couple of years when you've heard that, for me it hasn't really resonated because we don't have a really agile development cycle. We stand up an application and it runs for years. I've got some application teams that are asking for that, but we are feeding out a lot more information to mobile apps and those things change quite a bit. So the digital transformation for us is trying to become more proactive to deliver that content to our employees, our customers, our partners, and that's what it's gonna mean to us. Service excellence. Yeah. Yeah, great. Excellent, we gotta leave it there. Gentlemen, thanks so much for coming to theCUBE. Quick last words, your thoughts on 2016 Discover? 2016 Discover has been incredible. I think for me, the market and most importantly, our customers have voted with their confidence in their business and their dollars with HPE, especially on the storage side. So for me, that's been just a great realization of a strategy that we put into place a number of years ago. So for me, it's just every presentation we give, every panel, every interview, we bring a customer along with us to tell the story, not coming from me, coming from folks like Michael, so that's really important. Charlie Parker or LP? Well, I'm a Coltrane guy. Yeah. And Michael, your thoughts on the event here, your things you've learned, things that have excited you? I think it's been great. It's really good to see where HPE is taking their business and steering customers and they seem to be listening to what the needs are for their customers and being able to provide solutions for those. Great. All right, we're here live at theCUBE. Thanks very much gentlemen for coming on. You're at HPE Discover 2016. We'll be right back right after this short break.