 Okay, we're back, this is Dave Vellante, and I'm with Stu Miniman. We're live at theCUBE, SiliconANGLE.tv's coverage of Dell Storage Form. This is day two for us. This is the second year we've done the Dell Storage Form. Last year it was down in Orlando. It was really the first consolidated Dell Storage Form, bringing together equal logic and compelling customers as well as the Dell customer base. Year two, Stu, you know, a lot more polished I think. Last year was great, but this year is really get the feeling that Dell is serious about storage. Not that they weren't last year, but they're really putting forth, I think, a good show for customers. And speaking of customers, we're here with another customer segment. Chris Hansen is a system manager, sorry, at Gordon College up in Wenham Mass on the North Shore. And you're going to go to Fenway with us tonight? Yes, sounds like a great event tonight. Try to hit some homers? Yep, absolutely. You're not from around here, are you? I'm from Minnesota originally, actually, so... I'll be sharpening up my Boston accent as you can see. There you go, yeah. Chris, thank you for coming on theCUBE and welcome. Yeah, thank you. So were you at Orlando last year? I was not at Orlando, no. Okay, well it's pretty convenient for you here. Oh yeah, we're right up here, so it's great to be able to participate in the event. How long have you been a Dell customer? I've been with Gordon College for just about four years now. And Gordon has been using Compellent for about five or six years. Okay, so you were one of the original Compellent customers. They jumped on board pretty quickly, yeah. Yeah, cool. So tell me a little bit about your role there at Gordon. I'm the system manager, so I manage the... We have three data centers on campus. I manage the stores, the replications, backups and recovery, as well as doing a little bit of the application level with managing Active Directory, SQL Exchange, some of the backend pieces to those applications. And we're leveraging Compellent and all three of our data centers. We have cross-site replications set up between those, which has been very effective for us. So you do a little bit of everything. A little bit, yeah. Do you consider yourself an IT generalist or no? Yeah, well we have a few different IT departments at Gordon and we're in the network systems department. So yeah, we're in the broad range of doing all of the backend infrastructure. Do you have a storage admin team or no? I am the storage admin team. Okay, as part of the networking group. Yes, yes. I mean it's somewhat rare in IT, isn't it, that the storage in the... I mean, I've seen certainly network people that look after everything, but you've got a storage admin team inside of the networking group. Right, and absolutely. And just like we saw last night at the keynote, where they're kind of converging, they had a storage admin, their IT admin, their server admin and their network admin bringing those together. And that's really kind of what we're doing at Gordon, bringing those together. And it's really helpful for us to communicate on those levels. Especially when we're introducing technologies like iSCSI, we've got to be really tightly close-knit with networking switches to make sure we can utilize those correctly. So Chris, tell our audience a little bit more about Gordon College. What's the claim to fame and the specialty there? We're a private Christian college. We offer a four-year undergrad degree, and we also have some grad programs as well. And we're in Wenham, Massachusetts. It's a really nice area. Oh, it's beautiful out there. Oh yeah, it's a great, great location. Really nice. We have a couple of little ponds that are exclusive there for our students. We have a 12-to-1 student ratio, so we really try to emphasize small classrooms and giving those tools to the students to be able to get the most out of their education while they're there. And tell us about the IT environment there. So our IT department, we have, like I said, we have three data centers. Our primary data center is in one of our newer buildings that we have recently all physical servers that we just converted over to all being virtual machines running on VMware. But we have three data centers we're replicating across. Most of our applications, all of our tier one applications are in our primary data center. Everything gets replicated to a secondary site, and we're also running a VMware cluster on that secondary site. And then we have a third location that's still on campus, and that's sort of our DR spot, and that's where we're getting a third tier of redundancy, and we're implementing Commvault Simpana for that backup location. So you have three sites all sort of within synchronous distance? Yes. They all have a fiber connection on site. And so the data is all stored synchronously at these, or do you async to the third site? It's asynchronous replication. And synchronous to the second site? We're not doing async, we're doing all asynchronous replication. Are you doing any synchronous stuff locally, or snapshots? We're doing snapshots. So everything, yeah, we're doing 15-minute snapshots locally in our primary data center, and then that's an asynchronous replication to our secondary site. Okay. And are you utilizing the tiering functionality in the propellant? Oh, absolutely, yeah. So we have all of our tier one applications like SQL, Exchange, Active Directory, we want to have those having the highest capacity possible. So yeah, obviously we have those in a tier one structure. And then with the replication typology, we have that replicating down to tier three as those blocks are available to be written down. So I just want to paint a deeper picture if I can. How many physical servers? Physical servers, we were at around 100 physical servers, and then last summer I migrated all those to VMware, so our primary data center only has eight physical hosts. Okay, and then how many virtual machines are you running? About 100 virtual machines, so everything was a complete conversion. Okay, and then the storage is predominantly compelling, all compelling? Yes, all three sites are compelling. How much storage? Primary sites about 25 terabytes. Our secondary site is 15 terabytes, and then our third DL location is 10 terabytes. And you've taken 15 minute snapshots using compelling software, right? Yeah, the compelling data insert replay, not across the board. You know, our tier one applications, we're doing shorter snapshots, and then some other applications we can leverage longer distance snapshots for those. So you basically are offering the snapshots as a service that varies by application? Yes. And then, I want to come back to that actually, and then, okay, then you async it off to the second site, and then how does it get to the third site? So the third site, like I said, we're leveraging Compel & Simpana, and that's actually doing an application. Compel & Simpana. I'm sorry, Compel & Simpana, and that's doing that. Unless Dell made another acquisition that I did. They just announced recently, you heard it here first, no, our Compel & Simpana is our third level of backup, and that's doing an application level backup. So things like SQL where it's actually pausing IO on those SQL databases, and then grabbing that data, it's storing that in our third site. So there's no Commvault on the main site? Commvault is, it's running from that main site, copying it to our third site. Great, okay, so it goes from, so Commvault interacts with the first, the main, the production site, and the third site. Yes. Not the second site. Correct, the second site we're leveraging all your repellent replication. Got it, okay, and then how do you determine the RPO for each app? Is that a discussion that you sit down and have with your constituencies and the quote-unquote line of business? Right, right, so that's actually, every application's going to have some slightly different needs, and largely what we have to do is we're saying what's the core goal of our institution, and what are we trying to accomplish here, what's the most beneficial for us to put our money in. So things like SQL where obviously everything is referencing a SQL database at one point or another, most of our applications are hitting a SQL database. Things like your public website, obviously we need to have that be one of our most critical, have the most uptime on that machine, which is also leveraging a SQL database to provide all of this content. So we're kind of making a couple decisions of what are the most important pieces here that are the most visible, that would hurt us the most if they had an outage, and kind of going from there. Obviously Exchange, we just recently migrated to a hosted Exchange environment, so we're having a little bit less pressure on the Exchange servers, but obviously a local Exchange environment, that was a very high, highly noticeable event if there's anything that needed to pause operations on those Exchange servers. How's the hosted Exchange working for you? It's been excellent. We migrated about two years ago to a hosted Exchange solution, and it's been a really nice solution. So the college doesn't want to be in the business of managing email? We're managing it, we're not hosting it ourselves. Yeah, hosting email. Chris, I'm wondering from the student population, how does IT get involved with the storage of working and things like mobile technology kind of exploding? What's the impact been on your environment? Oh, yeah, absolutely. So with being in a college, students, they want to, we're doing bring your own device, obviously every student wants to bring every device they can think of. Our joke is kind of your average student has a cell phone, a laptop, a PDA, a tablet, their Xbox, their Wii, their PlayStation, everything, and they need it all on the network. Oftentimes they want it on the wireless network. Sometimes they can use a wired connection. So obviously that's hitting us with an impact, not just for storage resources, but network resources as well, getting all those devices connected. So we're doing a few security policies to make sure that all those devices are in compliance with our network before they can join, and that we have a policy key that kind of does a quick scan, make sure there's never a virus on their machine, make, you know, confirms that running the appropriate updates for the operating system, that sort of thing. And I'm just making sure, yes, we want them to all have their devices here. We want them to enjoy their devices while they're on campus, but we want it to be a safe place for all of them at the same time. We don't want viruses spreading from one device to the other just because of something that we could have prevented in the first place. And to clarify, the backup is all snapshot-based, disk-based backup? Is that right, or do you have... We're doing an application-level backup with Commvault and then we're also going to use it as just going to be the snapshot for applications. Okay, and the target for the Commvault is what? That's just an ice-gozzy storage solution, it's made by Drobo, and that's a separate solution that we're using. Okay, so you just pick a... No offense, Drobo, but pick a dumb target, use Commvault. Yeah, it doesn't need to be anything fancy, and then Commvault kind of works its magic. It can do de-deplication to that target. And not doing any tape? I've had a lot of discussion around Apashore. Is that something that you're looking at? Yeah, this was the first that I've actually seen anything in depth about that, but absolutely. So why would that interest you? If you're already doing snapshots and you've already got sort of the situation covered, what... I don't know if you've looked at it closely enough, but I'm interested from a practitioner's point of view, what more value would that bring than what you're doing today? Well, I think just kind of seeing the convergence of seeing if it can be provided from our storage array, and if it's coming from that aspect, having that data awareness can really be beneficial. Right now, we're starting to see a little bit of that convergence with Commvault and Dell, but seeing that come right from Dell as a source, I think, is really beneficial. And in that instance, you'd maintain Commvault for the DR piece? Well, it depends on... I mean, I know it's hypothetical. Right. It would depend on what's involved at the time, what the offering is. But I imagine we'd still use Commvault as that last tier of archival since that can offer the de-deplication and a final backup point. So it's the integration that appeals to you? Absolutely. Is that right? If it were just a point product, you probably wouldn't be as interested. Right. Well, absolutely. That's the thing. Seeing that it's integrated, like they were talking about last night, having an end-to-end solution, and just seeing that all of this product can be combined for the central purpose of providing that solid structure. What's the value of that to you specifically? You know, from an implementation standpoint, that's one of the highest rankings. It's just how well... When you're going to implement something for the first time, how well is it going to share and react with the rest of the resources we have? So that's huge. Being able to know that this is already tested, it's tried and true, it's working together with these other components that we already have. That's a huge piece when you're looking at implementing a new product like that. So, Chris, you've lived through the transition of Compellent being acquired by Dell. Yes. And understand you have really good experience with the co-pilot support that Compellent has. Can you give us a little bit of flavor as to what your experience has been kind of before, during, after the acquisition? Yeah, absolutely. So yeah, co-pilot has been one of the best resources. One of the reasons that we're just rock solid with Compellent, why they're such a great resource for us. Co-pilot has been anytime anything goes wrong. They're on the phone. They're saying, hey, did you know this just happened in your environment? If we spot something we call and say, hey, what can we do to get this worked out? An interesting story. My very first night as a system admin at Gordon, our primary sand ran out of space, went into what they call conservation mode. This is my very first night, 2am after I had just done one complete day as a system admin. And obviously I'm like, okay, what the heck do we need to do? How are we going to work this out? The guy that called me, his name was Bill, called me at 2am, just said, do you know this is what's going on? And I said, this is my scenario. I'm not familiar with Compellent, what I need to do to make this change, but I'd love to work with you. He stayed on the phone with me and he said, we'll work through this together. We're going to get there end to end until we had everything worked out. It was a matter of just working out a couple of replications that we had targeted more than they needed to be. But it would have been really easy for Bill to just call up, do his 24-7 monitoring, say, Chris, did you know there's a problem? You better get that worked out. Have a good night. You could have emailed you. Yeah, I could have gotten an alert. I could have gone to the next day once there was a catastrophic failure. You know, there were so many possibilities of how he could have handled that. But for him to call up and say, I'm not hanging up on the phone until we get this worked out, you know, that's a solid investment, being able to leverage Copilot in that way. They really just stuck with it until this solution had been achieved. So yeah, that's where we're really, I mean, just thrilled about the support that Copilot can offer. Has there been any change since the acquisition of Dell? So the change, yeah. They're obviously hiring a lot of new people. We're seeing a little bit of that where we're getting, you know, different tiered structures based on who you're talking to. One of the nice things, though, is with that acquisition, you're also seeing that Copilot is offering, you know, that the Dell support as a whole is offering more, providing support for more resources that we have. So for example, our QLogic HBAs can also be supported are, you know, we're using brocade fiber switches in our environment, seeing that those support structures are also involved with Dell now has been very helpful. Seeing those come together a little bit. Whereas before, you know, if you have an HBA fail, you need to call QLogic or Emulix. You need to get a hold of that provider. If you have a server fail, you need to call that company. You need to get in contact with each of those. Seeing them converge is really helpful. Having one central location for that support has been very helpful. What are the top items on Dell's to-do list? You know, Michael, let's say you're sitting down with Michael Dell. Chris, just talk to me. I'll get it done. One thing that I love, they touched on this a little bit last night was doing some caching at the server level. They were talking about eliminating that bottleneck between the storage array and the server itself. That's where you need to see an end-to-end solution. You can't just have any server provider with any storage array provide something like that. So having Dell, they touched on that a little bit of actually eliminating that bottleneck between your storage array and your server on a solid-state disk as close as you can to that processor, to that ESXi host in our case. Is that right? Project Hermes, I believe. That's what the Greek god of IT is. So yeah, that was really nice to see them touch on that. Only if you're providing an end-to-end solution like that, if they're behind the storage and behind the server itself, could they provide something like that. So that's exciting to see that. That would be a really nice piece and one of the things on the forefront of our agenda is how can we get that performance to a level that it needs to be. And especially in a tuning structure, how can you get those certain SQL applications to have that really fast read-write capabilities that they need. Chris, my last question. Lots going on here. You got cloud, you got big data, you got all these new terms coming out. Big changes going on in technology. For a lot of people, it's a complicated situation. Trying to get done, trying to deal with data growth, trying to figure out how to protect data, secure it, deal with cloud. How would you summarize what you would advise? What I would say is really embrace some of the changes that are occurring. There's several approaches to how you can implement a cloud structure, implementing a virtual infrastructure. But the resources, especially when you can look at a company like Dell and get support from all of these angles, are really leveraging those using our vendor Winslow Technology Group implementing with them has been an excellent resource as well. But yeah, taking a look at that and just seeing how you can leverage some of these great resources that are coming up, VMware has been a great solution that we've been able to implement and really cut down some costs and implementation times on those. So that's been excellent. We're excited to see where things are leading and we're glad to be partnering to be continuing on this journey with Dell. Excellent. Chris Hansen from Gordon College, thank you very much with your journey. Thank you very much. I appreciate your time. All right, keep it right there. This is Silicon Angles theCUBE. We'll be right back from Boston right after this.