 Live from VMworld 2011, this is theCUBE, creating space for big ideas to grow, from siliconangle.com and wikibond.org, a special spotlight backup and recovery with support from EMC, cloud meets big data. Your host, John Furrier, we're special guest, Link Anlender from Lone Star College Systems. And now, here's Dave. Hi, everybody, we're back. Dave Vellante live, silicon angle, continuous coverage of VMworld 2011. I'm here with Link Anlender, who's the associate vice chancellor at Lone Star College, Link. Welcome back to theCUBE. Oh, thanks for having me again. Good to have an alum on, and I think most people here have been on theCUBE before, so we're happy to have you. So thank you. So we're going to talk about backup. This is a deep dive on backup. Really, these spotlights are designed to help practitioners like your peers understand some of the challenges that they might have to go through, maybe give them some advice. And so, let's start with a little bit of background. Tell us about Lone Star College and your role there as vice chancellor. Okay, well, the Lone Star College system is the largest community college system in the Houston area and soon to be the largest in Texas. In fact, actually, we're going to bake another milestone and probably roll past 100,000 students. So you can imagine that challenge. I mean, students are coming in from 2008 until now, we've gone from 63,000 up into the 100,000 range. So it's a challenge. From the IT aspect, we have even more of a challenge. We're facing the ability to meet the current demand of students to have our enterprise systems be able to handle that load. And we're doing that across the distributed environment. We have 1,400 square miles. We now have six campuses online. So it's a daunting task to make sure that everything is working at all times. So in the college system, I provide services from the infrastructure side, but then I also provide the operational supports that's needed at the campus level. So in the organization, we face a lot of different IT challenges. Everybody does. So we love sports analyses here at theCUBE. We've been called the ESPN of Tech. So let's break down the lineup. So tell me about your IT environment. What are you guys running? How virtualized are you? Talk a little bit about your apps. Just paint a picture for us. Okay, well, this has been the best thing. The best opportunity that you can ever have is the opportunity to rip and replace everything. Total demolition from 2008 to present. And I mean total demolition. So we had the opportunity to implement a strategy and a plan and execute on that plan. In 18 months, we implemented a whole complete centralized VM environment, a new wide area network structure that can support that environment. And we went from 5% virtualization to our current 93%. And we're going to cross 98% and pretty much hold about 98% by the end of this year. So virtualization is king for us. How do I have flexible services? How can I do things differently? So on that lineup right there, that's important. The next part is, is how do you take your ERP to the next level? And that's what we did with the virtualization strategy. We virtualize our ERP in order to meet the demand of the students. You know, students are coming in. You have 90,000 plus students come in to register and they want to check their grades. They want to check this. They want to make sure the classes are right. Where do I have to be? All in a very short window. So by leveraging private cloud, we're now allowed to make it elastic. So we can stretch everything out. We can provide the resources. The student gets the same experience right now at Peach as they received when it was slow and they were doing the preregistration. So it's enabled a lot of functionality for the business side to bring students on board. So what was the driver for this Rippemer Place? The best way to put it was, our IT systems cannot meet what they needed. The college system was growing and the IT systems were falling behind. So we had a new CIO come on board and immediately, underneath the leadership of him and our chancellor, we basically immediately changed everything. And so the opportunity was there. So the chance to take that opportunity to the max to change everything, to run IT as a business in higher education, which is a lot different. But then also to look at how do we enable innovation in the classroom? How do we provide a core level of service that instructors and students are confident with? So rather than try to bleed blood from a stone and do patchwork, he said, all right, we're just going to do it right. So where a lot of customers we talk to, they run into a problem with backup after they virtualize. So you anticipated one, is that right? And or anticipated the backup requirements of virtualization and then implemented it back up accordingly in advance, proactive. Is that right? Absolutely, fully proactive. Talk a little bit about that. That's rare. Well, we had, like I said, we had an opportunity. We were told what was going to happen and what we needed to accomplish. So as we, we resigned the infrastructure, we were able to look at the infrastructure and say, in a virtualized world, what are the backup solutions we need to implement? In the storage, how do we need to manage the storage differently? How can we leverage all these tools? And so we baked off everybody. We looked at everybody out there and said, who can give me instant recovery? Who can I trust that this backup is happening, that the data is protected, but that I can also actually meet the objectives of the time, which was five nines. How do I, how do I make sure I can ensure that quality? So... This was 08-ish? Ah, yes. Okay. Yes, 08 was the planning, planning phase for three months, centralization, reorganization. Then we immediately began redesigning everything by September, virtualization first by January, new virtualized infrastructure, new data center operations, whole change in practice by that summer. We were 80% virtualized in the first six months of that year. But you chose to go with a client-side T-duke impression. Absolutely. Correct? Can you talk a little bit about why was it the virtualization adoption or other reasons? Well, it was a combination. As we were looking at everything that was out there at the time, one of the challenges we have is our colleges are located across 1400 square miles. So we're pulling data from everywhere. We needed a centralized backup solution that was fast and agile. And by doing it at the client side, we're moving small bits of deltas across and we're able to do the exact same on the recovery side. So at the time we were doing the reviews, it really was what is the best in our environment across the large college system with all these small, I believe in small, 100 to 200 servers in a farm out there that are virtualized. How do I get that data protected? How do I bring it back to a central organization? How do I replicate it out to a disaster recovery site? And to move that kind of data, you have to have that. You have to have a D-dupe at the target. Everything's got to be D-duped. It's got to be fast and it's got to be instant. So you're pushing less data over the network. Now at the time, the integration of backup and VMware was certainly not as matured as it is today here. For the last two or three years, seen a lot of emphasis on that. I mean, you've made the decision and you're going with it. I don't know how much you pay attention to sort of what's happening outside of your little world, but I'd be interested in your opinion. Do you feel like the whole industry is just getting better at solving that problem that you were trying to solve, or does client side D-dupe still bring that advantage is even when you have like a change block tracking and some of these other integrations that you're seeing? Well, we're always keeping track of what changes are happening. We're not going to become stagnant. That's what happened before. We became stagnant. The organization was stagnant in IT and that created that failure. So we're constantly looking at different strategies for everything we do, but we're not looking at change. We're looking to validate. And as we've reviewed options, in fact, I just had one sent from somebody recently said, can we do this? I'm like, yes, we can beat that. And it was on backup. It was on recovery. Can I beat their recovery time? They promised me they could do X. I'm like, yeah, that's not a problem. So we always go back and review everything we've done. We've got to. Otherwise, you stagnate in the organization, you just say, oh, I'm comfortable with this. One of the challenges out there is, is that, yes, it's matured now. The difference is that the maturity was there, in our case with Avomar, at the time we were going down our virtualization strategy. So while the maturity was there then, they're constantly improving and moving us forward. It's not like they've stagnated in their design and letting everybody else try to catch up to them. We finally jumped into their things and as the technologies are changing, we're able to get all those pieces together at the same time. What's on the to-do list? I mean, as a customer, what are you looking for from a supplier? What do you need out of, take an EMC Avomar. What are you pushing those guys toward? You know, that is a pretty tough question to ask right now because the last time we reviewed everything, we really looked at, what is our next strategy in backup? And of course, what you hear a lot of is, how do I get to a better client side backup? How do I take those senior executives, make sure I'm backing up their laptops without them worrying about it? Right. I would say that's a different strategy for us, but at the same time as we're looking more leveraging cloud type services and virtualization of their desktop. So when we're looking at all these different strategies right now, Avomar fits perfectly with what we're doing. And the biggest thing I look for is constantly getting from them a roadmap, where they're heading next, what is the next maturity that's going to happen so I can keep my eyes on what's happening in the industry, but also know where they're going at the same time. So that partnership makes a big difference for us. But when it comes down to it is they're delivering exactly what we need. I know I've got backups, I don't have to worry about my backups. You know, tape still exists, sad to say it does, retention and all those things, but the reality is if I have to recover anybody's data, I've got it. Have you done a recovery from tape and since you've moved over to this system? No. Right. We do tapes, we have to, legal requirements, everything else that's still out there for us, but the reality is if I have a recovery of something like that, I'm pulling it right off of the grid. You're pulling it off the grid and is it typically fresh data? I mean, less than, I don't know, week or so old? Oh, absolutely. Usually when we're in a recovery situation of anything, we're usually within five to three to five days, I would say typically. Sometimes it's pretty much the same day that we're dealing with. So whatever we look at during that process, it could be recover point, it could be the snaps, but typically if we have to go to Avamar, we can pull it very quickly. So you're running recover point? Absolutely. How do you look at recover point and CDP in the context of backup? Like where do you help us understand, paint a picture of the, sort of put it to context, if you will, the sort of the traditional sort of backup that I guess Avamar is not traditional backup, but let's put it in that camp and then there's sort of new CDP type of approach with recover point. How do they fit? How should practitioners think about that? To us, a product like recover point was critical for our tier one services. Absolutely, I need immediate restoration. We measure that on a five nines environment. So recover point was absolute. We've got a huge private cloud that transitions and shifts servers back and forth and around in services. So the customer doesn't know where it comes from and they don't care. Recover point allows me to have that agility that if I have a major failure, I've got services back immediately. There's no delay, there's no hassles, nothing happens differently in the organization and the customer does not know. I have had major hardware situations happen. Hardware events you can't prevent. You can only plan to be able to handle them. At the same time is because of the strategy of deploying recover point, how we've done it with Avamar, everything else, we're able to mix it. So Avamar is ideal in the environment for everything we have and it still does cover. We use Avamar and then we tape that off to network for file retention. And in addition, you'll put recover point CDP on top of that Avamar layer. So you layer Avamar across the entire application portfolio and tape, I presume. Not tape across everything, only what I have to retain. Only tape based on the organizational edict. Yes. Okay, and then the CDP you target based on the value of the application. Absolutely. Okay. If it's a five nines application, I've got to have that. I've got to have that ability. I've got to have that instant recovery. I can't, I don't have that time. You've got six minutes. That's it. Do you do a business impact analysis? Absolutely. On a regular basis you do, even for your backup. And we report metrics out back to our board of trustees. We measure ourselves, we have strict KPIs and that's exactly how I'm evaluated. So I mean, it's important as we run it like it's a business. We are a support, instruction is the mission. It's a big difference. This notion of a backup window has held us hostage in this industry for a while. Talk a little bit about your backup window and have you been able to essentially eliminate that gun to your head? We smashed our backup window, let's see. I guess that would have been somewhere right around February of 2009, 2010. Right after you implemented this. Backup window is not something that holds you hostage anymore. We have one legacy system that we're still running because we're still transmitting, bringing data out of our old ERP and yes, we still have a little bit out of that, but that's all. Link, I got to go get in the hook here. I really appreciate you coming on theCUBE. We got a great panel coming up. Link Alander, Lone Star College. Great to have you back on. Enjoyed it. My pleasure.