 Live from Las Vegas, Nevada. It's theCUBE, covering EMC World 2015. Brought to you by EMC, Brocade, and VCE. Okay, welcome back everyone. You are watching theCUBE here live in Las Vegas for EMC World 2015. This is theCUBE, our flagship program. We go out to the events and they instruct us to do the noise. I'm John Furrier with my co-host, Dave Vellante, our next guest is Blake Smith, Systems Director, Enterprise Infrastructure, Christus Health, Christ US. I said earlier, but it's a Christian-based Christus Health. Welcome to theCUBE. Thank you, thank you. So tell us what you guys are, what you're doing, and some of the tech that you're rolling out. We are a integrated delivery network. We have about 25 hospitals and about 150 clinics, as well as a health plan based in Texas. We also have international facilities in Chile and in Mexico. And we are living a life with healthcare IT. So what's it like? Healthcare is the biggest opportunity right now for companies they're saying to make money. Obviously, we hear all the time, oh, that's a big upside. But actually, from a tech perspective, there's some refresh going on, some regulations appear to. Give us the current reader's digest of top-line points of where the opportunities are for disruption and innovation. Yeah, I think the cost is being driven down by the business. There's really to the keynote this morning, they want speed and agility out of the business, but they want the infrastructure costs to drop. And that is providing all kinds of opportunities and disruptors and adding cloud services and cloud service providers, both internal and starting to explore external, which is a regulatory nightmare for us. Is there an appetite for cloud? Or is it like we have to go there eventually? There is a business appetite for the cloud because they think that it's faster and easier. There is a absolutely regulatory fear of the cloud due to the loss of control of the data. Let's talk about that. What is the biggest challenge right now? Technically, on the data stuff. And what do you see as an opportunity from a solutions standpoint to deal with that? I think one of the biggest challenges with the data is ensuring that you can encrypt it in the cloud, that you own it, and you can extract it as necessary if you have a cloud service provider. Many times the contracts are not very clear on that. So, talk about healthcare, the business. I mean, it's interesting to note you've got overseas facilities, which I think is somewhat rare, isn't it? It's rare in healthcare in the US. We're one of the few international providers. Yeah, so you not only have to deal with US regulations, but you got to figure out what's going on. And you said Chile and Mexico, right? Which probably have different set of regulations as well. In the US you got Affordable Care Act, AKA Obamacare. You got meaningful use. So you've got to show adoption of technology. It's got to be real. You want it to be real because it drives productivity. So talk about that dynamic and maybe how that differs from the folks in overseas. Interestingly, we keep our overseas data separate. That has a lot to do with the Patriot Act and some of the disclosures we might be required to perform for the government. And because it's healthcare and protected health information, we want to keep that data separate at this time. In terms of the adoption, the Affordable Care Act and just the Obamacare in general, data's adding a significant cost to our delivery. But it's also provided opportunities for us to get into big data and provide things that we haven't been able to provide to the business before. Because once you make that investment, the playing field's huge. So do you have a Chief Data Officer? We do not yet, but that position is being investigated. Okay, I mean, it's not uncommon for healthcare to be one of the leaders there. What is being discussed is a conversation. Talk about the consolidation of data centers because the Cloud House is an economic benefit, but at the same time, consolidation doesn't mean reduce costs and not invest. So talk about the, as you consolidate the data centers, what are the dynamics for actually continuing to invest in a data modeling, data backup? So we have, in the last four years, we've closed over six data centers. We've consolidated out of multiple regions into our two primary facilities and that has produced a huge capital savings, but it's also allowed us to reinvest in basically internal or private cloud technologies. So we have a very robust 95% virtual infrastructure that's fully replicated with SRM to our backup data center. So you guys like to stand there one day, tell them we're going to consolidate all these data centers and like, then what? Or was it more of you follow through saying, okay, if we do this, we have to have a hit the ground running plan. And could you take us through the mindset of what was the conversation then? Really, the plan was produced initially out of a capital refresh plan. When we looked at the cost of running the remote data centers, the cost we were providing services there, we were actually in co-low facilities, we were paying monthly rents. And from that really grew a plan to build a private cloud and consolidate to our central data centers. And was virtualization a key part of that? It's a key part of it. Four years ago, we were 30% virtual. And then how did that affect your data protection strategy? When you went into the virtualization. Totally changed our data protection strategy. Can you talk about that a little bit? Absolutely. We were traditional tape backup with 10 year old tape libraries four years ago. We were backing up hundreds and hundreds of tapes and ejecting them and setting them off to tape faults that we were maintaining in these remote data facilities. We'd have runners that would take the tapes to those locations and pop them up. That was popular, right? Here's the best part. And when a hurricane would be coming to South Texas, we'd get someone to take those tapes, put them in a car and drive ahead of the hurricane to the recovery site. That was our plan. That was your backup and recovery system. That was our backup and recovery plan. Put someone out there in harm's way and hope he makes it through. So it was not popular. All this manual work involved. What did you guys do technically? Was there automation involved? We took two steps. The first is we put a virtual infrastructure in our remote data centers. We actually added data domains in those remote data centers and began replicating that virtual infrastructure back to the San Antonio data center so that we could use that as recovery and eliminated the tape backups in the regions. No tape in the regions. We have one tape library left and that's maybe not after the 9,500 was released today. Okay, so the goal is actually to eliminate all tape, even as a last resort, no tape. And then what changed in terms of, how did it change your operations? Talk about any results. Operations changed significantly. Our availability went up. One of our primary systems, which was running like a 98% availability after it was 100% virtualized, we've been running at 99.99 or better for over two years. The largest change we did to our tier one protection strategy differences is we used to use a 30F replication between four different VMAX systems. We actually reduced that to two. We went to recovery point and now we do continuous data protection of that primary, our tier one applications. And then with the addition of site recovery manager, we just did something we've never been able to do before we brought up and are still running a complete copy of our tier one application, Meditech, which is our primary hospital information system. So we're doing ICD-10 testing or billing testing in that live environment or that copy of a live environment. We've been doing that for about two months. So that's a recovery site. That's not a recovery site. And you can test that recovery site now. So you couldn't test before. Yeah, we'd tested before but it was always kind of a circumstantial. We brought it up. We could prove that we could log into it but we never did transaction work. We're actually doing transactional work. Right. And it was too much effort. And it was VMware, the virtualization. Yes, VMware, virtual platform. And so your RPO and RTO obviously were affected. Oh, yes. The original RPO was 12 hours and the RTO was 24. The test we just performed, the RPO was two hours and the RTO was four hours. And you're taking, so you're taking snapshots how frequently? We're doing local data protection with RecoverPoint and remote data protection. So we're going to change the snapshots but at the time, metatectives support them. You guys happy with RecoverPoint? It worked for you? It's worked much better than we thought. I'll be honest with you. We were a little skeptical. It's a lot of data with high transaction. Until you really put it in production, you don't know. Skeptical eyes and fingers crossed. Was EMC like on supporting you guys? Oh, absolutely. EMC has been great. But we were kind of bleeding edge. We were one of the largest customers for metatec to ever. So everyone was happy then. It worked. It did. That's my approach. And you're doing, yeah, right. Pray some more. And you're using Vplex as well? How are you using Vplex? We have Vplex. We originally purchased Vplex because we were hoping to do active-active. Metatec hasn't released support for that. So what we've done with it is we've totally virtualized our storage infrastructure in our primary data center. And with that, we're using Vplex to split the data for SRM. And so that's all of our primary VMC drives that your standard data stores are all replicated through RetarverPoint using Vplex. So you have a lot of EMC products. We do. I think we probably have every EMC product they've made. Really? You got a VNXE? I don't, but I have four VNXs. We have six VNXs. I'm going to try to stop you. Yeah. Yeah. Guy Churchwood talked about it. Stephen Manley was on, talking about it. We had some of the product guys. So from your standpoint, you've got a lot of different products. How are they being integrated? And what do you like about what's going on? What could EMC do better? That is absolutely the hardest challenge we have facing us right now. We have all these products. They're all at a different release cycle, upgrade cycle, performance cycle. We are struggling with bringing all of those technologies to full use, you know, if you will, that full, perfect private cloud. We're very interested in the EMC cloud infrastructure offering that 28 days to bring all those tools up to a fully functional. Are you using converged infrastructure, V blocks at all? We have built our own V blocks basically. Yeah, it's okay. So you've done that in-house. Somebody, which servers are you using? UCS. So okay, so you got UCS, so essentially. We have 248 UCS blades running our VMware infrastructure. So what, looking forward, what are the big things on your agenda that you can talk about? Microsegmentation for the network NSX. Getting to that truly fully deployed cloud infrastructure with self-provisioning, Vipers on that path list. Really? In the data protection, we want to expand our testing of site recovery manager to the entire site. Most importantly, we probably are going to refresh it. Well, we know, I think we can say it, my EMC rep knows, we're going to refresh our Vmaxes with Xtreme IOs and then start replicating with Xtreme IOs technology. So no Xtreme IOs in there today? Yeah, two Xtreme IOs, or two four brick Xtreme IOs. Take a little taste of that and you'll like it. Yeah. Well, we got it for the VDI environment. Worked great. Oh, right, okay. Now we're really just pushing all our SQL server loads over there and actually pushed all of our C drives for our Meditech over there to reduce our patch window. So we can now reboot a thousand Meditech servers in 15 minutes. The application is down for 15 minutes a month for patching. Great. Well Blake, great to have you on the queue. I want to ask you one final question. What's the big learning that we're magnified out of this process? Obviously, huge task. I mean, your old process was, a lot of people go through that. There's a lot of manual work and certainly sending runners in front of hurricanes and putting them in harm's way is a great, highlights the critical nature of how this is part of your operations. What did you learn the most out of us? What taught you about this exercise, this transformation? You did a lot of work. You consolidated, you virtualized, you changed your workflow and now you're kicking butt and taking names. So what's the big magnification? I think we learned, because we had some struggles along the way. We've learned you've got to have a strong partner. You're, in this case, EMC and VM work, have to be strong partners at the table with you. And also, and I'm not the person, but I wish my team was here. You've got to hire champions and you've got to put champions in the roles to do their job the best they can. You mean internally? Internally. Internally to your group. And advice for folks out there looking at the same road that you had. What would you, what scar tissue can you share? And what would, you know, what would you do? Frey and hire champions. Okay Blake, thanks for coming. Blake Smith here inside theCUBE sharing about his journey with RecoverPoint, all the transmission from tape into automation, getting cloud ready, virtualization, state of the art in an important area. Of course, it's theCUBE. We'll be right back after this short break.