 This is Dave Vellante, and welcome to CUBE Conversations, a series that we're doing on data protection. And today's CUBE Conversation is with Healthcare Realty, a real estate investment trust, specializing in properties associated with healthcare delivery. And it's our pleasure to have with us today Robert Dillard, who's the Vice President of Technology Services at Healthcare Realty. Robert, welcome to CUBE Conversations. Thanks for coming on. Good morning, thank you. Pleasure to be here. So let's start with Healthcare Realty. Tell us a little bit about the organization and specifically your role there. So Healthcare Realty is a publicly traded New York Stock Exchange listed real estate investment trust. As you said, we specialize in owning, developing, acquiring, managing and operating healthcare related real estate, primarily in medical office buildings. We have just north of $3 billion worth of real estate spread around the country. And the nature of our business, we not only own the real estate, we manage that real estate. So about 90 of our employees are located here in our corporate office in Nashville. We have about 160 employees spread around the country. And my job is managing the IT program management and applications and development teams that provide technology related services to our employees as well as to our tenants and to our properties nationwide. Okay, so you got about 250 employees. They're somewhat distributed. They're actually significantly distributed in nature. So that obviously puts pressure on the backup environment. So I wonder if you could talk about data protection. I mean, I've been in this business a long time. Data protection has always been a challenge for practitioners like yourself. But take us back to the beginning of your recent journey. Talk about some of the pressures that your business was putting on you and some of the challenges that you were facing, specifically in the context of data protection. Okay, yeah, for us, you know, as you said, we do have a very decentralized environment. And so a couple of the challenges that we face are how do we protect the data that the company has while at the same time giving all of our users rapid access to that data. And so one decision that was made was to centralize all of the data into our two data centers. We have one in Nashville, one in Denver, Colorado. And when that was done, we were still running on a tape-based backup platform. And so we were encountering a number of the issues that are inherent with tape. Specifically, two of the main issues are speed with which we could perform backups and ensuring that we were capturing our full data sets within backup windows. The time to recovery, both in terms of the time it takes to actually pull the data back off the tapes, as well as just the time to pull tapes back in offsite should that be required. And then finally, you know, the issue related to the integrity of the tapes. You know, as we all know, tapes can degrade over time and, you know, it may have backed up okay, but, you know, that's really with tape, no guarantee that it's going to restore correctly. So those were some issues that we had in our environment. As they say, backup is one thing. Recovery is everything. So if I understand it correctly, you had. Exactly. You had, they always had the backup window issue. And the problem is if you can't get the backup done in your window, you spill over and you either lose services or you go unprotected. So that's unacceptable. And if I understood it correctly, you also had a recovery time issue. Tape was taking too long to recover. And then of course, the third piece was the longevity. So talk about what you did to address those problems and paint a picture today of what the solution looks like. So the first step that we took to address those problems is, you know, rewind back in history and our backup system was, it backed everything up in series versus parallel. So we implemented a solution that allowed us to back up multiple systems in parallel. And then probably the biggest shift was transitioning from tape to disk. You know, so it's as we began looking at things, it was, why are we dumping to disk? Well, I'm sorry, why are we dumping to tape? Well, it was so that we could get those tapes off site. Okay, we've got two data center. Why not just dump to disk one facility and clone those backups to another facility? And so that's what we did. Originally, we started backing up to disk on our sand and then we then made the shift to backing up to a dedicated backup and recovery appliance. Okay, so you got a highly virtualized infrastructure, as I recall, and you're an EMC shop heavily. So what's the backup solution look like? So the backup solution for us today, we're running Networker 8.1. We are highly virtualized, we're 100% virtualized on VMware and both of our data centers. And so our backup target, we're backing up to data domain appliances. And for us, really the next strategic shift, so first strategic shift was going from tape to disk. For us, the next strategic shift was going from a backup admin citric environment where the backup admin controlled the backup of the systems, the recovery of systems, development of policies, what have you, which one thing that we found with that is for the backup admin to configure a system to be backed up, he has to be told that that system has been created. So the backup admin was really great at developing policies, but at times there was a breakdown in communication or a time lag between when a system was brought online and needed to be backed up and when the backup admin was actually getting that system backed up. So it's with the integration between Networker 8.1, Data Domain, our VMware environment sequel, what have you, we've really been working on transitioning from backup being centered around the admin to the creation of policies that support business rules. That's the backup admin's job. And then moving back up and restore into more of a service type environment where the individuals that own and provision the VMs and the databases, making it so that they're applying those policies as databases and systems are being provisioned to support business applications, which really for us, the first transition ensured that we were protecting the data from an availability standpoint in regards to tape versus disk. Now then we're improving our compliance with business rules around that data and ensuring that data is protected at the time it's created. Okay, so step one, addressed your backup window problems, your RTO and RPO and presumably your tape longevity problems. You're saying step two, you're really moving to a backup as a service. Before we get into that, so let me ask you, so are you also replicating data offsite? How does that work? We are, data is being replicated offsite. So it's basically the way that works is it's quite simple. You know, the backup jobs run here in Nashville at our primary facility. And then once those jobs complete, they are automatically cloned to our backup facility in Denver, Colorado. And that is validated as the process is performed. It's reported on its track from an audit standpoint. Okay, so now let's talk about the whole IT as a service. No matter where we go, we hear about cloud and IT as a service. And you're talking about data protection as a service fitting into that service catalog. There's a lot of tension oftentimes, between, for example, DBAs and the backup admins. And what you're saying, if I understand it correctly, is you're really making this backup capability a self-service offering for the DBAs, the VMware admin, whomever. And you're changing the role of the backup admin. Is that right? And can you talk about that a little bit? Yeah, no, that is exactly right. The backup admin's job is to understand what the requirements are from a business standpoint for protecting the data and configuring the policies for protecting that data and making those policies available to the SQL admins and to the VMware admins. And then ensuring that there's a base level compliance across all of our systems and across all of our data so that anytime a new database or a new VM is created, then it's protected at a base level immediately. And then the SQL admins and the VMware admins within their native environment. So SQL Server Management Studio and within vSphere, they can then, based on the business rules, because they know the applications that they're building and that they're supporting. So based on the business rules that have been defined by our business analyst, they can then apply higher level backup policies to those systems. And then in the event that those systems need to be restored, they can affect those restores themselves in a timely manner and then immediately validate those restores. So it has really reduced the amount of time from the request for a restore, need for a restore coming in and that restore being completed and that business service being brought either back online or the data being restored to a state that the users require. I love this conversation as we've been talking for a long time now at Wikibon just about backup as a service and backup as the sort of expensive insurance policy and what you're doing is driving it into an IT service catalog and getting more value out of the individuals and the infrastructure. So congratulations on taking that move. My last question is going forward, what are you working on in 2014 and beyond and specifically, what's the role of the vendor community in helping you do that I guess generally and of course you're an EMC shop, so EMC specifically. I was gonna say for us, really just like we've been talking about today, we're focusing on driving automation and compliance and really enabling the people that own the systems to do what they need to do with the systems when they need to do it. And so really what we see is that happening across the board with all of our systems. So if you're an end user and you need an application instead of putting in a service desk ticket to have an application installed on your PC, having a web portal where users can request that application, if they're based upon their role within our system tracking or human resources system, if they're automatically approved for that application and it's pushed, if not, then it goes to their manager for approval to have that application provision, but really enabling the business to drive those decisions and as soon as the business has made the decision, the system automatically enabling that user with that software. Same thing for our developers with spinning up dev environments, we don't want them to have to come to a VMware admin and say, I need an IIS server, I need a SQL server, whatever with the orchestration tools that are built into VMware. We want to automate that process for them and so that's a focus this year. So really it's all about automation, but at the same time, one of our goals for that automation is ensuring compliance with corporate policies. So that's our focus for this year. And as far as the relationship with the vendor community, and this is one reason why we choose EMC specifically is to enable that level of orchestration and automation, it requires integration. And so that's one piece that we really appreciate about our relationship with EMC, VMware, RSA, and their relationships with companies such as Microsoft is that the combined solution allows us to have the integration so that we can very simply and easily automate the processes to enable the business. Yeah, we're seeing that as a big theme here in the industry within the Wikibon practitioner base. The more you can integrate, the more value you can deliver. Robert Gillett, thanks very much for sharing your insights and coming on theCUBE, we're out of time. Love to have you back on and double click on some of this stuff and unpack it even more. So thank you very much, really appreciate it. Thank you. All right, thanks for watching everybody. This has been CUBE Conversations with Dave Vellante. We'll see you next time.