 From the noise, it's theCUBE, covering VMworld 2015. Brought to you by VMware and its ecosystem sponsors. Now your host, Stu Miniman. Hi, I'm Stu Miniman with wikibon.com and welcome back to SiliconANGLE TVs, live wall-to-wall coverage from VMworld 2015. Excited to have a practitioner on the segment, Bob Dousseau, who's the Information Security Officer with Hallmark Business Connection and also we've got Greg Smith, who's the Senior Director of Product and Technical Marketing from Nutanix. Gentlemen, first time on theCUBE, thanks for joining me. Thank you. Thank you. All right, so Bob, let's start with you first. Tell us a little bit about just your background and what your role is at Hallmark. Well, my background, I've been doing this about 30 years now. I'm an infrastructure guy from way back and Hallmark brought me in to work at the B2B arm of the organization to bring in some, maybe a little bit of forward thinking in the infrastructure space. On top of that, information security is extremely important to us and so I take a critical look at that as well. All right, Greg, just for your first time on the program, we had your CEO on this morning, Giroj, you know, regular on our program. What's your role with Nutanix? I help with the product and technical marketing organization. So our responsibilities help produce some of the solutions guides, reference architectures that hopefully our customers use when deploying enterprise applications on Nutanix. All right, so Bob, in your open you said that you were brought in to help kind of do some new things, help clean up, you know, I guess, you know, infrastructure a lot of times is held back. The operation of IT and, you know, boy security is always a top of mind issue even if it's not necessarily the top funded item. So maybe we can talk a little bit about what the infrastructure looked like before and what some of the major pain points were that you were dealing with. Sure, like most infrastructures, that grow organically, they tend to become somewhat convoluted, complex and difficult to manage. They're not very scalable, they're very expensive, and the staff required to maintain them has to be very focused. At Hallmark Business Connections, are really, and it's not unusual, but we're under mandate to kind of do more with less. And so one of the things we needed to do was kind of untangle that ball of twine that had evolved over the course of about eight or nine years, the traditional three tier infrastructure, compelling HP, et cetera, and figure out a way to make that more scalable, make it more web friendly. And this concept of DevOps, which we all talk about so much, how do we leverage the concepts of DevOps that would hold back our development agility, right? The development staff have become very agile over the last several years, but infrastructure can't keep up with that typically. So how do we kind of reconcile that so that we can realize our dream of that DevOps value chain? And so that's why it was brought in. All right, so you mentioned a couple of the products. Can you paint a picture for us? So what's the IT staff look like kind of before, and were you expecting to do some organizational changes as part of moving to the new, I mean DevOps isn't necessarily something that your typical storage guy has been doing for years, so tell us about that. That's exactly right. So when I arrived about three years ago, HBC had your typical technology stack. We had a virtualization engineer that was very compute focused. We had a network engineer who was very network focused. We had a storage engineer very storage focused. And then we had a handful of ITIL help desk kind of support staff. And what I had presented when I came in was, why don't we try to take these folks and move them more towards a service delivery, a service management kind of platform where we are really trusted brokers of IT services? How do we do that? How do we show value, business value with the IT infrastructure? Because in the end, the traditional IT shop is becoming rapidly irrelevant and businesses can take their IT dollars and move them outside of that shop. And so how do we as an IT staff show value and maybe even take it to the point where we're a revenue center instead of a cost center? All right, so Bob, let's get into kind of the transition you went through. I mean, did you sweep the floor, just build a new data center, throw everything out that you have? Or were there some initial use cases and projects that helped you move? I'm assuming Nutanix is part of the solution. Absolutely, so yeah, obviously we don't throw away what we have. You hear it said all along, you have legacy. Legacy has to be retained. We have a lot of capital tied up in infrastructure and it's not going to go away until the book value is gone and then probably some time after that. So what we did was we looked at certain technologies that we thought were going to help us shrink our footprint, help us improve performance, help us streamline management, and then we looked for use cases where that might be relevant so that we could get some minimal investment from the organization to try the mountains in kind of a experimentation process. Can you give us a couple of examples? Absolutely. So initially we looked at VDI. VDI is a really great fit for hyper-converged infrastructure and because I wanted to bring that in, I looked at that first. The problem is we didn't have an appetite for that. We have a lot of unique workloads and so it wasn't a really good fit. Management came to me in the fall of 2013 and said, you know what? We want to do something around analytics. We want to do something around the data that we have in-house, show some efficacy of the programs that we manage for our clients around employee engagement, customer engagement, wellness engagement, and use that in the sales process to try to drive some revenue. And so we need to do that and we need to do that in a controlled way. We need to do that in a way that's hyper-forming and so we want an infrastructure dedicated to that and so that was my opportunity, that was my end. So at that point we started looking at a couple of options and we selected Nutanix through Approve a Concept and we rolled that out in a Microsoft SQL Stack, BDI types of SQL workloads and it proved out really, really well and at the end of that I went back to our senior leadership and I said, hey, look, this is it. This shows this is the future. How about we do a three-year roadmap and that's kind of what we're working against right now. All right, so Greg, let's bring you into the conversation. It's interesting how Bob said, VDI wasn't a fit and there are those out there that say, oh, Nutanix, it's just that VDI company that does some piece there. So how much is this kind of a typical example of trying to get in the environment and then spread out to be a larger piece of the infrastructure platform? Yeah, that's a good question. I think if you go back to maybe even three years, VDI was a very ideal insertion point for a hyper-converged infrastructure like Nutanix because it typically represented a new workload in the environment and gave them an opportunity to deploy new infrastructure. Today for Nutanix and perhaps for the larger hyper-converged industry, VDI is still a sizable portion of the workload but it's a minority. Nutanix today, most of our business is running production databases, big data, a lot of collaboration like Microsoft Exchange, a lot of Microsoft's apps, a lot of server consolidation. So today VDI is about 30% of our workloads. It's an ideal workload in that it demands a lot of resilience, high performance, a lot of density. But what we found is even the customers that started with the virtual desktops would be VMware View or Citrix Zen desktop is they got comfortable with the technology, they fell in love with the efficiency and the performance and they've expanded the hyper-converged footprint with Nutanix to run some of their more business critical tier one applications. So the other thing I'm wondering if you can comment on is it wasn't, you know, my traditional storage was just too expensive. It's agility and specific organizationally how they want to do things. And I heard very clearly, we know we need to change because what we're doing just isn't working for us but you know, and just going through kind of the standard refresh cycle of upgrading what you have. So do you need that change engine inside? Is that pretty typical and do you know, maybe comment on some of those other aspects? Yeah, the first thing you said is the cost. I think that if you go into a Nutanix infrastructure or hyper-converged looking to save money out of the bat, you may not always meet your goals. I think the real savings comes on the financial side on the dramatically reduced operational expenses, the drop in those over time. The primary reason people are adopting hyper-converged infrastructure is like you said, it's for the agility. The mere fact that I can stand up infrastructure not in weeks or days, but in hours. And so what we're able to do in Nutanix and what we did for Hallmark is in a same day that we're able to deliver converged infrastructure, they can start provisioning workloads that same day. And so if it's VDI or production of a database, we just collapse, we really consolidate that time to value for organizations. All right, so Bob I'm wondering, can you give us, do you have any concrete examples of kind of life before and life after rolling in the Nutanix either from the infrastructure standpoint or the operational organizational standpoint? Yeah, absolutely. We've got a couple of really concrete examples that I like to show. The very first example is, we talked a little bit here about you're not necessarily going to realize the financial impact directly when you buy the equipment. But what you do find is that as you move workloads from the traditional three-tier infrastructure to the Nutanix infrastructure, your footprint gets dramatically reduced and so our co-location costs begin to drop dramatically. We are going to save over the next three years roughly $2,200 per rack that we reduce per month. That's real numbers. We expect a return of about 65% over those three years. The next piece we saw from a performance standpoint was during the proof of concept around the analytics, we had the development team come to us and say, hey look, we're trying to do these continuous integration builds. We're trying to do 12 builds a day, but these builds are now starting to take almost two hours. And so one stops, the next one starts, there's not a lot of time for development. And so what we did was we moved that build workload which was using Final Builder on a .NET C-Sharp application. We moved that, no changes, just a V-motion into the Nutanix environment. And we went from about an hour and 40 minute build time to 22 minutes. It's about 67%, or 76%, I think it was, that we saw in terms of improvement in that workload. That's great. So Bob, we all know in IT, first time you do something it doesn't necessarily go as smooth as you want or basically if I have hindsight, I've learned something. So for your peers out there that are looking at a solution like this, what surprised you or what would you do different if you had to, if you're going into this is for the first time. I'll tell you what didn't surprise me and that was that it just worked because when I did my research, I knew that this was going to work. I just felt it. So what we learned was it really was that simple to instantiate a cluster of Nutanix and what really took us the longest was to get people on board with the idea that it was that simple. So if I had to do this over again and I am going to do this over again, what I would do is I would prep the engineering staff a little bit better. I would get them involved a little bit earlier in the architecture discussion so that they were kind of stakeholders in the process because after all, if your engineers aren't really on board with the architecture, it's not going to succeed. So you handle security also for HPC. So how does Nutanix change the security discussion? What's the impact of changing your infrastructure? Well, short-term, it really doesn't change the security discussion, but what it does do with some of the capabilities that are inherent with Nutanix, and I'm going to butcher the marketing term for it, but self-encrypting drives is probably going to help us in terms of performance. Today, we encrypt our at-risk data using Microsoft technologies, inherent within the SQL stack, and we do endpoints through BitLocker, those kinds of things. I think what this is going to allow us to do is take security to the next level, the same way we are with infrastructure, and we're going to start looking at it in terms of software defined. Everything we do now is with an eye toward software defined. If we can't script it, if we can't monitor it in an automated fashion, then we don't want to invest in it. Yeah, and some of that goes back to your comment about DevOps. Absolutely. Talk a little bit about, I mean, were you involved with DevOps before you came there? How do you kind of, it's really more of a mindset and there's tools involved, typically, but for those that aren't familiar, share your DevOps experience. Yeah. Absolutely, yeah. I've been doing what we call DevOps today long before it had a name. I was with a company called Plato Learning years ago, educational software company SAS provider and we knew back then that in order to make a transition from a client-server environment with perpetual license to a software as a service, that you had to change your mindset. And we knew that development and operations needed to work together. And so the then CTO invited me to move over from operations into development where I managed a development team for about five years. And it was through that experience that I started to see that there's a value chain in getting the software being developed out into the infrastructure and how you remove the obstacles in that value chain is what differentiates you from your competitors and that really is the DevOps value chain. Now what you said earlier about, well, there are tools involved and there are skills involved, but it is a mindset, right? It's about how do we recognize the constraints and how do we remove those constraints in the most efficient way possible using those tools and those skills? Yeah, so Greg, I have to say I haven't heard kind of the DevOps story from Nutanix. I obviously am familiar with kind of the heritage of the company, the web scale architectures and simplicity should help enable some of that, but how does DevOps fit into kind of the deployment and customer engagement from a Nutanix standpoint? Yeah, I think at least three different ways. One is by the simplicity of the platform, we really remove a lot of the complexity and help break down those silos within organizations, whether it be a reduction in the need for specialized storage skills, where the virtualization team, the business apps team, can work directly with the infrastructure team. Two is the exposure of a set of RESTful APIs, so the entire Nutanix solution is built upon a very open, extensible RESTful APIs, so we do that. Then largely just simplicity. It's the idea that I can provision infrastructure very, very quickly so that I can put in the hands of my developers such that they can start developing applications and move them through that lifecycle very, very quickly without having to go through multiple teams and get buy-offs as they go from staging and pre-production and production. All right, so Bob, we're here at VMworld. First of all, have you been to the show before? I have not. All right, but you're a company, you're a VMware customer? We are. All right, so tell us from your standpoint, what's the experience of being both a Nutanix and a VMware customer? I'm sure if you've read some of the industry magazines, you've been a watcher of the industry for a while, your take on that engagement. Yeah, so really we are heavily invested in VMware and it certainly is on our roadmap for the future in terms of our production workloads and it was one of the determining factors of us selecting a hyper-converged provider. They had to support VMware, ESX and Nutanix did that and Nutanix did that and they did it in such a way that it didn't invade the vSphere API and that was appealing to me. We are going to continue to use VMware. We are going to continue to look at alternatives from a cost perspective. What Nutanix is doing around allowing you to move applications around within that sphere, if you will, is exciting to us and so we just are nothing but excited about that. Great and the last question I have for you is how does the cloud environment fit with you? Do you consider your Nutanix environment? Is that part of the cloud strategy? Do you work with the public cloud? You've got background in cloud so how does that fit in in your organization? Well, I think it absolutely is cloud, right? What we do today, I would refer to as a private cloud. Our traditional infrastructure was not, it was very much a traditional infrastructure but what we do today is private cloud and it's service oriented and it's all of those things. We do have a selection of applications that are cloud enabled. We are looking at moving some of our workloads into the public cloud. We're looking at perhaps using Azure for DevTest. We're key partners with Microsoft as well so there are lots of opportunities, particularly around the area of DevTest and even for performance capacity kind of spikes. Yeah, Greg, you and I have talked for a few years, just some of the terminology out there we talk. What is cloud? What is software defined? Hyperconverge, heck we have a term server sand when Nutanix first came out, it was the no sand company even though you guys are the market leader in the space so our terminology doesn't match but what do you hear from customers? What's the driving conversations that's important for the IT staff and the C-suite of customers that you talk to? I mean what our customers tell us increasingly is that they need IT to really work for the business that they're spending too much time, too much money, not serving the business but maintaining the infrastructure. So increasingly they went like Bob said, they went infrastructure that just works and so when Derish was on the show earlier and he talked about invisible infrastructure, I mean he really meant that infrastructure that can elevate IT so that Bob and his colleagues can really focus on the applications without doing a lot of care and feeding of the infrastructure and I think that's a value proposition, those are technological attributes that appeal not just for the practitioner but to all levels of the business. All right, Bob, I just want to give you the final word your first time here at the show. What excites you about being at the show? What advice would you give to your peers kind of at this state of the industry? Well, I tell you, what really excites me about this show and others is just the concentration of intelligence. The smart people, the people that are innovative, it spreads, it's like a happy disease if you will and what I would advise folks who come to these things is to absolutely keep an open mind. Don't come in with a predisposition for a certain technology or other. Just go in and listen and don't be afraid to get your badge scanned and to listen to a spiel, that's what you're here for. If you don't open your eyes to what's being put in front of you, you're never going to know what the next big thing is and I'll have to tell you that selecting Nutanix and moving in the direction that we've done has been nothing but great for us and we would have never done that had I not been to a conference similar to this and listened. Yeah, well Bob, thank you so much for coming and sharing. Greg, thank you for joining us. Yeah, I can tell you just to reiterate what you said, I was at the Nutanix dot next conference in Miami and in many ways the vibe reminded me of VMworld is people excited, people contributing and exciting times to learn the industry. So thank you so much for joining us. We'll be right back with lots more coverage here from VMworld 2015, thanks for watching.