 Live from Chicago, Illinois. It's theCUBE. Covering VeeamON 2018, brought to you by Veeam. Welcome back to Chicago everybody. You're watching theCUBE, the leader in live tech coverage and we are covering VeeamON 2018, hashtag VeeamON. My name is Dave Vellante and I'm here with my co-host, Stuart Miniman. Duncan Epping is here, Chief Technologist, Storage and Availability at VeeamWare and the world's number one blogger in virtualization, yellowbricks, yellowbricks.com. Duncan, thanks very much for coming to theCUBE. Good to see you. No problem, my pleasure. It's been a while. I actually hope to be on the show probably six, seven, eight years ago. I don't know how long it is, but I've watched many episodes so it's great to be part of it. Well great. One of the biggest problems, you're so busy. Every year at VMworld, you were totally booked up. So no, thanks so much. We're so glad we could do this. No problem. So Stu and I, I remember the peer insight we did many, many years ago, back when, well we had Ed Buñonan recently and he was talking about when, when VMware sort of created virtualization, it pushed the bottleneck around to create a lot of stress on the storage systems. And then VMware for years dealt with that through API integration and the like and very well sort of covered. But I wonder if you could take us through your perspectives of the journey of storage at VMware and generally or specifically in virtualization generally. Yeah, it's a good question. I think everyone that has been part of the community has faced all of the different challenges from a storage perspective. I mean, Stu, you know what kind of problems EMC had when VMware first started doing virtualization and I think the key reasons for these were fairly straightforward. When we started virtualization and we started leveraging shared storage systems, those shared storage systems were never designed with virtualization in the back of their minds. They were designed for physical workloads, maybe one, two machines connected to it, you know, in larger environment, maybe 10 or 15, but not 10 or 15 physical hosts with hundreds of virtual machines. So what we started noticing is that from a performance perspective systems were lagging. We were doing all sorts of different things to the storage systems that they weren't expecting. Virtual machine snapshots, they were seeing IO patterns that they had never seen before. Instead of sequential IO, we had a lot of random IO. So we had to start doing different things from a storage perspective. So as you said, we started with APIs. We had the vSphere APIs for IO filtering. We have the Divi APIs, the array integration, so that we can upload some of the functionality. But of course, on top of that, what we started doing within VMware is we started exploring what we can do smarter from a storage point from our stands. So not just looking at how we can help the ecosystem, but also what we can do from our perspective. So there were two main efforts over the past couple of years. The first one is virtual volumes. It has taken a while before the adoption ramped up. I think part of that is mainly because a lot of our customer base was still on vSphere 5.5. Now that we're starting to see broader adoption of vSphere 6.0 and actually 6.5 and 6.7, we're starting to see the adoption of stuff like virtual volumes go up as well. That is also due to the fact that our partners like Pure Storage, Nimble, HP with Repar has been pushing, or have been pushing vVolves tremendously. So they've done a great job and we're starting to see a lot of customers adopting vVolves and that way we're getting around some of the limitations that we had from a traditional storage perspective. Explain that. What are customers telling you about the benefits that they're getting out of vVolves and vVol adoption? Well, there's two main things. It kind of depends on what kind of problems you're facing, but a lot of customers come to us with management issues and scalability issues. From a scalability perspective, we have larger customers that literally have thousands of volumes. If you look at an e6 cluster today, you're limited in terms of the numbers of volumes you can connect to a cluster. So that's one thing. As soon as they start moving to vVolves, now they're not managing those individual volumes anymore but they're managing the storage system as a whole and they start creating policies and that's where the management aspect comes into play. So it becomes a lot easier to manage because instead of having thousands of volumes to select from, they don't no longer need to look at a spreadsheet, for instance, figure out where to place a virtual machine. Now they simply pick a policy and the policy engine will figure out where to place that virtual machine. And that's- It sounds like cloud. It actually is the cloud version or cloud-ified version of storage, I would say. It brings a lot of benefits. And the funny thing is that we've been talking about policies and policy engines for a long time, even in the cloud, but try to come up with one cloud that actually has a decent policy engine. I don't think anyone has that today. From a storage perspective, I think the storage policy-based management framework that VMware has is quite unique. We're now starting to see that popping up in other areas and that's the strange thing about it. I'm always back to the software mainframes, Stu. Yeah. And Duncan, one of the things we've really seen a transition for, it took us about a decade to try to fix storage in a virtualized environment. Today, most things are built either understanding virtualization or at least that's part of the puzzle. And of course, V-Balls led us into was the ability for V-SAN. So maybe help us kind of transition that threshold as to how that's just kind of a given underneath for V-SAN and other solutions like it. Yeah, if you look at V-SAN, it has been around for a while. The beta was in 2013, as you guys know. We have a large adoption. At least we saw a large increase over the last couple of years, I would say the last two years. You guys have spoken with Yang Bing before, so you know about the business side of V-SAN. I'm not going to cover that. But if you look at it from a technology perspective, we started developing this 2008, 2009. That's when we started thinking about what we could do different from a storage perspective. There were all these companies doing something in the hyper-convert space and we figured we could do something significantly different than they were doing. They had a storage solution that's had on top of the hypervisor. We own the hypervisor so we can create something that sits within the hypervisor. And that's when we started looking at including these different technologies. So we started looking at how can we introduce things like de-duplication and compression? What can we do for robot solutions? Can we do something like stretch clustering in an easy way? There are a lot of stretch cluster solutions out there, but if you look at the stretch clustering solution today, it typically takes weeks to implement that. If you look at something like V-SAN, it was our aim to actually be able to deploy something like that from a storage perspective within hours instead of weeks. And we've been able to achieve that. And that has been a huge undertaking, but I think it's fair to say that it has been rather successful. All right, Duncan, help connect the dots to where we are here at Veeamon. It's funny. I think Veeam started out heavily in virtualization, still heavily involved in virtualization. They got a V in the beginning of their name. When I hear the keynote this morning, a lot of hyper reminded me of before we had, before hyper-converge fully took over, VMware tried to call it a hypervisor and a converged system around VMware. So talk to us a little bit about data protection, the Veeam relationship, and how that fits into things like V-SAN and V-Sphere. Yeah, I think of, you know, I talk to a lot of customers as a chief technologist. It's part of my role to talk to customers and have discussions about what's on top of their mind. Data protection is always one of those things that comes up. I would say it's always in the top three. And whenever you talk to a CIO, CTO protection of the data, availability of data, resiliency, reliability, it's fairly important. Veeam, of course, for us is a great partner, primarily because of the simplicity of the features and the products that they offer. Whenever I talk to a customer and they explain how difficult it is to manage their backup and recovery solutions, I always point them to a partner like Veeam, simply because it makes their life, it's going to make their life a lot easier, if you ask me. And I can see that Veeam is slowly transitioning. As you mentioned, the Veeam is in front of the name. The Veeam is in front of our name as well, but we know that it's not. It's not, the whole world isn't just Veeam, the whole world isn't just virtual. There's a lot of different solutions out there. And naturally, Veeam is looking at other revenue streams as well. I would argue, though, that if you look at something like the Edge Space, which I think that, more or less, exploring the looking at things like IoT, there's going to be some form of virtualization in that, whether that's Veeam-based or another solution, of course, is going to be the question. That's something that we'll need to figure out in the upcoming years. But I think there's a big opportunity out there. If you ask me, the keynote was really interesting. I kind of missed the in-depth details. I'm hoping that the last, the closing keynote is going to give some more details about what they will be doing in the IoT space, how they see their solution evolving from that point of view, because it's a market that's still being developed. But that's definitely going to be interesting. So Duncan, it's interesting to hear you say that when you talk to customers, data protection is in the top three, even amongst CIOs. It used to not be that way. Data protection was always a bolt-on. It was an afterthought. It was going to have one size fits all. What's changed? Well, I think the importance of the data has changed. If you look over the last 10 years, whenever you talk to any company out there that has lost any significant amount of data, they understand what the value was of the data that they were hosting. I think the big difference over the past 10 years is in the past we had applications like email and maybe some file services and that's it. But now everything revolves around applications. And that's also the shift that I'm seeing in the industry, also from an IT perspective. In the past, over the past decade, I think everyone has been focused on the infrastructure layer. If you look at something like VBlog, very much infrastructure focused. If you look at something like hyper-covert solutions, very much infrastructure focused. But now whenever we talk to customers, customers are more or less more or more interested in what we can do for the application layer. What kind of benefits do we have for exchange, for Oracle, for SAP, you name it. I think that's also a big change that's happening in the industry right now. One of the things from a technical perspective, there may be others, but when VMware really became prominent, it was wonderful. But we were reducing the number of physical resources and the one workload that took a lot of physical resources was backup. And then sort of Veeam swept in and took advantage of that sea change. What's the technical constraint now with when you think about things like multi-cloud and SAS and IoT, data's much more distributed. It's out of the control necessarily of a single platform. So from a technical perspective, what's the big challenge and sort of the gate to architectures today? Well, as you said, the distribution of data is the big challenge as it stands right now from a technical perspective. The biggest challenge that most of the players in this space and not just Veeam other players as well will have is trying to figure out how to control and manage their data. Other platforms are facing similar challenges and no one really has solved this problem yet. We're starting to see some players in the space that have solutions that sit out in Azure, that sit out in Google Cloud. But it's a very challenging solution. And I think if you ask me, and it's something that I've said internally as well, the company that it's capable of managing and owning the data is the company that's probably going to be most successful in the cloud war that's now happening. I think that's the most critical aspect. Workloads can move around, but data is very difficult to move around and own as well. Duncan, there's one of the discontinuities we've seen in the marketplace that you mentioned earlier. I wonder if you can talk to. In the enterprise and the data center you talk, how do we get them to get to that next version? Comfortable with it, it's stable, it works. If I look at the cloud, I'm running Microsoft Azure or AWS, I'm running the version that they want. How do we help close that gap? Because from a security standpoint, from a feature standpoint, we need to move there, but it seems to be just one of the greatest disconnects we see between kind of my data center and somebody else's cloud. Yeah, that's a great question. I think we had a lot of challenges in the past. I think it's fair to say with vSphere 5.0, it was a great release, 5.5. Or 5.5.1 great releases, but the challenges that we had from an upgrade perspective was typically vCenter and all of the components connected to it. It's not just the vSphere platform, but if you look at the vSphere platform, the challenges that we had were all of the components integrating with it, whether that's something like VR Ops, VRAs, or VR Realize Automation, but it could also be something like Evermar or maybe Veeam. So there were so many different components we had to take into account. So what we started doing within VMware is simplifying the architecture from a vSphere perspective. If you look at vCenter, for instance, it used to be a solution where we had multiple functions spread out across different virtual machines. So I'm now trying to bring that back into a single virtual machine again, actually dumbing it down, making it easier to upgrade. So that is something that is actively happening within VMware. That is something that we started with 6.0. And that's also the reason why we see the adoption of from 6.0 to 6.5 and 6.5 to 6.7 is at a much faster pace than in the five code streams. So five to five dot one, for instance, took a lot longer for a lot of customers, or five dot one to six dot zero took extremely long for a lot of customers. It's the key reason is complexity from our infrastructure stand. Why we're changing that and we're evolving that in the upcoming years. Duncan, that's the last question here because we got to jump, but as the technologists, things that you're looking at that are exciting to you, that get your juices flowing. Yeah, that's an interesting one because it's something that I've been thinking about recently. I've been doing vSphere for the last, well, it wasn't even called vSphere back then, but I've been doing this for the last 12 years, virtualization, 13 years, maybe something like that. At least as a consultant and then as a technologist and technical marketing. But recently I'm starting to look more and more into the edge space for computing, IOT. I think that's a really interesting space, especially because there isn't really significant market. Well, there is a significant market out, but there isn't really one player out there that really stands out. No one has really figured out what customers would like to do with it and how our customers are going to use it. So the edge computing space and IOT is a really interesting thing and especially because of the distributed aspect is one of the things that I've been always been passionate about. The vSphere clusters, which is a distributed mechanism. So distributed computing is definitely something that has my interest. All right, if you care about virtualization, VMware, follow the yellow brick road, yellowbricks.com. Duncan, thanks very much for coming on theCUBE. Thanks for having me guys. You're welcome. All right, keep it right there, everybody. We'll be back with our next guest. You're watching theCUBE live from Chicago, Veeamon 2018. We'll be right back.