 from London, England, extracting the signal from the noise. It's theCUBE, Cover, Discover 2015. Brought to you by Hewlett Packard Enterprise. Now your hosts, John Furrier and Dave Vellante. Okay, welcome back here, and we are here live in London for HP Discover, HPE Discover. This is theCUBE's Silicon Angles flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier with my co-host Dave Vellante. Our next guest is Patrick Osborn, Director of Product Management at HP Enterprise Storage and Neal Wogan-Hendrickson with COO of 99X. Welcome to theCUBE. Good to see you, Patrick. Thank you very much. Thank you. So talk about the digital transformation as it respects the storage. You guys are doing some work with 3PAR. Neal, tell us what you guys are doing, what's some of the things you're working on, and what you're talking about here at HP Discover. We're very excited about the RMC integration between the 3PAR and the store-bonds backup system. So we're just rolling that out now in our environment, and it's looking very promising so far. So you guys are a service provider, and you cover Pan Europe, or maybe tell us a little bit more about 99X. Yeah, yeah, we are a service provider based in Oslo, Norway. We are mostly the Norwegian market, but we do have some shipping customers with Central Asia as well, so we do have to handle a lot of offices around the globe as well. And so maybe give us a little color around sort of what's going on in the marketplace in Scandinavia, and what some of the business drivers are that you're seeing, what some of the pressures are in your business, and then how that relates to your IT infrastructure. Well, more for less is also well in Scandinavian market, I guess. You working on that, Patrick? Yes, every day, every day. So being able to deliver more high quality services is a very vital asset of what we think about for the next year. So what's your take on what these guys are doing? Yeah, so from my perspective, the more for less is it's actually how we design the product, right? So we did a panel yesterday talking about some of the challenges of being able to, they have a very, very heavily virtualized environment as a service provider. So for them, the provisioning task is pretty easy, but all of a sudden now you've got thousands, tens of thousands, approaching hundreds of thousands of VMs, right, that they're managing. So they need to keep their spend pretty flat in terms of the people that manage that business. So being able to manage your primary storage and secondary storage, the data protection services that they offer as part of their portfolio are very, very important for them. So being able to do that with three par combined with store wants in kind of an invisible type of format is really helpful for them from a management perspective, to keep those costs down. So, you know, tell us more about your infrastructure. Can you paint a picture of sort of where you've come from and where you are today? Obviously virtualization, you went through that journey that allowed you to do more with less. Talk about sort of how your infrastructure has evolved and what it looks like today. Well, we are highly virtualized, as Patrick said, and we're also very segmented. We have a lot of medium-sized customers probably in the European or American protected small customers. So we have a lot of network segments and backup is sort of painful in that scenario where we have a lot of firewalls and routers and so on to traverse. So being able to, with a recent position from HPE, we're able to back up servers without traversing the network as much as we used to do. And we're also able to back up server without first server configuration. So the management overhead is going down quite a bit. So talk a little bit more about that. So prior to the recent, you're talking about bringing in store once, right? But so what was backup like before that project and then take us through the project? Yeah, well, I think we do backup more or less every way there is, including agent-based, including using the APIs of the hypervisors. But even so, it's still traversing a lot of devices before it ends up on the backup storage. So getting down on the number of hops the data has to, or points it passes through before it lands, it's a vital asset as well. Because all those network segments and firewalls and the links and routers and whatever they are, they all have to be sized according to the data passing through them. So moving data with fewer hops and more directly is very keen in the way we look forward on data protection. So what was the problem that you were trying to solve was backup windows were being hard to get met, was it too expensive? Some apps weren't getting backed up as well as they should and some maybe were spending too much, all that sort of common problems. Yeah, absolutely, we're running back at more or less 24-7. We're moving on average two gigs every second, every hour, every day for a week. Back it was error prone, small changes in DNS or in the service date or in the service could affect backup. And we had a lot of people managing and debing those errors. So that's kind of what we were hoping to do. Okay, so you weren't able to offer the SLA that you wanted for your customers. Okay, so calls Patrick says, okay, help. I got some solutions for you. So okay, talk about sort of what you guys did together, the solution, what the project looked like. Yeah, so the design point of what we're trying to do with the portfolio is exactly what Neal said. We find a lot of customers, backup touches everything. So you have to have, it's an app-centric view, it touches virtualization, it's very taxing on the network, especially the Ethernet IP network. You've got a bunch of different back ends from a storage perspective. So if you have any minute change or something gets affected in that whole stack, it's very, you have to understand the whole stack to be able to troubleshoot that. So it's very error prone. So for us, what we're trying to offer is that if you have this platform, very scalable platform in three-par that a lot of service providers use to run their business, being able to make the secondary storage part of the initial conversation when you're architecting for a service provider type environment is it's very helpful to mitigate all of these issues around networking, the application stack, it's built in from a virtualization standpoint. So the design point of the product and the solution together is to bring primary and secondary together. That's what you call flat backup. Exactly. Meaning it's not bolted on. Exactly, you don't have to traverse up and down through apps and media servers and agent base. It's really sort of an endemic property of your primary infrastructure. Okay, and what does that result in from a operation standpoint? How did things change? Well, we haven't had this in production for a really long time yet, but so far what we're seeing is that we're able to back up larger amount of data in a lot shorter time, as well as we haven't had really had any real errors so far. It's just been working. So that is a large change for us. Okay, so the quality of the backup is better. And then, how about recovery? When you go back to the previous sort of situation, did you have to do a lot of recoveries and it was just a horror show? And I presume you haven't had to do any recoveries yet, but maybe you're testing. We have tested it obviously. Yeah, the speed has been sufficient. There is still some granularities we can improve on, but in my environment, we mostly think if you really need recovery scenario, you're using a HA or DR solution or it's not restored from backup for large data sets. So it's the speed we're seeing so far more than sufficient. So are you able to change your monetization strategy as a result of this and then change the way you charge customers offer more services, charge more, charge less in certain cases, more granularity, maybe you could talk about that a little bit. Yeah, we're able to go to the market with a lower price and higher quality and we're also able to direct our focus more to the protecting data set that really matters instead of debing all the error-prone backups of the data set that aren't really that important. It's working now and that's really good. So Patrick, is this how you generally approach the market not trying to just sell a bespoke backup solution but trying to sell it as part of an overall storage architecture or is this more unique? So for us, I think in this space in particular in data protection, nobody likes backup, right? It's a TCO model, right, more than anything. Now he's laughing. Yeah. I love backup. If we can give him the ability, for example, to make the 80% of his environment that is maybe business critical, not mission critical, just part of the infrastructure. He doesn't have to worry about it and you can charge maybe a lower class of service for that and then, but really sort of talk about the SLA on your most important mission critical data and you can wrap BC, DR around those type of scenarios and really provide a high level service. So for us, it's enabling those use cases. Do you like your paying your life insurance premium every year? I don't because someone else is going to derive utility from that, right? It's a model where you have to do it, right? And if we can find some ways for them to monetize it in different classes of service, that's great. But the expectation from customers around data protection is way more than it has been in the last couple of years. They want it to go away. They want it to be invisible. They want it to be a piece of the infrastructure. So now what's new for you guys? What's on the horizon? Any new projects that you're excited about? Yeah, we have new customers coming in starting next year. They will be put on this new system. We're also going to migrate out some existing storage show it to the three par and four ones. So we have a busy in future. All right, Patrick, we'll close with you. So Hula Packard Enterprise brand new balance sheet. I've been talking about that all week, looking good. I said the deck is stacked for HPE. What gives you confidence that your strategy is going to work in storage and you're going to continue to thrive in the marketplace and grow share? So from my perspective, I see a couple of things. The technology, I'm a technology guy, right? I think we have some great intellectual property and we started down this road five or six years ago where we really focused on our own IP, right? And it's, I think that's where really when they say innovation matters, I really feel it matters because we can put great solutions into the market. We can get that virtuous cycle of great customer partnerships focusing on engineering and you see that across the board, right? With storage, obviously three-par store warrants, a lot of things we're doing there. You see that with synergy, right? On the compute side. I think one of the things you'll see is moving the, when we talk about compute in storage, whether it's in an external storage array or now you're starting to get server sand and DAS, that's going to start to come together very quickly. It's very disruptive in the marketplace. We think so. I know you do. So for us, you know, it's where we're poised. We're in a great spot to take advantage of that market because the IP for us is all in the software, really at the end of the day. Patrick, talk about the battleground because we see the management software being a real tier where you guys can add a lot of value with your solutions. Is that a big battleground? You guys see that as strategic and what do you guys have a quick update on the management side? Where's that going? Yeah, so some of the things we see on the management side are two angles, right? You have being able to manage your physical environment at scale. So we have a great, great solution in one view and a lot of customers who have our composable infrastructure and storage together being managed with one of you on the physical side. That's a great solution and customers love it, especially when you have a really large compute population. And on the management side, the provisioning layer, being able to plug into frameworks like ServiceNow and Jenkins and Kubernetes and Docker and some of the tools that Neal uses for provisioning and agility within the enterprise is something that's coming. And from our design points, when I sit with my product management team, that's in the intellectual property right from the get-go. You have to have a well-known API that's to be published. It has to be, you know, supportable and plugable into those kind of frameworks. Well, guys, thanks for coming on theCUBE. Really appreciate it. Top of the morning to you guys. This is day two, early morning here in theCUBE. Day two of wall-to-wall coverage of HPE Discover. We'll be right back with more theCUBE action here in London after this short break.