>> Announcer: Live from Miami Beach, Florida, it's theCUBE. Covering VeeamON 2019. Brought to you by Veeam. >> Welcome back to the Fontainebleau, Miami, everybody. My name is Dave Vellante, I'm here with my co-host for this segment, Justin Warren. Justin it's great to see you. This is theCUBE, the leader in live tech coverage, day two of our coverage of VeeamON 2019 here in Miami. Jason Buffington, @Jbuff is here, he's the vice president of solution strategy congratulations on the promotion and great to see you again, my friend. >> Thank you very much. >> Dave: And Nathan Hughes who is the IT director at Flex-N-Gate. Great to see you, thanks for coming on. We love to get the customer's perspective, so welcome. >> Great to be here. >> Okay, so, Jason let me start with you. Former analyst, you've been at Veeam now for long enough to A, get promoted, but also, get the Kool-Aid injection, you're wearing the green, and, what are the big trends that you're seeing in the market that are really driving this next era, what do guys call it? Act two of data protection? >> Sure. So, I preached on this even before I joined Veeam that every 10 years or so, when the industry shifts the platform of choice, the data protection vendors almost always reset, right? The people that lead in NetWare don't lead in Windows. The people that lead in Windows didn't lead in Vert. The next wave is we're moving from servers to services. Right, we're going from on prem into cloud and so, and every time the problem is the secret sauce doesn't line up, right? So you got to reinvent yourself each time. And what we saw in the past generations, what we learned from, is, you can't be so busy taking care of your install base that you forget to keep innovating on what that next platform is and so for us, act two is all about cloud. We're going to take everything we know about reliability but we're moving into cloud. The difference is, that in virtualization there was one hero scenario. VMs, right? This time around it's IaaS, it's SaaS, it's PaaS, it's using cloud storage, it's BaaS and DRaaS, there's not a single hero scenario which means we have a lot more innovation to do. That's round two. >> And you made that point today, you used the Archimedes quote, give me a lever and a fulcrum and I'll change the world. You used the analogy of backup as now becoming much more than just backup, it's data protection, it's data management, we're going to get into that. And test some of that with Nathan. So, Nathan, tell us about Flex-N-Gate what does the company do and what does your role as IT director entail? >> Okay, so Flex-N-Gate is a tier one automotive supplier. Which means that we provide parts, most of the things that go into a car besides electronics and glass, to the final automotive makers. So most of the companies that you're familiar with when you go to buy one. >> Okay, so you guys are global, I think you've got what, 24,000 associates worldwide, 64 locations. So what're some of the things that are, fundamental drivers of your business, that are rippling through to your IT strategy? >> Well, our business is varied in the sense that we do a lot of different things in house so, we do, obviously, manufacturing, that's a big part of what we do. And then, even that is broken down into different kinds and then beyond manufacturing we have advanced product development and engineering so we do a lot of that in house. >> Dave: You support it all? >> Yes. >> So you've got diverse lines of business, you've got different roles and personas, you know, engineers versus business people versus finance people. And you got to make 'em all happy. >> We've got to make 'em all happy. >> So, one of the things I love about manufacturing examples, is if you think about it it's the two extremes of high tech and low tech, right? On the low tech side of things you've got this manufacturing floor and it's just producing real stuff, not the zeros and ones that we live with, but real things come off this line. And then you have the engineering and R and D side. Where they're absolutely focused on stuff that comes out of some engineer's head into a computer, which is truly unique data, so, one of the things I love about the story is, talk about the downtime challenges you have around the manufacturing floor. Because I learned some things when we first met, that I think is phenomenal when it comes to manufacturing things that I didn't realize. >> Sure. So, we have a lot of different kinds of manufacturing environments. Some of them are more passive and some of them are more active. The most active environments are, a form of manufacturing known as sequencing. And it's sort of where you bring final assembly of parts together right before they go to the customer. The way that customers order up parts these days, it's not like they used to back in the 70s and 80s. Where they would warehouse huge volumes of everything on their site and then just draw it down if they needed it. And you just kept the queue full. Now they want everything just in time delivery. So they basically want parts to come to the line right when they're needed and actually in the order they're needed. So, a final car maker, they're not necessarily making, 300 of the same thing in a row, they're going to make one of this in blue and one of that in red and they're all going to be sequenced behind each other, one right after the other on the assembly line. And they want the parts from the suppliers to come in the exact right order for that environment. So, the challenge with that from our perspective is that we have trucking windows that are between 30, maybe 60 minutes on the high end, and if anything goes badly, you can put the customer down. And now you're talking about stopping production at Ford, Chrysler, GM, whatever. And that's a lot of money and a lot of other suppliers impacted. >> Dave: So this is a data problem isn't it? >> Yeah, it definitely is. And it's an interesting point, 'cause, you talk about sequencing. Veeam has their own sequence about how customers use the product and they start with backup, everything starts with backup, and then they move further to the right so that you get, ideally, to fully automated data protection. So, what are you actually using Veeam for today? And where do you see yourself going with Veeam? >> So, right now, we're using Veeam primarily as backup and recovery. It's how we started with it. We came from another product that was, great conceptually, but in the real world it had terrible reliability and its performance was very poor as time went on and so, when Veeam came on the scene it was a breath of fresh air because we got to the place where we knew that what we had was dependable, it was reliable. We got to understand how the product worked and to improve the way that we'd implemented it. And so, one of the key features in Veeam that really actually excited us, especially in those sequencing environments are these instant recovery options, right? So, we were used to the idea of having to write down a VM out of snapshot storage. And then being put in a position where it might take an hour, two hours, three hours before you could get that thing back online now, or again, to be able to launch that right out of snapshot storage was a blessing in the industry we're in. >> Yeah, did you see the tech demo yesterday where they were showing off how you could do an instant recovery directly from cloud storage? >> Yes, yeah. >> Did that get you excited? >> Yes. That is exciting. >> Are you using cloud at the moment or is this something that you're looking to move towards? >> Cloud is something we're sort of investigating but it's not something that we're actively utilizing right now. >> So this instance recovery, you guys obviously make a big deal out of that, I was talking to Danny Allan yesterday offline about it. He claims it's unique in the industry. And I asked him a question, I said specifically, if you lose the catalog, can I actually get the data back? And he said yes. And I'm like, that sounds like magic. So, so I guess my question to maybe both of you is, instant, how instant? And how does it actually work? (he laughs) >> It just works, isn't that? >> It just works! >> It's just magic, new tagline? >> I guess we don't have to get into the weeds but when you say, when I hear instant recovery, we're talking like, (fingers clicking) instantaneous recovery with, very short RTOs? >> To us what that means is that in practice, we can expect to have a VM from snapshot data back into production in about a five minute window. >> Dave: Five minutes? Okay. >> And that is sufficient for our needs in any environment. >> Okay, so now we're talking RTO, right? And then, what about, so we said 64 sites across the world, 24,000 associates, is Veeam your enterprise wide data protection strategy or are you rolling it out now? Where are you at? >> Yes, no. Veeam, we started with it in a handful of key sites. And we were using it to specifically back up SharePoint and a few other platforms. But once we understood what the product was capable of, and we were sort of reaching the end of our rope with this former product, yeah, we began an active roll out and we've now had Veeam in our facilities for five, six years. >> So you swept the floor of that previous product. And how complicated was it for you to move from the legacy product to Veeam? >> It was a challenge just rethinking the way that we do things, the previous product, one thing that it really had going for it, if this could be considered a positive, I guess, is that it was very very simple to set up. So, you could take an entry level IT administrator and they just next, next, next, next, next. And it would do all the things that they needed it to do. But the problem was that in the real world, that was sort of the Achilles' Heel, because, it meant that it wasn't very well customized and it meant also that, the way that they've developed that product, it became performance, it had poor performance. >> So the reason I ask that question is because, so many times customers are stuck. And it's like they don't want to move, because it's a pain. But the longer they go, the more costly it is, down the road. So I'm always looking to IT practitioners like, advice that you would give in terms of others, things that you might do differently if you had a mulligan, I don't know, maybe you would've started sooner, or maybe there were some things that you'd do differently. What would you advise? >> Yeah, I mean, if we'd understood, the whole context of what was happening with that other product, we would've moved sooner. And the one thing that I will say about Veeam is, it's not click and point. It does involve a little more setup. But the Veeam team is excellent when it comes to support. So there's nothing to fear in that category because they stand behind their product and it's very easy to get qualified technicians to help you out. >> Is that by design? >> I don't know if it's. Well, the being great to work with, yes, that's by design. >> Yeah, but I mean. >> I was talking to Danny yesterday and asked about the interface thing. Because there is always that tension between making it really really simple to use but then it doesn't have any knobs to change when you need to. >> That's what I'm asking. >> But it can't be too complex either. >> Our gap actually comes a little bit later in the process, right? So, you asked earlier about, in what ways do you use Veeam? And we think about Veeam as a progression, right? So, everybody if they're using Veeam at all, they're using it for Veeam backup and replication and because foundationally, until you can protect your stuff, right? Until you can reliably do that, all the other stuff that you'd like to do around data management is aspirational and unattainable at best, right? So, we think the journey comes in at yeah, it is pretty easy, to go next, next, next, finish. Just a few tweaks, right? To get backup going. But then when you go beyond that, now there's a whole range of other things you can do, right? So Danny, I'm sure, talked about DataLabs yesterday. The orchestration engine, those are not, next, next, next, finish. But anything that's worthwhile takes a little bit of effort, right? So as we pivot from, now that you've solved backup, then you can do those other things and that's where we really start going back into something which is really more expertise driven. >> Well, and it's early days too and as you get more data and more experience you can begin to automate things. >> Yeah, absolutely. So Justin was asking, Nathan, where the direction is. Today it's really backup. You've seen the stages where, talking about full automation. Is that something that, is on the horizon, it is sort of near term, midterm, longterm? >> I mean, coming to the conference, our experience with backup, or Veeam, is primarily backup and recovery operations but, I've seen a lot of things in the last few days that have piqued my interest. Particularly when it comes to the cloud integration. That's being actively baked into the product now. And, some of the automated, API stuff, that's being built into the product. Any place where I can get to where we simplify our procedures for recovery, that's a plus. So I'm really excited about the idea of the virtual labs, being able to actively test backup on a regular basis without human intervention and have reporting out of that. Those are things that I don't see in any other product that's out there. >> You know, there's another piece of the innovation that we should think through, and, so we've talked about the sequencing side which is where we focus on RTO, how fast can you get back and running again? And when you and I talked earlier, the example that we worked on was think of a zipper, right? You've got the bumpers coming in to a line of cars and if either side slows down, everything breaks, and at the end, by the way, is the truck, right? And everything has to come at the same time at the same rate, if there's downtime on either side of the source, you're done. But that's an RTO problem. The engineering side, for high tech, is an RPO problem, right? You have unique stuff coming out of somebody's brain into a PC and it'll never come out that way again. And so, when we look at backup and replication, that should be the next pieces to go on. And then as you mentioned, DataLabs becomes really interesting and orchestration, so. >> Well speaking of human brains, and you kind of touched on it, Nathan, that you came here to learn some things and you've learned things from different sessions. So, what is it about coming to VeeamON that is worth the time for IT practitioners like yourself? >> I think it's all those, I mean we were talking about Veeam, doing backup and recovery operations, fairly straightforwardly, in terms of getting in, but once you see some of this stuff here at a conference like this, you get a better sense of all the more, elaborate aspects of the product. And, you wouldn't get that >> See the possibilities. >> I think, if you were just sitting in front of it using it conventionally, this is a good place to really learn the depth and the level that you can go with it. >> And you're like most of your peers here, is that right, highly virtualized, is that right? Lot of Microsoft apps. And, they say, mid-sized global organization, actually kind of bumping up into big. >> Nathan: Sure. >> Yeah, cool. I asked about the data problem before, it sounds like the zipper's coming together, that's some funky math that you got to figure out to make sure everything's there. So, talk about the data angle. How important data is to your organization, we know much data's growing, data's the new oil, all those promides but, what about your organization specifically as it relates to a digital strategy? It's a buzzword that we hear a lot but, does it have meaning for you, and what does it mean? >> Data is vital in any organization. I mean, we were referencing earlier, how you've got low tech in manufacturing, or at least people think of it as lower tech. And then high tech in R and D, and how those things merge together in a single company. But the reality is all of that is data driven, right? Even when you go to the shop floor, all your scheduling, all your automation equipment, all this stuff is talking and it's all laying down data. You're putting rivets in the parts, you're probably taking pictures of that now with imagers when you're in manufacturing. And you do that so that if you get 300 bad ones you can see exactly when that started and what happened at the machine level, right? So, >> That's a good one. >> We're just constantly collecting massive volumes of new data, and being able to store that reliably is everything. >> Well, and the reason I'm asking is you guys have been around for a while and your a highly distributed organization so, in the old days, even still today, you'd build, you'd get a server for an application, you'd harden that application, you'd secure that box and the application running on it, you'd lock the data inside and, my question is, can, the backup approach, the data protection approach, the data management, or whatever we want to call it, can it help solve that data silo problem? Is that part of the strategy or is it just too early for that? >> I'm, sorry, I'm going to get you to repeat that question in a slightly different way. >> Yeah so, am I correct that you've got data in silos from all the years and years and years of building up applications and-- >> I mean, we have-- >> And can you use something like Veeam to help unify that data model? >> Draw that all together? Yeah. I think a lot of that has, it's more on the hosting side, right? So it depends on how those systems were rolled out originally and all that kind of thing. But yeah, as we've moved towards Veeam, we've necessarily rebuilt some of those systems in such a way that they are more aggregated and that Veeam can pick them up in an integrated kind of way. >> You see that as a common theme? Veeam as one of the levers of the fulcrum to new data architecture? >> We're getting there, so here's the trick. So, first you got to solve for basic protection, right? But the next thing along the way to really get towards data management is you got to know what you got, right? You got to know what's actually in those zeros and ones. And so, some of the things that you've already seen from us are around what we do around GDPR compliance, some of the things we do around sanitization of data for DevOps scenarios and reuse scenarios. All of that opens up a box of, okay, now that the data is curated. Now that it's ingested into our system, what else can you do with it? You know, when I talk to C-level execs, what I tell them is, data protection, no matter who it comes from, including Veeam, is really expensive if the only thing you do is put that data in a box and wait for bad things to happen, right? Now the good news is, bad things are going to happen, so you're going to get ROI. But better is don't just leave your data in a box, right? Do other stuff with that data, unlock the value of it and some of that value comes in, now that I'm more aware of it, let's reduce some of the copies, let's reduce some of the compliance mandates. Let's only put data that has sovereignty requirements where it goes, but to do all of that, you got to know what you got. >> Go ahead, please. >> There was some impressive demo yesterday about exactly that, so, we have the data. You can use the API to script it and you can do all kinds of, basically, you're limited by your imagination. So it's going to be fascinating to see what customers do with it once they've put it in place, they've got their data protected. And then they start playing with things, come to a conference like this and learn, ooh, I might just give that a try on my data when I get back home. >> That's right. >> We'll give the customer the last word, Nathan. Impressions of VeeamON 2019? >> It's been great. And like I say, if you're a company that's been using Veeam even for a while, and you have your entry level setup for backup and recovery and I think there's a lot of, probably, companies out there that use Veeam in that kind of way, this is a great place to have a better understanding of all that's available to you in that product. And there's a lot more than just meets the eye. >> And it's fun, good food, fun people. Thanks you guys for coming on, really appreciate it. >> Yeah, thank you. >> Alright, keep it right there, buddy, we'll be back with our next guest, you're watching theCUBE, Dave Vellante, Justin Warren, and Peter Burris is also here. VeeamON 2019, we'll be right back. (electronic music)

>> Announcer: Live from Las Vegas, it's theCUBE! Covering VMworld 2017! Brought to you by VMware and its ecosystem partners. (futuristic music) >> Welcome back everyone, we are live, here, in Las Vegas for VMworld 2017, I'm John Furrier, my cohost Dave Vellante, eighth year with theCUBE, proud to have two great guests, Josh Epstein, CMO of Kaminario and Eyal David, CTO of Kaminario, great to see you guys again! >> Likewise, great to be here! >> You guys had a great event in Boston recently, what's going on with you guys? Give me an update on the company. >> Sure, I'll go first. Kaminario's been around for awhile, but we've been, first of all, moved the headquarters over to east coast US, outside of Boston, Massachusetts, opened up a great new office space there. Got a lot going on from a product perspective, a lot going on from a go-to-market perspective, you see a lot happening in the all-flash space and the storage space in general, and just, really excited to take it to the next step. We see a lot of things happening here. >> It's a pretty big week this week. We saw Scott Dietzen from Pure Storage become the Chairman and Jean Carlo, ex-CISCO MNA guy from Silver Lake come in to be CEO, so Dave and I were speculating, All flash, a lot of, what's going on! A lot of people saying, woah, is it growing? Still a need for flash. What's the big hubbub about? >> So, we definitely see a change in the market, and the emergence of two different models. The way people used to buy storage, and the way next-generation application, cloud-scale application, software-to-servers, e-commerce, online businesses, need to buy storage. And their need for simplicity, performance and cost-efficiency at scale is still driving the need for flash storage, and we'll talk about this yet to come some more. >> And you guys see those as really distinct opportunities, is that right? Can you add some color to that, Josh? >> Yeah, I think that we see the flash space made up of two different markets, one is just the massive stocking function of traditional enterprise data centers, making the move en masse to flash. And there you have, obviously, the incumbent vendors with their flash solutions, you know. That's a dogfight, there's a lot of competition in there. There's this other market which we see growing more healthily, more organically, which is the growth of these cloud-scale applications. As Eyal said, flash provider, or, software-to-server providers, e-commerce providers, fintech, healthtech, these large, highly-scalable, database-driven cloud-scale applications. That means a different type of of scale, so that's where we see less competition from the incumbents and more opportunity -- >> What's different about that market, what's the requirement, what are they looking for that makes this a good engine for them? >> So one of the key requirements is agility and flexibility. One of the current characteristics is they don't really know what is going to be the next workload, how their workload is going to change in scale over time. So they need an infrastructure that can change and adapt to their needs, still deliver the same level of performance, still deliver the same level of simplicity. But have that flexibility to address their changing needs in capacity and performance, to address growth in customers, changing in workload application, without too much pre-planning. >> So I'd ask the question to you guys, I get this all the time. So since you guys are the gurus in the area. I get this question a lot, what is a modern data center? With all the action on private cloud happenings, true private cloud, they truly point out, people are re-tooling their data centers to operate like cloud, it's still on-premise. That's kind of the gateway to hypercloud, very clear. Public cloud, workloads, all bursting, that stuff's great. What's a modern architecture, what's a modern data center? When I hear that term, what do you guys mean? >> That's a great question. So the modern data center, or even the next generation data center is exactly that, one that allows enterprises to achieve the same levels of scalability, efficiency, as the hypercloud, but on-premise, or in a hybrid fashion. But it allows them to have that level of control against operation simplicity that's hard to come by, but on their own terms, adapting to their own needs. >> So without the need to build out a massive engineering team to build this from the ground up. >> So are the buyers different, are those two worlds coming together? I wonder if you could address that. >> Yeah, I think the buyers are, in fact, different. I think, now, you see a convergence over time as the classic enterprise data centers start to look more like a private cloud. But we see this growth in large-managed private cloud providers really exciting, and they come in different forms. You have the Telcos getting into the business, you have the outsourcers getting into the business, you have the traditional channel getting into the business. We have a great partnership with Vion, a big federal reseller, and using Kaminario as a flash service offering. And they start looking like a cloud provider, and they're thinking like a cloud provider. >> And what's the benefits then? Cause I was just looking at the gov cloud impact, I was just at the Amazon Public Sector Summit. Huge traction right now because it's so fast, you can get into the government cloud quickly. Why is that unique, why, as a service, and why are you guys really driving that? >> One, it fits with our architecture perfectly. But I think from a customer standpoint, the ability to procure, like, procuring from the cloud, but also to get the kind of services, you know, as people start re-engineering applications thinking about dev-ops, cloud-data-type applications, leveraging the same kind of utilities that they might get from an Amazon or an Ajer, from a managed private cloud provider, it becomes really important. >> And Al-fed ramp is there, you get all the federal information stuff going on around it. >> So I wonder how you deal with this problem, it's a relatively small company, you're up against the big guys, you say, it's like a rock fight. But you have an affinity to, let's say, SAS players. They like your product and it fits better with their vision. But then you have this big whale, saying, okay, I'm going to buy my HR software from, you know, some SAS provider, I'm going to do some, whatever, 70,000-person deployment, but, as a quid pro quo, you've got to buy my all-flash array. So you must see that all the time. When you peel back the covers, underneath that SAS provider, what do you really see? Like, they fence off, sort of, legacy-vendors' stuff, and they really drive in their core business with your modern platform? Or is it sort of just a mishmash? >> No, I think we're seeing a shift. I think what we're seeing is, some of the legacy architectures are running up against boundaries. Boundaries in terms of complexity, boundaries in terms of agility. Kaminario was built to scale from the get-go. It was built for performance and it was built for scale. And I think what we're seeing is, the main value of these SAS providers, as they're reaching scale, is the ability to deliver consistent performance, consistent cost-efficiency, and really, our predictability. The ability to sort of forecast in the future what cost structure's going to look like in order to continue to deliver high-performance to their own users. >> So the hypothetical example I gave, I'm sure you see it, but are you, you know, winning head-to-head in those environments, and your piece is growing, and that's sort of just a static one-time deal? >> That's exactly what we're seeing, so our main growth, our main focus is on these software-to-service companies or software-to-service departments within existing companies building these types of offerings to deliver this as a service consumption model. And you were asking about the back-end, in the back-end, these are often large-scale databases operating mixed types of workloads, for example, transaction processing, analytics, all at the same time. And the need to support these types of workloads requires an infrastructure that can deliver at-scale, consistent performance. And when we face off the legacy vendors in those environments, we win out. >> You have to be substantially better as a small company. You are, otherwise, you're out of business. >> Absolutely. >> And so, interesting thing about the flash market it, a lot of the big guys realized right away, wow, I'm way behind, so they went out and they bought a lot of startups. What happened, did they sort of pollute them, through the integration, or ... (laughing) >> I think the marketshare statistics are a little bit confusing, but what we see is, you know, the bulk of the legacy vendors, you know, push in what we call retro-fit flash, basically taking their old legacy architectures, their scale-up or scale-out architectures, and cramming flash into it, and basically, then, they don't bring the same kind of simplicity, same kind of agility, same kind of scalability as a built-for-flash-offering like Kaminario. >> Right, what about, you guys have some announcements this week? >> Yup, take that? >> Yeah, two weeks ago we announced our next-generation platform, K2.n, which is based on a fully-converged, NVIO mean over fabric back end. This is basically taking our core operating system, Vision OS, which is a mature and robust storage software stack with all the data services and enterprise features that enterprises need. And deliver it on an NVIO fabric backend which leverages the existing capability to aggregate capacity and compute, and take it to the next level, delivering a very scalable and agile storage cluster that allows you to mix and match different types of resources, to add and remove resources very dynamically, and make your data center responsive in minutes and not hours or days or even months. >> You guys are familiar with our service and research, and we're very excited about NVIO over fabric, because we've been talking about it since probably, maybe 2008, 2009, some type of ability to scale and to communicate, and that's here today, finally. How close are we to actually having a product in the field that I can actually deploy? >> We will actually be shipping this in Q1, the K2.N They added another layer on top of that, We also announced a new software platform called Kaminario Flex, which is a orchestration platform which rides on top of K2.N, and allows you to dynamically compose virtual arrays out of these NVME-connected resources. So I really take that, looking ahead, that the classic notion of a monolithic shared-storage array, is going to die over time. >> Well, here's the numbers. I mean, it's automatic, go ahead. >> Well no, this is the whole debate that we've been clearing up with the true, private cloud report. I mean, guys, no-brainer, check, as a service, as the future, so you're good there. (laughs) The true pilot board, too bad it shows the on-prem stuff is declining in general, that's settlement for buying boxes, and the old way of doing things. Labor's being automated away and shifted, that's pretty obvious. Enter your business model, right? I mean, this is perfect for any cloud deal. >> Right. >> The question is, track record, bulletproof, reliability, security, the table stays all shift, data protection, all these details, that's what they care. You guys check that box ... (laughing) >> So the disability takes vision away, so I'm going to take it to the next generation. Technology is what actually allows us to do that. Whether it's in a hypercloud or we're going into a managed cloud provider, that is becoming a very desired consumption model for a lot of the ads of service members, allows them to build such a flexible architecture, based on a mature software step. >> So you guys, really, from what I see is your strategy is, get this out there quickly from a tech standpoint, software, flex, and integration with cloud is critical. Because you can offload a lot of that heavy lifting on those unique requirements to the cloud guys, where the pre-existing tech exists. Did I get that right? >> Yeah and I think what we see is these managed cloud providers are going to want to have a say in it, they want to actually be part of the evolution of the platform, right? >> Yeah, go ahead, fine, it's your stop! You can always buy the servers more flash! (laughing) >> So talk about your channel, and you go to market, help us understand that a little bit better. >> Yeah, I think it's all about focus for Kaminario. I mean, let's face it, the flash space is competitive, right, if we're going to go head-to-head with everyone, kind of, pull one of these growth-at-all-cost models. And you see what the market values those types of companies. So we've been really focused in two ways. One, SAS providers, next-generation business. I mean, if we opportunistically find a VDI deal, okay, that's great, we have a great solution for VDI, but it's not something that we're going to go out and hunting day to day. The second is really to focus on channel partners. We've got a channel first model, really, effectively 100% of our new business in 2017 will come through a channel partner. Most of those channel partners are looking at developing some type of managed services offering as well, so you know, it's not just about the margin on the deal, it's about the longterm -- >> Cause they're trying to respond to the market transit and value. >> Exactly, so it's about focus on a relatively small number of channel partners that get it, that like our model, and again, it's just -- >> Hey, you'll make money from it, cause that's all, at the end of the day, you've got to get that leverage, because that's your David and Goliath story. >> Exactly, yeah. >> And, global footprint? Is it primarily US and Europe or -- >> Yeah, so it's been, we started in Israel, US has been a good focus, last year we opened up the UK and France, end of the great we opened up Korea, we're now in Singapore, we're moving into China through partners, and so yeah, this is a global story. Clearly, US is the, in terms of adoption of these server infrastructures, US is really the furthest ahead, but it's a global phenomenon. >> What do you make of the VMwear momentum? Because two years ago, VMwear was, the stock was sort of in the tank and there was no growth, and now it's on fire, the data center's on fire, you can't get data center space! (laughing) >> From my perspective, the fast adoption that VMwear had for new technologies, for adopting containers, for adopting cloud paradigms, for adopting this new delivery model, and enabling a fuller stack aligns very well with the kind of demands of the next-generation data system we talked about, where the management plane, the orchestration plane, is becoming more and more important in optimizing the way in this infrastructure gets delivered. So that's, I believe, what is driving that forward. >> Josh and Elay, thanks so much for coming out, coming our way, you guys, company watch, love the business model. The tech comes home, you get it with that integration, man there's not a leverage there, congratulations on your success! (laughing) Great business. TheCUBE bringing you the CUBE as a service, all flash content here! Back with more VMworld coverage after this short break. (futuristic music)