(upbeat instrumental music) >> Welcome to the special CUBE Conversation here, in Palo Alto, I'm John Furrier, host of theCUBE. I'm here with David Richards the CEO of WANdisco, CUBE alumni, been on many times. WANdisco continues to make the right bets. The bet they recently made has been on cloud many years. We've covered it certainly on theCUBE. But live data is the new hot thing. Multiple clouds is turning out to be the trend. That's your friend. David, good to see you. >> Great to be back. >> Thanks for coming on. So we talk all the time about how you guys have always evolved the business and continued to stay out front in all the major waves. Now again, another good call. You've certainly bet on Cloud. We've talked about that, Open Source, Big Data, Cloud, you saw that coming, positioned for that. But now you got some great momentum and resonance with customers around live data, which is not a stretch, given what you guys have done with replication, things in the past, the core intellectual property. Give us the update. You guys have been in the news lately. >> So, thanks and I think you enumerated the past history over the past two or three years, which we like to say that we're living in dog years. Everything's happening seven times faster than it would do normally. So of course, we started out life by making a prediction that storage arrays would change. People are beginning to store, companies beginning to store structured and unstructured data, mammoth sizes that we've never seen previously. We're going to have to resort to Open Source software, running a commoditized hardware that we'd already seen the social media companies move to. Then we've seen, we began to see a problem emerge, even in that marketplace, where spike computes all the applications which were going to be heavily compute, would need to run in Cloud and Cloud environments where you have complete elastic compute at remarkably low cost. And that leads to a problem. So this iceberg kind of that we like to talk about underneath the oceans, so moving data for static archival data really simple problem. And that's not live data, that's archival data. You just FTP it from point A to point B. But if we're talking about transactional systems where 10, 20, 30, 40, 50 percent of the data set changes all of the time, that creates a humongous problem in moving data from one premises to cloud, either for hybrid cloud or between clouds for multi-cloud. And that's the precise problem that WANdisco solves. And we've seen customer attraction, recently we've just announced the deal, jointly with Microsoft Azure. Where a big healthcare company, who 12 months ago were not talking about cloud suddenly they got over that hump where security keys could be managed by themselves within the cloud, were able to move petabytes-scale data from their on-premise systems into the cloud, without any interruption to service, without any blocking. That's a trend that we're seeing our pipelines now full of companies, all trying to do that. >> It's like you hit the oil gusher with data, because the data tsunami has been there, and we've documented certainly on theCUBE, and our Research team at Wikibon, have been talking about it for years, and now you're starting to see it, and you guys are getting the benefits of it, is that people figured out that it's moving data around is expensive. And it's hard to do so you push compute to the edge, but you still got to move the data around because the key part of the latency piece of the cloud. So how do you do that at scale? So this is the thing that you guys have, and I want you to explain what it is. You guys have live data from multi-cloud. What does that mean? What is all the hubbub about? What's the buzz? Why is this such a hot topic, live data from multi-cloud. >> Okay so let's just take a step back and talk about what multi-cloud actually is in today's definition, which is the vendor's definition, which is very convenient. So what they mean is, moving, putting applications into a container, Kubernetes or whatever, picking it up and shifting it somewhere else. And hey presto, I've got applications running, the same applications running in two different clouds. That is not multi-cloud because you're forgetting about the data, and the iceberg underneath the ocean of this colossal amount of data. If I've got petabyte-scale, multi-terabyte-scale data sets, and I need to run the same applications, or different applications but against the same data set, I need guaranteed consistent data, and that is, by definition, a data consistency problem. It is not a data replication problem. So all of the stuff that we used to use in the past for gigabyte-scale data, for traditional, relational database problems, none of that stuff works in a live data world. And by live data, we're talking about multi-terabyte, petabyte-scale data. Data sets that are so large that we've never seen them before running in end cloud locations. It's different or same applications, but guaranteed consistent data in every location. >> So you guys have had this core composite around integrity around the data, whether it's in replication. Sounds like the same thing's true around moving data. >> Yep. >> You guys are managing the life cycle of end-to-end of data movement. >> Yep. >> Point A to point B. >> Yep. >> The other approach is to move compute to the data. >> Yep. >> We're just seeing Amazon do a deal with VMware on-premise. So there's two schools of thought. When should customers think about each approach? Can you just kind of debunk or just clarify those two positions? >> So it's not really a chicken and egg because we know which comes first. It's definitely the chicken. It's definitely the data. So if I'm going to rebuild my application infrastructure, in the cloud, I'm going to do it piece-by-piece. I can't do lift-and-shift for a thousand applications that are running against this data set and just hope that the data that block for six months because I've got petabyte-scale data, and wait for it to all arrive in the cloud, or put it to the back of you know, use a snowmobile or some physical device to move the data. I need to do this, I need to kind of build the aircraft while it's taking off and flying and that's probably a good analogy. So what we see, is companies the first step is to get consistent data on-premise to cloud, or between different clouds. Then what that enables me to do of course, is to piece-by-piece then rebuild my application infrastructure at the pace that I want to. I mean there's a great add that I keep on seeing on t.v. Where it's migration day. As though I can press a button and then suddenly you know, in this Alice in Wonderland magical world, everything just appears. Realistically, and I saw the CEO of VMware a couple of years ago talk about being in a hybrid cloud scenario for 20 years. I think that's probably accurate. We've got billions of applications. A mix of homegrown stuff, a mix of, you know, actuarial applications in the insurance industry that are impossible to build overnight. This is going to take an elongated period of time. >> I was talking on Twitter with a bunch of thought leaders. We were talking about hybrid cloud and multi-cloud, and the kindergarten class is hybrid, right? >> Yeah. >> So you got some public cloud, then you got some on-premise data center. So getting that operational thing nailed down is great. But as you get old, you know, you progress in the grades, and get smarter, as you increase your I.T. I.Q., you're dealing with multiple, potentially multiple data centers or bigger on site, or an IOT edge, and multiple clouds. >> Yep. >> So that sounds easy on paper, but when you have to move data around the different work loads, that's the core problem that people are talking about today. How do you guys address this problem? Because I buy multi-cloud, I can see that certain tools and certain clouds the right work load and the right cloud, I get that. >> Yeah. It makes a lot of sense to me. The data is the problem. >> Yep. >> So how do you guys address that? This is the number one concern. >> So the closest, people ask me all the time about competition. The closest is Google. Google have got a product called Google Spanner. And Google Spanner is a time-sensitive, active-active WAN-scope data replication solution. That looks on paper very close to what WANdisco does. It enables them to keep active data in all of their different geolocations that they've built for their add services years and years and years ago. The trouble with that is, it only works on their own proprietary network, against their own proprietary applications because they launched a satellite and stuck it in the sky, they put dark fiber under the ocean, and they put GPS atomic clocks on every single one of their servers because it uses time and time accuracy in order to synchronize all of their data. We can do all of that over the public internet. So we're not a hardware solution. This is a pure software solution that can work over the public internet. So we can do that for any cloud vendor, and any provider of applications. And that's what we do. We're licensing our I.P. all over the place at the moment. >> So which clouds are, I imagine there's a great uptake for the clouds. Which one are you working with now? Can you talk about the deals you've done? >> We're very close. We announced the Azure partnership with Microsoft, and their Azure product, and we've been very impressed with the traction that we're seeing with them, particularly an enterprise cloud. I mean the early stage of cloud obviously was dominated by Amazon, Amazon Web Services. And they did a fantastic job of really bringing cloud to the market by accident kind of inventing cloud and then bringing it to market very very quickly. The fastest ever company to, if it's and independent company to 15 billion dollars, but most of those applications and projects and companies were born in the cloud. I mean a lot of the modern companies today were actually of course, you have Airbnb et cetera, were born in the cloud. So that, the second inning of cloud is certainly enterprise. We've also been impressed with the traction that we've seen from Google GCP as being extremely impressive. And of course Amazon continued to thrive. In cloud we also have an OEM deal with Ali, with Alibaba with their cloud as well. So they're really the only full. >> If Google has Spanner, how do you differentiate between Google Spanner? >> So Google Spanner only works on their proprietary network. Which is great for Google and between their data centers, but what about 99.9 percent of the rest of the problem, which is the rest of us right, who operate on the public internet. So we can do what Google Spanner does active-active, geo, one scope replication of data but over the public internet. >> So you guys have been talking active-active for many times. We've had many conversations here on theCUBE. So I get that. How has your business changed with cloud? You had mentioned prior to coming on camera. You made a bet on cloud. It's paying off obviously. People who have made the right bets on cloud at the right time, it's certainly paying off. You're one of them. How does the live data in the multi-cloud change your business? Does it increase your trajectory? Is there a pivot? I mean what does it mean for WANdisco? >> So the very, so my thesis or the company's thesis, I won't take the credit for it, but the company's thesis was really simplistic, which is our bet was in the small data world of gigabyte-scale data, in order to do data replication, small data equals small outage. When you get data sets that are growing exponentially, and you get, you know, data sets through a thousand or a million times greater than what we've seen previously, what was a small outage or small blocking of client applications will become an elongated blocking of client applications that we're talking about, you know, six months to move 20 petabytes of data. You can't block applications, business critical applications for six months. That was the bet that we made. We expected initially to see that happen on-premise in the data like world, in the Hadoop world if you will. That didn't quite happen, or has not happen to date. We don't think that's probably going to happen. We're certainly seeing a huge desire of companies moving those data lakes into cloud, and we've actually innovated, we've got some new inventions coming out that enable you to move in a single pass, massive quantity of data that will be exponentially faster than anything else, and just doing a unidirectional data move into clouds. That was our bet that we said "Okay, companies in order to achieve the kind of scale "that they need to achieve, "they're going to have to do this in cloud." "In order to get to cloud, "they're going to have to move that data there, "and they're not going to be able to block even for a day "in order to move that data to cloud." And that was the bet we made, and it was the right bet. >> Talk about where you guys go from here. Give a company update. What's the status of the company? Get some new personnel? Any changes, notable updates? >> So we, really interestingly, my Co-Founder and Chief Scientist is a genius, Dr. Yeturu Aahlad, Ph.D. from UT, and undergrad from IIT, a new VP of Engineering Sakthi, IIT, Ph.D. at U.T. under Draxler. This fantastic Ph.D. program they did there. My new Head of Research came from, was Chairman of Computer Science at the University of Denver. He's was an IIT undergrad, Ph.D with Aahlad at UT. And I said jokingly to Aahlad: "There must be a fourth guy "that we can bring on board here "that went through the same program." He said, "We can but we can't hire him, "because he's the CTO of Microsoft, so." That was, he was the forth guy. Joel, who I know, is going to be coming on theCUBE shortly. He also has joined us from IBM to run Marketing for us. So we've made some fantastic new hires. The company's doing really well. You know cloud certainly has played a big part in the second half of last year. I think it's going to play a big part. It's definitely going to play a big part in 2019. We've seen a pivot in pipeline, that's moved away from possibly even disaster recovery, data lake in the first half of last year. We pivoted to more of a reliable subscription revenue in the second half of the year. We announced some pretty big deals, big healthcare companies. We've got really good public reference with AMD. We announced a motor vehicle company one of the new used cases there is four petabytes of data per day they're generating. That all has to be moved from on-premise to cloud. So we've got some ginormous deals in pipeline. We'll see how they play out in the coming weeks and months. >> It's great to see the change, and certainly on theCUBE. We've been talking, I think we've known each other for almost, this is our tenth year. >> Yeah. Ever since we first met. It's fun to see how you guys entered the market at Hadoop, staying on the data wave and thinking enterprise, integrity of the data, active-active, the key I.P. And how cloud is just assumed data, and it's not just data, it's large scale. So if you look at the new people you hired, you've got jobs in large scale systems. >> Yep. >> We're talking about a large systems, now data is just given. So you're really nailing the large scale, moving from an enterprise nice feature, certainly table stakes for fault tolerance, and active-active. Just add recovery to mission critical >> Yep. >> Ingredient in large scale cloud. >> Well it's ironic isn't it because our value actually increases with the volume of data. So we're an unusual company in that context where the larger the data site, the greater the problem, and the greater the problem that we solve. See we made a pretty good bet, the active-active replication, that live data would be a critical component of both hybrid cloud and multi-cloud. And that's playing out I think really well for us. >> And certainly a lot more changes to come. Great to have you on. >> Yeah. >> Cloud and multi-cloud. Certainly cloud has proven the economics proven large scale value of moving at cloud speed but now you have multiple clouds. That's going to change the game on applications, work loads. It's not going to change the data equation. There's still more tsunami of data that's not stopping. >> Exactly. >> I think you've got a good wave you're riding. >> Yeah. >> Data cloud wave. David Richards, CEO of WANdisco here in CUBE Conversations here in Palo Alto. I'm John Furrier, thanks for watching. (upbeat instrumental music)