 Silicon Valley, Santa Clara Convention Center. I'm John Furrier, the founder of SiliconANGLE, and this is theCUBE, our flagship program. We go out to events, extract the signal from the noise, and I'm joining my co-host, Jeff Kelly, chief analyst at Wikibon's Big Data Practice. And our next guest is CUBE alumni, Roger Levy, who's the VP of marketing at SkySQL. Welcome back to theCUBE in a new role. Well, great, always, always a pleasure to be here. I'm also always amazed that you keep inviting me back, even after the prior performances. You did a great job at HP Cloud. You read HP, and we had many discussions. You had a great career and continue to have a great career. And Dave and I always joke, we love having you on because you bring a great veteran perspective while still in the trenches making things happen. So, you know, we always love to get that extract, that signal from your brain and share with the audience out there. So I want to get your take on a couple things. One is, give us a quick update on what SkySQL's doing and your role there, and a little bit about the company, because folks may or may not know what's going on with SkySQL. Great, I mean, I joined SkySQL just now. Oh, it's almost my ninth week. So, you know, really I know everything there is to know about the company. The overall direction of Sky is to be the commercializer of the enterprise grade distribution of MariaDB. MariaDB is the next generation of MySQL. It was created by the same person by Monty Widnes who did MySQL. He really wanted to continue the innovation and the community-driven, community-led development and innovation around relational databases. And then Sky was formed in association with that to essentially be the company that can bring that out in a commercial form. What we're doing today is a great week. We announced MariaDB 10, which is the community-produced version of the database on Monday. And simultaneous with that, as SkySQL or SkySQL, for those who like to spell things, we brought out what we call MariaDB Enterprise 2, which is our commercial enterprise grade distribution. So that includes a subscription for support, as well as a whole basket of other services and bundle tools that come with it and available from SkySQL on a yearly subscription basis. So give us a take on the cloud market. I see that HP Cloud, you were one of the key architects from scaling that up into an operation now, going a whole other level now. At HP, I see you got Amazon, you're familiar with what's going on the cloud market. Cloud is the innovation engine. You're hearing, that's mainly the, I'm called the printing press of this modern era. But still, you need to have enterprises come in and manage that. But what's under the hood of that engine of innovation is the data. And we coined the term data first here in the cube, some say mobile first, some say cloud first, but there's no doubt cloud is going to be part of the future. But under the hood, there's still some things that's work in progress. Like databases, like Flash, that Gary already seen on earlier. What's your take on the cloud market and the transformation happening at some of the levels that we're seeing in the databases, in the Flash, in some of the components? It's a great question. Actually, there's just so much interconnectivity between what's happening in cloud and what's happening in data management. One of the motivators for me personally to make the transition from running the HP public cloud into the database market is, quite frankly, it's all about data. Every time I would give a talk at the Executive Briefing Center at HP, we always had one slide about every day four exabytes of new data is being created. And while a lot of that data will never be looked at again, will never be managed, will never be searched, there is a huge amount of it that needs management and that creates tremendous opportunity for data management across all of its forms. Whether it's transactional, relational, whether it is what's being called NoSQL, whether it is big data in terms of Duke derivatives or data warehousing, there's just such tremendous need for efficient, effective data management solutions to help continue to drive the cloud. The cloud itself is providing great innovation. I mean, OpenStack, where I've spent a lot of time, I think is a tremendous example of hundreds of companies, thousands of developers all coming together in one large cooperative effort to move innovation forward in the cloud. And then companies like HP, companies like Rackspace and others building commercial offerings, building operational systems around that. And sort of one of the other, I think great notes is as we look at how we link these two, the Trove project, for instance, and OpenStack is currently based upon relational database technology, lots of opportunity for us to continue to drive innovation between both data and cloud in a cooperative way. Talk about the developer's market because you can't bounce around Silicon Valley or around the world these days without hearing about everyone wants to win the developer community. Absolutely. MySQL obviously has been a big part of the evolution. We talked with Gary Orson earlier about a quarter of the multi database environments that you're seeing out there have a lot of MySQL programmers, hundreds of thousands of developers live with MySQL all kind of wanting to do more. So SQL's not going anywhere. People are putting SQL on top of NoSQL to roll up analytics, et cetera, automating it dev ops, what do you want to call it? But there's an issue around oversaturating the developer market. What is the key developer issues that you're seeing out there right now for the innovation cycle that we're seeing? You're not seeing mobile, you're seeing app developers. Okay, that's cool. But really there's some stuff going on at the data level that's pretty strategic. Can you comment on that? Yeah, I mean I think a great example, and everyone talks about NoSQL, a lot of people look at the example of Mongo and there's a lesson to be learned there. In my opinion what has made MongoDB successful is really thinking about the core needs of the developer. Someone who's building a mobile app, someone who's building a web scale app and really coming at it from their perspective, providing a technology that provides the feature functionality but also makes it very much a part of what a developer is used to. Being able to communicate with a database in JSON, looking at it as more of a large data structure than needing to know schemas and other bits and pieces. And I think there's lessons for us to be learned within the relational market. As we've launched MariaDB 10, we've done a number of things now to start bringing more and more interoperability between the relational database space and the non-relational database space. We've introduced, for instance, what I'll call a gateway. It's actually called the Cassandra Engine but it's a gateway between MariaDB and Cassandra. So if you have a primary transaction-based application running in MariaDB but you happen to be using Cassandra for another part of your application, you can actually read and write to your Cassandra database through MariaDB using this gateway. So we're beginning to bring interoperability. The analog I love is we are bringing hybrid database. Just the way in HP we did a lot to really be pioneers in a concept of hybrid cloud between public and private. One of the things that we really see at SkySQL is a great opportunity and it really brings us back into providing the developers the full set of tools they need is being able to better allow the developer to use the right tool for the right thing. So talk about the connecting to the NoSQL. Obviously the news I've seen with MariaDB is to connect to NoSQL. That's pretty big. So explain why you guys see that as a big deal. Well, it is big. I mean, I think that for many people who go after a NoSQL solution, one of the things they learn quickly is there is an awful lot of work to do to make that function appropriately with your application if you need any form of real transaction. And so what we offer is we offer a hybrid capability where you can build a non-relational, non-structured capability within, for instance, Cassandra, build your transaction oriented part of your application in Maria and bridge them together through the gateway. And this really is the right tool for the right job. You don't use a wrench to try to hammer an L and although I've done that a few times, I think we all have, it doesn't give you the best results or finish on the wood that you're working on. And then within the database itself, we've introduced concepts such as dynamic columns. Classical relational database, the number of columns per row is fixed. But we now have a capability that allows you to have essentially a dynamic number of columns associated with the different rows. So you can put more unstructured data in directly into your MariaDB database. And really, just back to the more general concept of what the developers are looking for, there's a number of things. And MariaDB 10 I think is really going far to address them. One of them is scalability in HA. So many new features that have gone into the database, parallel slave replication that allows you to use multiple threads for schema, multi-source replication, and many, many other features that really get to this aspect of scalability. And last week there was a great announcement in the community from the combination of Google, Facebook, Twitter, LinkedIn around web scale SQL. And I think that puts to rest any questions about whether relational database technologies to scale. Because if you've got a scale issue greater than a Google or a Facebook, we've yet to meet those sorts of challenges. A billion user, million and millions and millions of row per second insert into a database that are being achieved on relational database technology today. So developers don't have to go and think about losing the benefits of ACID, losing the benefits of a transactional oriented database just for the sake of worrying about will they scale. Roger, is there kind of an unnecessary war going on in the developer community, no SQL versus SQL? I mean, are you finding that more education needs to happen? There's a lot of talk on Twitter and elsewhere and the social web about kind of the two camps. But it sounds like clearly it's about picking the right technology for the right job and mapping it to the business requirement. Is there, what are you sensing in the developer community? Is that tied, was there kind of this irrational exuberance around no SQL and then now it's kind of people kind of coming to the realization that actually it's gonna be a hybrid model going forward? I think it's the same thing we see around many technologies, right? There's the excitement of a new shiny object that's gonna solve all your problems and we have quite frankly, it solves a number of very specific things. But then when you really get down into the details you find, as with most technologies, the hard parts, the things that they didn't tell you when you went and committed and bought the thing in the first place. So I honestly think there's a war but there's a great need for education out there in terms of what are the appropriate use cases? What are the design principles around the different technology bases? Whether it's relational or unstructured technologies such as Mongo. So what are you doing to kind of keep your finger on the pulse of that developer community and kind of help with that job of educating developers in the wider community around when to use which technology and then also how you respond to some of their needs. You mentioned earlier, you've got to put the developer first. So how do you do that? How do you keep your finger on the pulse of what's important to them? So I mean we have as do many companies a significant number of developers ourselves that are out there in the community that are keeping in touch with what's going on across different database technologies. And quite frankly, we also do it through our customers, right? Many of our customers use multiple technologies and by working with them, by working with the developers, the engineers within those companies, we learn and they learn. And we continue to reach out to developer communities. Opportunities like Percona Live are great. In the talks and the tutorials we're doing here, we get to reach out to a very significant portion of the highly influential developer and engineering community around databases. So yeah, so talk a little bit more about the community around MariaDB. What is the state of that community today? And you're somewhat new to it having just joined us guys. Well, what are your initial impressions? My initial impressions is extremely vibrant. I mean, first of all, the real core of the community are people who have been working on MySQL for, you know, since it existed, right? It starts with Monty, who is the founder of MySQL and now the creator of MariaDB. So, you know, it's sort of the patriarch and, you know, 25, 30 plus of the key engineers who really created MySQL and have now taken their passion for database work into MariaDB. And then the extended community, you know, we're working with companies, you know, like Google, like Kaibo over in China, that, you know, are very, very active in, you know, the community and adding to it. And, you know, I contrast it to the Linux community, right? When right now, Linux is dominated by one or two, maybe three very, very large companies. This is still a community which has a strong vibrancy. I mean, for instance, we've just, you know, in MariaDB 10 introduced something called the spider engine which is for sharding. I mean, that was done by one individual in Japan, you know, not associated with Oracle, not associated with Sky, just someone who, you know, had a real interest in building this technology. And we're getting, you know, more and more and more of those types of people wanting a place to innovate. You know, after MySQL had been acquired and with some of the changes in policy and licensing that occurred, you know, some folks that had been very active in community and open source contribution and the databases, you know, really didn't have a place to go. Now with MariaDB, the MariaDB Foundation, you know, there's a whole new community here that people have begun to engage in. And I think it's, you know, it's really a start. You know, it's a relatively, you know, early stage but very, very, very encouraging and very promising. Roger, give us a quick update. I want to end the segment and get to the last word on the status of the company funding. Can you share some specifics? Sure. What are you guys trying to do? Your objectives, number of employees? Give us a quick data dump on that. Sure, I mean, you know, really, really simply, the company is roughly around 70, 75 employees. A very, very large number of those are engineers either in development or within our consulting and support organization. We just recently, you know, raised a 20 million funding round a couple of months ago from, you know, some very strong organizations such as Intel themselves, OpenOcean, which is a VC that funds heavily into open source-based companies, California Technology Ventures. So we've got some very, very good backing. We've got, you know, the money to go where we're really heading is having the best of the best. So our view is MariaDB will be the best of the best choice within the relational database market. We will bring in our own innovative contributions and we'll also seek out work that's going on within MySQL, work that's going on within WebScale SQL and other places and bring them all together, hardened, tested, enterprise grade and curated. And, you know, we're building our business model around the subscription business for that. Tell us, through prediction or future vision, and this is, I'm not going to hold it to you because it's a future prediction. Well, I'll hold it to you at the end of the day. But you've been around the block. You've seen many innovation cycles coming. Inflection point we're seeing now is clearly the key. Data first is the term we coined here in theCUBE. Mobile first is what's buzzed about. Cloud first, Microsoft CEO is talking about. But all the actions happening at the data, this data layer, what people are doing with data, how data is hand-built, how Facebook are using data, how Twitter and the, you know, WebScale SQL is showing that the companies that are using data as an asset, as oil, if you will, to power the generators of innovation. What's your vision? You've seen things come and go. What's your take out there around this data-centric, developer-focused, open-source collision? I think at the beginning it's all about data. Computing is a tool, data's asset. And I think we're just at the very beginnings of people realizing the incredible amount of power there is in data. I mean, we're seeing it now with the explosion in data analytics, data science, and what people are able to do for their businesses, for nonprofit organizations, for governments, by harnessing what's there. And the need then of having great solutions to allow you to collect, manipulate, store, and manage that really is where the future is. And I think database community, database vendors, really are at the center of all of that. And hence my excitement about what we are, our recent release of products, and the future for Sky, and quite frankly, for the industry as a whole. Talk about enterprise-grade cloud. How soon are we there? Are we there yet? Are we almost there? Is it early days? What's your take on that? And everyone's talking about enterprise-grade Amazon, certainly pumping that message hard. You've got Rackspace with OpenStack and HP, among others, driving that. IBM's got a cloud, got Cloud Foundry out there. What does enterprise-grade mean? Well, back at HP, we pretty much put a differentiator on our cloud as enterprise-grade. And that's really because the customers at HP serves were the top 2,000, end 1,000 companies in the world that demanded a high level of availability, a high level of reliability. And quite frankly, one of the other things enterprise-grade means is, what's the total customer experience? And that's really where we were putting a premium on at HP. I think the reality is, when you move past the emotion and all of the interesting marketing, and you look at Amazon, you look at HP, you look at Rackspace, most of the clouds, in many cases, are actually operating at or above what many enterprises are able to achieve themselves in their data center. Security issues in public clouds are well-publicized. Security issues in private data centers are not. Same thing about performance issues in downtime. And so it's kind of a one-way evaluation. The public clouds are very much out there. They're under constant scrutiny. And doing a good comparison between what's really available today and what people get in their private data centers are tough. But my own view is we are very much approaching the point where these clouds are at or surpass what you can do in your own data center. It's a software-led infrastructure, software-defined data center, software-defined clouds, software-defined applications. Data is the new asset. We're here with Roger Levy, VP of Product Management, SkySQL. This is theCUBE, our flagship program throughout the advanced extract, I see them from the noise. I'm John Furrier with Jeff Kelly. Dave Vellante couldn't make it on this trip. He's out down these coast flying around talking to customers and getting some stuff done, Jeff. So we miss you, Dave. Dave, if you're watching, we'll see you next week. We'll be right back with our next guest after this short break.