 to the world. This is Stu Miniman. Thanks for watching theCUBE. Hi, I'm John Furrier. Thanks for watching theCUBE. Live from San Francisco, it's theCUBE covering Oracle OpenWorld 2016. Brought to you by Oracle. Now, here's your host, John Furrier and Peter Burris. Hey, welcome back everyone. We are here live at Oracle OpenWorld 2016. This is SiliconANGLE Media's theCUBE, our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, the co-CEO of SiliconANGLE with Peter Burris, head of research for SiliconANGLE Media as well as general manager of Wikibon Research. Our next guest, I'm excited to have back because he's a product guy and we'd love to go deep in the process. Cube alumni Juan Luisa, Senior Vice President of Database Technologies. Veteran of Oracle, welcome back to theCUBE. Great to see you. Thanks, great to be here. Love talking to the product guys in development side because we get to go deep into the roadmap and we're going to try to get as much information out of you as possible. But you'll do your best to hold back like you did last year. I'm only kidding. I'd never hold back. Okay, no. You must have me confused with some of you guys. Maybe that was Larry Ellsmore. He hasn't been on yet. Larry, we'll get you on. No, he's not so good at holding back either. That's why we want him on. That's why they won't let him on, I think. That's, Larry would be too comfortable in theCUBE. No, in all seriousness, joking aside, the hottest areas right now is in your wheelhouse. Engineered systems, which is going to be a real enabler for Oracle on the performance side. And as you make your own chips and you get Spark and Exadata and all this other cool stuff, it's going to go faster, faster, faster, lower cost, higher performance. The database, security, I mean, amazing stuff. But the database is where the crown jewel is for Oracle. Always has been before you put web logic on it, make it sticky, but now you get the cloud. The cloud is a environment for great opportunity for the database, business and other databases. Some Oracle, not Oracle. What's going on with the database and the cloud? Can you take a minute to explain the current situation? Yeah, so that's a big question. What's going on? So we're going to start with database, or you want to start with cloud? Let's start with the database? What's going on with the database? And what does that mean for customers? Has it moved to the cloud? Yeah, so database, we're in the process of releasing our next big database release. We don't release database releases very often. It only really happens every few years. It's a very big deal. So what we're trying to do with our next generation database is modernize the whole infrastructure just to a lot of the big transformations that are happening in the marketplace. So among those are things like big data. Where do we go with big data? So with our new generation database, we're making big data and database work seamlessly together. So we have something called big data SQL where you can query data regardless of whether it's in Hadoop, no SQL Oracle, it's completely transparent. So customers no longer have these silos of information. Another big thing in database is the data types are changing. So new generation wants JSON. It's called JSON, which is the new data format which was used in JavaScript. So web developers develop in JavaScript. They represent data in JSON. And then they say, hey, I don't want to take my JSON data and convert it to relational data. It's a big pain. So one of the things we've done in our new generation tools of database is we said, hey, take that JSON. We'll put that directly in the database. We'll allow it to be queried. We'll make it highly available. Highly available to our... Without a scheme, without any kind of scheme. Just throw it in the unstructured. Just throw it in there. That's right. That's right. So we've made it very simple for new age developers to use JSON with databases. That's another really big thing that's happening. So tell us what, just to double down on that for a second. JSON has been a big trend in API based systems. A lot of people using JSON endpoints for user experience, whether it's mobile or web, very prevalent now, pretty much standard. Yes. How does that get rendered itself from a customer's perspective? Are you saying that Oracle will just onboard it into the database itself? Or is it a separate product? Or is it, I mean... Directly in the data. So we have native JSON directly in the data. We've essentially added JSON as a data type. We've added it to SQL. We have SQL extensions. You can access... So I can run SQL queries on JSON. You can, exactly right. You can very simply write SQL queries on JSON. And what's the impact of the customer? All the stuff that comes with that. And what does that solve? What problem does that solve? It solves two problems. One is, people like that data type. So new age developers, they're writing in JavaScript. They have JSON and they just want to use it. So they don't have to convert. Which by the way, everyone's running in JavaScript. That's right. That's the big programming language. And the other big thing is unstructured data. So data that's not structured initially, that every piece of data has its own structure. So it's a representation for saying a dynamic unstructured representation that's very standard in the industry. It's a little bit like XML used to be before. JSON is kind of the new XML, the new age XML. How about the data lake concept? Because Hadoop as a market, didn't make it, right? I mean, Hadoop is out there. Yes, Spark is certainly relevant because you have that kind of use case of memory and faster processing. But the real power is that that's a batch oriented data set. As things like Hadoop and Spark evolve, how does that relate to Oracle's product roadmap? Yeah, so we have our own Hadoop as big data appliance where we run a Pladera-based Hadoop product. And what we're trying to do is make those work seamlessly with existing databases. So there's certain kinds of workloads and applications that Hadoop is really good for. Kind of a frivolous example is if you want to find cats in pictures, you're not going to do that with an Oracle database. So here's a billion pictures, find all the pictures that contain cats. That's not a good application for Oracle, right? On the other hand, if you're running analytic queries against relational data, that's perfect for Oracle. So we see that these technologies can coexist. So there's certain kinds of applications that are really good for the Hadoop kind of workload and certain kinds of applications that are really good for relational. And what we need to do is make sure that these things work seamlessly. What's the glue between those two layers? Well, that's it. There's even more applications that are going to want to use both. That's right. That's right. So eventually everybody goes to both, right? Yeah, so what is that glue? Well, there's a number of glues that we built, which is one is called big data SQL, it's a query seamlessly across them. We also have connectors that let you move data seamlessly between them. So those are kind of the main glues between them. So one of the things that we've observed is that to John's point, there's been a lot more downloads of Hadoop than we've seen go into production. It's become a very, very complex ecosystem and it's got some limitations, batch oriented, et cetera. The challenge that businesses have is they try to run pilots around Hadoop as they find themselves piloting the hardware, Hadoop, the clusters, all the way up to the use case. And a lot of the times they end up failing. How does something like the big data clients facilitate piloting? Because it looks like it should reduce the complexity of the infrastructure and give people an opportunity to spend more time on the use case. And you've got it exactly right, which is, there's some people that are hobbyists, right? Like there's people that wanna build their own log cabin. They wanna cut their own trees, kind of build their own blanks and put together their log cabin. And that's kind of how Hadoop started. It was kind of the hobbyist model, right? And Hadoop has kind of moved to the next level. Now it's people that wanna get stuff done and it's like, I don't wanna chop trees. I wanna live in, just give me a house, okay? Actually, I wouldn't say hobbyists. I mean, y'all who had a need, they needed log cabins. Right. So they built from, you know, but it was a use case, the web scaler guys needed it unstructured. But a lot of people were very much kind of in to build your own. So now a lot of people want a solution. They're like, I don't wanna be building this. So that's where big data planes come in. Because it's a complete solution, it includes the hardware, it's been pre-tuned, pre-optimized, it includes the cloud data software, it includes all our connectors, and it includes support for the whole thing. Because that's the other part of it. When you put together your own house, who are you gonna call when it leaks? Right? You're on your own when it leaks, right? If Oracle puts it together, we can support the entire stack when you have any kind of issue, any kind of problem. And that's the kind of stuff enterprises want. It's not a hobby anymore, once it becomes an enterprise. So given that we're in a big data universe right now, where we've got use cases that are proliferating very fast and we have limited experience about them, that the technologies underlying that we're deploying to build those use cases are also proliferating very fast, is it gonna be possible for the open source model that presumes download, try by, not salespeople, not a lot of learning, not a lot of hand holding to make it possible to fix that whole thing or make it all come together? Or is a company like Oracle gonna have to step in and take some responsibility for guiding how the market evolves? Yeah, so open source and Oracle can work together. I mean, we have Linux distributions, we own MySQL, so Oracle and open source. You're not a nods. That's right. In fact, we're one of the major open source companies in the world. But like I said, real businesses aren't in it as a hobby. They want a solution. They're looking at this as a tool and a lot of times they want somebody that can support it, that can basically ensure that it's gonna work for them and they have somebody that can call. It's not just, hey, I'm gonna post a message on a message board and hope that somebody responds. I mean, when you have airplanes in the air, when you have dollars flying across the network, you need a solution. You need somebody you can call that you can guarantee it's gonna solve the problem. And also that can ensure that the technology moves in the right direction, takes into account what users want and that the certain level of quality assurance and all that is built into it. So let's build on that. When you look at the future of database, what do you see? Well, there's a lot of different things. So database is in a very big change. Big change is happening in the database world right now. More than probably ever before. So one thing we've been talking a lot about is this sort of big data Hadoop. Another thing is JSON. Another area is in memory. It's a very big change that's happening in database. The whole moving into in memory into these different kinds of formats. Along with that, Oracle's pioneering, moving database algorithms directly into the chips, into the chip technology to make it run dramatically faster, make it more available, make it more secure. That's another big thing. Building multi-tenancy directly into the database. That's another big area that Oracle's pioneering. Instead of having it kind of cloudify the database directly natively inside the database. Another big area that we've been working on is putting native sharding of databases directly into the database. How about data protection? Well, that's a multi-tenancy, right? Take me through the multi-tenancy a little bit. How is multi-tenancy inside the database going to work? Well, okay, so that's what we call our multi-tenant database. A little bit like VM. So VMs say, hey, it looks like I have a physical machine, but in fact I have a fraction machine. It looks to me like a physical machine. In fact, it's a virtual machine. We're doing the same kind of thing with the database. So it looks like I have a physical database to the application, but in fact, you're sharing a database among many users. So what is the advantage of that? The advantage of that is we don't have one database, thousands of databases anymore. So many of our customers have deployed thousands of databases. It becomes a huge maintenance headache to have thousands of databases, especially in today's security world where you have to constantly patch and update these things. You can't just kind of leave them alone anymore. So I have a small number of physical databases, lots of virtual databases that completely saves costs. It's more agile, OPEX lower, CAPEX lower. That's the new world of multi-tenant cloud databases. Also it's friendly with appliances, and I want to give you thoughts on, last year, the big rage that I liked was this zero data loss recovery plans, CDLRA. That's right, you've got it right. What's the, I mean, no, I'm very fascinated, basically zero data loss technology. So is that still out there? What's going on with that? Zero data loss recovery plans is our fastest growing of plans right now. It is. Yes, easily. It's been very well received by the market. We have some of the biggest banks now running it, financial institutions, retailers. Why? Because it's a very simple value proposition, which is, hey, I want to protect my data in a way that it's constantly protected, I don't lose any data. In a way that it's scalable, in a way that it offloads my production databases. So it's a very simple situation. So that's a breach saving situation, right? So like, the people who have these security breaches is where that fits, where's the use case for CDLRA? CDLRA is not security, it's about availability. Okay, so if someone basically shuts the data center down. Right, if that database is corrupted, if there's some natural disaster, if there's a bomb, if there's whatever, is my data protected? Will I lose anything? Nobody can afford to lose data anymore. In the old days, when you did a backup, you did a nightly backup, and then if something happened, you'd restore. Well, guess what? That doesn't work anymore. We're too dependent. So nobody wants to lose their airline records, nobody wants to lose their bank records, nobody wants to lose their retail records. You can't afford to lose data anymore. Okay, I want to get- So you need a solution that's zero data loss. I'm surprised you guys aren't, there's not more fanfare in the show about that. I was really impressed last year, I'm glad to hear it's doing well. Containers, database containers. This is something that we talked about a little bit last time. That's the same as multi-tenants. It's kind of a different terminology for that. All right, now cloud-based databases, now we'll get to the cloud. Where does all this go to the cloud? Okay, so, you know, traditionally customers deploy on-premises. What we're doing now is we're taking the Oracle database that we've developed for the last 40 years. It's the most sophisticated database in the world, and moving it onto the cloud. So what does a customer get? They get, they can provision it instantly. So you go onto our website, say, I want a database, here's the size, here's the number of CPUs I want, boom, they get it. They pay monthly instead of paying up front. They don't pay for the licenses, they just pay us a monthly fee. And then Oracle operates the whole thing. It's like, I don't want to manage it, I just want to use it, right? So that's the benefit of the cloud. I go somewhere, I need a database, I get it right away, I don't have to mess with it, and I pay monthly. So the Oracle, on your Oracle cloud, you would then deploy all those other goodness, ZDLRA, all the other technology behind the curtain, so to speak, was it a box? So we have a range of offerings in our cloud. So we have a regular database service, we call it enterprise service, and then we have a high end service, the Exadata cloud service, right? That runs our Exadata technology, super fast, super available, and then we have something called Exadata Express, which is the lowest cost cloud database in the world. So we have kind of three things, depending on what the customer wants. They want a smaller database for really low cost, they want a super mission critical, high performance database, or they kind of want something in the middle. So we span the whole range, and by the way, our high end is higher than anybody else, our low end is lower cost than anybody else. So we span a bigger range than anyone else. You know, Juan, next year, we're going to need an hour with you. Yes. To cover all the tech. It's a lot of topics. You're a great guest, and you have a lot of experience, and we appreciate the insight. I'll give you the final word, I want to get one more answer out of you, because you're awesome, Sharon, great insight. For the folks watching, what's the one thing or one or two, three things they should know about Oracle Cloud, the technology, the database, the things going on at Oracle that they may not be hearing about, could be the best selling things, some things that's not on the mainstream, press reporting. Well, you know, our Oracle Cloud is pretty simple. I mean, the main thing to understand is, it's 100% compatible with databases on premises. So it's very easy to move workloads back and forth. That's the main thing. And the other thing is, we use the exact same infrastructure. So we've been developing, for example, our Exadata product, which is kind of the precursor to Cloud. It's a very specialized database system run on premises, and now we're running that in the Cloud. So again, the customer can get the exact same thing. And our latest offering is Cloud at Customer. So we take those same Cloud attributes, and we can put them inside the customer data. So we have a Cloud machine, an Exadata Cloud machine, and a big data Cloud machine. So customers got all the choices of Oracle. That's right. So the customer has full choice. They can move to the Cloud if and when they want, at the speed they want. They can move back and forth. They can do disaster recovery in the Cloud. They can do backup in the Cloud. They can do development in the Cloud. So all these range of offerings, all these range of choices are now the customers. So true or false, Larry Ellison's the master at the long game. Larry thinks long term, absolutely. Of course, true. And yes, absolutely. I mean, he's brilliant, and he's shown it over and over again. I agree. Big fan. Although yesterday's keynote, Larry could've done better. But he was too busy getting all those announcements out that he was mailing in at the end. There was so many announcements. It's hard these days because Oracle, there's so much happening at Oracle. There's so much happening at Oracle. Juan, thanks so much for spending your valuable time with us on theCUBE. We really appreciate it. This is SiliconANGLE Media's theCUBE. We go out to the events, extract the silhouettes. I'm John Furrier. I'm the way it's anybody's president, database platform services, live in San Francisco. We'll be right back.