 from San Jose in the heart of Silicon Valley. It's theCUBE, covering Big Data SV 2016. Now your host, John Furrier and Peter Burris. Okay, welcome back everyone. We are here live in Silicon Valley at Big Data Week, Big Data SV, as part of Strata Hadoop. This is theCUBE's SiliconANGLE's flagship program. We go out to the events and extract the signal from noise. I'm joined by my co-host Peter Burris, our new head of research at SiliconANGLE Media, Wikibon, our next guest is Mitt Wally, who's the EVP and Chief Product Officer for Informatica. Welcome back to theCUBE, CUBE alumni, great to see you. Good to be here. So the world is shifting right on your doorstep right now. And you guys went private and we've talked at Informatica World. You're on Big Data NYC in 2015. You guys are a big presence here at the show. Certainly the vibe and the heart is pumping hard with Informatica right now in Big Data. But the world is moving right to the conversation we had. You guys saw this engagement day to the challenges of customers. What's the update? I mean, you guys see the same picture. What new data are you seeing in the market that you can share around the opportunity that your customers have? Sure, and I think we were just chatting. We had that, so, I mean, there are three things that are happening. We see that. One is clearly the explosion of data and we've talked about that. But concrete examples are, we obviously had transaction data and in the last five years social media and in the world of social has changed that to adding a ton more social data, like you and me have data around us. We are like walking data, right? Geolocation data, our Facebook data, our LinkedIn data. And so that's caused a huge explosion. And the third one that I think is just getting started is the whole IoT, the whole industrial data or what I said data from the sensors which is also tied to consumers and enterprises. So huge explosion of data. Second thing that's happening is underlying all of this changing types of data is the data persistence layer is getting fundamentally fragmented. That's happening at a rapid scale. SQL, Hadoop, you're talking about, no SQL, IoT, that's going to change persistence. And at the top layer, the applications are getting massively fragmented. There are 3,000 plus cloud enterprise apps. So what is happening is that the data layer becomes the clue, not the app or the persistence and that's what customers are coming to the realization, right? Lots of data, but bringing, having a data layer and having the understanding and intelligence there is going to be the key to the success. Software enabled by software. So if you believe that data is the glue, which by the way we do, we think it's an asset, Peter's got great research coming out on that. So if data is the glue, it's got to have software around. You mentioned middleware. What does that software picture look like and how do you look at that from a product perspective? Because is it open source? Is it something that you guys are adding value to? So how does that software wrap around the glue? Because you have to make it enabling. You have to deal with multiple data sources. It's got to be open to completeness which means foreign data or other data. Yep, yep, now I think you're right. So the thing is that look, nothing happens in isolation and nothing takes over. No new technology trend completely subsumes the old trend. I mean, we've all lived, mainframes were there and then client server, web and everything, everything gets tuned. Our view is, and I think that's what we see large customers that look, we've had traditional data warehouses, the discussion you were having and data is there and now you can create new data, you know, data lakes leveraging Hadoop. It's much more scalable, it's cheaper, it's easier to do, but they have to exist together. I mean, otherwise there's no value, you will create another island and connecting the two is not a trivial job, right? And so just to me, technologies coexist. When you leverage, now if you think about that problem you want to bring data in, there is a whole integration, ingestion, transforming issue which is non-trivial, connecting to all the sources that have existed all the way from mainframe days. Creating quality, I mean, with all the data coming in you don't want to make it a data swamp, you want to make it intelligent, you want to make it smart and then you want to secure and govern it, right? Security is a huge concern because you're going to bring all this data in and you're going to leave it as is and secure it. So our view is that a platform has to be, have integration, governance, security and mastering in it and it will coexist with the old and the new. It has to, it cannot be an island, a separate island in itself. Well, we don't need it to be, right? I mean, the goal here is try to leverage the expertise within the business as much as we leverage the tools. And so there's no doubt going to be a large number of individuals that, you know, cut their teeth in the data warehousing in other worlds that are going to find their way into this new world and can provide an enormous amount of value to business as business tries to bring greater operational disciplines to how they manage that data layer. And you're exactly right. I think the other thing that we need to help as you talked about, our job as technology leaders is to help take that skill set that has existed who understand the complexity of data and how can we take the same skill set of people and help them move to the world of Hadoop or big data, right? So we need to take away that abstract complexity of technology away. Technology is going to come and go. We cannot let technology come in the way of big data creating value and that's a responsibility at least we've taken upon ourselves. Big data, how do you convert it to big value, remove all of that complexity, connect the dots, let business create value rapidly. I mean, I want to ask you about your customer's customer. That's how you guys stay in business. You guys have made some big bets. Going private was one of them, but outside of that product investments and also the business model of Informatica. What is the impact that is having to your customers? And if you could share some color around recent wins, some value you guys have provided and then what you guys are doing here at this event to continue that. Yeah, no, so we have a huge presence here. Big data is a huge vector of growth for us. Obviously, traditional data we invested in big data before the curve we've invested in cloud and cloud and big data lines are also blurring as we see. So we have a huge presence at Strada. In fact, tomorrow I'll be hosting a exact panel of fireside chat with our customers and our partners. We have Trans America is a big customer of ours. We'll have US Bank, Western Union is another customer of ours. They'll be there. We'll have obviously Cognizant as a partner. We'll have some analysts. We'll talk about what these customers are doing to solve real business problems. I'll give you a great example. Take the example of fraud and detection. That's a huge problem in the world of leveraging big data. Western Union is a customer of ours. 200 countries, they do wire transfers. It's a huge fraud detection problem for them. They're leveraging obviously underlying cloud era and they leverage Informatica to bring that data together and make real time decisions. You know, Trans America. I mean, they are looking at you and me and all of our data. They have traditional data. They want to marry with all other external data. Huge big data problem and be able to service their customer better. So these are the kind of use cases that our customers are solving. I love interviewing you. I'll tell you why, because you're a product guy and that's the hardest job right now to be a product executive because it's a moving train relative to specs and requirements. You've got waterfall, development changes to agile DevOps. These are the kind of things we always talk about, but you mentioned the fragmentation of the apps. That's a huge driver right now in the landscape. Does that affect your business and how has that changed the customers? Because you can also argue that pre-package apps and the analytics is going to be very domain specific. So it's hard to kind of render a broad product market opportunity for, say, jamming in a feature and that's different in thousands of different use cases. So how do you look at that? You mentioned data is the goal. I want to expand on that. How do you look at that fragmentation as an opportunity, a technical challenge, both? Actually it's a huge opportunity for us. The more the fragmentation, the better it is. And that's where the realization of having a data layer becomes important to customers and which is why we invested in intelligent data platform. 4,000 apps, lots of persistence across the board. Customers can't rely on one app today. See, 20 years ago, I lived in the world of one app, one database, can't do that anymore, right? So, and customers want the ability to pick and choose apps. If I want to do something today, two years from down the line, I can go to another app, just like we use it on the iPhone. But I want to carry my data wherever I go. So they want to decouple data from the apps and the persistence layer and be able to do all kinds of machine learning, all kinds of analytics on that data, irrespective of whichever app and persistence they have. What's the impact of the developer? Because now a lot of the business people, the CXOs or business people, are making huge investments in developers. And then I'll see on the infrastructure side, you got the geeks who are like figuring out, okay, how do I make the infrastructure, DevOps work, infrastructure as code? How do you guys balance the two trends of DevOps, making scalable infrastructure, which might not be in the interest of, say, the developer on the front end? Yeah, so I think there are two different ecosystems. When we look at the large enterprise customers, right? So in their case, obviously, they are not necessarily developer oriented. But the developer there is the guy in IT who wants to experiment and figure out whether something works. But very quickly, where they've realized is that, boy, that experiment can work in its own isolation, the discussion you were having, scaling it is not easy. And it has to connect with what we got today. So that's where they turn to us and say, look, boy, experiment works, help us scale it out, help us make it connect to the enterprise. The rest of the business. That's the business, because that's where value gets created. So we benefit from that. Let me build on this question, because every single significant transformation within the technology industry, I think we're in the midst of one right now, has been accompanied by the emergence of a developer ecosystem that created value for the business. One of the things that's interesting about what you're saying, and I agree with it, is that as we start to separate that notion of the control layer and the data layer, that the traditional way of thinking about how we unlock value, which was through the applications, is now going to see how do we liberate the data? What does that do to development? Are we going to see a different class of ecosystem? Are developers going to have to learn how to do things differently? What do you think? I think the opportunity for developers there is that once data gets decoupled, actually, in some ways, that's democratization of data, right? Now, the developers can create, in my opinion, true value-added applications for the enterprise. To me, that is democratization of application development. Now, if I'm a developer sitting at my home, I can actually create a true app because I can link to a separate data layer versus data set in a monolithic app that I had no way to go in. So my thinking is that that'll create a lot more innovation once the data gets decoupled. I mean, I want to get your thoughts on three things, three trends that we're seeing in theCUBE, and Peter's got a set of research that's going to be coming out with George Gilbert. It's going to go in much more detail around this. But three things I want to get your reaction to. One, completeness. Two, integration with something innovative, whether it's experimental to production, so integration, and then hybrid cloud. Number one, completeness. What does that mean for this world? Because that's the holy grail, to get to some level of data completeness. Where are we with that your thoughts on what that means? You know, I mean, that's always a journey. But I think completeness to me is where customers think about completeness is, take the context of Hadoop. I want to create some level of Hadoop Lake or whatever people want to call, but I want that to be connected to everything I have so I can bring data from there into that for me to do something special, whether it's analytics, whether it's fraud detection. So being able to be connected to everything, that to me is the first degree of completeness. Second is, can you create some kind of analytics or intelligence on top? Whether it's metadata driven intelligence, so you can understand, you have some kind of machine learning baked into it. So once I have that level of completeness of data, what does the data tell me? It's a lot of data. I mean, you can't really look it. You have to have pattern matching. So metadata and machine learning comes into to make sure that the complete data tells you something. That's going to always be iterating. It's always going to be iterating, right? Because think about this one. Today we have social media apps like Facebook, Twitter, LinkedIn and all that stuff. Tomorrow there'll be five more. Today Snapchat is more of a teenage kind of app. Who knows what kind of app goes out there that you want to link in to get data into your lake to get some more intelligence on. Let's say John, you. The crowd chat thing is exploding. So, so. You guys are going wild last night, by the way, I have to say. I appreciate that. That was good, good chat on the crowd. Go to crowdchat.net slash strata. Do if you want to join the conversation. Okay, word two, integration. This is, seems to be where the bar is now to get into the enterprise. Eat some table stakes and it's not always clear for startups to get into the enterprise because this huge integration needs to production. What's your thoughts on today's market of integration and the challenges there? Well, I mean we've lived that market from day one so we understand that. I think that's where the difficult infrastructure software work happens and people kind of gloss over it. But think about that. We go back or companies have to go back to the enterprise of 30 years ago. Mainframe, DB2, IMS, Sybase, blah, blah. Keep going, right? And you have to connect to all that stuff to be able to bring all that stuff into some place. And that's not easy to do. You can connect to the one last thing and you feel like you have some connections and that's where enterprises to create value look at solutions that have that breadth. That ties to completeness. Okay, final thing, hybrid cloud. Is that the engine? Is that the compute? What's the role of hybrid cloud or cloud in general for this data world that we're living in? We live in a hybrid world today. I think people don't realize it. And to me hybrid, by the way, is two definitions. Hybrid is on-prem in cloud. Cloud to cloud. I mean, we're in a hybrid cloud world. I mean, so I think hybrid has many meanings and I think we are in it. We were in the world of hybrid infrastructure on-prem. Cloud is basically make that a different kind of hybrid. We were mainframe client server. Now we have on-prem, cloud, cloud, cloud. So the hybrid world is here. We live in it. And I think anybody who does not realize that is basically passing. Yeah, they're going to be driftwood and it's going to get in front of that next wave is Pat Gelsing would say, okay, let's talk about how you're partnering because the ecosystem, as Peter was talking about, community-driven opportunity. Informatica has a partnering strategy. What is the update there? Can you share some of the strategies and successes of partnering? Because you have to partner. The apps are fragmented, there's no doubt, but there's developers to deliver it. I mean, we've always said this. We're the Switzerland. We're the Switzerland of data. We absolutely partner with everybody. Even if you say we've partnered with every persistence provider, right? All the way, Oracle, IBM, in the current world, cloud era, Hortonworks. We partner with everybody on the persistence layer. We partner with everybody on the app side, whether it was the SAP apps, Oracle apps, Salesforce.com in the cloud world. We partner with the AWS's, the Azure's of the world. Very good partners with them. So our belief is if we have to create value for our customers, we gotta be that Switzerland and connect anything to everything. And we've lived with that. And that strategy remains double down and every release you'll see we connect to everything. Whatever comes new or whatever is there in the old, we connect to that. So what are you guys announcing at the show here? Because obviously the things you've been mentioning, I see you guys are global. You mentioned earlier making things more intelligent. That it seems to be a theme of your entire product strategy and platform. What's being announced here at Big Data Week at Strata Hadoop and Big Data SV? So think of it three things. One is we obviously have announced the fall last year, walking into this year, the complete Big Data platform for IT to invest in integration, quality and governance and security. Because we believe in the world of Big Data. It's not just about integration. You gotta have look at governance and security as a part of an infrastructure stack. Second is bring that platform together from a metadata point of view. So basically you can not only bring our platform together, but it connects to everything else within your enterprise. But the most important is when we say Big Data, Big Value, we are basically taking it to the business user. So we just announced a brand new product called the Intelligent Data Lake. By the way catered completely for the business. You log in, it's an Amazonian experience, right? And a Facebook kind of experience. But behind the scenes and you can search and do any kind of analytics. We have our metadata is the Google for the enterprise. But behind that, full governance for IT. So as IT, now I feel very comfortable giving access to hundreds of business users to this lake. Because I know who's there, what data are they touching at any given point in time. I have governance, but business is liberated. So Intelligent Data Lake is a big thing we've announced here. Strata allowing business users to partner with IT and get value. Well I want to say congratulations. It's been really fun to watch Informatica really transform and really got a nice group swing going on. The products, the presence, certainly the vision. I think you guys are right on the right fault lines of this disruptive tectonic plates that are going on. And thanks for the insights that you shared here. Before I leave, this is the 10th year of Hadoop. So it's a happy birthday Hadoop in a way. I think we've seen how much this journey has transpired. Hadoop has come to now truly being able to create value for the enterprise. 10 years, you're just seeing what's happened pre 10 years ago. Facebook, I mean Twitter just celebrated our 10 year anniversaries. So it's amazing. So I want to get your thoughts kind of on it. Take your Informatica hat off. Put your experienced business manager and entrepreneur hat on. A lot of people out there trying to figure out which side of the street to land on old way, new ways. Obviously that's happening. Whether it's a startup, entrepreneur, there's a ton of opportunity, but there's a lot of noise. What advice would you share with folks out there? Being a product guy. Again, product market fit right now is the critical factor of most things that are going on on the invention and innovation side. Thoughts on and advice for folks out there who are really committed to delivering some value, come into market with new ways to do things, whether it's a horizontally scalable software glue layer at the data layer or new product that they want to launch or company. I think some things don't change, but I would say number one is build the best product. Products win. I mean we know product, product, build the best product, get the product market fit very quickly, iterate. I think we fall into that trap and we learn every day and we say look, let's not try to over engineer a product. If you've got the right idea, build something, get it in the hand of a customer, let the customer play with it, and the customer give you feedback. And that customer driven feedback, I think doesn't matter a large enterprise or your startup. You've got to basically go customer driven feedback. And you recommend agile development? Oh, absolutely. 100%. I mean we do releases every quarter now. For big data world, every quarter, cloud we do every quarter. I mean we do releases every month, but we roll it up in a quarter because customers obviously can't really make changes every month in the enterprise. How's that going for you guys? I mean it's great. I see from an innovation point of view, it's great internally also. See our engineers, obviously the people we've had, they want to innovate rapidly, right? They don't want to go slow. Customers like it because they can pick, they also get a much faster innovation. So we like it, customers like it. So on the new curves, it's like quarterly we do a release now. It's good to be a product guy right now, isn't it? It's challenging but yet fun. I mean technology, right? It's changing in front of us and it's creating value. And that's what at least as a product guy I get excited about. Not science projects are great, but how can we create value for our customers? That at least where we focus on. Really appreciate the CUBE insights here, Informatica on the CUBE. And remember we're going to Dublin for Hadoop Summit. So check out the CUBE, we'll be there. We're going to be getting on the plane going to Ireland. We'll be toasting a few pints of Guinness out there with the Hadoop community. This is the CUBE here. We'll be right back with more after the short break.