 Live from Las Vegas, Nevada, extracting the signal from the noise. It's theCUBE, covering Informatica World 2015. Brought to you by Informatica World. Okay, welcome back everyone. We are live in Las Vegas for theCUBE, SiliconANGLE's flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, the founder of SiliconANGLE. I'm joined by co-host for the segment, George Gilbert, Wikibon analyst, covering the big data space. And our next guest is Amit Walia, SVP GM of the data integration security group. I'm Amit from Atica, welcome to theCUBE. Thank you, well, good to be here. So, SVP GM, data integration and security. So you've got big data, big data analytics, and security, I know they're tied together, but let's start with big data analytics. This show here, you guys have a nice platform. You're not database dependent. So you have a large customer base. What is the key thing that you're seeing on the big data side of it, in terms of being put into use for the analytics and for your customers? That's a great question. So much is happening in the world of big data, right? I mean, in some ways I would say we just got started because so far the big data story has just been, can I just experiment? Can I get started? Can I just put it into some use to use cheaper processing? But the real value of big data is how do you do more predictive real-time analytics, right? We talked about this this morning when I demoed Project Sonoma. And our story there is, look, we have this intelligent data platform that first of all, as you said, is database agnostic. In fact, I would say in the new world, data agnostic, whether it's structured data, unstructured data, machine data, streaming data, social data, whatever data, we can bring it into any kind of big data lake, as we call it, lake, hub, reservoir, whatever terms customer like to use, lake is a very commonly used term. So you can bring any kind of data into the lake. But that's just the first step. Then you want to do analytics. But just to stop you at that point, machine readable or machine extracting structure from the data lake is tough, generally. Data, so far we've used a lot of data wrangling tools to do that. And then once they're structured, the machine takes over. How do you manage that doing it the other way around? Well, see, that's the power of Informatica. We take data from, let's take any sensor. Let's take, when you're using this laptop to watch any streaming movie. We take that machine data, which is completely unstructured, semi-structured. And then we can parse the logs to see, locate the name, the time, the location. And we can convert that unstructured, a semi-structured data to a structured data, which once you then put it into any kind of lake, it's easier to do SQL queries, easier to bring it together, join it with existing customer data. So then you can do meaningful analytics. That's what we do with machine data. So you're leveraging the earlier Informatica technology for extracting structure out of the data? Yeah. I mean, we are actually extending our technology. So when we announce the intelligent data platform, that's what it does. So to us, any kind of data, we're agnostic. We pick any kind of data and we, so customers don't have to worry about the type of data. We can bring that data, put structure to it, and then put it in any place they want to put it, to then do analytics. And then we take it a step forward. The concern we hear from a lot of business users today is that there is so much data, petabytes, zettabytes of data. How do I even look for the right data? And that's where we talked about Project Sonoma today, that we can bring all the data in. But then your ability then to truly search, and we, you know, enterprise-wide, not just a direct search, a semantic search, where we are inferring what the user wants to do in a very Amazonian way, recommending data to those customers. Bringing that to the data analyst is powerful. Talk about the real-time piece, because you know, there's two challenges. You just mentioned one. What to look for. And again, if you know, we were saying earlier on theCUBE that the thesis, if you believe that everything's connected together now, then everything in thesis, in theory, could be measured. If that's the case, then there's going to be a five-nines reliability of data available, so there should be 100% accountability for all actions. Now that's kind of a dream state, but that means for managers to know what to look for. So knowing what to look for is one thing. But then actually having real-time value is another challenge. How do companies balance that? Because they are moving fast to real-time with the apps, there's more data, geo data coming in. And how do they move their infrastructure and transform it fast to get access to the data and make it real-time? Yeah, I think that's a great question. Take an example where you want to do real-time offers, you want to know, so you want to know what your customers are doing real-time. So with the ability for us to bring all kinds of data within the lake, then with the ability of our project, Sonoma, where we have the library of map technology, we can bring the power of all that data. We can have the analyst or the scientist be able to search that data to do some real-time analysis. So if you're worried about, gee, what is my best customers doing today this week? So I can give them certain offer going forward so I don't lose them. You can bring real-time data and bring all the data together in the lake and allow an analyst to kind of then search for it, bring it together, do some real meaningful analytics, and then you can put the next best offer in front of your customers, whether it's defensive or it's often. I mean, whichever kind of offer you want to put in, but that's real-time. But that's in the iterative cycle because you're always bringing more data in for richer context, and then you're presumably operationalizing it by putting it into that model, which then works with your application without human intervention. Meaning, your analyst is always finding new data and new structure in the data, and then that improves the model, and the model works automatically. That's the real context. So after the analyst has done something, then we can operationalize that, because then you can put it in real cadence, and then the power of machine learning comes into play. Then basically, as you put those models in, they keep learning by all data that we have already connected to it coming in, and you learn, and you keep getting better, and then you can throw recommendations back to the data scientists or the data analysts, things that human mind cannot do, but computing power can take over. That's where machine learning comes in. Talk about context and contextual relevance. I mean, go back to the old internet infrastructure search days. You got behavioral and contextual. We're always the two killer variables. But now you got context. Talk about engagement, age of engagement. So those are the new concepts we're hearing here. Age of engagement is now ushering in the new era from productivity to engagement. What is engagement data, and what does it mean towards context? Because that threatens the real time piece. That's a great question. Think about it this way. You and me as consumers, every company has our data, if you are a customer of a bank, they have all the transactional data on you, your name, your social security number, the balance you carry. But in the world of engagement, there is a lot more about you that they would like to know. You're in Las Vegas today. You are hanging out today. You want to do something today evening. You just tweeted that, hey, let's bunch of us available, let's go somewhere. All that data is social data, the world of engagement, engagement data, that if they had access to, then they can say, gee, I know John. I know his spending trends. What if I put him in offer and say, look, if you're a pick your favorite bank, a customer, here is a great offer for you to go to the Steakhouse. Or the Steakhouse can leverage that data and say, how do I draw those customers in? So in the world of engagement, context is a lot, there is a lot more to context, where you are, what you're doing, what you're saying. If you marry that with transactional data, you just had a lot more richer analytics. But you never, you never stop adding context. You never do, you're right. But the thing is, that's where the power of also a data platform comes into play, right? We all want to add a lot more context, which means a lot more volume of data, which means a lot more variety of data. So you need to have a platform that allows you to do that without worrying about, gee, what if my skill sets go away? What if a new data set comes into play? You don't have to worry about that. We provide that abstraction layer so that bring in anything. Worry about the business, don't worry about the data. So I've got to ask you about the industry because we love this moment in time, it's so much change. I mean, we're talking about data like, it's like the computer revolution. When you go back 10 years, only the data geeks would go crazy like us. But it's a great opportunity. Talk about from your experience, why is this moment in time so important? Why is the shift and inflection point happening? And why is the cloud, mobile, social, where now we're talking about engagement data that's just going to be game changing? What is it about this time? So a lot of people haven't had the industry experience that you've had and seen the cycles. Why is it so powerful right now? So you made a good point. We are in the middle of the biggest change in the world of IT. So not only computing is changing from on-prem to cloud, but then there is social and mobile data that's available for you, again in the cloud. Then you also have machine and sensor data available to you. So much different types of data, on-prem and in the cloud available to you. And that what is calling the two things that are happening. First of all, a computing change is happening. Number two, data change is happening. Those two changes didn't happen at the same time ever before. This is the first time that those two changes are happening at the same time. Now, the third one is, the business value creation timeline has shrunk. Businesses are very much just in time. So with these two changes, and businesses wanting to drive value on a real-time basis, it's a massive inflation point. Every theater of innovation is exploding. You've got the technical theaters, you've got the business theater, you've got the social theater. And so like this is like the management side is also changing too. So transformation, that's the big buzzword, digital transformation. I love the buzzword, I mean I love everything digital, but like, what does it mean? I mean what does a food customer, I mean they don't say, hey I'm transforming digitally, they just are kind of doing it. But what does that mean from your perspective and from Informaticus? So look at this, we need to talk to customers. Customers are living in this world where while they have the traditional computing on-prem, it's not going away, but they are moving to cloud at the same time, right? Whether it's SaaS, PASS, or infrastructure as a service, they are all leveraging all kinds of social media and mobile platforms as well. So in that world, customers need some level of what I call sanity. So they're looking to us to provide the data platform. So one is having a data platform that can provide them the abstraction any kind of data and provide them, make data available for them to the analytics. Second thing is worried about security. I mean, traditional security is getting totally disrupted. I mean, we've had all kinds of security for the last so many years, but if you see the number of attacks have just kept going up. And the thing is what I would argue is that what customers really want to do is secure the data. But unfortunately today, the approach is indirect, going from the network to the device to get to data, but they cannot still get to data. The other change that we are obviously bringing to customers is that focus on the data from a security point of view. And that's why we're bringing security source. I got to ask about security because this is big. So we always, it's now ratified on the cube multiple times by many execs. I'll just say it, security's a do-over. What is that do-over right now? Because you perimeter, talk about that going away and cloud APIs, you know, real time, new CMSs, how people are interacting on apps, asynchronous, all this stuff's happening. Great, what does it mean for security? What's the do-over? And what do you guys see? Because there's going to be breakthroughs in security. Obviously big data will be part of identifying patterns, predictive analytics, but how do we protect the data? How do we protect the environments? It's a huge problem on a lot of our customers' mind. So I would just echo what you just said that traditional security is being turned upside down. The thing is that 10 years ago, the size of data was very limited. So you had one perimeter, you had one network, you can put all the data in a wall then you can protect it, right? Data wasn't proliferated. You did not have dispersed geographies, outsourced environments, you did not have the whole insider attack situation, right? And the data privacy and governance laws have exploded. None of those things existed five, 10 years ago. They're all the most important things today. So I would argue security is going through its biggest transformation ever in the 25-year history. And how can you layer your management of the data structures and the integration over a new security model? How do you leverage the two together? So we can, we have a data security intelligence platform. The idea is that from this vast pool of data that a customer has, we can identify the true sensitive data that they actually want to protect. That to me is first we find the haystack and then find the needle in the haystack. Because you understand the medium of structure. Because we understand data, we understand, we profile data, we classify data, but then is you're interested in where is this data going? The proliferation of data. Who is accessing it? Who's touching it? Where is it going? Those are all very important things to know. That our data platform is able to give to customers that no infrastructure security platform can provide you. So governance falls out of the management of sort of the structure of the data. Exactly. You need to know about your data first before you can do something about your data. Whether you want to govern data, protect data, mask, encrypt, but unless you know something, you're kind of blindfolded. So we kind of almost taking the blindfold away from our customers. Okay, I mean, we're running out of time, but I want to get you to final word and get you a comment on a phrase. I said, what's the hottest thing last week as the emcee world and we were talking about, you know, the hottest things in big data and extreme IO guy said, what are the top three conversations that you're involved in? He goes metadata, metadata and metadata. Why this metadata focus and what's, and why is it so important now? I mean, metadata is great, it tells you about data, but if you have to know your data, that brings up the point. Describe that the importance of metadata and how that will impact the future of intelligence of data. So I would repeat those three words again, metadata, metadata, metadata. And that's what is, metadata is data about data and with live data map and a metadata technologies, that is the secret sauce behind the new data-centric security, the new data, big data analytics, any data intelligence. When you have the context about the data, who is touching it? Where it's going? Which device it sits on? Which reports it touches? When you have that context, you are a lot richer and intelligent. Final question, I want to just do a follow-up on that because I want to just quickly follow-up. For the folks that aren't in the data space who kind of understand the value of the data, explain to them the order of magnitude of breakthroughs that we'll see, new breakthroughs, securities to do over, new stuff's going to enable there, you guys have an enabling platform, metadata, metadata, what kind of order of magnitude, kinds of breakthroughs, just in terms of volume, will we see in the next few years? For the folks who don't know. Significant, it's exponential. So we see exponential changes right now. Analytics is going through an exponential change. Security is going through a fundamental transformational change. The whole fragmentation of data storage stack, application stack, just look at that fundamental change that's happening. So again, computing, analytics, security, the number of transformational changes that are happening at the same time, which is why I was talking to this large customer, they said in this world of insanity, the only sanity they have is Informatica. So that's what we want to provide our customers. Okay, we've got to break it there, metadata, a lot of breakthroughs coming, we'll be right back into this short break. More data from theCUBE here at Informatica World 2015.