 Live from Seattle, Washington, it's theCUBE at Tableau Conference 2014 brought to you by headline sponsor, Tableau. Here are your hosts, John Furrier and Jeff Kelly. Welcome back everybody, we're here live in Seattle at the Tableau Conference 2014. I'm Jeff Kelly, you're watching theCUBE. So of course, a big focus for Tableau is their partner ecosystem. On the big data side, of course, there's a number of partners that are important to Tableau. One of them, of course, is Cloudera. We've got David Tishgart, who is the Director of Product Marketing from Cloudera, joining us now on theCUBE. Welcome. Hey, thanks for having me, Jeff. So we were just talking before we went on. So you came over to Cloudera from the Gazang acquisition, which was just a few months ago. That's right, it was in June. So yeah, talk a little bit about that experience, coming from Gazang, where you were focused on very much on the Hadoop security question. That's right. When we went over to Cloudera, certainly security for Hadoop is important. About the time that acquisition was around at Hadoop Summit time, we were there with theCUBE and there was a lot of talk around enterprise-grade features, security being one of them. Talk about the experience kind of coming from Gazang and now being part of Cloudera. So I think really the timing was really interesting because what it shows was really maturity in the Hadoop market. So this is not a new technology, it might seem new by technology years, but it's really been around since the late 2000s, so 2006, 2008, or I guess the late-oughts is the right way I'd say it. But what's happened now is Hadoop pilots have started to move into production. Those nosy guys from InfoSec and Compliance have started to butt their heads into the mix and said, okay, great. So where is the state again? What's Hadoop? Are you going to secure this? Who has access to it? All these questions start getting asked and all of a sudden the architects who in the developers were so excited to put this Hadoop project into place are now saying, ooh, I should probably take a look at what I'm really doing here. So around that time Gazang had really formed partnerships with a number of the NoSQL and Hadoop distributions, Cloudera, obviously being the biggest and certainly for us we had the most common customers. These are companies that are really, really large financial services organizations and large healthcare organizations that have requirements around HIPAA, around PCI, some organizations over in Europe have data privacy regulations that they needed to meet and the chief requirement for meeting all those things is making sure that the data is encrypted at rest and that the keys are managed separate from the data. Cloudera saw this as an opportunity to really differentiate their product by saying, okay, you know what, this is something that across the board as we go into production we recognize that our customers are going to need. So let's, and Cloudera was leading on security with Century and a number of other products so they pulled Gazang into the product mix and now encryption is available to Gazang customers and key management is available to Gazang customers and that's really, really unique for Hadoop frankly at this point. So actually I want to dig into that a little bit more. So for our viewers who don't, maybe you aren't, you know, security experts, we say key management is critical for the key to be managed separate from the data. What does that mean? Break that down for us. Yeah, yeah, the best way I can explain key management is if you think about what you do when you lock your house or you lock your car, you don't leave your key in the door, you take it with you, you secure it. I mean, granted, you're securing it in your pocket but it's still a way of doing key management, so to speak. The way your people are securing data right now is they'll encrypt it and then they'll put the key either, you know, in a file on the same server. They don't know or it's in a spreadsheet that's unmanaged. They don't know how to manage it properly. So an unmanaged key, a key that's not encrypted itself and stored on a separate server is incredibly vulnerable and if your key is breached, you know, you may as well just hand it over your data to a hacker. So that's what we talk about when you say key management. Okay, I mean, clearly with everything happening in the news with the Home Depot data breach now kind of top of mind, it's clearly, you know, an issue that people have to start thinking about and it has the ability. I mean, when you think about security and not being able to, you know, if you don't secure your data, it's going to be a kind of a monkey wrench for the whole big data space, right? Because, you know, you've got jurisdictions around the world that have different rules and if you can't meet those requirements, all those POCs aren't moving to production. Right, it's really hard and just navigating the ins and outs of security. And by the way, encryption and key management is just one part of a very, very comprehensive security strategy that you have to have. So just navigating that knowing, okay, so my company is based in Switzerland, but I have customers in China. What are the requirements there? You know, if I'm in California and I've got customers over in Europe and Germany, especially Germany is very, very rigid about their privacy rules. You know, how do I, how's the data, where do I need to keep it? So really understanding all these nuances is really, really important. And you're going to get those questions, you can't dodge them. So as you're thinking about big data and how you're going to stand up your infrastructure, you need to have that in the back of your mind because you can't put production level sensitive data, names, addresses, phone numbers, social security, numbers, things like that. You can't put that into a production level Hadoop database unless that data is secure. So you want to bring in, so you need somebody that knows how to talk to InfoSAC, you need somebody that knows how to talk to compliance. That needs to be a requirement as you're moving into production with Hadoop. Yeah, I think it's one of those areas for the, when I look at the Hadoop ecosystem and the vendor community, they've got to, you know, that the conversation is all about, you know, the guts of Hadoop for so long. Yeah. And you know, if you, you know, in order for vendors like Cladair and others to really be enterprise grade software vendors, you've got to be able to speak the security language, the governance language and all the other really, maybe not so sexy stuff with big data analytics, but critical. Right. I mean, it's hard work to get there, but you need to, you need to be thinking about, you need to be thinking about all those steps, right, because you don't want to move to 99% and then all of a sudden, you know, your project gets kicked to the curb because it, you know, you didn't have the right access controls configured. You didn't have that governance and lineage. You didn't have that encryption. So it's just, it's just something to think about. And interestingly, you know, we're at the Tableau conference right now and I love being here because we are so used to talking to IT developers, architects, and these are the end users that are going to benefit from a integration with Tableau and Cladair, and these are guys that are doing just amazing analytics, but they're not going to get a chance to see all this data and all these great visualizations that the data is not secure on the back end. So, you know, in all the conversations I've had today, that's come up a number of times great. Talk to me about security. I love what you guys are doing, talking about security. So for the end users, it's really, really important. Yeah, absolutely. So, as you mentioned, we're here at Tableau. So we go from kind of the enterprise grade at the foundation of Big Data up to the top of the stack with the visualization. Talk a little bit about what you guys are doing with Tableau. You know, Tableau is in this interesting position where they are, you know, obviously a great product, but they're also the beneficiaries of this kind of Big Data wave, and they're kind of sitting on top and serving as that visualization layer for any number of underlying infrastructure. Talk about Cladair's approach to visualization, what you guys are doing at Tableau specifically. Yeah, so what we're all about in Cladair is really bringing all the data to bear for the analysts and for the end users. So if you've got data right now that's moving from one siloed database to a data warehouse to another siloed database, whatever it might be, you want to make sure that the analysts are able to get access to it without any sort of impediment. They should be able to have unfettered access to the data they need to ask bigger questions and improve decision making and things like that. So what we're talking about is landing as much of your important data or all your data, however you want to set it up, as you can into our enterprise data hub. And we integrate with all sorts of data warehouses and with relational databases and really any kind of system that's in your existing infrastructure. And we do this so that as customers need to query their data to get that business intelligence out of it, that's really what's important to us. So we've set up this data hub that integrates with Tableau, integrates with your common ETL tools on the back end to allow this really nice and easy flow of information to the end users. So if we were talking earlier, so you're managing marketing for a lot of the software partners, so that includes the BI layer, data integration. I'm curious, maybe not Tableau specific, but just generally speaking, what is the approach of the BI vendors to big data? Because actually maybe even leaving Tableau out because they're kind of the newer modern approach to BI. You've got kind of the legacy vendors that grew up in this world of a structured rigid data model in the enterprise data warehouse, and they kind of fit very nicely on top of that, like a puzzle. And they answer questions that you've already kind of asked. You know what the questions are before you model that in your warehouse. And then the BI applications were built for that world. How are they adapting in your opinion to this new world of Hadoop, big data, where you're loading all sorts of data? Some have no structure whatsoever. You're asking questions that are very ad hoc. How are the BI vendors adapting? I think it's really exciting. So we work with a number of other vendors. We work with the BI vendors. Like I said, we work with the data warehouse guys too. And what we're trying to tell them is, hey, you know what? Just think of us as you think of your other systems, right? We're another piece to the puzzle, allowing you to get access to data that you couldn't have access to before. So you can take these structured rigid data sets that you're normally used to querying, and now you can bring in your click streams and your social data, your machine logs, what have you, and combine it in there to get some interesting insights. So I was just in a session with one of our joint customers. I won't name the name, but it's a travel company, business, corporate travel. And one of the cool scenarios that he talked about was, we all have been through SFO at one time or another. And I think almost 100% of us who've traveled in SFO have gotten fogged out at some point in time. The airport's closed and they had to go be shoveled over to Oakland or something like that to catch another flight. Well, what if there was a way to predict fog rolling in as opposed to finding out an hour later that your flight's been canceled and it's not going to be rescheduled. So what if we could predict that weather event and rebook you on another flight and figure out a way to show for you to the nearest airport based on what the traffic looks like to get from SFO to that airport. So you're now pulling data from a bunch of different sources. You've got your weather data. You've got your traffic data. You've got flight information and that point of sale system that allows you to book flights. So all of these disparate systems they have to pull data from and then model it and they model it in Tableau and try and figure out, okay, what's the optimal way to get you from point A to point B in the least amount of time? That is a really cool Hadoop problem that involves a number of disparate systems all working together. And so we kind of, I like to think that we're the center of that. We can help make that happen. It doesn't mean that we are replacing any of the other systems in there. In fact, we're optimizing them to make them work that much better. So let's dig into the relationship with Tableau a little bit more. What, I mean, do you see eye to eye do you think with the kind of the philosophy of Tableau they're very much focused on customers? I know that's right. Obviously an important part of Clutter's approach as well to be very customer focused. What's the partnership? What are the dynamics of the partnership? How deeply do you go in terms of the technology, reseller arrangements? What's the real nitty gritty? Yeah, so what we find is that a customer, a customer that's running Tableau on top of Clutter is almost 100% of the time a incredibly happy customer. When I talk to people inside our organization and our sales team and they've got a Clutter customer and they're looking to see what's next for them. Hey, you know what? Get that customer to use Tableau to check it out. That is a great way to look at your data. So the things we have with them, we've got an ODBC connector, but more importantly, we've got a tool called Impala. You might have heard about that. We launched it a couple of years ago and this is a way to do really, really high speed high speed queries on data. A lot of folks are used to running SQL queries. This is like doing that, but on all of your data. So it's a really fast way to get analytics out of Clutter and into Tableau. When Christian gave his keynote earlier today, he talked about speed being really, really important. And that's what we're enabling for Tableau customers is the speed to get data out of the data source and into Tableau. Yeah, well, I think performance is really important when it comes to, if you're going to try to do iterative analytics, you're just going to try to do iterative visualizations to try to answer a problem. You don't want to have to sit there. And as Christian said, I'm doing that, you know, kind of watching the wheel turn. You want kind of that real time interactive capabilities and the whole SQL on Hadoop approach just really bringing that to, well, bringing that to Hadoop in the past. It was really not possible to do that. So I guess looking forward, what's kind of your strategy? Now that you're part of the Clutter organization, you've had a few months, a few months I'm sure you're totally up to speed now, right? Got it all. Right, so what's on the roadmap for you guys? I mean, how do you look at this market from the BI and the data integration point of view and what are some of the key things for you, key initiatives for you and your organization over the next six, 12 months? Yeah, so we're creating an open partner ecosystem. We want to make sure that every vendor out there that is in the data management space or is any sort of, is on the periphery of data management understands what we do and the value we can bring as this data hub that's now getting stood up in a lot of enterprise data architectures. So we're working with a number of vendors across the board, large and small, to enable them to extract value from the data that companies are storing in Cladera. So I mean, that's one thing. We're just going to continue to work with partners. Do these high speed connectors so that company is not just using Tableau, but that might be using SAP, SAS, what have you, all these other systems and certainly all the data integration vendors. We want to make sure that they've got super fast connections into Cladera. As a company, we're really focused on growth. We had a really big year this year, obviously, with the announcement of the enterprise data hub early in the year and then taking down the money from Intel, which has really infused the organization, the Xang acquisition earlier in the summer. And now we're looking towards Hadoop World because there's a lot of big announcements that I don't want to spoil, but we'll be hearing about them in a month or so. So there's lots already happened and there's a lot more to come. Yeah, absolutely. And I guess kind of last question, this could be a question for Cladera or for Tableau. Maybe you can answer for both or maybe give some advice to Tableau. They're in some of this hyper growth phase. Cladera, likewise, with the cash injection from Intel, among other things, just the general market, the pace of the market, both companies growing quickly. And I mentioned being very customer focused. So remaining customer focused when you're growing at this speed can be a challenge. What is your approach from Cladera's perspective to maintaining that really close contact with the customer and then any advice to Tableau is they're kind of going through a similar period. And I'll tell you our customer interactions have been key to where we are today, right? We don't get there. We don't get to that nine out of 10 customer satisfaction rating by not talking to customers. So we've got a number of different folks in our organization that interact with customers. It's hardly just sales. We've got a great support. We've got a great technical services organization. We've got training. One of the cool things actually, if you don't mind, there's an interesting story on Tableau. So we do predictive and proactive support. And what I mean by predictive support, which is actually really cool, I don't think we talk about it enough, is so Cladera Manager, which is a management tool that allows you to do configurations and monitoring of your cluster, generates a ton of data, a ton of log data, right? And so customers who opt in to allow us to capture that data, we can compile it and model it. So we have our own enterprise data hub stood up in our environment. So I'm from Texas, so I'll say we drink our own margaritas, right? Or we drink our own tequila shots. So we use our enterprise data hub to collect all of this data from Cladera Manager about customer environments. And we can tell, we visualize everything up in Tableau and we can actually tell how a cluster is running, if a customer is running well or if it needs some work, if it's about to go down, if there's some sort of urgent thing that needs to happen, we'll proactively call customers and say, hey, by the way, you need to do this, this, and this to get your cluster back in working order. And when we've done that, we've seen like a 35 to 40% drop in meantime to recovery for any cluster that goes down. So great story about enterprise data hub about customer support and about using Tableau to service our customers. That's a great story. I mean, talking to Tableau during the day today, they talked about using Tableau to build the product, to build Tableau and you're using, similarly using Cladera's enterprise data hub to support your customers. I think that's a good approach because obviously, we're talking about sources of big data. Well, guess what? Hadoop clusters actually produce a lot of big data. They produce a lot of data, which by the way needs to be secured. Exactly. So just to go and visualize. And so just really, let's close the loop here. Yeah, exactly. Exactly. All right, David Tischkart from Cladera. Thanks so much for coming on theCUBE. Appreciate it. We'll see you again soon, maybe even in New York. Absolutely. A lot of action going to be happening there in about a month. So again, thanks for coming on. You're watching theCUBE. We'll be right back live from Tableau Conference 2014 in Seattle.