 the Cube. At Big Data SV 2014 is brought to you by headline sponsors WAN Disco. We make Hadoop invincible and Actian accelerating Big Data 2.0. Welcome back. Jeff Frick here at the Cube. We're at Big Data Silicon Valley 2014. Santa Clara, California is just across from the Santa Clara Convention Center and the Stratocons that's going on. We're over at the Hilton in the Yosemite Room. If you're close by, we invite you to stop by, take a look at the Cube, take a visit, and we're happy to always see our friends come on by. We are joined in our next segment by Mark Taranzoni, the CEO at Squirrel, Cube alumni. Mark, welcome back. Thanks, Jeff. Thanks for having me. So you just came over from Stratocon. What's kind of the vibe over there? Yeah, it's a good crew, a combination of biz-deb, but a lot of end customers and a lot of good sessions too. We had a talk yesterday that was very well attended also by our CTO. Awesome. So let's just jump in. We've had Squirrel on a number of times. Sure. Why don't you give us just kind of a quick update on the company? Sure. So we finished off 2013, had a plan on customers, had a plan on bookings, doubled the staff, moved the offices, closed the series A. So it was an exciting year for Squirrel, for sure. And we're not pausing. We're stepping right into 2014 with still some pretty big lofty goals ahead of us. Now, those are great. So those are all great business metrics, which is kind of another indicator where the state of this whole thing is, you weren't talking about speeds and feeds or anything but really business metrics and adoption in the area with customers. We've got great technology. There's no question about it, but technology, from my perspective, is really a basis for solving problems that customers have, and that's what we're really attempting to do. So we targeted some very specific verticals, banking and financing, healthcare and telecommunications, and we were able to penetrate all those verticals in 2013. So we talked a little bit before the lights came on in terms of some of the customer stories that potentially you could or couldn't share, and I know you guys really spun out of NSA, so it's some heavy lifting technology and stuff that's off to kind of kept under the covers. But I still wonder, without naming specific customers, if you can get into some of the use cases where people are able to do things before that they just couldn't do before or do it so much better? Yeah, absolutely. I'll give a couple examples of that. So we have a number of customers that are looking to put together a very secure data store. One of our major claims to fans and differentiators is our security modeling that we've built around, not only a cumulative, it's the open source project that's a foundation of our technology, but a lot of the tech that Squirrel has put together to enhance that security model. So a number of our customers are looking to really to manage their entitlement process and they have very complex security rules and requirements. We're seeing a transition from role-based security where it was really aligned to who the user is and then mapping that user to what access rights they have to more of an attribute-based access control, where it's not only who the user is, what location are they logging in from? Are they at their home office or are they in the office? What time of day is it? Is it a remote site? And those attributes now dictate what security policies are aligned to those people. So it's becoming much more complex and we're really seeing, we really feel like we're the leaders in that from an information management standpoint and a lot of our customers are starting to roll out products based on those requirements. That's interesting. So let's dig down a little deeper on the attribution. Is it because the same person has different rights if they're sitting at their home computer or working off their mobile or is it because it's really more of a check and balance is to get more attributes to verify who they are and what they really do have? I think it's a combination of both, Jeff, but in some cases that, and I'll use an example, you know, maybe in the financial services market, some trades just cannot be placed from a broker when they're outside of the office versus inside the office. So they'll log in from their home computer, there's certain things that they can't do that they're able to do when they're in the office. So that's kind of an example of a fine services aspect of it. There's other cases in health care where if they're accessing, you know, for remote location, you know, personalized health records, then, you know, they really can't access them from those places versus inside the firewall. Interesting. So, yeah, starting to see more and more of that for sure. And what's kind of the driver behind that? Is it the regs? Or is it the bring your own device? Or is it just because now they can? We had a guest on earlier talking about, wow, what if you could do these things? How would you look at the world differently? Or is it really just because they can now they do? I think it's a combination. The there's enabling technology, for sure. Regulations are getting more stringent. I think, you know, some of the news stories we've all seen and read about where, you know, personal information is being, you know, mishandled or misused, creating a lot of concern in companies and in, they're trying to figure out better ways to manage that and control that. Yeah, interesting. So I wonder if you can talk about kind of what some of the future challenges are that you guys are looking to take down? What are some of the next hills that you're you got your eye on? Absolutely. So, you know, we, we think we've locked down the security aspect of it. And we're really a building an enterprise ready, you know, no sequel data store that sits on a dupe. And we feel that we have enough traction market with that story and that solution. But customers want more, right? Okay, great. I have all my data in one place, and it's secure. I want to be able to do things with that when we do able to do things. So we were we're now starting to see a movement towards this operationally analytics. Well, really, it's, you know, in the old world, you know, OLAP and OLTP, you know, we're separated and now there's this combination of, you know, I'd like to be able to use this store in an operational way. But I also want to run some analytics on top of it. So we've built in some of those analytics capabilities, I think we'll still will certainly through this year grow those capabilities. We've got full tech search capabilities, we've got graph search capabilities, we're building in some graph based analytics, which are very unique and differentiated. Customers are very interested in understanding, you know, what entities are in their data and what are the relationships and those relationships are really, you know, best suited to understand in a graph model. And it gets pretty complex if we go multiple hops and multiple levels in those relationships. So those are some of the things that we're working on today. And, you know, we've got a very hefty roadmap. I've I'm making a lot of investments in the engineering team. So we've I've got some announcements over the next 30 days, we've brought in some some some more senior talent that are going to help us help us grow the technology over the next year. So stay tuned for that. I'd love to be able to talk about it today, but we're not ready to have a break. Come on, we love breaking news on the cubes. And then, you know, one of the things we're doing, I think last time I was on here, I talked about, you know, maybe, you know, we're big supporters of Accumulo. We have a lot of committers and contributors. We're, we're working on co-sponsoring an Accumulo event and would certainly love to have the cube there as well. Excellent. And then we've we've targeted June of this year. And really, you know, to help get the word about word out about Accumulo, what what, you know, attributes and differentiators it has, and bringing that development community together, you know, to talk about different problems that are being solved with the technology. Right. Well, and it's an interesting use case where you've got an open source kind of core, and yet you guys are super high security, you have the whole NSA roots. Talk a little bit about because that seems contradictory from the outside looking in and how you can manage both to have an open source core as well as having really high security application. Yeah. Well, so the the core product was developed by the NSA was kept internal for a long time. The value of having that open core for us is that, you know, it's been proven it's been running in a, you know, a pretty mission critical environment within the NSA. And, you know, I certainly don't know how they're using it. And most people, if you told me, you'd have to keep me. Exactly. Don't tell me anything. But I think one of the things we do know is that, you know, there, there's a tremendous amount of data in their back office, and they're managing things in a real time way. And this technology has been been proven to scale pretty, pretty large and certainly greater than most things out in the marketplace. So we have that ability to take that open core. And we've built a lot of tooling around that to make it usable in the enterprise. Right. So those are those are the, you know, things that we've built today. And we want to extend those capabilities in an analytic way. And and actually into solutions that we can drop in that customers can utilize to solve problems, you know, without a lot of heavy lifting. Now, I know, you know, there's a lot of tenants on big data, you know, velocity, variety, veracity, value, depending on who you talk to, they'll come up with a slightly different set. But I know one of the big ones is volume. And I know that you guys have some special capabilities around high volume ingest. I wonder if you could speak a little bit about that capability one and two, why that's a value and how people are taking that that capability and putting it to good use. Yeah, absolutely. So, you know, our ingest rates are this has been some statistics put out there, but our ingest rates are much greater than most products out in the marketplace. You know, so we're able to almost keep up in in stream data into the system in real time way. And I think that that really is based on the technology around accumulo and the way they built it to be able to massively ingest lots of different data, multi structured, semi structured, unstructured data feeds. And then not only in some of the technology that squirrels built is to be able to index that on the fly as it comes in. So it's available to the application real time as it comes, it streams into the system. It's a unique capability. It's certainly hard to do. We also apply these security labels on the data as it's being ingested. So, you know, we've got those security model applied to the data as it comes into the system and becomes available to the application of real time. So there are a number of large customers in, you know, telecom arenas is probably a good area for that that the amount of data that they collect in from a log perspective is massive. And it's happening all day, every day in real time. And they need to be able to store that information in a place that they can react to some, you know, network outages or any particular anomalies that are happening on the network in where we think we have a great solution for that. Right. But then you always move to your next point of failure, right? So now if you've got a super rapid ingest at night at the indexing, it's not going to be able to do something with it or else you're just building up a big store. Right. Right. So, yeah, and those that those become those real time applications on top of the on top of the data store. So, you know, we've built a very rich API on top of the system. We allow programmers to develop on Java, Ruby on Rails, Python, people who roll apps out very rapidly. And yes, and it's all about, you know, what type of insights the customer is looking for, what problems they're solving. So, you know, we have a number of cases where, you know, customers are building their own applications. They're able to roll them out very rapidly. And one of the reasons they're able to do that is because they don't have to worry about the security model. We've basically taken the security model, applied it to the vault. We've integrated to all their security systems, all their access control systems, all their policy engines, all their key management systems from an encryption standpoint. And we give them an auditing capability that can tie into the compliance modules. So, once you've established a model that you can apply the security to the vault, the development of those applications become much more rapid. Because you're not worried about that very, you know, complicated security model and what regulations may require to change it in the future. Hey, Mark, just wanted to ask you a few questions around industry landscape. You guys are out there. You're grinding away as a startup, always trying to innovate. And, you know, we're in an innovator die situation. And at Hadoop Summit, we saw each other last at the team. We saw them again last night at the Cube Party. A lot's going on. What's changed in your mind? You're out there, you're talking to customers. Are these real problems being solved? What's changed since Hadoop Summit last year to now in your world? Yeah, definitely. I think this year, I mean, a lot of people are figuring out in the last couple of years what their big data strategy is, you know, and what tools they're going to use to bring all this data together in one place. I really believe this year, and in some case into next year, is going to be all about value of that data. Creating use cases that solve real problems in real time and in a lot of cases, create monetary value. We've got another interesting use case in the medical field. Actually, I can talk about this one because we did an announcement with the company. The company's called Med Year. They're a private health exchange startup. And with some of the regulations out with the Affordable Care Act, as we all know, we can now shop for our own health care. We can pick, you know, different providers and we have the freedom of choice of who we want to use. One of the problems associated with that is how do I get my health records, you know, together in a place where I can utilize them and move them around to whoever I need to? Well, this company, Med Year, has created an application, a cloud-based application on top of our data store, where they're basically allowing their customers to store personal health records in this system and be able to parse it out to whoever, whatever providers they want to. So it's a, you know, some of these, they didn't exist because the regulations didn't allow for them. And the technology wouldn't allow for the security applications associated with them. Yeah, you know, one of the things I like about your company, the squirrel guys, is you guys have one of the smartest teams I've seen in the business. You guys, certainly on the IQ depth chart is pretty deep. So it's good to see what you're working on. And so I got to ask you, what are some of the things that you're working on that are really hard problems that you're solving with the team? Obviously, it's like shooting fish in the barrel. It's a lot of hard problems to go after. But in particular, you guys have a good team and, you know, tend to be a good barometer, it might be, you know, some of the cutting edge issues that the large customers want to solve. What are some of those hardest examples that you're just going after? So, I think, you know, we talk a lot about the security. I think we've kind of baked that model down. But now it's, you know, you want to apply that security into almost anything you do. So if we're including analytics, well, that security model has to be applied to the analytic layer, has to be applied to the indexing layer. There's a lot of complications associated with that. But the good news is that was really the forefront of the architecture that we developed from the beginning. So we've built a lot of that implementation in and, you know, we're happy to say that it works and I think the challenge of the future will be around increasing the analytic capabilities. You know, some of the graph stuff that we're doing I think is going to be, you know, absolutely on the forefront of technology. I think, you know, the speeds that we're trying to accomplish it at and the performance factors that we're looking at are going to be, you know, we've got lofty goals in that area, so there will be challenges. But I think the performance, you're talking about the graph and database on the performance. Yeah, in large, I mean, we've got, you know, distributed data, you know, multiple petabytes and you want to provide, you know, analysis of that information in an instant, you know, in, you know, web time. We have a meal coming on from Neo Technology, going back to, I mean, 2007, he and I chatted it. At the time, graph databases was something like that, what? But now, essentially, distributed networks is graph. Right. So that's a really, really key area to develop on what things are happening in that world that you can highlight for the folks who might not be following deep into the trenches of some of the data issues around, you know, the different database architectures that you guys are attacking. Yeah. So, you know, I think there's a couple of things that are going to be interesting, right? So if you think about the store itself, you know, the fundamental, you know, features we have in that store provide a lot of value to our customers. But then you look at on both sides of it, you know, one of the challenges, you know, data comes in many shapes and sizes in many, in many different speeds. So what we're trying to do is make sure that we have, you know, very easy to use tools of getting that data into our system, you know, whether it's coming from, you know, a real time streaming engine, whether it's coming out of a data warehouse, whether it's coming, you know, out of log files that are on the network. So you're building those models to bring that data in and then having a data model that, you know, solves problems for the customer in real time. One of the things that we do that I think is fairly unique is that we build a real time aggregation framework. So as information is coming in, if you know the types of dashboards that you're looking for from a customer perspective and the types of queries you want to run, you can build that in, in just time. And we'll keep those aggregates up to date as data streaming into the system. So it really becomes a powerful tool if you, for the queries that you know you're looking for. So you guys are startups. I want to ask you the startup question. I'll see where I'm in the hallways at Strata Conference. There's a lot of new names out there I've never heard of. It's like I'm not really that impressed with some of the batch of startups. You guys have been out there for a while. What's your take on the startup ecosystem? It is not going to, it's not going to be as green field as it used to be. You guys got in early, have good position. I'll see good tech, good IP and great team. And for the folks entering in the market, what's your take on one, the situation, the startup situation, and then kind of the competition? I mean, it's pretty brutal. It's still a noisy space, right? There's, you know, it's hard to find what people really do and everyone talks about solving the same problems in the same use cases. So if I was going to give some advice to companies today, I would say that you find your differentiation and you know, don't, you know, don't do the same things and don't say the same things everyone else does because I think it gets lost in the noise. Pick something that you think you're going to be really good at and use that as your mantra. And what's the update on with you guys? You're sort of a team last night all looking good, all in uniform. What's, what's the update? So, yeah, the update is that we're going to continue to invest in tech and talent on the engineering side. We know we've got a very good lock on some unique capabilities and it resonates with the customers today. But our goals are very lofty and we're going to continue to invest in engineering. We're going to continue to invest in a go-to-market strategy and we're going to continue to invest in a partner strategy. So I think if we can execute on all three of those, then I think 2014 similarly to 2013 will be a great year for squirrel. Well, it's great following you guys that love the momentum. Great to see the team. What's your, what's your success here at the show? What can you point out to the folks out there what you guys are announcing and talking about? So for the show, for us, I think, you know, we're always looking at, you know, making sure that the relationships with our partner ecosystem are strong. We've got everyone in the room together. So we have a lot of discussions there. And then you really understand, you know, what the end users are looking for. So we had some really good customer discussions around problems that they have and how we can be used to solve them. So, you know, I think the show has been great for us and, you know, it's great to be out here and it's certainly great to be here on theCUBE with you guys as well. OK, one final question. Summarize for the folks out there are watching who aren't here at the Big Data SV event and also the Stratoconference which is going across the street that we're covering. What is the moment in time right now? What's happening? What's the core story right here for Big Data SV? What is the big thing at this moment that's happening? What's the big story? I think, you know, a couple of things. One is that, you know, Hadoop is certainly for real in mainstream and being deployed, you know, in almost every, you know, corner of the world. And I think, you know, the theme around that is, OK, it's going to be here as a data layer. Now we're going to solve problems with it. People don't want to, you know, we've got two choices with the data that goes in there. We can have a data lake that provides value or we can have a data landfill. And I think most customers want data lakes. So we've got to solve problems on top of that data. I mean, we like data ocean. That's what's more like an ocean with rip currents and all kinds of tsunami. So a bunch of plastic things. Yeah, so, I mean, I just don't like the lake term. I don't know where that came from. Gartner, I hate data lake. It's a data ocean. Massive. I've got a data ocean. Icebergs everywhere in North Atlantic. Caribbean waters. And that's where you guys just say, well, great to see you guys. Squirrel Hot Startup continue to grow and add to the roster of partnerships and success. Thanks for coming on theCUBE. Always great to see you guys. This is theCUBE. We'll be right back live in Silicon Valley. Here at the Hilton, I'm John Furrier with Jeff Frick. We'll be right back.