 Okay, we're back here live in Las Vegas here for the Splunk conference.conference 2013. This is theCUBE, our flagship program. We extract this into the field of noise. This is the Splunk conference. We're there spinning data exhaust into gold, which is a massive growth market. That's the tagline. No limits, amazing conference here. We have wall-to-wall coverage. Day one of two days of live coverage on theCUBE. I'm John Furrier, the founder of SiliconANGLE. I'm joined by my co-host. Hi everybody, I'm Dave Vellante. I'm with wikibon.org. Rich Collier is here. He's a principal solutions architect at a company called Pre-Lert. It's an independent software vendor inside of the Splunk community. And ecosystem, Rich, welcome to theCUBE. Thank you. Thank you very much. So tell us more about Pre-Lert. You guys are developing a pretty cool app. Tell us about it. Sure, basically the idea is that we're bringing sort of machine learning, machine intelligence to machine data. And the whole idea behind it, of course, is to help the mere mortal, right? To help humans extract more value out of the data using some pretty advanced statistical techniques. The idea being that, you know, not everyone is a data scientist, right? So bringing a lot of the same kind of techniques that data scientists use, but package it in such a way that it makes it easier for the sort of average user to get value out of their data. So we've been hearing a lot of that. That's the thematic. Where do you pick up and where does Splunk leave off? Sure, well, you know, it's kind of one of those things where, you know, your success with, you know, sort of getting insight from your data very often is somewhat of a dependent upon your skill level, right, as a user and sort of how much time you want to put into that. And so very often, you know, people that are, you know, pretty good experts at Splunk can, you know, pretty much somewhat approximate what we do. We have our analytical techniques, but there are certain aspects of what we do that are simply impossible to do with Splunk just because of the limitations of the way that certainly the search pipeline works in Splunk. Being able for us to, for example, look at, you know, multiple kinds of analysis simultaneously, so it's a scalability factor there. But, you know, ultimately, you know, again, it's trying to, you know, provide the user with a set of tools that, you know, enhance, you know, the way that they look at data and sort of, I guess in a similar kind of way that statistics have helped other things, right, like, you know, weather forecasting and election predictions, for example, you know. Do you have any particular industry focus or is it more horizontal? It's really across the board. So, yeah, we've got users that are, you know, social media companies, security companies, banks, insurance companies, it's really across the board. It's basically anybody, again, trying to get a lot more value out of their data with less human effort if you want. So you beef up Splunk, you make it more scalable, you make it more functional, deeper function, there's a performance aspect of what you do, you just make it better. Yeah, is that right? Yeah, that's pretty much it, yeah. So how do you do that? Because Splunk's moving forward, you're moving forward, so you just got to, you're like, just keep ahead of it. You understand where the curve is and you want to fill this white space. Yeah, yeah, yeah. So we certainly spend a lot of time with Splunk on a corporate level, understanding things. We work a lot with the product management team and have some discussions there about how we can work together better. We're very much in tune with the engineering roadmap. In fact, Splunk themselves were so impressed with what we've done with our own app, given that it's pretty complicated, some of the things that we've done, right? They wanted us to be heavily involved with the new Splunk 6 app framework because all the new changes that are involved there. So, in fact, our CTO, Steve Dotsons, who's here at the conference, I think is doing some sort of interview with them now, right, on that very topic. How, what's your head count at pretty roughly? I mean, you know, how many people work there? Yes, we're still pretty small, so we're still under 20 people. Okay, and you've got offices in Boston and London, right? Offices in Boston and London. So you're really specialists, right, in this space, right? Yeah, so our development team, frankly, I mean, they're really steeped in computational statistics. I mean, that's our core intellectual property. So, when we first started, we had a standalone version of the app and quickly realized that we didn't want to write a lot of sort of data collection mechanisms to get data to us, right? We wanted to piggyback on top of existing tools that already had a wealth of data that was potentially untapped, right? So, you know, we've done, obviously, this partnership with Splunk. We've done an OEM with a very prominent APM tool in the marketplace. So, again, sort of under the mantra of taking existing tools and data and making it better for humans. We had Nate Silver on a few weeks ago. He was on the Tableau conference. We were asking about social data and he was sort of half empty on that topic, basically saying that there wasn't enough good data, the patterns weren't there yet. Do you see that or do you see that social data actually has the ingredients from a computational statistics basis to actually start to provide answers in direction and trends? You know, I don't necessarily think that, you know, any kind of data set in itself necessarily is a silver bullet. At least what I've seen from our customers is when you sort of bring a lot of disparate data that may be loosely coupled and bring it together, that's where you can get more of a better picture of things. So, I mean, certainly if you focus on one data set, you can sort of eke out some interesting things out of it on its own. But very often a lot of our customers say, well, that's good, but I want to correlate my network data with my authentication data or I want to correlate my app server data with my call center data or whatever. Painting more of a sort of holistic picture of what's going on. I think that's more sort of timely in a lot of ways. Now you guys don't resell Splunk, right? So you go into Splunk customer sites, you go, but you co-sell, how's that all work? So, I mean, yeah, in general, obviously since this app requires Splunk to be there in the first place, we have to sort of be on the tail end of things. But very often, even in some types of competitive situations, Splunk is trying to interest somebody in their platform. They see us as another augmentation to their platform and they will bring us in even on sales calls sometimes. Really in some ways to demonstrate how well that Splunk is a platform and that a third party can come along with some expertise and make their framework look good to the customer. Now, are you typically selling to the IT function or increasingly the business function? Talk about who the customer is. So I would say that when we first released our app, we were pretty much in the sweet spot of, I would say the IT ops and performance management crowd. So sort of traditional monitoring, sort of data center monitoring, infrastructure, application monitoring. With some of the stuff that we've done in sort of our most recent release, we're also getting a lot of traction around security. So anomaly detection in sort of the security context. We do have some customers that even use us on business data, but I would say that relatively speaking, the big push right now is obviously in the security side, but we've got a lot of sort of critical mass for people that are doing more traditional IT ops and APM stuff. So what's driving the big push in security? Just the increased awareness on it, Prism, NSA, Cloud, what's the driver there? I think on some level, some security experts are sort of realizing the limitations of sort of being behind the persistence and threats being behind the eight ball, right? Not trying to get in front of the users, of the bad guys. Always searching for the unknown unknowns, right? If you can't search for it, you don't know to search for it. How do you know it's out there? So I think a lot of security experts are now coming to grips with using more of a statistical method to find what's different in the data instead of relying on traditional sort of rules and the known threats, the known knowns, so to speak. So you're saying the emphasis is now on trying to figure out what you don't know? Is that right? Yeah, or even in some cases, finding things at scale, right? Which is, I've got a lot of data, I don't know what I'm looking for, at least just help me narrow it down to something that I can focus on as a human. Right, and okay, so really, as the old saying, you can't take the humans out of the, humans are the last mile, can't take the humans out of the equation. So speaking of equations, you got this equation on your shirt. Oh yeah, yeah, yeah. So it's this math equation, it says I'm not normal. What is that equation describing? It's actually, this is a equation for a Poisson probability distribution. Okay, so. Poisson distribution? We're in statistics, so. I remember my math days, you know. So the idea is that, and the reason why we say I'm not normal, because sort of it's a sort of traditional fallacy that a lot of machine data is normally distributed. I mean, it forms a sort of. This randomness certainly is normal. It's a randomness, but there's a lot of data that doesn't follow that normal distribution. Yeah. It follows more of a Poisson distribution or a log normal distribution. And one of the cool things about the way that our statistics works, is we use machine learning to automatically figure out what is the best fit model for the data. If you fit the model to the data better, then you actually get better outlier detection. You get better results, and you get less false positives and false negatives. Well, I mean, listen, listen. Math and statistics have been a great boon for businesses in the brick-and-mortar world. I mean, Poisson distribution, some of the math we're talking about have shaped how businesses organize and their operations, whether it's a drive-thru, you know, in and out burger, where they locate things, and just random probabilities for some randomness, but yet some predictability. Now with machine learning, you can scale that to a whole other level. So have you seen the business impact some of your clients, and where do you see that going? When you look at some of the math involved, where you get some sample data, extrapolate, use the predictive analytics, it's going to impact the value chains of a lot of the businesses out there. Have you guys gotten to that level of analysis yet? Yeah, I think one of the things, at least I'm seeing from a lot of our customers, is that they're scaling Splunk. They're scaling their data collection in some ways. They're doubling it every year, right? And they realize they're not doubling their staff every year, right? So they need to figure out ways that they can scale their analysis of their data in an efficient way that doesn't require the human aspect of things. So they're already recognizing that they need something to help, right? They need something that helps them get better. Talk about the automation piece, because one of the things that I've seen a lot of great things like you guys doing, I mean by the way, great work by the way, this is cutting edge, bleeding edge, and really great pioneering work that's relevant. But you're automating, you're using the technology to create some automation that would normally require a lot of manual process. We actually see that with Splunk with log files, there's no brainer, right? You're wrapping through log files. We all know what that means, but when you start looking at the automation of things, that's really where the impact is, because then people can do their job. So with that in mind, that automation, the next leg of, the next domino to fall, so to speak, or next leg of the journey is what? The skills, what's your take on that? I'll see, you get the automation in place, it's some geekiness, math, data science, data modeling. After that, what's next? Yeah, it's a great question. I think one of the things that we want to do is make our technology so ubiquitous that everyone can use it, right? So I think it's broadening out to not just the expert users, but also to the novice users, right? So that everyone can take advantage of this kind of technology. In other words, again, not everyone is a data scientist. So you're bringing a lot of the same techniques, to data mining, but do it in such a way that it's easy for everyone. So I don't know if they answered your question, but... No, I mean, it's kind of an open question because it really is about the future, right? I mean, as Dave always says, one penguin jumps in the water, they all jump in, and certainly with analytics, you're seeing the business benefits, it's kind of a no-brainer. You don't really got to do a quick ROI to say, hey, you get some good analysts, you get some data scientists, you can do some new things. Yeah, and I think it's a general trend. I think it's just sort of natural if you see the progression of how statistics has changed, even baseball, right? If you've seen Moneyball, it's sort of a, it got to a point where people needed to figure out better ways to optimize, how do I build a better team, even though I don't necessarily have the money to do it? Is there different ways I can focus and optimize players? Let me take a step back here and let's talk about the conference here in Splunk. I mean, it's the two questions. One, talking to the folks out there kind of describe the scene here. It's still early day, when the company went public, they're growing, they're on a big rocket ship ramp, a lot of success, a lot of passionate partners and customers. So talk about the environment here, what's it like? And two, why are people so excited about Splunk? You know what, it's great, I think, to see the kinds of people that do show up here. There's sort of an interesting mix of really hardcore geeks. I think that comes from sort of Splunk's roots. People that are more sort of business professionals. One of the things I love about this conference is that it's very, very education-focused. People come here to actually learn stuff. They have whiteboards in the hallways. Exactly, yeah. It's not just I'm here for a boondoggle in Las Vegas. There's a bit of that, but the fact that it's very much education-focused, when people come to our booth or any of the vendor's booth, they want to learn something. I think that's sort of an intrinsic sort of aspect of this conference. You know, Dave and I always talk about this market that we're in, it's an inflection point of, epic proportions, wealth creation, innovation, business value, human value, people value. And we kind of talk about the revolution of the PC revolution. And I got to say, these events like this remind me of the early days of the PC industry where people would look around at you, young people, building an industry, a whole new industry. Not like just disrupting an old one, taking over another, but basically recasting a new business. So it's super exciting, congratulations. I'll let you get the final word in. Two Northeastern boys, by the way. I don't know if you knew that. Riches attended Northeastern, John. Yeah, Northeastern, yeah, and you. Go Huskies. Everybody wants to go Northeastern now because it's the co-op program. Nobody can get in. I went to RPI too, by the way. RPI. Yeah, so I did spend some time in the capillary. Enjoy. Northeastern alumni's here. One of them's the host. That's me, guest, Northeastern alumni. We are inside theCUBE. I'm John Furrier with Dave Vellante, Rich Coley here. Inside theCUBE, we're right back with our next guest. Here at the Splunk Conference, where they are spinning data exhaust into gold, no limits, listen to your data. That's their theme. This is the future. This is a great opportunity, great market. We're right back with more exclusive coverage here live in Las Vegas. This is theCUBE, we're right back.