 everybody to theCUBE, SiliconANGLE's premier video broadcast. We are here at .conf 2012 at the Swank's annual user conference for in Las Vegas at the beautiful Cosmopolitan Hotel. We will be broadcasting all day today and going to have us some great guests. Continuing now, I'm here joining the course with my co-host Jeff Crick. Thank you Jeff, SiliconANGLE. Thank you. Yes, again, we've got a great show and a lot of customers today, a lot of people using the technology. So next up we have Rick Yetter from the Apollo Group, probably better known as the parent company of the University of Phoenix, which is a large and very big growing institution. Welcome to theCUBE. Thanks. We were talking a little bit offline before we got started and you were sharing a little bit of numbers about the growth of your guys' use of Splunk. So my question to kind of open us up is explain to us a little bit about, you know, why you have such a big data flow, kind of how does the Apollo Group use all this data, you know, what's generating all this, and then to kind of how you got started with Splunk and how because of the you can explain why, you know, the use has grown so much of the time. So a little bit of background, I started at Apollo Group about two and a half years ago. We were at a 50 gigabyte license and the primary use case was to get security information and be able to do some research analysis on security threats and so on and so forth. When I started using Splunk we were very quickly interested in seeing what else we could put into Splunk. From there we started ingesting all of our low balancer information and our forensic teams, they started using it a lot heavier. So our search times for low balancer information went from a five day process down to about a 15 minute process on some surges. Five days to 15 minutes. Yes. So we're talking needle in a haystack search that, you know, for a unique IP address for different fraud analysis. And our developers started getting involved in Splunk and they were very interested in putting their application logs in Splunk. So from there we took the step up to about a half terabyte of log data. So then we were putting in information from systems, all the security data, all the low balancer information and we quickly started realizing that there was a lot of business intelligence located in these logs. So the combination of security information, low balancer information, painted a true picture of how our users were behaving. Not only our customers but also how the employees of Apollo Group were behaving as far as, you know, internet usage and so on and so forth. So from there we had, we had launched our iPhone application for University of Phoenix and then we were working on launching the Android application. So that was our time to shine with Splunk. We took the Android application and we installed it throughout the entire application stack. And on launch day, we had zero faults. We had no outages, it was a perfect launch. However, we were able to circumvent some problems before it affected any of the users. Well, at the same time, we had three levels of visibility. We had an executive view that showed, you know, critical timeouts, any sort of dependency timeouts and then Android versus iPhone usage. And our executives walked into the war room and saw that they were like, wow, we need this. We need this on everything. All of our applications need this kind of visibility. So that pushed us to the five terabyte mark. When was that? When was that event? So that event was in October of last year. Okay. And it was an exciting time. It was a lot of fun to see the adoption of Splunk go from just a small group of security, a small group of application analysts or developers all the way up to a broad spectrum of business users, reporting users, system admins across the board. So we now have about 450 users in Splunk. That's impressive. I mean, I particularly like the part of the story where the executives came in and kind of were wowed by the things you were doing. And that kind of brings up the question that we hear a lot in big data scenarios is kind of, and this is not necessarily unique to big data, but the communication between the IT side and the business side is sometimes strained when it comes to technology, to say the least. But in big data, it's particularly important because especially with a tool like Splunk where you're discovering all sorts of different ways to solve business problems, not necessarily just IT problems. So how do you approach that at the Apollo group in terms of making sure that you're getting the most out of your big data infrastructure and tools and applications Splunk included between the business and the IT? So it's a socialization process. You start socializing Splunk from the lowest levels all the way up. It's very difficult to find an organization that has an adoption from the top down for a tool like Splunk. And just like any other monitoring tool that you have out there, HP OpenViews, you have Tivoli Monitoring. If you don't have a top down mentality, it's very hard to swim upstream. So our battle has been getting it socialized among the developers, managers and directors as something that is necessary for the operations of the company. And so our operations staff, they have multiple dashboards up of Splunk that shows critical information on how many users are logging in from around the world. We have geographic IP information that is displayed in our operations center. And so when there's an outage, Splunk is the first tool that they go to. They start looking at logs in real time to figure out what happened, why is the situation occurring. So that kind of visibility helps out with the executive staff saying, this is a great tool. An outage that would have taken an hour or two hours to circumvent has now been dropped down to 15 minutes. So that's the real big thing is that if you're using Splunk in the correct manner, it will get the right visibility. Interesting. So in terms of, it sounds like you've seeded it pretty well and people are happy and they're seeing some business value. In terms of extending forward and kind of looking forward now that you've got some experience with it and you've got some executive buy in and everyone kind of buys it. This is a cool tool. Where do you see it going next? Are you doing any of your own application development on top of it? Do you see some some neat things here in the partner pavilion that you would like to to leverage or kind of what's your long term vision? Because we're in the internet age, there's no super long term vision. Right. You know, the next six months, next 12 months, where do you see taking this tool to the next level? For us, I think our vision is going into Splunk for VMware, Splunk for Active Directory, Splunk for Exchange. Those three apps right there will provide a lot of insight to some of our problematic areas right now that our service desk does a lot of work in Active Directory and it's hard to see everything in one view. And so Splunk will give that visibility and the ability to have an audit trail of any changes that are made to Active Directory as well as Exchange. So the applications that are developed by Splunk as well as the partners are very good. And in my presentation today, I say, you know, they're about 80% there. So when you drop an application into your environment, it takes some configuration to customize it to what you needed to do. But, you know, for the most part, they're very well written. So yeah, so talk a little bit more about the community that's, you know, built up around Splunk. I think they, Splunk really makes a concerted effort to include their customers in both the development process, but also just kind of in the whole culture of the company. Right. Talk a little bit about what you're seeing here at the show. But, you know, more broadly, how does that community help you do your job better and make the most of Splunk and the other tools and applications you're using internally? The company itself, so Splunk as a company is a great company to work with and to partner with. But the community is very supportive of everything. So if you run into a problem, you can go to Splunk Answers and there is usually an answer for everyone of your problems, whether it be a small problem or a big problem. If you need help parsing out a line of code, just throw it out in Splunk Answers and someone will say, hey, just do this, this, and this, or try this. And so the collaboration between the community is better than I've ever seen from any sort of tool out there. So, you know, a little bit of my background I've been doing, I did Tivoli for about 12 years. Tivoli was a very difficult product to work with, but the community wasn't as supportive as Splunk's community. So like at Splunk Lives, you can sit there and talk to people, different customers that are running in the same issues or have similar environments and you just bring up a topic and they're very open to talking to you about it. So it's a very cool environment to be in and it's a very exciting product to work with. Yeah, it's an interesting combination of having a new way to do software in terms of the download to get started and then also supporting it with this community is really get out and you never used to hear that about old classic ERP or big applications. They were very siloed. There wasn't really a lot of community. It's like, here it is, it's hard, buy some training, but there wasn't this, let's get together over beers at an event like a Splunk Live and really hash through some of the issues I'm having, what have you seen, et cetera, and I don't know how much of that comes from the social aspect of the software world that we're all involved with, with Facebook and this or that, but it is a concerted effort to put that into the culture and to put that into the ecosystem with the partners and with the service providers and ultimately the customers as well. Yes, well also internally, at Apollo Group, we're seeing a widespread adoption from the developers and so they're developing their own dashboards, they're developing their own searches and it's great to see that because they actually will collaborate with each other on helping each other out on their individual problems. So you might have an application that is having an issue with JVM and another application and instead developers might be having the same problems. So they'll actually get together and start hashing out their own problems looking through Splunk and working together on that to come to a solution. So it's actually a really good, I would say it's more of a social developing tool than anything else. Yeah, interesting. So in terms of the disruption that Splunk is creating in the business intelligence data warehousing market, I mean it's really, big data in general, Splunk specifically, it's a new way of analyzing and processing storing data. It's not as rigid as the kind of the old way, it's a much more flexible and agile to adapt to changing circumstances. I'm curious to get your perspective as you've been in the industry for a long time. What kind of disrupting patterns are you seeing, not just Splunk, but big data in general and this whole notion of let's take advantage of all our data, let's do it on the fly, let's not have to model it ahead of time. How have you seen that transition and how does that fit into what you guys are doing? So from a disruptive standpoint, I'd see that Splunk is probably the most disruptive technology out there for big data. A lot of the traditional big data warehouses work on specifically formatted data that fits into a nice little box. Well, big data by definition doesn't fit into a small box, it fits everywhere. It comes from everything. Every computer has it, every device has data. We combine all that, it doesn't fit into a little box. So when you have something like Splunk where you can just do a search for a keyword and come up with 15,000 lines of related information, that makes it disruptive because you can actually correlate and you can view all that information in one view. You know, at traditional big data warehouses, you don't have that ability to do that. For example, if you have a student that logs in at the University of Phoenix and you type in their username, you can see every step that they take through the application from firewall to load balancer to application to database all the way back out and get individual timings all the way back. So that kind of information is very important to our developers to know what classes the users are taking, where they're from, what's their retainment, how long have they been logging in. So that kind of information is really good for Apollo Group to know how their students are doing and what their demeanor is and how long they're staying in the classes. Interesting. I mean, are there any examples you could share with us, some really interesting correlations or insights you've been able to gather because now you have this view and can basically cross check data against many different sources of other types of data. We're some of the real kind of insights that you just never could have imagined before. So a really good use case was showing the adoption of Android versus iPhone and also showing which classes were more widely adopted online versus or on mobile platform versus traditional online. And so we can actually see that humanities and the non technical classes are mostly being taken on mobile platform versus from a desktop. So our business people can actually sit there and say, OK, well, let's market this towards this group of people or let's market this towards this group of people. And this is not just the registration for the classes. This is the actual taking of the classes. Absolutely. And you can see that breakdown by type of class versus platform that they're actually consuming the content. Yeah, that's pretty awesome. I mean, we can we can do a lot of things with Splunk. Yeah, that is that is amazing. It's funny that, you know, again, we're here at the cube at Splunk 2012. We invite you to join the conversation. The hashtag is data journey. And it's it's amazing to me, all the people that we've had on Jeff have gone on a data journey. The fact that as you just said that you can kind of discover things that you couldn't discover before that you type in a couple of keywords and you get some interesting correlations that come out and that you can then again, go on this discovery journey to find out what is meaningful, what is it meaningful. Yep. It seems to be kind of a spot on a spot on theme for the show that really illustrates what people are doing and what the technology enables. Absolutely. And that example you just provided, I think is a good one because it shows the you know, these aren't just necessarily it decisions. These are business decisions that it's impacting, right, which is, you know, taking it to a whole new level. Yes. You know, certainly not to diminish the importance of the IT decisions. But when you get into the point where, you know, other groups within the company are making key decisions based on a tool that was initially brought in potentially for security or IT monitoring specifically. I mean, that's just getting them much more value out of your investment and really helping the whole business. Yes, it is definitely. And are you seeing to kind of the other, you know, kind of knock on big data, I guess, not really knock, but limitation is, you know, who can really use it, who can really interact with it, who can really drive it. And, you know, you're explaining pretty simple, straightforward tech searches and getting access to this stuff. So our aren't probably one of our another big steps that we're going to take is integration into Hadoop, as well as some of our business, our data warehousing. So we're looking at integrating with multiple databases, pulling information out, and also integrating with our CMDBs. So we have a lot more intelligence behind behind the logs. And so, you know, our data journey has gone quite a distance in the last two and a half years. But over the next two and a half years, you can do some amazing things. Wow. That's great. So thanks again, Rick. I appreciate you coming on. So again, Jeff and Jeff on the cube at the Splunk Conf 2012. The cube is SiliconANGLE's premier video event where we go out to the tech events. We talk to the people you want to hear from. We get the information that you want to hear. We go right to the insight beyond the press release, beyond the pre-packaged stuff and just have a conversation with people and learn about how they're using these technologies. So we are going to take a break and we're informed by the crew that they're going to feed us. So that's good, Jeff. I don't want you to pass out for lack of nourishment. So we're going to take about a 20 minute break, but we do have a bunch of exciting folks lined up for the balance of the afternoon. So we hope to see you back later.