 Live from Washington, D.C., it's theCUBE. Covering.conf 2017, brought to you by Splunk. I think I've seen you guys every year. And welcome back here on theCUBE. John Wall is at Dave Vellante. We're in Washington, D.C. for .conf 2017. Splunk's annual get together coming up to the nation's capital for the first time. This is the eighth year for the show and 7,000 plus attendees, 65 countries. Quite a wide menu of activities going on here. We'll get into that a little bit later on. We're joined now by a couple of gentlemen, Michael Arcella, who is the Vice President of Engineering at Lassian. Michael, thank you for being with us. Thank you. Actually, as Director of Business Development. Oh, Director of Business Development. My apologies. Three's doing a great job. I don't need that. Very good. And Brooke Gavit, who I believe is the VP of Engineering and the Chief Software Architect at 4850. No promotions or job assignments. I've got you on the right path there. Good deal. All right, thank you for joining us, both of you. First off, let's just set the stage a little bit for the folks watching at home. Tell us a little bit about company descriptions, core competencies, and your responsibilities. And then we'll get into the intersection of why the two of you are here. So Michael, why don't you lead off? So at Lassian, in our simplest form, we make team collaboration software. So our goal as a company is to really help make the tools that companies use to collaborate and communicate internally. Our primary focus and kind of our bread and butter has always been making the tools that software companies use to then turn around and make their software, which is a great position to be in. And then increasingly, we're seeing ourselves expand into providing that team collaboration software products like JIRA, Confluence, Bitbucket, and now the new introduction of a product called Stride, which is a real-time team collaboration product, not just for technical teams, but we're really seeing a great opportunity to empower all teams, because every team and every organization needs a better way to communicate and get things done. And that's really what at Lassian's core focus is all about. Gotcha. Brooke, if you would. So 4DF50 Labs, we're the software development and DevOps-focused subsidiary of Airstore Systems based out of Atlanta. And we focus primarily on four key partners, which would be Atlassian, Splunk, QA Symphony, and Red Hat. And primarily, we do integrations and extensibility around products that these guys provide as well as hosting, training, and consulting on DevOps and Atlassian products. I mean, so the ideal state in your worlds is you've got true DevOps, agile, infrastructures code, I'll throw all the buzzwords out at you, but essentially you're not tossing code from the development team into the operations team who then hacks the code, messes it up, points finger, all that stuff is in part anyway, what you're about eliminating and getting to value sooner. Okay, so that's the sort of in-state nirvana. Many companies struggle with that, obviously Gartner has this term bimodal IT, which everybody criticizes, but it's sort of true. You've got hybrid clouds, you've got different skill sets. What is the state of agile development, DevOps? Where are we in terms of organizational maturity? I wonder if you guys could comment. Yeah, I'll start with that. I think even though we've been talking about DevOps for a while and companies like Atlassian and Splunk, we live and breathe it, I still think when you look at the vast majority of enterprises, we're still at the early stages of effectively implementing this. I think we're still really bringing the right definition to what DevOps is, right? We kind of go through those cycles where there are buzzword gets hot, everybody glams on to it, but no one really knows what it means. I think we're really getting into that, truly understanding what DevOps means. I know we've been working hard as Atlassian to really define that strong ecosystem of partners. We really see ourselves as kind of in the middle of that DevOps lifecycle and we integrate with so many great solutions around monitoring and logging, testing, other operational softwares and things of that nature to really complete that DevOps lifecycle. And I think we're really just now finally seeing it come together and finally starting to see even larger organizations, very large Fortune 100 companies talk about how they know they've got to get away from waterfall, they've got to embrace agile, and they've got to get to a true DevOps culture. And I think that's where Atlassian is very strong. Devs have loved us for a long time. Operations teams are really learning to embrace Atlassian as well. And I think we're really in a great position to be at that mesh of what truly is DevOps as it really emerges in the next couple of years. And Brooke, people come to 4850 and they say, oh, you teach me how to fish in this DevOps world, is that right? Yeah, absolutely. I mean, one of the challenges that you have in large enterprises is bringing these two groups of people together. And one of the easy ways is to go out and buy a tool. I think the harder and more difficult challenge that they face is the culture change that's required to really have a successful DevOps transformation. And so we do a little bit of consulting in that area with workshops with folks like Gene Kim or Gary Groover, Jess Humble that we bring in who are sort of industry icons for that sort of DevOps transformation to assist based on our experiences ourselves in previous companies or engagements with customers where we've been successful. So the sort of cloud native guys, people who are doing predominantly cloud or smaller companies, tech companies, presumably have glommed on to this. What about the sort of, you know, the Fortune 1000 or the Global 2000? What are we seeing in terms of their adoption? I mean, you mentioned waterfall before. You talked to some application development heads will say, well, listen, we got to protect some of our waterfall because it's appropriate. What are you seeing in the sort of traditional enterprise? We see the traditional enterprise, you know, really embracing agile in a very aggressive way, right? And obviously they wouldn't be working with Atlassian if they weren't, right? So our view is probably a little bit tilted because the companies that engage with us are more open to that. But we're definitely seeing that be far and away the vast majority. And the reports that we get from our partners like 4850 Labs is that increasingly larger and larger companies are really aggressively looking to embrace agile, bring these methodologies in. And the other simple truth is with the way Atlassian sells our, with the way we sell our products online, we have always sort of grown kind of bottoms up inside a lot of these large organizations. So where officially IT may still be doing something else, there are always countless smaller teams within the organization that have embraced Atlassian or using Atlassian products. And then a year down the road or two years down the road, we tend to then emerge as the de facto solution for the organization after we've kind of spread through all these different groups within the company. It's a great growth strategy, a lot of trying to replicate it. Okay, what's the Splunk angle? What do you guys do with Splunk and how does it affect your business? Yeah, do you want to start? Yeah, sure. So we're both a partner of Splunk, a customer of Splunk and we use it in our own products in terms of our hosting and support methodologies that we leverage at 4850. We use the product day in and day out. And so with Atlassian, we have pulled together a connector that is, one half of it is a Splunk app that's available on Splunk base and the other part is in the Atlassian marketplace, which allows us to send events from Jira service desk, ticketing events over to Splunk to be indexed. So you have a data model that ties in and allows you to get some metrics out of those events. And then the return trip is to based on real time searches or alerts or things that you have, you're very interested in reports. You can trigger issues to be created inside of Jira. Yeah. Well, yeah. I think the only thing to add to that, so definitely that's been a great relationship and partnership. And we're seeing an increasing number of our partners also become partners with Splunk and vice versa, which is great. And then the other strong side to this as well is our own internal use of Splunk. So we, as a company, we always like to empower our different teams to pick whatever solution they want to use and embrace that and really give that authority to the individual teams. However, with logging, we were having a huge problem where all of our different teams were using a whole host variety of different logging solutions. And frankly, not to go into all the details, it was a mess. Our security team decided to embrace Splunk and start using Splunk and really, really got a lot of value out of the solution and fell in love with the solution, which says a lot because our security team doesn't normally like much of anything, especially if it's not homegrown. So that was a huge statement there. And then quickly, Splunk now has spread to our cloud team, which is growing rapidly as our cloud scales dramatically. Our developers are using it for troubleshooting our SREs and our support team for incident management. And it's even spread to our marketplace, which is one of the larger marketplaces out there today for third-party apps. And then the new product Stride for team collaboration is going to be very dependent on Splunk for logging as well. So it's become that uniform fabric. I even heard a dev use a term, which I've never heard a dev talk about logs since talk about log love, which is no PR. That is the direct statement from a developer, which I thought was amazing to hear. Because they just want to code and make stuff. They don't want to deal when it actually breaks and have to fix it. But with Splunk, they've actually, they're telling me they actually enjoy that. So that's a great feedback. So that's more than the answer is in the logs. That's, there's value in our logs. Right? Yeah, a ton of value, right? Because at the end of the day, these alerts are coming in and then we use tools like the 4850 Labs tool to get those tickets into JIRA. I mean, those logs and things are coming in. That means there's an issue and there's something to be resolved and there's customer pain. So the quicker we can resolve that, that log is that first indicator of what's going on in the cloud and in our platforms to help us figure out how do we keep that customer happy, right? So I mean, this isn't just work and just a task, right? This is about delivering customer value and that log can be that first indicator. Sooner you can get something resolved, the sooner the customer's back to getting stuff done and that's really our focus as a company, right? How do we enable people to get things done? Excuse me, when you are talking about your customers, like what are their pain points today? I mean, big data's getting bigger, right? And more capabilities. You've got all kinds of transport problems and storage problems and security problems. And so what are the pain points for the people who are just trying to get up to speed, trying to get into the game and the kind of services you're trying to bring to them to open their eyes? Right, I think if you look at the value stream mapping and time to market for most businesses, where Splunk and Atlassian play in is getting that fast feedback, the closer in to the development side, the left-hand side of the value stream that you can pull in key metrics and get an understanding of where issues are, that actually it's much less expensive to fix problems in development than when they're in production, obviously. And rolling things like Splunk that can be used as a SEM to do some security analysis on whether it be product code or business process early, rather than end up with a data breach or finding something after it's already in production. That kind of stuff, those are the challenges that a lot of the companies are facing, especially when the news, if you look at all the things that are going on from security perspective, taking these two products and being able to detect things that are going on, trends, any sort of unusual activity and immediately having that come back for somebody in a service desk to work on either as a security incident or if it's a developer finding a bug early in the life cycle. And augmenting your sort of infrastructure as code, the build out of the infrastructure itself, being able to log all of that data analysis and look at the metrics around that to help you build more robust, enterprise class platforms for your teams. We've been sort of joking earlier about how the big data, nobody really talks about big data anymore. Interestingly, Splunk, who used to never talk about big data, is now talking about big data, because they're kind of living it. It's almost like same wine, new bottle with machine learning and AI and deep learning are all kind of the new big data buzzwords. But my question is, as practitioners, you were describing a situation where you can sort of identify a problem, maybe get an alert, and then manually, I guess, remediate that problem. How far away are we from sort of the machines automating that remediation? Thoughts on that? Yeah, do you think it's the first step? You guys kind of did that first step. We've done a lot of automated remediation. It's closed loop remediation, you know, it's what you call it. And the big challenge is, it's a multi-disciplinary effort, right? So you might have folks that need to have expertise between network and systems and the application stack, maybe load balancing. There's a lot of different pieces there. So step one is you got to have folks that have the capacity to actually create the automation for their domain of expertise. And then you need to have sort of that cross-platform DevOps mindset of being able to pull that together in the coordinator role of let's orchestrate all of the automations. And then hopefully out of that, combined with machine learning, some of the stuff that you can do in AWS or with IBM's guide out, you can take some of that analysis and be a little bit smarter about running the automation, you know, in terms of whether that's scaling things up or when, for example, if you're in a financial industry and you've got a webpage that people are doing bill pay for, if you have a single website down or web server down out of a farm of a thousand, you know, in a traditional knock, that would be kind of red on a dashboard. And so it's high, it's low priority, but it's high visibility and it's just noise. And so leveraging machine learning, being able to do that in Splunk, to really refine what actually shows up in a knock, that's something that I think is compelling to customers. How are devs dealing with complexity? Obviously, you know, collaboration tools help, but I mean, the level of complexity today, versus when you think back at client server, is the orders of magnitude greater for admins and developers, and now you've got to throw in containers and microservices and the amount of data, is the industry keeping pace with the pace of escalation of complexity? And if so, how? I think we're trying. I think that's where we come into play. I mean, as this complexity increases, really the only way you can solve it is through better communication and better tools to make sure that teams have the right information at their fingertips. You know, the other challenge too, is that now in the world of the cloud, these teams need to be on 24-7, right? But you've got to kind of roll across the globe and have your support teams in different time zones, so you don't always have all the right people online at the same time to be able to address, and you can't always talk directly, so that's where having the right tools and processes in place are extremely important so that that team can know and know where. What did the team earlier do? How did they resolve this? Where's the runbook for this issue? And if this happens, how do we resolve it and how do we do so quickly? I think that tooling is key. And also too, this complexity is also, as you guys were talking about before, being solved through some automation as well, and we're increasingly seeing that to where if this occurs and a certain thing occurs, then JIRA can automatically start to trigger some things for you and then report back as to what it did. And you're going to see more and more of that going forward as these models become more intelligent and we can redeploy, or if capacity is low, let's pull back resources and let's not spend all of this money on cloud computing platforms that we may not need because utilization is low. You're seeing all of those things start to happen and JIRA as that workflow engine is that engine that's making those things happen in either an automated way at times or just enabling people to communicate and do things in a very logical fashion. As ecosystem partners, how do you view the evolution of Splunk? Is it becoming an application platform for you? Are you concerned about swim lanes? I wonder if you could talk about that. Yeah. I personally, I don't see any real concerns of overlap between Splunk and Atlassian. And our view at Atlassian is we tend to work very closely with people that kind of fit into that front of me category and they're definitely a partner that we overlap with, I think, in very, very few ways. And if and when we ever do, in a way, that's kind of something we always embrace as a company. One thing we'll say a lot is overlap is better than a gap because if there's a gap between us and a partner, then that's going to result in customer pain. That means there's nothing that's filling that void. I'd rather have some overlap and then give the customer the power to choose how do they want to do it? I mean, Splunk says you could probably do it this way. Atlassian says you could do it this way as long as they can get stuff done and that's always, it's not a cliche from us. I mean, that's a core message from Atlassian. Then we're happy. And regardless if they completely embrace it our way, a little bit, a little deviation, that's not what really matters. Too much better than too little. Exactly. That's what it comes down to. Gentlemen, thanks for being with us. Thank you. We appreciate the time today. I look forward to seeing you down the road and looking as your relationship continues not only between the two companies but with Splunk as well. Definitely. Great, thank you guys. Continue to Cube does live from Washington DC here at .conf 2017. Back with more in just a bit.