 from the computer museum in the heart of Silicon Valley, extracting the signal from the noise. It's theCUBE, covering OpenStack Silicon Valley 2015. Brought to you by Morantis. Now your host, John Furrier. Hey, welcome back everyone. We are here live in Silicon Valley for theCUBE. This is our flagship program from SiliconANGLE Media, where we go out to the events and extract the signal from the noise. I'm John Furrier, the founder of SiliconANGLE. I'm joined with Lachlan Evanson, team lead cloud platform engineering at Lithium, and Joe Sandoval, director of the cloud platform. Welcome to theCUBE guys. Thank you. Lithium, great company, been following for a while, doing really well, but grew out of the forum software. Now you guys doing community software, advocate software, whatever they call it these days, there's always a buzzword for it. But basically you guys are powering software through the largest enterprises out there for community, social data, interactions, persona, for folks who are driving blogs and marketing and all that stuff for companies. Customer advocacy, did I get that right? Yes, definitely. We're connecting companies with customers. Engagement data, so all this stuff is awesome. So we talk to the big players like Bapachiana, who's SVP at IBM, reports to the CEO. And he talks about systems of records, systems of engagement. This is the hottest thing in the planet right now, systems of engagement. But there's a little nuance. It has to be real time. It has to be fresh, not stale. So cloud can be a real opportunity. So how are you guys deploying cloud at Lithium? You guys do some very interesting things around hybrid, public and private. Share with what's happening with your deployments. What are some of the things going on? And how are you guys tackling this? You have business right now that you're doing with large customers. Yes. And you sell software as a license and probably some other SaaS products. Yes, yeah. Multi-tenancy, performance, real time. You guys are in the wheelhouse of all the hot action. So what's going on? Yeah, you know, Lithium, I mean, it's all the things that you described. I mean, there's definitely some challenges, you know, as we're trying to scale. I mean, I think it's a lot of the same things that, you know, other people in like operations are kind of dealing with. But we definitely have really, you know, taken an approach where, you know, there's, as the cloud space has matured, we really wanted to be able to kind of leverage. Like we knew what our use cases were and we wanted to be able to, you know, leverage, you know, public cloud as a vehicle to really, you know, help, you know, drive forward and move quickly. But as well as like we had, you know, some of our product evolving in the data center that, you know, we understood that private cloud, you know, OpenStack was a way that could help us to be able to transform and really, you know, enable our devs and be able to provide that self-service platform. So we kind of took in a really like, let's really take kind of like the best of both worlds and like try to leverage like what, you know, what is leading in public and as well as like, what's leading in private cloud in the data center. So you guys got some cutting edge DevOps going on, right? And I've been asking every guest on the program here yesterday and today and I'll ask you guys to answer individually, does hybrid cloud really exist? And if it is, is it a product or is it just a combination of public and private? Because you guys are doing both Amazon and OpenStack, which essentially you could argue is hybrid cloud, but you don't buy hybrid cloud, you engineer hybrid cloud, I'm guessing. But what's your take on it? Does hybrid cloud exist? Is that a marketing buzzword in your mind? Is it an outcome? Is it like distributed computing? What's your take? Sure, well, you know, I mean, I know, I know early on when I got into like this OpenStack journey, there was a lot of marketing stuff around it. I mean, there's a lot of like messages about like, hey, you could cloud burst, you could kind of get out there and you know, use, use, you know, use Amazon as a way to kind of burst and deal with those spikes. Really when we started really getting into it and started kind of understanding some of the challenges there, I think we really started, we stepped away and just realized that each one has a really specific use case for us. And it was like, let's, let's use the right tool at the right time is how we kind of looked at it. So yes, we are taking kind of a hybrid approach. We will use both and we are constantly evaluating like, you know, because we're very data driven and we want to make sure that, you know, we're very efficient in regards to like, you know, if it makes more sense for us to lean more towards public cloud, we will, we will allow ourselves to kind of like leverage that. So I think it's really more of that for anybody in operations that you should really take a really holistic look at it, you know, like is that public cloud platform is strong. But, you know, as far as like for us, we just felt that it's just something that we had to be very discreet about, like where we decided to deploy stuff. But you engineered that, that's, you essentially have a hybrid cloud now, but if you want to call it that, I mean hybrid, it's a little public, a little private. Yeah, yeah. I think one of the things to add to what Joe's saying, I think it's really picked up again with the introduction of containers. And if you actually see what Solomon Hike said, you know, you extract the platform. So each of these platforms have different idiosyncrasies and different ways to provide network compute and storage. But when you actually overlay containers on top of them, it levels the playing field and you're no longer worried about the platform. So things like hybrid cloud and moving workloads, when you really look at a hybrid cloud, is it bursting? Is it having workloads running in both? Is it, you know, having workloads that can be moved from one to the other? I think the container story actually picks it up a notch again and allows you a single pane of glass with things like Kubernetes to actually span multiple clouds. So orchestration is really key there. The developer framework with Docker enables you now to have that traversal, if you will, but then the two environments. How does that change your cost perspective? I mean, obviously Amazon, some say it's more expensive as you get bigger in it, but then you want to break it to private. Some people have been all in on Amazon then brought it on premise and then go back to Amazon, some have hybrid. Like you guys, do you guys find an advantage of having the hybrid? Does it help you guys from a workload standpoint, performance and cost or one or the other or both? Or can you guys share some insight on that? I mean, that's kind of been a really evolving thing for us. I mean, because we've wrestled with that constant, you know, my boss has done a great job at really like he crunches the numbers and constantly challenges me of like, you know, like, hey, I think our arc is this direction or maybe going back this direction, depending on like the size. But one of the like the biggest gating factors that I found on like this journey was just, a lot of the tooling around it was really that limiting factor. Like we weren't even ready for that discussion. So that's kind of where I was kind of pushing back. It's like, until we started seeing the emergence of like containers and Docker where we then I was like, wow, this hybrid vision that we kind of like thought about where then we really could allow ourselves to kind of leverage like ones being cheaper than the other. It provided that platform to give us that. You know, I think about your business a lot because we have the crowd chat container. I showed you guys that last night's analytics and it's all built in Amazon. It's all in the clouds, all real time. Sure. And you guys have a different business model in the sense that you're doing a lot of enterprises where it's a license and they have their own data. So, and I know a couple of your board members, Pete Sonsini and Peter Fenton, both are like, do more, do more in larger scale. So you guys are actually setting up pretty nicely from a scale standpoint. But now you got to get into the kind of competitive advantage strategy of saying, okay, I have a customer called Big Enterprise. They have their community data. It's their advocates. It's their influencers. It's their customers. They don't want that customer data in the main cloud, maybe not, but you might want to run real time analysts with cloud or whatnot. You have cloud and some other cool things at Lithium. So do you find that flexibility to be one of the driving factors or is it just you're looking at it from a pure futuristic standpoint? Do you guys actually have those conversations like, hey, we have to mind the multi-tenancy issue? And do you do that on public or do you do that on private? So this is kind of the nuances that you guys face as a cutting edge platform. So is that a conversation or is that just more of a future scenario? I mean, I think it's one of those things where us being able to kind of have that choice really gives us flexibility with our customers and knowing that we have multi-tenancy and that we want to make sure that we secure and isolate and that's kind of what having open-stack. You probably want to do that on private cloud. Right, and that's exactly what open-stack kind of gives. We can do that. So depending on our business requirements or things that go, yes, we don't really think so much on that level as much, but it is considerations that we had like all our decision-making. We want to have that ability where we have that choice to be able to move data internally where that's the driving factor. We're seeing a lot of companies, and I'm only going to imagine this scenario because I talked to a lot of other companies where they say, hey, you know what? I got the on-prem thing or private cloud because we have business issues where there's multi-tenancy, customer data, some compliance thing, but then they say, hey, I can go into Amazon and spin up basically a mainframe like functionality, massive compute to do some analytics, to do some things in real time. So this is the kind of the new way we're kind of hearing hybrid being discussed. Are we there yet? I mean, you guys are probably one of the most advanced. We've talked to you with this hybrid. How do you see that? Is that a nirvana or is it we close or? Sure. I mean, it's an interesting proposition. And if you take a look at all these container orchestration environments, what they actually do is schedule resources and maybe they are Hadoop jobs, maybe they're production workloads. But the real secret source in there is if you know where your resources are and how they're running, you can schedule around them and traditionally you build infrastructure to run Hadoop and you build infrastructure to run your cloud. And they've always been segregated, but with the container paradigm and looking at container orchestration, we can actually say where are these workloads running? Where have we got free cycles? Run your Hadoop workloads in those free cycles or schedule them at these time, but you actually bring them into a common infrastructure. So, you know, you have a lot more leverage of this larger resource pool than smaller discrete resource pools. So it becomes a lot more opportunity. I'm sure you can reduce costs too. I mean, you don't need a lot more sysadmins, probably reduce personnel count if you're going to be orchestrating that level of workload management. It's almost like push button IT. Yeah, yeah, no, definitely that's, those are the net gains that we get is that, you know, the team is able to be able to, you know, scale out large, you know, into a larger infrastructure, but without like increasing the headcount. I mean, we want to stay, you know, very efficient and optimal. Okay, so here's the most important question I want to ask. So, is there a way I can jump my cloud score up? And you guys go in and database and move me from a 66 to a 89. I think you best ask Justin Bieber because I think he still holds the highest cloud score. So if you can get a response from him, just tweet him, I'll show you get a response. Okay, I'm sure you guys probably might not, if you can't answer this question, still case they can't answer it. But a lot of people are saying, you know, cloud scores represent just frequency, how many times you tweet and influencers are a big thing right now. And you guys have probably the most comprehensive software in the influencer market because you have customers that run their advocacy programs on Lithium, here's the number one, I think the number one vendor in that space right now. So as the world moves into the crowd, as move worlds into Twitter and multiple platforms, influence has context. Sure. How do you guys look at that? Is that in your department? You guys look at that? I mean, that's, you know, if you want to talk DevOps, you want to talk cloud, I mean, those things I could talk all day. That's a little bit outside of my realm. Yeah. Yeah. Okay, so let's talk DevOps then. Okay, just want to get out there. So follow up on that with the Lithium folks. DevOps. So Amazon has elastic beanstalk, great products, doing well. Is there alternatives out there on the private cloud that beanstalk could be there? And on the database side, DynamoDB on Amazon is pretty solid. You got beanstalk, you got Redis and normal stuff in there. Is OpenStack ready for prime time in that kind of container model, application container, or is that going to be a Docker or is it homegrown? I mean, well, you know, what I'm seeing is like, you know, like even for us is, you know, yeah, Amazon has like just a great set of services. We, I mean, we feel like, hey, it's a great platform to be using. Honestly, I mean, if I had to give, you know, advice to like my peers and people like making this journey, you know, I would really encourage them like, you know, use public cloud. If you have no, no getting factors meaning you don't have to have data residency in the data center or these factors, you know, use those platforms. But, you know, for, you know, for us as well, like when we get it, when we started bringing things back into OpenStack, you know, it kind of like also, I think we wanted to not be so benderlocked as well. We wanted to be able to have some choice that. Well, you have business criterias that you have to make. You have to be mandated. I mean, you got data governance problem, each, you know, these customers you have. You know, definitely customers in Amia, you know, we have to, there's considerations there and who knows how these things are going to change. And so at least us having that competency that we know that we've kind of made that journey. Now we're in a position to really, you know, pivot quickly when as needed, depending on like business. Okay, so what are you guys locking your teeth into right now on OpenStack? Because OpenStack is growing fast. We've seen some successes here. You guys are certainly a great case study here in terms of folks on the trenches, front lines doing some great work. What are you guys sinking your teeth in now? What are you guys working on now? What's the next step for you guys in your DevOps, OpenStack, Private Cloud, Hybrid Cloud rollouts? Yeah, so, you know, being in the container journey right now, so we're on the forefront, we're focusing on leveraging the platform we've built in OpenStack and using that as shoulders to build our container platform. So, you know, we're heavily looking at containers and how we can utilize OpenStack and the frameworks it provides to make a great container experience for our developers, which are our customers. And how about your internal developers and they're using OpenStack and Amazon? Yeah, so we have, you know, like depending on like what part of the product it is, we have some using Amazon, some using OpenStack. A lot of times, you know, depending on like where we're at with like, because we tend to run very, you know, efficient. So we find that, you know, hey, Amazon's a great place to like kickstart a new project, a new service, and then we could decide later, maybe at times that it may need to roll back into the data center, as well as like, you know, as our markets expand, you know, we like that. There's a way that we have to, you know, we have some key data centers in place. There may be other regions that emerge where we can really, now that we have kind of a framework to kind of like jumpstart an AWS, or we have a template really to get going with OpenStack, that, you know, we may not have to build another data center. So we like that we have that as an option. It's great flexibility too. You guys have a competitive advantage now. Yeah. To spin up new apps, new functionality. Yeah, yeah. And get close to our customers. All right, and final question. You guys kicking the tires on Kubernetes at all? Looking at the orchestration side, you guys using Kubernetes? Absolutely, yeah. I just gave a presentation on how we're using Kubernetes and how we're utilizing it with OpenStack. So I'm excited to see how that project goes, but you know, we're on a path to working heavily with Kubernetes. And your take on Kubernetes, positive, negative, neutral? Absolutely positive. Yes. In the three months that I've been working with it, it's absolutely been fantastic. All right. Well guys, thanks for joining us on theCUBE. Thanks for the insight. Congratulations. Lithium, hot company growing company in San Francisco, powering a lot of the community's advocates offer out there. This is theCUBE powering the data and the signal here at OpenStack. So we'll be right back after this short break.