 Live from San Francisco, extracting the signal from the noise. It's theCUBE covering Nimble Storage, the power of predictive analytics. Now your host, Jeff Frick and Stu Miniman. Hey, welcome back everybody. Jeff Frick here with theCUBE. We're in downtown San Francisco at the Nimble Storage Predictive Flash Launch. A couple of hundred people really excited about really applying big data and analytics to storage and infrastructure. So we want to come up, get the latest information, find the smartest people we could find and pick their brains. So I'm joining this segment by Stu Miniman from Wikibon and Scott St. Clair, Senior Storage Analyst from ESG. Welcome. Thank you, thanks for having me. This is great. So what did you think of the presentation? We had a couple of talks a few minutes ago. What was your takeaway? I loved it. You know, it's a huge launch. There's so much stuff going on. You know, I've joked with some people here that just one or two of the sub-bullets could have been a launch for other companies. I mean, you're talking about predictive flash and all-flash array. You're looking at it to scale out technology. It's got encryption, deduplication. It has the ability to pool with Nimble's hybrid arrays. There's so many different things that are going on here. Great launch, really interesting technology. It's a great blend of like flash, plus softwares eating the world, right? Because at the end of the day, it's that software that really enables the opportunity to have that out there. So I guess, Scott, if we help the people watching this, this is not a me too. This is not the, oh wait, we are now the 14th company that has an all-flash array. What do you see is really differentiating Nimble and why is this just not kind of catching up to some of the other people? Well, and that's an excellent point. You can really tell that Nimble tried not to do the, oh well, I guess we got to do an all-flash array because everyone seems to be doing it. They really tried to do something different and I think they achieved it. A couple things that really stood out is the ability to create a data pool where you're able to mix and match the all-flash arrays with their existing hybrid arrays. And what that allows is you can actually take workloads that exist on their hybrid array and move them while the entire system and data's online over to the all-flash array. So, simplifying management, speeding up performance and really adding flexibility. So if you're in an environment and you're not really sure do I want to go hybrid, do I want to go all-flash, I don't have to default to the most expensive one. I can go hybrid and move to all-flash if I need to. So they move it back and forth or just making a kind of a migration move to better performance? From what they talked about today, it looked like you can move it back and forth. Back and forth based on whatever the demands are. Yeah, based on the demands and based on the needs. Yeah, that's pretty good. What I find is interesting is we like to have our definitions, especially those of us in the analyst world. We have to have here's our bucket and here's what a technology is. Well, Nimble might have been a hybrid company from their technology, but now they do both. And even that ability to migrate and move between different environments reminds me of what the hyperconverge guys have. So I can have a capacity optimized solution. I can have a performance optimized solution and I can move between those nodes. So some of the same value propositions that I see from the simplification of hyperconvergence I'm seeing from what Nimble can do here. Exactly, and that's what we're talking about is it's all about simplification and flexibility. And that's what they're doing with their software. I mean, all flash is great. Everybody loves all flash and our research has shown that too. And it makes sense. But the flexibility and the simplicity really adds to so much benefit or provides so much benefit. Yeah, so Scott, what does your research show you as kind of the economics here? Do you expect to see wildly swinging towards the all flash or is it a slower move between the hybrid and the all flash? So we've looked at, we actually did some research, 2015, looking at the storage market. And we looked at a number of different technologies including flash storage. In all flash, what we can say is all flash is an established market but it's not a mature market. So a lot of people may look at Nimble's launch and say, hey, are they late? Did they miss the curve? I mean, I think our numbers showed that around of the organizations that we interviewed North America, I think we saw about a third of organizations had already deployed in all flash. So that's fairly significant. But those numbers looked to only increase. In addition to that, our research has shown when we ask people, what's holding you back from deploying flash? What we find is there is a complete consensus in the fact that everyone loves flash. The only thing that holds them back is price. And if you look at what Nimble's trying to do here is trying to move down the effective cost of flash. They talk about, I think they announced here somewhere between a 33 to, or a one third to two thirds reduction in TCO. Now, assuming that those numbers are holding, what you're seeing is Nimble's really helping bring down the effective cost of flash. And if you can do that, that's just gonna increase the market size because it's gonna lead to more organizations being able to adopt flash for their environment. And do you see customers pre-building out ahead of the curve knowing that Moore's Law is Moore's Law. It's been cranking along for a while. That flash is going to come down. There's going to be more efficient ways to use flash. And building applications in advance of that, just anticipating the better ROI as those prices come down, the effective prices come down. I think we are seeing that. I think it's a combination of, I don't know if people are doing it purposely, but there's so much being done in the application space and using data, how we can use data more intelligently through big data analytics or business intelligence. Companies are being forced to use their data smarter to be competitive. And so what that's doing is it's putting more pressure on the storage environments and companies like Nimble are responding. So whether or not it is someone getting ahead of the curve and saying, hey look, these technologies are coming, I think we can make our apps more demanding or they're being driven by their competition to just be smarter in how they can use data more to drive their business better and they need a solution. Whatever is driving it, the bottom line is organizations are finding that they need flash and they're deploying as much as they can afford. So Scott, where do you see Nimble really fitting in the ecosystem as a whole? I guess from my standpoint, I like what they said about it's lowing costs and making a better experience using those analytics that they have. But storage is only a piece of the puzzle and talking about how they fit in converged infrastructure, their cloud play is mostly that they are supplier to some service providers like iLand who they have on there. So where do you see them competing against, fitting into some of these big waves of change that are going on in IT? In terms of the larger storage ecosystem, I still see them as a transactional workload centric storage provider, right? And some people call all flash arrays or hybrid arrays or SAN arrays or NAS arrays, whatever it is, the bottom line is it's an array designed to handle transactional workloads. Now, one of the things that surprised me in here in addition to some of the things we talked about is they added a scale out kickboard. So while I would have said, hey, that's great, they have a very usable, very effective all flash array, when you add scale out in there and the ability to get to a million IOPS plus, whatever that is, it actually opens you up to some net new markets in what type of workloads you can go after. All that being said, you're still in that transactional space. You're not in a cloud space, converge, yes, hyperconverge, no, some other spaces, maybe not. But at the end of the day, that space I just described, that transactional workload storage, that still is pretty much the dominant space in the industry right now. Another area they touched on is the financial transactions. So Nimble says that they're allowing customers to either buy from a CapEx model or move to more of a storage on demand model, which sounds a lot more like what we expect to see in the public cloud, some of the service providers. What are you hearing from users? What do you think of what Nimble's announcing here? Is this game changer just part of the story overall that's interesting? There are a number of organizations that have gone this direction. I love it, by the way. I love flexibility. Simply because what we've seen is, you would think everyone would love the cloud model. It's about OpEx, I only pay for what I need. But what's interesting is, when actually talking to customers, some organizations are actually not budgeted to handle transactions that way. I mean, everybody loves it until the CFO says, wait, I didn't plan for that, right? Exactly, exactly. And while just their budget cycles are saying, no, I don't know how to bill you on a monthly basis. You have a quarter of a million dollars this quarter that you have to spend. That's how it works. And so if you're, depends on your budget cycles, based on how your business handles budget cycles, you make, prefer one versus the other, so providing flexibility tends to be the better way to do it. Yeah, as always, we say in the industry that the technology tends to lead the way because you've all got all those political and operational issues, the inertia in the enterprise. So I guess, let me talk about that. The inertia in the enterprise, there was a bold statement from Gartner saying that you cannot keep doing things the way that you're doing it. And something we say often, but for Gartner to say that is the status quo is no good. So our enterprise is getting it to the customers that you're talking to. What does your research say about companies moving forward, being, I guess, more aggressive into finding new ways of leveraging their infrastructure and having IT be a tool to drive the business, not just something that was a drag on the business? Wow, bunch of points in there. So number one, I think they're getting it, right? And what I mean by that is that the current status quo is unsustainable. The big siloed arrays are kind of just having this general purpose 15K spinning drive array doesn't make as much sense anymore, especially at scale. And what we've seen is, despite the fact that a number of organizations are getting that, inertia still exists. For example, one of the big things that I think has really held back some adoption of flash technology is there still is this paradigm of evaluating storage on a dollar per gig basis. Oh, well, disk is X dollar per gig, but flash is a higher dollar per gig. When in fact they're ignoring the IOPS per dollar that they're getting. And what they can do with those IOPS. And the fact that if you eliminate the bottleneck that storage provides, servers can do more. Your applications can do more. And actually you get, and our research has actually shown you're actually, after deploying flash storage, a number of organizations are actually able to get this ripple effect to where they get more benefits through the remainder of their IT environment. And so organizations are starting to get it, but still it's always important to remember that for whatever the analysts like myself say, or Gartner, or anybody else, the storage industry is still very conservative space. You're talking about people's data. And IT organizations will probably move a little slower than they probably should. I mean, both storage and networking, the thing you always say is, if you save the company a million dollars you get a pat on the back. If you lose something or the company goes down you're out of a job. The company might be out of business. So we understand a little bit why there's some hesitancy, but we're at a point. I mean, you talked about these technologies, while they may not be prevalent everywhere, they're established, they're understood. Lots of companies have done it. Nimble's got 7,500 customers. So this isn't some unknown startup that we say, ah, can I trust it? It's now something I can trust these technologies. And that's an excellent point, is if you're afraid of using Flash Storage because you can't trust it, well then you are in the vast minority. Everyone is already using this. And honestly, that should not be an excuse anymore. Yeah, I'm curious about what your take is. Nimble said that we were about to enter a new era of predictive Flash. Do you think that's the next wave? I think Nimble would like it to be. That being said, I love the predictive Flash technology and I think a number of organizations are predicting themselves as the next wave. I mean, you see this a lot. I think predictive Flash creates a tremendous amount of value. I definitely think that this moves Nimble into a leader's section of not only all Flash players, but transactional storage system players. That being said, I think the next wave of storage is a mismatch of the cloud software defined, hyperconverge and Flash, all basically we have in storage a plethora of options that we didn't have before. If you think about it, eight years ago, the fact that we had an array that could do both San and NAS was a big deal. Like that was huge. Now we have cloud, we have software defined, we have object, we have converge, hyperconverge, all Flash, hybrid, a huge number of different options. And I think if you want to say what the next wave of storage is, it's going to be just the fact of marrying all those different options up with all the different workload demands. Exactly. I mean, that's what the smart guys, you guys got to figure out. It's, you know, how do I start doing my costing on that piece of infrastructure that's powering that application versus just the cost per gigabyte, cost per terabyte, it's interesting. So you've mentioned before, like with Amazon, you have to change the way people buy storage. I'm just buying a stack of capacity. I'm not buying gigabytes. I'm not buying terabytes. So how are the house, I always think of cost accounting 101, right? You never really know how much anything costs. So how do people start to rethink the infrastructure to match it more with the applications and really bring those into alignment? So the appropriate infrastructure, appropriate storage is used for the appropriate application, which is a higher value application than this other application. So yeah, it warrants the higher performance flash infrastructure. Well, and to add on to the challenge that you just brought up is, to a large extent, it requires you to predict the future because you're not just predicting what your apps are doing today or understanding them. You got to know what they're going to need three, five, you know, two, three, five years from now. So that's where the flexibility comes in. You know, I kind of started talking about that, right? Is the ability to mix and match hybrid and all flash. What you need is, and I think this is where Nimble's driving to deliver, is solutions that allow you to easily adapt your infrastructure as the needs change. Because at the end of the day, no matter how good you are at planning, you're not going to get it 100% right. So the more adaptable and predictive analytics that you can get out of your infrastructure like what Nimble has, especially when you include what they offer with InfoSight and the analytic capabilities from that and the consulting capabilities of that, the better you're able to tune your environment, the better off you'll be. All right, well, I think we're about out of time. I was going to give you the last word, but that was a pretty good last word in the last little bits that you're going to be working on the next six months. What are you excited about? You know, I'm just excited about everything that's going on in storage. Excited about this, we're working on, I'm doing a tremendous amount of analysis on the software-defined storage space, which I also think is very interesting, but honestly, just an exciting time to be in storage. Awesome, well Scott, thank you very much. Absolutely, thanks for stopping by. It's been a few minutes. All right, I'm Jeff Freik with Stu Miniman, Scott Sinclair, Senior Storage Analyst from ESG. You're watching theCUBE. We're at the Nimble Predictive Storage Launch. We'll be back with our next guest after this short break. Thanks for watching.