 Live from Chicago, Illinois. It's theCUBE, covering Veeamon 2018. Brought to you by Veeam. We're back in Chicago, Veeamon 2018. You're watching theCUBE, the leader in live tech coverage. We go out to the events, we extract the signal from the noise. My name is Dave Vellante, and I'm here with my co-host Stu Miniman, day one of our two day coverage of Veeamon, our second year. Von Stewart is here, he's the vice president of technology at Pure Storage. Cube alum, good friend. Great to see you, man. Thank you, Dave, Stu. Thanks so much for coming on. It's great to be back. So Pure, I remember when you joined Pure, and you're like, Dave, this is going to be the rocket ship of a lifetime. It's turned out to be the case, first company since NetApp to hit a billion dollars in the storage business. It's like independent storage companies are back. Yeah. So give us the update, what's happening at Pure? So, fantastic year, wrapped up in the January, right? So, first independent storage company hit a billion dollars. Actually, we're kind of on the cusp of maybe being the fastest infrastructure company, if not the fastest tied with being the fastest to hit a billion dollars. So the growth rate's been great. The products obviously have been off the charts, whether you're looking at, from an analyst perspective, we know the Gardner reports, the IDC market scape. So if you look at, from the customer's perspective, with the NPS scores, right? Just crushing in terms of the products, the customers stating that we're not overstating the capabilities, and we make some pretty bold statements. But when you kind of bowl it back to where are we at now? I think our focus is helping customers adopt a data-centric architecture as a part of their IT or data center modernization plans, right? This is, we've kind of gone through this phase of like virtualize everything, now everything to the cloud, and now we're starting to mature a little bit, and we're all looking at this tsunami of data that's being created, and it's more around, where's your data going to reside? Because there's going to be some gravity around it, and bring the compute to where the data should reside. And so our products and our strategies to help customers, again, this data-centric architecture, adopt new technologies that are going to help them radically shift how they operate, changing the cost of operations, changing the complexity to either let an existing storage team scale to a larger capacity per full-time, you know, FTE, or to allow actually the application teams to private cloud teams just manage their infrastructure stack, right? We're seeing kind of growth in both areas. I think beyond that, our technology with our evergreen storage as a subscription model has also been able to be transformative for enterprise about how do they acquire, refresh, and introduce new technologies within the storage space, and so it's been pretty exciting. So let's talk about some of that. I remember I've been around a long time, Stu, as you know, so Al Shugart, the legend, once told me when I was just a pup, trying to understand the business. He said, Dave, the storage business is simple. Customers want it to be dirt-cheap, rock-solid, and lightning-fast. This is the days of spinning disk, which were kind of dirt-cheap, and they really weren't that rock-solid, and they really weren't that lightning-fast. So you guys actually delivered on that promise, but you added some other things. Simplicity, the business model of reduced friction, you mentioned evergreen. So it's not, obviously, not just about flash. Everybody said, oh, flash, Pure Storage Flash, we have flash too. It's so much more than that, the way you positioned it just now, the company, in terms of data centricity. And as I say, the business model innovations have really worked well for you. You've been able to stay ahead of the competition. I wonder if you could comment. Sure. So for your audience, I think it's important to maybe back up a bit. Pure was born in the wave of a number of all flash arrays. Yeah. And a fair amount of them were acquired by large existing storage vendors. And I think now that the dust has settled a bit, we were kind of the Phoenix that rose through those ashes, if you will, within the storage space. And I think really the key driver there was, it wasn't about performance. Flash makes everything faster. It was about a combination of the business model, the operational simplicity, but also what I would call the tier one feature sets. You had to deliver on six nines of greater availability. You had to have all the data management capabilities to plug into a large partner ecosystem like Veeam, which, you know, we're here at Veeam Live. And that was kind of what I would call act one of pure, which was, you know, flash array based, you know, storage for your traditional enterprise apps. Last year, we introduced FlashBlade, right? This was a radically different architecture. It was a scale out, a scale out blade based storage platform that scaled capacity and performance linearly. And the adoption in that space has been these next generation apps, which are just the data sets are growing, you know, out of control and beyond what we would have ever expected with an enterprise app. So whether it's AI, machine learning, deep learning, analytics, or this new use case we're seeing, which is a rapid data recovery, the FlashBlade because of the combination of its low latency and massive parallel throughput has really been a big growth vector for us. And it's kind of act two, if you will, of pure storage. Let's go ahead, Steve. Yeah, Vaughn, I'd love to hear more of your thoughts on kind of that application proliferation. So I think back, you know, you and I lived through that wave of virtualization. While I love virtualization, one of the challenges I had with it was I could take my old application that was probably already too long in the tooth and stick it in a VM and then keep it for another five to 10 years because it didn't care about the hardware, their OS and all that same point. Today, talk about cloud native apps, talk about IoT and analytics and all of these microservices and everything like that. It's a huge impact on infrastructure and how we build things. Brings up to speed as how we bridge from kind of the old world and the new world. Yeah, I'm glad that you asked this question. I wish you would be coming to our conference next week. Well, Dave will be there. Because we'll have a session discussing this and it'll include an internal case study and so that's all the details I can give right now. For a long time, I think a lot of IT vendors, particularly those who made hardware products, tried to position this on-prem versus the cloud, right? And it was really the wrong mindset. Cloud is just one more deployment model for an organization to look at. The question that I think organizations have is what fits where? And I think to your point, if you're looking to build a new application or replatform an existing, what you have today versus in the past was you had a contained set of APIs and interfaces to work with, right? If you were building on, say, a database vendors enterprise business suite, that was the tools that you got to use. Today you look at what's available in open source or in the public cloud space and you get to build a massively disaggregated application that's comprised of functions and microservices, right? And it gets to leverage these notions of scaling on demand and being very elastic. What I would share with you and what we discussed with customers is your development team will want to go as fast as they can and leverage all these new tools and they're iterating very quickly and the cloud is an ideal platform for that. But you need to plan and look forward as to around what's the volume of data that you may be dealing with? What's the access requirements to that data over time and where's it going to be a better position? Should it sit in the cloud? Should it sit on-prem? Should it sit in a private to public cloud hybrid type of architectural model and leverage, say, the compute and all the software agility within the cloud and yet still have stewardship over your data and not have to deal with maybe unforeseen things like charges per API call or egress charges, things of that nature. And Edge adds a whole number of things. So I'm grossly simplifying a lot of this, but these are the conversations when you get within the enterprise, which is really our sweet spot. These are real considerations that they have, right? They're past the, is cloud secure? They're past the data sovereignty type of concerns. They're more around, how's this going to scale long term? Because for example, I'll give you an example. So we rolled out Meta, which is our AI as part of our support for our product. It's all getting ahead of the customers and predicting faults, getting them. This is what helps us achieve greater than six nines availability across the entire fleet for the last two and a half years, right? It's getting ahead of the problems. When we work on looking at some of the AI that we create around Meta and we want to test it, we have to download a year's worth of phone home data from the cloud. That takes 45 days to download today. And it's not going to get any faster as the install base gets larger, right? And so those are challenges that you have to look at and say, maybe I started in the cloud but maybe I need to look at something in a hybrid model because it's going to impact my business agility. And so these are conversations that we can have and our architects have with customers based on whatever their criteria or forecast look like. So just about a year ago, Scott Deetson stepped down as CEO, brought in Charlie, new leader. It was kind of interesting. It was right in the heels of Frank Slutman doing something similar. Frank Slutman just stepped down as chairman. So how's the new leadership going? What has Charlie brought? I can't wait to interview him next week on theCUBE but give us your take as somebody who's been an industry observer and obviously a long time peer employee. Yeah, so great question. So just for the audience, Deetson is still with us, right? So he stepped down for being the CEO and is now the chairman of the board. And I owe a large gratitude of debt to Scott. Scott brought me into Pure and I've always encouraged when, you know, every now and then you get that direct email from him, you know, keep being a thorn in someone's side pushes forward. That was a little self-service, so I apologize. But what I like about Charlie is, and understand, I was with NetApp for 13 years, right? And so we did this large growth cycle, not as early as with Pure, but going through a lot of the same growth pains and whatnot that we have today. But we did all that growth under Wormenhoven before they changed over. What was nice about Scott is he told me on day one he didn't know how far he would take Pure, but when he was apparent to him that he had to take it as far as he could, he would find his air. And obviously that Charlie was the choice. And what Charlie's brought in has been a lot of structure, right? The formation of business units, a lot of accountability, a lot of what I would say, that maturation phase from startup, right? That's kind of grown to the maximum output of your current organizational structures to looking forward into a structure that is going to allow us to scale better over time, right? Continue to grow, as well as, I think Charlie would be the first to tell you, you know, Pure's on a trajectory to hit $2 billion and can do that on inertia in the current products, right? Charlie's focus, or one of Charlie's focuses over the last handful of months is, what are we going to become two years from now? And what investments do we need to start making in the near term to get prepared for two years from now? So I brought up Frank Slutman who's in service now because I know Frank and Scott are close, right? There's some board action going on there over the years. They're part of the Silicon Valley mafia with Amir Bhushri, and we can joke about that, but there's a culture of succession that has really taken hold in certain parts of the valley. And again, very similar to what we saw at service now, where the new guy was brought in to take them to the next level. And the existing CEO, you know, mature enough, you know, maybe worked so hard for all these years too, maybe felt like they needed a little break, but still, mature enough to say, okay, I know my limitations and I want to bring somebody else in. So it's been sort of this new thing. And I want to tie it back to something we were talking about before on theCUBE. I mean, you guys hit escape velocity. When you look back at the sort of the virtualization craze with 3PAR and Isilon, Data Domain, Compellent, yeah, they kind of hit billion dollar status, you know, they hit unicorn, but they never hit billion dollar revenue. And so now, and then the other thing you talked about was some of the bigger players decided to buy up flash companies. And they said, you know, rather than pay $2.5 billion for Data Domain or 3PAR, we'll spend a billion dollars or in some cases, hundreds of millions of dollars, and then we'll organically grow that internally. Some, did it work? Maybe, you know, maybe some of it, maybe not. But you guys stayed the course and are now on track to do 2 billion. So here's my question. Long-winded sort of narrative babble, sorry about that. I used to question Wurman-hoven all the time, Gucci even, can you stay independent? That was the big question, you know, because Converged is coming, but now it looks like being an independent is actually, best of breed, is actually still a viable business model. So obviously, I'm not on the inside of whatever the board decisions may be. So let's- Well, you're an observer who knows this business. Let's take that opportunity- We're kind of talking about like Von the prognosticator or analyst, if you will. What I think is different today, and Stu and I were talking about this because we ran into each other over in the corner with Duncan. You know, the emergence of all the flash vendors and them getting acquired and really what's happened by and large is just the same old products just got flash injected into them and, you know, the vendors hope to get another decade out of them. But okay, they're faster, but it doesn't fundamentally change your business model or your operations. Sometimes that's a good thing, right? For some customers, right? They're changed adverse. Right, they don't want that disruption. Yeah. For us, we're trying to usher in now this next wave of shared accelerated storage and it's a disaggregated model, right? Start to look up it at, you know, in a commercial sense, if you will, what are the hyper scalers, you know, delivering? You know, they're not running that direct attached storage or not doing HCI, right? They've got pools of compute and pools of storage and it's either disk and cold or it's flash, flashing hot. And, you know, they've got network and it's all over ethernet, so it's greatly simplified. We're trying to help our customers with that type of architecture, whether they're looking at simplifying their private cloud or extending the private cloud to the public cloud or what's even more interesting as they look at like their data pipelines. You know, a lot of, you know, there's AI and analytics in every organization of every size. They may or may not sit inside the IT department, but they tend to follow that model of, hey, it's in software, so I'm just going to do it on DAZ and then I'm going to build this siloed cluster and, you know, it must be cheap regardless of whatever the efficiency I get out of it and what we're trying to help large organizations look at is data pipelines and flow and the flexibility that you gain by separating compute from storage and not having to worry about the performance issues or constraints of disk based systems from a decade ago because technology is like flash and now with non-volatile memory express and non-volatile memory express over fabrics, right? You're getting direct memory to memory communications from the servers to the storage. So you're getting all the benefits of pooling and sharing your storage with all the benefits of if it was on a local bus in terms of speed and performance. And so it can change, particularly at large volumes of data, it can change your agility. So that is certainly a tailwind for you but it was a tailwind for a lot of companies and you have the product. Let's assume best product just for sake of argument. I'm sure you would agree but best product doesn't always win, right? So what I'm hearing is there was business model innovation. Yeah. Obviously, you had very strong go to market. Yes. You guys knew where all the skeletons were buried all the reps that you guys hired. But there were other factors involved in your ascendancy which maybe is independent of the structure of the industry because the industry structure is changing. It's going from now remote cloud services into this digital matrix and somehow you have to fit into that digital matrix and participate in that. Yeah, it's, I think you brought up two points. So I think if you're going to be a startup to be successful, it's not just technology. You've hit the head on the nail there. Pure had, the technology had to deliver. Pure had that. The business model was innovative. The marketing was off the hook, right? For a startup, we were punching above our weight but you also have Salesforce execution and you never know what you get when you walk into a startup but if you don't hit on all four of those dimensions then you don't achieve escape velocity. In terms of shifting from startup to becoming mainstream, not only do we achieve a billion dollars last year, we were cashflow positive for the year and we were profitable for Q4, right? So that puts a lot of wind in ourselves as we go forward with at the end of last year, a half a billion dollars in the bank and now a billion dollars in the bank for us to go figure out what we're going to grow and grow into. I think moving forward and being independent, I think we'll see, right? I think there's always a tick tock in our industry, right? Things are distributed, they're centralized, they're distributed. I want one throat to choke, I want best to breed. I think with all the distributed apps and all the analytics platforms that are going to start to become more important than what we're used to in the X86 space, I think best to breed is starting to rise up right now and so I think the runway for pure to stay independent is there. Don't give a wrong, we're going to have to do our works with bubbling into clouds, right? And all those ecosystems because customers want a transparent experience but we'll be sharing some news on that I think next week. Excited to hear that. The cartel will continue to suck up startups, no question about it. But we love companies like Pure, put Nutanix in that mix and it was sad to us to see all that run of the virtualization companies that just disappeared because if it's just the cartel building new products, you're not going to have the level of innovation that you get with VC funded startups in the Valley, you just not. Well in the U.S. you're seeing, I mean you're in the U.S. you're seeing VC investment starting to diversify a bit, right? Colorado's getting hot. The Boston area has been there for a while but it's getting hot. For IoT and security. And that's been the great thing about IT and U.S., right? Is we've been an innovative landscape. I think the barrier has probably forced some innovators out based on just the cost of living. So who knows what the mix will look like a decade from now but yeah, we're still going to be Silicon Valley centric for the near term. So I love talking to you because we can have these conversations. We were joking off camera, we could go for 90 minutes which we easily could. We got to go soon. Let's talk about Veeam, the relationship with Veeam. You guys are kind of birds of a feather in a lot of ways but take us through that. Yeah, so the opportunity to partner with Veeam was a no-brainer. There were synergies there, right? Pure and Veeam, both trying to disrupt legacy markets, doing it through simplicity, right? Riding the wave of virtualization as a primary business focus but not exclusive. Our net promoter scores with both companies are off the charts, right? The customers love it and we're multiples higher than any of our competitors. And so bringing the technology together were real simple. So last month we announced, four or five weeks ago we announced and released a new set of solutions and integrations that was comprised around three areas of benefit, right? Accelerating backups, increasing the speed at which you could recover data and adding a new level of agility within your ecosystem. And so delivering those three value props were based on us supporting their universal API adapter so now that they can offload some of the backup process to array-based snapshots and that preserves the performance, makes the window collapse faster, that's when production data sits on the flash array. We've also certified putting the flash blade behind the Veeam servers as a backup data repository. And the benefits of that from a backup window are faster data ingestion times across your real estate, obviously smaller footprint, lower cost within the data center. The bigger impact on both of these is on rapid data recovery. So with Veeam, through their Explorer integrations, you can pull files, disks, VMs, applications, right out of the array snapshots. If the array is still online, but someone's just munched the data, if the array's no longer there and you need to pull from the flash blade, flash blade gives them a capability that they never had with disk, which is they can start, because you know how Veeam recovers, right? They actually start the data services and recover them from the backup repository and then live migrate it back to the production environment. With the data repository being all flash, now they can bring up a significant, if not all of your data back online and then trickle restore it back to the production data sets. We had a customer with a large distributed database that was on a more traditional disk backup system that was really focused on ingest, right? Make the backup window, not so much focused on the restore times. It took them in excess of 36 hours to put back their database and this was the mission critical database of the organization. We've come in and replaced that, 36 hours is now 30 minutes. So is all flash as a repository for your backup for everyone? Maybe not for every organization, but we're seeing a big growth ramp on that in the enterprise. The last piece that we've brought to market together and integrations is integrating with their data labs. That's their environment to be able to on-demand create test and dev infrastructures for you and that pairs really well with all flash arrays and snapshots because it's instantaneous, consumes no new storage and our automatic QoS preserves the resources for the production environment from the lab and so those are our three areas. Accelerate backups, rapid restores and give you some agility with your test dev. Okay, and the agility in the ecosystem is oftentimes underappreciated, right? I'm amazed at the customers that I, large enterprise customers, right? Revenues in the tens of billions of dollars that you still meet with today where they have staffs that their job is to restore an Oracle database to an Oracle developer and that's all the guy or gal does 40 hours a week. It's amazing. Vaughn, great to see you again. Thanks so much for coming to theCUBE. See you next week at Pure Accelerate at San Francisco where they're Wednesday, I believe, we're broadcasting. So look for all the things that Vaughn teased. He showed a little leg on some stuff so we'll be covering that next week. We're back here tomorrow. Stu and I will be kicking off at 9.30 with Peter McKay so don't miss that. We're out for today, Veeamon 2018, theCUBE. See you tomorrow.