 Hi everybody, this is Dave Vellante. We're here at wikibon.org headquarters in Marlboro, Massachusetts, inside the cube where we bring in guests, all the smartest people and extract knowledge and share that with you. We're here today with Ash Uthosh of Activio. He's the CEO, welcome Ash. Thank you Dave. And founder, former CTO, you are an alpha geek. You know that how the technology works. You've been a VC. You've run a number of companies, deal maker, interesting combination, and your latest baby is a new company based in Waltham, Massachusetts, Activio. Tell us a little bit about what Activio does. Thanks Dave. So about three years ago when we first put Activio together, it started out with a very simple premise where we saw that organizations and people spent about five times or even more on making and managing copies of data than the production data itself. It was very interesting to see that you ended up with the tail becoming much bigger than the dog itself. And we looked at two big trends that helped us change that whole ratio. One was a virtualization or adoption of virtualization across the board. Second was the movement of data predominately from disk to desk, as opposed to disk to other kinds of media. Taking those two, we tried to put together an economic model that said, is it possible for us to bring virtualization into data management? And that's what Activio is all about. So this week we announced a new class of storage called Protection and Availability Storage, which is purpose built for optimizing management to copies of storage and sits alongside your production systems completely independent of what's happening on the production side itself. So in the last 10 years or so, we've just seen an explosion, not only of data, but in copies of data. And that's certainly part of this, but there's replication software, there's snapshot software, there's backup software, and these are all siloed point products largely. Now some large companies have purchased smaller companies and sort of creating this marketing umbrella, but from a customer perspective, I've got to manage these individual points. And so talk a little bit about why we have so many copies and you mentioned you've got this sort of virtualization approach to managing data. Talk a little bit more about how you're solving that problem, why we're there and how you're solving that problem. I think it's a slight twist to the old ad age. We talk about the only constant in businesses change. Turns out there's a small twist to that. I think the only constant is a faster velocity of change. And so people and businesses have now a much bigger requirement to go back and manage this change much more faster than they were used to do before. And that requires a whole new way of how you craft your IT infrastructure, the whole new way of looking at how you run your operations and virtualization absolutely becomes a foundation of making that happen. And the biggest barrier to that was the ability to come back and leverage information so that you can really run your business as fast as possible. And what happened in the process was you had your core set of data that was literally being copied for different kinds of reasons. Sometimes for data protection and therefore you had a separate tool for that. Sometimes because you want to run some business in analytics, a separate tool for that. Sometimes for disaster recovery, a separate tool for that. So all these tools were a result of the business requiring more and more information out of the data itself. And over time you got to the point where you have a plethora of these tools. You had a cackle of these things that you have to manage across the board. In fact, I think we talk about the classic case of a hazing program for an IT organization being a backup administrator. The new guy comes in, you go do backup. It's a thankless job. It's one of those jobs and you never get credit for doing a great job but you certainly get fired if you make one mistake. And so we decided to go back and address this underbelly of the storage industry, bring in virtualization and bring in a notion of the consumerization of this whole thing so that you truly love to go back and manage these copies of data and bring about a new paradigm of how you can really make your business much more agile at a completely different price point. So we have DAS, we have NAS, now we have PAS, Protection Availability Storage. So it's not just backup, right? Talk a little bit about the use cases. It's very simple. I think when you make copy, it turns out that people make something like a backup software. Makes a copy every night, keeps it around for a few months. Something like Snapshot makes a copy every couple of hours, keeps it around for a few, maybe days. You have this very simple notion of how all these tools that are making copies of data follow a same set of fundamental data management primitives. They all copy, they store, they move, they restore. The only problem is they do them in individual silos. So if you can bring up a virtualization notion where you bring all those silos together with a single platform, virtualize all the disks underneath, not only have you optimized the underlying infrastructure, you also optimize the entire operational process. So now you come back and say, you captured the data once, you reuse it many times, whether it is recovering a file from yesterday because you lost one, or because your test and dev organization requires 14 copies of the database from two weeks ago, or you want the disaster recovery from a compliance perspective, we test it so often. The notion of having a virtualized platform for keeping copies that can instantly recreate these copies, just like your storage system does when you're accessing a file. Think about pass as a storage system that accesses history. So you have a notion of time added to this entire set of data that's kept up there. And most importantly, that data is now managed optimally, both in terms of capacity and the network. So you get a huge financial impact from a CAPEX, a completely different OPEX profile. But more importantly, what you do with the data, which is what most organizations want to come back and do, is just incredible in terms of the agility that brings about as a result of this virtualization. So as you write about the hazing program for IT professionals, data protection and generally in backup specifically has always been one of the biggest pain points for IT professionals. And a lot of the organizations we work with in the Wikibon community have very limited backup choices. They'll maybe do a daily incremental in a weekly full. And then for the mission critical applications, they might have some kind of replication software. And that's it. It's sort of a one size fits all. Can your system help break that mold? Absolutely. So talk a little bit about that. Absolutely. I mean, it's very interesting. We have one user about almost nine months ago who had been delaying the purchase of one of those tools because he was waiting for the management to come back with the definition of what the RPO and RTO was going to look like. Because based on that RPO and RTO, he was going to select a tool. And what, a complete refreshing change when we came back and said, here's a storage system that allows you to basically virtualize how you want to capture and how you want to restore. And you really don't worry about your RPOs and RTOs anymore. It reminds me of the pre-virtualization of the server days when you said, well, I got to run this application. I had to figure out which vendor I need to run this server on. You don't think about it anymore. It doesn't take you three and a half weeks to bring up an application anymore. It takes you literally 20 seconds. Click of a VMware virtualized environment or a virtualized server environment. And that's the same thing you're talking about here. You're not having the business. Go back and set a policy that drives down a choice of the product anymore. It's, you're converting IT into a service where it's a platform that allows the business to go back and pick the appropriate knobs and completely optimize how infrastructure is managed underneath. Now to be clear, you're not pushing or selling the production capacity, right? You're selling the backup, the business continuity, the disaster recovery, all those copies, right? So the production capacity is going to be what? It is any vendor. I mean, I think we're trying to very clearly isolate between two kinds of data in the organization. There's production data where your business runs. That's your golden copy of data. And then there's copies of this production data. And we are trying to optimize all these copies through a separate storage system. This production storage system could be all the favorite vendors you can pick your storage from. All of the favorite vendors that you pick on a daily basis. The important part is today you have this choice of production storage vendor. Absolutely a couple, your choice of your copies. And we are trying to decouple how you manage your data from where you stored it and have the SLA or your business requirements drive the class of storage and the kind of storage you want to use by having a data management engine that's virtualized. So your target can pretty much be any storage, any of the popular storage. Our target could be any of the storage vendors, any of the storage that's available, in fact, even a cloud storage, if you want to go back and use. The notion is, the way you manage your data should be dictated by the application and the SLA. Your choice of storage should be completely yours, based on this SLA. And then it could be the favorite vendor you have, it could be some other new vendor, but you have a completely different way of thinking about how you go back and choose this kind of storage. So one of the things you hear about, especially a backup land, you talked about disk to disk, deduplication was a huge trend, still is, but really took the world by storm and we all know the story of data domain and others. You guys provide deduplication capabilities, correct? That's right. And so there's a storage efficiency play, that you've got to do that, if you're really replacing copies. Talk a little bit about what customers are doing with the software. Are they able to actually replace a lot of this replication software and backup software? Are they sort of running that in parallel? Yeah. So there are two dimensions to optimization. One is within a bucket, make sure you optimize what's in the bucket itself. Then there's optimization of the fact that this unique set of stuff is across several of these buckets, several of these pipelines. And so we're taking a combination of, yep, we've got to go back and make sure we optimize within a bucket, but more importantly, collapse all these silos. So Actifio's virtual data pipeline is more about taking those individual silos, flipping them over and creating a pipeline of how data lives through its lifecycle. At the same time, you're having all the traditional optimization way of deduplication, compression, and what we call dedupe async, a whole new class of replication that is just dramatically an improvement over how data is replicated today. So you have a lot of the traditional tools that are put into a pipeline, and you have people who come back and start with a simple problem. It might be something as simple as, you know, I have about 12 terabytes of data and it's taking me well more than 24 hours to back it up. And somebody suggests, maybe you switched to a backup to this device. The reality is the real dog of the problem is the backup software. The notion of copying data so that you can put it into a tape was the reason why you had backup software. And now you remove the tape with something that acts like a tape and maybe it's a little faster, but when you want to restore it, it's going back to the backup software. So you have a little dumb device sitting there optimizing the capacity, but the intelligence lives in the backup itself. We are trying to come back and say, well, if you're moving from disk to disk, there's a much better opportunity to move the data in a raw format, put it back in the raw format and use a storage device to make it happen. And therefore you can start with something as simple as, I want to solve the backup in no problem, I want to do DR, or maybe, you know, I just wanted to do business continuity, or in some of the larger organizations, we've seen people come back and said, well, I got two petabytes of SAP data, and turns out I only have 111 petabytes of production database and 18 copies, two of them for backup and restore just in case, and 16 of them because my developers want a copy. In that, that's a scenario where we free up one and a half petabytes of storage because you have 111 terabytes of this production data and about 150, 160 terabytes of the copies, which are then virtualized to create these 18 copies any time you want. It's a completely different way of looking at how you go back and use storage, but more importantly, how you manage the data when and where you need it. There's not just virtualization and capacity, it's also virtualizing where you want the data to be accessed, what format you want it to be accessed in, so that the whole notion of the intelligence and software and box being dumb, you combine all of that into a box that is much more smarter. It's an object-oriented file system that allows you to get the context of this data for the first time. So that brings me to my next question, which is you're absolutely right. There's this heterogeneous siloed mess out there of different point products and customers want Nirvana, which is essentially the scenario that you're putting forth. How do they get from point A to point B and what do they have to do to adopt the product? I mean, you mentioned object-oriented file systems. Sometimes that scares people away to talk a little bit about how a customer goes from where they are today to what you're proposing. Yeah, and the beauty of Actifio from a philosophy perspective is to drive home the notion of simplicity. In fact, our tagline is radically simple. I mean, it's the notion that we want to make IT as simple as it can get and bring a notion of consumerization. So certainly we don't want our customers to start thinking about what's inside this box. We want them to think about the fact that it's a beautiful controller that shows up, connects to the power, connects to the network, and you set a discovery process of all the applications you have up there and assign SLAs. And that's all we want them to know. We want them to know two concepts. What apps they want to protect and what's the SLA they want assigned to this application. Now the question is where do you start? And you have multiple entry points. You have the ability to come back and say, you know, I've been looking for a DR solution and my vendor tells me I've got to buy another box of the exact same kind, some replication software, some brand optimization, or I just bring in an Actifio controller on both sides and it does take care of all this stuff. And by the way, it also happens to do backup. Also happens to do business continuity. Also happens to do, this is the power of virtualization. The things you can do when you take a function out of a sheet metal and make it into a file, you start making it do a lot more things. Or there are some situations, like I said, somebody wants to use a budget for this quarter because they have a terrible test end of problem. It's taking them an enormous amount of time to make copies of the data for their developers. You bring in an Actifio controller, exactly the same thing. It captures it and then reuses it for whichever specific application you want to apply it to. It turns out you can reuse it for other things too. But those are all part of the entire platform itself. So you mentioned consumerization of IT, bringing that trend to your ethos, really. Something we've talked about here at Wikibon, Silicon Angle and on theCUBE numerous times, what do you think is doing a good job in the industry? Do you think anybody is doing a good job of bringing that Apple iPhone-like consumerization mentality to enterprise products? Or do you feel like you're one of the first? I think from a data management perspective, we clearly are the first. I mean, I was an investor before I started this project and I looked at the landscape for a long time to look at where the pain points are and where the IT industry is. And I think in many cases, the incumbents absolutely have a desire to keep the complexity because that's part of the entire process and the business model itself. But more importantly, even if they had wonderful notions, it is very hard to come back and take a clean approach to say, I got 16 products. In fact, sometimes they may be from 16 companies and I'm gonna have to pull them together to do something. I mean, I worked in large companies where it's hard enough to get within an organization to get something done, let alone across an organization. So I think there are a few companies you will see especially early stage ones. Some of the notions that companies like Ecologic started off pretty early on. Some of the stuff that XIV has done in terms of user interface perspective. And they were the first baby steps. I think we took a big giant leap in going back and addressing one of the biggest pain points, the largest footprint in terms of data, and one that's considered an absolute unrebelly that nobody wants to mention. So I think we're setting a tone in terms of bringing about this consumerization and virtualization, at least as far as the storage part of the IT industry is concerned. Let's talk about some of the basics of Actifio, the company, so you were founded in 2008, is that correct? 2009. Okay, officially in 2009 you raised, I believe, 24 million? That's 24 million. And what's your head count today? I've got about 74 people now. And you're not going to tell me revenues, I presume, I remember we're on theCUBE if you want to, that's fine. But maybe you could talk about customers. I mean, we'll be talking a handful, dozens. We're just short of 100 customers. Typical range has been users across all kinds of verticals, starting about 10 to 20 terabytes of data. So you're talking about mid-size, lower end of the mid-range, to the enterprise class of customers. And the pain in terms of managing 10, 15 terabytes of data starts to become very palpable when you find that it takes two and a half weeks to recover your data or I'm a graphics artist in a marketing company and it's going to take me longer to get my old file that I accidentally deleted than for me to just recreate that entire file. That's when you know I've got to go back and change my IT. When your IT starts to become a bottleneck for business, that's when you bring virtualization, that's when you bring Actifio. So I wanted to shift gears a little bit and pick your brain. You've been in the business for a while now and you're a very experienced entrepreneur and executive. There's an interesting dynamic going on. And now you're very much tied to the virtualization trend and the disk to disk trend as the other piece, but VMware specifically is owned by EMC, which is a very strange situation. I mean, it's almost somewhat unnatural. And they're a storage company. So how do you, as a small company, feel about sort of a large company like EMC and VMware and maybe some of these other large companies like a NetApp or an IBM or an HP, sort of in that little collaboration, we've called it a cartel even, in quotes. How does a small company get access to SDKs and all the API knowledge that it needs to compete? How do you do that? I think we are living in a very, very wonderful world. I mean, you got the Arab Spring going on and there's democracy everywhere. The good part is that as an IT and as a business owner, my ability to come back and drive my business has very little to do with who you are and what your size is. Is it about relevance to me? I mean, that's what every organization has come back with. And you've seen that. I mean, VMware was a two-person company at some point, but they weren't relevant. And so is Actifio. So I doubt it's a two-person company and it's the question of am I adding relevance to your business? Am I making sense to your business? There are some attributes to being big, but absolutely I think you start seeing that as you get to the consumerization part, the notion of just sheer size is not gonna help out for too long. It's about innovation. It's about making sure that you're really able to make relevant sense for the business. And we've seen companies that were large. NCR was large, one of the largest. So was AT&T, because all there's left is a brand. So I think just the large size itself doesn't matter. If you're not relevant to the business, then it doesn't matter who you are, what size you are. But if you are relevant, we expect absolute market success. Can large companies innovate? Some can in pockets. Majority, it's very hard. I think majority have a foundation of a business model and a product set that they've built on which made them successful, made them big. And I think I'm a big fan of Clayton Christensen in terms of innovators dilemma. And it's a problem most companies have a tough time getting out of. You're successful, you write on a model that you think is what's built the company. And it becomes harder and harder as you grow so large that you're unable to move. A lot of that answer depends on how you define innovation, I guess. And I think our definition would, at least encapsulates massive disruption. I mean, I would certainly put, for instance, a data domain in that category. That feels like disruption to me. And I think it's hard to actually come up with the examples of large company innovations as under that definition. Absolutely, I think you said it right. I mean, innovation comes in various forms. I think one of the best examples, we just celebrated a hundred years of IBM. And that's a company that has completely changed the business model. And they do tremendous amount of innovation but not in the sense that we think of as a startup company. You see that across companies like Dell who go back and change their business models constantly. So I think for a larger company, the notion of innovation is probably relatively different from what a smaller company thinks about in terms of product. But in general, when process becomes a foundation of you're building a company and the spark of innovation that comes from people becomes much smaller part of the entire picture. Inevitably you just have to keep innovating the process. And nothing to do with the technology, nothing to do with the product itself. All right, Ash, well thanks for coming on theCUBE. Actifio, clearly disruptive, doing a lot of innovation. A company to watch. Love to see the East Coast startup innovation going on and appreciate you taking time out to join us. Thank you very much, Jay. Thank you. All right, thanks for watching everybody. And check out more videos on siliconangle.tv and check out wikibond.org, siliconangle.com. If you got questions, hopefully we've got answers. Thanks for watching.