 Inside theCUBE in a while to the home of Cloudera where we sit in siliconangle.com I'm John Furrier, the founder of Silicon Angle. I'm here with Chris Gladwin, the president and CEO of Cleversafe. Chris, welcome back. It's an easy name to remember, but it's a compound word that people have a surprising amount of trouble with. We're here at theCUBE. We want to talk about your business and some of the climate opportunity out there for folks and your customers and challenges. So first start off with what's going on with you guys in Cleversafe and what's the update on the company? Well, John, thanks for having me here. I appreciate it. It's a pleasure to talk about these topics. So the some of the latest developments with Cleversafe. At a high level what we're seeing is there's so much interest in cloud computing that it's really pulling through a lot of interest and awareness in cloud storage because obviously if you're going to compute you need to store. And that's just been really fortunate for us. Whereas two, three, four years ago it was just difficult to get awareness with these kind of issues in a broad sense. Now it's kitchen table conversation in a lot of places, particularly within the storage industry. It's a topic. So that's helped us quite a bit. We've recently had some pretty key customer announcements, one with the US intelligence community. And then we also just completed a big round of funding and internally we're getting a lot of traction and momentum and excitement. What was the round of that size of that round? $31.4 million. Nice. Congratulations. Thank you. Fresh money. Yeah, that's important. For technology expansion, sales, marketing, any particular focus all of the above? It's all the above. I mean the stage of the company now is we're about five and a half years old. And we began delivering product kind of on a limited basis a year or two ago. And really the middle of last year is when we began deploying production larger scale systems. So from money point of view it's, you know, we're still kind of expanding the capabilities but now we're also from sales and marketing and all the other things that we operate a business for. We're here with Chris Gladwin from the CEO of Cleversafe. You guys are based in Chicago. We are based in Chicago. You don't have to be in Silicon Valley to be successful. You guys have been doing very well lately around storage, a different kind of storage and the emphasis on security. Can you talk a little bit about that aspect of your business in particular? Well what we saw early on and there were a lot of surveys that we had access to about what's important to customers. And in a cloud environment you're essentially giving physically someone else your information. So obviously security is a big concern. And that fortunately for us is something that we do well. So in addition to the capability of cloud which is flexibility limitless scale, things like that, you also have to do that in a way that's secure. And there's really three dimensions to that. The classic textbook definition of information security includes confidentiality. Means that the bad guys can't see it. It means integrity which means that the data is perfect to the people that should see it. It's no good if they can see bits and the bits aren't right. And then finally availability meaning it's always available to people that should see it. It's kind of an easy acronym to remember because it's CIA. And you have some clients that are government related. We recently announced a strategic partnership with Incutel. And Incutel is the business development arm of the US intelligence community. They work a lot with CIA. So we have to differentiate it on the slide. But really when we talk about information security we mean confidentiality, integrity and availability. And that's essential. It goes without saying but if you have a system data is the lifeblood of the business. It's the assets of an individual. Those things have to just be there at a very high level. It can't be compromised. It's got to be right. And it's always got to be there. You guys have been around for five years. Obviously a lot of the clients are like the government and the CIA with the funding from Incutel. But as an entrepreneur life has some surprises. The world kind of spins your way once in a while. For you guys it seems to be the case where the world has spun your way with the focus on obviously the adoption of cloud and privacy and security. That being said what are you seeing out in the marketplace for the key trends and how security is one we'll dive into. But outside of security what are you seeing out there as macro trends in the marketplace? One of the macro trends is I would say the cloud thing. And there are reasons why they're driving it. You see this again and again and this is part of the cloud folklore practically about electrical power generation. I think their own that it became centralized and more efficient. It used to be that most companies had their own physical data centers. Now everyone uses common hosting facilities. When I was starting my career I was a network person. I did some storage work too. I had a large aerospace company where I was responsible for the networking and the storage products. And back then when TCPIP first came out we had TCPIP but it was our network. I mean it wasn't an internet. We had our own lease lines. So it kind of started out as a private thing but eventually it became much more efficient to have that aggregated as a public service which is just kind of aggregating a bunch of private needs into one thing. So the same thing is happening with kind of some of the basics of computing. Computing itself storage things like that are going through that transition now. And is that really kind of the whole focus of private cloud where IT has had their own kind of private environment in the lease lines as you mentioned in the old days. You're dealing with public infrastructure, HAA, public cloud. Just like you saw in the 80s with networking where those technologies came out they first were used for private networks within enterprises and then became public. We see the same thing happening even though there's a lot of visibility to public cloud storage most of the business today is private cloud. So that's mainly what we've been selling. It's private clouds that we sell to government agency bank, a big social networking site. That's mainly what we're seeing. Now definitely longer term public is going to be probably the larger segment of the market but today most of the action is private clouds. From your chair here or your office looking at the industry where are we with the whole public private cloud there's a collision course between the benefits of public what inning are we in so to speak and what are the key challenges for this kind of collision. I see everyone's talking about test ops and test and dev ops are kind of going to the cloud and mission critical but emphasis of production where the costs are heavier and scale issues are there. You get companies that have scale operations and they have to kind of move there. What are you seeing in the market for that transition? Well I don't ultimately see it as a collision. Like a lot of things it'll become both you'll take the requirements of public and private clouds and when those can both be met then it'll kind of merge to one. If you look at networking today and people still VPN all the time over the internet to make essentially a virtual private network on top of the public infrastructure you're definitely going to see the same kind of thing happen with cloud storage. We already have the capabilities to deliver public, use of public infrastructure but with the privacy, auto-vability, manageability control that defines a private cloud so you know a private internal requirement so it's already possible today to kind of merge those two things. Obviously very similar paradigm of the 80s with networking, seeing that now. So you have this converged networking landscape, storage guides, the networking guys and the compute guides, the servers kind of collisioning together. There's not a lot of talk going on around the compute side. It seems to be an abundance of compute. Storage seems to be the key linchpin in all the conversations whether it's a Hadoop, which we're in the cloud era's office here, to kind of industrial strength storage. And I wrote a post this morning called The Storage Cloud How is storage changing in that equation, in the conversion networking equation? Now I'll say it's more relevant. I mean data's not going away it's only getting bigger and bigger but latency is an issue, speed, application support multiple applications. How would you kind of summarize that whole dynamic? Is it a tsunami of change? Is it going to be something that's going to be slowly rolled in? I think cloud is a part of the first real change in the dominant architecture paradigm for computing since the 80s. When the personal computer paradigm really became the dominant paradigm. From a storage point of view where you basically had a hard drive and that primarily defined your data. As a local device it was associated with you as a person and it was typically in your PC. The internet delivered new kinds of applications but you still had your data as your own. Clearly we're going through a big transition where we're moving to a new paradigm and that's a big, you know, cloud is a part of that where you say now I want my data to exist independent of a particular device that I might have because in fact I may have ten devices. I may have a computer in my office, one at home, one in my car, different sizes for different functions so that answer has to become independent of all those devices I have my storage and that's my information and that transcends those devices as they come and go during my life which those devices always will the data persists. So that's a big driver to cloud storage that we see and that's a big shift. So a part of that is the cloud which is kind of the back end bulk store and then clearly what you're going to see local storage effectively just becomes a cache off that and you know for certain devices they'll cache a lot more stuff, you know other devices they'll cache less because they're smaller, more mobile but you know the net result application specific kind of scenarios right? Yeah but the net result is it'll just feel like your data is always there no matter what. And clearly that's something people want. They don't want this idea of well if I use this device I can see this and if I use that device I can see that and I've got to try to sit there and figure out what goes where. That's way too complicated. So that's a lot of what's going to drive that paradigm and that's a big part of cloud and then you look at you know the technologies of cloud, you know flash mobile devices, multi-device kind of lifestyles. I mean that's all converging to this new architecture. We're here with Chris Gladwell, the presidency of Cleversafe from MIT he's a geek, developers got hundreds of patents here on the solution. Question for you from your definition, big data. That's the big buzzword these days. I'll see data, big data, fast data, small data. Data in general, we'll get the Stratoconference next week which we'll be broadcasting live so that they angle the queue. What's your definition of big data? Big data, you know the, well let me first say that all the growth of data is big data. This is prior to 1999 most data, and when I say most data I mean cut open a random internet cable or break open a random hard drive and count the bits. Prior to 1999 most data was structured data like data in a database and the data type itself was numbers or text. 1999 was when it switched over and that was driven by digital music distributed over the internet. Which you have some history with. I'll get to that in a second. And that's when it, so image and audio data then cross 50%, so these large unstructured objects. Since then with the emergence of digital video, it's now in the 65, 70% range of all data is this big data primarily digital content. And then going forward, you know the longer you go out in time, the more it approaches 100% of all data. So you're still going to have structured data in a database, but it's just going to be a sideshow from a bit point of view compared to big data. So that's how we would characterize just large unstructured data objects and it's already dominating storage. I mean the internet is pretty much already turned into a video distribution machine and it's going to become just more so a big video distribution machine. Digital data is obviously big right now we see iTunes and you have a little history in it. Talk about your background and the folks out there who don't know the history you have with music. Which is the most popular digital asset out there. It is with video. It was really happening. So like you said, I went to MIT. I was an engineer and then I worked for a large aerospace company as a buyer of technology. Networking products, storage products. I did that for a while and then I switched to the vendor side where I made computer products and I did different jobs you know engineering, product strategy, marketing and then from there I started starting companies. And the first company I started was Cruise Technologies which at the time I thought you know the way to have a great company is to do something that's so hard no one else can do it but the reality is you can be ahead of your market and that was way ahead of the market. You know that was the 90's, mid 90's and we were making wireless thin clients like an iPad. And needless to say we were a little ahead of the market. A little bit a couple of years. So you know we ended up selling that IP and then I started a digital music company called Music Now which was the leading business, the business supplier of digital music services. So Best Buy, Clear Channel Charter, Earthlink. They had digital music services, download stores, internet radio, music subscription services. But it was just their brand and it was all us. We aggregated all the rights operated all the servers, figured out this penny goes to this guy and that penny goes to that guy. Micro payments dealing with DRM kind of. All the licenses and all that stuff. So you know that's obvious. It's complex. It was not simple. Yeah, it was, you know, there was a lot of interesting complexity. That's certainly one aspect of this technology. Back then that was big storage to store all music. Not so much anymore, but back then it was kind of a big storage system. Plus that rights distribution was extraordinarily complicated. More so legally and just the reality of getting all those licenses. That was remarkably complicated. Building a system to administer and pay them and audit and all that stuff. That was fairly complicated too. Let's talk about Cleversafe and your current project. Obviously you got a lot of funding. $34 million. No one's just writing checks. The guys in the street holding their hand out. It's going to be a significant amount of capital. You guys have a five year track record. Where did you get the idea that rate is dead? I mean, not the canned answer, you know, data is growing. You must have had some kernel of, you know, we're going to do things a little differently. The world's maybe different in vision. Talk about where the inspiration was and the idea behind Cleversafe's technology. Well, in my prior company music now we built the system to store all music. And, you know, we kind of experienced the state of the art as a customer. And honestly, I thought it was pretty bad. You know, I mean, there were times when, you know, whole arrays would fail and they would never come back. I mean, it just was beyond me. Like, how can this be? How can this be state of the art? There's got to be a better way. A lot of my background in addition to storage products is in wireless products. And, you know, when you talk to a storage guy, you know, they talk about a bit error rate of every ten to the fourteen, every ten to fifteen bits on a drive. In the meantime, the failure, you know, of, you know, a thousand, a couple thousand hours or something like that. They have no idea what failure and bit error rates are compared to wireless. Like, when you're driving 60 miles an hour with a mobile phone and you're talking away, the amount of retransmission, drama that's going on, packets are getting lost. You're driving by a building with a steel wall, a connection drive on and on and on. And there are techniques, or like, you know, when a spacecraft goes to Mars and it's talking back and forth. I mean, these techniques, you know, these advanced coding techniques, erasure coding, Reed Solomon, Ford air correction, whatever you want to call them, they've been around for thirty years to wireless guys. And so as a wireless guy, you look at storage and you say, they're not doing this? It's like a radio. Talk to each other, right? Yeah, I mean, you look at, you know, state of the art in storage with rate is parity. And, you know, maybe you have one or two-dimensional parity bits. And it's like, okay, that's great. That was like the sixties in wireless. And, you know, there's other ways to do this. Now, the reason why this hasn't happened before has to do with Moore's law. Back in, let's say, 1990, 1991, a CPU was a lot slower than it is now. You know, microprocessors, you know, were maybe a million times slower. So to do some extra math would slow things down. But things have changed. In 1991, you know, the average hard drive, for example, took about sixty seconds to read or write all the data on the drive. Drives have gotten a lot bigger, but they haven't gotten much faster. So now it takes somewhere around sixteen hours to read or write the data on a two terabyte drive. So now this trade-off, and meanwhile CPU's got a million times faster. And abundance, tons of abundance of compute power. Yeah. So now you can employ some advanced, you know, math is your friend. You know, and Moore's law basically is the cost of math. And the cost of math is a million times cheaper. So you can do some pretty interesting stuff without really noticing performance degradation or without really noticing cost. And so that kind of transition, that crossover from, you know, driven by Moore's law, enables these kind of advanced coding techniques for storage, which, to a wireless person, is like hard to believe the industry isn't already doing this. And so you had the idea. So you start sketching out some ideas, and you form a company around. Is that kind of how it went? And then boom, you've got a product, some patents, and... Yeah, you know. Well, having started a couple companies, one thing I wanted to do, which we did here, was I wanted the first thing, or as soon as possible, to sell it. Even if it's not built. Like go try and sell it and see what the reaction is. So the very first thing I did is I kind of wrote just a little very basic, long sense-replace prototype just to prove, okay, this is going to work. Alright, once that was done, the first thing I did is I worked with one of the other founders who also went to MIT as an interaction designer. And so she created what looked like a completely functioning system. Yeah. Was it a vase kind of thing? Behind the curtain? Yeah. So essentially what she did is she went to the Yahoo Small Business site and screen scraped every screen. And then pasted on there, like as a graphic artist, not as a developer, what looked like other menu options. And so we had both done a lot of focus groups. And so we knew how to pretend to be focus group moderators. So we ran a bunch of focus groups, like normal pay people 50 bucks, and they sit there and ask these questions. And we ran focus groups on what looked like a completely done system. And, you know, the things that we found were you know, I've done a lot of focus groups and you look for certain things. You look for unaided and aided awareness of the problems, people willingness to pay, satisfaction with alternatives. And data, and information, it's just off the charts. I mean, the value of storage, which really is your information as a business, as a person, is immensely valuable. And it's growing in its value. I mean, it doesn't depend that we are on digital information. So that's I mean, if you want to sell something to someone, they better value it. Okay, so that's highly valued. There was a great deal of aided or unaided awareness of problems. Like, oh, when my hard drive dies, my data is gone. And there was a very high level of dissatisfaction with the currently available alternatives. So you know, that's the combination. So you sold it, and it's like, let's go build it, and you did, right? Yeah, we ran focus groups. And, you know, you score, you know, these different, and we did, you know, enterprise, small business, consumer, and you look at these results and you're like, man, this is it. So then we said, alright, now we gotta go build it. And building a disperse storage system, it's been, these advanced math techniques have been known since the 50s, and they've been used in other industries like wireless. And there has been some use of them in storage, not so much in the commercial world, but in incredibly secure environments like the storage of cryptographic keys or weapon launch codes or things like that, they've used these kinds of techniques before. And there's been a lot of academic work in these areas over 20 years. So it's been known to some degree but to commercialize it. I mean, just to give you a sense, so we had algorithms working in like 2005. The algorithms were working, and that's interesting. But, you know, a customer doesn't buy algorithms, a customer buys a system that can be installed and upgraded, tuned and managed and configured in value. And secure and reliable and fast. And so building that whole ecosystem around, you know, that, well that's, you know, taken us another five years. And, you know, we knew it was a long project, other companies tried, and they really couldn't get it to go through deep tech. That's not what the company, your company CleverSafe, so you got some funding. How many employees do you have? I think right now we have around 48. So you guys are growing? Yeah. So where do you guys stack up in the marketplace? So your business compared to others in terms of who you're competing with and then just overall approach to the business. I'll say there's different approaches these days. You talk to anyone in storage and, you know, there's a big opportunity with a lot of other players being taken out with acquisition. Everyone knows about 3-part investors and others. So there's a huge, big vacuum. Yeah, there's not a lot of mid-sized storage companies left. There's not even a lot of, like, bigger small storage companies left. What's the core strategy for you guys right now? Got the funding? Are you going to go out and sell on the differentiated features of it? Talk about that. Well, there's kind of two phases to how we're building this business. The phase that we're in now is a phase where just having a product that's operational and ships and finishes QA and all that stuff is part of the answer. What a customer buys is a result. You know, they have a business result like they want the system to operate and deliver certain benefits. So... So what happens is, you know, the thing you want to look for is a customer that buys an initial system and then comes back and buys a second order that's maybe 10 times bigger. That's the best validation you have. That means it really works and they want more. So we're really focused on that and then having that customer be thrilled. That they're referenceable, that they score you high in terms of customer satisfaction, and we really want to focus on getting that done, not necessarily with 1,000 customers, but currently with the largest most discriminating kind of customers, lighthouse accounts in our key market, and that's really what we're focused on for this year. So you're not seeing a lot of broad marketing where we want to have 1,000. We've got a lot of very focused execution. Next year is when we will say, alright, let's scale it up because we've done this enough that we can kind of bottle it and repeat higher sales forces and go crazy. But we can't underestimate that step and even though we know it works, if we stumble, so for example let's say we have the world's best algorithms and the install program doesn't work. And the customer has a problem and it could be some trivial thing in an install program. It won't matter that it was a trivial thing in the install program. The reputation will be honest after the work. So you just have to get everything exactly right and that's what we're focused on. How are you guys competing against the EMC's in the world, not the big players? The IBM, HPE, EMC, the big whales out there, and we're going to delve into that now. What's the story there? Well the closest competitor to us is the EMC other product called Atmos, and they're using similar techniques. They got a little bit later start than us, so they're not very far along in terms of the maturity of the product. But it's, you know, they're a smart company and they're on their way and they'll certainly work through these issues. And that's good for us. I mean actually we want them to succeed because it validates a comparable in the market. You've got to compare the benchmark against it. Yeah, I mean our issue is not, I mean we go head to head with them every once in a while, but mainly what we compete against in terms of an actual live customer situation is a customer that's thinking about building it themselves. That's the number one thing we compete with in terms of a customer. Sometimes they're looking at like Hadoop and some other stuff and learn it together. That's normally what we compete against. And that still shows that our issue is kind of getting the market going. Not so much, you know, A versus C versus the D competition. When we get to that point we've succeeded in kind of moving the needle, but we're still in the early stages. Chris Gladlin, CEO of Cleversave. Chris, thanks for coming on the YouTube. Question about the tech and the product, the product angle. What's the key differentiation for you guys in the product? I mean you mentioned security and we'll drill a little bit down into that, but what are the key differentiates for you right now? Well that's the one we wrap it around is secure cloud storage. So that's what we sell as secure clouds. Secure storage clouds. And as I said before confidentiality, integrity, and availability. Those are the things that we lead with. Now there's other advantages. Cost efficiency, which is kind of the same thing as power efficiency, space efficiency. You just need less servers, less electricity, less capital. And for some customers it's cost for some others it's power and for others it's space. And so those are some of the key things that we really focus on. And there's some other benefits we could talk about like scalability, you know, the system can scale to forever if you're using a object interface for example, which is unique. I mean you could literally have a namespace with 10 exabytes in it. There's a lot of simultaneous riders in that same container. No one else can do that. But you've got to kind of pick your thing to focus on. And we really focus on secure cloud storage. What do you think about the efficiency message? I mean obviously every vendor has a storage efficiency because you know storage is getting bigger and bigger. I mean everyone has to have an efficiency story. What's your take on the storage efficiency messaging in general from other folks in the marketplace? And then what's your strategy and message around efficiency? What's where dispersal has real advantages? The traditional approach people have used to create reliability is to create copies as well as to use parity bits which is RAID. So in an enterprise environment it's not at all unusual for someone to have three copies. And it's not at all unusual for them to have many more than three. But if you just say you know a person's got three copies maybe two in one location and one remote or something like that and they're using parity which means it's in a RAID array which it pretty much already is always is. So you add up all the bits you end up with and there might be some other inefficiencies introduced by you know file system or kind of how they lay things out. But at a minimum for every bit of original information you end up with five bits to store. And now you got five bits to air condition and five bits to you know buy equipment for and five bits worth of floor space and so on and so on and there's just no way around it. That's what drives your cost. Dispersal allows you to get you know way higher reliability. So a typical talk about dispersal. We didn't we didn't get into that. That's your core different approach. Dispersal approach. Dispersing the data and then having multiple copies. Just take us through that real quick. Yeah so as I said normally to create reliability you maybe have three copies and so the value of that would be I could tolerate two simultaneous failures because I got three copies and if I put those in three locations then I can handle two location failures. If I put those in one location I can handle two server failures. So you know you can figure that based on your business requirements. But that's a very expensive way to get reliability and it has security problems because what you want to do from an availability point of view is make multiple copies and put them in multiple places. That's exactly what you don't want to do from a security point of view because you're basically putting your attack surfaces all over the place. And there's other things that doesn't address like data integrity. There's other ways to create reliability besides making copies and that's what dispersal does. So a typical example would be let's say I have a file or an object to store or a block of data. Instead of just writing those copies meaning I'm writing ones and zeros that are the data. It might be encrypted. It might be compressed. It doesn't matter. I'm just writing it's real data and then I might put a parity bit at the end if I want to rate a rate. There's a different way. There's other ways to represent information. And involves a little bit more math but math is cheap because the Morse law math is kind of approaching free. So instead of let's say so the way a typical dispersed environment would work is I'll take 10 numbers at a time of that data and what I'm going to do is I'm going to take those 10 numbers. I'm going to do a little linear algebra and I'm going to make 16 new numbers that have the property that if I have any 10 of them I can turn it back into 10 original numbers technically how that happens if you remember an algebra. If you have three equations and three unknowns you can solve. What we're really doing with those 10 numbers is we're making 16 equations with 10 variables and that's what we're storing is the sums of those equations. If I get any 10 of them I get 10 unknowns. And if I put each of those equations sums on 16 different servers and I could put them in 1, 2, 4, 6, 12, 16 locations I could create incredible reliability. So for example I could create the simultaneous value of 6 things because I have 16 I only need 10. So I get the reliability kind of like 7 copies but in terms of how much I'm physically storing I'm not storing 7 times what I started with I'm storing 1.6 times what I start with. And that doesn't include the parity, the rate that I would use in a copy based system. So you end up a pretty typical environment you end up with a million times the reliability with a third or a fourth times the bits. And that's a good trade off. You mentioned securities and you mentioned earlier about the storage guys not knowing the bit errors and things of that nature. We've seen over the past 15, 20 years as a client server has been enrolled into an internet environment that converges between network guys and developers, app developers. Completely different breeds, network guys and things provision. And security and storage we see the same kind of dynamic. As you just mentioned in your example you have multiple points of stuff being stored. You have more attack points. So a storage guy's optimized storage on paper but ultimately is exposed security wise. Talk about the dynamic you're seeing the collision between security and storage from a people standpoint. Because there's security staffs out there and there's network staffs and storage staffs we have to live through the old today's world of network and app kind of blending together. What's that status of the security storage guys? Are they coming together or is this still siloed out? It's been certainly upon the organization. What we found is that we're bringing advanced security capabilities to the market. But at the same time you don't want to present it as something completely new and different. Any organization that cares about security can have certain standards. Like they have a certain encryption level or encryption type that they're going to use or something like that. It's a very difficult conversation to go in there and explain to them why that's a bad idea and how you should do something better. It's a much easier conversation to say that's a great idea. We're going to still do that and we're going to add to it. And that's how we approach it. So even though at some level what we find is the person that drives the purchase of a private storage cloud or even a public storage cloud is the security guys I'm sorry the storage guys are the ones that are really doing it but they have to check in with the security guys. There's compliance and all kinds of other hurdles in the deal. So the way you have to address that is you can deliver a lot of benefits to the storage organization and they're really ultimately they typically drive the purchase and operate the system. But you have to do it in such a way that when they take that to their security representatives they can say it's what you've already established as a standard and more. And that's how we've designed the system and that's how we communicate those benefits. We're here with Chris Gladwin the CEO of Cleversafe Chris tell us talk about let's get into the sharing side and wrap this up. What advice would you give CIOs if they came to you and say hey Chris we're really looking at cloud seriously we've done some toe in the water kind of things but I really got to expand my storage footprint we got to have a strategy I need to get my arms around it I need a framework what should I be thinking about what should my what approaches should I be looking at. Well the things that you want to look for are you want to have the good things about the system that you have now and more you don't want to have a situation where you have to give up so for you know going back to security for example you don't want to have to suddenly relax your security standards but there's other issues so for example auditability control tuneability manageability those kinds of things that they're used to in their internal systems they need to demand those same things because you shouldn't have to step back you should get the benefits that they have from their own private resources and get more the cost effectiveness scalability whatever the kind of cloud elements that are important to them they should be able to get but they add in more so that's what they should ask for. What kind of things would you say to the CIO if I asked you if I was a CIO and I said Chris what kind of benefits should I be seeing from all this new tech clever save technology and or other tech around us and we're converging consumerization of IT however you want to call it but I still need to comply and I still need security what things should that be enabling what environment what things in my environment should be enabling and where those benefits should I be seeing well in our for the customers that we have now we always link it to something that's essential to business as we grow in the future you know we'll have a lot of customers but for now we focus on customers where something that clever save can uniquely provide is essential to their business and that's that's a pretty intimate relationship with the customer so it depends upon the customer so you know in the government environment you know security is incredibly important but resiliency you know if you look at the internet itself the internet itself was designed initially with government requirements in mind to be extraordinarily resistant in a war time environment you know that there are still parts of the government that consider that those kind of requirements so for them you know ability to tolerate six simultaneous losses or a higher number because you can configure it that way ten simultaneous losses still operate the confidentiality integrity that really drives them in other markets it can be it can translate into something about their business so for other customers of ours what it really means is your bottom line is going to have you know three more cents of earnings every quarter and that's a huge huge deal for them so it really depends you know but it you know technology capabilities are the same but it translates into different things for different customers. We're Chris Gladwin the CEO and president of Cleversafe you're a tech eagle at MIT you have a lot of experience as an entrepreneur you're running a big business now growing big financing you just had I say you've been to MIT East Coast we're here in California a lot of folks out there developing stuff from MIT good tech coming out of these big academic institutions like Stanford MIT and cloud is massively growing market math is our friend math is free ultimately free with the computer available what advice would you give those guys at a starting company not like you know I want to build an iPad out I mean there are some deep tech going on here in CloudAir we have some startups here in CloudAir labs and so open angle labs developing some tech what advice would you give them I mean on your journey five years you took a different approach than some just getting some venture capital what advice would you share aspiring entrepreneur or multi-term entrepreneur well the first thing I look for is a very significant benefit to the customer it's customers are not generally going to buy something from someone new and different it's 20% better you know that's not going to work and a competitor can respond to 20% better you've got to find just fundamental transform their business or you know transform their life kind of benefits so that's the first thing you've got to have you can't underestimate the competitors I mean they're not going to let you have their market I mean it's got to be fundamentally better to be the new entrant so that's the first thing you look for the second thing the number one business reason most new businesses fail is they don't last long enough they're usually right but they're wrong about the timing and they're always wrong on the optimistic side so you can't believe your own PR just because you're introducing this thing and it's great it doesn't mean that everyone's ready to buy it and it doesn't mean they're going to buy it tomorrow so you also have to look at timing and I like to find businesses where the mega trends are on your side like in our case with clever safe you know if you just look at hard drive storage which is the bulk of big storage and eventually this will be true of flash as well hard drives on a more logistic basis get slower every year eventually it will take a month two months to read or write a whole hard drive worth of data which makes those systems less and less reliable because if they fail it'll take a month to rebuild that that's good for us because we solve that problem we go to make that go away so if the pain isn't great enough for you as a customer with a 12 hour rebuild time well we'll be back because it's soon going to be 24 hours and it's soon going to be a weekend so that's just a matter of time and there's other things that we look for that where we just know where you know this is going to happen and you know sometimes the customer is ready to go down sometimes they'll be back in three years but it's going to happen so that's what you look for how does an entrepreneur handle the timing because you're right I mean I would agree with you entrepreneurs are very visionary and they tend to be early and a lot of the web bubble ideas were dead on they're just ahead of time how does an entrepreneur handle that timing issue because you said you mentioned in your you have some scar tissue where you've experienced that in one of your first startups in this one you're taking a little different approach what should an entrepreneur do how do you keep developing how to keep the idea alive and put on the shelf you know who test marketed well you got to get at least close you know with the wireless tablet thing I was off by you know 15 years you're not going to last that long so you got it you can be a couple years off but you got to be within a couple years and then you got to just you got to manage it like a business at the end of the day what businesses do is deliver a value to customers and get paid for it and get paid enough that they can make money that's what businesses do and that's never going to change so it's okay to have a you know front end business model keep you alive and keep you in the market getting data requirements you okay with that yeah a lot of times what you're doing in the early stages of a startup is you're learning you're learning how to describe the great thing that you've created you're learning who should be the person you know type of customer that you should approach there's a lot of learning that goes on and you've also got to be careful to not you know there's times when you you might think I got to like really start spending more money on every on this this this and this and you know if this whole thing's got to happen in the next three months we got you know you got to be careful not to do that too early when that time comes you got to hit it but you also if you hit that wrong and you hit that early yeah it's extremely difficult to recover from it's a little it's easy to recover from a little bit of hesitancy on that but if you don't wait too long but if you think the market's going to go to you know soon and you start spending you'll spend yourself out and you'll never come back question for you obviously you've done some really really good tack and you've written number of patents hundreds of patents the question I always get from entrepreneurs is when do I file a patent if do I file a patent I'll see some of the internet stuff time makes critical and maybe maybe patents will work or not it depends on patents I mean you have a lot of them it's obviously on the on one end you know it's deep tech to protect it when you're not sure no like okay I got to start filing patents or provisional patents or whatever it depends I mean it's always good to file I mean your patent is not arbitrary I mean the patent has very defined standards for what's patentable it has to really be a genuine invention it has to be usable it has to be non-obvious right those are the three criteria so you have to have patentable stuff it's just you don't just get a patent because you want it degree so obviously that's the first qualification it's got to be patentable which is not necessarily an easy standard you know because a lot of things you know there's a lot of prior art you know so that's I mean that's legally a minimum requirement in terms of the if you now have stuff that is patentable do you file you genuinely want to it has a lot of benefits it will give you credibility in the market and it'll help validate you to your customers to press to potential acquires to potential partners to investors so it's useful in that sense you shouldn't expect as a kind of a new company that you're going to collect royalties because the time in which that could happen is you're not going to get the patent for four years and you know the idea that you're going to sue somebody and get money that's very rare you sue some companies and they're going to sue you back with a thousand of their patents and you're the dead end I mean yeah so it's not so much about I'm going to monetize this and I'm just going to sit there and collect royalties but it does have value in all those other things the credibility very long-tracking investors showing some defensibility yeah some of those key requirements of investment in our business more than likely as you said you're not going to use it offensively more than likely you're not going to use it other than to kind of give you some credibility I mean you're not going to use it in a short term but if you do the most likely way will be defensive is that you'll get sued by somebody else and you'll need it we're here with Chris Gladwin the president CEO Cleversafe talking about his business at Cleversafe the innovative approach with dispersal talking about lessons for entrepreneurs CIOs final question for you Chris is should the arrow forward five years or so you talked earlier about how we've seen in the 80s the shift and how things changed what's going to happen over the next five years in your mind that's not so much from a Cleversafe statement just from a market what's going to change what's going to be different in five years what will continue what might shift just speculate a little bit I get very specialized so for the past five years I've thought a lot about cloud storage so I can talk a lot about that when you get me outside of that I get really stale really fast but in the world of cloud storage it's going to take at least five years for the industry to broadly transition I mean look you know everyone thinks like mobile telephony happened overnight because of their experience of adopting mobile telephony happened overnight one day they got a phone but that's not how it worked the time from the first working mobile phone call wireless phone call to the first commercial wireless phone call was about 15 years took a little while for that to become obvious it's very obvious now but it wasn't so obvious in the 70s the same thing is true I mean you know you get in Silicon Valley and you know come to the Cleversafe office and it might seem very obvious what's going to happen in cloud storage but that doesn't mean every single storage administrators on the page for next quarter's purchases so five years is not that much time and you know I think five years is really a realistic time of how long it will take for the industry to broadly adopt and deploy cloud storage congratulations thanks for coming on Chris really appreciate your very smart and clever and got the credentials and great tech with Cleversafe I'm John Furrier with siliconangle.com and siliconangle.tv we're inside the Cube in Palo Alto at the home of big data cloud era where we sublease some space and we have our silicon angle labs thanks for coming on and look forward to next time