 That's so very, very simple. So, demystify the zero client, because we've had HP on earlier about talking about thin clients. We've had folks talking about VDI and interesting VDI and desktop virtualization. It's kind of makes your head spin a little bit. So just clarify, and the Citrix guy talked about specifically what zero client means. What is it? I mean, how does it differ from all these other thin clients? Sure, I mean, at the most fundamental level, it differs because it doesn't have a processor in it. Zero really means no processing occurring at the endpoint, none of any type. And people who were using things like thin clients or PCs to act as the endpoint for a desktop virtualization system are in fact giving themselves twice the work because they're not only gonna be managing the desktop in the cloud or in the data center, they're also gonna be managing a desktop at the endpoint of the network. I know a fair amount about thin clients because prior to coming to Pano, I was the CEO of WISE, which I did for about four years. So I understand what thin clients are good for and I understand what their limitations are. And sort of made that decision. Tell us what are those limitations? Were they good for or were they not good for? Well, they're great for terminal services. That's what they were built for. And as we all know, terminal services is a fabulous application paradigm. But in point of fact, it only addresses a certain percentage of the enterprise applications that people might want to use. So it's terrific for that. The problem is any time you try to retrofit an architecture, especially a hardware architecture, to something new, you end up kind of trying to force the square peg in the round hole. And more or less that's what all these guys do. Can you give an example of the square in the round hole? Specifically like what would be that, what retrofitting you're referring to? Well, sure. I mean, look, when you're doing desktop virtualization, what you're in fact trying to do is move the desktop from the endpoint and put it in the network someplace, right? And you're trying to manage it in the network and you're trying to do that for all these good reasons that our friends at Citrix and everybody else talks about. Decreasing the amount of administration, improving security and all these other things. But if while you're doing that, you're actually leaving a computer at the endpoint, you're not actually taking advantage of all of what desktop virtualization is gonna bring to bear. So you're taking something that was designed to run part of an application or an application stub or management or something else and you're trying to dumb it down. But at the end of the day, you can't dumb it down because it's got a processor, it runs a protocol, it does all the other stuff that you really don't want to have happen in a pure VDI implementation. So one of the other components, it doesn't have a processor, it doesn't have storage. What about a system bus? Talk about the system bus and how that changes. Sure, well, what we have in fact done at Pano is we figured out how to stretch the virtual machine system bus across the network. So in point of fact, if you think about a PC, what a PC is, is a system bus that has a bunch of stuff hung off of it. There's processors, there's memory, there's IO devices, there's all this stuff. And that system bus has certain characteristics. But the way the processor communicates, for instance, with the memory is not using any kind of display protocol. It uses a PCI protocol of some type, which is a very, very low level sort of way of communicating. What we have figured out how to do is effectively fool the virtual machine into believing that that Pano is on its local system bus. And so it treats it exactly the same way. We don't need remote display protocols because we're not doing remote display. As far as the computer is concerned, it's local display. And so that gives us a ton of advantages, including any USB device you plug into a Pano will operate just as if it were plugged into a local PC. We don't need to duplicate drivers. We don't need to run fancy protocols. It just works. Yeah, it just works. So what's this cost? $389 and that's all in. That's all the software, everything you'll ever need. So there've been people that speculated that ultimately this becomes a bundled giveaway. Like a lot of telcos will bundle in phones. Do you see that happening? Open a bank account, get a Pano. Yeah, well, there you go. You know, the nice thing about that is it won't break. But yes, I mean the point, one of the things that we're doing as a company is continually working to drive the cost of these devices lower and lower. Because for VDI, ultimately to really take off, the endpoint has to be commoditized down to a price where it's really insignificant for someone to get into it. John, this is exciting for me personally. I love this era because it's got a geek, gadgety feel to it, but more importantly, it takes our whole theme of, we think that the word desktop's a horrible name because when you have this kind of innovation, you're going to get smaller, faster, cheaper devices. You're looking at wearable computers. You're looking at computers everywhere. I mean, the whole Internet of Things kind of comes into play, not just from a sensor network standpoint. I mean, how do you see that whole trend? I mean, you're essentially talking about a connected device. Not so much consumer electronics. It's essentially a computer. Correct. The way we talk about a Pano was we call it actually a consumption point. It's where people consume computing that's actually going on somewhere else. And it's because of the fact that we have been able to take all of the expensive parts out of the endpoint, that you're going to be able to see this driven down and you're going to be shifting what used to be the cost of the PC into the network. And that's going to allow the entire sort of, the entire ecosystem to be delivered at a much lower price point. So agree with you completely. I mean, the difference with this, and this is the thing people have a hard time sort of getting over, is it's not a computer. It is a part, it's a part of a computer. And the computer is actually the network plus the back end system. It acts like a computer though. As far as the end user is concerned, he's logged into whatever he or she is used to logging into. But that device is simply the IO subsection. It's sort of the card that we used to put in the old big IBM. So from your conversation, obviously you're old school, you're seeing the movies before, multiple generations of innovation. The iPad and the iPhone now Android is kind of given the consumer a taste of this user experience heroine we call it. And it's good. People want more this profound user experience and it's becoming the preferred user experience. So here at Citrus they're talking about the user experience, et cetera, et cetera. But this really kind of changes the game because what you're essentially doing is taking that phenomenon of the iPad that which will accelerate the user adoption of more computing at the edge. How do you see the future evolving? I mean, right now the internet of things has always been discussion around sensor networks. But if you can connect to a network in whatever form, how do you see that playing out? I mean, what's your vision of this edge becoming more intelligent? I mean, mobility is obvious an example, but beyond mobility. We see interestingly, I don't think of the edge as ultimately becoming more intelligent. I think that what's going to happen to edge devices is that they're going to follow the same commoditization patterns that we've seen in the PC world. So over time, people are going to look for ways to make them cheaper and cheaper and cheaper. And the truth is, they're eventually going to get to the point that we got to in the PC world, say five or six years ago, where you realize that you were about as low as you could go as long as you needed memory, a hard drive, a processor, and a case. So edge devices ultimately will follow a paradigm similar to what we've been able to demonstrate with this edge consumption point that we have. So what needs to happen is the networks need to continue to get more and more robust, and the back end systems continue to need to get more and more robust. That's happening as we speak. There are whole parts of the world right now where you can go on wireless networks which are extremely robust, have low latency, have extremely high bandwidth, et cetera. And that is what we're going to see proliferated everywhere, and when that happens, you're not going to have the need to have all of this processing packed into an edge device. You'll rather pack it into the back end system. What's the biggest mega trend that you're seeing that will fuel that vision? I mean, at the network level, honestly, some have said, hey, oh, the network's so fast, carrier grade, blah, blah, blah. I mean, obviously, wireless has changed in the past three, four years alone. What is the biggest trend that you see that needs the most work to enable that kind of innovation with the consumption point? There's two things that need to happen. One is we have to learn to manage massive numbers of virtual machines. And to date, the largest installations aren't measured in millions. They're measured in thousands or tens or perhaps a hundred thousand. If this is going to go into kind of a mainstream deployment, we have to manage millions and tens of millions and hundreds of millions. Nobody's built that data center yet. Second thing that has to happen is network bandwidth has to become far more sort of reliably available and latency has to draw wirelessly. Wire line, we've already seen in the States things like Fios and AT&T U-Verse and things bringing in 10 to 25 megabits to the home. That's going to change everything once we get that rolled out ubiquitously across our market. There are markets already around the world, as I said, Singapore, Germany, others, where you've already got this. And it's changing how people are thinking about not only communication but also about how they do compute. I mean, if you think about those two things you just mentioned, it really makes you think about how early this really is. And I talk to entrepreneurs all the time and Dave and I are both entrepreneurs, both run an analyst firm and we run a media company together. It's so early and entrepreneurs, they're at the ground floor of this. I mean, any advice of folks out there who are thinking about starting companies, whether it's a new programmer and doing some cool stuff with apps to someone who's trying to re-architect a computer out there or networking. What's your advice to those entrepreneurs about are we that early? How early is it? Is it purely embryonic? Are we at a good stage? Or what's your take on that advice? Well, I think where we are right now is we're at sort of the elbow in the catalyzation curve for this stuff. We're not quite there yet. We're all waiting for the big lighthouse accounts to come on. We're waiting for sort of the first massive surge of adoption. Actually, this show is one of the things that you could use to convince yourself that we've actually hit the elbow. Yeah, it's feeling like it's getting there. It is, it certainly is. And so I think we're there. To an entrepreneur, I would tell them, what you really need to do is keep your eyes fixed on the endpoint. It's gonna get there. Nobody could have predicted in 2008 that we'd go into a recession and that slowed a lot of stuff down for a lot of us. Good companies figured out how to sell their ways through it. Great entrepreneurs figured out how to deliver services ahead of the market and they're hanging on now, waiting for it to happen. So John, talk a little bit about Pano at the company. So you guys are founded in 2007. You've had a long history, you mentioned WISE and you also work for Ellison. Yes. And of course, those of us who are old enough to remember Ellison in the mid 90s, Ellison McNeely, very articulately talking about the thin client model and challenging Microsoft. And it's taken a little while, but we're kind of finally there. Talk about the company, where you're at. You obviously got some good backers. I mean, Goldman's a backer, Mayfield Foundation Capital. Give us a little background, the vital statistics and share with our audience what you can. Sure. As you said, the company was founded in 2007, really went to market in 2008, sort of right as the market was crashing. We ended up as a business deciding that we were going to sell product to those people who were buying, which meant that we sold mostly to the counter cyclical verticals like education, healthcare and government, which is I think what you'd hear from everybody here. Because quite frankly, those were the verticals buying virtualization back then. And financial services, right? Although back then, no. Now, maybe. Yeah, I was just going to say that over the last year, what we've seen is the cyclicals coming up, so manufacturing, retail, et cetera, and financial services is thank God, finally investing again. Because in past sort of sea changes in technology, they led the curve. They were always the ones you waited for one of the big three or four to come up and say, we've embraced client server computing or we've embraced a certain network topology. In this case, this market change was led from the bottom. Because the people that were buying were actually the smaller guys. And they ended up with all of the experience about how to actually implement VDI. And that has been fascinating to watch. Because now what happens is we talk to a Fortune 500 company and they want to go talk to a small regional bank because they've been doing VDI for three years. And this just never happened. It always used to be exactly the other way around. And so we have, the company's done very well. We did well through the downturn. And we're now, we're doubling year on year. We'll probably do more than that. What's your head count? We're at about 70 people. 70? Nice. And how much did you raise? The company in the last round raised $20 million. $20 million? Led by Mayfield. Mayfield, and you said foundation was in? Foundation was our first investor. And Goldman led the V round. Yep. Nice. So yeah, it's a good, I think the consumption point is really good. I love the messaging of the consumption point. It kind of makes it, defines a new concept that's pretty consistent with what people do. But the question I'm getting from some of the folks out there in the Twitter sphere is multimedia panel. Obviously multimedia is great. Status, where is that at? How's that working? Multimedia, meaning rich media. Like, you know, videos and so on. All of it works great. We just put out, there was this report that came out from the Sorrel Group, which looked at the display characteristics and user sort of experience of the pano versus some wise devices versus a PC. And interestingly, we beat the thin clients and they found us to be on par with the PC in terms of experience. That included video, that included Skype, that included webcams, that included all that stuff. The trick here and the thing people can't imagine is that it is possible to do 25 or 30 frames per second full screen video without a processor at the end point. They just can't believe it's possible to do and all you have to do is go to our booth. How does Intel feel about this? Well, quite frankly, the more panos that are sold, the more server hardware they sell, the more infrastructure they sell. Just pushing mips around. And Dave, in our research too, we talk to the Intel folks, we've been doing our own independent digging on the analysis. If you look at what Intel's done, their business on the cloud side is booming. They made a good call a few years ago to really double down on the cloud. So, I totally agree with your point. On paper you can say, oh, they must be pissed because no processor, but they're moving that growth in another sector. That's what you're saying. And I think the advantage of the zero client here is that what it allows big companies to do is to shift customer spend from the end point back into the data center or cloud, exactly what you're saying. And when you do this, not only are you shifting low margin business for high margin business, you're shifting to a part of the stack where you get higher services attach rates. And so, for companies like the larger system vendors, that's how they're measuring themselves today. Because they realize that hardware, all of us who have been in hardware know that over time, everything commoditized. And you're right, too. The trend on that, you know, Dave and I are launching a new section called Services Angle. And we talked to a lot of boutique consultancies and the big guys like CSE. And there's a lot of services changes in innovation in the services business. But the consistent theme that we're hearing is for desktop virtualization at the edge, it's about reducing costs. That's it. And there's not a lot of services that people play in there. The services are dropping back at the cloud side and or the re-architecting side. So yeah, it's an element of the overall package, but not fundamental, but still key. And from a cost perspective, it's got to get down, but important for the end user. So, John, talk about your total available market in the use cases, right? Because you got to be plugged in. Right. So that narrows some sectors. But where are the sweet spots? Well, right, yes. You have to be plugged in today. I mean, there are going to be products that'll be released in the not too distant future, which will push that on below. Right now, I would say, if you were to look at the overall enterprise market of connected devices, it's about 300 million PCs. Of that 300 million PCs, the bottom 10 to 20% are well-served by terminal services. They really are. These are disposable employees that you put equipment in front of and you don't really care what the user experience is. I know a lot about those because I used to sell to those. You're saying that's why. This is market, essentially. It is, it is. And then there's the group of applications in the middle, which are really the office productivity applications. And I would say that's probably 65, 70% of the market. And that's where we're targeting. So these are connected, land-based desktops that are servicing sort of the general office worker. Then there's the upper 10% where they really need super high-quality video. You're doing video editing or cat cam drawing where you just really need to have a supercomputer under your desk in order to make it work and work efficiently. So we're going off after that middle. And as I said, if I were to sort of generalize where we are seeing the greatest traction right now, it would be around compliance. So HIPAA compliance, PCI compliance, SOX compliance, other things, where security ends up being a real big deal. We're also seeing it around sort of places where it's difficult to administer. So remote office sales, things like that, where people are very, very quickly figuring out that it's just too expensive to support PCs. Retail too? Yeah, in retail, absolutely. Well, they have to in retail because as you know, because of PCI requirements now, they can't keep credit cards local in the stores, which is basically what all those old point of sale devices did. So they need something like this. Ideally for retail, perfect. Exactly. This device is interesting because there's a little light on the top of it. It's not lit right now, but that button constitutes the only thing that support has to do. That light glows one of three colors. It actually glows blue if everything's good, yellow if there's no VM attached, and red if there's no network. Our support guys basically have a, you know, it's what colors the light. There is one other possibility, there's no light like it is now, and then they tell you to plug it in. But other than that, it's good. And for retail, for harsh environments, for anywhere where you don't necessarily have the expertise on site to do a lot of sort of management, these things are perfect. It's a big load off the IT guys. I mean, the old standard desktop, as we would call it, that middle ground, that 65% that office productivity kind of concept, you know, that's the old roll out the desktops and refresh the operating system. I mean, that's been a blocking and tackling, and it's been a pain in the butt for IT, and it's a cost center for them. So, you know, IT guys don't like to roll out and service PCs like they used to. So I think clearly that's an advantage to these guys. You know, they have management software, they can look at a light, and they can roll out software in the cloud. They can move on to other things and develop applications. I think, you know, our customers would tell you that simply the ability to only have to modify an image or an operating system or upgrade flash or whatever at one place in the network is huge. Also, you know, not having to go there when it breaks and roll a truck or send a person is huge. How do you go to market? I mean, how do you go after those 200 million or so? Current, how do you go out the fat middle? Well, you know, the truth is, it depends upon the size of the customer. We are committed to selling exclusively through channels. So the question here is, what is the channel that you're going to? Small, you know, sort of businesses, we go through resellers. Mid-size, we go through MSPs or system integrators. And at the very high end, we're using OEMs, like Fujitsu, who's a partner of ours, who's already embedded this technology in a range of devices, and they're taking it and rolling it out to their enterprise-class customers. So the Fujitsu I get, right, that's embedded. The others, there's a particular skill set you would think you would need because it's not just, you're not just shipping widgets, right? You're changing the mindset of the buyer, essentially, or maybe they're already bought in, but talk about that a little bit. Sure, I have yet to meet anyone who doesn't believe that ultimately the PC gets sucked into the network. The question is, what's left? And that's really what, that's what this whole thing is about. Everyone wants to get rid of the PC. The problem is that the PC has a tremendous amount of inertia. So, and people are still worried about the risk associated with successful deployments of alternative technologies. So that's starting to change now. So I'll tell you, we see probably of the customers we talk to, there are a good 30 or 40% who have already decided, before they talk to us, that they're jettisoning the PC. That could be for one of a million reasons, but they've already said, we're not going to use PCs on the desktop anymore. We want to move it into the network. That's a pretty easy sale for us. It's the folks that say, I'm not sure. Or I think, you know, I'll try it, but I'm... Prove a concept, blah, blah, blah. And that's what I, that sort of speaks exactly to what I talked about, the elbow of the catalyzation curve. So we're seeing the early adopters definitely in there and definitely in there in a big way. We haven't yet seen mainstream adoption, I would expect by this time two years from now, you'll see a lot more. And there's enough desktop virtualization going on in there that you can leverage that infrastructure. Is that right? Or is it still got to be more of a critical mass for you guys to get up there? Well, yeah, I think it really comes down to, is there a critical mass in the kind of business that the customer is talking about? We have over 1,300 customers and they're scattered all over the world. They're big and small. We have plenty of, you know, we can usually find you someone to talk to that's your size company. And I think that does give people a lot of comfort. But what really gives them comfort is when we go in with a partner who knows how to make that successful. And interestingly, that group right now tends to be the managed service providers who have been doing desktop deployments for a long time and have figured out that desktop virtualization is simply a much more efficient way economically for them to do what they've been doing all along. So they're coming to us and that's a great feeling because it means that we've gotten something right. You talk about big integrators, you talk about more local guys, cars. Well, I'm talking about both. I mean, it can be anything from an MSP like a Fujitsu or an HP or someone who does large scale deployments who takes on 100,000 desktops for original or local. Or it could be a telco. You'd be surprised how many telcos have small business arms where they run the infrastructure. You guys are in red with city too, right? You're local, local here. Okay, great, John. Great to have you on theCUBE. We could spend another hour with you but we got to get on schedule but really appreciate the time spending with us inside theCUBE. Great knowledge, great content. Thanks for sharing. Congratulations on all your success. We love the Pano. Love to get one for our office because it looks cool. Great, coolest product I've seen. Love it, love the trend. Thanks for coming on theCUBE. John Keshe, CEO of Panologic. Thanks very much for coming on. Great, thanks very much.