(upbeat electronic music) >> Welcome back, everyone, to this special Cube conversation. I'm John Furrier in the Cube's Palo Alto Studios. My next guest is Dr. Tom Bradicich, he's a friend of the Cube, works at Hewlett Packard Enterprise, heads up the IOT. He's general manager and vice president of servers, converged edge, IOT systems. But we're here to talk about, not so much HPE but really that work that Tom's done in a topic called First Mover, a book that he's writing. It hasn't come out yet, so we'll get an early preview of what it's like to create a category innovation and how to use process to your advantage, not make it your enemy. (chuckles) How to use creativity and how to motivate people and how to sell it through organizations, whether it's venture capitalists or managers. Tom, you've got great experience, thanks for spending time to come into the studio. >> Great to be here, thanks for having me and I'm happy to have this discussion. >> If you go back to the Cube videos, folks watching that know you, seen all the videos at HPE Discover or HP Discover, back in the day, you had a great career. You were an engineer, built the first notebook computer with IBM, you've done a lot of groundbreaking things and I like the topic of your book, First Mover, 'cause it speaks to your mindset. Entrepreneurial, innovative, breaking through walls, you probably got a lot of scar tissue. So, I want to talk about that. Because this is what the opportunity many entrepreneurs have as you know, in the Cube, we really believe that a renaissance in software development is coming. It's so obvious, open source is growing at a extraordinary pace, reuse of code. >> Right. >> You got IOT. >> You're involved in, you got crypto currency, block chain, all these new waves are coming all at once. >> Yes. >> I wish I was 22 again. >> Because this is a great opportunity to innovate. But this improving things, what are some of those things? Let's jump in, what do you see as the playbook? What have you learned and what can you share? >> Well, sure, I've been blessed, I've had a career where I've been able to do a lot of innovation but also, I like to separate the notion of innovation from differentiation. Now see, it's possible to be innovated and not different. Like it's possible for you and I to have the same new suit. It's new, it's innovative, but it's not different. And differentiation is really where one can have a first mover advantage because differentiation by definition is new, is innovation. But it's not always the other way around. So, I always tell my teams and I always focus on, how can we be two things, both different and better. It's possible also to be different and not as good. You can have the highest failure rate in the industry, you're different but that's not good. >> Right? >> Yeah. >> So, the concept here is how do you be different, not just new and innovative but how to be different and how do you be good. And I've actually faced three risks in mostly the big corporate culture that we've had to innovation. And the first risk is, of course, the obvious one, will customers buy it, that's called market risk. Is it something that's good enough to be purchased at a profit? The second risk is, can it be manufactured at quality and at a rate of consumption. The third risk is your own company, does the company have what it takes, actually, to take on the risk of a brand new product category, not just a new product. But a new category of products that, by definition, have never been done before. And when one can do that, when one can figure that out, and I've had some significant experience with this, you can catapult your careers, you can catapult your company and your customers to new levels because you enjoy the benefits of the first mover. That's the name of the book, The First Mover. >> Well, I'm looking forward to seeing it. But I want to ask, this is super important because a lot of people are really good at something and they run hard, they break through a wall but might have missed something. So, you kind of bring up this holistic picture. What are some of the things that folks should focus in on? Say I have a breakthrough idea, I have a prototype I've been running, it's in market, I think it's the best thing since sliced bread, I'm pushing it hard, people are just going to lap this up, this is going to be great, I know it's innovative but no one else knows it. >> Right, right, yeah. >> What do I do? >> What's the process, what do you recommend? >> Well, what I like to do is portion the benefits into two categories. There's supply side benefits that's to your company. Why is this good for your company to do this? And then there are demand side benefits. Meaning, why is it good for the customer? Most people tend to focus mostly on the demand side. Oh, it's solves this problem and the customers will love it and that's important and I would call that a necessary but not a sufficient condition. The other condition is why is this good for your company? And many times, when it's a brand new product category, those inside a company aren't quite in tune with why it's good for the customer. Because, again, it's a new thing, it's a new product category. Why is an automobile better than a horse and buggy, right? Why is a laptop computer better than a desktop computer? These are the ideas where it may be intuitive, it may be instructive to talk about that but when you can get a business model first and start with that, well, the reason is, we can enjoy this margin. The reason is, we can enjoy this particular first mover advantage, the halo effect, the reputation of being the leader. The reason is because we can penetrate a new market. The reason is we can now overcome a falling revenue in a shrinking tam. Now we can accelerate in another tam, perhaps, as well. So, by coming up with both the demand side and the supply side, you have a better case to go forward for support and funding inside a big corporation. >> There's always product market fit, I hear the buzzwords, I got to get the cashflow positive, break even. There's always a motivating force to get something done. How should someone organize the order of their operations to get something done, to the market, if it's an innovative, groundbreaking, differentiating? Because a lot of the big challenge is, some people call it landing span, I heard that buzzword too but you get a champion inside a company and that champion embraces it and most people think, oh man, I got a customer. But then that person has to sell it through and then it has to be operationalized, meaning, people got to get used to it. These are really challenges. >> They are, yes. >> What is your view of how an entrepreneur or a business executive or practitioner to get through that? >> Well, you have to get people on your side and it's really important. Somebody's got to believe in, either, you not even understanding what you're proposing but they'd say, well, you have a track record. For some reason, I believe what you're saying. And then, secondly, getting customers. So, I have personally never done anything major without a customer that I call an inspiration customer. That's a name I just made up. So, a customer, by definition, is an end user that will buy something from you, that's the definition of a customer. And an inspiration customer is one that will help you that is okay with seeing your dirty laundry, okay with mistakes you might make because they see the value in it and they also see the value in them being a first mover. And I like to tell my team, we want to be a first mover and a trendsetter, so our customers can also be trendsetters in their business as well. So therefore, by getting that customer support, and that's in the form of POCs or in trials or in just customer testimony, combine that now with a second dimension called the analyst community, which you're team resides in as well, also saying well, I think this is good as well, brings a lot credibility because there's a saying, a verse in the bible that a prophet is not without honor except in his own home town. Now, if you think about that, a lot of times, you're own company that you reside in has a lower point of view because it's very consumed with, indeed, what is next and doing the right thing, by the way. I have to make this quarter, right. We have to protect the brand. We have to keep the cashflow coming in. These are all important things, so how do you get someone to focus on that? Many times, it's not you anymore, it's outside. And I call that the second C. The first C is internal, the company. The second C is your customers and the community. That also could include, by the way, analysts, the media, other experts, consultants, those type of Cs around there. Now the third C is the competition. This is a little bit controversial. What happens when the idea is now exploited by the competition first; sometimes that is a motivator for a company to jump on it as well and make the market. But, again, if you follow the competition, you're not the first mover, you don't enjoy the benefits of first mover advantage. Higher margin, the halo effect of being the innovators and also, learning, that's an important one. When you're a first mover, you're out there learning so that you can respond to the second generation in a better way. >> I like the notion of differentiation and innovation as two different variables. >> Yes. >> Because it's super important. You can be different and not innovative. You can be innovative and not different. Again, it's all contextual but I want to get back to the pioneering of the first movers. So, statistically speaking, a lot of the best entrepreneurs are first movers and they're often "misunderstood", you hear that all the time. >> Yes. >> Or being a visionary is the difference being 10 years in the future versus an hour, can make the difference between success. (chuckles) We are crazy on one end and you're brilliant on the other because the time to value catches up with that profit, if you will. So the question is that, how does first movers continue to win 'cause I've seen situations where first movers come in, get a position and win and stay, keep the lead. Other times, first movers come in, set the market up, create all the attention and then have arrows on their back. >> And a second mover enjoys the benefit. >> Yeah, so the second mover comes in, bigger scale, so this competition, competitive strategy overlaid on this. Which even complicates it even further. >> Indeed, yes. >> So, your thoughts on that. >> Yes, indeed. Well, one way to look at this is the way to move forward is again, when you can get some momentum that's not you. That's the number one as a... >> John: Market growth, number of subscribers, doing the internet as a trend. >> Yes. >> Mobile users. >> Yes. >> And a third party consultant who's highly respected, a greaser, an analyst. I ran into an analyst recently in a coffee shop who agreed with some of this first mover work we're doing and converged edge systems, which is a new class of products as well. But it's really important that you can't be discouraged, let me point this out. What I tell my team, and I tell students, I lecture at universities and I've been edge professor, those younger in their career, is if you cast and vision and you have an idea and nobody gets it, don't be discouraged, that's a good sign. That's sounds a little funny. Why is it a good sign? Because if everybody gets it right away, it's likely not that novel, it's likely rather ordinary, it's likely been thought of before as well. So, by the very nature and definition that the average person might think it's discouraging. Oh, nobody understands me, nobody gets this idea, should be an encouragement, and a motivation. Now the risk here, is people not getting it is also a sign of a stupid idea. So, usually, when people don't get it, it's either, really not good. >> Or really good. >> Or really amazing that, eventually, they'll come around to it. I had a boss in one of my career opportunities told me to stop working on a product. I don't want to give too much detail, but he literally told me that. And I said, I didn't want to be insubordinate to a boss, we have them and I said, can I please just keep working on it, okay, don't let it interfere with the other stuff. Dah, dah, dah. Today that market is a nine billion dollar market as well. >> Of that product that you-- >> Of that very product that I was told by a very astute person, one of my colleagues, my bosses, that I don't see the future in this, let's not do this, you know, as well. But, being able to have a second thing. So, number one is don't be discouraged by people not getting it. By definition, that's supposed to happen. >> Yeah. >> When you have new-- >> Good point, you want to finish that? >> I just want to get-- >> Get one more thing. >> If I may add a second one. And as you're moving forward with this as well is seek out and find those who do agree with you and stick with them very, very closely. And I have, I can say a couple of names. There's one, we've created this new product class called Converge Edge Systems. Alan Andriole is senior vice president at HP. >> Cube alumni. >> And he's a Cube alumni. >> Super smart. And I'm pointing him out because he has publicly taken on this idea that this product category can really, really work and he's worked-- >> John: Cloud Nine? >> Oh, the converge edge system called Edgeline. >> Okay, got it. >> The Edgeline product brand. >> You know it as well. So therefore, when you find someone who had authority-- >> Eagles fly together, you want to get a good peer group. >> Absolutely. >> Here's a question for you. >> One of my experiences, and I want to just get your reaction and add on to it, your thoughts is, most entrepreneurs or pioneers are misunderstood, so I agree, don't be discouraged, but also, keep validating and be a data seeker, get the data. But a lot of the times, just getting something in the market or getting it going creates movement and inertia to get rolling and sometimes the original idea is actually the big idea turns into it as you get more data. An example is like Air B&B wasn't... What it is, it was basically air mattresses and selling cereal. >> Yes, yes. >> That was the original story, right. And then it turned into, but conceptually, it was the same thing, so you don't have to be 100% right on the semantics. >> It's well known that most startups don't end up being successful with the product they start with. That's well known fact but that's true also in large companies with a product idea as well. So, you have to have this interesting balance. It's very interesting as I've thought about this in study. You have to have deep philosophical and conviction of principles. And here's why: If you don't, you will be swayed by everybody's opinion and you'll never get anything done because oh, well, that's a good idea, maybe I should do this well, that's a good idea, maybe I should do this. Now, I'm not saying that's bad to listen to others but if you don't have a grounding of principles. Example, we established the seven principles of the IOT over two years ago, and we've held on to them and created the success we have based on those principles. Now that's not to say we didn't modify them a little bit but the point is, we were convicted with something and when somebody would come up with a counter to it, we had a way to defend our convictions, if you will, in internal debates and external debates as well. And then, secondly, you got to be also okay with being the sole inhabitant of that field of discourse. Being a visionary can be a very lonely job because of that, right. And, again, it's because you are and your team is, it's not always a lone person right, the team is actually creating something that literally nobody's ever seen before. Nobody understand before. >> What process do you wrap around this? Because Dave Alonzo and I always talk about this on the Cube and after the Cube is that the process has to be your friend, not your enemy. It has to work for you. >> I always say that, yeah. >> Also says that as well on Amazon. But also Charlie Munger, Warren Buffet's partner always says I'm not a big fan of master plans, meaning, because become a slave to the plan rather than the opportunity. >> Yep, yep. >> So these are process kind of things, right. So how does an innovator that's a first mover that wants to create a category, 'cause categories killers or category creators are huge opportunities financially. So they create a lot of value wealth and opportunity. What process is best? Is there a view, is it conditional on certain things? What's your thoughts on... >> Well, let me say, I'm going to give you a big company or a medium size company context, not a startup, I think they're distinctly different. I have limited experience with a startup but I've had significant experience with bigger, medium and large, now, companies as well. You can't try to change the system because now you have two variables. You got this new product that nobody's ever heard of and now you're trying to change the whole system. Now, again, this is just advice for bigger companies. So be careful how many things you want to change, how many things you want to stop. So you want to take this new thing and align it with existing processes and existing core competencies as much as you can, even though it's new, it has to have some alignment; I'll give you an example. When we built the converged edge systems, the Edgeline brand, we aligned it with compute. It's not only compute, but we aligned it with compute, why? Because HPE or HP, at the time, was and is and now, number one in compute when it comes to data center. Compute systems when it comes to high performance computing and mission critical, right. So therefore, that was easy to understand so you're okay, you're familiar with this, but now, let me tell you this new twist on it. And I would assume, and I don't know this for sure, but I would assume Steve Jobs and the Apple team that was thinking of this smartphone concept, the iPhone as well, they had to align it with some level of compute capabilities, right. And if you notice, as it emerged, it also included something that already exists called the iPod which was already aligned with their laptop computers and their desktops, right. Your music would be downloaded as an app to connectivity, but now you can take it with you and by the way, now I'll add a phone to it and so this incrementally built and by the way, you ain't seen nothing yet, I'm going to add a GPS system, I'm going to add a camera, your flashlight, your wallet, I'm going to add all that in. So, I think, by incrementally moving but not upsetting the system, like you said, in a large company really, really helps because you can't change everything too quickly. You got to be okay being alone-- >> Well, I want to interrupt you there for a second. Peter Buress and I talk all the time; I love his quote, Peter Buress, head Cube on research says, the iPhone was a computer that happened to make phone calls. Okay, and that's the smartphone, it's category creator and we know what happened, the rest is history. However, you mentioned talking to customers, having an inspiration customer, I love that concept. Because you need a muse as an innovator. You got to have someone you can trust that knows what you're trying to do that understands the mission. If Steve Jobs went into the marketplace and did market research, he would have probably had the customer feedback to build the best Blackberry. A better Blackberry or another device. Instead, he used is gut, was on his mission and then he understood the inspirational customer, whether it was real or not, he was going down a different road. It takes guts but also some discipline. >> I hear you and I agree with this 100%. When I had the great fortune of leading a team that created the first enterprise blade server or converge system, and today that is pushing about a 10 billion dollar market opportunity, and not one customer asked me for it. Now, that doesn't mean I didn't listen, okay. But I had to bring it to them. So here's the difference, we're not responding to trends, this is a key point, we're creating a trend. And what I tell my team is, you must create trends, not follow them. Many of competitors, are by the way making good money and doing good business, I'm not knocking that, but I'm saying they're not creating a trend, they're actually following one. They're in an exploding tam. >> Pretty lucrative trend. >> It can be. >> Very mature, big market. >> Dave Thomas with Wendy's followed a trend called hamburgers and he did pretty well. He didn't create the hamburger market but he followed one. Now, this is really rather interesting. So when you come in, and then you're saying I want to actually set a trend and create one, it really gives you this opportunity to redefine what is happening. So now, quick story, you may have heard this, maybe your viewers have heard this. A manager of a shoe company sends two guys to an island. He says, I want you to sell shoes on this island. They get to the island, the first guy calls back and says, boss, this is terrible, everybody is barefoot. There's no opportunity to sell shoes. This is terrible, I'm coming home. The second guy calls and says, boss, you're not going to believe this, there's not a shoe on this island and I have a tam that's 100% of the market to sell shoes. I believe, as you pointed out, Steve Jobs didn't go and say well, what apps do you own on your Blackberry. What he did is reversed it and this is what we're doing, we're reversing, we're saying, if you could watch a full length high definition movie in your hand, would you? Well, I can but I can't do it on this device. But if you could, right. So now, in the IOT, I hear this all the time from my competitors and even some colleagues out in the industry, well, we ask them what apps they run at the Edge. We ask them what they do at the Edge. That's good, that's necessary but not sufficient. You have to say, but if you had this product, wouldn't you, for example, run an entire database? Would you compile your machine learning models at the Edge, do it in the cloud now, wouldn't you do that, if you had it? Well, I never thought of that because I don't have that capability, just like, well, I never thought of being able to take pictures and watch full length high definition movies 'cause I never had it. But what if you did, would you do it? So you always got to be setting that trend, not responding to it only. >> That's awesome. >> Dr. Tom Bradicich, writing a book called First Mover really about being innovative. Give you the final word, thanks for coming in, appreciate you sharing the advice. What's going on with HPE and your IOT work? Take a minute to talk about what's happening at HPE. >> Well, thanks, pretty exciting, we've been able to move forward with some really great customer wins. I'm hoping to go public with them. We're in many ways, I know this is an abused term, but we're revolutionizing the industrial IOT in particular and manufacturing floors. We have the large auto-manufacturer that has chosen Edgeline as the standard to produce more and more vehicles per day. That's their goal, how many more vehicles can I get into my customer's hands per day. We have snack company making potato chips. Looking at what we're doing with sulfur, defining operations. We have even, we've talked about this before, space travel, engage with what the space edge is all about. In many ways, we're potato chips to space ships. >> Data centers on Mars. >> Data centers everywhere. >> And then, also, converging OT, just like the smartphone converged the camera and the GPS system, we're converging control systems, data acquisition systems. It's pretty exciting, I've been fortunate to have a company and our new CEO, Antonia Neery, has been very supportive, I was with him this morning and we talked about that new, first-of-a-kind product that we have at this auto-- >> So, is Antonio going to let us come in and do an exclusive interview since he's been a Cube alumni multiple times? >> Yes, I think he should. >> Tell him we said hello. >> I will, I will. >> Tom, great to see you. >> Thanks for having me. >> Tom Bradicich, great thought leader, really around category killers, category creators, being innovative and different, that's the key to success. Thanks for sharing. This is the Cube Conversation here in Palo Alto, I'm John Furrier, thanks for watching. (upbeat electronic music)

>> Narrator: Live from Madrid, Spain, it's theCUBE, covering HPE Discover Madrid 2017, brought to you by Hewlett Packard Enterprise. >> Welcome back to Madrid, Spain, everybody. This is theCUBE, the leader in live tech coverage, and this is day two of our exclusive coverage of HPE Discover 2017. I'm Dave Vellante with my co-host Peter Burris. Last night was a great night of customer meetings. We stumbled into the CIO meeting, we were at the-- >> And were quickly ushered out. (both laugh) >> We were at the analyst event, and of course we met our good friend Dr. Tom Bradicich at the analyst meeting. This is the man who brought a lot of the IOT Initiative into HPE. He's the general manager of the IOT and Systems division. Great to see you again, Dr. Tom. Thanks so much for coming on. >> Thank you Dave and Peter, it's great to be here at theCUBE, great to be here at HPE Discover Madrid. Lots of great things happening, I can't wait to tell you about 'em. >> So we're very excited to have you on. John Furg and I interviewed you in the very early days after you came over from your previous company, and you had this sort of vision of, you know, bringing the HPE into the intelligent edge. >> Yes. >> And we're like okay, this sounds really complicated. You got ecosystem, you got all kinds of technologies that you gotta develop. Hardware, software. And you're making it happen. It's become a meaningful portion of HPE's business, so I know you got a long way to go, but congratulations on the progress so far. >> Thank you. Give us the update on the-- >> Well, first of all, thank you for that, I appreciate it. I must give credit to my team, I tell them all the time that if you don't execute and do the work, I'm just a science fiction writer. (interviewers laugh) And the vision has come about, and we have real customer deployments of course that the, you know, the proof of it. >> Right. >> At first we had no products and no customers, now we have these products that we'll talk about, and we have the customer deployments, and we're changing things for businesses at the edge, and again the edge is just not the data center. And the manufacturing floor, we'll talk about refineries, oil rigs, those type of edges. We're doing a lot of work there. And it's been exciting to see the ideas that we have get adopted by not only customers, but the industry, so we're seeing other analysts pick up on two dimensions: computing at the edge, and a little more complicated one, a little more difficult to grasp, is converged OT and IT at the edge, the two worlds of operational technology converging with IT. We were on theCUBE talking with an OT partner, National Instruments, a long while ago, and now we literally have those products in the market in the hands of customers. National Instruments is reselling the Edgeline 1000, the Edgeline 4000 products, as well as of course us selling it, and it's pretty exciting to see this happening. >> Well what I love about that conversation is, you know, when we first started to talk to you, we said okay, let's play the skeptic, analysts are skeptic. >> Sure. >> And we said one of the big problems you're gonna face is bringing the organizations together, OT and IT. They're just different worlds, oil and water, you know, you got hardcore engineers and you got IT guys, and then subsequent to that conversation, you bring on National Instrument, right? >> Yes. >> And we have that conversation. Okay, so we sit down, I check that box, at least they're having conversations. Can you talk about how that convergence is actually occurring, and what's in it for the customer? >> Well great. To talk about this convergence, the best thing to do is say it can happen at several levels. It can happen at a solutions level, it can happen at a software level and a hardware, physical level. Let's talk about a physical level, it's a little more tangible to understand. Let me use the smartphone, which everybody has. Like Peter, you have one there. If you hold that up, you will notice inside the manufacturer of that phone converged, or integrated, those are synonyms, many consumer devices. Such as what? A music player, of course, the phone, of course. But also many other things. A GPS system. >> Camera. >> A camera. The list goes on, right? We can go on. Oh, the flashlight, and by the way, your wallet. Maybe not your wallet, but a millennial and younger's wallet-- >> Yeah, sure. >> Is in that phone. >> My wallet's in it. >> My wallet's in it. >> In it, and-- >> Venmo, baby. >> That's right. (all laugh) >> I have my kids' wallets in there too. >> Oh that's great, you've done that switch. So what is happening there obviously is the notion of we're, you know, software defining and we're converging. Now the benefits of that are irrefutable. One thing you buy, it's less energy. One thing to manage, the convenience of carrying it around. Let's take that metaphor and impute it at, let me say a manufacturing floor edge. There's lots of edges out there. We go to a manufacturing floor edge, we see several devices, just like the early pioneers of the smartphone saw a consumer with a camera around his neck, a GPS on his belt, text, right, a flashlight, a wallet, and all this. We see all these devices out there, and what are they? Some of 'em are OT, as you mentioned. Operational technology devices such as control systems, such as data acquisition systems. >> Real-time systems. >> Real-time systems, industrial networks. CAN, PROFIBUS, SCADA solutions and networks. And the second thing we see is some IT. Most of it's closed, so this is important. It's good IT, meaning computing and storage, but a lot of it is closed systems. It's not the open EXEDY 6 architecture that we so enjoy in the data center. So those things are out there. We looked at 'em and we put them all in one box, just like the smartphone is one device. What are the benefits? Lower space, there's not a lot of space at the edge. Lower energy, there's not a lot of energy, right, at the edge. But the more profound benefits that we're seeing, and we have a large auto manufacturer who has deployed this on their manufacturing line, is it keeps uptime higher. In other words, it reduces downtime. So if the manufacturing line stops, there's nothing worse than a manufacturing line stopped, except perhaps an empty one. But the point is, when a manufacturing line stops, you can't put out product. You can't put out product, you can't recognize revenue get it in the consumer's hands. It's very obvious. It's an air-tight business case, actually. So we're able to reduce any downtime, why? Because first of all, everything's together, and secondly, we're able to manage it just like we're managing the data center because it's an open EXEDY 6 architecture. >> So you're converging tasks as well as hardware. >> As well as hardware, and then the next step is software, you know, as well. We just launched a new class of software called the Edgeline Services Platform, and this is OT software. So we're talking OT functions like aggregators and things that do OT technologies and some IT, but because we have so much compute power and it's open, it's EXEDY 6, it can run software like VMware, Microsoft Products, even database products as well. But because we have that, we're able to software define. When you software define, and I'll use the wallet again. You don't have a billfold with your license anymore. Plastic and leather has been software defined, and therefore it's less to deal with. It's much more efficient. So that announcement of our software strategy along now with our hardware strategy is very exciting for us, and customers are very much interested in it. >> So do you have some examples, you know, some real world examples? Customers that you can talk about where you're bringing together OT and IT disciplines? >> Yeah, you bet. Yeah, you bet. Let me talk about a large global beverage and snack company, and they make snacks, and in this case, potato chips. So a potato chip is a product, and the idea of having them come out of the line in the bag and be a higher quality is important. So we took an Edgeline System, the EL 1000, and we put it at the edge, and we were able to software define several of their IT and OT components and get it to a consolidation and integration in one box. Now what that did is it allowed the, and will do, is allowed the foods to move faster. So if they move across the conveyor belt faster, you can bag them faster, get 'em out to the consumer. The second thing is because it's so powerful, this is interesting. Now they can use video cameras to inspect the quality. Now think about that. That's not necessarily a new idea, but what is new is the notion that you can take video, which I think you'd agree is the largest data, is that right? A video is big, big data. >> We know that well. >> Especially if it's high, Yeah, especially if it's higher resolution, and your hosting costs are telling you that as well, right? Of all these videos. But if it's high resolution, and because you're looking for, you know, defects, indeed, one has to process that not only in high resolution, massive data, number one. Number two, quickly, because the thing is moving, and you wanna know to knock it off or stop or whatever the case may be. So what has happened there is my team and I did not think of that. Our customers thought that, well because you gave us this platform, we can now enhance it with a new type of sensor called a camera, with a new type of data, called video, to enhance our quality and keep our process moving faster. >> So keeping this converged notion going, you're converging the hardware, which is, you know, important. You're converging a lot of the administrative tasks. >> Yes. >> Which reduces the likelihood of any single human failure bringing the whole system down, but now you're talking about, in the whole sense, infer, and act loop that typifies what happens at the edge, you're converging new technologies into that loop by being able to add new data type, bring modeling, machine learning, analytics, in the infer, and then being able to act right there, which allows you to think about new invention, new innovation very, very rapidly because you have the processing power to converge all that new function as it becomes better understood. Have I got that right? >> You got it right. I serve as an adjunct professor at university, so let me position it in an easy way to learn. You said sense, infer, and act. Let's just call 'em the three A's. Acquire, analyze, and act. >> Okay. >> It's just easier to remember. And let me talk to that too, but this is actually just synonyms. So the acquisition of the data is through sensors in D to A conversion, or let me say A to D, analog to digital. Because most of these phenomenon, video for example, it has to be, is a light phenomenon. Moisture, pressure. At Duke Energy, for example, the second largest energy provider I worked on that industrial internet of things solution, and vibration was the thing that needed to be acquired and then analog to digital. Now the analysis has to take place. There are seven reasons to analyze at the edge. There are seven reasons not to send the data to the cloud. In the past, we have talked about it. One of them's latency, one of them's cost, one of them's bandwidth, another one is security, another one is reliability, another one is geofencing and policy, another one is duplication and security, you know, hostile or just, you know, reliability drop packets. There's a lot of issues to do that analysis there. But because we have a non-compromised full EXEDY 6, in fact, 64 in one box. 64 Xeon, Intel Xeon product in one box. We don't have to compromise the stack. We can take it directly out of the data center and run things like artificial intelligence, machine learning algorithms. We can virtualize, we can containerize, we can run Citrix applications at the edge to have better access to the data and of course the application. But you're absolutely right, and then the second thing in this point is we move from the middle A, analysis right, to the action. The reason, I've learned this doing many IOT deployments. The reason people do an IOT deployment is to act. Yes, it's exciting to collect data. It's also exciting to analyze it. But have you ever been in a business meeting where you sit and you analyze data and you give tremendous insights, and one conclusion is pit against another conclusion and it cancels out all conclusiveness, and then you talk and you analyze, and you walk out and nothing happens, there's no action. Many of us have been in that. That's the idea here. You can't stop at the analysis, even though artificial intelligence, deep algorithms, moving averages, signatures that we can compare are very powerful. Well, what do you do when you do that? Because we have control and actuation systems built into Edgeline, we literally in a physically space, as well as in a logical process, as you pointed out, close that loop. >> Right. >> Acquire, analyze, act, acquire, analyze, act. Yes, connect to the cloud or the data center if we need to, but the issue is you don't have to. Now here's what's profound about that. This system at the edge can be managed and run the same stacks as any cloud or data center. I'm gonna use those as synonyms because a cloud is just a data center that nobody's supposed to know where it is. So a data center far away on the corporate campus or in a public or private cloud somewhere, is managed the same way. When that happens, we are revolutionizing workload management. Now, I spent a lot of years in my former time in IT and building data centers and building some of the first clouds, workload management's a big deal. How do you shift the workload to the free server? >> Peter: Right. >> Or to the free resources, right? To optimize, obviously. And it's a packing problem many times in the data center. Well now we've introduced another place to workload manage. >> Right. >> It's called the edge, it's far away. So we workload managed in the data center, then the cloud was invented, that's the first off premises. The next off premises is now the edge. So the other off premise is the edge. So now we have a workload management capability. Do you wanna do 100% processing at the edge where the action is, and where the acquisition is? Do you wanna do 100% in the cloud? That's still possible. Do you wanna do 50-50? Would you like to do 10-90? Would you like to do 30-70? You get my point. >> Totally. >> I can shift this, and depending on the season, depending on issues like disaster recovery, depending on your workloads, you can now do that, and again, you can do this with the Edgeline 1000, the Edgeline 4000, because of the processing power and the converged OT inside it. >> Well our observation is that it's not about bringing your business to the cloud, it's about bringing the cloud to your business. >> Yes. >> So bringing that sense of workload management. You know, you might say the cloud is just a virtualized data center when you come right down to it. So bringing all those capabilities and bringing them to wherever the data requires it. And there's gonna be a lot of instances where the data is gonna be at the edge, stay at the edge, but that doesn't mean you don't want all the benefits of how you run computing data at the edge where that data is. >> Yeah, and we're not obviating, we're offering choice. >> Right. >> But again, there are seven reason I went over why you do it here, but I've had a customer say none of those seven matter. So okay, we send everything to the cloud, and we have great cloud hybrid IT products that do that. >> Yeah. >> And we've envisioned a three-tier data model, you know, real time at the edge. >> Yes. >> Maybe you don't persist everything, but like you said, there are a lot of reasons not to move all the data back. But there is maybe a spot where you aggregate some of that data from discrete devices, and sure, if you wanna do some deep modeling in the cloud, go for it. And that cloud might be the public cloud, it might be your own private cloud. Does that seem reasonable to you? >> Very reasonable, and another reason for a cloud is it's an aggregation point for other, in this case, manufacturing lines where other smart cities to come together, because you're not gonna connect every city, every plant, any to any. You'll have a hub and spoke model where the cloud serves as that hub. So there are always reasons, and that's why, you know, if you look at our company, the pillars of our company, Pointnext services, the second pillar is hybrid IT, primarily focused on cloud and data centers, and the third is the intelligent edge. And those all play very, very closely together, in fact we have edge to core strategies, we have edge to core offerings with partners like NVIDEA, with partners like SAP, with partners like SAS, we have edge to core. For example, Schneider as well, Schneider Electric. All of them are looking at this idea, GE, Microsoft Azure, let's go to the edge. And two years ago, that was not the case, right? Let's go there, when you go to the edge, what are you gonna run it on? Well, let's not force our software partners to re-architect like they used to have to to run at the edge, which is like I'd call that drive-by analytics. You just have to cut out everything because it only ran on a wimpy core somewhere or a little device. No, let's move the entire data center capability out to the edge, when I was presenting this to one of our partners, the CEO of the company, I was presenting this vision, and he was texting during my talk 'cause I was boring. (interviewers laugh) And then I said this, this is a very powerful company, I won't mention names. Then I said, we're gonna move data center class technology out to the edge. It's not gonna be in compromised cores or limited memory or a little bit of storage. It's the very things in the data center we'll harden called Edgeline. We'll add controls systems and data acquisition, we'll put it out at the edge. He stopped texting. Then he looked up at me and said, "Wow, you're really moving a data center out to the edge." and you just said that, right? It's the cloud is coming. It's almost a reverse idea of what was happening before. >> Well you wrote a blog recently. >> Yes. >> About the space edge. So I wanted to ask you about that. What's going on in the space, and that's the ultimate edge, I guess. >> The infinite edge. >> The infinite edge. Explain what you guys are doing there and why it's important. >> Well, this is exciting. Space travel for exploration and eventually colonization, if you would believe that, is happening. We have the first supercomputer technology in a NASA spaceship now. It has orbited the Earth well over 1,000 times and it is doing thousands of benchmarks and is doing very well, isn't failing. Now, why is that profound? Because again, that edge is so far away and the ability to push that back to Earth now, which we could call the data centers on Earth, is limited. It takes minutes, sometimes even longer. There's issues with reliability as well. So we were able to do that, and then we've created a new thing called Project Extreme Edge, where we're going to build Edgeline systems that will fit better with lower energy, smaller size in spaceships, and eventually in colonization, but we're just going into space travel and exploration right now. And I'd like to mention that HP Labs is a great participant in this because they're working on a technology, and the name of it is called the Dot-Product Engine. And dot-product is a mathematical operation needed in high-performance computing and artificial intelligence. But we're able to use that technology because it's small, it's fast, faster than we believe anything else on the market, and also it has a low energy profile. And those are all any edge, obviously, but it's also great for the space edge, and I like to quote Frank Sinatra when he said if I can make it there, I can make it anywhere, New York, New York. (laughs) Well, if we can make it in the space edge, these Earth edges will benefit as well. Some of the same challenges. >> All right, we're out of time, but I gotta ask you. Meg stopped by yesterday, and was giving great support for the intelligence. >> She has, yes. >> The company's now reporting the intelligent edge is gonna be one of the main areas. What about the new guy? Antonio. >> Antonio Neri. >> You know, what's your relationship with him, experience? Has he been focused on this area? >> Support? >> He's been great, he supports in three ways, let me just sum up in three ways. Number one, he supports in customer visits. He and I have been on customer visits together, it's always wonderful to have the president and now the new CEO with you affirming what we're doing. That's number one of three, number two of three, he supports the work we're doing with our new global IoT innovation labs, in fact our first grand opening, the first one in Houston, we will have one in Singapore opening in February, and then we'll have one in Europe and perhaps one in India, we're opening these labs for innovation, but my point is, the one in Houston, our first grand opening, Antonio Neri came personally and did the ribbon cutting and sponsored that as well. And then third, he is of course funding my business unit, and he's been very, very supportive and I'm really happy that he's staying with us and he'll be CEO. >> Excellent, Dr. Tom, thanks so much for coming on theCUBE. Congratulations, as you say, I know there's a long way to go, but looks like you're off to a great start and have some real traction. >> Tom: Thank you very much. >> So we appreciate your time and your insights. Okay, keep it right there buddy, we'll be back with our next guest. This is theCUBE, we're live from Madrid. Be right back. (upbeat electronic music)

>> Voiceover: Live from Las Vegas, it's theCUBE, covering Discover 2016, Las Vegas. Brought to you by Hewlett Packard Enterprise. Now, here are your hosts, John Furrier and Dave Vellante. >> Okay, welcome back everyone. We are here live in Las Vegas for SiliconANGLE Media's theCUBE. It's our flagship program, we go out to the events to extract the signal from the noise, we're your exclusive coverage of HP Enterprise, Discover 2016, I'm John Furrier with my co-host, Dave Vellante, extracting the signals from the noise with two great guests, Dr. Tom Bradicich, VP and General Manager of the servers and IoT systems, and Eric Starkloff, the EVP of Global Sales and Marketing at National Instruments, welcome back to theCUBE. >> Thank you. >> John: Welcome for the first time Cube alumni, welcome to theCUBE. >> Thank you. >> So we are seeing a real interesting historic announcement from HP, because not only is there an IoT announcement this morning that you are the architect of, but the twist that you're taking with IoT, is very cutting edge, kind of like I just had Google IO, and at these big conferences they always have some sort of sexy demo, that's to kind of show the customers the future, like AI, or you know, Oculus Rift goggles as the future of their application, but you actually don't have something that's futuristic, it's reality, you have a new product, around IoT, at the Edge, Edgeline, the announcements are all online. Tom, but you guys did something different. And Eric's here for a reason, we'll get to that in a second, but the announcement represents a significant bet. That you're making, and HP's making, on the future of IoT. Please share the vision, and the importance of this event. >> Well thank you, and it's great to be back here with you guys. We've looked around and we could not find anything that existed today, if you will, to satisfy the needs of this industry and our customers. So we had to create not only a new product, but a new product category. A category of products that didn't exist before, and the new Edgeline1000, and the Edgeline4000 are the first entrance into this new product category. Now, what's a new product category? Well, whoever invented the first automobile, there was not a category of automobiles. When the first automobile was invented, it created a new product category called automobiles, and today everybody has a new entry into that as well. So we're creating a new product category, called converged IoT systems. Converged IoT systems are needed to deliver the real-time insights, real-time response, and advance the business outcomes, or the engineering outcomes, or the scientific outcomes, depending on the situation of our customers. They're needed to do that. Now when you have a name, converged, that means somewhat, a synonym is integration, what did we integrate? Now, I want to tell you the three major things we integrated, one of which comes from Eric, and the fine National Instruments company, that makes this technology that we actually put in, to the single box. And I can't wait to tell you more about it, but that's what we did, a new product category, not just two new products. >> So, you guys are bringing two industries together, again, that's not only just point technologies or platforms, in tooling, you're bringing disparate kind of players together. >> Yes. >> But it's not just a partnership, it's not like shaking hands and doing a strategic partnership, so there's real meat on the bone here. Eric, talk about one, the importance of this integration of two industries, basically, coming together, converged category if you will, or industry, and what specifically is in the box or in the technology. >> Yeah, I think you hit it exactly right. I mean, everyone talks about the convergence of OT, or operational technology, and IT. And we're actually doing it together. I represent the OT side, National Instruments is a global leader. >> John: OT, it means, just for the audience? >> Operational Technology, it's basically industrial equipment, measurement equipment, the thing that is connected to the real world. Taking data and controlling the thing that is in the internet of things, or the industrial internet of things as we play. And we've been doing internet of... >> And IT is Information Technologies, we know what that is, OT is... >> I figured that one you knew, OT is Operational Technology. We've been doing IoT before it was a buzzword. Doing measurement and control systems on industrial equipment. So when we say we're making it real, this Edgeline system actually incorporates in National Instruments technology, on an industry standard called PXI. And it is a measurement and control standard that's ubiquitous in the industry, and it's used to connect to the real world, to connect to sensors, actuators, to take in image data, and temperature data and all of those things, to instrument the world, and take in huge amounts of analog data, and then apply the compute power of an Edgeline system onto that application. >> We don't talk a lot about analog data in the IT world. >> Yeah. >> Why is analog data so important, I mean it's prevalent obviously in your world. Talk a little bit more about that. >> It's the largest source of data in the world, as Tom says it's the oldest as well. Analog, of course if you think about it, the analog world is literally infinite. And it's only limited by how many things we want to measure, and how fast we measure them. And the trend in technology is more measurement points and faster. Let me give you a couple of examples of the world we live in. Our customers have acquired over the years, approximately 22 exabytes of data. We don't deal with exabytes that often, I'll give an analogy. It's streaming high definition video, continuously, for a million years, produces 22 exabytes of data. Customers like CERN, that do the Large Hadron Collider, they're a customer of ours, they take huge amounts of analog data. Every time they do an experiment, it's the equivalent of 14 million images, photographs, that they take per second. They create 25 petabytes of data each year. The importance of this and the importance of Edgeline, and we'll get into this some, is that when you have that quantity of data, you need to push processing, and compute technology, towards the edge. For two main reasons. One, is the quantity of data, doesn't lend itself, or takes up too much bandwidth, to be streaming all of it back to central, to cloud, or centralized storage locations. The other one that's very, very important is latency. In the applications that we serve, you often need to make a decision in microseconds. And that means that the processing needs to be done, literally the speed of light is a limiting factor, the processing must be done on the edge, at the thing itself. >> So basically you need a data center at the edge. >> A great way to say it. >> A great way to say it. And this data, or big analog data as we love to call it, is things like particulates, motion, acceleration, voltage, light, sound, location, such as GPS, as well as many other things like vibration and moisture. That is the data that is pent up in things. In the internet of things. And Eric's company National Instruments, can extract that data, digitize it, make it ones and zeroes, and put it into the IT world where we can compute it and gain these insights and actions. So we really have a seminal moment here. We really have the OT industry represented by Eric, connecting with the IT industry, in the same box, literally in the same product in the box, not just a partnership as you pointed out. In fact it's quite a moment, I think we should have a photo op here, shaking hands, two industries coming together. >> So you talk about this new product category. What are the parameters of a new product category? You gave an example of an automobile, okay, but nobody had ever seen one before, but now you're bringing together sort of two worlds. What defines the parameters of a product category, such that it warrants a new category? >> Well, in general, never been done before, and accomplishes something that's not been done before, so that would be more general. But very specifically, this new product, EL1000 and EL4000, creates a new product category because this is an industry first. Never before have we taken data acquisition and capture technology from National Instruments, and data control technology from National Instruments, put that in the same box as deep compute. Deep x86 compute. What do I mean by deep? 64 xeon cores. As you said, a piece of the data center. But that's not all we converged. We took Enterprise Class systems management, something that HP has done very well for many, many years. We've taken the Hewlett Packard Enterprise iLo lights-out technology, converged that as well. In addition we put storage in there. 10s of terabytes of storage can be at the edge. So by this combination of things, that did exist before, the elements of course, by that combination of things, we've created this new product category. >> And is there a data store out there as well? A database? >> Oh yes, now since we have, this is the profundity of what I said, lies in the fact that because we have so many cores, so close to the acquisition of the data, from National Instruments, we can run virtually any application that runs on an x86 server. So, and I'm not exaggerating, thousands. Thousands of databases. Machine learning. Manageability, insight, visualization of data. Data capture tools, that all run on servers and workstations, now run at the edge. Again, that's never been done before, in the sense that at the edge today, are very weak processing. Very weak, and you can't just run an unmodified app, at that level. >> And in terms of the value chain, National Instruments is a supplier to this new product category? Is that the right way to think about it? >> An ingredient, a solution ingredient but just like we are, number one, but we are both reselling the product together. >> Dave: Okay. >> So we've jointly, collaboratively, developed this together. >> So it's engineers and engineers getting together, building the product. >> Exactly. His engineers, mine, we worked extremely close, and produced this beauty. >> We had a conversation yesterday, argument about the iPhone, I was saying hey, this was a game-changing category, if you will, because it was a computer that had software that could make phone calls. Versus the other guys, who had a phone, that could do text messages and do email. With a browser. >> Tom: With that converged product. >> So this would be similar, if I may, and you can correct me if I'm wrong, I want you to correct me and clarify, what you're saying is, you guys essentially looked at the edge differently, saying let's build the data center, at the edge, in theory or in concept here, in a little concept, but in theory, the power of a data center, that happens to do edge stuff. >> Tom: That's right. >> Is that accurate? >> I think it's very accurate. Let me make a point and let you respond. >> Okay. >> Neapolitan ice cream has three flavors. Chocolate, vanilla, strawberry, all in one box. That's what we did with this Edgeline. What's the value of that? Well, you can carry it, you can store it, you can serve it more conveniently, with everything together. You could have separate boxes, of chocolate, vanilla, and strawberry, that existed, right, but coming together, that convergence is key. We did that with deep compute, with data capture and control, and then systems management and Enterprise class device and systems management. And I'd like to explain why this is a product. Why would you use this product, you know, as well. Before I continue though, I want to get to the seven reasons why you would use this. And we'll go fast. But seven reasons why. But would you like to add anything about the definition of the conversion? >> Yeah, I was going to just give a little perspective, from an OT and an industrial OT kind of perspective. This world has generally lived in a silo away from IT. >> Mm-hmm. >> It's been proprietary networking standards, not been connected to the rest of the enterprise. That's the huge opportunity when we talk about the IoT, or the industrial IT, is connecting that to the rest of the enterprise. Let me give you an example. One of our customers is Duke Energy. They've implemented an online monitoring system for all of their power generation plants. They have 2,000 of our devices called CompactRIO, that connect to 30,000 sensors across all of their generation plants, getting real-time monitoring, predictive analytics, predictive failure, and it needs to have processing close to the edge, that latency issue I mentioned? They need to basically be able to do deep processing and potentially shut down a machine. Immediately if it's an a condition that warrants so. The importance here is that as those things are brought online, into IT infrastructure, the importance of deep compute, and the importance of the security and the capability that HPE has, becomes critical to our customers in the industrial internet of things. >> Well, I want to push back and just kind of play devil's advocate, and kind of poke holes in your thesis, if I can. >> Eric: Sure thing. >> So you got the probes and all the sensors and all the analog stuff that's been going on for you know, years and years, powering and instrumentation. You've got the box. So okay, I'm a customer. I have other stuff I might put in there, so I don't want to just rely on just your two stuff. Your technologies. So how do you deal with the corner case of I might have my own different devices, it's connected through IT, is that just a requirement on your end, or is that... How do you deal with the multi-vendor thing? >> It has to be an open standard. And there's two elements of open standard in this product, I'll let Tom come in on one, but one of them is, the actual IO standard, that connects to the physical world, we said it's something called PXI. National Instruments is a major vendor within this PXI market, but it is an open standard, there are 70 different vendors, thousands of products, so that part of it in connecting to the physical world, is built on an open standard, and the rest of the platform is as well. >> Indeed. Can I go back to your metaphor of the smartphone that you held up? There are times even today, but it's getting less and less, that people still carry around a camera. Or a second phone. Or a music player. Or the Beats headphones, et cetera, right? There's still time for that. So to answer your question, it's not a replacement for everything. But very frankly, the vision is over time, just like the smartphone, and the app store, more and more will get converged into this platform. So it's an introduction of a platform, we've done the inaugural convergence of the aforementioned data capture, high compute, management, storage, and we'll continue to add more and more, again, just like the smartphone analogy. And there will still be peripheral solutions around, to address your point. >> But your multi-vendor strategy if I get this right, doesn't prevent you, doesn't foreclose the customer's benefits in any way, so they connect through IT, they're connected into the box and benefits. You changed, they're just not converged inside the box. >> At this point. But I'm getting calls regularly, and you may too, Eric, of other vendors saying, I want in. I would like to relate that conceptually to the app store. Third party apps are being produced all the time that go onto this platform. And it's pretty exciting. >> And before you get to your seven killer attributes, what's the business model? So you guys have jointly engineered this product, you're jointly selling it through your channels, >> Eric: Yes. >> If you have a large customer like GE for example, who just sort of made the public commitment to HPE infrastructure. How will you guys "split the booty," so to speak? (laughter) >> Well we are actually, as Tom said we are doing reselling, we'll be reselling this through our channel, but I think one of the key things is bringing together our mutual expertise. Because when we talk about convergence of OT and IT, it's also bringing together the engineering expertise of our two companies. We really understand acquiring data from the real world, controlling industrial systems. HPE is the world leader in IT technology. And so, we'll be working together and mutually with customers to bring those two perspectives together, and we see huge opportunity in that. >> Yeah, okay so it's engineering. You guys are primarily a channel company anyway, so. >> Actually, I can make it frankly real simple, knowing that if we go back to the Neapolitan ice cream, and we reference National Instruments as chocolate, they have all the contact with the chocolate vendor, the chocolate customers if you will. We have all the vanilla. So we can go in and then pull each other that way, and then go in and pull this way, right? So that's one way as this market develops. And that's going to very powerful because indeed, the more we talk about when it used to be separated, before today, the more we're expressing that also separate customers. That the other guy does not know. And that's the key here in this relationship. >> So talk about the trend we're hearing here at the show, I mean it's been around in IT for a long time. But more now with the agility, the DevOps and cloud and everything. End to end management. Because that seems to be the table stakes. Do you address any of that in the announcement, is it part, does it fit right in? >> Absolutely, because, when we take, and we shift left, this is one of our monikers, we shift left. The data center and the cloud is on the right, and we're shifting left the data center class capabilities, out to the edge. That's why we call it shift left. And we meet, our partner National Instruments is already there, and an expert and a leader. As we shift left, we're also shifting with it, the manageability capabilities and the software that runs the management. Whether it be infrastructure, I mean I can do virtualization at the edge now, with a very popular virtualization package, I can do remote desktops like the Citrix company, the VMware company, these technologies and databases that come from our own Vertica database, that come from PTC, a great partner, with again, operations technology. Things that were running already in the data center now, get to run there. >> So you bring the benefit to the IT guy, out to the edge, to management, and Eric, you get the benefit of connecting into IT, to bring that data benefits into the business processes. >> Exactly. And as the industrial internet of things scales to billions of machines that have monitoring, and online monitoring capability, that's critical. Right, it has to be manageable. You have to be able to have these IT capabilities in order to manage such a diverse set of assets. >> Well, the big data group can basically validate that, and the whole big data thesis is, moving data where it needs to be, and having data about physical analog stuff, assets, can come in and surface more insight. >> Exactly. The biggest data of all. >> And vice versa. >> Yup. >> All right, we've got to get to the significant seven, we only have a few minutes left. >> All right. Oh yeah. >> Hit us. >> Yeah, yeah. And we're cliffhanging here on that one. But let me go through them real quick. So the question is, why wouldn't I just, you know, rudimentary collect the data, do some rudimentary analytics, send it all up to the cloud. In fact you hear that today a lot, pop-up. Censored cloud, censored cloud. Who doesn't have a cloud today? Every time you turn around, somebody's got a cloud, please send me all your data. We do that, and we do that well. We have Helion, we have the Microsoft Azure IoT cloud, we do that well. But my point is, there's a world out there. And it can be as high as 40 to 50 percent of the market, IDC is quoted as suggesting 40 percent of the data collected at the edge, by for example National Instruments, will be processed at the edge. Not sent, necessarily back to the data center or cloud, okay. With that background, there are seven reasons to not send all the data, back to the cloud. That doesn't mean you can't or you shouldn't, it just means you don't have to. There are seven reasons to compute at the edge. With an Edgeline system. Ready? >> Dave: Ready. >> We're going to go fast. And there'll be a test on this, so. >> I'm writing it down. >> Number one is latency, Eric already talked about that. How fast do you want your turnaround time? How fast would you like to know your asset's going to catch on fire? How fast would you like to know when the future autonomous car, that there's a little girl playing in the road, as opposed to a plastic bag being blown against the road, and are you going to rely on the latency of going all the way to the cloud and back, which by the way may be dropped, it's not only slow, but you ever try to make a phone call recently, and it not work, right? So you get that point. So that's latency one. You need to time to incite, time to response. Number one of seven, I'll go real quick. Number two of seven is bandwidth. If you're going to send all this big analog data, the oldest, the fastest, and the biggest of all big data, all back, you need tremendous bandwidth. And sometimes it doesn't exist, or, as some of our mutual customers tell us, it exists but I don't want to use it all for edge data coming back. That's two of seven. Three of seven is cost. If you're going to use the bandwidth, you've got to pay for it. Even if you have money to pay for it, you might not want to, so again that's three, let's go to four. (coughs) Excuse me. Number four of seven is threats. If you're going to send all the data across sites, you have threats. It doesn't mean we can't handle the threats, in fact we have the best security in the industry, with our Aruba security, ClearPass, we have ArcSight, we have Volt. We have several things. But the point is, again, it just exposes it to more threats. I've had customers say, we don't want it exposed. Anyway, that's four. Let's move on to five, is duplication. If you're going to collect all the data, and then send it all back, you're going to duplicate at the edge, you're going to duplicate not all things, but some things, both. All right, so duplication. And here we're coming up to number six. Number six is corruption. Not hostile corruption, but just package dropped. Data gets corrupt. The longer you have it in motion, e.g. back to the cloud, right, the longer it is as well. So you have corruption, you can avoid. And number three, I'm sorry, number seven, here we go with number seven. Not to send all the data back, is what we call policies and compliance, geo-fencing, I've had a customer say, I am not allowed to send all the data to these data centers or to my data scientists, because I can't leave country borders. I can't go over the ocean, as well. Now again, all these seven, create a market for us, so we can solve these seven, or at least significantly ameliorate the issues by computing at the edge with the Edgeline systems. >> Great. Eric, I want to get your final thoughts here, and as we wind down the segment. You're from the ops side, ops technologies, this is your world, it's not new to you, this edge stuff, it's been there, been there, done that, it is IoT for you, right? So you've seen the evolution of your industry. For the folks that are in IT, that HP is going to be approaching with this new category, and this new shift left, what does it mean? Share your color behind, and reasoning and reality check, on the viability. >> Sure. >> And relevance. >> Yeah, I think that there are some significant things that are driving this change. The rise of software capability, connecting these previously siloed, unconnected assets to the rest of the world, is a fundamental shift. And the cost point of acquisition technology has come down the point where we literally have a better, more compelling economic case to be made, for the online monitoring of more and more machine-type data. That example I gave of Duke Energy? Ten years ago they evaluated online monitoring, and it wasn't economical, to implement that type of a system. Today it is, and it's actually very, very compelling to their business, in terms of scheduled downtime, maintenance cost, it's a compelling value proposition. And the final one is as we deliver more analytics capability to the edge, I believe that's going to create opportunity that we don't even really, completely envision yet. And this deep computing, that the Edgeline systems have, is going to enable us to do an analysis at the edge, that we've previously never done. And I think that's going to create whole new opportunities. >> So based on your expert opinion, talk to the IT guys watching, viability, and ability to do this, what's the... Because some people are a little nervous, will the parachute open? I mean, it's a huge endeavor for an IT company to instrument the edge of their business, it's the cutting, bleeding edge, literally. What's the viability, the outcome, is it possible? >> It's here now. It is here now, I mean this announcement kind of codifies it in a new product category, but it's here now, and it's inevitable. >> Final word, your thoughts. >> Tom: I agree. >> Proud papa, you're like a proud papa now, you got your baby out there. >> It's great. But the more I tell you how wonderful the EL1000, EL4000 is, it's like my mother calling me handsome. Therefore I want to point the audience to Flowserve. F-L-O-W, S-E-R-V-E. They're one of our customers using Edgeline, and National Instruments equipment, so you can find that video online as well. They'll tell us about really the value here, and it's really powerful to hear from a customer. >> John: And availability is... >> Right now we have EL1000s and EL4000s in the hands of our customers, doing evaluations, at the end of the summer... >> John: Pre-announcement, not general availability. >> Right, general availability is not yet, but we'll have that at the end of the summer, and we can do limited availability as we call it, depending on the demand, and how we roll it out, so. >> How big the customer base is, in relevance to the... Now, is this the old boon shot box, just a quick final question. >> Tom: It is not, no. >> Really? >> We are leveraging some high-performance, low-power technology, that Intel has just announced, I'd like to shout out to that partner. They just announced and launched... Diane Bryant did her keynote to launch the new xeon, E3, low-power high-performance xeon, and it was streamed, her keynote, on the Edgeline compute engine. That's actually going into the Edgeline, that compute blade is going into the Edgeline. She streamed with it, we're pretty excited about that as well. >> Tom and Eric, thanks so much for sharing the big news, and of course congratulations, new category. >> Thank you. >> Let's see how this plays out, we'll be watching, got to get the draft picks in for this new sports league, we're calling it, like IoT, the edge, of course we're theCUBE, we're living at the edge, all the time, we're at the edge of HPE Discovery. Have one more day tomorrow, but again, three days of coverage. You're watching theCUBE, I'm John Furrier with Dave Vellante, we'll be right back. (electronic music)

>> Announcer: Live from Madrid, Spain, it's The Cube covering HPE Discover Madrid 2017. Brought to you by: Hewlett Packard Enterprise. >> Welcome back to HPE Discover, 2017 in Madrid. This is The Cube, the leader in live tech coverage, my name is Dave Vellante, I'm here to rap with my co-host, Peter Burris. >> Hey, Dave. >> Dave: Good couple a days. >> Oh, you know what I just discovered. I discovered The Cube is the antidote to jet lag. (laughs) >> That's right, when you get interesting people on. >> Oh, man. >> It pumps you up. >> Totally. Just unbelievable, exciting and it's all framed by... Well let's start where we talked about yesterday, we proposed that increasing what we're seeing in the industry is the new model of computing being established by Amazon and then the other poll, where it was known, we know that it's not all gonna be one cloud, it's not all gonna be a central cloud model, or essentialize cloud model. There's gonna be other places where data's gonna need to be processed. >> Dave: Well, that's what we believe. >> That's what we believe, and... There's physics behind that statement. There's legal regulations about data residency, behind that statement. But, we didn't know who was gonna step up and lead that other side and it's nice to see this conference indicate that HPE is in a position to help demonstrate, or help show the industry how cloud truly can go from centralized down to the edge. >> Yeah, and I think as I said a number of times, the strategy's coming into focus, you could debate it. You could say, "well, splitting it up was the wrong thing to do. "They lost their supply chain." But, Meg's argument, and then Antonio's argument always was, "look, we're gonna be more focused, "it's gonna allow us to do "a better job for our customers. "Yes, we're gonna be service's lead." They didn't say this. "Our margines are gonna be lower, "you don't have software anymore, "but that's okay, we can learn how to make money at that." And you know, the old HPE went through a similar transition. Kinda, got out of the HPEX business and got out of building it's own OS, and relying more on Microsoft and Intel and it made a lot of money. In those days. >> Peter: It did well. >> Did very well. It didn't invest under the herd regime the way it could have or should have and that hurt and then it spun out and made a lot of missteps but... Meg, to her credit, didn't make a lot of missteps. There was the initial entrance into the public cloud, they pulled back fast, they failed fast on that, good. Yeah, maybe there was some organizational issues early on but in general, the acquisitions have been solid, the strategy... >> And well integrated. >> And well integrated, absolutely. >> Peter: They've gotten value out of 'em. >> The strategies has been... I think clear internally, it wasn't always clear externally but they stayed calm about that, they didn't freak out about that. Helped that the stock price was going up a little bit, 'cause it was pretty depressed for a while. >> And shareholders weren't incontestable like they were for many years. >> That's right, and so, that gave them a little bit of time to bring it all together... It's finally here and I think Meg is stepping down at absolutely the right time. >> Or at a... She's stepping down at a good time, she's leaving a company that is much stronger than it was when she took it over. >> And that's what you want, one of the things I'm personally proud of when I left IDC it was in really good shape when I left, it wasn't a mess that I handed to somebody else. Had a lot of messes and IDC that I turned around as you well know. So, I think, I feel as though the company's in good shape and good hands. And, again, I think the... I don't know if you're a stock analyst or if you're pounding the table saying "buy this stock." 'Cause it is a relatively low margin business and there's a lot of competition, there's knife fights out there, it's not a high growth business, but on the flip side, it's clean, it throws off a lot of cash, they got a decent balance sheet and the customers love 'em. >> And that's the most important thing, it's the customers. Look, I... Disclosure, I actually did a significant consulting stint, here at HPE, right around the time of the compact acquisition and I saw what happened and for many years, the senior manager and team of HPE behaved as though they presumed that scale was it's own reward. If we get bigger, we'll find efficiencies, we'll find opportunities. Just being big, is the objective and I think that they have wandered in the desert trying to find those opportunities, that were the consequence of just being big and they never materialized. >> They weren't there. >> They never... It was like mirages on the horizon, they never materialized and I think if there's anything to your point that Meg has successfully done, is she's gotten the company to say, "don't chase the mirages, chase the customer. "Let's come back to what made HPE great for so long." And the idea that, if we stay focused on the customer and focus on technology, we can put them together in unique and interesting ways that will bind us to what customers are doing. And if you take a look at this event and the new messaging, and the things that they're focusing on it feels like, to me, that HPE is no longer wandering in the desert. You and I are smart guys, we are... Typically we can look at a company and we can see whether or not they know what they're doing and when you said, "well, you know what. "Maybe they had it all figured out inside, "and the rest of us couldn't see it." No, that's not the case. It was not figured out inside and that's what we saw but under Meg, it has become increasingly more figured out and the consequence of that... And it's been very, very plan full. She first was figured out and then she told Wall Street and Wall Street was happy with the numbers, and then she figured out and she started talking to customers when customers were there and now she's figuring it out, she's telling a broader market place. >> Well, and when she stopped by- >> And Antonio's got a great big story to tell. >> And both of those guys stopped by to see us. Meg spent 10 minutes with us, we were chatting here on the open mics and she was very good. Meg, one on one situation, in a small crowd is phenomenal. I've always said that about Meg. Not the greatest presence on stage, not a super dynamic speaker, she's not a Steve Jobs, obviously nobody is, but... But, man, is she credible in a one on one situation. One of the things she said to us was, "Y'know, we kinda got lucky..." My words, "with Aruba, we bought him "because we thought we could compete with "Cisco better, we bought him obviously "because it was a great business, a growth business," and boom all of a sudden, this intelligent edge thing hit. You sprinkle in a little Dr. Tom Bradicich and boom, off you go and you've got not only a great business, you got something that is becoming increasingly strategic for organizations. Great example, I mean the nimble acquisition. We heard, yesterday, Bill Philbin talking about, "well, when we got nimble-" was it Bill Philbin, no it was somebody else today it was... Alain Andreoli. He said, "we picked up nimble 'cause it was a great "flash company, but then we saw this inside thing, "we said, wow, we can spread this thing "across our entire portfolio." That's where- >> And the example he gave was: in six months, it's not running on... >> On three par and then it's gonna run... His goal, he says, "I'm not committing to this, "but my goal is by the end of the next year "it's gonna be running across the entire "server and storage and networking line." That would be a major accomplishment. If in fact, we'll see how much of this stuff is actually impactful to the business, how much it can actually save money you know, anticipate failures, I don't know. We'll see, it's AI, it's a perfect application. You guys have written a lot on the Wikibon team about AI for ITOM. >> Oh yeah, look... >> Dave: And this is a good example. >> I'm not the kinda guy, as you know, that gets all excited about technology for technology's sake. I like thinking about technology and how it's gonna be applied, more problems are gonna be solved and so as we, in Wikibon, started running around and getting all excited about AI, my challenge to the guys was: Well, show me the two concrete cases, where it's gonna have a material business impact and one of the most important cases is, it's got a material business impact and how IT runs itself because you cannot... IT cannot reduce the number of people it's got and take on these increasingly complex application, problems, and portfolios unless they get a lot of help and the best, most likely source of that help is by bringing a lot of these new AI technologies that are capable of taking concrete, real time action in response to what's happening within the infrastructure and the applications at any given time. >> Yeah, now... Couple other things, just observations. Ana Pinczuk came on, great leader, woman in tech, big proponent of advancing women's causes, especially in tech. She had mixed feelings about Meg stepping down, obviously you have a woman leader, I thought her comments there were... Were quite interesting, but she said, "But I am up for the challenge "to continue the mission." Which leads me to Antonio. Antonio is outwardly a humble guy, he may have a big ego I don't know, he's been on The Cube a number of times, but he certainly doesn't come across as a guy who's looking to get credit. He's a quiet but very competent leader, he knows the business very well. Really interesting to see what his relationship- >> Peter: Homegrown. >> Homegrown, which is 22 years at HPE, technology background, not a U.S... Born individual, now living in the U.S. obviously. But, somebody with international experience which is always been an attribute that's valued at HPE. Gonna be interesting to see what his relationship is with Wall Street. Will he be sort of a quiet leader that lets the CFO take front and center, which would be fine. Or will he slowly sort of advance, he's not been sitting on the earnings calls. I'm interested to see how he handles it, or he may just say, "you know what, "I'm gonna go execute in the business "and let the results speak for themselves." So, I'm kind of curious as to how that all... All plays out. It's a big job, it's a big role as you pointed out with me the other day. Big role for him, big job for him. Serious opportunities to make a mark in the industry. >> Again, and you raise a really great point. Meg had a very good reputation on Wall Street, the knock on her when she came on, was she didn't know customers. Antonio's got a great reputation with customers, you're asking the question: is he gonna get to know Wall Street? A great CEO has to be able to take care of customers and owners He seems very... Look, this is a, this whole simplification of how they're gonna bring cloud technologies to where their data's gonna require is apparently, based on what we heard, in large part Antonio's brain child. He conceived it, he invested in it, he nurtured it, he took risks for it, he put some skin in the game and now it's coming to fruition, that's great, and he's got customers lining up behind it. We'll see, this is another place where we'll see, but I don't think that there's... There's no reason to suspect, just looking at Antonio's track record, why Wall Street would abandon him. On the contrary, there's reasons to suspect that he will also be able to develop that set of skills that Wall Street needs to do their job. But, clearly this is a guy that's gonna turn on a lot of customers. >> Yeah, and as I say, it's gonna be interesting to see what his relationship, like look at a guy like Frank Slootman, who had a great relationship with Wall Street, everybody loved him 'cause he just performed but he's a hard-driving, in your face kinda guy, who developed close relationships with the street. It's gonna be, as they say, I gotta watch that, to see how Antonio interacts with them. I think it's important to have a relationship with... >> Peter: With your ownership, yeah it usually is. >> And I think that's the one big question mark here is, where has his presence been there but so we'll watch and I'm confident he'll step up to that. Okay. Let's see, The Cube... Next week? Cube-con? >> Peter: Yeah. >> Next week in Austin. Right, so development. You'll see The Cube expanding way beyond it's original infrastructure route, so obviously HPE Discover, big infrastructure show. But we're at Amazon Reinvent this week, it's our big cloud show. We obviously... All the IBM shows are being consolidated into one show called Think. This year The Cube will be there. But CES is gonna be January, we were there last year, likely be there again. Cisco live is on the radar, we're gonna be at Cisco live I think both in Barcelona and most likely in the states this year, so that's another big thing. A lot of developer shows, Docckercon, Kubecon, working with the Linux Foundation, developers are really the lynch pin, developers in cloud. Really big areas of growth. IOT, some IOT conferences that we're gonna be doin' this year. Obviously, our big data heritage we still do a lot of work there, so. It's been an unbelievable year, I think a 125 shows for The Cube. TheCube.net, new website, our new clipper tool, you see the clips that come out, so. A lot of innovation comin' out of Siliconangle Media, check out Siliconangle.com. Peter, the work that your team is doing on the Wikibon side, Wikibon.com. Unbelievable amounts of research that you guys are crackin' out. Digital business, AI, AI for ITOM stuff that we talked about, we still do some stuff in infrastructure, true private cloud. >> New computing architectures, memory based computer architectures. >> So, fantastic work there and... Yeah, so we're looking forward to another great year. Thanks everybody for these last two days, thanks to the crew, great job. Everybody at home. We're out. Dave Vellante for Peter Buriss from Madrid. Thanks for watching. (upbeat music)

>> (Narrator) Live from Madrid, Spain. It's The Cube covering HP Discover Madrid 2017, brought to you by Hewlett Packard Enterprise. >> We're back at HPE Discover Madrid 2017. This is The Cube, the leader in live tech coverage. My name is Dave Vellante and I'm with my co-host for the week, Peter Burris. Cat Graves is here, she's a research scientist at Hewlett Packard Enterprises. And she's joined by Natalia Vassilieva. Cube alum, senior research manager at HPE. Both with the labs in Palo Alto. Thanks so much for coming on The Cube. >> Thank you for having us. >> You're welcome. So for decades this industry has marched to the cadence of Moore's Law, bowed down to Moore's Law, been subservient to Moore's Law. But that's changing, isn't it? >> Absolutely. >> What's going on? >> I can tell Moore's Law is changing. So we can't increase the number, of course, on the same chip and have the same space. We can't increase the density of the computer today. And from the software perspective, we need to analyze more and more data. We are now marching calls into the area of artificial intelligence when we need to train larger and larger models, we need more and more compute for that. And the only possible way today to speed up the training of those modules, to actually enable the AI, is to scale out. Because we can't put more cores on the chip. So we try to use more chips together But then communication bottlenecks come in. So we can't efficiently use all of those chips. So for us on the software side, on the part of people who works how to speed up the training, how to speed up the implementation of the algorithms, and the work of those algorithms, that's a problem. And that's where Cat can help us because she's working on a new hardware which will overcome those troubles. >> Yeah, so in our lab what we do is try and think of new ways of doing computation but also doing the computations that really matter. You know, what are the bottlenecks for the applications that Natalia is working on that are really preventing the performance from accelerating? Again exponentially like Moore's Law, right? We'd like to return to Moore's Law where we're in that sort of exponential growth in terms of what compute is really capable of. And so what we're doing in labs is leveraging novel devices so, you've heard of memristor in the past probably. But instead of using memristor for computer memory, non volatile memory for persistent memory driven computer systems, we're using these devices instead for doing computation itself in the analog domain. So one of our first target applications, and target core computations that we're going after is matrix multiplication. And that is a fundamental mathematical building block for a lot of different machine learning, deep learning, signal processing, you kind of name it, it's pretty broad in terms of where it's used today. >> So Dr. Tom Bradicich was talking about the dot product, and it sounds like it's related. Matrix multiplications, suddenly I start breaking out in hives but is that kind of related? >> That's exactly what it is. So, if you remember your linear algebra in college, a dot product is exactly a matrix multiplication. It's the dot in between the vector and the matrix. The two itself, so exactly right. Our hardware prototype is called the dot product engine. It's just cranking out those matrix multiplications. >> And can you explain how that addresses the problem that we're trying to solve with respect to Moore's Law? >> Yeah, let me. You mentioned the problem with Moore's Law. From me as a software person, the end of Moore's Law is a bad thing because I can't increase their compute power anymore on the single chip. But for Cat it's a good thing because it forced her to think what's unconventional. >> (Cat) It's an opportunity. >> It's an opportunity! >> It forced her to think, what are unconventional devices which she can come up with? And we also have to mention they understand that general purpose computing is not always a solution. Sometimes if you want to speed up the thing, you need to come up with a device which is designed specifically for the type of computation which you care about. And for machine learning technification, again as I've mentioned, these matrix-matrix multiplications matrix-vector multiplications, these are the core of it. Today if you want to do those AI type applications, you spend roughly 90% of the time doing exactly that computation. So if we can come up with a more power efficient and a more effective way of doing that, that will really help us, and that's what dot product engine is solving. >> Yes, an example some of our colleagues did in architectural work. Sort of taking the dot product engine as the core, and then saying, okay if I designed a computer architecture specifically for doing convolutional neural networks. So image classification, these kinds of applications. If I built this architecture, how would it perform? And how would it compare to GPUs? And we're seeing 10 to 100 X speed up over GPUs. And even 15 X speed up over if you had a custom-built, state of the art specialized digital Asic. Even comparing to the best that we can do today, we are seeing this potential for a huge amount of speed up and also energy savings as well. >> So follow up on that, if I may. So you're saying these alternative processors like GPUs, FGPAs, custom Asics, can I infer from that that that is a stop-gap architecturally, in your mind? Because you're seeing these alternative processors pop up all over the place. >> (Cat) Yes. >> Is that a fair assertion? >> I think that recent trends are obviously favoring a return to specialized hardware. >> (Dave) Yeah, for sure. Just look at INVIDIA, it's exploding. >> I think it really depends on the application and you have to look at what the requirements are. Especially in terms of where there's a lot of power limitations, right, GPUs have become a little bit tricky. So there's a lot of interest in the automotive industry, space, robotics, for more low power but still very high performance, highly efficient computation. >> Many years ago when I was actually thinking about doing computer science and realized pretty quickly that I didn't have the brain power to get there. But I remember thinking in terms of there's three ways of improving performance. You can do it architecturally, what do you do with an instruction? You can do it organizationally, how do you fit the various elements together? You can do it with technology, which is what's the clock speed, what's the underlying substrate? Moore's Law is focused on the technology. Risk, for example, focused on architecture. FPGAs, arm processors, GPUs focus on architecture. What we're talking about to get back to that doubling the performance every 18 months from a computing standpoint not just a chip standpoint, now we're talking about revealing and liberating, I presume, some of the organization elements. Ways of thinking about how to put these things together. So even if we can't get improvements that we've gotten out of technology, we can start getting more performance out of new architectures. But organizing how everything works together. And make it so that the software doesn't have to know, or the developer, doesn't have to know everything about the organization. Am I kind of getting there with this? >> Yes, I think you are right. And if we are talking about some of the architectural challenges of today's processors, not only we can't increase the power of a single device today, but even if we increase the power of a single device, then the challenge would be how do you bring the data fast enough to that device? So we will have problems with feeding that device. And again, what dot product engine does, it does computations in memory, inside. So you limit the number of data transfers between different chips and you don't face the problem of feeding their computation thing. >> So similar same technology, different architecture, and using a new organization to take advantage of that architecture. The dot product engine being kind of that combination. >> I would say that even technology is different. >> Yeah, my view of it we're actually thinking about it holistically. We have in labs software working with architects. >> I mean it's not just a clock speed issue. >> It's not just a clock speed issue. It's thinking about what computations actually matter, which ones you're actually doing, and how to perform them in different ways. And so one of the great things as well with the dot product engine and these kind of new computation accelerators, is with something like the memory driven computing architecture. We have now an ecosystem that is really favoring accelerators and encouraging the development of these specialized hardware pieces that can kind of slot in in the same architecture that can scale also in size. >> And you invoke that resource in an automated way, presumably. >> Yeah, exactly. >> What's the secret sauce behind that? Is that software that does that or an algorithm that chooses the algorithm? >> A gen z. >> A gen z's underlying protocol is to make the device talk to the data. But at the end of the system software, it's algorithms also which will make a decision at every particular point which compute device I should use to do a particular task. With memory driven computing, if all my data sits in the shared pool of memory and I have different heterogeneous compute devices, being able to see that data and to talk to that data, then it's up to the system management software to allocate the execution of a particular task to the device which does that the best. In a more power efficient way, in the fastest way, and everybody wins. >> So as a software person, you now with memory driven computing have been thinking about developing software in a completely different way. Is that correct? >> (Natalia) Yeah. You're not thinking about going through I/O stack anymore and waiting for a mechanical device and doing other things? >> It's not only the I/O stack. >> As I mentioned today, the only possibility for us to decrease the time of processing for the algorithms is to scale out. That means that I need to take into account the locality of the data. It's not only when you distribute the computation across multiple nodes, even if we have some number based which is we have different sockets in a single system. With local memory and the memory which is remote to that socket but which is local to another socket. Today as a software programmer, as a developer, I need to take into account where my data sits. Because I know in order to accept the data on a local memory it'll take me 100 seconds to accept my data. In the remote socket, it will take me longer. So when I developed the algorithm in order to prevent my computational course to stall and to wait for the data, I need to schedule that very carefully. With memory driven computing, giving an assumption that, again, all memory not only in the single pool, but it's also evenly accessible from every compute device. I don't need to care about that anymore. And you can't even imagine such a relief it is! (laughs) It makes our life so much easier. >> Yeah, because you're spending a lot of time previously trying to optimize your code >> Yes for that factor of the locality of the data. How much of your time was spent doing that menial task? >> Years! In the beginning of Moore's Law and the beginning of the traditional architectures, if you turn to the HPC applications, every HPC application device today needs to take care of data locality. >> And you hear about when a new GPU comes out or even just a slightly new generation. They have to take months to even redesign their algorithm to tune it to that specific hardware, right? And that's the same company, maybe even the same product sort of path lined. But just because that architecture has slightly changed changes exactly what Natalia is talking about. >> I'm interested in switching subjects here. I'd love to spend a minute on women in tech. How you guys got into this role. You're both obviously strong in math, computer backgrounds. But give us a little flavor of your background, Cat, and then, Natalia, you as well. >> Me or you? >> You start. >> Hm, I don't know. I was always interested in a lot of different things. I kind of wanted to study and do everything. And I got to the point in college where physics was something that still fascinated me. I felt like I didn't know nearly enough. I felt like there was still so much to learn and it was constantly challenging me. So I decided to pursue my Ph.D in that, and it's never boring, and you're always learning something new. Yeah, I don't know. >> Okay, and that led to a career in technology development. >> Yeah, and I actually did my Ph.D in kind of something that was pretty different. But towards the end of it, decided I really enjoyed research and was just always inspired by it. But I wanted to do that research on projects that I felt like might have more of an impact. And particularly an impact in my lifetime. My Ph.D work was kind of something that I knew would never actually be implemented in, maybe a couple hundred years or something we might get to that point. So there's not too many places, at least in my field in hardware, where you can be doing what feels like very cutting edge research, but be doing it in a place where you can see your ideas and your work be implemented. That's something that led me to labs. >> And Natalia, what's your passion? How did you arrive here? >> As a kid I always liked different math puzzles. I was into math and pretty soon it became obvious that I like solving those math problems much more than writing about anything. I think in middle school there was the first class on programming, I went right into that. And then the teacher told me that I should probably go to a specialized school and that led me to physics and mathematics lyceum and then mathematical department at the university so it was pretty straightforward for me since then. >> You're both obviously very comfortable in this role, extremely knowledgeable. You seem like great leaders. Why do you feel that more women don't pursue a career in technology. Do you have these discussions amongst yourselves? Is this something that you even think about? >> I think it starts very early. For me, both my parents are scientists, and so always had books around the house. Always was encouraged to think and pursue that path, and be curious. I think its something that happens at a very young age. And various academic institutions have done studies and shown when they do certain things, its surmountable. Carnegie Mellon has a very nice program for this, where they went for the percentage of women in their CS program went from 10% to 40% in five years. And there were a couple of strategies that they implemented. I'm not gonna get all of them, but one was peer to peer mentoring, when the freshmen came in, pairing them with a senior, feeling like you're not the only one doing what you're doing, or interested in what you're doing. It's like anything human, you want to feel like you belong and can relate to your group. So I think, yeah. (laughs) >> Let's have a last word. >> On that topic? >> Yeah sure, or any topic. But yes, I'm very interested in this topic because less than 20% of the tech business is women. Its 50W% of the population. >> I think for me its not the percentage which matters Just don't stay in the way of those who's interested in that. And give equal opportunities to everybody. And yes, the environment from the very childhood should be the proper one. >> Do you feel like the industry gives women equal opportunity? >> For me, my feeling would be yes. You also need to understand >> Because of your experience Because of my experience, but I also originally came from Russia, was born in St. Petersburg, and I do believe that ex-Soviet Union countries has much better history in that. Because the Soviet Union, we don't have man and woman. We have comrades. And after the Second World War, there was women who took all hard jobs. And we used to get moms at work. All moms of all my peers have been working. My mom was an engineer, my dad is an engineer. From that, there is no perception that the woman should stay at home, or the woman is taking care of kids. There is less of that. >> Interesting. So for me, yes. Now I think that industry going that direction. And that's right. >> Instructive, great. Well, listen, thanks very much for coming on the Cube. >> Sure. >> Sharing the stories, and good luck in lab, wherever you may end up. >> Thank you. >> Good to see you. >> Thank you very much. >> Alright, keep it right there everybody. We'll be back with our next guest, Dave Vallante for Peter Buress. We're live from Madrid, 2017, HPE Discover. This is the Cube.

>> Announcer: Live from Madrid, Spain, it's theCUBE, covering HPE Discover Madrid 2017, brought to you by Hewlett Packard Enterprise. >> Welcome back to Madrid, everybody, this is theCUBE, the leader in live tech coverage. My name is Dave Vellante, I'm here with Peter Burris, this is day one of HPE Discover Madrid. Olivier Frank is here, he's the Worldwide Senior Sales Director for Alliances for IoT at HPE, and Kurt Bayer, otherwise known as Bager in English, in America. He's Vice President of IoT Solutions for EMEA PTC, did I get that right? >> Yeah you did it. >> Bayer? All right, well thank you for sharing that with me. Welcome to theCUBE, gentlemen. Olivier, let me start with you. The relationship between PTC and HPE is not brand new. You guys got together a while back. What catalyzed that getting together? >> Yeah, it's a great question, and thank you for inviting us, it's great pleasure to be on theCUBE, and for me the first time, so thank you for that. >> Welcome. >> Yeah, you know, the partnership is all about action and doing things together, so we did start about a year ago with, you may remember flow serve and industrial pump that we showcased, and since then we've been working very closely together to actually allow our customers to go an test the technology themselves. So I would say the partnership has matured, we now have two live environments that customer can visit, one in Europe, in Germany, in Aachen, with the RWTH University, and one in the US, near Houston, with Texmark who you know because you also came to the show. >> Right, okay, Kurt give us the update on PTC. Company's been in business for a long time, IoT is like a tailwind. >> It is, that's right. PTC is mostly known for CAD and PLM, so for 30 years they made 3D CAD software for when you design and make an aircraft or car engine. But over the last five years, PTC have moved heavily into IoT, spent a billion on acquiring and designing software platform that can connect and calculate and show in augmented reality. >> So let me build on that, because PTC as a CAD company, as a PLM company, has done a phenomenal job of using software and technology to be able to design things to a level of specificity and tolerance that just wasn't able to be done before, and it's revolutionized how people build products. But now, because technology's advanced, you can leverage that information in your drawings, in your systems to create a new kind of an artifact, a digital twin that allows a business that's working closely with you to actually render that in an IoT sense and add intelligence to it. Have I got that right? >> You got it exactly right. So making the copy. We can draw it and we can design the physical part, and we can make the digital twin of the physical part with sensors. So in that way you can loop back and see if the calculation, the design, the engineering you have made is the right fit, or you need to change things. You can optimize product with having the live digital twin of the things that you've designed physically. >> So it's like a model, except it's not a model. It's like a real world instantiation. Model is an estimate, right? A digital twin is actual real data. >> It's feeded by live data, so you have a real copy of what's going on. And we use it for not only closing the loop of designing products, but also to optimize in the industrial fold, to optimize operation and creating manufacturing of things, and we use it to connect things, so you can do predictive maintenance or you can turn products to be a service, instead of selling an asset, the company can buy by click, by use, plus the product are connected. >> I want to really amplify this, Dave, 'cause it's really important, I want to test this with you, 'cause the whole concept of using technology, IoT technology to improve the operational efficiency, to improve the serviceability, to evolve your business models, your ability to do that is tied back to the fidelity of the models you're using for things that are delivering the services, and I don't think the world fully understands the degree to which it's a natural leap from CAD and related technologies, into building the digital artifacts that are gonna be necessary to make that all work. Have I got that right? >> You got it completely right. So it is moving from having live informations from the physical object. So if you go to augmented reality, so you have the opportunity to look at things and get live information about temperature, power, streaming of water, and all these things that goes on inside the product, you also have the opportunity to understand if there's something wrong with the product, you can click on it and you can be directed on how to change and service things like when the augmented reality, all built by the CAD drawing in the beginning that is combined with sensor information and >> And simulate, and test, and all the other things that are hard, but obviously to do that, you need a whole bunch of other technology, and I guess that's where HPE comes in. >> Exactly. >> Absolutely. In fact to bounce on that thought, we talk a lot about connected operation, where you know, we are showing the digital twin, but one of the new use case that we're showing on the floor here is what we call smart product engineering. So we're basically using the CAD environment of (mumbles), running on that edge line with edge compute, you know, enterprise compute capability, manageability and security, and running on that same platform then, simulation from companies like Ensys, right, and then doing 3D printing, print prototyping, and basically instrumenting the prototype, we're using a bike, the saddle stem of a bike showcase, and they are able to connect and collect the data, we're partnering with National Instruments who are also well-known, and reinject the real data into the digital model. So again, the engineers can compare their thought and their design assumptions with the real physical prototype, and accelerate time to market. >> PTC's been a leader in starting with the CAD and then pulling it through product life cycle management, PLM. So talk about this is going to alter the way PLM becomes a design tool for digital business. If I'm right. >> You're right, it becomes industrial innovation platform from creating the product to the full life cycle of it. >> Peter: All the way up to the business model. >> All the way up to the business model. And talking about analytics, so if you have a lot of data and you want to make sure you get some decision made fast about predictive maintenance, that's an area where we are partnering with HP so we have a lot of power close in the edge, close to the products that can do the calculations from the devices, from the product, and do some fast results in order to do predictive maintenance and only send the results away from the location. >> So what are some of the things you guys are most excited about, Olivier? >> Well, really excited about making those use cases, being the smart product engineering, or the predictive maintenance, you know, work for our customers so behind the scenes we have great solutions, now we're partnering on the sales front to kind of go together to customers, we have huge install base on both sides, and picking the right customers interested in this digital transformation, and make it real for them, because we know it's a journey, we know it's kind of the crawl, walk, run, and it's really about accelerating, you know, turning insights into information and into actions, and that's really where we are very much excited to work together. >> So it's not just, so the collaboration's extending to go to market is what I'm hearing. And so what's the uptake been like, what are customers, customers must be asking you, "Where do I start?" What do you tell them? >> Before you start, it's important that you have a business case, a business value, you understand what you wanted to achieve, by integrating an IoT solution. That's important. Then you need to figure out what is the data, what is the fast solution I need to take, and then you can start deciding on the planning of your implementation of the IoT. >> Can I go back one step further, >> Yep. >> You tell me if I got that. And that one step further is, look, every... Innovation and adoption happens faster when you can take an existing asset and create new value. >> Kurt: Exactly. >> So isn't PTC actually starting by saying, hey, you've already got these designs, you've already got these models. Reuse them, create new life, give 'em new life, create new value with 'em. Do things in ways that now you can work with your customers totally differently, and isn't that kind of where it starts? >> It does, and you already have a good portion of what you need, so in order to make a fast value out of your new product or the new thing you can do with the product, connecting the products, then PTC and HP is a good platform to move on. >> Yeah but the pretesting, precertify, packaging, the software with the hardware, is allowing our customer to go faster to proof of concept and then to production. So we have a number of workshops, customers can come, again as I mentioned at the beginning, in Germany, in Aachen or in Houston at our Texmark facility, where we can basically walk the talk with customers and start those early POCs, defining the business success factors, business value they want to take out of it, and basically get the ball rolling. But it's really exciting because we have, we're touching really some of the key digital transformation of our enterprise customers. >> And don't forget that you need a partner that can do a good job in service, because you need a organization that can help you get it through, and HP are a strong service organization too. >> Well this idea of the intelligent edge has a lot of obviously executive support at Hewlett Packard Enterprise, that keeps buzzing at theCUBE today, Meg Whitman's in the house, she's right next door, and we're gonna do a quick cutaway to Meg, give her a shoutout, trying to get her over here to talk about her six-year tenure here, but you know, that top-down executive support has been so critical in terms of HPE getting early into the edge, IoT, intelligent edge you call it, Tom Bradicich obviously a leader, he's coming on. You mentioned National Instruments, PTC, you guys were first, really, from a traditional IT business to really get into that space. >> We're also the first to converge OT and IT, so we're showing on the floor what we're doing in end of line quality testing for automotive for example, taking PX higher standard, which is like instrumentation and real-time data position into our converged systems. So what I found is really amazing. You take the same architecture, and we can do it edge to core to cloud, right, that's very powerful. One software framework, one IT architecture that's pan out. >> Peter: Not some time in the future, but right now. >> Yeah, right now. >> So we talk about a three, maybe even a 3A, four-tier data model, where you've got data at the edge, real time, maybe you don't persist all of it or a lot of it. >> We call it experience data or primary data at the edge. vet data, or secondary data, and then business optimization data at the top level, that's at the cloud. >> So let's unpack that a little bit and get your perspective. So the edge, obviously you're talking about real time decision making, autonomous cars, you're not gonna go back to the cloud to make that decision. That, well you call it core, that's what did you call it? >> The hybrid IT. >> The vet, the vet. That's an aggregation point, right, to collect a lot of the data from the edge, and then cloud maybe is where you do the deep analysis and do the deep modeling. And that cloud can be on-prem, or it can be on the public cloud. Is that a reasonable data model for the flow of data for edge and IoT? >> I believe it is, because some of these products generate a lot of data, and you need to be able to handle that data, and honestly, connectivity is not for free, and sometimes it's difficult if it's in the industry floor, manufacturing floor, you need good connectivity, but you still have limitations. So if you can do the local analytics and then you only send the results to the core, then it's a perfect model. And then there's a lot of regulations around data, so for many countries, and especially in Europe, there's boundaries around the data, it's not all that you can move to a cloud, especially if it's out of the country. So the model makes a good hybrid in between speed, connectivity, analytics and the legislation problem. >> Dave: And you've both got solutions at each layer? >> Absolutely, so in fact... So PTC can run at the edge, at the core or in the cloud, and of course we are powering the three pillars. And I think what's also interesting to know is that with the advance in artificial intelligence, as was explored during the main session, there it is pivotal you need to keep a lot of data in order to learn from those data, right? So I think it's quite fascinating that we're going to store more and more data, probably make some useful right away, and maybe store some that we come back to it. That's why we're working also with companies like OSIsoft, an historian, which is collecting this time stamp data for later utilization. But I wanted also to say that what's great working with PTC is that it's kind of a workflow in media, in terms of collecting the data, contextualizing them and then visualization and then analytics. But we're developing a rich ecosystem, because in this complex world of IoT, again it's kind of an art and a science, and the ability to partner ourselves, but also our let's say friendly partners is very, very critical. >> Dave: Guys, oh good, last word. >> I will say we started with a digital twin, and for some companies they might be late to get the digital twin. The longer you have had collecting data from a live product >> The better the model gets >> The stronger you will be, >> Peter: Better fidelity. >> The better model you can do, because you have the bigger data. So it's a matter of getting the data into the twin. >> That's exactly what our research suggests. We've got a lot of examples of this. >> It's the difference between sampling and having an entire corpus of data. >> Kurt: Exactly. >> Kurt, Olivier, thanks very much for coming on the theCUBE. >> Thank you. >> Thank you so much. >> Great segment guys. Okay, keep it right there everybody, Dave Vellante for Peter Burris, we'll be back in Madrid right after this short break.