(electronic music) >> Live from Toronto, Canada. It's the Cube. Covering Blockchain Futurist Conference 2018. Brought to you by the Cube. >> Hello, everyone. Welcome back to the Cube coverage here in Toronto, Canada. We're in Ontario to the Untraceable Blockchain Futurist Conference. This is day two of two days of coverage. I'm John Furrier, your host. Our next guest is serial entrepreneur, Chris Harper, the CEO of Jereki and Mapogo Ventures. Welcome. >> And ZippedScript. >> And ZippedScript. >> Yeah, yeah. >> Hey, lot of balls in the air, a lot of irons in the fire. Welcome to the Cube. >> Hey, good to have me, man. I'm excited to be here. This is awesome. >> So before we get into, you're a serial entrepreneur, what are you working on now? Take a quick minute, explain. >> Yeah. >> Where the names came from, what do they mean, what are you doing? >> So, first things first, we'll start with Jereki. Jereki is a Japanese proverb which means to achieve enlightenment through one's own efforts. So Jereki is a shell company and we currently run this company, it's called Chase your Drink. It basically is replacing pop and juice as a mix for any hard liquor. No sugar, no calories, nothing artificial and the kicker, we got 90 vitamins in these to combat your hangover the next day. So if you're not drinking with Chase, you're drinking wrong. (laughs) >> It's a chaser. >> Chaser, exactly. >> Yeah. >> And these are in stores like, it'll be in Sobeys, Farm Boy in Canada, GNC soon. And it's going really well. >> Okay, how 'about the venture firm? >> Yes, so the next company is Mapogo Ventures. Mapogo is actually, it comes from a group of lions in the African Savanna that were only six lions, but they dominated the savanna for their whole life span, which is super rare, and they took down animals like giraffes, rhinos, and became legend. It was a folk legend about these Mapogo lions. So Mapogo is a venture firm. We specialize in food and beverage companies. If you're doing something epic, we want to talk to you. And then we also specialize in blockchain cryptocurrency and anything that's on the forefront of what's going on in the tech space. So if anybody's interested, they think they have a great idea, you can reach out to us wherever you guys put contact information, I don't know. >> We'll put it up there. >> Yeah, yeah, definitely. >> What's the website? >> Mapogoventures.com >> Okay, got it. >> Yeah, Mapogoventures.com. >> And how much are you guys investing? What's the kind of round size you guys do? >> So it totally depends. We almost don't have a limit or a minimum. It's all about the team, the idea, where you're going, and what you need. We'll get you what you need. >> Is it a new firm, are you making business? >> It's a new firm, it's a new firm. So we have two companies that we're looking at right now, but we don't have any companies in the portfolio, we're looking to add. >> Great, awesome. >> Yeah, yeah. >> Well any great ideas, check it out. How about crypto? What's your seeing, what's your thing, what are you seeing on crypto? What kind of deals? Obviously the flight to quality right now is starting to see the ICOs kind of burning out here and there, but the ones that are solid are standing and growing in a build-out mode. >> I mean, the whole space right now, everybody's worried about it, right? If you're an outsider, you're looking at it like it's all down. But one thing I did want to say during this interview was this is a great event. Untraceable, they sent up an incredible event and even if you're not into cryptocurrency, if you're a business person, crypto's only been around for, you know, six, seven years. So everybody in this room did something before crypto. Right? So they're all multi-faceted individuals and if you're not in crypto, if you're scared of crypto, if you're hesitant about crypto, if you don't understand it, you should be here. You should be at these events because it's priceless networking and who knows where you can go. >> Plus, starting companies on a down, on down the bottom of the market-- >> Yeah. >> Is when the best companies get built. >> 1000%, you know. What did Warren Buffet say? Be fearful when others are greedy, be greedy when others are fearful. Looks to me that the market-- >> Yeah. >> Is incredibly fearful. So maybe you should consider being greedy right now. >> For the people that aren't here, what's the vibe of the show? What's your take, what's the hallway conversations like? >> Yeah, I mean, the vibe of the show. This is actually one of the best conferences I've been to. I've been to a few in New York. This one is incredible. Everyone's so friendly. You can come here, don't know anyone. >> Yeah. >> People will say hi to you. They'll introduce themselves to you. Next thing you know, you had an idea, now you have funding. But it's up to you to make this situation a great situation. >> What's interesting is this sector, blockchain and crypto. >> Yeah. >> Attracts alpha entrepreneurs, alpha engineers. >> Okay. >> You mentioned-- >> Mapogo. >> Smart people are in this world. They've done things before, so this is really interesting. >> Yeah, like people always forget that. They see crypto and they get nervous 'cause like I don't know anything about it. Remember guys, this is a new industry. And we're only in, you know, the first couple innings. This is going to be huge. So come, learn, and surround yourself with killers. >> Alright, what's the coolest thing you've seen so far here? >> The coolest thing I've seen so far. You know, I'm going to be completely honest with you. Larry King. I was so happy to see Larry King and it's awesome that a guy like that is supporting the community, you know. >> Yeah. >> Because this is really a revolutionary technology, the blockchain technology. >> You've done a lot entrepreneurial things since you were 10, you were talking before we came on. >> Yeah. >> How does that help you right now navigate this scene and looking at deals and your own deals and you're building out, you're investing. Other entrepreneurs are coming in, sometimes first time entrepreneurs, how does that help you and what advice would you give other entrepreneurs? >> So I started really young, not knowing where I was going to go. It was kind of just like in my blood. But, you know, you got to get out, you got to talk to people, you know. I always say no deal happens on your couch. You got to jump off the porch. You got to go out, you got to network, you got to meet people. And I started doing that at a young age which got my conversation skills a lot more advanced, so now I can go in and close a deal in 10 minutes where, you know, back in the day, it might take me two hours and I probably wouldn't even close it. So what I would say. >> 10 minutes is a good metric. >> It is. >> That's hey. >> Hey, I don't need to say more or less. If it's an interesting idea, let's go. You should be able to tell me what it is. >> Yeah. >> We should be able to hammer something out. Yeah, yeah, that's pretty much what's going on. >> Awesome. And what's some of the plans that you have for your ventures? Let's go back, the zip line, what's that one? >> Oh, yeah, ZippedScript. >> ZippedScript, I'm sorry. >> So I can't talk too much about ZippedScript. It's launching in fall of 2018. ZippedScript is basically going, it is revolutionizing the higher education industry and the transcript section in that industry. And all I can say is we may or may not be using blockchain technology to do it. >> Got it, okay. >> Yeah. >> And how about the chaser, that sounds very cool. >> Yeah, it is really cool. And, I mean, you guys can go to chase your drink.com, check it out. You can head over to our Instagram, Chase Your Drink. It's taken over. You know, this cola flavor I've got here and tropical thunder is pineapple mango, but cola tastes just like Coca-Cola. >> Yeah. >> Without any of the bad ingredients. And it's really taken over. You know, our biggest problem is supply. >> Yeah. >> We just can't produce enough, but we're fixing that problem. >> That's a good problem to have. >> It's a very good problem to have, right. >> How did you get into the venture side? Just you're scratching an itch, you wanted to put some of your money to work, did you raise unlimited partners, how's that, how'd that develop? >> Totally. >> And what's the current situation? >> Yeah, so it was a group of fellow entrepreneurs and we're all working on our own companies, but we're all ADD, right? And we're like I'm doing this, I'm doing that, but we have all these contacts, all these different skill sets, and we're all great friends. So that's another very important thing that most people talk about. Surround yourself with like minded people, but you want them to have different skill sets. >> Awesome. >> I don't want a clone. I have a clone, we're not going to work well together. >> You want added value, you don't want to subtract value. >> Yeah, exactly. So we came together and we're like we have so much value in so many different spaces, we can walk companies through, you know, a proven concept in any industry, food and beverage, cryptocurrency, and basically you won't make mistakes that we made. That's the bottom line. So you'll accelerate your success by working with us. >> Well, Chris, great to have you on. >> Yeah. >> Congratulations on your success. >> It was amazing, man. >> Check out Chase, check out the fund if you've got a great idea, contact Chris, go the cube.net, you can find his information there. I'm John Furrier here in Toronto with all the action here at the Blockchain Futurist Conference where the future's being created, robust industry, people looking at the long term, this is where the action is. Thanks for watching. Stay with us for day two coverage after this short break. (electronic music)

>>don't talk mhm, >>Okay, thing is my presentation on coherent nonlinear dynamics and combinatorial optimization. This is going to be a talk to introduce an approach we're taking to the analysis of the performance of coherent using machines. So let me start with a brief introduction to easing optimization. The easing model represents a set of interacting magnetic moments or spins the total energy given by the expression shown at the bottom left of this slide. Here, the signal variables are meditate binary values. The Matrix element J. I. J. Represents the interaction, strength and signed between any pair of spins. I. J and A Chive represents a possible local magnetic field acting on each thing. The easing ground state problem is to find an assignment of binary spin values that achieves the lowest possible value of total energy. And an instance of the easing problem is specified by giving numerical values for the Matrix J in Vector H. Although the easy model originates in physics, we understand the ground state problem to correspond to what would be called quadratic binary optimization in the field of operations research and in fact, in terms of computational complexity theory, it could be established that the easing ground state problem is np complete. Qualitatively speaking, this makes the easing problem a representative sort of hard optimization problem, for which it is expected that the runtime required by any computational algorithm to find exact solutions should, as anatomically scale exponentially with the number of spends and for worst case instances at each end. Of course, there's no reason to believe that the problem instances that actually arrives in practical optimization scenarios are going to be worst case instances. And it's also not generally the case in practical optimization scenarios that we demand absolute optimum solutions. Usually we're more interested in just getting the best solution we can within an affordable cost, where costs may be measured in terms of time, service fees and or energy required for a computation. This focuses great interest on so called heuristic algorithms for the easing problem in other NP complete problems which generally get very good but not guaranteed optimum solutions and run much faster than algorithms that are designed to find absolute Optima. To get some feeling for present day numbers, we can consider the famous traveling salesman problem for which extensive compilations of benchmarking data may be found online. A recent study found that the best known TSP solver required median run times across the Library of Problem instances That scaled is a very steep route exponential for end up to approximately 4500. This gives some indication of the change in runtime scaling for generic as opposed the worst case problem instances. Some of the instances considered in this study were taken from a public library of T SPS derived from real world Veil aside design data. This feels I TSP Library includes instances within ranging from 131 to 744,710 instances from this library with end between 6880 13,584 were first solved just a few years ago in 2017 requiring days of run time and a 48 core to King hurts cluster, while instances with and greater than or equal to 14,233 remain unsolved exactly by any means. Approximate solutions, however, have been found by heuristic methods for all instances in the VLS i TSP library with, for example, a solution within 0.14% of a no lower bound, having been discovered, for instance, with an equal 19,289 requiring approximately two days of run time on a single core of 2.4 gigahertz. Now, if we simple mindedly extrapolate the root exponential scaling from the study up to an equal 4500, we might expect that an exact solver would require something more like a year of run time on the 48 core cluster used for the N equals 13,580 for instance, which shows how much a very small concession on the quality of the solution makes it possible to tackle much larger instances with much lower cost. At the extreme end, the largest TSP ever solved exactly has an equal 85,900. This is an instance derived from 19 eighties VLSI design, and it's required 136 CPU. Years of computation normalized to a single cord, 2.4 gigahertz. But the 24 larger so called world TSP benchmark instance within equals 1,904,711 has been solved approximately within ophthalmology. Gap bounded below 0.474%. Coming back to the general. Practical concerns have applied optimization. We may note that a recent meta study analyzed the performance of no fewer than 37 heuristic algorithms for Max cut and quadratic pioneer optimization problems and found the performance sort and found that different heuristics work best for different problem instances selected from a large scale heterogeneous test bed with some evidence but cryptic structure in terms of what types of problem instances were best solved by any given heuristic. Indeed, their their reasons to believe that these results from Mexico and quadratic binary optimization reflected general principle of performance complementarity among heuristic optimization algorithms in the practice of solving heart optimization problems there. The cerise is a critical pre processing issue of trying to guess which of a number of available good heuristic algorithms should be chosen to tackle a given problem. Instance, assuming that any one of them would incur high costs to run on a large problem, instances incidence, making an astute choice of heuristic is a crucial part of maximizing overall performance. Unfortunately, we still have very little conceptual insight about what makes a specific problem instance, good or bad for any given heuristic optimization algorithm. This has certainly been pinpointed by researchers in the field is a circumstance that must be addressed. So adding this all up, we see that a critical frontier for cutting edge academic research involves both the development of novel heuristic algorithms that deliver better performance, with lower cost on classes of problem instances that are underserved by existing approaches, as well as fundamental research to provide deep conceptual insight into what makes a given problem in, since easy or hard for such algorithms. In fact, these days, as we talk about the end of Moore's law and speculate about a so called second quantum revolution, it's natural to talk not only about novel algorithms for conventional CPUs but also about highly customized special purpose hardware architectures on which we may run entirely unconventional algorithms for combinatorial optimization such as easing problem. So against that backdrop, I'd like to use my remaining time to introduce our work on analysis of coherent using machine architectures and associate ID optimization algorithms. These machines, in general, are a novel class of information processing architectures for solving combinatorial optimization problems by embedding them in the dynamics of analog, physical or cyber physical systems, in contrast to both MAWR traditional engineering approaches that build using machines using conventional electron ICS and more radical proposals that would require large scale quantum entanglement. The emerging paradigm of coherent easing machines leverages coherent nonlinear dynamics in photonic or Opto electronic platforms to enable near term construction of large scale prototypes that leverage post Simoes information dynamics, the general structure of of current CM systems has shown in the figure on the right. The role of the easing spins is played by a train of optical pulses circulating around a fiber optical storage ring. A beam splitter inserted in the ring is used to periodically sample the amplitude of every optical pulse, and the measurement results are continually read into a refugee A, which uses them to compute perturbations to be applied to each pulse by a synchronized optical injections. These perturbations, air engineered to implement the spin, spin coupling and local magnetic field terms of the easing Hamiltonian, corresponding to a linear part of the CME Dynamics, a synchronously pumped parametric amplifier denoted here as PPL and Wave Guide adds a crucial nonlinear component to the CIA and Dynamics as well. In the basic CM algorithm, the pump power starts very low and has gradually increased at low pump powers. The amplitude of the easing spin pulses behaviors continuous, complex variables. Who Israel parts which can be positive or negative, play the role of play the role of soft or perhaps mean field spins once the pump, our crosses the threshold for parametric self oscillation. In the optical fiber ring, however, the attitudes of the easing spin pulses become effectively Qantas ized into binary values while the pump power is being ramped up. The F P J subsystem continuously applies its measurement based feedback. Implementation of the using Hamiltonian terms, the interplay of the linear rised using dynamics implemented by the F P G A and the threshold conversation dynamics provided by the sink pumped Parametric amplifier result in the final state of the optical optical pulse amplitude at the end of the pump ramp that could be read as a binary strain, giving a proposed solution of the easing ground state problem. This method of solving easing problem seems quite different from a conventional algorithm that runs entirely on a digital computer as a crucial aspect of the computation is performed physically by the analog, continuous, coherent, nonlinear dynamics of the optical degrees of freedom. In our efforts to analyze CIA and performance, we have therefore turned to the tools of dynamical systems theory, namely, a study of modifications, the evolution of critical points and apologies of hetero clinic orbits and basins of attraction. We conjecture that such analysis can provide fundamental insight into what makes certain optimization instances hard or easy for coherent using machines and hope that our approach can lead to both improvements of the course, the AM algorithm and a pre processing rubric for rapidly assessing the CME suitability of new instances. Okay, to provide a bit of intuition about how this all works, it may help to consider the threshold dynamics of just one or two optical parametric oscillators in the CME architecture just described. We can think of each of the pulse time slots circulating around the fiber ring, as are presenting an independent Opio. We can think of a single Opio degree of freedom as a single, resonant optical node that experiences linear dissipation, do toe out coupling loss and gain in a pump. Nonlinear crystal has shown in the diagram on the upper left of this slide as the pump power is increased from zero. As in the CME algorithm, the non linear game is initially to low toe overcome linear dissipation, and the Opio field remains in a near vacuum state at a critical threshold. Value gain. Equal participation in the Popeo undergoes a sort of lazing transition, and the study states of the OPIO above this threshold are essentially coherent states. There are actually two possible values of the Opio career in amplitude and any given above threshold pump power which are equal in magnitude but opposite in phase when the OPI across the special diet basically chooses one of the two possible phases randomly, resulting in the generation of a single bit of information. If we consider to uncoupled, Opio has shown in the upper right diagram pumped it exactly the same power at all times. Then, as the pump power has increased through threshold, each Opio will independently choose the phase and thus to random bits are generated for any number of uncoupled. Oppose the threshold power per opio is unchanged from the single Opio case. Now, however, consider a scenario in which the two appeals air, coupled to each other by a mutual injection of their out coupled fields has shown in the diagram on the lower right. One can imagine that depending on the sign of the coupling parameter Alfa, when one Opio is lazing, it will inject a perturbation into the other that may interfere either constructively or destructively, with the feel that it is trying to generate by its own lazing process. As a result, when came easily showed that for Alfa positive, there's an effective ferro magnetic coupling between the two Opio fields and their collective oscillation threshold is lowered from that of the independent Opio case. But on Lee for the two collective oscillation modes in which the two Opio phases are the same for Alfa Negative, the collective oscillation threshold is lowered on Lee for the configurations in which the Opio phases air opposite. So then, looking at how Alfa is related to the J. I. J matrix of the easing spin coupling Hamiltonian, it follows that we could use this simplistic to a p o. C. I am to solve the ground state problem of a fair magnetic or anti ferro magnetic ankles to easing model simply by increasing the pump power from zero and observing what phase relation occurs as the two appeals first start delays. Clearly, we can imagine generalizing this story toe larger, and however the story doesn't stay is clean and simple for all larger problem instances. And to find a more complicated example, we only need to go to n equals four for some choices of J J for n equals, for the story remains simple. Like the n equals two case. The figure on the upper left of this slide shows the energy of various critical points for a non frustrated and equals, for instance, in which the first bifurcated critical point that is the one that I forget to the lowest pump value a. Uh, this first bifurcated critical point flows as symptomatically into the lowest energy easing solution and the figure on the upper right. However, the first bifurcated critical point flows to a very good but sub optimal minimum at large pump power. The global minimum is actually given by a distinct critical critical point that first appears at a higher pump power and is not automatically connected to the origin. The basic C am algorithm is thus not able to find this global minimum. Such non ideal behaviors needs to become more confident. Larger end for the n equals 20 instance, showing the lower plots where the lower right plot is just a zoom into a region of the lower left lot. It can be seen that the global minimum corresponds to a critical point that first appears out of pump parameter, a around 0.16 at some distance from the idiomatic trajectory of the origin. That's curious to note that in both of these small and examples, however, the critical point corresponding to the global minimum appears relatively close to the idiomatic projector of the origin as compared to the most of the other local minima that appear. We're currently working to characterize the face portrait topology between the global minimum in the antibiotic trajectory of the origin, taking clues as to how the basic C am algorithm could be generalized to search for non idiomatic trajectories that jump to the global minimum during the pump ramp. Of course, n equals 20 is still too small to be of interest for practical optimization applications. But the advantage of beginning with the study of small instances is that we're able reliably to determine their global minima and to see how they relate to the 80 about trajectory of the origin in the basic C am algorithm. In the smaller and limit, we can also analyze fully quantum mechanical models of Syrian dynamics. But that's a topic for future talks. Um, existing large scale prototypes are pushing into the range of in equals 10 to the 4 10 to 5 to six. So our ultimate objective in theoretical analysis really has to be to try to say something about CIA and dynamics and regime of much larger in our initial approach to characterizing CIA and behavior in the large in regime relies on the use of random matrix theory, and this connects to prior research on spin classes, SK models and the tap equations etcetera. At present, we're focusing on statistical characterization of the CIA ingredient descent landscape, including the evolution of critical points in their Eigen value spectra. As the pump power is gradually increased. We're investigating, for example, whether there could be some way to exploit differences in the relative stability of the global minimum versus other local minima. We're also working to understand the deleterious or potentially beneficial effects of non ideologies, such as a symmetry in the implemented these and couplings. Looking one step ahead, we plan to move next in the direction of considering more realistic classes of problem instances such as quadratic, binary optimization with constraints. Eso In closing, I should acknowledge people who did the hard work on these things that I've shown eso. My group, including graduate students Ed winning, Daniel Wennberg, Tatsuya Nagamoto and Atsushi Yamamura, have been working in close collaboration with Syria Ganguly, Marty Fair and Amir Safarini Nini, all of us within the Department of Applied Physics at Stanford University. On also in collaboration with the Oshima Moto over at NTT 55 research labs, Onda should acknowledge funding support from the NSF by the Coherent Easing Machines Expedition in computing, also from NTT five research labs, Army Research Office and Exxon Mobil. Uh, that's it. Thanks very much. >>Mhm e >>t research and the Oshie for putting together this program and also the opportunity to speak here. My name is Al Gore ism or Andy and I'm from Caltech, and today I'm going to tell you about the work that we have been doing on networks off optical parametric oscillators and how we have been using them for icing machines and how we're pushing them toward Cornum photonics to acknowledge my team at Caltech, which is now eight graduate students and five researcher and postdocs as well as collaborators from all over the world, including entity research and also the funding from different places, including entity. So this talk is primarily about networks of resonate er's, and these networks are everywhere from nature. For instance, the brain, which is a network of oscillators all the way to optics and photonics and some of the biggest examples or metal materials, which is an array of small resonate er's. And we're recently the field of technological photonics, which is trying thio implement a lot of the technological behaviors of models in the condensed matter, physics in photonics and if you want to extend it even further, some of the implementations off quantum computing are technically networks of quantum oscillators. So we started thinking about these things in the context of icing machines, which is based on the icing problem, which is based on the icing model, which is the simple summation over the spins and spins can be their upward down and the couplings is given by the JJ. And the icing problem is, if you know J I J. What is the spin configuration that gives you the ground state? And this problem is shown to be an MP high problem. So it's computational e important because it's a representative of the MP problems on NPR. Problems are important because first, their heart and standard computers if you use a brute force algorithm and they're everywhere on the application side. That's why there is this demand for making a machine that can target these problems, and hopefully it can provide some meaningful computational benefit compared to the standard digital computers. So I've been building these icing machines based on this building block, which is a degenerate optical parametric. Oscillator on what it is is resonator with non linearity in it, and we pump these resonate er's and we generate the signal at half the frequency of the pump. One vote on a pump splits into two identical photons of signal, and they have some very interesting phase of frequency locking behaviors. And if you look at the phase locking behavior, you realize that you can actually have two possible phase states as the escalation result of these Opio which are off by pie, and that's one of the important characteristics of them. So I want to emphasize a little more on that and I have this mechanical analogy which are basically two simple pendulum. But there are parametric oscillators because I'm going to modulate the parameter of them in this video, which is the length of the string on by that modulation, which is that will make a pump. I'm gonna make a muscular. That'll make a signal which is half the frequency of the pump. And I have two of them to show you that they can acquire these face states so they're still facing frequency lock to the pump. But it can also lead in either the zero pie face states on. The idea is to use this binary phase to represent the binary icing spin. So each opio is going to represent spin, which can be either is your pie or up or down. And to implement the network of these resonate er's, we use the time off blood scheme, and the idea is that we put impulses in the cavity. These pulses air separated by the repetition period that you put in or t r. And you can think about these pulses in one resonator, xaz and temporarily separated synthetic resonate Er's if you want a couple of these resonator is to each other, and now you can introduce these delays, each of which is a multiple of TR. If you look at the shortest delay it couples resonator wanted to 2 to 3 and so on. If you look at the second delay, which is two times a rotation period, the couple's 123 and so on. And if you have and minus one delay lines, then you can have any potential couplings among these synthetic resonate er's. And if I can introduce these modulators in those delay lines so that I can strength, I can control the strength and the phase of these couplings at the right time. Then I can have a program will all toe all connected network in this time off like scheme, and the whole physical size of the system scales linearly with the number of pulses. So the idea of opium based icing machine is didn't having these o pos, each of them can be either zero pie and I can arbitrarily connect them to each other. And then I start with programming this machine to a given icing problem by just setting the couplings and setting the controllers in each of those delight lines. So now I have a network which represents an icing problem. Then the icing problem maps to finding the face state that satisfy maximum number of coupling constraints. And the way it happens is that the icing Hamiltonian maps to the linear loss of the network. And if I start adding gain by just putting pump into the network, then the OPI ohs are expected to oscillate in the lowest, lowest lost state. And, uh and we have been doing these in the past, uh, six or seven years and I'm just going to quickly show you the transition, especially what happened in the first implementation, which was using a free space optical system and then the guided wave implementation in 2016 and the measurement feedback idea which led to increasing the size and doing actual computation with these machines. So I just want to make this distinction here that, um, the first implementation was an all optical interaction. We also had an unequal 16 implementation. And then we transition to this measurement feedback idea, which I'll tell you quickly what it iss on. There's still a lot of ongoing work, especially on the entity side, to make larger machines using the measurement feedback. But I'm gonna mostly focused on the all optical networks and how we're using all optical networks to go beyond simulation of icing Hamiltonian both in the linear and non linear side and also how we're working on miniaturization of these Opio networks. So the first experiment, which was the four opium machine, it was a free space implementation and this is the actual picture off the machine and we implemented a small and it calls for Mexico problem on the machine. So one problem for one experiment and we ran the machine 1000 times, we looked at the state and we always saw it oscillate in one of these, um, ground states of the icing laboratoria. So then the measurement feedback idea was to replace those couplings and the controller with the simulator. So we basically simulated all those coherent interactions on on FB g. A. And we replicated the coherent pulse with respect to all those measurements. And then we injected it back into the cavity and on the near to you still remain. So it still is a non. They're dynamical system, but the linear side is all simulated. So there are lots of questions about if this system is preserving important information or not, or if it's gonna behave better. Computational wars. And that's still ah, lot of ongoing studies. But nevertheless, the reason that this implementation was very interesting is that you don't need the end minus one delight lines so you can just use one. Then you can implement a large machine, and then you can run several thousands of problems in the machine, and then you can compare the performance from the computational perspective Looks so I'm gonna split this idea of opium based icing machine into two parts. One is the linear part, which is if you take out the non linearity out of the resonator and just think about the connections. You can think about this as a simple matrix multiplication scheme. And that's basically what gives you the icing Hambletonian modeling. So the optical laws of this network corresponds to the icing Hamiltonian. And if I just want to show you the example of the n equals for experiment on all those face states and the history Graham that we saw, you can actually calculate the laws of each of those states because all those interferences in the beam splitters and the delay lines are going to give you a different losses. And then you will see that the ground states corresponds to the lowest laws of the actual optical network. If you add the non linearity, the simple way of thinking about what the non linearity does is that it provides to gain, and then you start bringing up the gain so that it hits the loss. Then you go through the game saturation or the threshold which is going to give you this phase bifurcation. So you go either to zero the pie face state. And the expectation is that Theis, the network oscillates in the lowest possible state, the lowest possible loss state. There are some challenges associated with this intensity Durban face transition, which I'm going to briefly talk about. I'm also going to tell you about other types of non aerodynamics that we're looking at on the non air side of these networks. So if you just think about the linear network, we're actually interested in looking at some technological behaviors in these networks. And the difference between looking at the technological behaviors and the icing uh, machine is that now, First of all, we're looking at the type of Hamilton Ian's that are a little different than the icing Hamilton. And one of the biggest difference is is that most of these technological Hamilton Ian's that require breaking the time reversal symmetry, meaning that you go from one spin to in the one side to another side and you get one phase. And if you go back where you get a different phase, and the other thing is that we're not just interested in finding the ground state, we're actually now interesting and looking at all sorts of states and looking at the dynamics and the behaviors of all these states in the network. So we started with the simplest implementation, of course, which is a one d chain of thes resonate, er's, which corresponds to a so called ssh model. In the technological work, we get the similar energy to los mapping and now we can actually look at the band structure on. This is an actual measurement that we get with this associate model and you see how it reasonably how How? Well, it actually follows the prediction and the theory. One of the interesting things about the time multiplexing implementation is that now you have the flexibility of changing the network as you are running the machine. And that's something unique about this time multiplex implementation so that we can actually look at the dynamics. And one example that we have looked at is we can actually go through the transition off going from top A logical to the to the standard nontrivial. I'm sorry to the trivial behavior of the network. You can then look at the edge states and you can also see the trivial and states and the technological at states actually showing up in this network. We have just recently implement on a two D, uh, network with Harper Hofstadter model and when you don't have the results here. But we're one of the other important characteristic of time multiplexing is that you can go to higher and higher dimensions and keeping that flexibility and dynamics, and we can also think about adding non linearity both in a classical and quantum regimes, which is going to give us a lot of exotic, no classical and quantum, non innate behaviors in these networks. Yeah, So I told you about the linear side. Mostly let me just switch gears and talk about the nonlinear side of the network. And the biggest thing that I talked about so far in the icing machine is this face transition that threshold. So the low threshold we have squeezed state in these. Oh, pios, if you increase the pump, we go through this intensity driven phase transition and then we got the face stays above threshold. And this is basically the mechanism off the computation in these O pos, which is through this phase transition below to above threshold. So one of the characteristics of this phase transition is that below threshold, you expect to see quantum states above threshold. You expect to see more classical states or coherent states, and that's basically corresponding to the intensity off the driving pump. So it's really hard to imagine that it can go above threshold. Or you can have this friends transition happen in the all in the quantum regime. And there are also some challenges associated with the intensity homogeneity off the network, which, for example, is if one opioid starts oscillating and then its intensity goes really high. Then it's going to ruin this collective decision making off the network because of the intensity driven face transition nature. So So the question is, can we look at other phase transitions? Can we utilize them for both computing? And also can we bring them to the quantum regime on? I'm going to specifically talk about the face transition in the spectral domain, which is the transition from the so called degenerate regime, which is what I mostly talked about to the non degenerate regime, which happens by just tuning the phase of the cavity. And what is interesting is that this phase transition corresponds to a distinct phase noise behavior. So in the degenerate regime, which we call it the order state, you're gonna have the phase being locked to the phase of the pump. As I talked about non degenerate regime. However, the phase is the phase is mostly dominated by the quantum diffusion. Off the off the phase, which is limited by the so called shallow towns limit, and you can see that transition from the general to non degenerate, which also has distinct symmetry differences. And this transition corresponds to a symmetry breaking in the non degenerate case. The signal can acquire any of those phases on the circle, so it has a you one symmetry. Okay, and if you go to the degenerate case, then that symmetry is broken and you only have zero pie face days I will look at. So now the question is can utilize this phase transition, which is a face driven phase transition, and can we use it for similar computational scheme? So that's one of the questions that were also thinking about. And it's not just this face transition is not just important for computing. It's also interesting from the sensing potentials and this face transition, you can easily bring it below threshold and just operated in the quantum regime. Either Gaussian or non Gaussian. If you make a network of Opio is now, we can see all sorts off more complicated and more interesting phase transitions in the spectral domain. One of them is the first order phase transition, which you get by just coupling to Opio, and that's a very abrupt face transition and compared to the to the single Opio phase transition. And if you do the couplings right, you can actually get a lot of non her mission dynamics and exceptional points, which are actually very interesting to explore both in the classical and quantum regime. And I should also mention that you can think about the cup links to be also nonlinear couplings. And that's another behavior that you can see, especially in the nonlinear in the non degenerate regime. So with that, I basically told you about these Opio networks, how we can think about the linear scheme and the linear behaviors and how we can think about the rich, nonlinear dynamics and non linear behaviors both in the classical and quantum regime. I want to switch gear and tell you a little bit about the miniaturization of these Opio networks. And of course, the motivation is if you look at the electron ICS and what we had 60 or 70 years ago with vacuum tube and how we transition from relatively small scale computers in the order of thousands of nonlinear elements to billions of non elements where we are now with the optics is probably very similar to 70 years ago, which is a table talk implementation. And the question is, how can we utilize nano photonics? I'm gonna just briefly show you the two directions on that which we're working on. One is based on lithium Diabate, and the other is based on even a smaller resonate er's could you? So the work on Nana Photonic lithium naive. It was started in collaboration with Harvard Marko Loncar, and also might affair at Stanford. And, uh, we could show that you can do the periodic polling in the phenomenon of it and get all sorts of very highly nonlinear processes happening in this net. Photonic periodically polls if, um Diabate. And now we're working on building. Opio was based on that kind of photonic the film Diabate. And these air some some examples of the devices that we have been building in the past few months, which I'm not gonna tell you more about. But the O. P. O. S. And the Opio Networks are in the works. And that's not the only way of making large networks. Um, but also I want to point out that The reason that these Nana photonic goblins are actually exciting is not just because you can make a large networks and it can make him compact in a in a small footprint. They also provide some opportunities in terms of the operation regime. On one of them is about making cat states and Opio, which is, can we have the quantum superposition of the zero pie states that I talked about and the Net a photonic within? I've It provides some opportunities to actually get closer to that regime because of the spatial temporal confinement that you can get in these wave guides. So we're doing some theory on that. We're confident that the type of non linearity two losses that it can get with these platforms are actually much higher than what you can get with other platform their existing platforms and to go even smaller. We have been asking the question off. What is the smallest possible Opio that you can make? Then you can think about really wavelength scale type, resonate er's and adding the chi to non linearity and see how and when you can get the Opio to operate. And recently, in collaboration with us see, we have been actually USC and Creole. We have demonstrated that you can use nano lasers and get some spin Hamilton and implementations on those networks. So if you can build the a P. O s, we know that there is a path for implementing Opio Networks on on such a nano scale. So we have looked at these calculations and we try to estimate the threshold of a pos. Let's say for me resonator and it turns out that it can actually be even lower than the type of bulk Pip Llano Pos that we have been building in the past 50 years or so. So we're working on the experiments and we're hoping that we can actually make even larger and larger scale Opio networks. So let me summarize the talk I told you about the opium networks and our work that has been going on on icing machines and the measurement feedback. And I told you about the ongoing work on the all optical implementations both on the linear side and also on the nonlinear behaviors. And I also told you a little bit about the efforts on miniaturization and going to the to the Nano scale. So with that, I would like Thio >>three from the University of Tokyo. Before I thought that would like to thank you showing all the stuff of entity for the invitation and the organization of this online meeting and also would like to say that it has been very exciting to see the growth of this new film lab. And I'm happy to share with you today of some of the recent works that have been done either by me or by character of Hong Kong. Honest Group indicates the title of my talk is a neuro more fic in silica simulator for the communities in machine. And here is the outline I would like to make the case that the simulation in digital Tektronix of the CME can be useful for the better understanding or improving its function principles by new job introducing some ideas from neural networks. This is what I will discuss in the first part and then it will show some proof of concept of the game and performance that can be obtained using dissimulation in the second part and the protection of the performance that can be achieved using a very large chaos simulator in the third part and finally talk about future plans. So first, let me start by comparing recently proposed izing machines using this table there is elected from recent natural tronics paper from the village Park hard people, and this comparison shows that there's always a trade off between energy efficiency, speed and scalability that depends on the physical implementation. So in red, here are the limitation of each of the servers hardware on, interestingly, the F p G, a based systems such as a producer, digital, another uh Toshiba beautification machine or a recently proposed restricted Bozeman machine, FPD A by a group in Berkeley. They offer a good compromise between speed and scalability. And this is why, despite the unique advantage that some of these older hardware have trust as the currency proposition in Fox, CBS or the energy efficiency off memory Sisters uh P. J. O are still an attractive platform for building large organizing machines in the near future. The reason for the good performance of Refugee A is not so much that they operate at the high frequency. No, there are particular in use, efficient, but rather that the physical wiring off its elements can be reconfigured in a way that limits the funding human bottleneck, larger, funny and phenols and the long propagation video information within the system. In this respect, the LPGA is They are interesting from the perspective off the physics off complex systems, but then the physics of the actions on the photos. So to put the performance of these various hardware and perspective, we can look at the competition of bringing the brain the brain complete, using billions of neurons using only 20 watts of power and operates. It's a very theoretically slow, if we can see and so this impressive characteristic, they motivate us to try to investigate. What kind of new inspired principles be useful for designing better izing machines? The idea of this research project in the future collaboration it's to temporary alleviates the limitations that are intrinsic to the realization of an optical cortex in machine shown in the top panel here. By designing a large care simulator in silicone in the bottom here that can be used for digesting the better organization principles of the CIA and this talk, I will talk about three neuro inspired principles that are the symmetry of connections, neural dynamics orphan chaotic because of symmetry, is interconnectivity the infrastructure? No. Next talks are not composed of the reputation of always the same types of non environments of the neurons, but there is a local structure that is repeated. So here's the schematic of the micro column in the cortex. And lastly, the Iraqi co organization of connectivity connectivity is organizing a tree structure in the brain. So here you see a representation of the Iraqi and organization of the monkey cerebral cortex. So how can these principles we used to improve the performance of the icing machines? And it's in sequence stimulation. So, first about the two of principles of the estimate Trian Rico structure. We know that the classical approximation of the car testing machine, which is the ground toe, the rate based on your networks. So in the case of the icing machines, uh, the okay, Scott approximation can be obtained using the trump active in your position, for example, so the times of both of the system they are, they can be described by the following ordinary differential equations on in which, in case of see, I am the X, I represent the in phase component of one GOP Oh, Theo f represents the monitor optical parts, the district optical Parametric amplification and some of the good I JoJo extra represent the coupling, which is done in the case of the measure of feedback coupling cm using oh, more than detection and refugee A and then injection off the cooking time and eso this dynamics in both cases of CNN in your networks, they can be written as the grand set of a potential function V, and this written here, and this potential functionally includes the rising Maccagnan. So this is why it's natural to use this type of, uh, dynamics to solve the icing problem in which the Omega I J or the eyes in coping and the H is the extension of the icing and attorney in India and expect so. Not that this potential function can only be defined if the Omega I j. R. A. Symmetric. So the well known problem of this approach is that this potential function V that we obtain is very non convicts at low temperature, and also one strategy is to gradually deformed this landscape, using so many in process. But there is no theorem. Unfortunately, that granted conventions to the global minimum of There's even Tony and using this approach. And so this is why we propose, uh, to introduce a macro structures of the system where one analog spin or one D O. P. O is replaced by a pair off one another spin and one error, according viable. And the addition of this chemical structure introduces a symmetry in the system, which in terms induces chaotic dynamics, a chaotic search rather than a learning process for searching for the ground state of the icing. Every 20 within this massacre structure the role of the er variable eyes to control the amplitude off the analog spins toe force. The amplitude of the expense toe become equal to certain target amplitude a uh and, uh, and this is done by modulating the strength off the icing complaints or see the the error variable E I multiply the icing complaint here in the dynamics off air d o p. O. On then the dynamics. The whole dynamics described by this coupled equations because the e I do not necessarily take away the same value for the different. I thesis introduces a symmetry in the system, which in turn creates security dynamics, which I'm sure here for solving certain current size off, um, escape problem, Uh, in which the X I are shown here and the i r from here and the value of the icing energy showing the bottom plots. You see this Celtics search that visit various local minima of the as Newtonian and eventually finds the global minimum? Um, it can be shown that this modulation off the target opportunity can be used to destabilize all the local minima off the icing evertonians so that we're gonna do not get stuck in any of them. On more over the other types of attractors I can eventually appear, such as limits I contractors, Okot contractors. They can also be destabilized using the motivation of the target and Batuta. And so we have proposed in the past two different moderation of the target amateur. The first one is a modulation that ensure the uh 100 reproduction rate of the system to become positive on this forbids the creation off any nontrivial tractors. And but in this work, I will talk about another moderation or arrested moderation which is given here. That works, uh, as well as this first uh, moderation, but is easy to be implemented on refugee. So this couple of the question that represent becoming the stimulation of the cortex in machine with some error correction they can be implemented especially efficiently on an F B. G. And here I show the time that it takes to simulate three system and also in red. You see, at the time that it takes to simulate the X I term the EI term, the dot product and the rising Hamiltonian for a system with 500 spins and Iraq Spain's equivalent to 500 g. O. P. S. So >>in >>f b d a. The nonlinear dynamics which, according to the digital optical Parametric amplification that the Opa off the CME can be computed in only 13 clock cycles at 300 yards. So which corresponds to about 0.1 microseconds. And this is Toby, uh, compared to what can be achieved in the measurements back O C. M. In which, if we want to get 500 timer chip Xia Pios with the one she got repetition rate through the obstacle nine narrative. Uh, then way would require 0.5 microseconds toe do this so the submission in F B J can be at least as fast as ah one g repression. Uh, replicate pulsed laser CIA Um, then the DOT product that appears in this differential equation can be completed in 43 clock cycles. That's to say, one microseconds at 15 years. So I pieced for pouring sizes that are larger than 500 speeds. The dot product becomes clearly the bottleneck, and this can be seen by looking at the the skating off the time the numbers of clock cycles a text to compute either the non in your optical parts or the dog products, respect to the problem size. And And if we had infinite amount of resources and PGA to simulate the dynamics, then the non illogical post can could be done in the old one. On the mattress Vector product could be done in the low carrot off, located off scales as a look at it off and and while the guide off end. Because computing the dot product involves assuming all the terms in the product, which is done by a nephew, GE by another tree, which heights scarce logarithmic any with the size of the system. But This is in the case if we had an infinite amount of resources on the LPGA food, but for dealing for larger problems off more than 100 spins. Usually we need to decompose the metrics into ah, smaller blocks with the block side that are not you here. And then the scaling becomes funny, non inner parts linear in the end, over you and for the products in the end of EU square eso typically for low NF pdf cheap PGA you the block size off this matrix is typically about 100. So clearly way want to make you as large as possible in order to maintain this scanning in a log event for the numbers of clock cycles needed to compute the product rather than this and square that occurs if we decompose the metrics into smaller blocks. But the difficulty in, uh, having this larger blocks eyes that having another tree very large Haider tree introduces a large finding and finance and long distance start a path within the refugee. So the solution to get higher performance for a simulator of the contest in machine eyes to get rid of this bottleneck for the dot product by increasing the size of this at the tree. And this can be done by organizing your critique the electrical components within the LPGA in order which is shown here in this, uh, right panel here in order to minimize the finding finance of the system and to minimize the long distance that a path in the in the fpt So I'm not going to the details of how this is implemented LPGA. But just to give you a idea off why the Iraqi Yahiko organization off the system becomes the extremely important toe get good performance for similar organizing machine. So instead of instead of getting into the details of the mpg implementation, I would like to give some few benchmark results off this simulator, uh, off the that that was used as a proof of concept for this idea which is can be found in this archive paper here and here. I should results for solving escape problems. Free connected person, randomly person minus one spring last problems and we sure, as we use as a metric the numbers of the mattress Victor products since it's the bottleneck of the computation, uh, to get the optimal solution of this escape problem with the Nina successful BT against the problem size here and and in red here, this propose FDJ implementation and in ah blue is the numbers of retrospective product that are necessary for the C. I am without error correction to solve this escape programs and in green here for noisy means in an evening which is, uh, behavior with similar to the Cartesian mission. Uh, and so clearly you see that the scaring off the numbers of matrix vector product necessary to solve this problem scales with a better exponents than this other approaches. So So So that's interesting feature of the system and next we can see what is the real time to solution to solve this SK instances eso in the last six years, the time institution in seconds to find a grand state of risk. Instances remain answers probability for different state of the art hardware. So in red is the F B g. A presentation proposing this paper and then the other curve represent Ah, brick a local search in in orange and silver lining in purple, for example. And so you see that the scaring off this purpose simulator is is rather good, and that for larger plant sizes we can get orders of magnitude faster than the state of the art approaches. Moreover, the relatively good scanning off the time to search in respect to problem size uh, they indicate that the FPD implementation would be faster than risk. Other recently proposed izing machine, such as the hope you know, natural complimented on memories distance that is very fast for small problem size in blue here, which is very fast for small problem size. But which scanning is not good on the same thing for the restricted Bosman machine. Implementing a PGA proposed by some group in Broken Recently Again, which is very fast for small parliament sizes but which canning is bad so that a dis worse than the proposed approach so that we can expect that for programs size is larger than 1000 spins. The proposed, of course, would be the faster one. Let me jump toe this other slide and another confirmation that the scheme scales well that you can find the maximum cut values off benchmark sets. The G sets better candidates that have been previously found by any other algorithms, so they are the best known could values to best of our knowledge. And, um or so which is shown in this paper table here in particular, the instances, uh, 14 and 15 of this G set can be We can find better converse than previously known, and we can find this can vary is 100 times faster than the state of the art algorithm and CP to do this which is a very common Kasich. It s not that getting this a good result on the G sets, they do not require ah, particular hard tuning of the parameters. So the tuning issuing here is very simple. It it just depends on the degree off connectivity within each graph. And so this good results on the set indicate that the proposed approach would be a good not only at solving escape problems in this problems, but all the types off graph sizing problems on Mexican province in communities. So given that the performance off the design depends on the height of this other tree, we can try to maximize the height of this other tree on a large F p g a onda and carefully routing the components within the P G A and and we can draw some projections of what type of performance we can achieve in the near future based on the, uh, implementation that we are currently working. So here you see projection for the time to solution way, then next property for solving this escape programs respect to the prime assize. And here, compared to different with such publicizing machines, particularly the digital. And, you know, 42 is shown in the green here, the green line without that's and, uh and we should two different, uh, hypothesis for this productions either that the time to solution scales as exponential off n or that the time of social skills as expression of square root off. So it seems, according to the data, that time solution scares more as an expression of square root of and also we can be sure on this and this production show that we probably can solve prime escape problem of science 2000 spins, uh, to find the rial ground state of this problem with 99 success ability in about 10 seconds, which is much faster than all the other proposed approaches. So one of the future plans for this current is in machine simulator. So the first thing is that we would like to make dissimulation closer to the rial, uh, GOP oh, optical system in particular for a first step to get closer to the system of a measurement back. See, I am. And to do this what is, uh, simulate Herbal on the p a is this quantum, uh, condoms Goshen model that is proposed described in this paper and proposed by people in the in the Entity group. And so the idea of this model is that instead of having the very simple or these and have shown previously, it includes paired all these that take into account on me the mean off the awesome leverage off the, uh, European face component, but also their violence s so that we can take into account more quantum effects off the g o p. O, such as the squeezing. And then we plan toe, make the simulator open access for the members to run their instances on the system. There will be a first version in September that will be just based on the simple common line access for the simulator and in which will have just a classic or approximation of the system. We don't know Sturm, binary weights and museum in term, but then will propose a second version that would extend the current arising machine to Iraq off F p g. A, in which we will add the more refined models truncated, ignoring the bottom Goshen model they just talked about on the support in which he valued waits for the rising problems and support the cement. So we will announce later when this is available and and far right is working >>hard comes from Universal down today in physics department, and I'd like to thank the organizers for their kind invitation to participate in this very interesting and promising workshop. Also like to say that I look forward to collaborations with with a file lab and Yoshi and collaborators on the topics of this world. So today I'll briefly talk about our attempt to understand the fundamental limits off another continues time computing, at least from the point off you off bullion satisfy ability, problem solving, using ordinary differential equations. But I think the issues that we raise, um, during this occasion actually apply to other other approaches on a log approaches as well and into other problems as well. I think everyone here knows what Dorien satisfy ability. Problems are, um, you have boolean variables. You have em clauses. Each of disjunction of collaterals literally is a variable, or it's, uh, negation. And the goal is to find an assignment to the variable, such that order clauses are true. This is a decision type problem from the MP class, which means you can checking polynomial time for satisfy ability off any assignment. And the three set is empty, complete with K three a larger, which means an efficient trees. That's over, uh, implies an efficient source for all the problems in the empty class, because all the problems in the empty class can be reduced in Polian on real time to reset. As a matter of fact, you can reduce the NP complete problems into each other. You can go from three set to set backing or two maximum dependent set, which is a set packing in graph theoretic notions or terms toe the icing graphs. A problem decision version. This is useful, and you're comparing different approaches, working on different kinds of problems when not all the closest can be satisfied. You're looking at the accusation version offset, uh called Max Set. And the goal here is to find assignment that satisfies the maximum number of clauses. And this is from the NPR class. In terms of applications. If we had inefficient sets over or np complete problems over, it was literally, positively influenced. Thousands off problems and applications in industry and and science. I'm not going to read this, but this this, of course, gives a strong motivation toe work on this kind of problems. Now our approach to set solving involves embedding the problem in a continuous space, and you use all the east to do that. So instead of working zeros and ones, we work with minus one across once, and we allow the corresponding variables toe change continuously between the two bounds. We formulate the problem with the help of a close metrics. If if a if a close, uh, does not contain a variable or its negation. The corresponding matrix element is zero. If it contains the variable in positive, for which one contains the variable in a gated for Mitt's negative one, and then we use this to formulate this products caused quote, close violation functions one for every clause, Uh, which really, continuously between zero and one. And they're zero if and only if the clause itself is true. Uh, then we form the define in order to define a dynamic such dynamics in this and dimensional hyper cube where the search happens and if they exist, solutions. They're sitting in some of the corners of this hyper cube. So we define this, uh, energy potential or landscape function shown here in a way that this is zero if and only if all the clauses all the kmc zero or the clauses off satisfied keeping these auxiliary variables a EMS always positive. And therefore, what you do here is a dynamics that is a essentially ingredient descend on this potential energy landscape. If you were to keep all the M's constant that it would get stuck in some local minimum. However, what we do here is we couple it with the dynamics we cooperated the clothes violation functions as shown here. And if he didn't have this am here just just the chaos. For example, you have essentially what case you have positive feedback. You have increasing variable. Uh, but in that case, you still get stuck would still behave will still find. So she is better than the constant version but still would get stuck only when you put here this a m which makes the dynamics in in this variable exponential like uh, only then it keeps searching until he finds a solution on deer is a reason for that. I'm not going toe talk about here, but essentially boils down toe performing a Grady and descend on a globally time barren landscape. And this is what works. Now I'm gonna talk about good or bad and maybe the ugly. Uh, this is, uh, this is What's good is that it's a hyperbolic dynamical system, which means that if you take any domain in the search space that doesn't have a solution in it or any socially than the number of trajectories in it decays exponentially quickly. And the decay rate is a characteristic in variant characteristic off the dynamics itself. Dynamical systems called the escape right the inverse off that is the time scale in which you find solutions by this by this dynamical system, and you can see here some song trajectories that are Kelty because it's it's no linear, but it's transient, chaotic. Give their sources, of course, because eventually knowledge to the solution. Now, in terms of performance here, what you show for a bunch off, um, constraint densities defined by M overran the ratio between closes toe variables for random, said Problems is random. Chris had problems, and they as its function off n And we look at money toward the wartime, the wall clock time and it behaves quite value behaves Azat party nominally until you actually he to reach the set on set transition where the hardest problems are found. But what's more interesting is if you monitor the continuous time t the performance in terms off the A narrow, continuous Time t because that seems to be a polynomial. And the way we show that is, we consider, uh, random case that random three set for a fixed constraint density Onda. We hear what you show here. Is that the right of the trash hold that it's really hard and, uh, the money through the fraction of problems that we have not been able to solve it. We select thousands of problems at that constraint ratio and resolve them without algorithm, and we monitor the fractional problems that have not yet been solved by continuous 90. And this, as you see these decays exponentially different. Educate rates for different system sizes, and in this spot shows that is dedicated behaves polynomial, or actually as a power law. So if you combine these two, you find that the time needed to solve all problems except maybe appear traction off them scales foreign or merely with the problem size. So you have paranormal, continuous time complexity. And this is also true for other types of very hard constraints and sexual problems such as exact cover, because you can always transform them into three set as we discussed before, Ramsey coloring and and on these problems, even algorithms like survey propagation will will fail. But this doesn't mean that P equals NP because what you have first of all, if you were toe implement these equations in a device whose behavior is described by these, uh, the keys. Then, of course, T the continue style variable becomes a physical work off. Time on that will be polynomial is scaling, but you have another other variables. Oxidative variables, which structured in an exponential manner. So if they represent currents or voltages in your realization and it would be an exponential cost Al Qaeda. But this is some kind of trade between time and energy, while I know how toe generate energy or I don't know how to generate time. But I know how to generate energy so it could use for it. But there's other issues as well, especially if you're trying toe do this son and digital machine but also happens. Problems happen appear. Other problems appear on in physical devices as well as we discuss later. So if you implement this in GPU, you can. Then you can get in order off to magnitude. Speed up. And you can also modify this to solve Max sad problems. Uh, quite efficiently. You are competitive with the best heuristic solvers. This is a weather problems. In 2016 Max set competition eso so this this is this is definitely this seems like a good approach, but there's off course interesting limitations, I would say interesting, because it kind of makes you think about what it means and how you can exploit this thes observations in understanding better on a low continues time complexity. If you monitored the discrete number the number of discrete steps. Don't buy the room, Dakota integrator. When you solve this on a digital machine, you're using some kind of integrator. Um and you're using the same approach. But now you measure the number off problems you haven't sold by given number of this kid, uh, steps taken by the integrator. You find out you have exponential, discrete time, complexity and, of course, thistles. A problem. And if you look closely, what happens even though the analog mathematical trajectory, that's the record here. If you monitor what happens in discrete time, uh, the integrator frustrates very little. So this is like, you know, third or for the disposition, but fluctuates like crazy. So it really is like the intervention frees us out. And this is because of the phenomenon of stiffness that are I'll talk a little bit a more about little bit layer eso. >>You know, it might look >>like an integration issue on digital machines that you could improve and could definitely improve. But actually issues bigger than that. It's It's deeper than that, because on a digital machine there is no time energy conversion. So the outside variables are efficiently representing a digital machine. So there's no exponential fluctuating current of wattage in your computer when you do this. Eso If it is not equal NP then the exponential time, complexity or exponential costs complexity has to hit you somewhere. And this is how um, but, you know, one would be tempted to think maybe this wouldn't be an issue in a analog device, and to some extent is true on our devices can be ordered to maintain faster, but they also suffer from their own problems because he not gonna be affect. That classes soldiers as well. So, indeed, if you look at other systems like Mirandizing machine measurement feedback, probably talk on the grass or selected networks. They're all hinge on some kind off our ability to control your variables in arbitrary, high precision and a certain networks you want toe read out across frequencies in case off CM's. You required identical and program because which is hard to keep, and they kind of fluctuate away from one another, shift away from one another. And if you control that, of course that you can control the performance. So actually one can ask if whether or not this is a universal bottleneck and it seems so aside, I will argue next. Um, we can recall a fundamental result by by showing harder in reaction Target from 1978. Who says that it's a purely computer science proof that if you are able toe, compute the addition multiplication division off riel variables with infinite precision, then you could solve any complete problems in polynomial time. It doesn't actually proposals all where he just chose mathematically that this would be the case. Now, of course, in Real warned, you have also precision. So the next question is, how does that affect the competition about problems? This is what you're after. Lots of precision means information also, or entropy production. Eso what you're really looking at the relationship between hardness and cost of computing off a problem. Uh, and according to Sean Hagar, there's this left branch which in principle could be polynomial time. But the question whether or not this is achievable that is not achievable, but something more cheerful. That's on the right hand side. There's always going to be some information loss, so mental degeneration that could keep you away from possibly from point normal time. So this is what we like to understand, and this information laws the source off. This is not just always I will argue, uh, in any physical system, but it's also off algorithm nature, so that is a questionable area or approach. But China gets results. Security theoretical. No, actual solar is proposed. So we can ask, you know, just theoretically get out off. Curiosity would in principle be such soldiers because it is not proposing a soldier with such properties. In principle, if if you want to look mathematically precisely what the solar does would have the right properties on, I argue. Yes, I don't have a mathematical proof, but I have some arguments that that would be the case. And this is the case for actually our city there solver that if you could calculate its trajectory in a loss this way, then it would be, uh, would solve epic complete problems in polynomial continuous time. Now, as a matter of fact, this a bit more difficult question, because time in all these can be re scared however you want. So what? Burns says that you actually have to measure the length of the trajectory, which is a new variant off the dynamical system or property dynamical system, not off its parameters ization. And we did that. So Suba Corral, my student did that first, improving on the stiffness off the problem off the integrations, using implicit solvers and some smart tricks such that you actually are closer to the actual trajectory and using the same approach. You know what fraction off problems you can solve? We did not give the length of the trajectory. You find that it is putting on nearly scaling the problem sites we have putting on your skin complexity. That means that our solar is both Polly length and, as it is, defined it also poorly time analog solver. But if you look at as a discreet algorithm, if you measure the discrete steps on a digital machine, it is an exponential solver. And the reason is because off all these stiffness, every integrator has tow truck it digitizing truncate the equations, and what it has to do is to keep the integration between the so called stability region for for that scheme, and you have to keep this product within a grimace of Jacoby in and the step size read in this region. If you use explicit methods. You want to stay within this region? Uh, but what happens that some off the Eigen values grow fast for Steve problems, and then you're you're forced to reduce that t so the product stays in this bonded domain, which means that now you have to you're forced to take smaller and smaller times, So you're you're freezing out the integration and what I will show you. That's the case. Now you can move to increase its soldiers, which is which is a tree. In this case, you have to make domain is actually on the outside. But what happens in this case is some of the Eigen values of the Jacobean, also, for six systems, start to move to zero. As they're moving to zero, they're going to enter this instability region, so your soul is going to try to keep it out, so it's going to increase the data T. But if you increase that to increase the truncation hours, so you get randomized, uh, in the large search space, so it's it's really not, uh, not going to work out. Now, one can sort off introduce a theory or language to discuss computational and are computational complexity, using the language from dynamical systems theory. But basically I I don't have time to go into this, but you have for heart problems. Security object the chaotic satellite Ouch! In the middle of the search space somewhere, and that dictates how the dynamics happens and variant properties off the dynamics. Of course, off that saddle is what the targets performance and many things, so a new, important measure that we find that it's also helpful in describing thesis. Another complexity is the so called called Makarov, or metric entropy and basically what this does in an intuitive A eyes, uh, to describe the rate at which the uncertainty containing the insignificant digits off a trajectory in the back, the flow towards the significant ones as you lose information because off arrows being, uh grown or are developed in tow. Larger errors in an exponential at an exponential rate because you have positively up north spawning. But this is an in variant property. It's the property of the set of all. This is not how you compute them, and it's really the interesting create off accuracy philosopher dynamical system. A zay said that you have in such a high dimensional that I'm consistent were positive and negatively upon of exponents. Aziz Many The total is the dimension of space and user dimension, the number off unstable manifold dimensions and as Saddam was stable, manifold direction. And there's an interesting and I think, important passion, equality, equality called the passion, equality that connect the information theoretic aspect the rate off information loss with the geometric rate of which trajectory separate minus kappa, which is the escape rate that I already talked about. Now one can actually prove a simple theorems like back off the envelope calculation. The idea here is that you know the rate at which the largest rated, which closely started trajectory separate from one another. So now you can say that, uh, that is fine, as long as my trajectory finds the solution before the projective separate too quickly. In that case, I can have the hope that if I start from some region off the face base, several close early started trajectories, they kind of go into the same solution orphaned and and that's that's That's this upper bound of this limit, and it is really showing that it has to be. It's an exponentially small number. What? It depends on the end dependence off the exponents right here, which combines information loss rate and the social time performance. So these, if this exponents here or that has a large independence or river linear independence, then you then you really have to start, uh, trajectories exponentially closer to one another in orderto end up in the same order. So this is sort off like the direction that you're going in tow, and this formulation is applicable toe all dynamical systems, uh, deterministic dynamical systems. And I think we can We can expand this further because, uh, there is, ah, way off getting the expression for the escaped rate in terms off n the number of variables from cycle expansions that I don't have time to talk about. What? It's kind of like a program that you can try toe pursuit, and this is it. So the conclusions I think of self explanatory I think there is a lot of future in in, uh, in an allo. Continue start computing. Um, they can be efficient by orders of magnitude and digital ones in solving empty heart problems because, first of all, many of the systems you like the phone line and bottleneck. There's parallelism involved, and and you can also have a large spectrum or continues time, time dynamical algorithms than discrete ones. And you know. But we also have to be mindful off. What are the possibility of what are the limits? And 11 open question is very important. Open question is, you know, what are these limits? Is there some kind off no go theory? And that tells you that you can never perform better than this limit or that limit? And I think that's that's the exciting part toe to derive thes thes this levian 10.

>>Good morning. Good afternoon. Good evening, everyone. I should thank Entity Research and the Oshie for putting together this program and also the opportunity to speak here. My name is Al Gore ism or Andy and I'm from Caltech. And today I'm going to tell you about the work that we have been doing on networks off optical parametric oscillators and how we have been using them for icing machines and how we're pushing them toward Cornum. Photonics should acknowledge my team at Caltech, which is now eight graduate students and five researcher and postdocs as well as collaborators from all over the world, including entity research and also the funding from different places, including entity. So this talk is primarily about networks of resonate er's and these networks are everywhere from nature. For instance, the brain, which is a network of oscillators all the way to optics and photonics and some of the biggest examples or meta materials, which is an array of small resonate er's. And we're recently the field of technological photonics, which is trying thio implement a lot of the technological behaviors of models in the condensed matter, physics in photonics. And if you want to extend it even further. Some of the implementations off quantum computing are technically networks of quantum oscillators. So we started thinking about these things in the context of icing machines, which is based on the icing problem, which is based on the icing model, which is the simple summation over the spins and spins can be their upward down, and the couplings is given by the G I J. And the icing problem is, if you know J I J. What is the spin configuration that gives you the ground state? And this problem is shown to be an MP high problem. So it's computational e important because it's a representative of the MP problems on NPR. Problems are important because first, their heart in standard computers, if you use a brute force algorithm and they're everywhere on the application side. That's why there is this demand for making a machine that can target these problems and hopefully it can provide some meaningful computational benefit compared to the standard digital computers. So I've been building these icing machines based on this building block, which is a degenerate optical parametric oscillator on what it is is resonator with non linearity in it and we pump these resonate er's and we generate the signal at half the frequency of the pump. One vote on a pump splits into two identical photons of signal, and they have some very interesting phase of frequency locking behaviors. And if you look at the phase locking behavior, you realize that you can actually have two possible face states as the escalation result of these Opio, which are off by pie, and that's one of the important characteristics of them. So I want to emphasize >>a little more on that, and I have this mechanical analogy which are basically two simple pendulum. But there are parametric oscillators because I'm going to modulate the parameter of them in this video, which is the length of the strength on by that modulation, which is that will make a pump. I'm gonna make a muscular. That'll make a signal, which is half the frequency of the pump. >>And I have two of them to show you that they can acquire these face states so they're still face their frequency lock to the pump. But it can also lead in either the zero pie face state on. The idea is to use this binary phase to represent the binary icing spin. So each Opio is going to represent spin, which can be >>either is your pie or up or down, >>and to implement the network of these resonate er's. We use the time off blood scheme, and the idea is that we put impulses in the cavity, these pulses air separated by the repetition period that you put in or t R. And you can think about these pulses in one resonator, xaz and temporarily separated synthetic resonate Er's If you want a couple of these resonator is to each other, and now you can introduce these delays, each of which is a multiple of TR. If you look at the shortest delay it couples resonator wanted to 2 to 3 and so on. If you look at the second delay, which is two times a rotation period, the couple's 123 and so on. If you have any minus one delay lines, then you can have any potential couplings among these synthetic resonate er's. And if I can introduce these modulators in those delay lines so that I can strength, I can control the strength and the phase of these couplings at the right time. Then I can >>have a program. We'll all toe all connected network in this time off like scheme. >>And the whole physical size of the system scales linearly with the number of pulses. So the idea of opium based icing machine is didn't having these o pos. Each of them can be either zero pie, and I can arbitrarily connect them to each other. And then I start with programming this machine to a given icing problem by just setting the couplings and setting the controllers in each of those delight lines. So now I have a network which represents an icing problem thin the icing problem maps to finding the face state that satisfy maximum number of coupling constraints. And the way it happens is that the icing Hamiltonian maps to the linear loss of the network. And if I start adding gain by just putting pump into the network, then the OPI ohs are expected to oscillating the lowest, lowest lost state. And, uh and we have been doing these in the past, uh, six or seven years and I'm just going to quickly show you the transition, especially what happened in the first implementation which was using a free space optical system and then the guided wave implementation in 2016 and the measurement feedback idea which led to increasing the size and doing actual computation with these machines. So I just want to make this distinction here that, um the first implementation was on our optical interaction. We also had an unequal 16 implementation and then we transition to this measurement feedback idea, which I'll tell you quickly what it iss on. There's still a lot of ongoing work, especially on the entity side, to make larger machines using the measurement feedback. But I'm gonna mostly focused on the all optical networks and how we're using all optical networks to go beyond simulation of icing. Hamiltonian is both in the linear and >>nonlinear side and also how we're working on miniaturization of these Opio networks. So >>the first experiment, which was the four Opium machine it was a free space implementation and this is the actual picture of the machine and we implemented a small and it calls for Mexico problem on the machine. So one problem for one experiment and we ran the machine 1000 times, we looked at the state and we always saw it oscillate in one of these, um, ground states of the icing laboratoria. Yeah, so then the measurement feedback idea was to replace those couplings and the controller with the simulator. So we basically simulated all those coherent interactions on on FB g A. And we replicated the coherent pulse with respect to all those measurements. And then we injected it back into the cavity and on the near to you still remain. So it still is a non. They're dynamical system, but the linear side is all simulated. So there are lots of questions about if this system is preserving important information or not, or if it's gonna behave better Computational wars. And that's still ah, lot of ongoing studies. But nevertheless, the reason that this implementation was very interesting is that you don't need the end minus one delight lines so you can just use one, and you can implement a large machine, and then you can run several thousands of problems in the machine, and then you can compare the performance from the computational perspective. Looks so I'm gonna split this idea of opium based icing machine into two parts One is the linear part, which is if you take out the non linearity out of the resonator and just think about the connections. You can think about this as a simple matrix multiplication scheme, and that's basically >>what gives you the icing Hamiltonian model A. So the optical loss of this network corresponds to the icing Hamiltonian. >>And if I just want to show you the example of the n equals for experiment on all those face states and the history Graham that we saw, you can actually calculate the laws of each of those states because all those interferences in the beam splitters and the delay lines are going to give you a different losses. And then you will see that ground states corresponds to the lowest laws of the actual optical network. If you add the non linearity, the simple way of thinking about what the non linearity does is that it provides to gain, and then you start bringing up the gain so that it hits the loss. Then you go through the game saturation or the threshold which is going to give you this phase bifurcation. >>So you go either to zero the pie face state, and the expectation is that this the network oscillates in the lowest possible state, the lowest possible loss state. >>There are some challenges associated with this intensity Durban face transition, which I'm going to briefly talk about. I'm also going to tell you about other types of non their dynamics that we're looking at on the non air side of these networks. So if you just think about the linear network, we're actually interested in looking at some technological behaviors in these networks. And the difference between looking at the technological behaviors and the icing uh, machine is that now, First of all, we're looking at the type of Hamilton Ian's that are a little different than the icing Hamilton. And one of the biggest difference is is that most of these technological Hamilton Ian's that require breaking the time reversal symmetry, meaning that you go from one spin to on the one side to another side and you get one phase. And if you go back where you get a different phase, and the other thing is that we're not just interested in finding the ground state, we're actually now interesting and looking at all sorts of States and looking at the dynamics and the behaviors of all these states in the network. So we started with the simplest implementation, of course, which is a one d chain of thes resonate er's which corresponds to a so called ssh model. In the technological work, we get the similar energy to los mapping. And now we can actually look at the band structure on. This is an actual measurement >>that we get with this associate model and you see how it reasonably how how? Well, it actually follows the prediction and the theory. >>One of the interesting things about the time multiplexing implementation is that now you have the flexibility of changing the network as we were running the machine. And that's something unique about this time multiplex implementation so that we can actually look at the dynamics. And one example >>that we have looked at is we can actually go to the transition off going from top a logical to the to the standard nontrivial. I'm sorry to the trivial behavior of the network. >>You can then look at the edge states and you can also see the trivial and states and the technological at states actually showing up in this network. We have just recently implement on a two D, >>uh, network with Harper Hofstadter model when you don't have the results here. But we're one of the other important characteristic of time multiplexing is that you can go to higher and higher dimensions and keeping that flexibility and dynamics. And we can also think about adding non linearity both in a classical and quantum regimes, which is going to give us a lot of exotic oh, classical and quantum, non innate behaviors in these networks. >>So I told you about the linear side. Mostly let me just switch gears and talk about the nonlinear side of the network. And the biggest thing that I talked about so far in the icing machine is this phase transition, that threshold. So the low threshold we have squeezed state in these Oh, pios, if you increase the pump, we go through this intensity driven phase transition and then we got the face stays above threshold. And this is basically the mechanism off the computation in these O pos, which is through this phase transition below to above threshold. So one of the characteristics of this phase transition is that below threshold, you expect to see quantum states above threshold. You expect to see more classical states or coherent states, and that's basically corresponding to the intensity off the driving pump. So it's really hard to imagine that it can go above threshold. Or you can have this friends transition happen in the all in the quantum regime. And there are also some challenges associated with the intensity homogeneity off the network. Which, for example, is if one Opio starts oscillating and then its intensity goes really high. Then it's going to ruin this collective decision making off the network because of the intensity driven face transition nature. So So the question is, can we look at other phase transitions? Can we utilize them for both computing? And also, can we bring them to the quantum regime on? I'm going to specifically talk about the face transition in the spectral domain, which is the transition from the so called degenerate regime, which is what I mostly talked about to the non degenerate regime, which happens by just tuning the phase of the cavity. And what is interesting is that this phase transition corresponds to a distinct phase noise, behavior So in the degenerate regime, which we call it the order state. You're gonna have the phase being locked to the phase of the pump as I talked about in the non the general regime. However, the phase is the phase is mostly dominated by the quantum diffusion off the off the phase, which is limited by the so called shallow towns limit and you can see that transition from the general to non degenerate, which also has distinct symmetry differences. And this transition corresponds to a symmetry breaking in the non degenerate case. The signal can acquire any of those phases on the circle, so it has a you one symmetry. And if you go to the degenerate case, then that symmetry is broken and you only have zero pie face days I will look at So now the question is can utilize this phase transition, which is a face driven phase transition and can we use it for similar computational scheme? So that's one of the questions that were also thinking about. And it's not just this face transition is not just important for computing. It's also interesting from the sensing potentials and this face transition. You can easily bring it below threshold and just operated in the quantum regime. Either Gaussian or non Gaussian. If you make a network of Opio is now, we can see all sorts of more complicated and more interesting phase transitions in the spectral domain. One of them is the first order phase transition, which you get by just coupling to oppose. And that's a very abrupt face transition and compared to the to the single Opio face transition. And if you do the couplings right, you can actually get a lot of non her mission dynamics and exceptional points, which are actually very interesting to explore both in the classical and quantum regime. And I should also mention that you can think about the cup links to be also nonlinear couplings. And that's another behavior that you can see, especially in the nonlinear in the non degenerate regime. So with that, I basically told you about these Opio networks, how we can think about the linear scheme and the linear behaviors and how we can think about the rich, nonlinear dynamics and non linear behaviors both in the classical and quantum regime. I want to switch gear and tell you a little bit about the miniaturization of these Opio networks. And of course, the motivation is if you look at the electron ICS and >>what we had 60 or 70 years ago with vacuum tube and how we transition from relatively small scale computers in the order of thousands of nonlinear elements to billions of non linear elements, where we are now with the optics is probably very similar to seven years ago, which is a tabletop implementation. >>And the question is, how can we utilize nano photonics? I'm gonna just briefly show you the two directions on that which we're working on. One is based on lithium Diabate, and the other is based on even a smaller resonate er's Did you? So the work on Nana Photonic lithium naive. It was started in collaboration with Harvard Marko Loncar and also might affair at Stanford. And, uh, we could show that you can do the >>periodic polling in the phenomenon of it and get all sorts of very highly non in your process is happening in this net. Photonic periodically polls if, um Diabate >>and now we're working on building. Opio was based on that kind of photonic lithium Diabate and these air some some examples of the devices that we have been building in the past few months, which I'm not gonna tell you more about. But the OPI ohs and the Opio networks are in the works, and that's not the only way of making large networks. But also I want to point out that the reason that these Nana photonic goblins are actually exciting is not just because you can make a large networks and it can make him compact in a in a small footprint, they also provide some opportunities in terms of the operation regime. On one of them is about making cat states in o pos, which is can we have the quantum superposition of >>the zero pie states that I talked about >>and the nano photonics within? I would provide some opportunities to actually get >>closer to that regime because of the spatial temporal confinement that you can get in these wave guides. So we're doing some theory on that. We're confident that the type of non linearity two losses that it can get with these platforms are actually much higher than what you can get with other platform, other existing platforms and to >>go even smaller. We have been asking the question off. What is the smallest possible Opio that you can make? Then you can think about really wavelength scale type resonate er's and adding the chi to non linearity and see how and when you can get the Opio to operate. And recently, in collaboration with us. See, we have been actually USC and Creole. We have demonstrated that you can use nano lasers and get some spin Hamiltonian implementations on those networks. So if you can't build a pos, we know that there is a path for implementing Opio Networks on on such a nano scale. So we have looked at these calculations and we try to >>estimate the threshold of a pos. Let's say for me resonator and it turns out that it can actually be even lower than the type of bulk Pippen O pos that we have been building in the past 50 years or so. >>So we're working on the experiments and we're hoping that we can actually make even larger and larger scale Opio networks. So let me summarize the talk I told you about the opium networks and >>our work that has been going on on icing machines and the >>measurement feedback on I told you about the ongoing work on the all optical implementations both on the linear side and also on the nonlinear behaviors. And I also told you >>a little bit about the efforts on miniaturization and going to the to the nano scale. So with that, I would like Thio stop here and thank you for your attention.

>> Live from Orlando, Florida. It's theCUBE. Covering Microsoft Ignite. Brought to you by Cohesity, and theCUBE's Ecosystem partners. >> Welcome back, everyone, to theCUBE's live coverage of Microsoft Ignite here in the Orange County Civic Center in Orlando, Florida. I'm your host, Rebecca Knight, along with my cohost, Stu Miniman. We are joined by Raphael Meyerowitz, he is the VP Office of the CTO at Presidio, And Jake Smith, who is the Director Data Center Solutions and Technologies at Intel. Thank you both so much for coming back on theCUBE. You're both CUBE alums. >> Thank you for having us. >> It's great to be back. >> So, I want to start by laying out for our viewers, why you're here, and if you're part of the Microsoft ecosystem: Intel, Cisco, Dell and others. Can you explain a little bit, to our viewers, the roll you play in this ecosystem. >> Well, for us, Microsoft is a long time partner. I mean, it's pretty well documented, we don't want to go there today, but at this particular event we're announcing a bunch of new product solutions. We're announcing new technology capabilities. And at four PM we're going to announce some world record results, for performance with an operating system in an application environment. So it's a very exciting time for Intel to be a part of this event. >> Well, this is quite a tease. (giggles) Can you give us a little-- >> You're going to have to wait 'til four PM. I will say, it has to do with Windows Server. It has to do with Xeon scale of a processor family. And, our future Optane products. >> Well, so, these are all great lead ins. And, before the cameras were rolling we were talking about all of these things. You want to go through, a little bit, where we are with each of those businesses? >> Yeah, at Presidio, we've mostly been partnering with Intel for a long time. And one of the things that we've seen also, is how Intel has developed their ecosystem of partners. The software, like today, if you look at today what was in our today with desktop as a service with citrix. That's something that we have been involved in, probably, for about 10 years. And now we actually seen that come to market. We're not just, the control plane is in the cloud. But, the actual, virtual desktops are in the cloud. And, we think that that's going to be a really good viable options for our customers with Office 365. >> Raph, maybe expand on that a little bit for our audience. You know, one of the things I always say is you talk in this multi-cloud heterogeneous world. You want to follow the apps. You want to follow the data. Well, you know, the desktop is part of where those applications and data live. So, how does that, you know, tie into all the cloud stuff we've been talkin' about, the last few years? >> So, for a lot of customers, one of the reasons they move to cloud is really for simplicities sake, alright. When you look at the desktop, the desktop is really not necessarily being the most simple thing in the world. Whether it's virtual, or whether it's physical desktop. By having the control plane in the virtual desktop in the cloud, where you can consume it with Office 365. And also through Microsoft. And you can buy it through a single entity. Customers are already going to see a lot of value in that. And we think it's really going to play in the market really, really well. Upper Enterprise customers and some Healthcare customers may take a little bit more time to adapt to. >> Jake, one of the things we talk, for years, we talked about people did their upgrades based on the tick-tock of the Intel fees there. >> Correct. >> Now we're talkin' about things like, you know, Windows as a service, going Evergreen. Maybe, how does that relationship, the old traditional Wintel versus the cloud era. Upgrades. You're talkin' about the new latest generation. How do we think about that? >> You know what, I'm not going to use that, the merged term, because that's, you know. The work that Windows does on Xeon scalable processor family has been amazing. But, typically, we've done a two to three year cycle on a server release. With our new road map, which we announced in August, which you were there for, so thank you. We're actually going to release a new CPU every year. We're releasing a new CPU every year because we have to deal with the fact that cloud customers, in Azure, want to have the availability to the latest and greatest technology, right now. And partners, like Presidio and Raph's team, have developed technologies, like Concierge, which he'll talk about, that give customers the ability to manage their hybrid cloud environments, both in the cloud and on premises. When you start giving customers that flexibility they want the choice to say, I want to deploy your latest Xeon scalable processor family, Skylake processors this year, and next year, I'm going to maybe skip a year before I deploy your next version. >> Yeah, thanks Jake. One of the things that we've done at Presidio, we've tried to innovate ourselves, and we listen to our customers, and we know where our customers pain points are. So, Presidio Concierge is something that we developed from the ground up, that provides both shared space applications, provides customers with the usage on their shared space applications, how they're consuming their licenses, and also provides them with an allessor sign, so the infrastructure's a service. A lot of customers, when you talk about multi cloud, it doesn't always necessarily always mean the Harper scalers, right. It could mean shared space products, as well. So, we developed this product from the ground up in combination with Intel, and it's something that our customers are starting to use a lot, and we think that there's going to be a great grow in their first product. Some of the features that we actually give to our customers are actually for free, because we know that our customers are really battling with figuring out their usage patterns, internally. >> Well, I want to hear about those pain points. What were the problems that you were trying to solve with Concierge? >> So, some of the pain points, you know, we have customers today that get invoices from some of the public cloud companies or their service providers or with their infrastructures service. And the invoices are 50 pages long. They can never actually figure out what their true costs are. So we, through a shared space platform, that we developed from the ground up, we can provide customers with all of those metrics around their licenses. Plus, also, their usage around infrastructure as a service, as well. >> And, what has demand been like? >> The demand's been really good. Actually, when we launched product about two, three months ago, we were already at 20 customers. And we've seen a lot of interest. Presidio has about 7700 customers nationally, that we call on today. And we've grown tremendously, we have about a three billion dollar infrastructure partner today that provides both on premises and public cloud services. >> Yeah, I like, you brought up the fact that customers are looking for simplicity. Unfortunately, today, cloud is no longer simple. You know, I would say if you said, okay, If I went to my server vendor of choice and wanted to configure something, versus I went to my cloud vendor of choice and try to configure something, cloud might even be more challenging for somebody to do. But, one of the areas that we're trying to help customers get some simplicity back, is if you look at solutions like Azure Stack. So, Rebecca and I interviewed Jeffrey Snover earlier today, and that was the goal they had, was to give, kind of, that operational model and even some of the services from Azure and put them in my data center. Was wondering if Intel and Presidio are both partnering with Microsoft on this. What are you seeing, what are you hearing from customers? Any proof points as to how the roll outs are going, on there? >> We at Presidio, we are one of the first Azure Stack partners. Probably, about a year and a half ago, when it was actually announced and when it went, yeah, I think it was June of last year, and we partnered with Cisco, Dell, and also HP in the space, and we seen demand from our customers creep up. Single node solutions. We've seen demand with Single node PLC solutions are being deployed today. And then, in the public sector, we're also starting to see customers that are interested in it because it will provide them with a gateway to the public cloud in the future. >> Yeah, we're seeing the exact same thing. Obviously, we've been partnering together for some time. The beauty of Azure Stack is it's optimized for Xeon scalable processor family, as well as Intel Optane technologies, both the SSDs and in the future, our persistent memory capabilities. What we like in our work that we've done on Azure Stack and Azure Stack development, is that customers have had a lot of releases to begin to determine where Azure Stack's going to fit in their overall portfolio. And that's how you really have to look at Azure Stack, is how do you manage your portfolio between the cloud and on premises. Azure Stack is a great tool for that. >> You know, leading up to the release of Azure Stack, I talked to a number of service providers that had pent up demand. Leading up to this show, I was hearing a lot of non-North American interest. Can you give us any characterization as to how the roll out's going? >> Yeah, I think when you look at non-North American interest, there's a lot of localization, that has to take place in a lot of those countries. Maybe there's not actually an Azure, a public cloud Azure in those countries today, which is something that Microsoft is building towards. So, customers want to get used to their API's, they want to keep their data local. And when they're the same API's, on premises versus in the public cloud for all of their applications. And that's why I think you see, especially in Europe, as an example, a lot of countries in Europe where actually, data sovereignty's a big issue, alright. The data's not allowed to leave the country that they're actually in. And the demand, I think will, I always say, Microsoft, version two or version three. They always get it right. I mean, we've seen this time and time again. They've proven to us, they get this right all the time. >> I want to follow up on something you were just talking about, though with, sort of, risk management being a really big, hot opportunity. The next generation of risk management and mitigation. Can you talk a little bit about what you're doing there, and what you're hearing from customers? >> Yeah, so, Presidio developed the next generation risk management framework, called NGRM. So, we found we do a lot of security with Cisco, Palo Alto. We have a lot of security vendors out there that we deal with, but what our CIO's were really looking for is they were looking for a single dashboard that could actually provide them with a scorecard: Green, Yellow, or Red. Basically saying this is where we're at in our security strategy and this is what we need to remediate right away. They can take that to their board, they can also use that internally for all of their CSO's and also all their internal IT infrastructure personnel that they have. So, it's something that we've seen customers adopt, because it provides that analysis and the remediation and it's not necessarily tied to a specific product. Again, this is a shared space platform that we developed from the ground up, because our customers are always saying, "Well, there's always security vulnerabilities. "How can we constantly check on this?" Right? And it doesn't matter whether you're running Azure, whether you have on-premises solutions, or whether you have some other cloud provider, we can provide that holistic view for customers today. >> One of the announcements that I think surprised everyone. I mean, things like Server 2019, we all expect. The open data initiative, the commentary that we had is if you talk about digital transformation. I mean, Microsoft, Adobe and SAP. Two companies at the center of it. What does it mean? When will customers see the benefits of this? And any commentary of digital transformation in general would be great. >> Well, typically, we've been involved in a lot of these open standards, and they typically take three to five years to work their way all the way through the system and build the proper ecosystem and standards. And then work their way into the product lines. I think, in this particular instance, there is a driver. We talked about the driver of cloud and why we, we Intel, are now producing chips every year, and you're not waiting for the three year release cycle. Well, the open data initiative, I think, falls into that camp. I think you're going to see an escalated transition to the open data initiative, because people have to be able to move their workloads. Presidio recognized it very early on in the process. We've been working with them for some time. But that's one of the values that they bring to customers, is their ability to do that. But, more and more customers and more and more data are being stretched and there has to be compatibility between file systems, file format, and data classification. The open data initiative is a start in that direction. >> Yeah, I mean, one of the examples that I could give you also is we always talk about IT transformation. We have a large customer that's actually a fleet truck company that underwent IT transformation, and they came to us and they said that they actually needed telematics on the trucks in the fleet of trucks. And the reason was because a lot of these trucks are breaking down and they would send it to a mechanic and the mechanic would diagnose it. So, we actually created, in partnership with Intel and with Microsoft, this telematic platform that actually can provide the customer, in real time, with what issues they actually have with the truck. And it saves the customer a lot of money. That's the type of information that customers are looking for. This customer has on premises data, plus, also in the public cloud, and I think stretching it and providing analytics around that is really important. >> And is it possible to take away the silos? I mean, you seem to be an optimist here. >> I'm very optimistic that we can take away the silos, but I'm also realistic. The only way to take away the silos is to develop new applications, new capabilities. And as my friends in Windows Server Team will tell you, we spend a lot of time trying to figure out, how do we use virtualization and container technologies to take old legacy data and carry it forward onto new modern IT infrastructure. And when you can do that, then you can extract value from the data. If you can not take it from an old, antiquated infrastructure to a new infrastructure as Presidio has done, you stranded the data. And that's where you have those silo breakdowns. So, I think we're developing the tools, but we're not all the way there. >> Yeah, you look at Windows 2019 coming out, there's Linux support in Windows 2019. Who would ever think that Microsoft would be releasing Linux support. >> Microsoft loves Linux. >> Microsoft loves Linux now, right? >> And they will in get it. >> And they'll get it now as well. Microsoft is really developed their ecosystem. Our partners also around the open API's and what they've been doing over the past few years. And I think customers are really starting to embrace that. And you look at even another feature that's coming with Windows 2019 with Storage Spaces Direct. Right, I think Microsoft, this is really going to be their entry into the Apple convert space. Customers are going to start building, they'll have to converge platform based on Windows 2019 Data Center. >> Wondering if you can give a little more color here, Raph. You and I lived through, kind of converged and hyperconvergence, when we wrote our original research at Wikibon, it was VMware is the one that's going to get everybody talking about it, but the one eventually that will be very important here is Microsoft. 'Cause, Microsoft owns the apps. They've got the operating systems, so absolutely, they can be critical in the HCI space. What are they doing and how does Presidio and partners go to market with this? >> So, I mean, when you look at Windows 2016, Windows 2016 was really the first iteration of Storage Spaces Direct. Windows 2019 has really improved upon that, and we're starting to see customers become more interested in that. The reason is because customers want a single platform that they can easily manage with a single operating system. So, there used to be the war, as you mentioned Stu, between VMware and Harper-V. ESXi and Harper-V. I don't really see that being talked about anymore. It's more around the features and the robust features that customers can actually get on as quickly as possible. I don't know if you have anymore. >> Well Raph, you're absolutely right on. I think people have taken virtualization for granted. We added virtualization technology in Xeon in 2006 and they've sort of taken it for granted. Obviously, VMware is a big partner for both Microsoft and Intel, but the reality is is that in a hyper convergent environment, you need a file system, you need an operating system, and you need apps. And Microsoft has all that capability. As you'll hear at four o'clock, we announce world record numbers and it's spectacular. And the reason for it is in our last version of Windows Server 2016, we delivered 16 million IOP's in a hyper converged environment. That got Raph and his team off the table saying, okay, you guys are legitimate. You have a legitimate platform now. But it's not good enough. We think this new instantiation that we've already started to announce in Windows 2019, and Jeff Wolsey announced it earlier today and started talking about the features in Project Honolulu. We think those kind of transitions are what it's going to take for Enterprise customers to begin to break down those silos that you discussed, and really start to look at their data holistically, build data lakes that can scale, and build frameworks that are, I don't even want to use the term convergent anymore, but hyper scalable. >> Yeah, I mean, to tie into that, right. You look at what Intel has developed around Optane and some of the storage platforms that they've come out with. 10 years ago? Intel wasn't really known as a storage company, right? But, you look at all the storage vendors out there today, they really are putting Intel aside. And when you start looking at what Storage Spaces Direct is going to deliver and some of the robustness around Optane, we really think that it's going to be something our customers are going to embrace with Windows 2019 and future versions and sequels. >> So, Raph, I got to give Presidio a lot of credit, though. We launched a program called Intel Select Solutions, and it really allowed us to take Windows and Storage Spaces Direct and create a solution that included both the CPU, the networking, the SSD's and the memory. And Presidio has led that. And so because we have these Intel Select Solutions for Storage Spaces Direct with Presidio, we have the flexibility now to give customers package solutions that are pre-configured. >> Great. Well, Jake and Raphael, thank you so much for coming on theCUBE. It was great talking to you. >> Thank you very much. >> I'm Rebecca Knight, for Stu Miniman, we will have more of theCUBE's live coverage of Microsoft Ignite coming up just in a little bit. (light tehcno 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 Madrid everybody. This is the Cube, the leader in live tech coverage, and this is day two of HPE Discover 2017. I'm Dave Vellante with my co-host, Peter Burris, Alain Andreoli is here. He's the Senior Vice President and general manager of the hybrid IT group at HPE. Great to see you again. >> Great to see you David, great to see you Peter. >> So, a lot of good energy here, the story Alain is coming together. >> Alain: Yes. >> We've seen it over the last five years but really fine-tuned the organization and seems like things are going well. >> We have more clarity on our strategy than I've ever seen in a company, and this was not easy to do because the market is changing so fast. We addressing $120 billion market in hybrid IT, we lead the market in compute, we lead the market in storage, we lead the market with private cloud, we have invented composable, we are ramping up our Harper converge offering, and now on top of the infrastructure, we building these layers of one sphere, which is managing a multi-cloud environment for the data, and we are adjusting our services to become advisory and consumption models. This is having such an impact on our customers, 74 percent of our customers are going for hybrid IT journey. So we have organized ourselves to make this journey to be basically the partner of choice for our customers as they go through that. >> I mean so cloud of the last five, seven years, cloud and open-source software have really disrupted our industry. You've had to respond to that, and basically bringing cloud-like operating models to your customers. >> Alain: Yes. >> How have you done that, how do you rate your progress and where are you to date in that regard? >> So the first decision we had to make is are we a neutral party to our customers? (laughing) >> Dave: Yeah. >> We need to redo it. (laughing) >> They're getting you back, right? So, I don't know if you can see that, alright? Alain came by on his scooter, here we go, let's catch this. Here we go, this is called payback. (laughing) During Dr. Tom's interview, Alain came by with his scooter. (laughing) >> I will get you, I will get you for this. (laughing) >> It's great fun on the Cube. >> We can kid, that's alright. >> That's good. >> So the decision we had to make is are we the partner for our customers to go to the cloud or are we saying on PRIM is better? >> Dave: Yeah. >> And we 'vedecided to be this partner. Because we believe there is value for everyone and we believe it is not a one-way street. And we see actually that 32 percent of the customers who have moved work loads to the cloud are bringing these work loads back on PRIM. So we had to advise them. We helped them go through this journey, we really mean it, we helped them to go on Amazon, we helped them to go on Azure, we helped them to go on Google, and we helped them make it work, and this is why it's a service-led journey. The problem if you go on the public cloud is that we don't really know how much it is going to cost you, and you don't really have a single pane of glass to have all your data being managed across, what is now an ecosystem. We enabled them to do that. And the market we are directly addressing on PRIM is not shrinking. We still see huge pockets of growth, in flash storage, in HPC, you've seen the results we have in HPC. In Mission-Critical X86, in Hyperconvert, so we are basically moving from the one-size fits all type of organization of freeing X86 and start off storage, to become a company that offers value to customers, in specialized pools of compute, of storage, of networking, and offering them the end to end journey across the different stack. What I think is going to make a huge difference, if you look at the five-year horizon, is the growth of The Edge and the fact that 70 percent of the data are going to come from The Edge, and then you will really see the power of our strategy of private IT which goes from The Edge, to the core, to the cloud, because we will be able to enable our customers to have their data moving seamlessly across this journey. And we have exactly organized the company that way. >> One of the obvious use cases from what I like to call machine intelligence or artificial intelligence is really infusing artificial intelligence into infrastructure for predictive analytics and predictive maintenance, IT operations management, Infocyte, you got through an acquisition of Nimble and have been impressed with the pace at which you pushed that throughout the portfolio, I wondered if you could address that. >> We've been almost surprised. We looked at, we wanted to become the flash company because we saw that the market over three years, would completely move to flash. And when there is such a pendulum shift, you want to be at the forefront. >> Dave: Right. >> So we looked at all these companies who were having very strong positions on flash and Nimble intrigued us because they had, by far, when we talked to their customers, the highest customer satisfaction, I think it was something like 87 percent. >> The NPS is off the charts. >> The NPS is off the charts, right? And then we peeled the onion and we saw Infocyte, which was almost enough to start south because it was not part of our list, right? Initially of our list of this is how we are gonna select a company we want to acquire, and when we got into Infocyte, how it works, how we can actually port easily these to three power and then to SimpliVity and then to the rest of the portfolio we felt this is the crown jewel that is going to be the foundation of us making >> Dave: And not just the storage portfolio. >> No, end to end so we're gonna do these for everything, now we cannot do it in one day. The priority was to give a seamless experience to customers going three power or Nimble, so we've done that very quickly. We acquired the company six months ago and it's already there for three power. Next one will be Simplivity, very soon in a few weeks, then we go to the whole computes platform as well, then finally to networking. I hope, it's not a commitment, but I hope that by the end of next year, and under a year, we will be done for the whole infrastructure portfolio. >> And explain the benefit to customers. >> And then the benefit is that you basically have, you eliminate the need for level one and level two support because it's proactively, now you have to be wanting to have your device calling home, right? Because otherwise, if you want your device to be in the data center and insulated from communicating with the network effect, that is not going to work, so but assuming you want your device to be connected centrally, so that it can be monitored centrally the artificial intelligence that is embedded in Infocyte is basically going to monitor the behavior of your device compared with hundreds of thousands of other ones and therefore anything that is deviant will be flagged as a potential problem and resolved before you even know about it. That's one. So when you end up having a problem eventually, which is becoming very, very rare, then you directly call the level three engineer who is an expert and who has, on the screen, the behavior of your device for the last month compared to others, and the resolution is in less than a minute. So it's a revolution in the way to do service. >> So, one of the things that we've observed as we've talked to customers is that the characteristics of the problems that they're now trying to solve have real world elements, and that's really what The Edge is about in many respects. For the first 50 years of IT, we were doing accounting, and HR, and supply chain, and we were able to define what the data models looked like, we could therefore say, the data's going to be here, the processing is going to be here, we could build data centers. Now as you said, 70 percent of the data is going to be coming from The Edge. It's not clear, necessarily where the best place to process that data is. Where's the compute going to be? How's it going to integrate with people? In many respects, hybrid IT is about diminishing the degree to which infrastructure dictates the way the problem gets solved. Would you agree with that? It's kind of like where does, let the data reside where it needs to reside, and make sure that the business is a natural infrastructure that reflects and corresponds to the work that needs to get done. >> I totally agree with your problem statement, and the way you position the question. In terms of semantics, I would just say we need to make infrastructure invisible. It's still there because it's all running on infrastructure. The iPhone is infrastructure, your PC is infrastructure, your camera is infrastructure, it's all there. >> A C.I.O said to me not too long ago... >> But you know what? We are having this interview, we are not thinking about what makes it happen. >> Peter: Right, right, right. >> Our business is to talk and communicate right now, this all has got to be seamless and that's how we need to make IT, seamless. >> I had a conversation with a C.I.O. >> Invisible. >> Yeah, who said that the value of my infrastructure is inversely proportional to the degree to which anybody knows anything about it. So, is that kind of what the HP promise is, is we're gonna let the data and the work loads define where the infrastructure goes and ensure we have those options? >> It's exactly right and the vehicle to do that, we call it autonomous data centers. Your phone is a data center. Your data center is a data center. Your off-frame cloud is a data center that you are subcontracting, right? So we want all of these to be autonomous, in terms of self-healing and everything else, and then the intelligence of where these data are being moved and how you use what and when is the single pane of glass that we are developing around one sphere. And how to get the customers to move their work loads and their business around that is what we do with point next with services. This is our strategy. >> So let me break that down a little bit. So, we've got devices that are powerful enough that we could put new types of control, new types of work loads there if we wanted to, we've got now the ability to package infrastructure, and have a single pane of glass, and have a common management framework. >> Right. >> But when you say the autonomous data center, it's we have a common business approach thinking about policy, thinking about value, thinking about how we're gonna do things, and we can put that into this entire vision, and let it actually execute how that manifests itself from a business standpoint. >> Exactly right. >> Have I got that right? >> It's exactly right. I love the way you put it. That's exactly what we are trying to do. it's not going to be done in one day, but that is our strategy, and we have organized, once again, the whole company around it, to execute this strategy and to make it happen for our customers. >> So if we think about what an HPE customer is gonna look like in, you know a really good HPE customer in 2023, what.. >> Alain: That's a long time. >> That's, five years, but I'm giving you that much run way, because you're right, it's not there yet and if it's too ambitious then so be it, but how is a business person going to think differently about working, about the role that IT is going to play in the business, and what it means to have a great partnership with a company like HP? >> Yeah, so we are basically, our motto is One size doesn't fit all, so we are first trying to understand the business of the customer, and then we will apply solutions to enhance this business, or to empower this business, right? So, we have the biggest brace of infrastructure that you can think of, think about this infrastructure becoming self-healing, but this infrastructure is more and more specialized, there is HPC, there is Mission-Critical, we just found Superdome flex, or SAP, we have all these specializations that, for those customers to optimize their business outcome. Then we have the single pane of glass that allows everything to seamlessly operate the data around, and then our point-neck services are going to work with the customers to architect their IT model in a way that their work loads are optimized. And one of the key is the right mix. The right mix of what you do yourself, what you got from multi-cloud, how much do you pay for it, how much do you anticipate that you're gonna pay for it, do you want this to be CAPEX, do you want this to be OPEX? And then how do you manage The Edge, and with Aruba and with Edgeline, and then with all your IT platforms that can manage the data across The Edge. We have the capability to also let the customer decide, do I want a lot of analytics and decisions to be made at The Edge, in my devices, and this is highly valuable depending on what customer business model we are talking about, or, do I want all the data from the analog world through the censors to come straight back to the ranch. All of these decisions, we are gonna have platforms to allow customers to make these decisions, to decide, kind of templates if you want, this is how I want it to run, and to be executed, and then to be automatically, autonomously operated. That's our vision of how we can help our customers moving forward. >> Last question, so the attendees of Discover, your customers, when they go back and he or she talks to their boss, what do you want them to say about Discover 2018? >> I invested two or three days of my time to come to HPE Discover. It was really exciting because I felt that it's like a new company, it's the company I know. I know they are customer first and customer last, and they are the ones who help me when I have a problem, whether they created it or not, they are here to help me. This is not going away, but they are taking us to the new world. They are gonna help us to build our hybrid IT model, and I think we need to trust them to have a seat at the table when we make these decisions, boss. >> Intimacy, innovation... >> Alain: Yeah, innovation. >> Trust. >> HPE's no longer wandering in the desert. (laughing) >> Alain Andreoli thanks so much for coming on the Cube, it is always a pleasure. >> It was a pleasure. Take care, thanks Peter. >> Keep it right there, everybody, Peter and I will be back with our next guest, right after this short break, we're live from Madrid. You're watching the Cube. (techno music)

live from the Fairmont Hotel in San Jose California it's the queue at big data sv 2015 hello everyone welcome back this is the cube our flagship program we go out to the events they strike this evil noise i'm john furrier we're here with IBM to talk about big data big data analytics and we're doing a first-ever crowd chat simulcast of live feed with IBM so guys we're going to try this out it's like go to crouch at dan / Hadoop next and join the conversation and our guests here Rob Thomas vice president product development big data analyst at IBM and beth smith general manager of IBM analytics platform guys welcome to welcome to the cube thank you welcome back and so IBM mostly we're super excited to next week as I was the interconnect you're bigger than you guys mashed up three shows for the mega shows and and Aerosmith's playing so it's going to say I'm from the Boston air so I'm really excited about you know Aerosmith and all the activities of social lounge and and whatnot but we've been following you guys the transformation of IBM is really impressive you guys certainly think a lot of heat in the press in terms of some of the performance size in the business but it's pumping right now you guys seem to have great positioning the stories are hanging together a huge customer base huge services so we're at the Big Data world which is tends to be startup driven from the past few years over the past phase one the big cuppies came in and started saying hey you know there's a big market our customers see demand and that so I got your take on on as we're coming in to interconnect next next week what is the perspective of big data asli Watson has garnered headlines from powering toys to jeopardy to solving huge world problems that's a big data problem you guys are not new to Big Data so when you look at this big data week here and Silicon Valley what's the take sure so I'll start often embedded Bethke night in so our big focus is how we start to bring data to the masses and we start to think in terms of personas data science and plays an increasingly important role around big data how people are accessing that the developer community and then obviously the line of business community which is the client set that I've been serving four years but the announcements that we've made this week around Hadoop are really focused on the first two personas in terms of data scientists how they start to get better value out of Hadoop leveraging different tools we'll talk about what some of those are and so we're really starting to change it about Hadoop results me about insight it's not about infrastructure infrastructure is interesting but it's really about what you're getting out of it so that's why we're approaching it that way it's how well it has naturally the IBM strategy around data cloud and engagement and data is really about using the insights which like Rob said it's about the value can get from the data and how that can be used in to transform professions and industries and I think when we bring it back to Big Data and the topic of a doob I think frankly it has gotten to a point that clients are really beginning to say it's time to scale they're seeing the value in the technology what it can bring how it gives them some diversity in their data and analytics platform and they're ready to announce scale on their workloads as a part of it so the theme is Hadoop next okay so that takes us right to the next point which is okay what's next is a phase one okay we got some base position validation okay this new environments customers don't want that so what so what is next i mean we're earring things like in memories hot aussie spark has proven that there's an action in member that that kind of says okay analytics at the speed of business is something that's important you guys are all over that and we've heard some things from you guys so so what's how do we get to the next part where we take Hadoop as an infrastructure opportunity and put it into practice for solutions at what what are the key things that you guys see happening that must happen for the large customers to be successful so I think that actually ties into the announcements we made this week around the open data platform because that's about getting that core platform to ensure that their standardization around it there's interoperability around it and then that's the base and that vendors and clients are coming together do that and to really enable and facilitate the community to be able to standardize around that then it's about the value on top of that around it etc it's about the workloads and what could be brought to bear to extend up that how do you apply it to real time streaming how do you add things like machine learning how do you deal with things like text analytics I mean we have a we have a client situation where the client took 4 billion tweets and were able to analyze that to identify over a hundred and ten million profiles of individuals and then by integrating and analyzing that data with the internal data sources of about seven or eight different data sources they were able to narrow into 1.7 million profiles that matched at at least ninety percent precision you know now they've got data that they can apply on buying patterns and stuff it's about that it's about going up the stack we're going to talk for hours my mind's exploding privacy creepy I mean a personas is relevant now you talk about personalization I mean collective intelligence has been an AI concepts we try not to be creepy okay cool but now so that brings us to the next level I mean you guys were talk about cognitives on that is a word you guys kick around also systems of engagement systems of records an old term that's been around in the old data warehousing dates fenced-off resources of disk and data but now with systems of engagement real-time in the moment immersive experience which is essentially the social and/or kind of mobile experience what does that mean how do you guys get there how do you make it so it's better for the users more secure or I mean these are hot button issues that kind of lead us right to that point so I'll take you to that a couple ways so so first of all your first question round head tube next so Hadoop was no longer just an IT discussion that's what I've seen changed dramatically in the last six months I was with the CEO of one of the world's largest banks just three days ago and the CEO is asking about Hadoop so there's a great interest in this topic and so so why so why would a CEO even care I think one is people are starting to understand the use cases of the place so that talks about entity extraction so how you start to look at customer records that you have internally in your systems are record to your point John and then you you know how do you match that against what's happening in the social world which is more or the engagement piece so there's a clear use case around that that changes how clients you know work with their with their customers so so that's one reason second is huge momentum in this idea of a logical data warehouse we no longer think of the data infrastructure as oh it's a warehouse or it's a database physically tied to something not tied to just what relational store so you can have a warehouse but you can scale in Hadoop you can provision data back and forth you can write queries from either side that's what we're doing is we're enabling clients to modernize their infrastructure with this type of a logit logical data warehouse approach when you take those kinds of use cases and then you put the data science tools on top of it suddenly our customers can develop a different relationship with their customers and they can really start to change the way that they're doing business Beth I want to get your comments we have the Crouch at crowd chat / a dupe next some commentary coming in ousley transforming industries billion tweets killer for customer experience so customer experience and then also the link about the data science into high gear so let's bring that now into the data science so the logical you know stores okay Nick sands with virtualization things are moving around you have some sort of cognitive engines out there that can overlay on top of that customer experience and data science how are they inter playing because this came out on some of the retail event at New York City that happened last week good point of purchase personalization customer experience hated science it's all rolling together and what does that mean unpack that for us and simplify it if you can oh wows complexing is a big topic you know it's a big topic so a couple of different points so first of all I think it is about enabling the data scientists to be able to do what they their specialty is and the technologies have advanced to allow them to do that and then it's about them having the the data and the different forms of data and the analytics at their fingertips to be able to apply that I the other point in it though is that the lines are blurring between the person that is the data scientist and the business user that needs to worry about how do they attract new customers or how do they you know create new business models and what do they use as a part of do you think we're also seeing that line blurring one of the things that we're trying to do is is help the industry around growing skills so we actually have big data University we have what two hundred and thirty thousand participants and this online free education and we're expanding that topic now to again go up the stack to go into the things that data scientists want to deal with like machine learning to go into things that the business user really wants to now be able to capture it's a part of it trying to ask you guys kind of more could be a product question and/or kind of a market question at IBM's Ted at IBM event in he talked about a big medical example in one of her favorite use cases but she made a comment in their active data active date is not a new term for the data geeks out there but we look at data science lag is really important Realty near real time is not going to make it for airplanes and people crossing the street with mobile devices so real real time means like that second latency is really important speed so active date is a big part of that so can you guys talk about passive active data and how that relates to computing and because it's all kind of coming to get it's not an obvious thing but she highlighted that in her presentation because I see with medical medical care is obviously urgent you know in the moment kind of thing so if you would what does that all mean I mean is that something custom Street paying attention to is it viable is it doable so certainly a viable I mean it's a huge opportunity and i'd say probably most famous story we have around that is the work that we did at the university of toronto at the Hospital for Sick Children where we were using real-time streaming algorithms and a real-time streaming engine to monitor instance in the neonatal care facility and this was a million data points coming off of a human body monitoring in real time and so why is that relevant I mean it's pretty pretty basic actually if you extract the data you eat yell it somewhere you load in a warehouse then you start to say well what's going on it's way too late you know we're talking about you know at the moment you need to know what's happening and so it started as a lot was in the medical field would you notice there's some examples that you mentioned but real time is now going well beyond the medical field you know places from retail at the point of sale and how things are happening to even things like farming so real time is here to stay we don't really view that as different from what I would describe as Hadoop next because streaming to me as part of what we're doing with a dupe and with spark which we'll talk about in a bit so it's certainly it is it is the new paradigm for many clients but it's going to be much more common actually if i can add there's a client North Carolina State University it's where I went to school so it's a if it's a client that I talk about a lot but they in addition to what they do with their students they also work with a lot of businesses own different opportunities that may that they may have and they have a big data and analytics sort of extended education business education project as a part of that they are now prepared to be able to analyze one petabyte in near real time so the examples that you and Rob talked about of the real world workloads that are going to exist where real time matters are there there's no doubt about it they're not going away and the technology is prepared to be able to handle the massive amount of data and analytics that needs to happen right there in real time you know that's a great exact point I mean these flagship examples are kind of like lighthouses for people to look at and kind of the ships that kind of come into the harbor if you will for other customers as you always have the early adopters can you guys talk about where the mainstream market is right now I'll see from a services standpoint you guys have great presence and a lot of accounts where are these ships coming into which Harper where the lighthouse is actually medical you mentioned some of those examples are bringing in the main customers is it the new apps that are driving it what innovations and what are the forces and what are the customers doing in the main stream right now where are they in the evolution of moving to these kind of higher-end examples so I mean so Hadoop I'd say this is the year Hadoop where clients have become serious about Hadoop like I said it's now become a board-level topic so it's it's at the forefront right now I see clients being very aggressive about trying out new use cases everybody really across every interest industry is looking for one thing which is growth and the way that you get growth if you're a bank is you're not really going to change your asset structure what you're going to change is how you engage with clients and how you personalized offers if your retailer you're not going to grow by simply adding more stores it might be a short term growth impact but you're going to change how you're engaging with clients and so these use cases are very real and they're happening now Hadoop is a bore group discussion or big day I just didn't see you formula we should have more Hadoop or is it you know I see I've seen it over and over again I'll tell you where you see a lot from his companies that are private equity-owned the private equity guys have figured out that there's savings and there's innovation here every company i worked with that has private equity ownership Hadoop is a boardroom discussion and the idea is how do we modernize the infrastructure because it's it's because of other forces though it's because of mobile it's because of cloud that comes to the forefront so absolutely so let's take Hadoop so I do bits great bad just great a lot of innovations going on there boardroom in these private equity because one they're cutting edge probably they're like an investment they want to see I realized pretty quickly now speed is critical right I would infer that was coming from the private equity side speed is critical right so speed to value what does that mean for ibn and your customers how do you guys deliver the speed to value is that's one of the things that comes out on all the premises of all the conversations is hey you can do things faster now so value on the business side what do you guys see that sure so a a lot of different ways to approach that so we believe that as I said when I said before it's not just about the infrastructure it's about the insight we've built a lot of analytic capabilities into what we're doing around a dupe and spark so that clients can get the answers faster so one thing that we're going to be we have a session here at strata this week talking about our new innovation big R which is our our algorithms which are the only our algorithms that you can run natively on Hadoop where your statistical programmers can suddenly start to you know analyze data and you know drive that to decision make it as an example so we believe that by providing the analytics on top of the infrastructure you can you can change how clients are getting value out of that so how do we do it quickly we've got IBM SoftLayer so we've got our Hadoop infrastructure up on the cloud so anybody can go provision something and get started and ours which is not something that was the case even a couple years ago and so speed is important but the tools and how you get the insight is equally important how about speed 22 value from a customer deployment standpoint is it the apps or is it innovating on existing what do you sing well I think it's both actually um and and so you talked earlier about system of engagement vs system of record you know and I think at the end of the day for clients is really about systems of insight which is some combination of that right we tend to thank the systems of engagement or the newer things and the newer applications and we tend to thank the systems of record are the older ones but I think it's a combination of it and we see it show up in different ways so I'll take an example of telco and we have a solution on the now factory and this is now about applying analytics in real time about the network and the dynamics so that for example the operator has a better view of what's happening for their customers they're in users and they can tell that an application has gone down and that customers have now switched all of a sudden using a competitive application on their mobile devices you know that's different and that is that new applications or old or is it the combination and I think at the end of the day it really comes to a combination I love these systems of insight i'm just going to write that down here inside the inside the crowd chat so i got to talk about the the holy grail for big data analytics and big data from your perspective ideas perspective and to where you guys are partnering I'll see here there's a show of rich targets of a queue hires acquisitions partnerships I mean it's really a frill ground certainly Silicon Valley and and in the growth of a big data cloud mobile and social kind of these infrared photography biz is a message we've heard so what is the holy grail and then what are you guys looking for in partnerships and within the community of startups and or other alliances sure you want to start with the Holy Grail me yeah so so you know I think at the end of the day it is about using technology for business value and business outcome I you know I really think that's what said the spirit of it and so if I tell you why we have for example increased our attention and investment around this topic it's because of that it's because of what Rob said earlier when he said the state that clients are now in um so that's what I think is really important there and I think it's only going to be successful if it's done based own standards and something that is in support of you know heterogeneous environments I mean that's the world of technology that we live in and that's a critical element of it which leads to why we are a part of the Open Data Platform initiative so on the on the the piece of analytics I was just cus our comment about our for example I was just mentioning the crowd chat I had Microsoft just revolution analytics which is not our which is different community is there a land-grab going on between the big guys of you know IBM's a big company what do you guys see in that kind of area terms acquisition targets yeah man I think the numbers would say there's not a land-grab I don't think the MMA numbers have changed at a macro level at all in the last couple years I mean we're very opportunistic in our strategy right we look for things that augment what we do I think you know it's related to partner on your comment your question on partnering but we do acquisitions is not only about what that company does but it's about how does it fit within what IBM already does because we're trying to you know we're going after a rising tide in terms of how we deliver what clients need I think some companies make that mistake they think that if they have a great product that's relevant to us maybe maybe not but it's about how it fits in what we're doing and that's how we look at all of our partnerships really and you know we partner with global systems integrators even though we have one with an IBM we partner with ISVs application developers the big push this week as I described before is around data scientists so we're rolling out data science education on Big Data university because we think that data scientists will quickly find that the best place to do that is on an IBM platform because it's the best tools and if they can provide better insight to their companies or to their clients they're going to be better off so I was so yes that was the commenting on and certainly the end of last week and earlier this week about that Twitter and it's a lot of common in Twitter's figured out and people are confused by Twitter versus facebook and I know IBM has a relation but we're so just that's why pops in my head and I was are saying HP Buddha's got a great value and so I was on the side of Twitter's a winner i love twitter i love the company misunderstood certainly i think in this market where there's waves coming in more and more there's a lot of misunderstanding and i think i want to get your perspective you can share with the folks out there what is that next way because it's confusing out there you guys are insiders IBM i would say like twitter is winning doing very well certainly we're close to you guys we are we're deeply reporting on IBM so we can see the momentum and the positioning it's all in line what we see is that is where the outcomes will end up being for customers but there's still a lot of naysayers out there certainly you guys had your share as as to where's as an example so what is the big misunderstanding that you think is out there around the market we're in and what's the next wave as always waves coming in if you're not out in front that next wave usually driftwood as the old expression goes so what is that big misunderstanding and this kind of converged from a hyper targeted with analytics this is all new stuff huge opportunities huge shifts and inflection point as Bob picciano said on the cube is its kind of both going on the same time shift and it point so what's misunderstood and what's that next big waves so let me start with the next big way is that I'll back into the misunderstanding so the next big wave to me is machine learning and how do you start to take the data assets that you have and through machine learning and the application of those type of algorithms you start to generate better insights or outcomes and the reason i think is the next big wave is it's it may be one of the last competitive motes out there if you think about it if you have a a corpus of data that's unique to you and you can practice machine learning on that and have that you know either data that you can sell or to feed into your core business that's something that nobody else can replicate so it becomes incredibly powerful so one example I'll share with you and I want to bring you my book but it's actually not getting published next week since so maybe next week but so Wiley's publishing a book I wrote and one of the examples I give is a company by the name of co-star which I think very few people have heard of co-star is in the commercial real estate business they weren't even around a decade ago they have skyrocketed you know from zero to five hundred million dollars in revenue and it's because they have data on four million commercial properties out there who else has that absolutely nobody has that kind of reach and so they've got a unique data asset they can apply things like machine learning and statistics to that and therefore anybody who wants to do anything commercial real estate has to start with them so I pointed you're starting to get the point where you have some businesses where data is the product it's not an enabler it's the actual product I think that's probably one of the big misunderstandings out there is that you know data is just something that serves our existing products or existing services we're moving to a world where data is the product and that's the moat I wrote a post in 2008 called data is the new development kit and what you're basically saying is that's the competitive advantage a business user can make any innovation observation about data and not be a scientist and change the game that's what you were saying earlier similar right that's right okay so next big wave misunderstanding what do you wait bet what's your take on what are people not getting what is Wall Street what is potential the VCG really on the front end of some of the innovation but what is the general public not getting I mean we are in shift and an inflection what's it what's the big shift and misunderstanding going on so so I I would tend to you know actually agree with with Rob that I think folks aren't yet really appreciating and I guess I would twist it a little bit and say the insight instead of just the data but but they're not realizing what that is and what it's going to give us the opportunity for you know I would retire early if I actually could predict everything that was going to happen but but you know yeah but if you think about it you know if you think about you know mid to late 90s and what we would have all fault that the internet was going to allow us to do compared to what it actually allowed us to do is probably like night and day and I think the the time we're in now when you take data and you take mobility and you take cloud and you take these systems of engagement and the fact the way people individuals actually want to do things is is similar but almost like on steroids to what we were dealing with in the mid-90s or so and so you know the possibilities are frankly endless and and I think that's part of what people aren't necessarily realizing is that they have to think about that insight that data that actually has some value to it in very different ways there's a lot of disruptive enablers out Dunham's there's a lot to look at but finding which ones will be the biggest right it's hard I mean you get paid a lot of money to do that is if you can figure it out and keep it a secret um but you didn't you machine learning is now out there you just shared with us out competitive advantage so everyone knows know everyone kind of new kind of in the inside but but not everybody's using it right i mean i think another example a company like into it has done a great job of they started off as a software company they've become a data company i think what you what i've observed in all these companies is you can build a business model that's effectively recession proof because data becomes the IP in the organization and so I don't I actually you know I think for us those are the live in the world we this is well understood I don't think it's that well understood yet yeah insiders mic right and you know when we first started doing big data research and working with thousands of clients around the world there were there were six basic use cases it started of course with the customer the the end customer and the customer 360 and that sort of thing and went through a number of different things around optimization etc but the additional one is about those new business models and you know that is clearly in the last 12 to 18 months has become a lot more of what the topic is when I'm talking to clients and I think we will see that expand even more as we go in the future we've a lot of activity on the crowd chatter crowd chatter net / Hadoop necks and I'll mentioned we can probably extend time on that if you guys want to keep it keep it going conversation is awesome and we did getting the hook here so we'll remove the conversation to crouch at totnes Esther Dube next great thought leadership and I can go on this stuff for an hour you guys are awesome great to have you on the cube and so much to talk about a lot of ground will certainly see it in to connect go final question for you guys is what do you guys see for this week real quick summarize what do you expect to see it unfold for a big data week here at Silicon Valley Big Data asked me so I think you know a lot of the what we talked about machine learning is going to be a big topic I think there'll be a lot of discussion around the open data platform that Beth mentioned before it's a big move that we made along with another group supporting the apache software foundation I think that that's a big thing for this week but it should be exciting alright guys thanks for coming out to be IBM here inside the cube we're live in Silicon Valley would be right back with our next guest after the strip break I'm Jennifer this is the cube we write back