(upbeat music) >> Welcome everyone to theCUBE's presentation of the AWS Startup Showcase Open Cloud Innovations. This is season two of the ongoing series. We're covering exciting start startups in the AWS ecosystem to talk about open source community stuff. I'm your host, Dave Nicholson. And I'm delighted today to have two guests from Weaveworks. Steve George, COO of Weaveworks, and Steve Waterworth, technical marketing engineer from Weaveworks. Welcome, gentlemen, how are you? >> Very well, thanks. >> Very well, thanks very much. >> So, Steve G., what's the relationship with AWS? This is the AWS Startup Showcase. How do Weaveworks and AWS interact? >> Yeah sure. So, AWS is a investor in Weaveworks. And we, actually, collaborate really closely around EKS and some specific EKS tooling. So, in the early days of Kubernetes when AWS was working on EKS, the Elastic Kubernetes Service, we started working on the command line interface for EKS itself. And due to that partnership, we've been working closely with the EKS team for a long period of time, helping them to build the CLI and make sure that users in the community find EKS really easy to use. And so that brought us together with the AWS team, working on GitOps and thinking about how to deploy applications and clusters using this GitOps approach. And we've built that into the EKS CLI, which is an open source tool, is a project on GitHub. So, everybody can get involved with that, use it, contribute to it. We love hearing user feedback about how to help teams take advantage of the elastic nature of Kubernetes as simply and easily as possible. >> Well, it's great to have you. Before we get into the specifics around what Weaveworks is doing in this area that we're about to discuss, let's talk about this concept of GitOps. Some of us may have gotten too deep into a Netflix series, and we didn't realize that we've moved on from the world of DevOps or DevSecOps and the like. Explain where GitOps fits into this evolution. >> Yeah, sure. So, really GitOps is an instantiation, a version of DevOps. And it fits within the idea that, particularly in the Kubernetes world, we have a model in Kubernetes, which tells us exactly what we want to deploy. And so what we're talking about is using Git as a way of recording what we want to be in the runtime environment, and then telling Kubernetes from the configuration that is stored in Git exactly what we want to deploy. So, in a sense, it's very much aligned with DevOps, because we know we want to bring teams together, help them to deploy their applications, their clusters, their environments. And really with GitOps, we have a specific set of tools that we can use. And obviously what's nice about Git is it's a very developer tool, or lots and lots of developers use it, the vast majority. And so what we're trying to do is bring those operational processes into the way that developers work. So, really bringing DevOps to that generation through that specific tooling. >> So Steve G., let's continue down this thread a little bit. Why is it necessary then this sort of added wrinkle? If right now in my organization we have developers, who consider themselves to be DevOps folks, and we give them Amazon gift cards each month. And we say, "Hey, it's a world of serverless, "no code, low code lights out data centers. "Go out and deploy your code. "Everything should be fine." What's the problem with that model, and how does GitOps come in and address that? >> Right. I think there's a couple of things. So, for individual developers, one of the big challenges is that, when you watch development teams, like deploying applications and running them, you watch them switching between all those different tabs, and services, and systems that they're using. So, GitOps has a real advantage to developers, because they're already sat in Git, they're already using their familiar tooling. And so by bringing operations within that developer tooling, you're giving them that familiarity. So, it's one advantage for developers. And then for operations staff, one of the things that it does is it centralizes where all of this configuration is kept. And then you can use things like templating and some other things that we're going to be talking about today to make sure that you automate and go quickly, but you also do that in a way which is reliable, and secure, and stable. So, it's really helping to bring that run fast, but don't break things kind of ethos to how we can deploy and run applications in the cloud. >> So, Steve W., let's start talking about where Weaveworks comes into the picture, and what's your perspective. >> So, yeah, Weaveworks has an engine, a set of software, that enables this to happen. So, think of it as a constant reconciliation engine. So, you've got your declared state, your desired state is declared in Git. So, this is where all your YAML for all your Kubernetes hangs out. And then you have an agent that's running inside Kubernetes, that's the Weaveworks GitOps agent. And it's constantly comparing the desired state in Git with the actual state, which is what's running in Kubernetes. So, then as a developer, you want to make a change, or an operator, you want to make a change. You push a change into Git. The reconciliation loop runs and says, "All right, what we've got in Git does not match "what we've got in Kubernetes. "Therefore, I will create story resource, whatever." But it also works the other way. So, if someone does directly access Kubernetes and make a change, then the next time that reconciliation loop runs, it's automatically reverted back to that single source of truth in Git. So, your Kubernetes cluster, you don't get any configuration drift. It's always configured as you desire it to be configured. And as Steve George has already said, from a developer or engineer point of view, it's easy to use. They're just using Git just as they always have done and continue to do. There's nothing new to learn. No change to working practices. I just push code into Git, magic happens. >> So, Steve W., little deeper dive on that. When we hear Ops, a lot of us start thinking about, specifically in terms of infrastructure, and especially since infrastructure when deployed and left out there, even though it's really idle, you're paying for it. So, anytime there's an Ops component to the discussion, cost and resource management come into play. You mentioned this idea of not letting things drift from a template. What are those templates based on? Are they based on... Is this primarily an infrastructure discussion, or are we talking about the code itself that is outside of the infrastructure discussion? >> It's predominantly around the infrastructure. So, what you're managing in Git, as far as Kubernetes is concerned, is always deployment files, and services, and horizontal pod autoscalers, all those Kubernetes entities. Typically, the source code for your application, be it in Java, Node.js, whatever it is you happen to be writing it in, that's, typically, in a separate repository. You, typically, don't combine the two. So, you've got one set of repository, basically, for building your containers, and your CLI will run off that, and ultimately push a container into a registry somewhere. Then you have a separate repo, which is your config. repo, which declares what version of the containers you're going to run, how many you're going to run, how the services are bound to those containers, et cetera. >> Yeah, that makes sense. Steve G., talk to us about this concept of trusted application delivery with GitOps, and frankly, it's what led to the sort of prior question. When you think about trusted application delivery, where is that intertwinement between what we think of as the application code versus the code that is creating the infrastructure? So, what is trusted application delivery? >> Sure, so, with GitOps, we have the ability to deploy the infrastructure components. And then we also define what the application containers are, that would go to be deployed into that environment. And so, this is a really interesting question, because some teams will associate all of the services that an application needs within an application team. And sometimes teams will deploy sort of horizontal infrastructure, which then all application teams services take advantage of. Either way, you can define that within your configuration, within your GitOps configuration. Now, when you start deploying speed, particularly when you have multiple different teams doing these sorts of deployments, one of the questions that starts to come up will be from the security team, or someone who's thinking about, well, what happens if we make a deployment, which is accidentally incorrect, or if there is a security issue in one of those dependencies, and we need to get a new version deployed as quickly as possible? And so, in the GitOps pipeline, one of the things that we can do is to put in various checkpoints to check that the policy is being followed correctly. So, are we deploying the right number of applications, the right configuration of an application? Does that application follow certain standards that the enterprise has set down? And that's what we talk about when we talk about trusted policy and trusted delivery. Because really what we're thinking about here is enabling the development teams to go as quickly as possible with their new deployments, but protecting them with automated guard rails. So, making sure that they can go fast, but they are not going to do anything which destroys the reliability of the application platform. >> Yeah, you've mentioned reliability and kind of alluded to scalability in the application environment. What about looking at this from the security perspective? There've been some recently, pretty well publicized breaches. Not a lot of senior executives in enterprises understand that a very high percentage of code that their businesses are running on is coming out of the open source community, where developers and maintainers are, to a certain degree, what they would consider to be volunteers. That can be a scary thing. So, talk about why an enterprise struggles today with security, policy, and governance. And I toss this out to Steve W. Or Steve George. Answer appropriately. >> I'll try that in a high level, and Steve W. can give more of the technical detail. I mean, I'll say that when I talk to enterprise customers, there's two areas of concern. One area of concern is that, we're in an environment with DevOps where we started this conversation of trying to help teams to go as quickly as possible. But there's many instances where teams accidentally do things, but, nonetheless, that is a security issue. They deploy something manually into an environment, they forget about it, and that's something which is wrong. So, helping with this kind of policy as code pipeline, ensuring that everything goes through a set of standards could really help teams. And that's why we call it developer guard rails, because this is about helping the development team by providing automation around the outside, that helps them to go faster and relieves them from that mental concern of have they made any mistakes or errors. So, that's one form. And then the other form is the form, where you are going, David, which is really around security dependencies within software, a whole supply chain of concern. And what we can do there, by, again, having a set of standard scanners and policy checking, which ensures that everything is checked before it goes into the environment. That really helps to make sure that there are no security issues in the runtime deployment. Steve W., anything that I missed there? >> Yeah, well, I'll just say, I'll just go a little deeper on the technology bit. So, essentially, we have a library of policies, which get you started. Of course, you can modify those policies, write your own. The library is there just to get you going. So, as a change is made, typically, via, say, a GitHub action, the policy engine then kicks in and checks all those deployment files, all those YAML for Kubernetes, and looks for things that then are outside of policy. And if that's the case, then the action will fail, and that'll show up on the pull request. So, things like, are your containers coming from trusted sources? You're not just pulling in some random container from a public registry. You're actually using a trusted registry. Things like, are containers running as route, or are they running in privileged mode, which, again, it could be a security? But it's not just about security, it can also be about coding standards. Are the containers correctly annotated? Is the deployment correctly annotated? Does it have the annotation fields that we require for our coding standards? And it can also be about reliability. Does the deployment script have the health checks defined? Does it have a suitable replica account? So, a rolling update. We'll actually do a rolling update. You can't do a rolling update with only one replica. So, you can have all these sorts of checks and guards in there. And then finally, there's an admission controller that runs inside Kubernetes. So, if someone does try and squeeze through, and do something a little naughty, and go directly to the cluster, it's not going to happen, 'cause that admission controller is going to say, "Hey, no, that's a policy violation. "I'm not letting that in." So, it really just stops. It stops developers making mistakes. I know, I know, I've done development, and I've deployed things into Kubernetes, and haven't got the conflict quite right, and then it falls flat on its face. And you're sitting there scratching your head. And with the policy checks, then that wouldn't happen. 'Cause you would try and put something in that has a slightly iffy configuration, and it would spit it straight back out at you. >> So, obviously you have some sort of policy engine that you're you're relying on. But what is the user experience like? I mean, is this a screen that is reminiscent of the matrix with non-readable characters streaming down that only another machine can understand? What does this look like to the operator? >> Yeah, sure, so, we have a console, a web console, where developers and operators can use a set of predefined policies. And so that's the starting point. And we have a set of recommendations there and policies that you can just attach to your deployments. So, set of recommendations about different AWS resources, deployment types, EKS deployment types, different sets of standards that your enterprise might be following along with. So, that's one way of doing it. And then you can take those policies and start customizing them to your needs. And by using GitOps, what we're aiming for here is to bring both the application configuration, the environment configuration. We talked about this earlier, all of this being within Git. We're adding these policies within Git as well. So, for advanced users, they'll have everything that they need together in a single unit of change, your application, your definitions of how you want to run this application service, and the policies that you want it to follow, all together in Git. And then when there is some sort of policy violation on the other end of the pipeline, people can see where this policy is being violated, how it was violated. And then for a set of those, we try and automate by showing a pull request for the user about how they can fix this policy violation. So, try and make it as simple as possible. Because in many of these sorts of violations, if you're a busy developer, there'll be minor configuration details going against the configuration, and you just want to fix those really quickly. >> So Steve W., is that what the Mega Leaks policy engine is? >> Yes, that's the Mega Leaks policy engine. So, yes, it's a SaaS-based service that holds the actual policy engine and your library of policies. So, when your GitHub action runs, it goes and essentially makes a call across with the configuration and does the check and spits out any violation errors, if there are any. >> So, folks in this community really like to try things before they deploy them. Is there an opportunity for people to get a demo of this, get their hands on it? what's the best way to do that? >> The best way to do it is have a play with it. As an engineer, I just love getting my hands dirty with these sorts of things. So, yeah, you can go to the Mega Leaks website and get a 30-day free trial. You can spin yourself up a little, test cluster, and have a play. >> So, what's coming next? We had DevOps, and then DevSecOps, and now GitOps. What's next? Are we going to go back to all infrastructure on premises all the time, back to waterfall? Back to waterfall, "Hot Tub Time Machine?" What's the prediction? >> Well, I think the thing that you set out right at the start, actually, is the prediction. The difference between infrastructure and applications is steadily going away, as we try and be more dynamic in the way that we deploy. And for us with GitOps, I think we're... When we talk about operations, there's a lots of depth to what we mean about operations. So, I think there's lots of areas to explore how to bring operations into developer tooling with GitOps. So, that's, I think, certainly where Weaveworks will be focusing. >> Well, as an old infrastructure guy myself, I see this as vindication. Because infrastructure still matters, kids. And we need sophisticated ways to make sure that the proper infrastructure is applied. People are shocked to learn that even serverless application environments involve servers. So, I tell my 14-year-old son this regularly, he doesn't believe it, but it is what it is. Steve W., any final thoughts on this whole move towards GitOps and, specifically, the Weaveworks secret sauce and superpower. >> Yeah. It's all about (indistinct)... It's all about going as quickly as possible, but without tripping up. Being able to run fast, but without tripping over your shoe laces, which you forgot to tie up. And that's what the automation brings. It allows you to go quickly, does lots of things for you, and yeah, we try and stop you shooting yourself in the foot as you're going. >> Well, it's been fantastic talking to both of you today. For the audience's sake, I'm in California, and we have a gentleman in France, and a gentlemen in the UK. It's just the wonders of modern technology never cease. Thanks, again, Steve Waterworth, Steve George from Weaveworks. Thanks for coming on theCUBE for the AWS Startup Showcase. And to the rest of us, keep it right here for more action on theCUBE, your leader in tech coverage. (upbeat music)

(upbeat music) >> Okay, Chloe, thank you very much. Hey folks, in here in the Cloud City We with Danielle Royston. Great to see you. Watching you up on stage, I got to say, as the CEO of TelcoDR, leader and chief executive of that company. As well as a great visionary, you laid out the vision. It's hard to debate that. I mean, I think there's people who will say that vision, is like freedom, no one can debate it. It's not going to happen. >> Yeah, there's still a lot of debate in our industry about it. There's a lot of articles being written about it. I've referenced one about, you know, should we let the dragons into the castle? For me, I think it's super obvious. I think other industries are like "Duh, we've made the move." And Telco is still like, "Hmm, we're not sure." And so, am I a visionary, I don't know. I'm just sort of just Babe Ruth-ing it a little bit. I think that's where we're going. >> You know you do, you have a lot of content, podcasts, you write blogs, you do a lot of speaking. You brought it all together on stage, right? That has got to feel good. >> Yeah. >> You've got a body of work and it came together very nicely. How did you feel up there? >> Oh my God, it's absolutely nerve wrecking. I sort of feel like, you know, could you tell if my hands were shaking? Right, could you tell that my heart was racing? >> It's a good feeling. >> I don't know. >> Come on! >> I'll be honest, I'm happy it's over, I'm happy. I think I did a really great job and I'm really happy >> Yeah, you did a great job, I love the dragon reference-- >> Have it in the can. >> Fantastic, loved the Game of Thrones vibe there. It was cool-- >> Totally. >> One of the things I wanted pick up on, I thought it was very interesting and unique was the iPhone reference 14 years ago, because that really, to me, was a similar moment because that shifted the smartphone. A computer that happened to make phone calls. And then we all knew who was the leader at that time, Nokia, Blackberry with the phones, and they became toast. That ushered in a whole another era of change, wealth creation, innovation, new things. >> Yeah. Well, up until that moment, carriers had been designing the phones themselves. They were branded with their logos. And so Steve Jobs fought for the design of the iPhone. He designed it with the consumer, with the user in mind. But I think what it really, I mean, it's such a big pivotal moment in our industry because it singled the end of voice revenue and ushered in the era of data. But it also introduced the OTT players, right? That came in through the apps and started a siphon approved from the carriers. And this is like, it's a pivotal moment in the industry, like, changed the industry forever. >> It's a step function, it was a step function change, it's obvious, everyone knew it. But what's interesting is that we were riffing yesterday about O-RAN and Android. So you have iPhone, but Android became a very successful open source project that changed the landscape of the handset. Some are saying that that kind of phenomenon is coming here. Into Telco with software, kind of like an Android model where that'll come in. What's your thoughts on that, reaction to that? >> Yeah, well the dis-aggregation of the hardware, right? We're in the iconic Erickson booth, right? They get most of their revenue from RAN, from Radio Access Networks. And now with the introduction of Open RAN, right? With 50% less CapEx, 40% less OPEX, you know, I think it's easiest for Greenfield operators like Dish, that are building a brand new network. But just this month, Vodafone announced they're going to build the world's largest Open RAN network. Change is happening and the big operators are starting to adopt Open RAN in a real big way. >> So to me, riding the dragon means taking the advantage of new opportunities on top of that dragon. Developing apps like the iPhone did. And you mentioned Android, they got it right. Remember the Windows Phone, right? They tried to take Windows and shove it to the phone-- >> Barely. >> It was a kin phone too. >> I try to delete it from my, look here, beep! >> I'm going to take this old world app and I'm going to shove it into the new world, and guess what, it failed. So if the Telco is trying to do the same thing here, it will fail, but if they start building 5G apps in the cloud and pick the cloud native and think about the consumer, isn't really that the opportunity that you're talking about? >> Well, I think it is, absolutely. And I think it's a wake up call for the vendors in our space, right? And I'm certainly trying to become a vendor with Totogi. I'm really pushing my idea. But you can't take, using your Windows example on the Windows Phone, you can't take a Windows app and stuff it onto a phone and you can't take these old school applications that were written 20 years ago and just stuff them into the cloud, right? Cloud is not a place, it's a way to design applications and it all needs to be rewritten and let's go write, rewrite it. >> It's not a destination as we always say. Let's take a step back on the keynote 'cause I know we just did a couple of highlights there, wasn't the whole thing. We were watching it, by the way, we thought you did a great job, you were very cool and calm under pressure. But take us through the core ideas in the keynote. Break down the core elements of what the talk was about. >> Yeah, I think the headline really is, you know, just like there were good and bad things about the iPhone, right? It killed voice, but introduced data and all these other things. There's good and bad things about the public cloud, right? It's not going to be smooth sailing, no downsides. And so I acknowledge that, even though I'm the self appointed queen, you know? This self appointed evangelist. And so, I think that if you completely ignore the public cloud, try to stick your head in the sand and pretend it doesn't exist, I think there's nothing but downsides for Telcos. And so I think you need to learn how to maximize the advantage there, ride he dragon, like spew some fire and, you know, get some speed and height, and then you can double your ARPU. But I think, going from there, so the next three, I was trying to give examples of what I meant by that, of why it's a double-edged sword, why it's two sides of the coin. And I think there's three areas, which is the enterprise, the network, and a relationship with subscribers. And so that really what the talk, that's what the talk is about >> The three main pillars. >> Yeah, yeah! >> Future, work, enterprise, transition, Open RAN. >> The network and then the relationship with the subscribers. >> Those are the structural elements you see. >> Yeah, yeah, yeah. >> What's the most important one you think, right now, that people are focused on? >> I mean, I think the first one, with work, that's an easy one to do, because there's not too much downside, right? I think we all learned that we could work productively from home. The reason public cloud matter there is because we had tools like Zoom and G Suite and we didn't need to be, I mean, imagine if that this had happened even 20 years ago, right? Broadband at the home wasn't ready, the tools weren't ready. I mean, it would have been, I mean a bigger disaster than it was, right? And so this is an opportunity to sort of ride this work from home wave that a lot of CEOs are saying, we're not coming back or we're going to have smaller offices. And all of those employees need fiber to their home. They need 5G at their home. I mean, if I'm a head of enterprise in a Telco, I am shifting my 5G message from like random applications or whatever, to be like, how are you getting big pipes to the home so your workers can be productive there? And that, I don't hear Telco's talking about that and that's a really big idea. >> You know, you say it's a no brainer, but it's interesting you had your buildings crumbling, which was great, very nice effect in the talk. I heard a executive, Wall Street executive the other day, talking about how, "My people will be back in the office. "I'm going to mandate vaccinations, they're going to be back "in the office, you work for me. "Even though it's an employee friendly environment "right now, I don't care". And I was shocked. I go, okay, this is just an old guy. But, and it's not just the fact that it's an old guy, old guard doing that because I take two examples of old guys, Michael Dell and Frank Slootman. >> Yeah. >> Right, Michael Dell, you know, hundred billion dollar company, Frank Slootman, hottest, you know, software company. Both of them, sort of agree. It's a no brainer. >> Yeah. >> Why should I spend all this money on buildings? And my people are going to be more productive. They love it, so. Why fight the fashion? >> Well, I think the office and I can talk about this for a long time and I know we don't have that much time, but on offices, it's a way to see when did you come in and when did you leave, and look over your shoulder and what we're working on. And that's what offices are for. Now, we tell ourselves it's about collaboration and all this other stuff. And you know, these guys are saying, "come back to the office." It's because they don't have an answer on how to manage productivity. What are you working on? Are you off, are you authentically working 40 hours a week? I want to see, I know if at least you're here, you're here. Now, you might be playing, you know, Minesweeper. You might be playing Minesweeper on your computer, but at least you were, your butt was at your computer. So yeah, I think this is a pivotal moment in work. I think Telcos could push it, to work from home. We'll get you the pipes, we'll get you the cloud-based tools to help manage productivity, to change in work style. >> Yeah, and we've covered this in theCube many times, about how software is going to enable this virtual first model, no one's actually built software for virtual first. I think that's going to happen. Again, back to your team software, but I want to ask you about software defined infrastructure. You mentioned O-RAN, and as software eats the world and eats infrastructure, you still need infrastructure. So, talk about the relationship of how you see O-RAN competing and winning with the balance of software versus the commodity argument. >> Yeah, and I think this is really where people get scared in Telco. I mean, authentically nervous, right. Where you're like, okay, really the public cloud is at that network edge, right? We're really going to like, who are we? It's an identity crisis. We're not the towers anymore. We're renting space, right? We're now dis-aggregating the network, putting the edge cloud right there and it's AWS or Google. Who are we, what do we do, are we networks? Are we a tech company? Right, and so I'm like, guys, you are your subscribers and you don't focus on that. I mean, it's kind of like a last thought. >> So you're like a therapist then too, not just an evangelist. >> I'm a little bit of a therapist. >> Okay, lay down on the couch, Telco. >> Let's talk about what your problems are. (laughs) >> They have tower issues. >> All seriousness, no but, the tower is changing is backhauling. Look at direct connects for instance. The rise of direct and killed the exchanges. I mean, broadband, backhaul, last mile, >> Yeah. >> Completely, still issues, >> Yeah. >> But it's going to software and so that's there. The other thing I want to get to quickly, I know we don't have a lot of time, is the love relationship you talk about with subscribers. We had Peter Adderton on, from a Boost Mobile, formerly Boost Mobile, earlier. He was saying, if you don't have a focus on the customer, then you're just selling minutes and that's it. >> Yeah. >> And his point was, they don't really care. >> Yeah. Let's talk about organizational energy, right? How much energy is contained within any organization, not just Telco, but any organization. To some of your people time is the hours they work per week. And then you think of that as a sack on how you're allocating your time and spending your time, right? And so I think they spend 50% of their time, maybe more, fighting servers, machines, the network, right? And having all these battles. How much of that organizational energy is dedicated to driving great subscriber experiences? And it just shrunk, right? And I think that's where the public cloud can really help them. Like ride the dragon. Let the dragon deal with some of this underlying stuff. So that you can ride a dragon, survey the land, focus on your subscriber and back to the software. Use software, just like the OTT players are doing. They are taking away your ARPU. They're siphoning your ARPU, 'cause they're providing a better customer experience. You need to compete on that dimension. Not the network, not the three Telcos in the country. You're competing again, WhatsApp, Apple, Amazon, Facebook. And you spent how much of your organizational energy to focus on that? Very small. >> And that's where digital platforms roll by, it uses the word platform, why? Because everybody wants to be a platform. Why do you want to be a platform? Because I want to be like Amazon, they're a platform. And you think about Netflix, right? It's not, you know, you don't think about Netflix UK or Netflix Spain, right? >> It's global. >> There's one Netflix >> Yeah, yeah. >> You don't think about their marketing department or their sales department or their customer service, you think about the app. >> Yeah. >> You know. One interface. And that's what digital platforms allow you to do. And granted, there's a lot of public policy to deal with, but if you're shooting satellites up in space, >> Yeah. >> You know, now, you own that space, right, global network. >> And what makes Netflix so good, I think, is that it knows you, right? It knows what you're watching and recommends things, and you're like, "Oh, I would like that, that's great." Who knows more about you than your mobile phone? Carry it everywhere you go, right? What you're watching, what you're doing, who you're calling, what time did you wake up? And right now all of that data we talked about a couple of days ago, it's trapped in siloed old systems. And like why do people think Google knows so much about you? Telco knows about you. And to start to use that to drive a great experience. >> And you've got a great relationship with Netflix. The relationship we have with our our carrier is to your admin, "can you call these guys? "I don't know, I lost the password, I can't get in". >> Right. >> It's like-- >> Or you get SIM hacked-- >> I don't have an hour and a half to call your call center 'cause you don't have a chat bot, right. >> I don't have time. >> Chat bot, right. I can't even do the chat bot because my problem is, you're like, I got to talk to someone. All of their systems are built with the intention of a human being on the other side, and there's all this awesome chat bot AI that works. >> Yeah. >> Set it free. >> Yeah, yeah, right. You almost rather go to the dentist, then calling your carrier. >> Well, we're going to wrap things up here on the keynote review. Did you achieve what you wanted to achieve? I mean, controversy, bold vision, leadership, also that came across, but people they know who you are now. You're out there and that's great news. >> Yeah. I think I rocked the Telco universe and I'm really, that was my goal, and I think I accomplish it so, very excited. >> Well, we love having you on theCUBE. It's great to have great conversations, not only are you dynamic and smart, you're causing a lot of controversy, in a good way and getting, waking people up. >> Making people talk, that's a start. >> And I think, the conversations are there. People are talking and having relationships on the ecosystem open, it's all there. Danielle Royston, you are a digital revolution, DR. Telco DR, thanks for coming to theCube. >> Thank you so much, always fun. >> Good to see you. >> Thanks. >> Of course, back to the Cloud City studios. Adam is going to take it from here and continue on day three of theCube. Adam in studio, thanks for having us and take it from here.

>>Hey, welcome to the Cubes coverage of Red Hat Summit 2021, the virtual experience, I'm lisa martin, I have two guests joining me. One is a cube alum kamal Shah is back, he's now the VP of cloud platforms at Brent had come on, it's great to have you back on the program. You're in a new role, we're going to talk about that. Thank you. And Kirsten newcomer is here as well. She's the Director of cloud and Death stickups strategy at Red Hat, Kirsten, Welcome and thank you for bringing the red hat vibe to the segment. >>Absolutely, very happy to be here. >>So looking forward to this conversation that we're going to be having in the next 20 minutes or so. We're gonna be talking about the last time come on, you were on, you were the ceo of stack rocks In January of 2021. The announcement that red hat plans to acquire stack rocks, it wouldn't be talking all about that. But I'd like to start with Kirsten, give us your perspective from red hats perspective, why is red hat a good fit for stack rocks? >>You know, there are so many reasons first of all as as you know, right? Red hat has been working with product Izing kubernetes since kubernetes one dato. Right, so so open shift three dato shipped with kubernetes one dot Oh, so we've been working with kubernetes for a long time, stack rocks embraces kind of is kubernetes native security embraces the declarative nature of kubernetes and brings that to security. Red hats, Custer's red hat enterprise customers, we have a great set across different verticals that are very security conscious and and during my five years at red hat, that's where I spend the majority of my time is talking with our customers about container and kubernetes security. And while there's a great deal of security built in to open shift as it goes to market out of the box, customers need the additional capabilities that stack rock springs. Historically, we've met those needs with our security partners. We have a great ecosystem of security partners. And with the stack rocks acquisition, we're now in a position to offer additional choice. Right. If a customer wants those capabilities from Red hat tightly integrated with open shift, we'll have those available and we continue to support and work with our broad ecosystem of security partners. >>Excellent customers always want choice. Come on. Give me your perspective. You were at the helm the ceo of stack rocks as you were last time you were on the cube. Talk to me about the redhead acquisition from your seat. >>Yeah. So as as Kirsten mentioned, we were partners of red hat. You're part of the red hat partner ecosystem. And uh, what we found is that was both a great strategic fit and a great cultural fit between our two companies. Right? And so the discussions that we had were how do we go and quickly enable our customers to accelerate their digital transformation initiatives to move workloads to the cloud, to containerized them, to manage them through kubernetes and make sure that we seamlessly addressed their security concerns. Right? Because it continues to be the number one concern for large enterprises and medium sized enterprises and frankly any enterprise that uh, you know, uh, working out today. So, so that was kind of the impetus behind it. And I must say that so far the the acquisition has been going on very smoothly. So we had two months in roughly and everybody and has been very welcoming, very collaborative, very supportive. And we are already working hand in hand to to integrate our companies and to make sure that we are working closely together to make our customers successful. >>Excellent. We're gonna talk about that integration in a second. But I can imagine challenging going through an acquisition during a global pandemic. Um but that is one of the things that I think lends itself to the cultural alignment. Kamal that you talked about, Kirsten. I want to get your perspective. We know we talk about corporate culture and corporate culture has changed a lot in the last year with everybody or so many of us being remote. Talk to me about kind of the core values that red hat and stack rocks share >>actually, you know, that's been one of the great joys doing during the acquisition process in particular, Kamal and and ali shared kind of their key values and how they um how they talked to talk with their team And some of the overlap was just so resonated so much for all of us. In particular the sense of transparency, uh, that the, that the team the stack rocks executive team brings and approaches. That's a that's a clear value for red hat um strongly maintained. Uh, that was one of the key things the interest in um uh, containers and kubernetes. Right. So the technology alignment was very clear. We probably wouldn't have proceeded without that. But again, um and I think the investment in people and the independence and the and the strong drive of the individuals and supporting the individuals as they contribute to the offering so that it really creates that sense of community um and collaboration that is key. Uh and and it's just really strong overlap in in cultural values and we so appreciated that >>community and collaboration couldn't be more important these days. And ultimately the winner is the customers. So let's dig in. Let's talk about what stack rocks brings to open shift Kirsten take it away >>man. So as I said earlier, um so I think we we really believe in continuous security at red hat and in defense and depth. And so when we look at an enterprise kubernetes distribution that involves security at the real core os layer security and kubernetes adding the things into the distribution, making sure they're there by default, that any distribution needs to be secured to be hardened, auditing, logging, identity, access management, just a wealth of things. And Red hat has historically focused on infrastructure and platform security, building those capabilities into what we bring to market stack rocks enhances what we already have and really adds workload protection, which is really when it comes down to it. Especially if you're looking at hybrid cloud, multi cloud, how you secure, not just the platform, but how you secure your workloads changes. And we're moving from a world where, you know, you're deploying anti virus or malware scanners on your VMS and your host operating system to a world where those work clothes may be very short lived. And if they aren't secured from the get go, you miss your opportunity to secure them right? You can't rely on, you know, you do need controls in the infrastructure but they need to be kubernetes native controls and you need to shift that security left. Right? You never patch a running container. You always have to rebuild and redeploy if you patch the running container the next time that container images deployed, you've missed, you've lost that patch. And so the whole ethos the whole shift left. The Deb sec ops capabilities that stack rock springs really adds such value. Right? You can't just do DEF SEc or set cops. You need to do a full infinity loop to really have def SEc ops and stack rocks. I'm gonna let Kamal tell you about it, but they have so many capabilities that that really drive that shift left and enable that closed loop. We're just so excited that they're part of our offerings. >>So can you take us through that? How does stack rocks facilitate the shift left? >>Yeah, absolutely. So stack rocks, which we we announced at summit is now being rebranded as red hat. Advanced cluster security was really purpose built to help our customers address the use cases across the entire application lifecycle. Right? So from bill to deploy to run time. So this is the infinite loop that Kirsten mentioned earlier and one of our foundations was to be kubernetes native to ensure that security is really built into the application is supposed to bolt it on. So specifically, we help our customers shift left by securing the supply chain and we're making sure that we identifying vulnerabilities early during the build process before they make it to a production environment. We helped them secure the infrastructure by preventing miS configurations again early in the process because as we all know, MIS configurations often lead to breaches at at runtime. Right? We help them address uh compliance requirements by ensuring that we can check for CS benchmarks are regulatory requirements around the C I P C I, hip hop and this and and that's uh you know, just focusing on shift left, doesn't really mean that you ignore the right side or ignore the controls you need uh when your applications are running in production. So we help them secure that at runtime by identifying preventing breaches the threat detection, prevention and incident response. >>That built in security is you both mentioned that built in versus bolt on Kirsten? Talk to me about that, that as really kind of a door opener. We talked a lot about security issues, especially in the last year. I don't know how many times we've talked about miS configurations leading to breaches that we've seen so many security challenges present in the last year. We talked to me a little bit Kirsten about >>what >>customers appetites are for going. All right now, I've got cloud native security, I'm going to be able to, I'm going to feel more comfortable with rolling out production deployments. >>It's, it's a great place to go. So there are a number of elements to think about. And if I could, I could, I could start with by building on the example that Kamal said, Right, So when we think about um I need to build security into my pipeline so that when I deliver my containerized workloads, they're secure. What if I miss a step or what if a new vulnerability is discovered after the fact? Right. So one of the things that stack rocks or redhead a CS offers is it has built in policy checks to see whether a container or running image has something like a package manager in it. Well, a package manager can be used to load software that is not delivered with the container. And so the idea of ensuring that you are including workload, built in workload, protect locks with policies that are written for you. So you can focus on building your applications. You don't necessarily have to learn everything there is to know about the new attack vectors that are really just it it's new packaging, it's new technology. It's not so much there are some new attack vectors, but mostly it's a new way of delivering and running your applications. That requires some changes to how you implement your security policies. And so ensuring that you have the tools and the technology that you're running on have those capabilities built in. So that when we have conversations with our security conscious customers, we can talk with them about the attack vectors they care about. We can illustrate how we are addressing those particular concerns. Right? One of them being malware in a container, we can look for stack. Rocks can look for a package manager that could be used to pull in, you know, code that could be exploited and you can stop a running container. Um, we can do deeper data collection with stack rocks. Again, one of the challenges when you're looking at moving your security capabilities from a traditional application environment is containers come and go all the time. In a kubernetes cluster nodes, your servers can come and go in a cloud native kubernetes cluster, right? If you're running on on cloud public cloud infrastructure, um, those things are the nodes are ephemeral to, they're designed to be shut down and brought back up. So you've got a lot more data that you need to collect and that you need to analyze and you need to correlate the information between these. Right? I no longer have one application stack running on one or more VMS, it's just things are things are moving fast so you want the right type of data collection and the right correlation to have good visibility into your environment. >>And if I can just build on that a little bit. The whole idea here is that these policies really serve as god rails right for the developers. So the it allows developers to move quickly to accelerate the speed of development without having to worry about hundreds of potential security issues because there are guardrails that will notify that with concrete recommendations early in the process. And the analogy I often use is that you know the reason we have breaks in our cars, it's not to slow us down but to allow us to go faster because we know we can slow down when we need to write. So similarly these policies are really it's really designed to accelerate the speed of development and accelerate digital transformation initiatives that our customers are embarking on >>and come on. I want to stick with you on the digital transformation front. We've talked so much about how accelerated that has been in the last year with everything going on in such a dynamic market. Talk to me Kamal about some of the feedback that you've gotten from stack rocks customers about the acquisition and how it is that maybe that facilitator of the many pivots that businesses have had to do in the last year to go from survival mode to thriving business. >>Yeah. Yes, absolutely. The feedback from all of our customers bar none has been very very positive. So it's been it's allowed us to invest more in the business and you know, we publicly stated that we are going to invest more in adding more capabilities. We are more than doubling the size of our teams as an example. And really working hand in hand with our uh the broader team at Red had to uh further accelerate the speed of development and digital transformation initiatives. So it's been extremely positive because we're adding more resources, We're investing more. We're accelerating the product roadmap uh based on uh compared to what we could do as a, as a start up as you can imagine. And and the feedback has been nothing but positive. So that's kind of where we are today. And what we're doing with the summit is rolling out a new bundle called open shift uh, Open shift platform plus, which includes not just Red hat A CS which used to be Stock rocks, but also red hat open shift hybrid cloud platform as well as Red hat advanced uh container cluster management, ACM capabilities as well as create the container registry. So we're making it easier for our customers to get all the capabilities that they need to for the drive digital transformation initiatives to get. It goes back to this whole customer centric city team that red hat has, that was also core value of stack rocks and and the winner and all of this, we believe ultimately is our, our our customers because that's where we exist to serve them, >>right. And I really like that if I could chime in kind of on top of that a little bit. Um so, so I think that one of the things we've seen with the pandemic is more of the red Hat customers are accelerating their move to public cloud and away from on premises data centers. Uh and and you know, that's just part partly because of so many people working remotely. Um it just has really pushed things. And so with Hybrid cloud becoming even more key to our joint customer base and by hybrid cloud, I mean that they have some environments that are on premises as they're making this transition. Some of those environments may stay that footprint may stay on premises, but it might be smaller, they may not have settled on a single public cloud. They could, in fact, they often are picking a public cloud based on where their development focuses. Google is very popular for ai and ml workloads. Amazon of course is just used, you know, by pretty much everybody. Um and then Azzurri is popular with um a subset of customers as well. And so we see our customers investing in all of these environments and stack rocks red hat A CS like open shift runs in all these environments. So with open shift platform plus you get a complete solution that helps with multi cluster management with a C. M with security across all of these environments, right? You can take one approach to how you secure your cluster, how you secure your workloads, how you manage configurations, You get one approach no matter where you're running your containers and kubernetes platform when you're doing this with open shift platform plus. So you also get portability. If today you want to be running an amazon maybe tomorrow you need to spin up a cluster in google, you can do that if you're working with the K s or G K E, you can or a Ks, you can do that with red hat a CS as well. So we really give you everything you need to be successful in this move and we give you back to that choice word, right? We give you the opportunity to choose and to migrate at the speed that works for you. >>So that's simplicity. That streamlining. I gotta ask you the last question here in our last couple of minutes. Come on, what's the integration process been like? as we said the acquisition just a couple of months in. But talk to me about that integration process. What that's been like? >>Yeah, absolutely. So as I mentioned earlier, the process has been very smooth so far, so two months in and it's largely driven by the common set of culture and core values that exists between our two companies. And so uh you know, from a product standpoint, we've been working hand in hand because I mentioned earlier, we were partners are working hand in hand on accelerating the road map the joint roadmap that we have here uh from a go to market perspective teams are well integrated. We are going to be rolling out the rolling out the bundle and we're gonna be rolling out additional uh options for our customers. We've also publicly announced that will be open sourcing uh red hat A. C. S. Uh formerly known as Stock Rock. So stay tuned for further news and that announcement. And, and so you know, uh, again two months and everybody's been super collaborative. Super helpful, super welcoming. And the team is the well settled and we're looking forward to now focusing on our primary objective is just to make sure that our customers are successful. >>Absolutely. That customer focus is absolutely critical. But also so is the employee experience. And it sounds like we both talked about the ethos and the and the core value alignment. They're probably being pretty critical to doing an integration during a very challenging time globally. I appreciate both of you joining me on the program today, sharing what's going on stack rocks now asks the opportunities for customers to have that built in cuBA and the security. Thanks so much for your time. >>Thank you. Thank >>you for Camel shaw and Kirsten newcomer. I'm lisa martin. You're watching the cubes coverage of Red Hat Summit, The virtual experience. Mhm

>>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.

>>Yeah, >>from the Cube Studios in Palo Alto and Boston. It's the Cube covering IBM. Think brought to you by IBM. >>Welcome back, everybody. This is Dave Vellante of the Cube, and you're watching our continuous coverage of the IBM think Digital 2020 experience. And we're really pleased to have Rob High here. He's not only an IBM fellow bodies. He runs the vice president CTO of the IBM Edge Computing Initiative. Rob, thanks so much for coming on the Cube. Good to see you. Which we're face to face, but yeah, that time to be safe and healthy, I guess. And did so edge obviously hot topic. Everybody has this sort of point of view would be interested in how IBM looks at edge. You define it and what your thoughts are on. It's evolution. >>Yeah, well, you know, there's ah really kind of two fairly distinct ways of thinking about the edge of the telcos. Our, ah, you know, they're creating edge capabilities in their own network facilities. We call that the network edge on the other side of the edge they that I think matters a lot to our enterprise businesses is there's remote on premise locations where they actually perform the work that they do, where the majority of people are, where the data that actually gets created is first formed and where the actions that they need to operate on are being taken. That is a lot of interest, because if we can move work workloads, Iot workloads to where that data is being created, where those actions are being taken Uh, not only can we dramatically reduce the late and see to those decisions, uh, but we can also ensure continuous operations and the failed in the presence of perhaps network failures. We can manage the growth of increasing demand for network bandwidth as Maura born data gets created and we can optimize the efficiency of both the business operations as well as the I t operations before that. So for us edge computing at the end of the day is about movie work where the data and the actions are being taken >>well, so this work from home, you know, gives a result of this pandemic is kind of creating a new stresses on networks and people are putting, you know, pouring money actually into beefing up that infrastructure is sort of an extension of what we used to think about edge. But I wonder if you could talk about some of the industries and the use cases that you guys we are seeing and notwithstanding, though assay that >>work from home pivot. Yeah, absolutely. So I mean, look, we have seen ah, the need for placing workloads close to where it is being created and where actions have been picking in virtually every industry, the ones that are probably easier for us to think about and more common in terms of our mindset. Our is manufacturing. If you think about all the things that go on in a factory floor that need to be able to perform analytic in, uh, in the equipment and the processes that are performing in the affection for, If you think, for example, production quality. Uh, you know, if you've got a machine that's putting out parts and maybe it's welding seams on metal boxes, uh, you know, you want to be able to look at the quality of that seem at the moment that is being performed, so that if there are any problems, you can remediate that immediately rather than having that box move on down the line and find that you know the quality issues they were created earlier on now have exacerbated in other ways. Um, you know, so quality, productive quality. Ah, inspection production optimization in our world of Covic Cover 19 and worker safety and getting workers back to work and ensuring that you know people wearing the masks and are exercising social distancing. This is on the factory floor. Worker Insight is another major use case that we're seeing surface of lake with a lot of interest in using whether that's infrared cameras or Bluetooth beacons or infrared cameras. Any variety of devices that could be employed in the work area to help ensure that factories are operating efficiently, that workers are safe. Ah, and whether that's in a factor situation or even in an office situation or e a r in a warehouse or distribution center. And all these scenarios the the utility, the edge computing to bring to those use cases is tremendous. >>And a lot of these devices are unattended or infrequently attended. I always use the windmill example. Um, you know, you don't want to have to do a truck roll to figure out you know what the dynamics are going on, that at the windmill s, so I can instrument that. But what about the management of those devices you know from an autonomous standpoint? And and are you? What are you doing? Or are you doing anything in the autonomous managed space? >>Yeah. In fact, that's really kind of key here, because when you think about the scale, the diversity and the dynamic dynamism of equipment in these environments And as you point out, Dave, you know the lack of I t resource lack of skills on the factory floor, or even in the retail store or hotel or distribution center or any of these environments. The situation is very similar. You can't simply manage getting the right workloads to the right place at the right time. In sort of the traditional approach is, you have to really think about another autonomous approach to management and, you know, let the system the side for you. What software needs to be placed out there? Which software to put their If it's an analytic algorithm, what models to be associated with that software and getting to the right place at the right place at the right Time is a key Part of what we do in this thing that we call IBM Edge application manager is that product that we're really kind of bringing to market right now in the context of edge computing that facilitates this idea of autonomous management. >>You know, I wonder if you could comment Robb on just sort of the approach that you're taking with regard to providing products and services. I mean, we've seen a lot of, uh, situations where people are just essentially packing, packaging traditional, you know, compute and storage devices and sort of throwing it over the fence at the edge. Uh, and saying, Hey, here's our edge computing solution and another saying there's not a place for that. Maybe that will help flatten the network and, you know, provide Ah, gateway for storing on maybe processing information. But it seems to us that that that a bottoms up approach is going to be more appropriate. In other words, you've got engineers, you know who really understand operations, technology, people, maybe a new breed of developers emerging. How do you see the evolution you know of products and services and architectures at >>the edge? Yeah, so First of all, let me say IBM is taking a really pretty broad approach to edge computing we have. What I just described is IBM Edge Application Manager, which is the if you will the platform or the infrastructure on which we can manage the appointment of workloads out to the edge. But then add to that we do have a whole variety of edge and Nevil enabled applications that are being created are global service of practices and our AI applications business all are creating, um, variations of their product specific to address and exploit edge computing and to bring that advantage to the business. And of course, then we also have global services Consulting, which is a set of skilled resource, is who know we understand the transformations that business need to go through when they went, take advantage of edge computing and how to think about that in the context of both their journey to the cloud as well as now in this case, the edge. But also then how to go about implementing and delivering that, uh and then firmly further managing that now you know, coupled out then with at the end of the day you're also going to need the equipment, the devices, whether that is an intelligent automobile or other vehicle, whether that is an appellate, a robot or a camera, Um, or if those things are not intelligent. But you want to bring intelligence to them that how you augment that with servers and other forms of cluster computing that resides resident with the device. All of those are going to require participation from a very broad ecosystem. So we've been working with partners of whether that is vendors who create hardware and enabling that hardware in certifying that hardware to work with our management infrastructure or whether those are people who bring higher order services to the table that provide support for, let's, say, data cashing and facilitating the creation of applications, or whether those are device manufacturers that are embedding compute in their device equipment. All of that is part of our partnership ecosystem, Um, and then finally, you know, I need to emphasize that, you know, the world that we operate in is so vast and so large. There are so many edge devices in the marketplace, and that's growing so rapidly, and so many participants in that likewise There are a lot of other contributors to this ecosystem that we call edge computing. And so for all of those reasons, we have grounded IBM education manager on open source. We created an open source project called Open Rise, and we've been developing that, actually now, for about 4.5 years just recently, the Linux Foundation has adopted Stage one adoption of Open arising as part of its Lennox Foundation edge LF edge, uh, Reg X Foundry project. And so we think this is key to building out, Um, a ecosystem of partners who want to both contribute as well consumed value and create ecosystems around this common idea of how we manage the edge. >>Yeah, I'm glad you brought up the ecosystem, and it's too big for any one company toe to go it alone. But I want to tap your brain on just sort of architectures. And there's so many diverse use cases, you know, we don't necessarily see one uber architecture emerging, but there are some characteristics that we think are important at the edge you mentioned sort of real time or near real time. In many cases, it has to be real time you think about autonomous vehicles? Um, yeah. A lot of the data today is analog, and maybe it doesn't have to be digitized, but much of it will be, um, it's not all gonna be sent back to the cloud. It may not all have to be persisted. So we've envisioned this sort of purpose built, you know, architecture for certain use cases that can support real time. That maybe have, you know, arm based processors. Ah, or other alternative processors there that can do real time analytics at the edge and maybe sending portions of the data back. How do you see the architectures evolving from a technologist? >>Well, so certainly one of the things that we see at the edge is a tremendous premium being placed on things like energy consumption. So architectures they're able to operate efficiently with less power is ah is certainly an advantage to any of those architectures that are being brought aboard. Um, clearly, you know x 86 is a dominant architecture in any information technology endeavor. More specifically at the edge. We're seeing the emergence of lot of arm based architecture chips out there. In fact, I would guess that the majority of the edge devices today are not being created with, um, arm architectures, but it's the you know, but some of this is about the underlying architecture of the compute. But also then the augmentation of that compute the the compute Thea the CP use with other types of processing units. Whether those GPS, of course, we're seeing, you know, a number of deep use being created that are designed to be low power consuming, um, and have a tremendous amount of utility at the edge. There are alternate processing units, architectures that have been designed specifically for AI model based analytics. Uh, things like TP use and infuse and and, uh, and set around, which are very purpose built for certain kinds of intellect. And we think that those are starting to surface and become increasingly important. And then on the flip side of this is both the memory storage in network architectures which are sort of exotically different. But at least in terms of capacity, um have quite variability. Specifically, five G, though, is emerging and five g. While it's not necessarily the same computing, there is a lot of symbolism between edge and five G and the kinds of use cases that five G envisions are very similar to those that we've been talking about in the edge world as well. >>Rob, I want to ask you about sort of this notion of program ability at the edge. I mean, we've seen the success of infrastructure as code. Um, how do you see program ability occurring at the edge in terms of fostering innovation and maybe new developer bottles or maybe existing developer models at the edge? Yeah, >>we found a lot of utility in sort of leveraging what we now think of as cloud computing or cloud computing models. Uh, you know, the idea of continue ization extends itself very easily into the edge. Whether that is running a container in a docker runtime, let's say on an edge device which is, you know, resource constrained and purpose built and needs to focus on sort of a very small footprint or even edge clusters edge servers where we might be running a cluster of containers using our kubernetes platform called open shift. Um, you know the course of practices of continuous integration, continuous delivery. What we write a Otherwise think of his Dev ops. Ah, and, of course, the benefits they continue. Realization brings to the idea of component architectures. Three. Idea of loose coupling. The separation of concerns, the ability to mix and match different service implementations to be opposed. Your application are all ideas that were matured in the cloud world but have a lot of utility in the edge world. Now we actually call it edge native programming. But you can think of that as being mostly cloud native programming, with a further extension that there are certain things you have to be aware of what you're building for the edge. You have to recognize that resource is air limited. Unlike the cloud where we have this notion of infinite resource, you don't have that at the edge. Find and constrained resources. Be worried about, you know, Layton sees and the fact that there is a network that separates the different services and that network can be and reliable. It can introduce his own forms of Layton sees it, maybe bandwidth constrained and those air issues that you now have to factor into your thinking as you build out the logic of your application components. But I think by building on the cloud native programming about me paradigm. You know, we get to exercise sort of all of the skills that have been developing and maturing in the cloud world. Now, for the edge >>that makes sense. My last question is around security. I mean, I've often sort of tongue in cheek said, you know, building a moat around the castle doesn't work anymore. The queen i e. The data has left the castle. She's everywhere. So what about the security model? I mean, I feel like the edge is moving so fast you feel confident or what gives you confidence >>that we can secure the edge. You know, the edges does introduce some very interesting and challenging concerns with respect to security because, frankly, the compute is out there in the wild. You know, you've got computers in the store you've got, you know, people walking around the kiosks you have in the manufacturing site, you know, workers that are, you know, in the midst of all of this compute capability and so the attack surface is substantially bigger. And that's been a big focus for us, is how to the only way validate in 30 of the software that was But it also takes advantage of one of the key characters with edge computing to bring to the table, which is, if you think about it. You know, when you've got personal and private information being entered into quote system, the more often you move that personal private data around, and certainly the more that you move it to a central location and aggregate that with other data, the more of a target becomes more vulnerable, exposed that data becomes and by using edge computing, which moves the workloads out to the edge where that did has been created in some sense, you can process on it there and then move it back. They need central location, you don't have to aggregate it. And that actually in itself is a counterbalance of all of the other issues that we also describe about security by essentially not moving the personal privacy and in protecting by keeping it exactly where it began. >>You know, Rob, this is an exciting topic. Is a huge opportunity for IBM and Ginny in and talk about the trillion dollar opportunity and hybrid cloud and the Edge is a multi $1,000,000,000 opportunity for IBM and, uh So you just got to go get her done. But I really appreciate you coming on the Cube and sharing your insights. That awesome topic in the best interest of the David. Yeah. Thank you. Thank you for the thank you. Stay safe and thank you for watching everybody. This is Dave Volante for the Cube. This is our coverage of IBM. Think 2020 the digital. Think >>we'll be right back after this short break? >>Yeah, yeah, yeah, yeah.

>> Announcer: From theCUBE Studios in Palo Alto and Boston, it's theCUBE! Covering IBM Think, brought to you by IBM. >> Hi everybody, welcome back to theCUBE's coverage of IBM 2020, it's the digital IBM 2020, the Think Event Experience. My name is Dave Vellante, and you are watching theCUBE. Steve Canepa is here, he's the global GM of Communications, of the Communications sector for IBM. Steve, how ya doing, good to have you on. >> Doing great Dave, thanks for having me. >> Yeah, you're very welcome. I mean, communications is sort of a broad term for the stuff you covered. Telco, Cable, Entertainment, Broadcast, Publishing, Satellite, Sports, Music, Games, I mean, Social, wow. You run the gamut. >> It's exciting times. >> Pretty big role, yeah, I'll say you've got exciting times. With so much going on in your space, and of course this pandemic is really, you know, hit the communications industry in so many different ways. Some tailwinds, some headwinds, and it's just crazy out there. What are you seeing and what are you guys doing to support clients? >> Well, first and importantly, our thoughts go out to everyone. As we're all dealing with this around the world. I have the opportunity to work with clients, in every geography, around the globe and each and everyone of them is busily dealing with how they make sure their employees are safe, how they're providing services to their customers. And, we're right there alongside them, helping them do that as well. For us in the telecommunication space, as you know, it's actually essential, it's an essential industry that's helping the world deal with this as we are all going virtual like this session we're having today. So we're working with clients to help them get their resources in place so that they can support their businesses, their network platforms, their media services in a way that they can keep the business running. Our telecommunications customers all around the globe had to get their resources and work at home environments, we work with many of them, in deploying real-time services. We also work with them in deploying call center chatbot capabilities so that they could answer questions from their customers, from other members of the community as they were coming in. So tremendous opportunity for us to help them respond to what's happening. >> It's actually quite amazing the response, when you think about telco, you think about telco infrastructure, what comes to mind is, it's hard and it's reliable, it works and all of a sudden you've got all these remote workers. The pace of the pivot has been actually quite astounding. I mean your thoughts on that. >> Yeah and it actually goes hand in glove in what way we've been preparing the industry for generally. I mean there's been this evolution that digital service providers that's been happening in the industry now for a number of years and in fact the center point of what we're doing now to help telcos virtualize and abstract those networks so that they're software based services platforms that can respond to these kinds of peak load demand issues. Not that anyone anticipated COVID, but the ability to have a platform, that can scale your business. And allow you to respond, move services where they need to be moved, be much more agile in the way you work. These are all playing in the ability to respond to it. >> Steve I want to ask you about something you said in a recent article in Forbes, you said, winners in the 5G and edge era will be those who embrace a hybrid multicloud approach." Well, first of all, I want to ask you, I mean five G can't get here fast enough, but still you're kind of predicting, my inference a five g and edge era coming this decade, like I said it can't happen soon enough. What are your thoughts on this coming era? >> In my view there's three fundamental things that are happening simultaneously. So first obviously, five g is emerging. It's showing up now. Most service providers around the world are starting to already deploy their private five g capabilities. We're seeing it show up in evolution form, in consumer marketplace. So five g is here and will continue to scale. The second key transformation that's happening is the telco network itself is becoming a hybrid cloud platform. What I mean by that is that just as when video abstracted as a service and it could be deployed over the top service platforms, enabling things like our interview that we're doing today to happen, that got loaded on top of now open IP platform. The same exact thing is happening in the network domain where the network services, data services, voice service, multimedia services are being put on an open platform architecture that allows you to respond. And then the third key thing that's happening in the market is this edge phenomenon. And this is all about the ability to move workloads, to move services out closer to where things happen and take advantage of those key five g features like ultra low latency, increased bandwidth, and of course the ability to slice the network down to dedicated to a specific application. This opens up a whole new set of services. >> Yeah I mean as I was sort of eluding to before, the reliability of telco networks has been the hallmark of that infrastructure. As we move to this more open sort of standardized environments Steve, I would imagine that one of the technical challenges is maintaining that level of reliability and predictability while at the same time being able to support remote workers, etc low latency workloads. Can you comment on that? >> Yeah so a couple of key points there. One is as you may know, IBM acquired Red Hat a little over a year ago. Red Hat has created an open platform for the telco's to modernize their core infrastructure. And the power of that is we can see is this enormous upstream community now and that community can help accelerate the rate pace of transformation is happening, bring innovation in. That's really powerful. The second is, once we go through an open platform, software based platform, we can infuse automation. Extreme levels of automation, and AI for intelligent predictive capability. And now think about the network becoming a living, breathing, responding platform where it's based on software. So we can deploy services and functions and we can automate those services and functions. That level of intelligence serves as the ability to then get out these services. >> So Steve, definitely we had I think a decent understanding of the Red Hat and the strategy around Open Shift and the container approach, hybrid multicloud. What I didn't realize is that there was specificity around the telco industry. Can you talk more specifically about what IBM is doing in that regard? >> Yeah, it's a great question. Red Hat has a very significant presence in 120 telcos around the globe. And so not only they're Red Hat Linux which is kind of a defacto standard in the marketplace, but their open stack architecture now we're moving after the Open Shift architecture. And as part of that the relationship with an enormous upstream community of talent, it's building on those platforms. And so we're able to really infuse into Red Hat the kind of requirements that are necessary for their software platform to serve as the platform, the open platform for the telcos as we go forward. It has been an incredible synergy. I think of it as kind of two puzzle pieces that fit together incredibly well. At IBM we've had the long standing relationship with all the service providers around the world and helping them transform their business and now with Red Hat we have the opportunity to really integrate what we're doing with automation and AI standpoint with all the power of that Red Hat Platform. >> So where do you see the edge fitting into this hybrid multicloud approach? Is it sort of an extension of cloud? Is it a new cloud? We know we are envisioning this seamless experience between on-prem, cloud, multicloud, and edge. >> Yeah I think of it in a kind of simple venn diagram where you have kind of this virtualized open software based telco network on one side and you have the edge on the other and in the middle you have this kind of combination where you do edge in partnership with the telco. And the idea here is that all industries are going to want to provide a next generation of insights to their customers and to their partners. The ability to move those workloads, so think about a manufacturing shop for an example. You know we've already had IoT centers, hundreds if not thousands of them. Now we can infuse video cameras and take a huge amount of data through the enhanced bandwidth of five g and bring that down to an edge platform and analyze that video data in real time, whether employees are in safe zones, maybe with COVID now even, whether or not they're taking the proper social distancing, and looking at actually everything that's coming off of that platform or manufacturing line, looking at the equipment itself and adding AI to that so that we can analyze it in real time. Edge allows us to take advantage of those five g attributes and to put it wherever that workload should run. Whether it's on the plant floor itself, in proximity to where that equipment is, or back at a central office location within the network of a telco. >> Well this is huge for the telcos because for years, I keep talking about their hardened network, but their cost per bit has been coming down. They're responsible for putting in that infrastructure, maintaining that infrastructure and then you got the over the top providers laying out content growing like crazy, has really disrupted that industry. This is going to change the way in which telcos are able to compete, is it not? >> It's a great point. Yes. If you think about the last generation of evolution you know when we went to four g and smartphones came out, think about the Apple App Store as an example. Folks started not going to the telcos anymore for those services, they went to that OTP capability to get those applications. Now think about about in this edge world as we essentially are creating platforms for innovation for businesses and all industries. And they can now innovate on those platforms and create incredible value in their business and the telcos now can add beyond just the transport capability, but artificial intelligence, automation, they can expose certain data capabilities, they can make those applications smarter, understanding proximity data, that could be applied to things like logistics or pricing or as I said operations like in manufacturing. So a tremendous new set of value in fact most analysts say a trillion dollars in value is going to be created here. And the opportunity I see is that the open network platform becomes a way for the service providers to not only capture value for themselves, but to accelerate the value for businesses in all industries. >> Well I think we're going to see some huge moves in the chess board. More M and A. I mean it's going to be a very exciting time and of course five g's at the heart of it, but Steve I wonder if you could give us IBM's point of view in terms of where we are with five g, I mean sometimes I see it pop up on my phone and I'm like come on, that's not real five g quite yet. We heard recently that Apple might somewhat delay it's new phones that maybe five g's involved in that, but it's going to take some time for that infrastructure to roll out but what's your point of view on sort of that time frame and the business impact that we can all expect? >> Yeah I know, it's a good question. And we will see it roll out over time. Some things are starting to roll out now. Think about stadiums or other venues where you have a manufacturing shop floor as an example, oil rig off the coast. I mean you have environments where you could create five g infrastructure in a private model today and then of course consumer models are going to roll out as cities continue to get deployed by the various service providers. But I think the important point which is what we've spoken about so far, is that as we start to create this platform capability around the edge and we start to transform those network themself to coincide telcos platform, we can start to capture those values today in a four g world and as five g comes along you just essentially evolve into the capabilities that that brings especially with regards to latency and bandwidth. Now some applications where slicing will be really important. Think about a medical operation where a doctor is consulting on a surgery in a remote location. Now if I know for sure that bandwidths going to be there, that doctor no longer has to be in the same location as that robotic equipment as an example. So the ability to have dedicated bandwidth which will come with five g, will be an important attribute that gets added. >> I mean the possibilities are really mind boggling. You mentioned stadiums. Now of course hopefully at some point we'll be able to go to football games again. But I mean the last decade was all about how big can you make the screen in the stadium versus this screen. This is where a lot of the action is going to be now. Replays and just the whole experience, ordering goods and services. And then of course hardened environments like oil rigs etc so really we're not just going to return to the last decade we've been talking about that a lot here. Go ahead please. >> An example that I like to mention just 'cause it kind of brings us all together, think about first responders. Now we're in the midst of the COVID thing but soon in California again unfortunately we'll probably get close to fire season. Think in a five g edge world what that might look like. So the minute that fire starts in some location in California, drones are in the air sending video down to an edge platform that's being analyzed to understand where that fire's going and importantly everything that's in it's path and how to best battle it. Sensors coming in from IoT centers in the area feeding in data. Our weather company app feeding in real time weather statistics, wind path, temperature changes, that are going to influence the way that that that fire performs. And now with the announcement we made Samsung just recently with our edge platform, the ability to have those first responders have sensors on them, Samsung devices that are measuring their vital signs and with the predictive models that are being built, we'll know whether that first responders' in distress or about to be in distress. The ability to scale our inbound communications capability digitally so that chat bots can handle this enormous increase in the amount of folks calling in to get information on what's happening in real time. And of course with the AI in that edge platform. Moving all of that physical equipment, the asset, humans, the first responders, in the optimal position at all times in order to get that fire out as soon as possible. I think it's a good example of how we can see these capabilities come together in a five g and edge world and allow us to get enormous value, saving lives, saving property, responding to an incident like that. >> I mean that's a great example of how you're going to put innovation into action 'cause you touched all points. Imagine the amount of data now that's being created and that example that you just gave, I mean it's just going exponential. Applying artificial intelligence, machine intelligence, and then the other phrase you used is real time. And we're talking about real time or near real time decisions actually being made potentially often times by the machines or in combination with humans so that these actions can be taken of course it's all occurring on an infrastructure that's sort of an expanding definition of cloud, not just on prem, not just hybrid, not just multicloud but now the edge. It's really going to be an exciting 10 years. >> You got it exactly right. And importantly, using that example, once that fire's put out, that edge platform can wind back down to where it was before the incident occurred. But all the intelligence that was gained during that, can be taken to the next incident has it happens. So this agility becomes really powerful 'cause we get the cumulative learning that happens in these models going forward. >> Amazing. So where can people go to get some more information on sort of IBM's edge approach? >> If you go to IBM.com, you'll see information on both on IBM edge solutions that we're putting forward into marketplace and what we're doing specifically with the telecommunication service providers to help them transform their networks to take advantage of this incredible opportunity. >> Well Steve, thanks so much for your time. Really great discussion. I appreciate you comin on and sharing with our community. >> My pleasure. Thank you. >> And thank you everybody. This is theCUBE's continuous coverage of IBM Think 2020, the Digital Event Experience. My name's Dave Vellante. Keep it right there, I'll be right back right after this short break. (relaxing music)

>>Hi and welcome back to Red Hat. Summit 2020. This is the Cube, and I'm your host. Stew minimum. We're talking so many topics. This event happening globally. We're treating our partners in the red hat executives where they are around the globe. And I guess right now is Nick Carr said, Who's the senior director of technology strategy with Red Hat, And Nic is coming to us. He's early in the Bahamas, speaking to us from his boat, though. Nick, pleasure to see you. And thanks so much for joining us. >>Very nice to meet you. Yeah, remote employees. And I enjoy that a lot. >>Absolutely. So we've been talking your team a lot. Of course. You know, many employees of Red Hat already were remote, but everyone now is working where they are. You're gonna be about a topic, of course, which is even more about riveted solutions. And where things are, we're going to talk about edge and five G before we get into the topic. It's a little bit about your background, how long you've been with red hat. And you know what? Your what your role is. >>So I joined right as a little more than five years ago after the acquisition. Off of all the companies that was working on open stack. Interesting technology. I've been in open source for the past 20 more years. Um, I was, uh, working miss of many distributions of Lennox over the years, so I consider myself in open source veteran. >>Excellent. I I remember that acquisition. We had the Cube at the open stack summit for many years on that, um, you know, new the company before the acquisition >>of the >>brand. And frankly, >>though, let's talk about it. First of all, you know, you talk about edge. Edge means different things to a >>lot of people >>are talking about it >>from a >>career perspective. You know, every customer in the Iot piece. Where does Red Hat into the whole notion of edge on? You know what kind of pieces of the portfolio? Yeah. >>So obviously, edge is about building an infrastructure that goes as far as possible to be as close as possible to where people are either producing or consuming data and building infrastructure as always, being the very heart off what Red Hat has been doing. And we've been growing. That's infrastructure capability. over time. So that means that today we feel the need to fulfill the requirements of those customers that want to extend their infrastructure to there. Because when we say the edge, we have to be countries that we're talking about. Like the layers of an onion more You dig into it. The more layers you find, the more particle case you have. There's no way there is a single. >>Yeah, no, you're absolutely right. So, you know, back in the open stack days, we talked a lot about in it, though Some of the barriers I know I've spoken to prizes. Who's, You know, we are of red Hat, though, you know, Maybe start there and help us understand. You know, where are we with the solution? Uh, talk about how five g fits into it. Of course. Everybody's talking about five feet. Well, that will take time, but help us understand where we are today. >>So, um, obviously for us, the edge years, just an extension of our open ivory clouds. Right? We have always been very vocal in saying that you need to be able to deploy the same workload in any place, and the edges are Justin extensions off these anyplace. So the same strategy that we've been developing first we use open stack, uh then with open shift and making open shift are both our development and our deployment platform for all types of workloads having open shift now, this report normally container based workloads, but also the visualization based workloads are exactly what we are doing at the edge. We want people to be able to deploy a single type of platform on various types of fruit brands managers globally. Ah is complete consistency so that there is no extra cost in maintaining those thousands, sometimes millions off added location into their existing infrastructure off course. In order to do that, we need to develop new tools to do the management to develop new AI or machine learning technology, to help people process not only the data coming from the platform, but also the management of the platform itself. We are reaching such scales that we wouldn't be able to do it. We've out. You are no from the platform yourself. >>Yeah, absolutely. And of course, scale is a relative before rise in. Yeah, I've talked to a couple of times. My understanding you've got news related. This that horizon went off? >>Yeah. This week Horizon has bean announcing ah, reinforce partnership between our two companies to help them heal their edge Platform. Ah, here we are talking about the first step in their edge platform which years? What we call the extension off the board we are talking about developing small data centers are going to be closer to the certainly. Um, And here we are talking about scales that, um can comprise to hundreds of data center, each having to 20 machines or more, um, to do all the processing of their future five g network and further, um, five g years, one off the enabler off edge. But it's also the reason for telcos to start deploying their edge network because the have a requirement to boot treatments off the information closer to where the five g antennas. And this is what we are developing. >>Alright. So, Nick, we talked. You've talked a minute ago about open stack and open fifth, help our audience understand a little bit. We've already talked a lot of customers. You don't. You can have one without the other, or you can layer off of the open stack when it comes to the the solution that you're talking about Verizon or ah, you know, other other service providers out there is it? Is it one is in both eyes that I've been there, Help us understand. >>So currently we have a complete shorts. We can do an edge platform. So Levi's open stack. You've got multiple customer doing that around the world. We can build an edge platform. We is open shift on top of the stack. But if we look at a future as we are, you know, designing it, we are looking at enabling simplicity and simplicity. Means deploying a single seeing open shift on to bare metal and have these bare metal platform deal we both vm and container so that you only have one AP I. You only have one management. You only have one thing to worry about. And since open shift and bark the OS, um, there is extreme simplicity in the methodology for updating or upgrading, and I think this is going to be a key point, making things simple, reducing the number of layers in your set. >>All right, that that really intrigued Nick, help us understand a little bit this ICO, Obviously any red hat doing is open source. It's how you're for that, you know, Red hat does. But you know how you're involved in the industry to help make the word that as edge solutions roll out that customers have flexibility in the first place. >>So you have multiple tee off partnership in this industry, you've got the partnership that are built around community and we are participating in numerous community, like the Lennox Foundation. Edge on many, many more. And this is where we are building the fundamental block off our future solutions we have. Partnership also is multiple vendor. Every time you're dealing with is a specific vertical. You will have a certain number of vendors that are going to be the one enabling 80% of the applications are going to be deployed, and that's okay for the edge. And then you have the partnership we made. We see our customers because the best source off requirements are always our customers. And that's something that we've now made a strong principle, which is to always find early adopters with when we are going to build a solution in a vertical sector on the horizon is one of them has been one of them. For what, a few years now and then replicate this success on to other customers of same sex. And we are reproducing this in the industry and manufacturing sector and in many other virtual. >>Excellent. Uh, you talked earlier about the open hybrid cloud. Obviously talk about they right, Wild help us understand, Nick. You know, edge and cloud. How do they actually go together? Many people. First of all, the people living article that was, you know, edge kills the cloud we've been talking about for a while. We know everything in i t is always additive. But how should customers on the surface but really be thinking about how edge cloud fit together >>in our design? The cloud and the edge is the same thing. You address the edge, you address the cloud, you should address your on premise art where the same way you use the same guy. And this driving FBI ease of communities, FBI, which we deliver through open ship. Um, soon. What is the difference? The difference is going to be who owns the edge, or we also machine running in your cloud who owns the machine running in your private data center. What network you're using, you're going to have Ah, a lot of constrained are going to be a bit more complex when you aren't yet. For example, you are sometimes going to go through the satellite connection. These huge delays in communication you're sometime going to put machines location that are absolutely not secure. So you need to have security layers. You're ensuring that nobody can remember these machines. These are you know it. But overall, once the deployment has done, we really, really on. People should consider that's their edge piece parts of their cloud or vice versa. >>Yeah, Nick, you brought up a lot of good points there. Security, of course. Critical. A one piece that I want to get your honest about. So we're spending a few years really looking at in a worker's process at the edge. What that's brought back core talk about AI work. Both generally understood praying things out at the edge. That's gonna happen. You know more of the core and then get out of the overall devices. What do you seeing where your customers But that overall, when it comes to their data. And >>from a technical perspective, data is the real real motivation about yet they are generating so much data that we are not able to process it anymore in a central location. So we have to process this data locally where it is generated. Or I suppose it's possible to where it is generated before sending, Let's say, a summary of these data or alerts or whatever the business process that pulls for to the center of operation. The use cases that we are demonstrating in this week, uh, that you can watch through the demo booth or you can watch increases. Ah, known presentation. Use the pays off manufacturer, which is installing sensors on many of the machine producing oh stuff. And when you have the right sensors like the vibration sensor or a temperature sensor, you can very easily develop knowledge off. Oh, this machine is going to break in a short amount of time. Maybe I should start scheduling some preventive maintenance on these machines, and you can do that by just actually leading the data and have humans read it. And you can do that a lot more efficiently. Training a machine learning algorithm This is what we are demonstrating that is processing the data and sending the alerts in real part when issues are discovered. Um, all this off course needs to be down in a very scalable fashion. Here we are talking about a use case where the customer may have 50 factories >>around the world. >>Are you updates all these machine learning models in all the factories when you have an update percent to learn about something you so data and data processing and now the eye. But big data are the heart off all of the use cases we, uh, discovered around old verticals for edge. And this is why we are now almost joining forces between the team working on AI. That's right out producing the open data hub and the team marking teams working on our solution. >>Yeah, I'm really glad you brought up manufacturing as the is one of the verticals. Look at their one of the turns and challenges we saw with all of that is yeah, some of the organization, Specifically, if you look at manufacturing, it could be an ot. Um, I'm curious is you're seeing solutions. Roll out your work, Aziz. How Customers are getting beyond those barriers. You know, some of the traditional silos where there was thoroughly collaborate. >>Well, it's always Ah, problem. Every time you introduce a change, you have to manage this in every project off, deploying something you anywhere well, fail if you do not account for the human factor and edge is no different in that. And when you're talking about the factory, if you're not directly talking with the people on the floor well, regarding their needs, you're only talking is a central guy. And you just arrived one day saying, Oh, everything is going to change. It's going to be a failure that the same way is a failure when the government make a decision without going through a consultative process before implementing it. So, um, nothing new, I would say. But as usual, And maybe because of the scale of edge, yeah, we will need to ensure that our customers are aware of those challenges that lay ahead of us. >>Alright, Well, next sounds Sounds like a lot of good progress. Been made definitely further breakout. What? From summit? You learn more. Thank you so much for joining. >>Thank you for having me >>all right. More coverage from the Cube at Red Hat Summit 2020. I'm screaming a man and as always what? Alright, Nick. Good stuff.

>>from the Silicon Angle Media office in Boston, Massachusetts. It's the Cube now here's your host, Dave Vellante. >>Hello, everyone. And welcome to this breaking analysis. We're here to assess the VM Ware v Sphere seven announcement, which is the general availability of so called Project Pacific. VM Ware has called this the biggest change to V sphere in the last 10 years. Now Project Specific Pacific supports kubernetes and natively in VM Ware environments. Why is this important? This is critical for multi and hybrid cloud because Kubernetes and its surrounding orchestration enable application portability and management. Yeah, as we've been reporting, VM Ware is one of the big players eyeing multi cloud, along with a crowded field of aspirants that include IBM with Red hat, Microsoft, Cisco, Google and a host of specialists in the ecosystem. Like how she and rancher as well play. Some players have focused in their respective stack swim lanes like security and data protection, storage, networking, etcetera. And with me to dig into this announcement is stew. Minutemen's Do is a senior analyst at Wiki Bond and co host of The Cube is too good to see you and let's get into it great to talk about this state. Okay, so the Sphere seven, what is being announced? And why is it relevant? >>Yes. So, David, as you said in the open, this is the general availability of what they talked about at VM World 2019 as Project Pacific. So it really is integrating kubernetes into V sphere. The VM ware, of course, will position this is that they're now enabling, you know, the 90% of the data centers around the world that have VM ware. Hey, your kubernetes enabled. Congratulations. You're cloud native. Everything like that. Only being a little facetious here. But this is very important. How do we get from where we were to live in this more cloud? Native environments. So containers in general and kubernetes specifically are being a first class citizen. There's a lot of work, Dave, and my understanding this has been going on for a number of years. You know, it's not like they just started working at this six months ago. A overhaul to how this works. Because it's not just we're going to stick a couple of containers on top of, you know, the guest operating system in the virtual machine. But there is a supervisor cluster for kubernetes at the hyper visor level. And there's a lot of, you know, in the weeds things that we're all trying to understand and figure out because you've got you know, we've got a hyper visor and you've got VM. And now you've got the containers and kubernetes on. Some of them are living in my data center. Some VM ware, of course, lives on multiple clouds like the VM ware on AWS. Solutions of this will go there on and, you know, how do I manage that? How does this impact my operations? You know, how did this change my application portfolio? Because, you know, the early value proposition for VM Ware always was. Hey, you're gonna put VM ware on there. You don't need to touch your applications. Everything runs like it did before you were running windows APS on a physical server. You move into virtual. It's all great. There's a lot of nuance and complexity. So when VM Ware says this is the biggest change in a decade probably is, I think back to you know, I remember when the fx 2.0, rolled out in V motion really changed the landscape. That was big V balls. Move to really ah storage. To really understand that architecture and really fix storage was was a huge undertaking that took many years. This this definitely stacks up with some of those previous changes to really change the way that we think about VM Ware. I think the advertising you have even seen from being where some places is don't think of them as VM ware their cloud where our container ware with like because vm zehr still there. But VM Ware is much more than VMS today, >>so this feels like it's bm were trying to maintain its relevance in a cloud native world and really solidify its because, let's face it, VM Ware is a platform that Pat Gelsinger's has ride. The Waves tried many times in many angles to try to ride the cloud wave, and it's finally settled on the partnerships with AWS specifically. But others on DSO really Is this their attempt to become cloud native, not get left behind and be cloud naive? His many say >>Yeah, great question, David. Absolutely. There's the question as to you know what's happening with my applications, you know lots of customers. They say, Well, I'm just going to satisfy the environments. Watched the huge growth of companies like service now workday. Those applications, well, customers don't even know what they live on. Do they live on virtualization? Environment is a containers I don't need to worry about because SAS takes care of that. If I'm building modern applications, well, I'm probably not starting with VMS. Containers are the way that most people are doing that. Or they might even be going serverless now if we take these environments. So how does VM ware make sure that they have the broadest application support? Kubernetes really won the container orchestration wars on. And this is a way that VM ware now can enable customers to move down that path to modernize their environments on. And what they wanna have is really some consistency between what's happening in the cloud and happening in the environments that they control >>themselves. Vm ware saying that containers in our first class citizen within v sphere what does that mean? Why is that important? First of all, are they really And what does that mean? And why is that important? >>Yes. So, Dave, my understanding is, you know, absolutely. It's their, You know, the nuances that you will put there is. You know, we're not just running bare metal servers with Lennox and running containers on top of it. It is. You're still sitting on top of the hyper visors. One of the things I'm trying to understand when you dig down is you know what? The device driver level VM ware always looked a little bit like Linux. But the people that use it and operate it, they're not letting people Dave, these, you know, the OS. The number one os that always ran on VM ware was Windows and the traditional applications that ran there. So when we talk about containers and we're enabling that in a kubernetes environment, there are some questions about how do we make sure that my applications get certified? Dave, you got a lot of history knowing things like s ap and Oracle. I need to make sure that we've tested everything in this works. This is not what we were running traditionally in VM ware and VM ware. Just thanks. Hey, v Sphere seven, turn the crank. Everything certified Well, I would tell customers make sure you understand that your application has been tested, that your Eyes V has certified this environment because this is definitely, as VM Ware says, a huge architectural change. So therefore, there's some ripple effects to make sure that what I'm doing in this environment stays fully supported. Of course, I'm sure VM Ware is working with their huge ecosystem to make sure that all the pieces or environment you mentioned things like data protection. We absolutely know that VM Ware is making sure the day one the data protection plugs in and supported in these environments when you're using the kind of kubernetes persona or containers solutions in V sphere. >>Well, this brings me to my next question. I mean, we were talking to Bernard Golden the other day and he was saying, You know, Kubernetes is necessary for multi cloud, but it's insufficient. And so this seems to me to be a first step and, as I say, VM ware maintaining and growing its relevance. But there's gonna be a roadmap here that goes beyond just containers and portability. There's other management factors you mentioned security of enabling the ecosystem to plug in. So maybe talk about that a little bit in terms of what's necessary to really build this out over the next >>decade. And actually, it's a great point. So, first of all, you know, V. Sphere, of course, is the core of VM Ware's business. But there's only a piece of the overall portfolio said this lives in. I believe they would consider this part of what they call their Tansu family. Tando is their cloud native overarching piece of it, and one of the updates is their product hands admission control. Which of the existing product really came out of the Hep D Oh acquisition is how we can really manage any kubernetes anywhere, and this is pure software. Dave. I'm sure you saw the most recent earnings announcement from VM Ware, and you know what's going sass. What's going subscription? VM Ware is trying to build out some of their software portfolio that that isn't kind of the more traditional shrink wrap software, so Tan Xue can manage any kubernetes environment. So, of course, day one Hey, obviously or seven, it's a kubernetes distribution. Absolutely. It's going to manage this environment and but also if I've got Cooper days from azure kubernetes from Amazon communities from other environment. Tanja can manage across all of those environments. So when when you're what VM Ware has always done. If you think back in the early days of virtualization, I had a lot of different servers. How do I manage across those environments? Well, VM ware was a layer that lived across them. VM Ware is trying to do the same thing in the cloud. Talk about multi cloud. And how do I manage that? How do we get value across them? Well, there's certain pieces that you know VM Ware is looking to enable with their management software to go across them. But there are a lot of other companies, you know, Amazon Google actually not Amazon yet for multi cloud. But Microsoft and Google absolutely spent a lot of time talking about that in the last year. A swell as you mentioned. Companies like Rancher and Hashi Corp absolutely play across What Lots of these multi cloud. Well, >>let's talk about the competition. Who do you see is the number one competitors >>Well, so the number one competitor absolutely has to be red hat, Dave. So you know, when I've been in the kubernetes ecosystem for a number of years for many years. When I talk to practitioners, the number one, you know what kubernetes you're using? Well, the answer for many years was, Well, I'm grabbing it, you know, the open source and I'm building my own stack. And the reason customers did that was because there wasn't necessarily maturity, and this was kind of leading edge, bleeding edge customers in this space. The number two besides build my own was Red Hat was because I'm a red hat customer, a lot of Lennox tooling the way of building things the way my application developers do. Things fit in that environment. And therefore, that's why Red Hat has over 2000 open shift customers leading distribution for Kubernetes. And you know, this seems purely directly targeted at that market. That red hat did you know it was a big reason why IBM spent $34 billion on the Red Hat acquisition is to go after this multi cloud opportunity. So you know, absolutely this shot across the bow because Red Hat is a partner of VM Ware's, but absolutely is also a competitive >>Well, Maritz told me years ago that's true. We're with everybody and you could see that playing out. What if you look at what VM Ware could do and some of their options if they gave it away, that would really be a shot across the bow at open shift, wouldn't it? >>Yeah, absolutely, Dave, because kubernetes is not free if you're enabling kubernetes on my Google environment, I, you know, just within the last week's awesome things that were like, Okay, wait. If you're testing an environment, yes, it is free. But, you know, started talking about the hourly charges for the management layer of kubernetes. So you know kubernetes again. A color friend, Cory Quinn. Communities absolutely is not free, and he will give you an earful and his thoughts on it s o in Amazon or Google. And absolutely, Dave, it's an important revenue stream for red hat. So if I'm vm ware and you know, maybe for some period of time, you make it a line item, it's part of my l. A. You know, a good thing for customers to look out for is when you're renegotiating your l a toe, understand? If you're going to use this, what is the impact? Because absolutely, you know, from a financial standpoint, you know, Pat Gelsinger on the VM Ware team has been doing a lot of acquisitions. Many of those Dave have been targeted at this space. You know, not to step Geo, but a bit NAMI. And even the pivotal acquisition all fit in this environment. So they've spent billions of dollars. It shouldn't be a net zero revenue to the top line of what VM Ware is doing in the space. >>So that would be an issue from Wall Street's perspective. But at the same time, it's again, they're playing the long game here. Do we have any pricing data at this point? >>So I still have not gotten clear data as to how they're doing pricing now. >>Okay, Um, and others that are in there and in the mix. We talked about Red Hat. Certainly Microsoft is in there with Arc. I've mentioned many times Cisco coming at this from a networking perspective. But who else do you see and then Antos with Google? >>Yeah. And you know, Dave, all the companies we're talking about here, you know, Pat Gelsinger has had to leverage his intel experience to how to balance that line between a partner with everybody but slowly competing against everybody. So, you know, we've spent many hours talking about the VM Ware Amazon relationship. Amazon does not admit the multi cloud a solution yet and does not have a management tool for supporting all of the kubernetes environment. But absolutely Microsoft and Google do. Cisco has strong partnerships with all the cloud environment and is doing that hybrid solution and Dave Justice nothingto expand on a little bit there. If you talk about V sphere, you say, Okay, Visa or seven trolling out Well, how long will it take most of the customer base to roll to this environment? There will be some that absolutely want to take advantage of kubernetes and will go there. But we know that is typically a multi year process to get most of the install base over onto this. And if you extend that out to where VM Ware is putting their solution into cloud environments, there's that tension between, you know, Is there a match actually, between what I have in my data center and what is in the managed environment managed by VM Ware and Amazon, or manage for to support some of the other cloud environment. So the positioning always is that you're going to do VM Ware everywhere, and therefore it's going to be consistent everywhere. Well, the devil's in the details because I have control on what's in my data center, and I might have a little bit less control to some of those managed services that I'm consuming. So absolutely something to keep a close eye on. And not just for VM, where everybody is having these concerns. Even if you talk about the native kubernetes distributions, most of the kubernetes services from the cloud providers are not, you know, immediately on the latest revision of kubernetes, >>right, So Okay, well, let's let's talk about that. Remember when open Stack first came out? It was a Hail Mary against Amazon. Yeah, well, the new Hail Mary and looks like it has more teeth is kubernetes right, because it allows portability and and and of course, you know Amazon doesn't publicly say this, but it's not. That's not good for Amazon. If you're reporting things, applications, moving things around, moving them out of the Amazon cloud, and that makes it easier. Of course, Amazon does support kubernetes right, But you've got >>alternatives. So, David, it's fascinating. So I've talked to many practitioners that have deployed kubernetes and one of the top reasons that they say that why they're using Kubernetes is so they have options with the cloud. When you also ask them what cloud they're running, they're running Amazon. Did they have planned to move off of it? Well, probably not. I had a great customer that I didn't interview with that one of the Cube con shows, and they actually started out with Azure just because it was a little further head with kubernetes and then for the services they wanted. They ended up moving to AWS and Dave. It's not a click a button and you move from one kubernetes to another. You need toe match up and say, Okay, here's the five or six services I'm using. What are the equivalent? What changes do I need to make? Multi cloud is not simple. Today, I mentioned Hashi Corp is one of those companies that help people across these environments. If you have haji solution and you're managing across multiple clouds, you look in the code and you understand that there's a lot of difference between those different clouds, and they simplify that. But don't eliminate it. Just it is not. There is not a way today. This is not a utility when you talk about the public cloud. So you know Kubernetes absolutely is existentially a little bit of a threat to Amazon but Amazon still going strong in that space. And you know that the majority of customers that have deployed kubernetes in the public cloud are doing it on Amazon just because of their position in the marketplace and what they're. >>So let's double click on that. So Jassy, an exclusive interview with John Furrier before last year's re invent, said, Look, we understand there's a lot of reasons why people might choose multiple clouds, you know, go through them in a developer preference. And I think I think, you know, people want o optionality and reduce lock in potentially. But I've always said, by the way, just as an aside, that that the risk of lock in it is far down on the list relative to business value, people will choose business value over over, you know, no lock in every time. About 15% of the customers you might not agree. Nonetheless, Jassy claimed that typically when you get into a multiple cloud environment, he didn't use the term multi cloud that it's it's not a 50 50. It's a premier primary cloud supplier. So might be 70 30 or 80 20 or even 90 10. But it's really that kind of, you know, imbalance. First of all, do you see that? And then what does that mean for how they approach of this space? Multi cloud and in particular. >>So I'm sorry. You're asking how Amazon should approach the space. And you've said that I don't think they'll >>eventually enter this market place. >>Yeah, you know, absolutely, Dave. You know, first of all, in general, yes, I do agree. It is not. There are certain financial companies that, you know, have always chosen two of everything. Because for regulation and you know certain we need to protect ourselves. We're gonna have to suppliers. We're going to keep them as even as possible. But that is a corner case. Most customers I have a primary cloud. That's what I'm doing. That what I t tries to get everybody on and you need to have Is there a reason why you want to use a secondary or tertiary cloud because there's a service that they need. Of course, Google. You often run it. It's like, Oh, well, there's certain data services that they're doing well And, of course, the business productivity solutions that Microsoft's doing where the relationship with Oracle that are driving people towards Microsoft. But just as we saw Amazon soften on their hybrid solutions, we spent a lot of time at re invent talking about all the various hybrid solutions. Um, since their customers are going to have multiple clouds on and even you take most of their customers that have M and a involved you buy another company, they might be using another cloud. As Microsoft's position in the marketplace has grown, you would expect that Amazon would have not just migration services but management services to match what customers need, especially in this kubernetes environment, seems that it seems a natural fit for them. It's possible they might just leverage, you know, partnerships with red hat VM ware, you know, in some of the other players for the time being. But if the market gets big enough and customers are asking for it, that's usually when Amazon response >>So let's let's wrap with what this means to the customer. And I've said that last decade really multi cloud was a symptom of multi vendor and not so much of the strategy that's changing. You know, clearly, jokes CIOs are being called in to clean up the crime scene on do you know, put in edicts corporate edicts around security and governance and compliance and so forth. So it started to become a complicated situation for a lot of companies. We've said that multi cloud is gonna it's gonna be they're going. People are going to put the right war load and the right cloud, etcetera, and this advantages to certain clouds. But what should customers be thinking specifically as it relates to v. Sphere seven? >>Yes. So, Dave, the biggest thing I would say that people need to look at it is that understanding in your organization that that boundary and line between infrastructure and application people have often looked at you looked at the ascendancy of VM Ware, Andi V. M's and then what's happening with cloud and containers. And we think of it from an infrastructure standpoint that I'm just changing the underlying pieces. This is where it lives and where I put things. But the really important thing is it's about my data and my applications, Dave. So if I'm moving an application to a new environment, how do I take advantage of it? You know, we don't just move it to a new environment and run it the same way we were doing it. I need to take advantage of those new environments. Kubernetes is involved in infrastructure, but the real piece is how I have my application, my developers, my app. Dev's working on this environment and therefore it might be that if VM Ware's the right environment, I'm doing a lot of it that the development team says, Hey, I need you to give me a pool and provisioned this for me and I can have my sandbox where I can move really fast. But VM Ware helped initially customers when they went from physical to virtual, move faster. From an infrastructure standpoint, what it needs to do to really enable this environment is help me move faster on the application side. And that's a big gap from VM. Ware's history is where the pivotal people and hefty O people and bit NAMI and all the new people are helping along to help that whole cloud native team. But that is a big shift from customers. So for this to be successful, it's not just, oh, the virtualization admin. He upgraded to the new thing. He made some changes and said, Okay, hey, I can give you a kubernetes cluster when you need it. It's really understanding what's going to happen on the application side in a lot of that is going to be very similar to what you're doing in cloud environments. And I think this is Dave often where your customers, they say, Oh, well, I did that cloud and it was too expensive and it was too hard, and I repatriated. Everything else is, well, you probably didn't plan properly and you didn't understand what you're getting yourself into. And you jumped into the deep end of the pool and oh, wait, I forgot how to learn how to swim. So you know, that is where we are. You know, Dave, you know the technology parts. Always the easiest piece. It's getting all of the organizational and political things sorted out. And you know the developer we know how important that is, we're seeing. It's great to see VM Ware pushing faster in this environment. Kudos to them for how fast they moved. Project Pacific to G. A. That is really impressive to see and can't wait to hear the customers roll out because if this is successful, we should be hearing great transformation stories from customers as to how this is enabling their business, enabling them to move faster on. You know, that has been what, one of the favorite stories that I've been telling with customers on the Cube last couple of years. >>The vast majority of VM Ware's business, of course, is on print, and essentially they're doing here is enabling developers in their customer base and the half a 1,000,000 customers to really develop in a cloud native manner. The question is, you know, from a ah, from a cultural standpoint, is that actually gonna happen? Or the developers gonna reject the organ and say, No, I want to develop in AWS or Microsoft in the cloud. I think VM Ware would say, We're trying to embrace no matter where they want to develop, but they're still going to be. That's interesting organizational tension or developer attention in terms of what their primary choices is. They're not. >>Yeah, Dave, Absolutely. We've been saying for years. That cloud is not a location. It is an operating model. So this is helping to enable that operating model more in the data center. There's still questions and concerns, of course around, you know, consumption on demand versus you know, whether whether you've bought the entire thing as more and more services become available in the public cloud, are those actually enabled to be able to be used, you know, in my data center hosted environment. So you know, this story is not completed, but we're definitely ready. I believe we're saying it's the multi clouds Chapter three of what? We've been watching >>you and you're seeing a major tam expansion yet again from VM Ware that started with the NSX. And then, of course, went in tow networking and storage. And now they've got a cloud security division. We're talking about the the cloud native capabilities here and and on and on, it goes to thanks for helping us break this VC seven announcement down and good job fixed. All right. And thank you for watching everybody. This is Dave Volante for stew Minimum. We'll see you next time on the Cube. >>Yeah,

>>long from Las Vegas. It's the Q covering a ws re invent 2019. Brought to you by Amazon Web service is and in along with its ecosystem partners. >>Hey, welcome back to the cubes. Coverage of AWS 19 from Las Vegas. This is Day two of our coverage of three days. Two sets, lots of cute content. Lisa Martin here with Justin Warren, founder and chief analyst. A pivot nine. Justin and I are joined by a couple of guests New to the Cube. We've got David Meister next to meet Global head of sales for Io Tahoe. Welcome. Eddie Edwards with a cool name. Global Data Service is director from Direct Energy. Welcome, Eddie. Thank you. Okay, So, David, I know we had somebody from Io Tahoe on yesterday, but I'd love for you to give her audience an overview of Io Tahoe, and then you gotta tell us what the name means. >>Okay. Well, day pie stir. Io Tahoe thinks it's wonderful event here in AWS and excited to be here. Uh, I, oh, Tahoe were located in downtown on Wall Street, New York on and I Oh, Tahoe. Well, there's a lot of different meanings, but mainly Tahoe for Data Lake Input output into the lake is how it was originally meant So But ah, little background on Io Tahoe way are 2014. We came out way started in stealth came out of stealth in 2017 with two signature clients. When you're going to hear from in a moment direct energy, the other one g e and we'll speak to those in just a moment I owe Tahoe takes a unique approach way have nine machine learning machine learning algorithms 14 future sets that interrogates the data. At the data level, we go past metadata, so solving that really difficult data challenge and I'm gonna let Eddie describe some of the use cases that were around data migration, P II discovery, and so over to you >>a little bit about direct energy. What, you where you're located, What you guys do and how data is absolutely critical to your business. Yeah, >>sure. So direct energy. Well, it's the largest residential energy supplier in the er us around 5000 employees. Loss of this is coming from acquisitions. So as you can imagine, we have a vast amount of data that we need some money. Currently, I've got just under 1700 applications in my portfolio. Onda a lot. The challenges We guys are around the cost, driving down costs to serve so we can pass that back onto our consumers on the challenge that with hard is how best to gain that understanding. Where I alter whole came into play, it was vainly around off ability to use the products quickly for being able to connect to our existing sources to discover the data. What, then, that Thio catalog that information to start applying the rules around whether it be legislation like GDP, are or that way gets a lot of cases where these difference between the states on the standings and definitions so the product gives us the ability to bring a common approach So that information a good success story, would be about three months ago, we took the 30 and applications for our North America home business. We were able to running through the product within a week on that gave us the information to them, consolidate the estate downwards, working with bar business colleagues Thio, identify all the data we don't see the archival retention reels on, bring you no more meaning to the data on actually improve ourselves opportunities by highlights in that rich information that was not known >>previously. Yes, you mentioned that you growing through acquisition. One thing that people tend to underestimate around I t. Is that it's not a heterogeneous. It's not a homogeneous environments hatred genius. Like as soon as you buy another company, you've got another. You got another silent. You got another day to say. You got something else. So walk us through how iota who actually deals with that very disparity set of data that you've night out inherited from just acquiring all of these different companies? >>Yeah, so exactly right. You know, every time we a private organization, they would have various different applications that were running in the estate. Where would be an old article? I say, Hey, sequel tap environment. What we're able to do is use the products to plug in a name profile to understand what's inside knowledge they have around their customer base and how we can number in. That's in to build up a single view and offer additional products value adding products or rewards for customers, whether that be, uh on our hay truck side our heat in a ventilation and air con unit, which again we have 4600 engineers in that space. So it's opening up new opportunities and territories to us. >>Go ahead, >>say additionally to that, we're across multiple sectors, but the problem death by Excel was in the financial service is we're located on Wall Street. As I mentioned on this problem of legacy to spirit, data, sources and understanding, and knowing your data was a common problem, banks were just throwing people at the problem. So his use case with 1700 applications, a lot of them legacy is fits right into what we d'oh and cataloging is he mentioned. We catalogue with that discover in search engine that we have. We enable search cross enterprise. But Discovery we auto tag and auto classify the sensitive data into the catalog automatically, and that's a key part of what we do. And it >>was that Dave is something in thinking of differentiation, wanting to know what is unique about Iota. What was the opportunity that you guys saw? But is the cataloging and the sensitive information one of the key things that makes it a difference >>Way enabled data governance. So it's not just sensitive information way catalog, entire data set multiple data sets. And what makes us what differentiates us is that the machine learning way Interrogate in brute force The data So every single so metadata beyond so 1,000,000,000 rose. 100,000 columns. Large, complex data sets way. Interrogate every field value. And we tell you what this looks like A phone number. This looks like an address. This looks like a first name. This looks like the last name and we tagged at to the catalog. And then anything that sensitive in nature will color coded red green, highly sensitive, sensitive. So that's our big differentiator. >>So is that like 100% visibility into the granularity of what is in this data? >>Yes, that's that's one of the issues is who were here ahead of us. We're finding a lot of folks are wanting to go to the cloud, but they can't get access to the data. They don't know their data. They don't understand it. On DSO where that bridge were a key strategic partner for aws Andi we're excited about the opportunity that's come about in the last six months with AWS because we're gonna be that key geese for migration to the cloud >>so that the data like I love the name iota, How But in your opinion, you know, you could hear so many different things about Data Lake Data's turning into data Swamp is there's still a lot of value and data lakes that customers just like you're saying before, you just don't know what they have. >>Well, what's interesting in this transition to one of other clients? But on I just want to make a note that way actually started in the relational world. So we're already a mess. We're across header genius environment so but Tahoe does have more to do with Lake. But at a time a few years back, everybody was just dumping data into the lake. They didn't understand what what was in there, and it's created in this era of privacy, a big issue, and Comcast had this problem. The large Terry Tate instance just dumping into the lake, not understanding data flows, how they're data's flowing, not understanding what's in the lake, sensitivity wise, and they want to start, you know they want enable b I. They want they want to start doing analytics, but you gotta understand and know the data, right? So for Comcast, we enable data ops for them automatically with our machine learning. So that was one of the use cases. And then they put the information and we integrated with Apache Atlas, and they have a large JW aws instance, and they're able to then better govern their data on S O N G. Digital. One other customer very complex use case around their data. 36 e. R. P s being migrated toe one virtually r p in the lake. And think about finance data How difficult that is to manage and understand. So we were a key piece in helping that migration happen in weeks rather than months. >>David, you mentioned cloud. Clearly weird. We're at a cloud show, but you mentioned knowing your data. One of the aspect of that cloud is that it moves fast, and it's a much bigger scale than what we've been used to. So I'm interested. Maybe, Eddie, you can. You can fill us in here as well about the use of a tool to help you know your data when we're not creating any less stated. There's just more and more data. So at this speed and this scale, how important is it that you actually have tooling to provide to the to the humans who have to go on that operate on all of this data >>building on what David was saying around the speed in the agility side, you know, now all our information I would know for North America home business is in AWS Hold on ns free bucket. We are already starting work with AWS connect on the call center side. Being able to stream that information through so we're getting to the point now is an organization where we're able to profile the data riel. Time on. Take that information Bolts predict what the customers going going to do is part that machine learning side. So we're starting to trial where we will interject into a call to say, Well, you know, a customer might be on your digital site trying to do a journey. You can see the challenges around data, and you could Then they go in with a chop using, say, the new AWS trap that's just coming through at the moment. So >>one of the things that opportunities I'm here. Sorry, Eddie is the opportunity to leverage the insights into the data to deliver more. You mentioned like customer words, are more personalized experience or a call center agent. Knowing this is the problem of this customer is experiencing this way. Have tried X, y and Z to resolve, or this customer is loyal to pay their bills on time. They should be eligible for some sort of reward program. I think consumers that I think amazon dot com has created us this demanding consumer that way expect you to know us. I expect you to serve us up things that you think we want. Talk to me about the opportunity that I owe Ty was is giving your business to be able to delight customers in ways that you probably couldn't even have predicted? >>Well, they touched on the tagging earlier, you know, survive on the stunned in the data that's coming through. Being able to use the data flow technology on dhe categorizing were able than telling kidding with wider estate. So David mentioned Comcast around 36 e. R. P. You know, we've just gone through the same in other parts of our organization. We're driving the additional level of value, turning away from being a manually labor intensive task. So I used to have 20 architects that daily goal through trying to build an understanding the relationship. I do not need that now. I just have a couple of people that are able to take the outputs and then be able to validate the information using the products. >>And I like that. There's just so much you mentioned customer 360. Example at a call centre. There's so much data ops that has to happen to make that happen on. That's the most difficult challenge to solve. And that's where we come in. And after you catalogue the data, I just want to touch on this. We enable search for the enterprise so you're now connected to 50 115 100 sources with our software. Now you've catalogued it. You profiled it. Now you can search Karen Kim Kim Smith, So your your your engineers, your architect, your data stewards influences your business analysts. This is folks can now search anything they want and find anything sensitive. Find that person find an invoice, and that helps enable. But you mentioned the customer >>360. But I can Also. What I'm hearing is, as it has the potential to enable a better relationship between I t in the business. >>Absolutely. It brings those both together because they're so siloed. In this day and age, your data siloed and your business is siloed in a different business unit. So this helps exactly collaborate crowdsource, bring it all together. One platform >>and how many you so 1700 applications. How many you mentioned the 36 or so air peace. What percentage? If you can guess who have you been able to reduce duplicate triplicate at center applications? And what are some of the overarching business benefits that direct energy is achieving? >>So incentive the direct senator, decide that we're just at the beginning about journey. We're about four months in what? We've already decommissioned 12. The applications I was starting to move out to the wider side in terms of benefits are oh, I probably around 300% of the moment >>in a 300% r A y in just a few months. >>Just now, you know you've got some of the basic savings around the story side. We're also getting large savings from some of the existing that support agreements that we have in place. David touched on data Rob's. I've been able to reduce the amount of people that are required to support the team. There is now a more common on the standing within the organization and have money to turn it more into a self care opportunity with the business operations by pushing the line from being a technical problem to a business challenge. And at the end of the day, they're the experts. They understand the data better than any IittIe fault that sat in a corner, right? So I'm >>gonna ask you one more question. What gave you the confidence that I Oh, Tahoe was the right solution for you >>purely down Thio three Open Soul site. So we come from a you know I've been using. I'll tell whole probably for about two years in parts of the organization. We were very early. Adopters are over technologies in the open source market, and it was just the ability thio on the proof of concept to be able to turn it around iTunes, where you'll go to a traditional vendor, which would take a few months large business cases. They need any of that. We were able to show results within 24 48 hours on now buys the confidence. And I'm sure David would take the challenge of being able to plug in some day. It says on to show you the day. >>Cool stuff, guys. Well, thank you for sharing with us what you guys are doing. And I have a Iot Tahoe keeping up data Lake Blue and the successes that you're cheating in such a short time, but direct energy. I appreciate your time, guys. Thank you. Excellent. Our pleasure. >>No, you'll day. >>Exactly know your data. My guests and my co host, Justin Warren. I'm Lisa Martin. I'm gonna go often. Learn my data. Now you've been watching the Cube and AWS reinvent 19. Thanks for watching

>>live from Orlando, Florida It's the cue covering Microsoft Ignite Brought to you by Cho He City >>Welcome back, everyone. And welcome to the cubes live coverage of Microsoft Ignite Here in Orlando, I'm your host, Rebecca Night, along with my co host Stew Minimum. We're joined by Nicholas Djerassi. Moto's He is a cloud computing evangelist at Red Hat. Thank you so much for coming on the Cube. It's a pleasure. Thank you. So tell us a little bit about what you do at Red Hat. >>So I work with a lot of red, have partners really trying to foster the ecosystem and build red have products and solutions that can actually be deployable, repeatable for different customers. So different verticals. Financial health care doesn't really matter. For the most part, I try and just focus on cloud computing and really just evangelizing a lot of our technologies that we have. >>Okay, so So what are the kinds of things you're doing here at ignite? >>So I've been spending a lot of time actually working with some of the partners, like a center IBM. We've been doing a bunch of different webinars a little bit of hands on workshops that kind of educating people about distributed computing edge computing on dhe some of the technologies that we've been working along with Microsoft. So, uh, co engineering of sequel server The man is service offering that we're doing with open shift, which is our enterprise great kubernetes platform along many other >>different things. So So, Nicholas, you know, it's been a couple of years now that we've gotten over some of the gas. Wait. Microsoft has not said that, you know, we're killing the penguins, you know, off on the side. I was in Boston for Red Hat Summit. Tatiana Della's up on stage there, you know, Red hat. You know he's not hiding at the show. So bring us inside. You know where customers deployments are happening where engineering efforts are working together. You know, we know we've been hearing for years red hats in all of the clouds and partnering all of the merit. So what? What, you know, different or special, about the Microsoft relationship? >>I mean, honestly, I think the relationship is just evolving and growing because our customers were asking for it right there, going towards hybrid and multi cloud type of strategies. They want to be able to take advantage of, you know, running rail within their own data. Centers were running rails specifically on top of Microsoft Azure, but they're also looking at other club service providers. I think it's gonna be mandated eventually at some point in time where customers are gonna start looking at diversification when it comes to running applications, wherever it makes sense, taking advantage of different you know, cloud end of service is different providers. So we've been getting a lot of time like understanding what their needs are and then trying to build the engineering to actually address those needs. I think a lot of that has really come from the co engineering that we have going on. So we have a red head engineer sitting alongside bikers, off engineers, spending a lot of time building things like the Windows distraction layer wsl things along those lines, All >>right, so I'll be a Q Khan in a couple of weeks and kubernetes still, a lot of people don't really understand where it fits Way have been saying in a Cuban eight is gonna be baked into every platform. Red hat, of course, is not really a major contributor but has a lot of customers on open shift. We had Microsoft, you know, this week, talking about as your arc is in preview. But you know, they're they're the David Taunton who does partnership, Engagement says. You know, this does not mean that we will not continue to partner with open shift in the best place to run open shift is on azure. It's the most secure. It's the best. So help us understand his toe. You know where this fits In the overall discussion of that multi hybrid cloud that we were talking about earlier. I >>think everybody wants kind of a single pane of glass for manageability. They want ability to actually look and see where their infrastructure is being deployed. One of the pitfalls of moving to the cloud is the fact that it's so easy to spend a resource is that a lot of times we lose track of where these resource is. Our or individuals leave companies, and when they leave, cos they leave behind a lot of leftover items and instances, and that becomes really costly over a period of time. Maybe not so bad if you have, you know, 100 or 500 instances. But when you talk to some of these enterprise customers that are running 110,000 instances and spending millions of dollars a month, it could get very costly. And not only that, but it could also be a security risk is well, >>so let's talk about security. What kinds of conversations are you having with regard to security and data protection at this conference? >>So you know, one of the biggest things that we've had a lot of customers asking about his redhead insights so ready in sizes away it's a smart management application that actually ties into looking at either workloads or configuration management. It could actually tell you if you have a drift. So, for example, let's say you install sequel server on well, and you miss configure it. You leave the admin account running on it, it can actually alert you and make recommendations for remediation. Or maybe in general, you're using you know, S E. Lennox is disabled. The things along those lines so insights can actually look into, uh, the operating system or the applications and tell you if there's miss configurations all right, >>a lot of discussion about developers here, You know, day to keynote was all about, you know, AP Dev And, like Sathya have been a lot of time talking about the citizen developer. Seems like that would be an intersection between what red hats doing in and Microsoft. >>Um, so I would say, you know, we're obviously very developer first focused right when we built things like Open Shift Way kind of. We're thinking about developers. Before you were thinking about operations, and later on, we actually had to build more of the operations aspects into it. Now, like, for example, in open shift, there's two different portals. There's one for the developer Focus and one for the I T admin focus with operations groups because they want to see what's going on. Developers don't really care specifically about seeing the distraction of where things are. They just want to deploy their code, get it out the door as quickly as they can, and they're really just not too concerned about the infrastructure component pieces. But all of these developers, they want to be ableto right there, applications right there code and deploy it essentially anywhere and everywhere and having the easiest process and We're really just trying to make that as simple as possible, like visual studio plug ins that we have for open shift, you know, Eclipse G and other things. So really, I mean, Red has always been very developer focused first, >>so does that seeing Microsoft Satya Nadella up on the stage talking about this developer first attitude that Microsoft is really embracing the developer. And, as you said at development for all that does seem like a bit of a cultural shift for Microsoft much more aligned with the red hat way and sort of open source. So are you talking about that within without your cut with your colleagues? That red hat, about the change that you've seen the evolution of Microsoft? >>Absolutely. I mean, if you look at, like Microsoft, the contributions that they're putting towards, like kubernetes or even contribution towards open shift, it's It's amazing, right? I mean, it's like the company's gonna complete 1 80 from the way that they used to be. There's so much more open the acquisition of Like Get Hub, for example, all these different changes, it's it's amazing. He's done amazing things with the company. I can't say enough positive things about all the wonderful things that he's done. So >>all right, so Nicholas Red Hat has an interesting position in the marketplace because you do partner with all of the clouds on the environment. While IBM is now the parent owner of Red Hat and they have a cloud, your customers touch all of them. I'm not gonna ask you to competitively analyze them. But when you're talking to customers that are choosing Azure, is there anything that calling out as to why they're choosing Microsoft where you know they have, you know, a advantage of the marketplace or what is drawing customers to them on then? Of course, redhead. With that, >>I think Microsoft is more advanced when it comes to artificial intelligence and machine learning. A, I and ML and computing. I think they're light years ahead of everyone else at this point in time. I think you know, Amazon and Google are kind of playing a little bit of catch up there, Um, and it's showing right. If you look at the power platform, for example, customers are embracing that. It's just it's fantastic looking at a lot of the changes that they've implemented and I think it's very complimentary toe the way that people are starting to build their applications. Moving towards distributed infrastructures, Micro Service's and then obviously cloud native service is as well >>in terms of the future will be. We are really just scratching the surface when it comes to to the cloud. What do you see 5 10 years from now in terms of growth rates and also in terms of the ways in which companies are using the cloud. >>So I kind of like Thio equate it towards, like, the progression that we've had with cars. I know it sounds so simple, but, you know, we went from steam engine to regular piston engines, and now we've gotten to a point where we have electric cars and there's gonna be self driving cars. I think we're gonna get to a point where code is gonna be autonomous in a sense, right self correcting ability to actually just write code and deploy it. Not really having to worry about that entire infrastructure layer. Everybody's calling it server lists. There's always gonna be a server per se, but I think we're gonna have a point where next 5 to 10 years that all of that is gonna be completely abstracted away. It's just gonna be focused on writing the code and machine learning is gonna help us actually evolve that code and make it run faster and make it run better. We're already seeing huge benefits. And when it comes to machine learning and the big data analytics and things on those lines, it's just natural progression. All right, >>love, you know what's top of mine from the customers that you're talking to Earth event. Any new learning is that you've had or, you know, things that have kind of caught your attention. >>I think the biggest thing, honestly, is really been them. The multi cloud Polly Cloud methodology that everybody seems to be embracing. It seems like every customer I'm talking to is looking at trying to avoid that vendor lock and per se, but still have that flexibility to deploy their applications wherever and still utilize cloud Native Service's without actually specifically having to, you know, go completely open source >>and one of the challenges there is every cloud. I need different skills to be able to do them. If I'm deploying it, it's the people and being able to do that. You know, we all lived through that era of trying to do multi vendor, and often it was challenges. So have we learned from what we've done in the past? Can multi cloud actually be more valuable to a company than the sum of its parts? >>I think so. And I think that's the reason why I, like Microsoft, is investing in art. For example, I think those methodologies way No multi clouds, tough. It's never gonna be easy. And so these companies need to start building in developing platforms for it. There needs to be be great if there were standard AP ice and such right, but they're never gonna do something along those lines. But I think the investments that they're putting forth now are gonna make Multiplied and Polly Cloud a lot easier in the future. And I think customers are asking for it. Customers ask for it, they're gonna build it. >>What does this mean for the workforce, though? In in terms of the kinds of candidates that cos they're going to hire because, as we said, it does require different skills and and different capabilities. So how what's your advice to the young computer scientists coming up in terms of what they should be learning. And then also, how do you think companies are making sensible of this? >>So I know from a company respectable. It's challenging a lot of companies. Especially, for example, I was talking to a very large financial institution, and they were saying that their biggest issue right now is hiring talented people to deal with Micro Service's kubernetes. Any time to hire someone, they end up getting poached by the big cloud companies. So you know, it's one of those things where people are gonna have to start diversifying their talents and look at the future. So I mean, obviously, Micro Service's are here. They're gonna continue to be here. I would say people should invest in that. But also look a server Lis, you know, I definitely think serverless these days towards the future. And then when it comes to like learning skills of multi club, I think cloud competing, that's just the number one growing in general. >>So since you didn't bring up server Lis, you know, today I hear serverless and most customers that I talked to that means a W s number two in the space probably is Microsoft, but there's efforts in to try to help, you know, give a little bit of open source and standardization there. Where's Red Hat? Stand on this. What do you see? What from Microsoft? What are you hearing from customers? >>Were heavily contribute all the different, you know, projects, trying to make server lists like easier to use and not so much specific vendors, Right? So whether that's, you know, Apache, spar or whatever you want to consider it to be, were trying to invest. Invest in those different types of technologies. I think the main issue we serve earless right now is we still don't really know how to utilize it effectively. And it's still kind of this gray area in a sense, right? It's cutting edge, bleeding edge emerging technologies. And it's just, in my opinion, it's not perfectly ready for prime time. But I think that's specifically because there's just not enough people that are actually invested in it. This point in time. So >>So what are you gonna take back with you when you head back to Phoenix from from this conference? What are the things that have sparked your interest the most. >>Gosh, I live, I would probably have to say, Really digging in deep on the Ark announcement. I think that's the thing that I'm most interested in, understanding how how we can actually contribute to that and maybe make that plug double for things like open Shift. You know, whether it's open shift on premise, open shit, running in the cloud on another, Well, architecture's, you know, things like insights. Being able to plug into that, I really see us trying to work with Microsoft to start building those things. >>Well, Nicholas, thank you so much for coming on. The cubit was really fabulous conversation. Thank you. I'm Rebecca Knight for Sue minimum. Stay tuned for more of the cubes. Live coverage from Microsoft ignite.

>> from Cambridge, Massachusetts. It's the Cube covering M I T. Chief data officer and information quality Symposium 2019. Brought to you by Silicon Angle Media. >> Welcome back to M I. T. Everybody. This is the Cube. The leader in live coverage. My name is David wanted. I'm here with my co host, Paul Gill, in this day to coverage of the M I t cdo I Q conference. A lot of acronym stands for M I. T. Of course, the great institution. But Chief Data officer information quality event is his 13th annual event. Lars to Maria's here is the managing partner of Brass Rat Capital. Cool name Lars. Welcome to the Cube. Great. Very much. Glad I start with a name brass around Capitol was That's >> rat is reference to the M I t school. Okay, Beaver? Well, he is, but the students call it a brass rat, and I'm third generation M i t. So it's just seen absolutely appropriate. That is a brass rods and capital is not a reference to money, but is actually referenced to the intellectual capital. They if you have five or six brass rats in the same company, you know, we Sometimes engineers arrive and they could do some things. >> And it Boy, if you put in some data data capital in there, you really explosions. We cause a few problems. So we're gonna talk about some new regulations that are coming down. New legislation that's coming down that you exposed me to yesterday, which is gonna have downstream implications. You get ahead of this stuff and understand it. You can really first of all, prepare, make sure you're in compliance, but then potentially take advantage for your business. So explain to us this notion of open government act. >> Um, in the last five years, six years or so, there's been an effort going on to increase the transparency across all levels of government. Okay, State, local and federal government. The first of federal government laws was called the the Open Data Act of 2014 and that was an act. They was acted unanimously by Congress and signed by Obama. They was taking the departments of the various agencies of the United States government and trying to roll up all the expenses into one kind of expense. This is where we spent our money and who got the money and doing that. That's what they were trying to do. >> Big picture type of thing. >> Yeah, big picture type thing. But unfortunately, it didn't work, okay? Because they forgot to include this odd word called mentalities. So the same departments meant the same thing. Data problem. They have a really big data problem. They still have it. So they're to G et o reports out criticizing how was done, and the government's gonna try and correct it. Then in earlier this year, there was another open government date act which said in it was signed by Trump. Now, this time you had, like, maybe 25 negative votes, but essentially otherwise passed Congress completely. I was called the Open as all capital O >> P E >> n Government Data act. Okay, and that's not been implemented yet. But there's live talking around this conference today in various Chief date officers are talking about this requirement that every single non intelligence defense, you know, vital protection of the people type stuff all the like, um, interior, treasury, transportation, those type of systems. If you produce a report these days, which is machine, I mean human readable. You must now in two years or three years. I forget the exact invitation date. Have it also be machine readable. Now, some people think machine riddle mil means like pdf formats, but no, >> In fact, what the government did is it >> said it must be machine readable. So you must be able to get into the reports, and you have to be able to extract out the information and attach it to the tree of knowledge. Okay, so we're all of sudden having context like they're currently machine readable, Quote unquote, easy reports. But you can get into those SEC reports. You pull out the net net income information and says its net income, but you don't know what it attaches to on the tree of knowledge. So, um, we are helping the government in some sense able, machine readable type reporting that weaken, do machine to machine without people being involved. >> Would you say the tree of knowledge You're talking about the constant >> man tick semantic tree of knowledge so that, you know, we all come from one concept like the human is example of a living thing living beast, a living Beeston example Living thing. So it also goes back, and they're serving as you get farther and farther out the tree, there's more distance or semantic distance, but you can attach it back to concept so you can attach context to the various data. Is this essentially metadata? That's what people call it. But if I would go over see sale here at M I t, they would turn around. They call it the Tree of Knowledge or semantic data. Okay, it's referred to his semantic dated, So you are passing not only the data itself, but the context that >> goes along with the data. Okay, how does this relate to the financial transparency? >> Well, Financial Transparency Act was introduced by representative Issa, who's a Republican out of California. He's run the government Affairs Committee in the House. He retired from Congress this past November, but in 2017 he introduced what's got referred to his H R 15 30 Um, and the 15 30 is going to dramatically change the way, um, financial regulators work in the United States. Um, it is about it was about to be introduced two weeks ago when the labor of digital currency stuff came up. So it's been delayed a little bit because they're trying to add some of the digital currency legislation to that law. >> A front run that Well, >> I don't know exactly what the remember soul coming out of Maxine Waters Committee. So the staff is working on a bunch of different things at once. But, um, we own g was asked to consult with them on looking at the 15 30 act and saying, How would we improve quote unquote, given our technical, you know, not doing policy. We just don't have the technical aspects of the act. How would we want to see it improved? So one of the things we have advised is that for the first time in the United States codes history, they're gonna include interesting term called ontology. You know what intelligence? Well, everyone gets scared by the word. And when I read run into people, they say, Are you a doctor? I said, no, no, no. I'm just a date. A guy. Um, but an intolerant tea is like a taxonomy, but it had order has important, and an ontology allows you to do it is ah, kinda, you know, giving some context of linking something to something else. And so you're able Thio give Maur information with an intolerant that you're able to you with a tax on it. >> Okay, so it's a taxonomy on steroids? >> Yes, exactly what? More flexible, >> Yes, but it's critically important for artificial intelligence machine warning because if I can give them until ology of sort of how it goes up and down the semantics, I can turn around, do a I and machine learning problems on the >> order of 100 >> 1000 even 10,000 times faster. And it has context. It has contacts in just having a little bit of context speeds up these problems so dramatically so and it is that what enables the machine to machine? New notion? No, the machine to machine is coming in with son called SP R M just standard business report model. It's a OMG sophistication of way of allowing the computers or machines, as we call them these days to get into a standard business report. Okay, so let's say you're ah drug company. You have thio certify you >> drugged you manufactured in India, get United States safely. Okay, you have various >> reporting requirements on the way. You've got to give extra easy the FDA et cetera that will always be a standard format. The SEC has a different format. FERC has a different format. Okay, so what s p r m does it allows it to describe in an intolerant he what's in the report? And then it also allows one to attach an ontology to the cells in the report. So if you like at a sec 10 Q 10 k report, you can attach a US gap taxonomy or ontology to it and say, OK, net income annual. That's part of the income statement. You should never see that in a balance sheet type item. You know his example? Okay. Or you can for the first time by having that context you can say are solid problem, which suggested that you can file these machine readable reports that air wrong. So they believe or not, There were about 50 cases in the last 10 years where SEC reports have been filed where the assets don't equal total liabilities, plus cheryl equity, you know, just they didn't add >> up. So this to, >> you know, to entry accounting doesn't work. >> Okay, so so you could have the machines go and check scale. Hey, we got a problem We've >> got a problem here, and you don't have to get humans evolved. So we're gonna, um uh, Holland in Australia or two leaders ahead of the United States. In this area, they seem dramatic pickups. I mean, Holland's reporting something on the order of 90%. Pick up Australia's reporting 60% pickup. >> We say pick up. You're talking about pickup of errors. No efficiency, productivity, productivity. Okay, >> you're taking people out of the whole cycle. It's dramatic. >> Okay, now what's the OMG is rolling on the hoof. Explain the OMG >> Object Management Group. I'm not speaking on behalf of them. It's a membership run organization. You remember? I am a >> member of cold. >> I'm a khalid of it. But I don't represent omg. It's the membership has to collectively vote that this is what we think. Okay, so I can't speak on them, right? I have a pretty significant role with them. I run on behalf of OMG something called the Federated Enterprise Risk Management Group. That's the group which is focusing on risk management for large entities like the federal government's Veterans Affairs or Department offense upstairs. I think talking right now is the Chief date Officer for transportation. OK, that's a large organization, which they, they're instructed by own be at the, um, chief financial officer level. The one number one thing to do for the government is to get an effective enterprise worst management model going in the government agencies. And so they come to own G let just like NIST or just like DARPA does from the defense or intelligence side, saying we need to have standards in this area. So not only can we talk thio you effectively, but we can talk with our industry partners effectively on space. Programs are on retail, on medical programs, on finance programs, and so they're at OMG. There are two significant financial programs, or Sanders, that exist once called figgy financial instrument global identifier, which is a way of identifying a swap. Its way of identifying a security does not have to be used for a que ce it, but a worldwide. You can identify that you know, IBM stock did trade in Tokyo, so it's a different identifier has different, you know, the liberals against the one trading New York. Okay, so those air called figgy identifiers them. There are attributes associated with that security or that beast the being identified, which is generally comes out of 50 which is the financial industry business ontology. So you know, it says for a corporate bond, it has coupon maturity, semi annual payment, bullets. You know, it is an example. So that gives you all the information that you would need to go through to the calculation, assuming you could have a calculation routine to do it, then you need thio. Then turn around and set up your well. Call your environment. You know where Ford Yield Curves are with mortgage backed securities or any portable call. Will bond sort of probabilistic lee run their numbers many times and come up with effective duration? Um, And then you do your Vader's analytics. No aggregating the portfolio and looking at Shortfalls versus your funding. Or however you're doing risk management and then finally do reporting, which is where the standardized business reporting model comes in. So that kind of the five parts of doing a full enterprise risk model and Alex So what >> does >> this mean for first? Well, who does his impact on? What does it mean for organizations? >> Well, it's gonna change the world for basically everyone because it's like doing a clue ends of a software upgrade. Conversion one's version two point. Oh, and you know how software upgrades Everyone hates and it hurts because everyone's gonna have to now start using the same standard ontology. And, of course, that Sarah Ontology No one completely agrees with the regulators have agreed to it. The and the ultimate controlling authority in this thing is going to be F sock, which is the Dodd frank mandated response to not ever having another chart. So the secretary of Treasury heads it. It's Ah, I forget it's the, uh, federal systemic oversight committee or something like that. All eight regulators report into it. And, oh, if our stands is being the adviser Teff sock for all the analytics, what these laws were doing, you're getting over farm or more power to turn around and look at how we're going to find data across the three so we can come up consistent analytics and we can therefore hopefully take one day. Like Goldman, Sachs is pre payment model on mortgages. Apply it to Citibank Portfolio so we can look at consistency of analytics as well. It is only apply to regulated businesses. It's gonna apply to regulated financial businesses. Okay, so it's gonna capture all your mutual funds, is gonna capture all your investment adviser is gonna catch her. Most of your insurance companies through the medical air side, it's gonna capture all your commercial banks is gonna capture most of you community banks. Okay, Not all of them, because some of they're so small, they're not regularly on a federal basis. The one regulator which is being skipped at this point, is the National Association Insurance Commissioners. But they're apparently coming along as well. Independent federal legislation. Remember, they're regulated on the state level, not regularly on the federal level. But they've kind of realized where the ball's going and, >> well, let's make life better or simply more complex. >> It's going to make life horrible at first, but we're gonna take out incredible efficiency gains, probably after the first time you get it done. Okay, is gonna be the problem of getting it done to everyone agreeing. We use the same definitions >> of the same data. Who gets the efficiency gains? The regulators, The companies are both >> all everyone. Can you imagine that? You know Ah, Goldman Sachs earnings report comes out. You're an analyst. Looking at How do I know what Goldman? Good or bad? You have your own equity model. You just give the model to the semantic worksheet and all turn around. Say, Oh, those numbers are all good. This is what expected. Did it? Did it? Didn't you? Haven't. You could do that. There are examples of companies here in the United States where they used to have, um, competitive analysis. Okay. They would be taking somewhere on the order of 600 to 7. How 100 man hours to do the competitive analysis by having an available electronically, they cut those 600 hours down to five to do a competitive analysis. Okay, that's an example of the type of productivity you're gonna see both on the investment side when you're doing analysis, but also on the regulatory site. Can you now imagine you get a regulatory reports say, Oh, there's they're out of their way out of whack. I can tell you this fraud going on here because their numbers are too much in X y z. You know, you had to fudge numbers today, >> and so the securities analyst can spend Mme. Or his or her time looking forward, doing forecasts exactly analysis than having a look back and reconcile all this >> right? And you know, you hear it through this conference, for instance, something like 80 to 85% of the time of analysts to spend getting the data ready. >> You hear the same thing with data scientists, >> right? And so it's extent that we can helped define the data. We're going thio speed things up dramatically. But then what's really instinct to me, being an M I t engineer is that we have great possibilities. An A I I mean, really great possibilities. Right now, most of the A miles or pattern matching like you know, this idea using face shield technology that's just really doing patterns. You can do wonderful predictive analytics of a I and but we just need to give ah lot of the a m a. I am a I models the contact so they can run more quickly. OK, so we're going to see a world which is gonna found funny, But we're going to see a world. We talk about semantic analytics. Okay. Semantic analytics means I'm getting all the inputs for the analysis with context to each one of the variables. And when I and what comes out of it will be a variable results. But you also have semantics with it. So one in the future not too distant future. Where are we? We're in some of the national labs. Where are you doing it? You're doing pipelines of one model goes to next model goes the next mile. On it goes Next model. So you're gonna software pipelines, Believe or not, you get them running out of an Excel spreadsheet. You know, our modern Enhanced Excel spreadsheet, and that's where the future is gonna be. So you really? If you're gonna be really good in this business, you're gonna have to be able to use your brain. You have to understand what data means You're going to figure out what your modeling really means. What happens if we were, You know, normally for a lot of the stuff we do bell curves. Okay, well, that doesn't have to be the only distribution you could do fat tail. So if you did fat tail descriptions that a bell curve gets you much different results. Now, which one's better? I don't know, but, you know, and just using example >> to another cut in the data. So our view now talk about more about the tech behind this. He's mentioned a I What about math? Machine learning? Deep learning. Yeah, that's a color to that. >> Well, the tech behind it is, believe or not, some relatively old tech. There is a technology called rd F, which is kind of turned around for a long time. It's a science kind of, ah, machine learning, not machine wearing. I'm sorry. Machine code type. Fairly simplistic definitions. Lots of angle brackets and all this stuff there is a higher level. That was your distracted, I think put into standard in, like, 2000 for 2005. Called out. Well, two point. Oh, and it does a lot at a higher level. The same stuff that already f does. Okay, you could also create, um, believer, not your own special ways of a communicating and ontology just using XML. Okay, So, uh, x b r l is an enhanced version of XML, okay? And so some of these older technologies, quote unquote old 20 years old, are essentially gonna be driving a lot of this stuff. So you know you know Corbett, right? Corba? Is that what a maid omg you know, on the communication and press thing, do you realize that basically every single device in the world has a corpus standard at okay? Yeah, omg Standard isn't all your smartphones and all your computers. And and that's how they communicate. It turns out that a lot of this old stuff quote unquote, is so rigidly well defined. Well done that you can build modern stuff that takes us to the Mars based on these old standards. >> All right, we got to go. But I gotta give you the award for the most acronyms >> HR 15 30 fi G o m g s b r >> m fsoc tarp. Oh, fr already halfway. We knew that Owl XML ex brl corba, Which of course >> I do. But that's well done. Like thanks so much for coming. Everyone tried to have you. All right, keep it right there, everybody, We'll be back with our next guest from M i t cdo I Q right after this short, brief short message. Thank you

>> from our studios in the heart of Silicon Valley, Palo Alto, California. It is a cute conversation. >> Everyone. Welcome to this Special Cube conversation here at the Palo Alto Cube Studios. I'm John for a host of Cuba here. Moritz man is the head of the product management team at Open Systems A G. Great to see you again. Thanks for coming in. >> Hey, John. Thanks for having me. >> So last time we spoke, you had your event in Las Vegas. You guys are launching. You have a new headquarters here in Silicon Valley. Opened up this past spring. Congratulations. Thank you. >> Yeah, it's a great, great venue to start, and we set foot on the Silicon Valley ground. So to make our way to >> I know you've been super busy with the new building and rolling out, expanding heavily here in the Valley. But you guys were in the hottest area that we're covering Security Cloud security on premise, security. The combination of both has been the number one conversation pretty much in the cloud world right now. Honestly, besides a normal cloud, native cloud I t hybrid versus multi cloud out. See, that continues to be the discussion I think there's no more debate around multi cloud in hybrid public clouds. Great people gonna still keep their enterprises. But the security equation still is changing this new requirements. What's the latest that you guys are seeing with respect to security? >> Yeah. So, John, what we see is actually that cloud adoption had happens at different speeds. So you have usually the infrastructure of the service. Adoption would happens in a quite controlled way because there's a lift in shift. Do you have your old data center? You you take it and you transferred into azure I W S O G C P. But then there's also uncontrolled at option, which is in the SAS space. And I think this is where a lot off data risk occur, especially the wake off GDP are on where we see that this adoption happens. Maurin a sometimes control, but sometimes in a very uncontrolled way, >> explain that the uncontrolled and controlled expansion of of how security and multi cloud and cloud is going because this interesting control means this this plan's to do stuff uncontrolled means it's just by other forces explain uncontrolled versus controls >> eso controlled specifically means the IittIe team takes as a project plan and aches servers and workloads and moves them in a controlled fashion or in a dedicated project to the cloud. But what happened in the business world of business I t is actually did use those share content at any time with any device at any at any time and in all locations. So this is called the Mobile Enterprise on the Cloud First Enterprise. So it means that the classical security perimeter and the controls in that are my past, actually, by the path of least resistance or the shortest path >> available. And this is the classic case. People use Dropbox with some, you know, personal things. They're at home, they're at work, a p I based software. That's what you're getting at the >> and the issue of this is that that the data that has bean, like contained an pera meters where, you know, as it Caesar, where your data is. This has bean deployed too many edge devices, too many mobile devices, and it's get it gets shared, a nun controlled way. >> We'll get a couple talk tracks would like to drill down on that, because I think this is the trend. We're seeing a pea eye's dominant. The perimeter on the infrastructure has gone away. It's only getting bigger and larger. You got I, O. T and T Edge just and the networks are controlled and also owned by different people. So the packets of moving on it that's crazy so that that's the reality. First, talk track is the security challenge. What is the security challenge? How does a customer figure out what to do from an architectural standpoint when they're dealing with hybrid and multi cloud? So first of >> all, um, customers or BC enterprises try need to re think their infrastructure infrastructure centric view off the architecture's. So the architecture that had been built around data send us needs to become hybrid and multi cloud aware. So that means they need to define a new way off a perimeter, which is in cloud but also in the covering. Still the old, so to say, legacy hyper data center set up, which has the data still in the old data center and at the same time, they need to open up and become the cloud themselves, so to say, and but still draw a perimeter around their data and they users and not and their applications and not so much anymore around the physical infrastructure. >> So taking, changing their view of what a security product is, Is that really what you're getting at? >> Yeah, So the issues with the product point solution was that they fixed a certain part off off a tactile issue. So if you take a firewall in itself, firewall back then it was like a entry door to a big building, and you could could decide who comes out goes in. Now. If the the kind of the walls of the building are vanishing or arm or more FIC, you need to come over the more integrated concept. So having these stacked appliance and stacked security solutions trying to work together and chain them doesn't work anymore. So we think and we see that, >> Why is that? Why doesn't it work? Because in >> the end, it's it's it's hardly two to operate them. Each of those points solutions have their own end off life. They have their own life cycle. They have their own AP eyes. They have their own TCO, as all that needs to be covered. And then there's the human aspect where you have the knowledge pools around >> those technologies. So as an enterprise you have to content to continuously keep the very scar security experts to maintain content continues the depreciating assets running right, >> and they're also in it. We weren't built for tying into a holistic kind of platform. >> Yeah, What we see is that that enterprises now realize we have data centers and it's not accepted reality that you can abstracted with the cloud. So you have You don't own your own servers and buildings anymore. So you have a PAX model to subscribe to Cloud Service is and we think that this has to happen to security to so shift from cap ex to our pecs and the same way also for operational matters >> securities. The service is a crepe is a small I want to ask you on that front you mentioned mobile users. How do you secure the mobile uses when they use cloud collaboration? Because this is really what uses expect, and they want How do you secure it? >> So be secured by by actually monitoring the data where it actually gravitates, and this is usually in the cloud. So we enforce the data that is in transit through, ah, proxies and gators towards the cloud from the endpoint devices, but also then looking by AP eyes in the cloud themselves to look for threats, data leakage and also sandbox. Certain activities that happened. There >> are the next talk talk I want to get into is the expansion to hybrid and multi cloud so that you guys do from a product standpoint, solution for your customers. But in general, this is in the industry conversation as well. How how do you look at this from a software standpoint? Because, you know, we've heard Pat Gelsinger of'em were talking about somewhere to find Data Center S d n. Everything's now software based. You talk about the premiere goes away. You guys were kind of bring up a different approaches. A software perimeter? Yeah, what is the challenge for expanding to multi cloud and hybrid cloud? >> So So the challenge for enterprise and customers we talked to is that they have to run their old business. Gardner once called it by motile business, and it's still adopting not one cloud, but we see in our surveys. And this is also what market research confirms is that customers end up with 2 to 3 loud vendors. So there were will be one or two platforms that will be the primary to their major majority of applications and data gravity. But they will end up and become much more flexible with have running AWS, the old Davis Center. But it was the G, C, P and Azure, or Ali Baba glowed even side by side, right tow cover the different speeds at what their own and the price runs. And >> so I gotta ask you about Cloud Needed was one of the things that you're bringing up that just jumps in my head. And when I got to ask, because this is what I see is a potential challenge. It might be a current challenges when you have kubernetes growing such a rapid rate. You see the level of service is coming online much higher rate. So okay, people, mobile users, they're using the drop boxes, the boxes and using all these FBI service's. But that's just those wraps. As a hundreds and thousands of micro service is being stood up and Tauron down in there, you guys are taking, I think, an approach of putting a perimeter software premieres around these kinds of things, but they get turned on enough. How do you know what's clean? It's all done automatically, so this is becoming a challenge. So is this what you guys mean when you say software perimeter that you guys could just put security around things at any time? Is that explain this? >> Yeah, So? So if you talk about the service match so really mashing cloudy but native functions, I think it's still in the face where it's, I would say, chaos chaotic when you have specific projects that are being ramped up them down. So we draw a perimeter in that specific contact. So let's say you have You're ramping up a lot off cloud a function AWS. We can build a pyramid around this kind off containment and look especially for threats in the activity locks off. The different component is containers, but from from a design perspective, this needs to be, uh, we need to think off the future because if you look at Mike soft on AWS strategy, those containers will eventually move Also back to the edge. Eso were in preparing that to support those models also cover. Bring these functions closer back again to the edge on We call that not any longer the when, ej but it will become a cloud at at actually. So it's not an extension of the land that comes to the data. It's actually the data and the applications coming back to the user and much closer. >> Yeah. I mean, in that case, you could define the on premises environment has an edge, big edge, because this is all about moving, were close and data around. This is what the new normal is. Yeah, So okay, I gotta ask the next question, which is okay, If that's true, that means that kubernetes becomes a critical part of all this. And containers. How do you guys play with that at all? >> So we play with us by by actually looking at data coming from that at the moment. We're looking at this from a from a data transit perspective. We But we will further Maur integrate into their eighties AP eyes and actually become part off the C I C D. Process that building then actually big become a security function in approval and rolling out a cannery to certain service mesh. And we can say, Well, this is safe for this is unsafe This is, I think, the eventual goal to get there. But But for now, it's It's really about tracking the locks of each of those containers and actually having a parent her and segmentation around this service mash cloud. So to say, >> I think you guys got a good thing going on when you talk about this new concept that's of softer to find perimeter. You can almost map that to anything you get. Really think everything has its own little perimeter workload. Could be moving around still in these three secure. So I gotta ask on the next talk Trek is this leads into hybrid cloud. This is the hottest topic. Hybrid cloud to me is the same as multi cloud. Just kind of get together a little bit different. But hybrid cloud means you're operating both on premises and in the cloud. This is becoming a channel most si si SOS Chief admission Security officers. I don't want to fork their teams and have multiple people coding different stacks. They don't want the vendor lock in, and so you're seeing a lot of people pulling back on premises building their own stacks, deploying in the cloud and having a seamless operation. What is your definition of hybrid? Where do you see hybrid going? And how important is it? Have a hybrid strategy. >> So I think the key successfactors of a hybrid strategy is that standards standardization is a big topic. So we think that a service platform that to secure that like the SD when secure service platform rebuilt, needs to be standardized on operational level, but also from a baseline security and detection level. And this means that if you run and create your own work, those on Prem you need to have the same security and standard security and deployment standard for the clout and have the seamless security primary perimeter and level off security no matter where these these deployments are. And the second factor of this is actually how do you ensure a secure data transfer between those different workloads? And this is where S T win comes into play, which acts as a fabric together with when backbone, where we connect all those pieces together in a secure fashion >> where it's great to have you on the Q and sharing your insight on the industry. Let's get into your company. Open systems. You guys provide an integrated solution for Dev Ops and Secure Service and Security Platform. Take a minute to talk about the innovations that you guys were doing because you guys talk a lot about Casby. Talk a lot about integrated esti when but first define what Casby is for. The audience doesn't know what Casby is. C. A S B. It's kicked around all of the security conscious of your new to security. It's an acronym that you should pay attention to so defined casby and talk about your solution. >> Eso casby isn't theory. Aviation means cloud access security we broker. So it's actually becoming this centralized orchestrator that that allows and defines access based on a trust level. So saying, um, first of all, it's between networks saying I have a mobile workforce accessing SAS or I s applications. Can't be it in the middle to provide security and visibility about Where's my data moving? Where's married? Where do I have exposure off off GDP, our compliance or P C. I or he power risks And where is it exposed to, Which is a big deal on it's kind of the lowest level to start with, But then it goes further by. You can use the Casby to actually pull in data that that is about I s were close to toe identified data that's being addressed and stored. So are there any incidentally, a shared data artifacts that are actually critical to the business? And are they shared with extra resource is and then going one step further, where we then have a complete zero trust access model where we say we know exactly who can talkto which application at any time on give access to. But as everything this needs to be is in embedded in an evolution >> and the benefit ultimately goes to the SAS applications toe, have security built in. >> That's the first thing that you need to tackle. Nowadays, it's get your sass, cloud security or policy enforced on, but without disrupting service on business on to actually empower business and not to block and keep out the business >> can make us the classic application developer challenge, which is? They love to co they love the build applications, and what cloud did with Dev Ops was abstracted away the infrastructure so that they didn't have to do all this configuration. Sister. Right? APs You guys air enabling that for security? >> Exactly. Yeah. So coming back to this multi protein product cloud would, which is not keeping up anymore with the current reality and needs of a business. So we took the approach and compared death ops with a great service platform. So we have engineers building the platform. That's Integrated Security Service Platform, which promotes Esti Wen managed Detection response and Caspi Service is in one on the one platform which is tightly integrated. But in the in the customer focus that we provide them on or Pecs model, which is pretty, very predictable, very transparent in their security posture. Make that a scalable platform to operate and expand their business on. >> And that's great. Congratulations. I wanna go back for the final point here to round up the interview for the I T. Folks watching or, um, folks who have to implement multi cloud and hybrid cloud they're sitting there could be a cloud architect that could be an I T. Operations or 90 pro. They think multi cloud this in hybrid club. This is the environment. They have to get their arms around. How? What >> should they >> be thinking about? Around multi cloud and hybrid cloud. What is it, really? What's the reality now? What >> should they be considering for evaluation? What are some of the key things that that should be on their mind when they're dealing with hybrid cloud and all the opportunity around it? >> So I think they're they're like, four key pieces. Oneness. Um, they think they still have to start to think strategic. So what? It's a platform and a partner That helps them to plan ahead for the next 3 to 5 years in a way that they can really focus on what their business needs are. This is the scalability aspect. Secondly, it's a do. We have a network on security, our architecture that allows me to grow confidently and go down different venues to to actually adopt multi clouds without worrying about the security implication behind it. Too much, uh, and to implement it. And third is have this baseline and have this standardized security posture around wherever the data is moving, being at Mobil's being it SAS or being on Prem and in clouds workloads, the fourth pieces again, reading, thinking off where did you spend most of my time? Where do I create? Create value by by defining this framework so it really can create a benefit and value for the enterprise? Because if you do it not right your not right. You will have a way. You will end up with a an architecture that will break the business and not accelerated. >> Or it's made head of product that open systems here inside the Cube studios. Um, great job. Must love your job. You got the keys. A lot of pressure. Security being a product. Head of product for security companies. A lot of pressure before we wrap up. Just give a quick plug for the company. You guys hiring you have a new office space here in Redwood City. Looks beautiful. Give a quick shared play for the company. >> Yeah. So open systems the great company to work with. We're expanding in the U. S. On also, Amy, uh, with all the work force. So we're hiring. So go on our website. We have a lot off open positions, exciting challenges in a growth or into workspace. Andi. Yeah. As you said, security at the moment, it's one of the hottest areas to be in, especially with all the fundamental changes happening in the enterprise and architecture. I d landscape. So yeah, >> and clouds securing specifically. Not just in point. The normal stuff that people used to classify as hot as hot as Hades could be right now. But thanks for coming on. Strong insights. I'm jumping with Cuba here in Palo Alto with more Morris Man is the head of product management for open systems. Thanks for watching.

>> Live from San Francisco, it's theCUBE covering IBM Think 2019, brought to you by IBM. >> Hey, welcome back everyone. We're here live at theCUBE in Moscone North in San Francisco, for IBM Think 2019. I'm John Furrier with Stu Miniman, talking to all the top executives, top people here at IBM, getting the scoop on cloud and AI. Our next guest, Jason Gardner, Vice President of Worldwide Sales for Hybrid Cloud at IBM, manages key product, which is part of the IBM Cloud Private, big part of the announcements, big Cloud story here. It's multi-cloud, it's hybrid. Welcome back. >> It's hybrid multi-cloud. Thank you, for having me back. >> CUBE Alumni been on as early, going back as 2012. Now, one big event. >> I can't believe it's been that long. But yeah, I'm happy to be back and I can't believe I've been on theCUBE for so long. >> Talk about your new role, and you had previous roles within IBM dealing with the kind of clients and integration. Your role now is worldwide sales. You're taking this Cloud Private offering, bringing the customers, being as the linchpin for integration. Talk about what you do and some of the engagements you have. >> Yeah, previously, I was really focused in on development and offering management on, point products and how they help clients move to the Cloud. Things such as our Pure Business, our Spare Business, and now I've actually been able to move into a much more horizontal role, where I have the portfolio across the Hybrid Cloud integration side, so everything from our Websphere family, which includes IBM Cloud Private, straight to the integration challenges that that brings as well as our digital business automation portfolio. >> Yeah, I have a personal joy. Stu knows I'm fanatic about Kubernetes, and when I heard Ginni Rometty say Kubernetes twice in a CNBC interview you know it's made it. >> Yes. >> Kubernetes is a big part of cloud native containers, really now has created the connective tissue to make cloud and multi cloud viable. This is a key part of it. I want you to talk about the context of these trends and unpack this Cloud Private offering. Because it's instrumental in seems in the news. >> It is, it is. >> What is it about? >> It is, it really creates that ubiquitous layer I think that we've all been searching for. That next generation of virtualization and connective tissue as you call it. And as you begin to unpack that it really kind of starts with the rise of microservices and the need to be able to pack them very tightly into containers. That's really the birth of Kubernetes, was the ability to orchestrate those containers. So Kubernetes becomes that ubiquitous layer in there. But, IBM Cloud Private takes that and takes it to the next level, right. And, really what it is, it's the services on top of that, the cloud services which enable those containers to work together. And, it is a lot of open source capabilities such as Helm, Prometheus, Kibana and some of those core services that those microservices require in order to be able to run efficiently. >> So, Jason, we know it's a multicloud world. Everybody out there would love to say, oh yes, there's one cloud, I can simplify it. I'd like to get to a nice scalable model that's simple. But, the reality is customers choose lots of different solutions because they have different needs. The Private Cloud piece is not really well understood. I'd love you to take us inside your users. Because they say okay, I'm using Amazon, I'm using Microsoft Business Services. There are certain data things that Google has. IBM has AI and business productivity and database offerings. That Cloud Private, what are the services, what are the use cases, what are the reasons why I'm buying this and being part of my overall portfolio. >> Yeah, Ginni called it Cloud 2.0, right. 1.0 was about lifting shift, it was about cloud native, and that really got us about 20% of the way there. It's at 80%, that's the real challenge, that's really where the complication comes into play. That's really what Private Cloud is about. Because not everybody can be able to take their applications, throw them away, build cloud native, or lift and shift them. If you think of big regulated industries like banking, insurance, healthcare, government. They really need to be able to have that level of security and assurances that they need within there. And, that's really where private cloud comes into play, is those really tough, challenging problems in the industry. >> Yeah, I love that. A trend I've heard from a number of customers, you talk about them getting to containerization and multifactor services, is, step one is, I've got to modernize the platform-- >> Absolutely. >> Then I can modernize the applications on top it. Is that the trend you're seeing? >> Yeah, definitely. We've been building on microservices and modernization, it's a journey right, and it's a journey of discovery I think for a lot of clients out there. And, we'd all love to be able to say, OK this is my platform and now I'm going to work on the applications. But really, sometimes the starting point may be one or the another, and it usually comes in a space of a digital requirement, and so they begin to out modernize the application and then realize, jeez! I need to be able to manage all of this, I need to be able to deploy it all, and that's when the platform comes into play and all the other services, I should say, that come along with it. >> Stu, I think you coined the term Private Cloud. I think wasn't it? >> The true private cloud. >> True private cloud. So the private cloud, again, it's all cloud operations, so I kind of disagree on this whole point about one cloud or multi-cloud. Because I think, yes multi-cloud, but you see people use cloud for workloads, right? So pick the right cloud for the right application. So this basically says, okay, if you want to use Amazon, use Amazon if that's what you want, but if you are going to use 365, maybe use Azure. >> Yep. >> If you are going to use G Suite, use Google. You guys kind of have the business apps nailed down. >> Right. >> So If you're going to use your business apps, maybe IBM. This is your opportunity. >> This is our opportunity. >> Talk about specifically the kinds of apps that you guys will power with your cloud, because multi-cloud certainly makes sense for you guys. It's multi-cloud, you won't that portability and interoperability, but the apps that you're going to power with IBM Cloud. Talk about what they are, how-- >> Yeah, if you look at, from a language perspective over the last, jeez it's been 23 years I think, since the rise of Java, right? And 1995, when the first app servers came out. Those app servers, that is really where ore applications really run on top of. And, it's those core Java applications, that are now needing that facelift, right? They need to be able to be injected with new forms of AI, new types of integrations, new types of personalization of that digital transformation that's driving it, and that's really the core suite, right? And if I look at that core suite in there, and then what do you do to modernize a Java application, and what kind of tools are available to you. How do you then manage, how do you distribute, and how do you scale those applications. It's very important. >> What is the adoption of the private cloud or the Cloud Private product. >> Yeah. >> Talk about some of the trends, how is it being used, be specific on how customers are using it. What are some of the use cases? >> Yeah, so the primary use case is to increase the agility, lower cost on the overall managing of them. But it's the increase in the agility, which is really hard to measure. Because clients want to be able to react very fast to it. And so as they build up microservices, microservices then become independent with one another. You can then update ones, very quickly and easily. They manage and they run independently, and they scale independently, and so Cloud Private provides you with all those services to able to run those microservices as containers, but then be able to tie them together in a much more comprehensive enterprise suite. You know, a core technology like Helm, I'm waiting for Ginni to say that one on stage. But a core technology like Helm, really provides that robust, enterprise class distribution for scalability and high availability of a microservice based application. >> Jason, can you bring us inside the organization of the customers your selling to? It used to be, it was the refresh cycle. It's like OK, my X86 refresh, or you know, the budget cycles that I had. Cloud is quite a bit different. >> It is. >> Private Cloud is kind of straddling between the old world and the new world. What are the dynamics you're seeing as to who controls the purse strings? Are they moving faster to that opex model. >> You know, there's no one person who owns the purse strings on it, but it does float between the infrastructure team, knows that they need to do something different, the developers or the application development team, and really the strategy, the Chief Strategy Officer, in that IT organization is really where it's coming together. Because one thing I think that we've all learned is that developers will find the easiest, fastest way to do something. No matter what rules or policies we put down. And this is about providing them with an environment that has guardrails, for them to be able to innovate as fast as they want, use the tools that they want, that their most comfortable with. Really, it's a grass roots kind of movement into these microservices, led by the developers. But the purse strings are still held at the CTO side. >> That's always a fascinating interest, because the developers they're going to go do it, but they're not usually the ones with the budget. >> That's right. >> But when do the ops people get involved, the business people, to make sure that IT manages it, gets rid of like stealth IT? >> And a lot of clients have learned to listen to the developers, because the early days of cloud, they didn't, and developers found ways through it, no matter what. And so that's really what it's about. It's like a game of bumper cars, right? You got to make sure they stay within the ring of what's safe. And, especially in this day and age of the security requirements that are out there, it's more important today than ever before. >> Jason, can you share some data around some observations that you've noticed on trends around industry uptake or is there any patterns in terms of the customer base? Obviously, people aren't going to going to cloud operations. Just, Ginni mentioned 60/40, 80/20, the ratios. What does that all mean? And, just share the trend data around adoption and patterns? >> Probably the biggest onE in there, is the 80/20, right? That there's still 80% of the applications left in the world are still locked behind the brick and mortar. That's probably our biggest piece of our opportunity, and providing clients with a way to lift them up and be able to modernize them. I think is where the huge opportunity is. But then looking at where do they land, it's not all going to public cloud, right. So private cloud it's a huge business. I think a lot of us underestimated how large that business really is, and depending on the industry, you'll see 50/50, 60/40, 40/60 split, depending on the regulations within that industry, that country, the geography, of where they really want to go to. And, a lot of our clients are asking us for solutions around that private side, but yet be able to have the flexibility to be able to-- >> So you're seeing friction on the public cloud, mainly that's inherent from either regulatory compliance, or just technical challenges. Is that kind of the vibe? >> That's probably the first one. I think there's still that regulatory requirements of data residency, and how do I get my data to application. I can build all the applications I want in the cloud, but how do I get my data there? How do I synchronize it? My lineage of my data. So they really challenged her on that. But, then on the other side of it, is around the cost, right. And, if you wanted to rebuild all of your applications, as true cloud native, from scratch. It will take you a very long time and be very, very expensive. And so, there's also a cost element and speed. You can modernize something much more quickly, and be able to get it to that same level of service, without having to start over. >> We had Arvind on earlier, yesterday, and I want to get your thoughts on the impact of the Red Hat acquisition news, because if you look at what Open Shift is doing with Cloud Private. Arvind was saying yesterday that, Arvind Krishna, he's like, this is really enabling a lot of the acceleration for the modernization of the new cloud stuff, and keeping the legacy stuff and/or transition out on different timetables. Your thought on that? >> Absolutely right, Open Shift is going to be a critical component for our overall hybrid strategy. I'm very excited about it and really looking forward to it. And, Cloud Private and the services that I talked about, run in Open Shift today. That was part of our partnership agreement. I think that you guys were at, that Arvind talked about at that time. But, it provides the platform, for all of those traditional applications, which we've modernized. And the interesting thing is that we've actually modernized ourselves. We've modernized our middle-ware. We've modernized some of those products that are you know, 10, 20 years old. Everything from WebSphere, to MQ, to BPM. They've all been modernized in that same fashion. >> Yeah, Jason, speaking of modernization. Bring us inside you're sales force a little bit. How do they keep up, and what's the skill set that you're looking for, on your team to sell on this. You know, they need to understand Helm and Kubernetes, and all these microservice architecture, where five years ago, it was a totally different world. >> Absolutely, you know I think that if I look at a, it's not a skill, it's passion, right? It's that never give up type of mentality, I think that we look for, in a sales force and I never give up attitude really provides you with that foundation, for never stop learning, right. If anything that you've guys have noticed here over the last ten years in your guys' journey, is that this industry just changes so repidly, all the time. And, so as a sales force, you can't just acquire skills. You don't go out and hire skills. You hire people and you hire passion, and you hire people with that never give up attitude. I've been going around. We've been doing our sales kick-offs. I've done two out of the three now, so far. I tell you they are energized. They love it. They are energized about the Red Hat Acquisition. It shows that IBM really gets it. They've been telling me, does IBM really get it? And now they're like wow, we really do get it? And, they're really energized, because all of the pieces are falling into place, around this modernization, and clients, and we're hitting the timeing. >> It's time to hit that pedal to the metal, put the gas on-- >> They always say, there's no speed limit on sales. >> (laughs) Exactly. OK, first of all great, great conversation, and thanks for waiting out our journey. Stu, I would say that the salespeople got to watch all theCube videos, because all of the best content is coming out of theCube here, and great to have you on. But, quick plug, I'll give you the last word. What's the pitch, share the pitch for the Hybrid Cloud, what your team is offering? What's the, the core pitch for your customers, when you go to them? >> I think the core pitch is around modernization. It's the journey that clients are on, from application development, to how you build your apps, and how you build the microservices. How you integrate those applications, what's your API strategy, how do you move that data around securely, and then how do you manage all of those pieces together in that new modern world. And then, really looking your overall processes, and can you modernize your overall processes, add AI capabilities into that. So, it's that modernization journey. That's really what I talk to them about, and you don't have to do everything, right? Start small, start as a pinpointed piece, and we'll help you along that journey. And it becomes a journey of self-discovery, but we're there the whole way. We're a partner, that's really what it's about. >> Jason Gardner, Vice President of Worldwide Sales with Hybrid Cloud at IBM. TheCube, bringing all the data here, from IBM Think 2019. This is day three, of four days of coverage, here in Moscone live in San Francisco. We'll be right back with more, after this short break. (upbeat music)

>> (Narrator) Live from Vancouver, Canada, it's The Cube, covering OpenStack Summit North America 2018, brought to you by Red Hat, the OpenStack Foundation, and its Ecosystem partners. >> Welcome back, I'm Stu Miniman with my co-host John Troyer and you're watching The Cube, worldwide leader in tech coverage, and this is OpenStack Summit 2018 in Vancouver. Happy to welcome to the program, first-time guest Melvin Hillsman, who's the governance board member of OpenLab, which we got to hear about in the keynote on Monday. Thanks so much for joining us. >> Thanks for having me. >> Alright, so Melvin, we were given, start us off with a little bit about your background, what brought you to the OpenStack community, and we'll go from there. >> Sure, yeah, so my background is in Linux system administration and my getting involved in OpenStack was more or less seeing the writing on the wall as it relates to virtualization and wanting to get an early start in understanding how things would pan out over the course of some years. So I probably started OpenStack maybe three or four so years ago. I was probably later to the party than I wanted to be, but through that process, started working at Rackspace first and that's how I really got more involved into OpenStack in particular. >> Yeah, you made a comment, though. The writing on the wall for virtualization. Explained that for a sec. >> So for me, I was at a shared hosting company and we weren't virtualizin' anything. We were using traditional servers, dedicated servers, installing hundreds of customers on those servers. And so, at one point, what we started doing was we would take a dedicated server, we would create a virtual machine on it, but we would use most of the resources of that dedicated server, and so what allowed that shared hosting was to tear stuff down and recreate it, but it was very manual process and so, of course, the infrastructure service and orchestration around that OpenStack was becoming the de facto standard and way of doing it, and so I didn't want to try to learn manually, or fix something up internally, I wanted to go where OpenStack was being highly developed a lot and people working on it in their day to day jobs, which is why I went to Rackspace. >> Okay, one of the things we look at, this is a community here, so it takes people from lots of different backgrounds, and some of them do it on their spare time, some of them are paid by larger companies to participate, so to tell us about OpenLab, itself, and how your company participates there. >> Sure, so I started, well I'm at Huawei now, but I was at Rackspace and that's kind of how I got more involved in the community and there I started working on testing things above the OpenStack ecosystem, so things that people want to build on top of OpenStack and during that process Huawei reached out to me and was like hey you know you're doing a great job here, and I was like yeah I would love to come and explore more of how we can increase this activity in the community at large. And so Oakland Lab was essentially born out of that, which the OpenStack community, they deliver the OpenStack API's, and they kind of stop there, you know. Everything above that is, you do that on your own, more or less, and so also, as a chair of the user committee, again, just being more concerned about the people who are using stuff, OpenLab was able, was available to facilitate me having access to hardware and access to people who are using things outside OverStack in use cases, et cetera, where we want to test out more integrated tools working with OpenStack and different versions of OpenStack. And so that's essentially what OpenLab is-- >> So in OpenLab, projects come together and it's basically, it's an Interop, boy, in the networking world, they've had the Interop plug and plug fest for a long time, but, in essence, projects come together and you integrate them and start, you invite them in and they integrate and start to test them. Starting with, I mean, I see, for this release, Terraform and Kubernetes. >> Yeah, so a lot of people want to to use Kubernetes, right? And as an OpenStack operator you essentially, you don't really want to go and learn all the bits of Kubernetes, necessarily, and so, but you want to use Kubernetes and you want to work seamlessly with OpenStack and you want to use the API's that you're used to using with OpenStack and so we work very heavily on the external cloud provider for OpenStack, enabling Cinder V3 for containers that you're spinning up in Kubernetes, so that they have seamless integration, you don't have to try to attach your volumes, they are automatically attached. You don't have to figure out what your load balancing is going to to look like. You use Octavia, which is load balancing service for OpenStack, very tightly integrated and things, you know, as you spin things up, they work as you as you would expect and so then all the other legacy applications and all the things you're used to doing with OpenStack, you bring on Kubernetes and you essentially do things the way you've been doing them before, with just an additional layer. >> Yeah, now I wonder if you can talk a little bit about the providers and the users, you know how do they get engaged, to and give us a little flavor around those. >> Yeah, so you get engaged, you go to OpenLabtesting.org and there's two options. One, is you can test out your applications and tools, by clicking get started, you fill that out. And what's great about open lab is that we actually reach out and we talk with you, we consult with you, per se, because we have a lot of variation in hardware that's available to us and so we want to figure out the right hardware that you need in order to do the tests that you want, so that we can get the output as it relates to that integration that will, of course, educate and inform the community at large of whether or not it's working and been validated. And, again, so as a person who wants to support OpenLab or for a provider, for example, who wants to support OpenLab, you click on the support OpenLab link, you fill out a form and you tell us you know, do you want to provide more infrastructure, do you want to talk with us about how clouds are being architected, integrations are being architected, things that you're seeing in the open source use cases that may not be getting the testing that they need and you're willing to work with engineers from other companies around that, so individual testers and then companies who may bring a number of testers together around a particular use case. >> Now, you're starting to publish some of the results of Interop testing and things like that. How is open lab, how does it produce its results, is it eventually going to be producing white papers and things like that or dashboards or what's your vision there? >> Yeah, so we produce a very archaic dashboard right now, but we're working with the CNCF to, if you go to CNCF.CI, and they have a very nice dashboard that kind of shows you a number of projects and whether or not they work together. And so it's open source, so what we want to do is work with that team to figure out how do we change the logos and the git repos, to how to driving those red and green, success or failure icons that are there, but they're relevant to the test that we're doing in OpenLab, so yeah. So we definitely want to have a dashboard that's very easy to decipher what tests are failing in or passing. >> Looking forward, what kinds of projects are you most interested in getting involved? >> Right now, very much Kubernetes, of course. We're really focusing on multi architecture, again, as a result of our work with Kubernetes and driving full conformance and multi architecture. That's kind of the wheelhouse at this time. We're open for folks to give us a lot of different use cases, like we were starting to look at some edge stuff, how can we participate there, we're starting looking at FPGA's and GPU's, so a lot different, we don't have a full integration in a lot of different areas, just yet, but we are having those conversations. >> So, actually, I spent a bunch of years, when I worked on the vendor side, living in an Interrupt lab, and the most valuable things were not figuring out what worked, but what broke, so what kind of things, you know, as you're working through this, what learnings back do you share with the community, both the providers and users? Big stumbling blocks that you can help people, give a red flag, or say you know, avoid these type of things. >> Yeah, exactly what you just said. You know, what's good is some of our stuff is geographically dispersed, so we can start to talk about if, what's the latency look like? You may, within that few square miles that you're operating and doing things, it works great, but when I'm sending something across the water how, is your product still moving quickly, is the latency too bad that we can't, I can't create a container over here because it takes too long, so one example of looking at something fail as it relates to that is we're talking with Octavia folks to see, if I spin up a lot of containers am I going to therefore create a lot of load balancers and if I create a lot of load balancers I'm creating a lot of VM's, or am I creating a lot of containers or are things breaking apart, so we need to dig a little bit further to understand what is and is not working with the integrations we're currently working on and then again we're exploring GPV, GPUs just landed more or less, that was a part of the keynote as well, and so now we're talking about, well, let's do some of that testing. The software, the code, is there but is it usable? And so that's one area we want to start playing around with. >> Okay, one of the other things in the keynote's got mentioned was Zul, the CIDT tool, how's that fitting into the OpenLab? >> Yeah, we use Zul as our gating, so what's great about Zul is that you can interac6t with projects from different SCM's, so we have some projects that live in github, some that utilize Garrote, some that utilize gitlab, and Zul has applicability where it can talk to different, it can talk across these different SCM's, and if you have a patch that depends on a patch in another another pod, so a patch on one project in one SCM can depend on a patch in another project, in a different SCM, and so what's great about Zul is that you can say, hey I'm depending on th6at, so before this patch lands, check to make sure this stuff works over there, so if it succeeds there and it's a dependency then you basically confirm that succeeds there and then now I can run the test here, and it passes here as well, so you know that you can use both of those projects together again, in an integration. Does it makes sense? Hopefully I'm making it very clear, the power there with the cross SCM integration. >> Yeah, Melvin, you've had a busy week, here, at the show. Any, you know, interesting things you learned this week or something that you heard from a customer that you thought, oh boy we got to, you know, get this into our lab or a road map or, you know-- >> The ARM story, the multi architecture is, I feel like that's really taking off. We've had discussions with quite a few folks around that, so yeah, that for me, that's the next thing that I think we're really going to concentrate a little bit harder on is, again, figuring out if there are some problems, because mostly it's been just x86, but we need to start exploring what's breaking as we add more to multi architecture. >> Melvin, no shortage of new things to test and play with, and every customer always brings some unique spins on things, so appreciate you giving us the update on OpenLab, thanks so much for joining us. >> You're Welcome. Thanks for having me. >> From John Troyer, I'm Stu Miniman, thanks so much for watching The Cube. (electronic music)