>> Announcer: TheCube presents HPE Discover 2022, brought to you by HPE. >> Welcome back to HPE Discover 2022, theCube's continuous wall to wall coverage, Dave Vellante with John Furrier. Keith Basil is here as the General Manager for the Edge Business Unit at SUSE. Keith, welcome to theCube, man good to see you. >> Great to be here, it's my first time here and I've seen many shows and you guys are the best. >> Thanks you. >> Thank you very much. >> Big fans of SUSE you know, we've had Melissa on several times. >> Yes. >> Let's start with kind of what you guys are doing here at Discover. >> Well, we're here to support our wonderful partner HPE, as you know SUSE's products and services are now being integrated into the GreenLake offering. So that's very exciting for us. >> Yeah. Now tell us about your background. It's quite interesting you've kind of been in the mix in some really cool places. Tell us a little bit about yourself. >> Probably the most relevant was I used to work at Red Hat, I was a Product Manager working in security for OpenStack and OpenShift working with DOD customers in the intelligence community. Left Red Hat to go to Rancher, started out there as VP of Edge Solutions and then transitioned over to VP of Product for all of Rancher. And then obviously we know SUSE acquired Rancher and as of November 1st, of 2020, I think it was. >> Dave: 2020. >> Yeah, yeah time is flying. I came over, I still remained VP of Product for Rancher for Cloud Native Infrastructure. And I was working on the edge strategy for SUSE and about four months ago we internally built three business units, one for the Linux business, one for enterprise container management, basically the Rancher business, and then the newly minted business unit was the Edge business. And I was offered the role to be GM for that business unit and I happily accepted it. >> Very cool. I mean the market dynamics since the 2018 have changed dramatically, IBM bought Red Hat. A lot of customers said, "Hmm let's see what other alternatives are out there." SUSE popped its head up. You know, Melissa's been quite, you know forthcoming about that. And then you acquire Rancher in 2020, IPO in 2021. That kind of gives you another tailwind. So there's a new market when you go from 2018 to 2022, it's a completely changed dynamic. >> Yes and I'm going to answer your question from the Rancher perspective first, because as we were at Rancher, we had experimented with different flavors of the underlying OS underneath Kubernetes or Kubernetes offerings. And we had, as I said, different flavors, we weren't really operating system people for example. And so post-acquisition, you know, one of my internal roles was to bring the two halves of the house together, the philosophies together where you had a cloud native side in the form of Rancher, very progressive leading innovative products with Rancher with K3s for example. And then you had, you know, really strong enterprise roots around compliance and security, secure supply chain with the enterprise grade Linux. And what we found out was SUSE had been building a version of Linux called SLE Micro, and it was perfectly designed for Edge. And so what we've done over that time period since the acquisition is that we've brought those two things together. And now we're using Kubernetes directives and philosophies to manage all the way down to the operating system. And it is a winning strategy for our customers. And we're really excited about that. >> And what does that product look like? Is that a managed service? How are customers consuming that? >> It could be a managed service, it's something that our managed service providers could embrace and offer to their customers. But we have some customers who are very sophisticated who want to do the whole thing themselves. And so they stand up Rancher, you know at a centralized location at cloud GreenLake for example which is why this is very relevant. And then that control plane if you will, manages thousands of downstream clusters that are running K3s at these Edge locations. And so that's what the complete stack looks like. And so when you add the Linux capability to that scenario we can now roll a new operating system, new kernel, CVE updates, build that as an OCI container image registry format, right? Put that into a registry and then have that thing cascade down through all the downstream clusters and up through a rolling window upgrade of the operating system underneath Kubernetes. And it is a tremendous amount of value when you talk to customers that have this massive scale. >> What's the impact of that, just take us through what happens next. Is it faster? Is it more performant? Is it more reliable? Is it processing data at the Edge? What's the impact of the customer? >> Yes, the answer is yes to that. So let's actually talk about one customer that we we highlighted in our keynote, which is Home Depot. So as we know, Kubernetes is on fire, right? It is the technology everybody's after. So by being in demand, the skills needed, the people shortage is real and people are commanding very high, you know, salaries. And so it's hard to attract talent is the bottom line. And so using our software and our solution and our approach it allows people to scale their existing teams to preserve those precious human resources and that human capital. So that now you can take a team of seven people and manage let's say 3000 downstream stores. >> Yeah it's like the old SRE model for DevOps. >> Correct. >> It's not servers they're managing one to many. >> Yes. >> One to many clusters. >> Correct so you've got the cluster, the life cycle of the cluster. You already have the application life cycle with the classic DevOps. And now what we've built and added to the stack is going down one step further, clicking down if you will to managing the life cycle of the operating system. So you have the SUSE enterprise build chain, all the value, the goodness, compliance, security. Again, all of that comes with that build process. And now we're hooking that into a cloud native flow that ends up downstream in our customers. >> So what I'm hearing is your Edge strategy is not some kind of bespoke, "Hey, I'm going after Edge." It connects to the entire value chain. >> Yes, yeah it's a great point. We want to reuse the existing philosophies that are being used today. We don't want to create something net new, cause that's really the point in leverage that we get by having these teams, you know, do these things at scale. Another point I'm going to make here is that we've defined the Edge into three segments. One is the near Edge, which is the realm of the-- >> I was going to ask about this, great. >> The telecommunications companies. So those use cases and profiles look very different. They're almost data center lite, right? So you've had regional locations, central offices where they're standing up gear classic to you machines, right? So things you find from HPE, for example. And then once you get on the other side of the access device right? The cable modem, the router, whatever it is you get into what we call the far Edge. And this is where the majority of the use cases reside. This is where the diversity of use cases presents itself as well. >> Also security challenges. >> Security challenges. Yes and we can talk about that following in a moment. And then finally, if you look at that far Edge as a box, right? Think of it as a layer two domain, a network. Inside that location, on that network you'll have industrial IOT devices. Those devices are too small to run a full blown operating system such as Linux and Kubernetes in the stack but they do have software on them, right? So we need to be able to discover those devices and manage those devices and pull data from those devices and do it in a cloud native way. So that's what we called the tiny Edge. And I stole that name from the folks over at Microsoft. Kate and Edrick are are leading a project upstream called Akri, A-K-R-I, and we are very much heavily involved in Akri because it will discover the industrial IOT devices and plug those into a local Kubernetes cluster running at that location. >> And Home Depot would fit into the near edge is that correct? >> Yes. >> Yeah okay. >> So each Home Depot store, just to bring it home, is a far Edge location and they have over 2,600 of these locations. >> So far Edge? You would put far Edge? >> Keith: Far Edge yes. >> Far edge, okay. >> John: Near edge is like Metro. Think of Metro. >> And Teleco, communication, service providers MSOs, multi-service operators. Those guys are-- >> Near Edge. >> The near edge, yes. >> Don't you think, John's been asking all week about machine learning and AI, in that tiny Edge. We think there's going to be a lot of AI influencing. >> Keith: Oh absolutely. >> Real time. And it actually is going to need some kind of lighter weight you know, platform. How do you fit into that? >> So going on this, like this model I just described if you go back and look at the SUSECON 2022 demo keynote that I did, we actually on stage stood up that exact stack. So we had a single Intel nook running SLE Micro as we mentioned earlier, running K3s and we plugged into that device, a USB camera which was automatically detected and it loaded Akri and gave us a driver to plug it into a container. Now, to answer your question, that is the point in time where we bring in the ML and the AI, the inference and the pattern recognition, because that camera when you showed the SUSE plush doll, it actually recognized it and put a QR code up on the screen. So that's where it all comes together. So we tried to showcase that in a complete demo. >> Last week, I was here in Vegas for an event Amazon and AWS put on called re:Mars, machine learning, automation, robotics, and space. >> Okay. >> Kind of but basically for me was an industrial edge show. Cause The space is the ultimate like glam to edge is like, you're doing stuff in space that's pretty edgy so to speak, pun intended. But the industrial side of the Edge is going to, we think, accelerate with machine learning. >> Keith: Absolutely. >> And with these kinds of new portable I won't say flash compute or just like connected power sources software. The industrial is going to move really fast. We've been kind of in a snails pace at the Edge, in my opinion. What's your reaction to that? Do you think we're going to see a mass acceleration of growth at the Edge industrial, basically physical, the physical world. >> Yes, first I agree with your assessment okay, wholeheartedly, so much so that it's my strategy to go after the tiny Edge space and be a leader in the industrial IOT space from an open source perspective. So yes. So a few things to answer your question we do have K3s in space. We have a customer partner called Hypergiant where they've launched satellites with K3s running in space same model, that's a far Edge location, probably the farthest Edge location we have. >> John: Deep Edge, deep space. >> Here at HPE Discover, we have a business unit called SUSE RGS, Rancher Government Services, which focuses on the US government and DOD and IC, right? So little bit of the world that I used to work in my past career. Brandon Gulla the CTO of of that unit gave a great presentation about what we call the tactical Edge. And so the same technology that we're using on the commercial and the manufacturing side. >> Like the Jedi contract, the tactical military Edge I think. >> Yes so imagine some of these military grade industrial IOT devices in a disconnected environment. The same software stack and technology would apply to that use case as well. >> So basically the tactical Edge is life? We're humans, we're at the Edge? >> Or it's maintenance, right? So maybe it's pulling sensors from aircraft, Humvees, submarines and doing predictive analysis on the maintenance for those items, those assets. >> All these different Edges, they underscore the diversity that you were just talking Keith and we also see a new hardware architecture emerging, a lot of arm based stuff. Just take a look at what Tesla's doing at the tiny Edge. Keith Basil, thanks so much. >> Sure. >> For coming on theCube. >> John: Great to have you. >> Grateful to be here. >> Awesome story. Okay and thank you for watching. This is Dave Vellante for John Furrier. This is day three of HPE Discover 2022. You're watching theCube, the leader in enterprise and emerging tech coverage. We'll be right back. (upbeat music)

(digital music) >> Welcome to this CUBE Conversation. I'm Lisa Martin. Bren Briggs, joins me next, the Director of DevOps and Cybersecurity at Hypergiant. Bren, welcome to theCUBE. >> Hey there, I'm glad to be here. >> You have a very cool background, which I wish we had time to get into your mandolin playing, but we don't. Tell me a little bit about Hypergiant this is a company that's new to me? >> So we are an AI and Machine Learning Company, and we had the slogan we talked about a lot, it's almost tongue in cheek, "Tomorrowing Today" where we want to build and focus on technology that advances the state-of-the-art and we want to, where this deep history and background in services, where we build custom solutions for companies that have data problems and that have AI and machine learning problems. And they come to us and we help them make sense of their data and we build a custom software solution from top to bottom. And we help them with their data problems and their really difficult problems that they have there in a very specialized way. And yeah, that's what we do. It's really fun. >> "Tomorrowing Today", I like that build T-shirts with that on that. (Bren chuckles) So talk to me about the work that you guys are doing with SUSE Rancher Government Labs. You're doing some very cool work with the air force, help me understand that. >> Sure, so about a year and some change ago, we had a government contract, an air force contract, to develop some new or just to basically write an experiment with some new sensing technology onboard a satellite. So we built this satellite, we were talking about how we're going to employ DevOps' best practices on the satellite and if that's even a thing that can be done. How we get these rights of space and really thinking through the entire process. And as we did this, we were getting more and more deeply involved with a very very new group. Actually, we kind of started at the same time. A new group within the air force called, Platform One. Platform One's mission is to bring DevSecOps to the DoD Enterprise. And so as we're kind of starting off together and getting to know each other, Rob Slaughter who started and ran Platform One for the first bit of his existence, he said, "hey, we're going to incorporate some Platform One stuff into this. Let's talk about just building an actual Platform One satellite and see what that looks like." And so that was kind of the start of this whole idea was what do we do and how do we do DevSecOps in low Earth orbit? Can we put Kubernetes on satellite and will it work? >> And tell me some of the results? So, I used to work for NASA, so I would geek out on anything that has to do with the space program. But talk to me about some of the things that you uncovered bringing Kubernetes, AI, machine learning to this, outer Edge of Earth? >> I think the first thing that we learned that I think, it's an understatement to say that space is hard. (Both laughing) But it really is. And that was the part that we learned about was it was hard in all of the ways that we did not expect. And a lot of it had to do with just government and logistics. We learned that it is difficult a lot of times to just to find a way to get into space and then once you're there, how you operate in the conditions that you're in and how you could even communicate with your satellite is it's just a logistical adventure on top of all of the other engineering problems that you have while you're on low Earth orbit? The other thing that we figured out was awkward things are difficult. While you're on orbit, they can be slow or fragmented and so it pays to get it right the first time but that's not the nature of modern software development is you'd never get it right and you're continually updating. So that was a problem that really nagged us for awhile was after we did the wider experiment, like how would we continuously update this and what would we do? And those ideas and questions fed into the experiment that became Sat One and then the follow one much bigger experiment that became the Edge One and Edge working group. >> Tell me a little bit about the wider experiment, give me some context of how that relates to Platform One, Sat One? >> I can't (laughing) I can't really go into details about what wider did or anything like that. It was not a classified mission, it's just not something that I can disclose. >> Okay, got it. >> Sorry. >> So talk to me about some of the work that you guys are doing together Hypergiant with SUSE in terms of pushing forward the next generation of Kubernetes to low Earth orbit and beyond. >> Sure, so SUSE RGS, specifically, Chris Nuber, like, one of the things that I have to do is I have to be a cheerleader for all of the amazing people that were on this project. And two people in particular, Chris Tacke and Chris Nuber, were instrumental in making this work. I was like almost tangentially involved where I was doing some input and architecture and helping debug but it was really Chris Tacke and Chris Nuber that made this thing, that built this thing and made it work. And Chris Nuber, was our assigned resource from SUSE RGS. And he said, "Obviously SUSE is going to prefer, or SUSE is going to prefer SUSE products." That it makes sense. But there's a reason because the products that he implemented and the patterns that he implemented and the architecture and expertise that he brought were second to none, I don't think that we could have done better with any other distribution of Kubernetes. He recommended a K3s is a very lightweight Kubernetes distribution that had really good opinions. It's a single binary. It was very easy to deploy and manage and update and it just, it really didn't break. That was the best thing that we were looking for (chuckles) it was one solid piece with no moving parts, relatively speaking. And so Chris Nuber was very essential in providing the Kubernetes architecture while Chris Tacky was the one who helped us write some of the demo applications and build the fail over and out of band interaction that we were going to have from the hardware on the satellite to the Kubernetes control plane. >> Very cool. It sounds like you had a great collaborative team there, which is essential in any environment. >> We deed. >> And I liked how you described space as a logistical adventure that reminds me very much of my days at NASA. (Bren laughing) It definitely is a logistical adventure to put it mildly. Talk to me a little bit about the work that you're doing to define the Edge for the Department of Defense? That sounds very intriguing. >> Yeah, so this was almost a direct result of what happened with the sat one experiment where Rob Slaughter and a few of the other folks who saw what we did with sat one, you know, were again, logistical adventure. We built this entire thing and we worked so hard and we're moving through fright flight readiness checks and as things happen, funding kind of went. And so you've got all this experience and this like, prototype that this really confident that it's space ready and everything and they said, "hey, listen, you know, we have the same problem on our flight with terrestrial environments, they're nearly identical the only difference is, you know, you don't have to worry about radiation nearly as much." (laughing) So then, you know, we joked about that and we started this new idea, this Edge One idea as part of the AVMs program, where they're figuring out this new, like battlefield communications pattern of the future. And one of the things that they're really concerned about is secure processing and how do you do applications at like where people are stationed, which could be anywhere in very remote locations. Then that's what turned into Edge One is, you know, we imagined initially Edge One as satellite one without wings and earth bound and that grew into, well, what about submarines? What about carriers? What about command and control squadrons that are stationed in cities? What about special operators that are far forward? What about first responders who are moving into, you know, hazardous environmental conditions? Can you wear a Kubernetes cluster with like super low power arm chips? And so we started thinking of all these different applications of what Edge could be anywhere from a five volt board all the way up to a data center in a box. And that caused us to realize that we're going to break Edge into really three categories based on the amount of material or resources needed to power it and how hard it is to get to. So we have the Near Edge, which is, you know, you have data center like capabilities, and it's easy to get to it, but you, because you have people stationed with it, but you may have reached back once every month or so. So think, you know, a shift that's underway or an air gap system or something like that. And then you have a Tiny Edge, which is exactly like kind of the more traditional idea that you think of when you think of Edge, which is really, really tiny compute, maybe it's on a windmill or something I don't really know, pick your thing to put Kubernetes on that should never have Kubernetes, that's the kind of thing. And then you've got Far Edge, which is, you know, if the control plane crashes, good luck, you'll never getting to it. And so that would be a satellite. And so the far it... so really a lot of these, it depends on the failure mode. Like what happens when it fails and that for the most part defines kind of what category you're going to be in. >> Tiny Edge, Near Edge and Far Edge. I think Sir. Richard Branson and his team went to the Far Edge (chuckles) low Earth orbit >> He did (laughing). >> This last weekend, I guess, yeah. That low Earth orbit does seem like it would be the Far Edge. Talk to me a little bit about, I mean, you talk about these applications then from a defense perspective that very dramatically, what are some of the important lessons that you've learned besides if it breaks in the Far Edge, you're not getting to it. >> Some of the important lessons that we learned. So I actually did this exact job in the air force. I was a combat communicator, which meant that we took, by pure coincidence I'm back in this, like, I did not intend for this to happen its pure coincidence, (Lisa laughing) but, you know, we communicate, we went out to the Edge, right. We went out to the Near Edge and we did all of this stuff. And the biggest lesson, I think learning from doing this or doing that and then going into this is that the world doesn't have to revolve around SharePoint anymore (Lisa laughing) because we can shape our own habitation (Both laughing) >> That is good to know. >> If it can be done on SharePoint, the air force and the army will do it in SharePoint, I promise you. They've done some actually terrifying things with it. All joking aside though, I think that one of the things that we learned was the difference between like something being complex and complicated when it came to systems engineering and management, like this is a very complex system it's actually orders of magnitude more complex than the current deployments that are out there which is effectively VMware and you're migrating virtual machines across multiple physical nodes in these remote data centers. But it's also complicated, it's really difficult to manage these deployments and the hardware. And I remember like when I was in combat comm, we had this 72 hour goal to get all of our systems up. And it was kind of like a 50-50, if we would make it, it felt like most of the time where you had priorities for getting things up and running. And obviously, you know, that certain applications weren't as important as others. So they were the ones that had to fall on the wayside if you're going to make your 72 hour mark. But I'm just thinking about like how difficult it was to deploy and manage all of this stuff and now with Kubernetes, yes, the complexity is far higher, but we can make it so it's not as complicated. We can offload a lot of that brain sweat, the people in the rear echelon, where they can connect in remotely after you come up and you get reached back, they push your config and your mission profile is there. And now you're focused on the mission you're not focused on debugging pods, and you're focused on the mission and not focused on, you know, why my virtual machine didn't migrate or something like that. And we can get applications that are built in-house and updated continuously, and we can verify and validate the sources of where these things are coming from. And all of these are important problems to everybody, not just the military, but the military tends to have the money and the ability to think about these things first, 'cause that's where these problems tend to get solved first. >> So interesting. You've sort of had this circular experience being in the air force, now coming back and working on projects like this, what are some of the things that Hypergiant has learned? And some of the things that are next next for Hypergiant as a company? >> I think that we are getting really good at being a small contractor in the Federal space where we actually were just awarded an IDIQ with a cap of $950 million in a small group of, I think, 23 other companies. And so that shows right there the investment that the Federal Government has in us and the potential that they see for us to build and deliver these highly tailored and specialized solutions. The other thing that we've learned is how to form like coalitions to collaborate with a lot of these other smaller companies. I think that the days of seeing the Defense Industrial Base dominated by the same four people or five people are over. And it's not that these people, I mean, they've been, they've basically been propping up most of the defense industry for a very long time and I think a lot of people would argue that, you know, this is a problem, right, you have this near monopoly of a very few people, but the other thing is that they're not as nimble, they grow by acquisition and we have this ability to be highly tailored and specialized and we don't need to do everything in the world to survive. We can go and form coalitions with other groups to go solve a particular problem. Like we're great at AI and ML, and we're great at DevSecOps, then maybe we're not so great at, you know, hardware or you know, things like that. Like we can go partner up with these people and solve problems together and we don't have to be a Boeing to do it and you don't have to go hire a Boeing to do this. And I think that's really, really great, no slight to Boeing, but I think it's really great that it's a lot easier for smaller companies to do this and we are navigating this new world and we're bringing Agile into the government and that's, yeah, in some cases we have to drag them, kicking and screaming into this decade, but, you know, that's what we're doing and I'm very excited to see that because when I was in Agile and DevOps, those were words you didn't say, you weren't allowed to do that. >> No. >> Now they've done a complete 180, it's really cool. >> That's cool. I have a minimum that brings in thought diversity, having more companies to work with, but to your point, the agility that you bring in as a smaller company helping them to actually embrace Agile, that's huge because to your point, that's kind of historically not what government organizations are used to. So it sounds like a little bit they've learned a tremendous amount from working with small companies like Pepperdine. >> I like the thing so. Platform One is a fantastic example. So it was really started as a what we're calling software factories within the air force and within the DOD and other DOD branches have now started to replicate the pattern. So we have several software factories within the air force and Platform One is like the DevSecOps Software factory, and we have the ski camp and space camping, Kobayashi Maru and you're noticing a theme here (laughing) and so they're very nerdy names, but so we have these software factories and there's all these projects are being worked. But one of the amazing things I noticed when I showed up to work on the first day was that I had no idea who was uniformed and who was civilian. It was a completely badge off rank, off situation. Very few people showed up in uniform and the ones that did typically had their blouse off so you had no idea what their rank was. Everybody went by first name and we behaved like a start-up. And these civilians were coming from other startups like Hypergiant or a Timo or other very small, very specialized groups and SUSE RGS, of course they were there too and they're embedded in several different teams. And so you have this, like this quasi company that got this startup really that got formed and the culture is very, you know, very varies, you know, bay area startup type in some ways, for both better and worse. There's, I mean, we're, definitely full tilt on (laughs) on the Agile train there, but it's just, it's like nothing I've ever seen inside the DOD. And they're not just learning from these small companies and from Agile companies, but they're behaving like them. And it's spreading, they're seeing what work is getting done and what can be accomplished and how you can continuously deliver value instead of working for, you know, six or eight months and then showing the customer something and them hating it and you sending it back and, you know, it's more of a continuous improvement type thing. And I think that they're embracing that and I'm very excited to see it. >> That's important 'cause changing a culture is incredibly hard but seeing and hearing that they're embracing that is exciting. And I'm sure there's going to be many more things you could talk about generally, but I got to ask you if somebody like SUSE gave you $250,000, and you could buy one of the tickets on Branson's next flight, would you do it? >> I mean, yeah, why would I not? Like, how can I pass up a trip, (Lisa laughing) you know, go to the Edge of space. >> The Far Edge. >> Like yeah, the Far Edge, maybe I'll just, you know, hurdle the satellite out the window, as you know, we're up there, you know, peak and probably could throw it quite that fast, but we'll see. (Lisa laughing) But yeah, no, I think I would take the trip, yeah, that'd be fun. >> You're brave. Brave than I'm, I don't know. Well, Bren it's been delightful talking to you. Thank you for sharing what you guys at Hypergiant and SUSE have been doing together, the Department of Defense, the exciting things going on there and for the new definitions and my lexicon of the Edge, it's been great talking to you. >> Thank you, have a great day. >> You too. For Bren Briggs, I'm Lisa Martin. You're watching a CUBE Conversation. (digital music)