>>You can put this in a conference. >>Good morning and welcome back to Dallas. Ladies and gentlemen, we are here with the cube Live from, from Supercomputing 2022. David, my cohost, how you doing? Exciting. Day two. Feeling good. >>Very exciting. Ready to start off the >>Day. Very excited. We have two fascinating guests joining us to kick us off. Please welcome Pete and Armando. Gentlemen, thank you for being here with us. >>Having us, >>For having us. I'm excited that you're starting off the day because we've been hearing a lot of rumors about ethernet as the fabric for hpc, but we really haven't done a deep dive yet during the show. Y'all seem all in on ethernet. Tell us about that. Armando, why don't you start? >>Yeah. I mean, when you look at ethernet, customers are asking for flexibility and choice. So when you look at HPC and you know, infinite band's always been around, right? But when you look at where Ethernet's coming in, it's really our commercial and their enterprise customers. And not everybody wants to be in the top 500. What they want to do is improve their job time and improve their latency over the network. And when you look at ethernet, you kinda look at the sweet spot between 8, 12, 16, 32 nodes. That's a perfect fit for ethernet and that space and, and those types of jobs. >>I love that. Pete, you wanna elaborate? Yeah, yeah, >>Yeah, sure. I mean, I think, you know, one of the biggest things you find with internet for HPC is that, you know, if you look at where the different technologies have gone over time, you know, you've had old technologies like, you know, atm, Sonic, fitty, you know, and pretty much everything is now kind of converged toward ethernet. I mean, there's still some technologies such as, you know, InfiniBand, omnipath that are out there. Yeah. But basically there's single source at this point. So, you know, what you see is that there is a huge ecosystem behind ethernet. And you see that also, the fact that ethernet is used in the rest of the enterprise is using the cloud data centers that is very easy to integrate HPC based systems into those systems. So as you move HPC out of academia, you know, into, you know, into enterprise, into cloud service providers is much easier to integrate it with the same technology you're already using in those data centers, in those networks. >>So, so what's this, what is, what's the state of the art for ethernet right now? What, you know, what's, what's the leading edge, what's shipping now and what and what's in the near future? You, you were with Broadcom, you guys design this stuff. >>Yeah, yeah. Right. Yeah. So leading edge right now, I got a couple, you know, Wes stage >>Trough here on the cube. Yeah. >>So this is Tomahawk four. So this is what is in production is shipping in large data centers worldwide. We started sampling this in 2019, started going into data centers in 2020. And this is 25.6 tets per second. Okay. Which matches any other technology out there. Like if you look at say, infin band, highest they have right now that's just starting to get into production is 25 point sixt. So state of the art right now is what we introduced. We announced this in August. This is Tomahawk five. So this is 51.2 terabytes per second. So double the bandwidth have, you know, any other technology that's out there. And the important thing about networking technology is when you double the bandwidth, you don't just double the efficiency, it's actually winds up being a factor of six efficiency. Wow. Cause if you want, I can go into that, but why >>Not? Well, I, what I wanna know, please tell me that in your labs you have a poster on the wall that says T five with, with some like Terminator kind of character. Cause that would be cool if it's not true. Don't just don't say anything. I just want, I can actually shift visual >>It into a terminator. So. >>Well, but so what, what are the, what are the, so this is, this is from a switching perspective. Yeah. When we talk about the end nodes, when we talk about creating a fabric, what, what's, what's the latest in terms of, well, the kns that are, that are going in there, what's, what speed are we talking about today? >>So as far as 30 speeds, it tends to be 50 gigabits per second. Okay. Moving to a hundred gig pan four. Okay. And we do see a lot of Knicks in the 200 gig ethernet port speed. So that would be, you know, four lanes, 50 gig. But we do see that advancing to 400 gig fairly soon. 800 gig in the future. But say state of the art right now, we're seeing for the end nodes tends to be 200 gig E based on 50 gig pan four. Wow. >>Yeah. That's crazy. Yeah, >>That is, that is great. My mind is act actively blown. I wanna circle back to something that you brought up a second ago, which I think is really astute. When you talked about HPC moving from academia into enterprise, you're both seeing this happen. Where do you think we are on the adoption curve and sort of in that cycle? Armand, do you wanna go? >>Yeah, yeah. Well, if you look at the market research, they're actually telling it's 50 50 now. So ethernet is at the level of 50%. InfiniBand is at 50%. Right. Interesting. Yeah. And so what's interesting to us, customers are coming to us and say, Hey, we want to see, you know, flexibility and choice and hey, let's look at ethernet and let's look at InfiniBand. But what is interesting about this is that we're working with Broadcom, we have their chips in our lab, we have our switches in our lab. And really what we're trying to do is make it easy to simple and configure the network for essentially mpi. And so the goal here with our validated designs is really to simplify this. So if you have a customer that, Hey, I've been in fbe, but now I want to go ethernet, you know, there's gonna be some learning curves there. And so what we wanna do is really simplify that so that we can make it easy to install, get the cluster up and running, and they can actually get some value out of the cluster. >>Yeah. Peter, what, talk about that partnership. What, what, what does that look like? Is it, is it, I mean, are you, you working with Dell before the, you know, before the T six comes out? Or you just say, you know, what would be cool, what would be cool is we'll put this in the T six? >>No, we've had a very long partnership both on the hardware and the software side. You know, Dell has been an early adopter of our silicon. We've worked very closely on SI and Sonic on the operating system, you know, and they provide very valuable feedback for us on our roadmap. So before we put out a new chip, and we have actually three different product lines within the switching group within Broadcom, we've then gotten, you know, very valuable feedback on the hardware and on the APIs, on the operating system that goes on top of those chips. So that way when it comes to market, you know, Dell can take it and, you know, deliver the exact features that they have in the current generation to their customers to have that continuity. And also they give us feedback on the next gen features they'd like to see again in both the hardware and the software. >>So, so I, I'm, I'm just, I'm fascinated by, I I, I always like to know kind like what Yeah, exactly. Exactly right. Look, you, you start talking about the largest super supercomputers, most powerful supercomputers that exist today, and you start looking at the specs and there might be 2 million CPUs, 2 million CPU cores, yeah. Ex alop of, of, of, of performance. What are the, what are the outward limits of T five in switches, building out a fabric, what does that look like? What are the, what are the increments in terms of how many, and I know it, I know it's a depends answer, but, but, but how many nodes can you support in a, in a, in a scale out cluster before you need another switch? What does that increment of scale look like today? >>Yeah, so I think, so this is 51.2 terras per second. What we see the most common implementation based on this would be with 400 gig ethernet ports. Okay. So that would be 128, you know, 400 giggi ports connected to, to one chip. Okay. Now, if you went to 200 gig, which is kind of the state of the art for the Nicks, you can have double that. Okay. So, you know, in a single hop you can have 256 end nodes connected through one switch. >>So, okay, so this T five, that thing right there inside a sheet metal box, obviously you've got a bunch of ports coming out of that. So what is, what does that, what's the form factor look like for that, for where that T five sits? Is there just one in a chassis or you have, what does that look >>Like? It tends to be pizza boxes these days. Okay. What you've seen overall is that the industry's moved away from chassis for these high end systems more towards pizza, pizza boxes. And you can have composable systems where, you know, in the past you would have line cards, either the fabric cards that the line cards are plugged into or interface to these days, what tends to happen is you'd have a pizza box, and if you wanted to build up like a virtual chassis, what you would do is use one of those pizza boxes as the fabric card, one of them as the, the line card. >>Okay. >>So what we see, the most common form factor for this is they tend to be two, I'd say for North America, most common would be a two R U with 64 OSF P ports. And often each of those OSF p, which is an 800 gig e or 800 gig port, we've broken out into two 400 gig quarts. Okay. So yeah, in two r u you've got, and this is all air cooled, you know, in two re you've got 51.2 T. We do see some cases where customers would like to have different optics, and they'll actually deploy a four U just so that way they have the face place density, so they can plug in 128, say qsf P one 12. But yeah, it really depends on which optics, if you wanna have DAK connectivity combined with, with optics. But those are the two most common form factors. >>And, and Armando ethernet isn't, ethernet isn't necessarily ethernet in the sense that many protocols can be run over it. Right. I think I have a projector at home that's actually using ethernet physical connections. But what, so what are we talking about here in terms of the actual protocol that's running over this? Is this exactly the same as what you think of as data center ethernet, or, or is this, you know, RDMA over converged ethernet? What, what are >>We talking about? Yeah, so our rdma, right? So when you look at, you know, running, you know, essentially HPC workloads, you have the NPI protocol, so message passing interface, right? And so what you need to do is you may need to make sure that that NPI message passing interface runs efficiently on ethernet. And so this is why we want to test and validate all these different things to make sure that that protocol runs really, really fast on ethernet, if you look at NPI is officially, you know, built to, Hey, it was designed to run on InfiniBand, but now what you see with Broadcom and the great work they're doing now, we can make that work on ethernet and get, you know, it's same performance. So that's huge for customers. >>Both of you get to see a lot of different types of customers. I kind of feel like you're a little bit of a, a looking into the crystal ball type because you essentially get to see the future knowing what people are trying to achieve moving forward. Talk to us about the future of ethernet in hpc in terms of AI and ml. Where, where do you think we're gonna be next year or 10 years from now? >>You wanna go first or you want me to go first? I can start. >>Yeah. Pete feels ready. >>So I mean, what I see, I mean, ethernet, I mean, is what we've seen is that as far as on the starting off of the switch side, is that we've consistently doubled the bandwidth every 18 to 24 months. That's >>Impressive. >>Yeah. So nicely >>Done, casual, humble brag there. That was great. That was great. I love that. >>I'm here for you. I mean, I think that's one of the benefits of, of Ethan is like, is the ecosystem, is the trajectory, the roadmap we've had, I mean, you don't see that in any other networking technology >>More who, >>So, you know, I see that, you know, that trajectory is gonna continue as far as the switches, you know, doubling in bandwidth. I think that, you know, they're evolving protocols. You know, especially again, as you're moving away from academia into the enterprise, into cloud data centers, you need to have a combination of protocols. So you'll probably focus still on rdma, you know, for the supercomputing, the a AIML workloads. But we do see that, you know, as you have, you know, a mix of the applications running on these end nodes, maybe they're interfacing to the, the CPUs for some processing, you might use a different mix of protocols. So I'd say it's gonna be doubling a bandwidth over time evolution of the protocols. I mean, I expect that Rocky is probably gonna evolve over time depending on the a AIML and the HPC workloads. I think also there's a big change coming as far as the physical connectivity within the data center. Like one thing we've been focusing on is co-pack optics. So, you know, right now this chip is all, all the balls in the back here, there's electrical connections. How >>Many are there, by the way? 9,000 plus on the back of that >>352. >>I love how specific it is. It's brilliant. >>Yeah. So we get, so right now, you know, all the thirties, all the signals are coming out electrically based, but we've actually shown, we have this, actually, we have a version of Hawk four at 25 point sixt that has co-pack optics. So instead of having electrical output, you actually have optics directly out of the package. And if you look at, we'll have a version of Tomahawk five Nice. Where it's actually even a smaller form factor than this, where instead of having the electrical output from the bottom, you actually have fibers that plug directly into the sides. Wow. Cool. So I see, you know, there's, you know, the bandwidth, there's radis increasing protocols, different physical connectivity. So I think there's, you know, a lot of things throughout, and the protocol stack's also evolving. So, you know, a lot of excitement, a lot of new technology coming to bear. >>Okay. You just threw a carrot down the rabbit hole. I'm only gonna chase this one. Okay. >>All right. >>So I think of, I think of individual discreet physical connections to the back of those balls. Yeah. So if there's 9,000, fill in the blank, that's how many connections there are. How do you do that in many optical connections? What's, what's, what's the mapping there? What does that, what does that look like? >>So what we've announced for TAMA five is it would have fr four optics coming out. So you'd actually have, you know, 512 fiber pairs coming out. So you'd have, you know, basically on all four sides, you'd have these fiber ribbons that come in and connect. There's actually fibers coming out of the, the sides there. We wind up having, actually, I think in this case, we would actually have 512 channels and it would wind up being on 128 actual fiber pairs because >>It's, it's miraculous, essentially. It's, I know. Yeah, yeah, yeah, yeah. Yeah. So, so, you know, a lot of people are gonna be looking at this and thinking in terms of InfiniBand versus versus ethernet. I think you've highlighted some of the benefits of specifically running ethernet moving forward as, as hpc, you know, which is sort of just trails slightly behind supercomputing as we define it, becomes more pervasive AI ml. What, what are some of the other things that maybe people might not immediately think about when they think about the advantages of running ethernet in that environment? Is it, is it connecting, is it about connecting the HPC part of their business into the rest of it? What, or what, what are the advantages? >>Yeah, I mean, that's a big thing. I think, and one of the biggest things that ethernet has again, is that, you know, the data centers, you know, the networks within enterprises within, you know, clouds right now are run on ethernet. So now if you want to add services for your customers, the easiest thing for you to do is, you know, the drop in clusters that are connected with the same networking technology, you know, so I think what, you know, one of the biggest things there is that if you look at what's happening with some of the other proprietary technologies, I mean, in some cases they'll have two different types of networking technologies before they interface to ethernet. So now you've got to train your technicians, you train your, your assist admins on two different network technologies. You need to have all the, the debug technology, all the interconnect for that. So here, the easiest thing is you can use ethernet, it's gonna give you the same performance. And actually in some cases we seen better performance than we've seen with omnipath than, you know, better than in InfiniBand. >>That's awesome. Armando, we didn't get to you, so I wanna make sure we get your future hot take. Where do you see the future of ethernet here in hpc? >>Well, Pete hit on a big thing is bandwidth, right? So when you look at train a model, okay, so when you go and train a model in ai, you need to have a lot of data in order to train that model, right? So what you do is essentially you build a model, you choose whatever neural network you wanna utilize, but if you don't have a good data set that's trained over that model, you can't essentially train the model. So if you have bandwidth, you want big pipes because you have to move that data set from the storage to the cpu. And essentially, if you're gonna do it maybe on CPU only, but if you do it on accelerators, well guess what? You need a big pipe in order to get all that data through. And here's the deal. The bigger the pipe you have, the more data, the faster you can train that model. So the faster you can train that model, guess what? The faster you get to some new insight, maybe it's a new competitive advantage. Maybe it's some new way you design a product, but that's a benefit of speed you want faster, faster, faster. >>It's all about making it faster and easier. It is for, for the users. I love that. Last question for you, Pete, just because you've said Tomahawk seven times, and I'm thinking we're in Texas Stakes, there's a lot going on with with that making >>Me hungry. >>I know exactly. I'm sitting up here thinking, man, I did not have a big enough breakfast. How do you come up with the name Tomahawk? >>So Tomahawk, I think you just came, came from a list. So we had, we have a tri end product line. Ah, a missile product line. And Tomahawk is being kinda like, you know, the bigger and batter missile, so, oh, okay. >>Love this. Yeah, I, well, I >>Mean, so you let your engineers, you get to name it >>Had to ask. It's >>Collaborative. Oh good. I wanna make sure everyone's in sync with it. >>So just so we, it's not the Aquaman tried. Right, >>Right. >>The steak Tomahawk. I >>Think we're, we're good now. Now that we've cleared that up. Now we've cleared >>That up. >>Armando P, it was really nice to have both you. Thank you for teaching us about the future of ethernet N hpc. David Nicholson, always a pleasure to share the stage with you. And thank you all for tuning in to the Cube Live from Dallas. We're here talking all things HPC and Supercomputing all day long. We hope you'll continue to tune in. My name's Savannah Peterson, thanks for joining us.

>>from San Francisco. It's the Q covering pager duty Summit 2019. Brought to you by pager Duty. >>Hey, welcome back there. Ready, Geoffrey? Here with the cue, we're pager duty Summit in the historic Western St Francis Hotel, downtown San Francisco. I think they've outgrown the venue. The place is packed to the gills. Standing, rolling, the keynote Really excited of our next guest. Someone who's been to this industry for awhile really done some super cool creative things. He's given the closing keynote. We're happy to have him here right now. That's Mitchell Hashimoto from Hachiko. It's great to see you. >>Good to see you too. Thanks for having >>absolutely so just a quick overview before we get into it on hot chic or for the people in our familiar >>Sure, so hospitals a company that what we try to do is help people adopt cloud, but more, more realistically, Adolfo multi cloud and hybrid cloud the real world complexities. That cloud isn't just a technical landing point, but it's really way You deliver software. You want to deliver more applications, you want to connect them faster. You want to do this in an automated away infrastructure is code of all these modern practices way. Build a suite of tools Thio provisions secure, connect and run those applications for separate products that we sell that you could adopt separately. Good mix and match. That's That's what we've been doing for a long time. Based on open source Software, Way started purely as an open source community and have grown into an enterprise cos that's that's That's the elevator pitch. >>No, it's great, but it's a great story, >>right? Europe, Europe in Seattle got some access to some cloud infrastructure and really solve your own problem. Figured out other people of that problem and then really built a really cool, open source kind of based software company. >>Yeah, I mean, I think the amount of people that had the problem I was facing personally was orders of magnitude more than I expected. I've told other people we never expected to start even a business around this. It was just scratching and building technical solutions. But a CZ, as we sort of worked at startups, started talking a bigger and bigger companies. It just kept everyone kept saying, Yes, I have that problem and it's only grown since then, right surprising, >>and the complexity has only grown exponentially before. The you know, years ago, there was this bright, shiny new object called a W s. I mean, I love Bezos is great line that nobody even paid attention. You have six or seven years. They've got a head start and kind of this Russian. Now there's been a little bit of a fallback as people trying to figure out what to go where now it's hybrid cloud and horses for courses. So a lot of great complexity, which is nothing but good news for you. >>Absolutely. I told this story before, but our first year incorporated company I actually got hung up on by an analyst because I said way we're trying to solve a multi cloud problem and they said that's not a real problem when it will never be a real problem. They hung up on me on it was a bet, then, and and I think they're the expectation that was it was gonna be Eight of us is gonna be physical infrastructure and the physical infrastructure days were numbered. It was gonna get acts out. It was just gonna all go to eight of us and our conviction was that you would have both forever and or for a very long time. And then people like Azure, Google and others would pick up and and that's been true. But I think what we didn't expect, the complexity that got introduced with things like containers in Kubernetes because it's not like Clouded option finished in the next started It all came at once. So now Riel Cos they're dealing with the complexity of their still trying to move the clouds. They're trying to get more out of their physical infrastructure, trying to adopt kubernetes. Now people are starting to peck at them about server list. So there the complexity is is a bit crazy and review our job trying to simplify that adoption make you get the most out of >>right. And that was before you could get a piece of Ian where inside of AWS, get a get a piece of the Google Data Center inside your own data center. So it just continues to get crazier. >>Yes, yeah, So you're giving a closing keynote on a >>new project. You're working on fault, and it's an existing project. Justin cry. They're old, but but I think you talked about before we turn the cameras on. It's really more of a kind of an attitude in a and a point of view and a way to go after the problem. So I wonder if you could kind of dig into a little bit of What did you see? How did you decide to kind of turn the lens a little bit and reframe this challenge? Yeah, >>I think the big picture of story I'm trying to tell him the keynote is that everybody? Anything You look around the technical nontechnical, this table, that glass. Like everything you look at, it trickles back to the idea of one or a small group of people, and it takes an army to make it show up on this table. But it starts by somebody's vision, and everything was created by somebody. So I'm talking about vault, something we made and, you know, why don't we create it? And why do we make it the way we did? And you know, another thing I say is people ask, Why did you start hot record for having this vision? Something I constantly told myself was wine on me. I get someone's gonna do it. Why not make it? Could be anybody, like I'll give it a shot. Why not? And Bolt was that way. We Armand. I'm a co founder. Way took security classes in college, but we don't have a formal security background. We didn't work in security in industry. So the odds of us launching a security product that is so prevalent today whether you know it or not, it's behind the scenes very prevalent were stacked against us. How did that happen? And that's that's sort of what I've been going to talk about. >>Let's go. But do >>dive into a little bit on the security challenge because it's funny, right? Everyone always says, Right. Security's got to be baked in and you've got these complex infrastructure and everything's connected with AP eyes, the other people's applications and, oh yes, delivered through this little thing that you carry around. And maybe the network's not working well or the CPS running low are You're running iPhone five. And of course, it's not gonna work on most modern app. Yeah, bacon security always do, but that's easy to say. It's much harder to do, you know. Still, people want to build moats and castles and drawbridges, and that's just not gonna work anymore. >>Exactly. So you exactly hit upon the two major issues that we recognize there felt we recognize. One was that a lot of people were saying it. Very few people were doing it on. The reality was it was hard to do. Everyone knew theoretically what they should do. No one, no one thought. Oh yeah, saving somebody's personal information in plain text in the daytime. It's a good idea. Nobody thought that Everyone said it should be encrypted, but encryption is hard. So maybe one day, so no one was doing it. And then the other side of it was the people that were doing it where the world's largest companies, because the solutions were catered towards his mindset of of castle and moats, which works totally fine in a physical tradition environment but completely breaks down in a cloud world where there is no four perimeters anymore. It's >>still there, There. >>You're one AP I call away from opening everything to the Internet. So how do you protect this? And we've seen a lot of trends change towards zero Trust and ServiceMaster Mutual feel like there's a lot of stuff that happened way sort of jumped on that. >>Yeah, so So you're using, like, multi level encryption, and I've read a little bit on the website. It's way over my head, I think. But, you >>know, the basics are just making kryptonite. Christian makes security, cloud infrastructure, security approachable by anybody and a core philosophy. Our company, Hashi Hashi Hashi. My name means bridge, and that is a core part of our culture. Which is you can't just have, ah, theoretical thing or a shiny object and leave people hanging. You gotta give them a bridge, a path to get there, right? And so we say, with all our technology, one of the crawl, walk and run adoption periods and with security it's the same is that to say you're secure means something totally different everybody for a bank to be secure, it's a lot more than for a five person started to be secure. So how do you give somebody a solutions they could adopt? Check the security box for themselves at every path of the lake, and bald is one of the tools that way have individuals using it, and we have the world's largest companies, almost 10% of the global 2000 paying customers evolved many more open source users on its scales the entire spectrum. >>Wow. So you keep coming up >>with lots of new, uh, new projects as we get ready to flip the counter to 2020. What are some of the things you're thinking about? >>I think the big one, you know, that our focus is right now is service. Miss Vault is we're big enough company now where we always have teams working on every every one of our projects we have release is going out. The thing we've been talking about the most is the service mess thing. I think Cloud as a mainstream thing, Let's say, has has existed for seven or eight years. It's since it's been released. It's been over in almost 15 but as a thing that people have, that is a good idea. Seven or eight years and you know we've touched security. Now we've touched how infrastructures managed touch developers. I think a place that's been relatively untouched and has gotten by without anyone noticing has been networking and network security. They're they're really doing things the way they've always done things, and I think that's been okay because there's bigger fish to fry. But I think the time has come and networking as a bull's eye on it. And people are looking at What is networking mean in a cloud world and service mash appears to be the way that is gonna happen. Way have our own service mess solution called Council on Our Approaches Standard Hasta Corp. It's nothing new. It's We're gonna work with everything containers, kubernetes, viens physical infrastructure. We're gonna make it all work across multiple data centers. That is our approach service fashion, solving that challenge. >>What's the secret sauce? >>I mean, it's not that secret, right? >>It's just building. Just execute. Better understand that this header >>JD is the problem, right? Right, I said, This is our keynote a couple weeks ago that there are a lot of service messes out there, and nine out of 10 of them are solving a solution for a single environment, whether it's kubernetes or physical environment. And I think that's a problem. But it's not the problem. The problem to me is how do I get my kubernetes instances pods to communicate to my NSX service is on my physical infrastructure. That is the problem as people, whether that's temporary, not and they intend to move the communities or whatever. It's that's the reality. And how do you make that work? And that is what we're focused on solving that problem >>just every time I hear service mess. I think there was a company a while ago that sold the CSC. Probably like 2013. Didn't really get into That is a as a good, happy story. But they were early on the name. Yeah. Yeah. So last thing pager duty were Pedrie. What? You guys doing a page of duty? >>Sure. So we've been I've actually been a paying customer pager duty since before we even made this company in my previous job was a customer wear now, still customers. So we still use it internally. But in addition to that way, do integration across the board. So with terra form our infrastructure provisioning tool way have a way to manage all pager duty as code and as your complexion pager duty rises instead of clicking through a u. I being able to version and code everything and have that realize itself and how he works very valuable from like a service MASH consul standpoint. Hooking in the monitoring to the alerting of Pedro duty is a big thing that we do so tying those together. So it's very symbiotic. I love pager duty as a user and a partner. There's a lot here. >>Yeah, is pretty interesting slide when Jennifer put up in the keynote where it listed so many integration points with so many applications with on the outside looking in and you're like how you're integrating with spunk, that making how you're innovating with service. Now that doesn't make any sense. How Integrated was in Desperate. These were all kind of systems of record, but really, there's some really elegant integration points to make. This one plus one equals three opportunity between these applications. >>Yeah, I think it's very similar to the stuff we do with Walton Security. It's like the core permanence. Everybody needs him like with security. Everyone is an auto. Everyone needs traceability. Everyone needs access control. But rebuilding that functionality and every application is unrealistic. And paging and alerting an on call and events are the same thing. So it's you'd rather integrate and leverage those systems that make that your nexus for that specific functionality. And that's where Page duties. Awesome way. Step in, >>which was always great to catch up. Good luck on your keynote tomorrow. And really, it's a really amazing story to watch that you got You guys have built >>Well, thank you very much. >>All right. He's Mitchell. I'm Jeff. You're watching the cue. Were paid your duty, Simon in downtown San Francisco. Thanks for watching. We'll see you next time.

(upbeat music) >> From Sand Hill Road in the heart of Silicone Valley, it's theCube, presenting the People First Network, insights from entrepreneurs and tech leaders. >> Hello everyone, I'm John Furrier with theCube. We are here in Sand Hill Road at Mayfield office here talking about entrepreneurship, People First, this is our co-created program with Mayfield. I'm John Furrier, your host, we're with Mitchell Hashimoto, who's the co-founder and co-CEO at HashiCorp. Great to see you, good to keep alumni, you're back on theCube . Thanks for joining me today. >> Yeah, thanks so much, I was here so long ago. (John laughs) >> Like five or six years ago. >> So, we've been really psyched about the program that Mayfield's put together called People First. They're celebrating their 50th anniversary as a venture capital firm, which is historic in the sense that it's kind of still a young industry. Think about it. And love to have entrepreneurs come on because you've been very successful. We talked years ago. I think, the first year you were formed and Cloud certainly has happened. Open Source continues to pump more value. I mean, you get things out there coming out of Google, some ridiculously amazing... The goodness in Open Source is certainly driving a lot of great software development. You've been a part of that so thanks for joining us. So I got to ask you, you guys are growing right now, you're Venture backed, you got a unique culture. Explain HashiCorp, 'cause you guys have a unique business. You're in Open Source, you're in Cloud, you're a distributed workforce. Take a minute to explain what you guys are doing. >> Yeah, so we are trying to build or have been building, sort of infrastructure software of the future. We've been saying that since we were founded and what's been interesting is the future has changed quite a bit in the past six years so there's been Cloud, that was the big thing when we were founded and then containers and now schedulers and Kubernetes and things like that. And while we're doing that, we're also sort of building what I think is sort of the company of the future, which is over 90% of our workfoce is fully distributed. Basically, unless there's legal reasons not to be distributed, we are distributed and we're in multiple countries, we're in over 40 states. All of our process is built remote first so everything happens, Slack, all our meetings are Zoom. Even our all hands, we present behind a camera, and things like that so I think that's all very unique, but only for now, I think that-- >> How do you do the all hands? That's interesting. Do you have a camera to a zoom or is it a camera live streaming? How do you do the all hands? >> Yeah, so we set up sort of an AV setup in our office because we have a few of the executives in the office that often are presenting on the all hands and we set up a camera feed so that whether you actually decide to go into the office or whether you're at home, we want that experience to be authentic to both sides. We don't want a great in-room experience and then one corner camera that makes it really hard to hear and stuff like that so yeah, you have to walk up to the camera and be part of the zoom to really be part of the all hands. >> So that people feel present and connected? >> Right, exactly, and we force questions to come through Slack. There's no in-person questions. You have to ask on Slack so everyone can see them and things like that, so-- >> That's awesome. Talk about the journey as you started. You have a co-founder. You guys have an interesting relationship. How did this all get started? What was the beginning genesis of HashiCorp like and take us through some of the early days. >> Sure, so I'm very lucky, I have a co-founder who before the company, we were best friends and after the company or during the company, we're still best friends so (laughs) which it isn't always the case, but in terms of HashiCorp itself, we're super lucky 'cause we went to the University of Washington, up in Seattle, and this was in sort of the mid-2000s and this is a good time to be up there, 'cause Cloud was starting to emerge and we were sort of equidistant geographically, across the lake, if you will, to Amazon, Google, and Microsoft and so, we were getting early access to what they thought was sort of the Cloud at the time and it was rapidly changing. We were getting access to the servers, with the APIs, and all this stuff and being a university without a lot of funding, my job there was sort of to help us utilize all these resources and so in the mid 2000s, Armand and I were already realizing, we're on the same team, Armand and I were already realizing that this is not a solved problem by any means, I mean this is a new problem and that eventually, years later, became the genesis-- >> and what was that problem that you saw immediately? >> It was sort of like multi-Cloud, resource management, deployment, security, it's funny 'cause it's... Over 10 years later and it's... It is the problem that Enterprise are hitting right now. >> Think about the early days of Amazon. I still have these memory flashbacks of EC2, long URLs, it's like, okay, now how to I redirect my web servers to this, like, so it was easy to stand up on EC2 instance, put a little S3 to it, then it's like, okay now what? >> Yeah, we're at the-- >> Little red scale, I put this in there, what's kind of there. So again, a little early, kind of build your own kind of a junkyard. You build a car out of some spare parts. But then it had to mature really fast. >> Yeah, we're are the Day Zero state then and now we're firmly in like Day Two. >> And so what was the next step. Can you peg the journey for us, because obviously, they grew up really fast and then they really kind of hit a tipping point around 2010, 11, 12, 13, and kind of grew like a weed >> Yup, yeah, so around that time frame you just painted to 2012 is when Enterprise was sort of adopting it. And I think a lot of that was single Cloud focus. There was very much like, this is our first Cloud so we're going to land purely on Amazon or something and focus on that and we're at the point now, about six years later, 2018, where the maturity around operating the Cloud is sort of well understood and companies are now starting to sort of use what's best for the job and also realized that there's multiple Clouds and we're keeping our private data centers and also, there's new things coming on the scene above Cloud sort of higher level, like Kubernetes, and how we're going to manage all this and so, we like to describe it as sort of the mindset is like the Cloud operating model. It's like you can't operate your resources in the Cloud the same way you do on Prim and people are starting to get that. That's like automation, very people-focused workflows, things like that, and companies are getting that and so now the challenge is these heterogeneous environments. >> So, the top conversation in our office and everyone loves when I bring this up, I want to get your definition and opinion, >> Okay. >> Is Kubernetes. >> Sure. >> Kubernetes, just, a lot of people love it. I've been having Kubernetes dreams these days 'cause there's so much Kubernetes conversation. (Mitchell laughs) you got Kubernetes, you got the notion of Service Mesh is right around the corner, StateFul applications with net problems really hard to work on. Stateless has been around for awhile. What's the importance of Kubernetes? What's the impact, in your opinion, expert opinion, why is Kubernetes important and what's the impact of Kubernetes? >> Yeah, I think the more abstract answer is the scheduler idea and Kubernetes are built on that and really, it's the idea of like, let's move away from looking at the individual machine and let's start moving higher level to just assuming resources are there. It's sort of like when you write, the transition of when you were writing software from having to know how much memory you had to just, let's just assume it's infinite and put whatever in there and it's someone else's problem and we're sort of moving into that data center, it's like, let's just assume we always have compute and storage and network and let's just deploy and what freedom does that give you and I think that's really what Schedulers give you and also, when you sort of take away huge operability challenges of placing the application and giving that to a computer to put in the right spot, you can now deploy so many more applications because-- >> so you're freed up? >> You're freed up in a lot of ways. It introduces a lot of new challenges, but that's a good problem. You want new challenges, you want to solve the old ones. >> What are some of the new challenges that you see emerging that kind of keep the evolution going? >> I think Service Mesh is a great example we could jump into, which is that the challenge of, we like to describe Service Mesh as three fundamental problems, which is discoverability, configurability, and secure connectivity. If you have two services, that is not a problem because you could hard-code the IPs, you could hard-code the configuration, and you could just hard-code TLS certificates, make it work. When you have thousands of services that are coming and going and people are trying new services all the time, that has to all be automated so the idea of Service Mesh is automating that and making it invisible, automatic, free, and that's new, that's a new problem. >> And that's a huge concept. This is a scalable, scale out, huge concept, and super important. >> Yes, yeah. >> This changes the game at many levels. What would you see that changing? What would some of the, for folks who are just now understanding, what does it change downstream or down the road for enterprises and for businesses? >> I think the biggest change is a mind shift change from sort of perimeter or host-based security to identity and service-based security. So, traditional sort of networking and security is very IP Space focused, it's like does this rack talk to this rack or no and things like that. And that has to all go away because that's restricting the placement, that's not allowing apps to go anywhere. We have to move towards this service can or can't talk to this service, don't care where it is or anything and sort of move from a perimeter to just the perimeter being the app itself so we have to sort of firewall and protect right at the app layer and that's hard to transition, that's tooling change, that's education change, that's team change. >> I want to ask you, I could talk about this forever, Cloud Automation is, I think, one of the most important things. That's only going to make AI more powerful and the data behind it, and as new data emerges, but I got to ask you about some of the new blood coming into the market place because traditionally, if you think about Service Mesh, oh it's a software problem, we'll just solve the software, but you actually got to have networking shops, you got to have to have a computer science or computer engineering. A new skill sets developing really fast in this new, I don't want to, maybe call it under the hood, I don't know what to call it, but maybe, it's an engineering mindset, where people, there's a huge demand for skills in automating. It's not your classic application developers, there's great role for that and there's tons of apps being built, but, I'm talking about a new kind of operator. >> Yes. >> What's your take on this new skill, this new opportunity for people to learn and develop a career? >> Yeah, I think the real way to look at it, I like to look at it, is sort of the difference between creating, sort of doing something once and creating a process to do something. And there's sort of two different tasks, righ. It's like when you get promoted for the first time from you know, to manager. It's like the big challenge is learning how to teach others process and enforcing consistent process, versus actually, you know, doing it yourself. And I think that's the difference between someone who is used to the slinging, let's go back to like the server automation, someone who's used to just manually clicking or slinging bass scripts to do one off task, you could be a wizard at that, but then, try to do that repeatably, safely, 9000 times out of 9000 times and now, that's a resiliency challenge. That's sort of understanding failure modes. It's very different and I think that's the biggest skill set to adopt is, I always sort of push anybody in their job to just what, how do you not do your job? Like, how do you move on to the next problem? >> How do you eliminate your job? >> Yeah, basically/ >> That's almost, like the way I think about it. >> Yeah, what's the process. Is it possible right now? And if it's not possible, what's sort of blocking it? >> So I want to ask you a question and I love this one, going to move on just from the business side in a second, but I want to get your thoughts because I've been having conversations lately with Cloud folks and engineers and developers around two words, replicating and reproducing. >> Okay. >> They're kind of two different concepts. Reproducing is doing the same thing over again. Make that spaghetti sauce, do it again, but did I write it down? Is there a recipe? Or I could just hand you the recipe and say, you make it yourself or automating it. So I think, replicating, I'll say has scale, reproducing requires the same components. Do you see dev ops evolving to a point where, do it once and it's replicated? Or is there some reproduction involved, reproducing things? Where is that, where do you see the tech happening? >> I think inevitably, you're sort of doing both, but my sort of dream world, where I think it'll be still, but I think it's sooner than we expect, but I think sort of like 10 years from now is a safe, sort of stage, it's sort of like every, it doesn't matter if you're Fortune 500 or a new company, sort of the way it infrastructure server management goes is you just start with one server. I like to call it the stem cell server. You just start with one server, you say what you want and just let it go and it's going to either replicate or reproduce, it's either creating something new or it's like creating more copies of itself, but it'll turn into any sort of scale face, book level scale that you would want in theory and I think that, that's sort of my long, you know, fence post, guiding fence post, that I always think about the problem. >> Talk about the culture of your company, you guys have a new CEO, you have a partner you've been best friends with so-- >> I don't think he's that new? >> Yes he is. (both laugh) Okay, he's been around for awhile? >> Couple years, yeah. >> Couple years, so you've had a co-founder dynamic. Did you guys look at each other and say hey, we got to bring a CEO in . Some people like to have one of the founders be the CEO. Talk about that dynamic 'cause that's a struggle for a lot of entrepreneurs to have the self awareness and or the need to do that. >> Yeah so Armand and I made the decision to look for a CEO, if possible, I think three and a half or four years ago, it took us almost two years to find Dave and our motivation is really, it's a few things, one was something our investors told us, which is, long term, you want to do for the company you want to give the company the biggest value you can and like, what do you bring to the company? For us, as founders, our skill set was product vision, engineering, sort of industry strategy, things like that and it wasn't the executive management, financing, building various teams like sales marketing, building out the corporate structure, that wasn't us and so we looked at it and thought, we could learn it, probably, but we would make mistakes and it would be hard, it's just not our passion, it's not what we want to do, or we could try to find someone who aligns with our culture and gets our vision, gets open source, things like that, bring them in and sort of scale to a way where we're giving our startup the best chance it has, which means we give it the value we do, which is engineering and product vision and the new person coming in gives it that sort of corporate maturity and that's exactly what Dave did. >> That's awesome and it's always hard to do that because you got to have real maturity to make that happen so congratulations. >> Thanks, yeah. >> You know, a lot of us have that problem. (chuckles) and then one of my startups like, I need a new CEO, the venture guys were pushing it on you, but it's a challenge, you know, you got to think about, you know... That we didn't have a business model back then, but it's different stories, but that's always a tough one. Now let's talk about the culture around where you started from and where you are now because a lot of the stories around entrepreneurship is team, culture, and how you're going to set up your future of work, which you guys have a good structure. Iterating and figuring out where the tail wind is. Are you at the spot where you thought you'd be at a few years ago when we first met? How has it evolved, where there a little bit of zigs and zags you had to make. What was that like and share some of the journey color commentary with us. >> Sure, I mean, as a company sizes, we're nowhere near where I thought we'd be. I think Armand and I came into it expecting failure most likely and so anything beyond that was just surprise. So that's great. I think the place we are where we thought we'd be is sort of the company culture and stuff and that's something we've been very fiercely protective of and we define our culture sort of as we published them, we call the principal of HashiCorp, which sort of revolve around kindness, honesty, humility, things like that, so it's who would we want to work with and let's put words to it because we don't want to be this nebulous thing and so we've held to that really strongly. We're over 300 people now and every... Something Armand says, which I totally agree with, is I come into work, come into work, I go to my remote office, but I come into work and I'm excited to work with everyone at HashiCorp, which is, in past jobs we've had, we'd come into work and we're excited to work with like two out of 10 people, you know, and that's not a good ratio to have and I think that's what I'm most proud of from the culture side, that the ways we've done that is like we have the principles. We also have something called The Tal, which has been incredibly successful for us, both internally and externally, which is how we view product development and design and that helps sort of align the type of engineer who could get behind our vision and put some words to our vision so it's not again nebulous, whatever the founders think. >> So they have expectations of what's going to be like? >> Mhmm. >> From a coding standpoint, contribution? >> Yeah, from how do you, I like to describe it as how do you build product and how do you... How do you handle people? We have the two sides totally published and we're pretty explicit about it. >> That's awesome. Talk about the role of open source and lots of changing and you're seeing a lot of things like the Linux Foundation, CNCF, massively commercialized, there's tons of money coming in there, but Linux Foundation has done a good job of keeping that pretty pure. Success in entrepreneurship and open source go hand in hand now, it's almost... It's really the perfect storm for creators. >> Yeah. >> But, there's a playbook, there's a way that's changed. Share your vision of how you think open source is today and where it needs to maintain and what could be changed for the better? >> Yeah, I think, so open source today is pretty much a default, expected, accepted, sort of a pattern, which is really nice. It gives you community so you could, you know, Groundswell, anyone could adopt your software, without having to go through a sales person or something like that, which is really important, anyone can contribute and make their mark on the software. It's a great way to sort of get careers started. I think it brings a level of transparency to software that is, you know, you could hide behind closed source. It's like we like to tell our customers, it's like if you don't believe us, not only try it, but go look at how it works. We're telling you the truth. And I think that's really important. I think there's still a lot of challenges around how do companies sort of build successful businesses around it? I think we're doing alright and things like that, but there's still low number of data points. >> Always the challenge is, from looking at your reaction on this, is that as companies get involved, the classic reaction was, oh we got the big companies now in this open source project, it's going to be land grabbed, they're going to put their fingers in there, need better governance. >> Yup. >> Things fracture. Where ideally, it's an upstream project, where everyone contributes for the better good and then people pull it downstream. I mean, that's the basic ethos of open source. That's the main, that's the playbook that we want and that's what you believe, that's the ideal scenario? >> I think that yeah, I think shared ownership is really important, but I also think that sort of unified vision is equally important. So, that's a healthy tension to me, which is that you have a huge community that wants to pull the project in different directions and I think if you don't, if you have a governance that's totally fair, what ends up happening, in my opinion, is you end up getting camels instead of horses, right, like you'd start pulling in all these different directions. You sort of need a slightly unfair governance model so there is somebody that says, this is the direction we're going. And that person needs to be someone that's trusted by the community. >> And Linux was very successful with that too, I mean, you know. >> Right, and I think Linux is an example of a project that like reaches a point where that's, the vision is obvious and clear and it reaches a point where, you know, Linux could step down for a bit and take a break and it still runs fine, but it's a-- >> in the early days, you need a benevolent dictator to say, look, we got to do this. >> Yeah, right, Linux is a 25, 30 year old project versus, you know, some of these CNCF projects are two or three years old and I think that's where you absolutely need strong leadership versus-- >> But we'll see. We'll look at the contribution. We look at that, we obviously follow that pretty heavily and learn to appreciate the Kubernetes commentaries. We think that's super important too. Obviously containers, it's pretty much voted, it's open now so. >> Yes, yeah, yeah. >> (laughs) We know that. Okay, so I got to ask you the final question. As an entrepreneur, access to capital is super important. How did you guys go about it? How did you raise money? How should people raise money today? I'll say your an entrepreneur in the ecosystem, you're out in the front lines building a company. >> Mhmm. >> How did you guys access the capital? How should people figure this out? >> Yeah, I mean you just, you got to tell people why, you know it's a marketing problem, in away, but you got to tell people why what you're working on matters because it's so obvious to you as the founder, that's easy, it's about how do you articulate that and tell people how and why it's important and not just to you, but to the market and how it's going to help people and we did that and I think our biggest challenge was we had to do that across six or seven products, which is, we had a lot of pressure to like, why don't you just do one thing, but it was because for us, what was important was not just what the product did, but the greater vision behind why are we doing six things. And we just, you'd say that and you'd find people who believe it and they help you. >> And as you guys, a great example of you're on a big wave with Cloud and Open Source. How should entrepreneurs and what do you guys do to do this, maybe it's more of advice or anecdotal observation, as you have the dynamics with investors, advisors, service providers, how do you get the most out of them and how do you manage that board dynamic, because when you have an emerging market, there's a danger of saying, we got to lock in a business model. >> Yeah. >> So in Open Source, I'll see a little bit more freedom there 'cause you're open source, but that's always a danger and it's that much more you got to balance that, okay, we got to move the needle, but let's not overdrive too hard. How should entrepreneurs handle the... Taking advantage of their investors and board and how should they manage them or work with them? >> Yeah, I think on one side you need sort of, it's like multiple pillars and on one pillar you need a strong vision, so you need, what won't you sacrifice on, sort of? What's the fence post in the distance and maybe the journey there is slightly different, but you know where you're sort of heading towards so that always grounds you. I think the second thing is sort of a level of pragmatism, like you need to have that vision, but you need to meet your customers where they are and so, you need to figure out what you need to give them today, but still head towards that vision. And when you have those two things, you have a board that is on board with both of those things, you have founders that are dedicated, and you have employees, as well, and everything sort of moves in the right direction. >> But you got to lay that out. >> You have to be pretty explicit about it, yeah. >> Alright, well, congratulations on all your success and looking forward to following up and seeing how you guys are doing. Thanks for coming in and sharing your thoughts today. Appreciate it. >> Thank you. I'm John Furrier here at Mayfield for the 50th anniversary, part of our People First network coverage. I'm John Furrier, thanks for watching. (upbeat music)