 Hello everyone, and welcome back to theCUBE where we are streaming live this week from KubeCon. I am Savannah Peterson, and I am joined by an absolutely stellar lineup of CUBE brilliance this afternoon. To my left, a familiar face. Lisa Martin, Lisa, how are you feeling end of day two? Excellent, it was so much fun today. The buzz started yesterday, the momentum, the swell, and we only heard even more greatness today. Yeah, yeah, absolutely. You know, I sometimes think we'd hit an energy cliff, but it feels like the energy just continues to build. I think we're going to slide right into tomorrow. Yeah, me too. I love it. And we've got two fantastic analysts with us today. Sarbjeet and Keith, thank you both for joining us. We feel so lucky today. Great being back on. Thanks for having us. Yeah, it's nice to have you back on the show. We were, had you yesterday, but I miss hosting with you. It's been a while. It has been a while. We haven't done anything since pre-pandemic, right? Yeah, I think you're right. The before times. The before times. Yeah, back in the day. I always enjoy hosting Lisa, because she's so well-prepared. I don't have to do any research when I come on. Lisa will bring up some, oh, sorry, Jeep, I see that in 2008 you won this award for being just excellent at it. Oh, yeah, yeah, yeah. All right, Keith, so. So did you do his analysis? Yeah. It's all done. Yeah, great. The only part was he's not sitting next to me, so we can't see it. So it's going to be like a magic crystal ball. So a lot of people here, you got some stats in terms of the attendees compared to last year. Yeah, Priyanka told us we were double last year, up to 8,000. We also got the scoop earlier that 2023 is going to be in Chicago, which is very exciting. Oh, that is nice. Yeah, we got to break that here. Excellent. Keith, talk to us about what some of the things are that you've seen the last couple of days, the momentum. What's the vibe? I saw your tweet about the top three things you were being asked. Kubernetes was not one of them. Kubernetes was not one of them. This conference is starting to, it still feels very different than a vendor conference. The keynote is kind of all over the place talking about projects. But the hallway track has been, this is maybe my fifth or sixth coupon in person. And the hallway track is different. It's less about projects and more about how do we adjust to the enterprise? How do we actually do enterprise things and it has been amazing watching this community grow. I'm going to say grow up. And, you know, they're not wearing ties yet, but they are definitely understanding kind of the friction of implementing new technology in the enterprise. Yeah. Subjee, what's been your take? We were with you yesterday. What's been the take today, the takeaways? Not much has changed since yesterday, but a few things I think I missed talking about that yesterday were that, first of all, let's talk about Amazon. Amazon earnings came out, it spooked the market. And I think it's relevant in this context as well because they are number one cloud provider. And I mean, almost all of these technologies on the back of us here, they are related to cloud, right? So it will have some impact on these. We have to analyze that. Like, will it make the open source go faster or slower in lieu of the fact that the cloud growth is slowing, right? So that's one thing. That's put that aside. I've been thinking about the future of Kubernetes. What is the future of Kubernetes? And in that context, I was thinking like, you know, I think when I put a pointer there and think in tangents, like what else is around this thing? So I think CNCF has been riding the success of Kubernetes. That was their number one flagship project, if you will. And it was mature enough to stand on its own. It was Google, it's Google's Borg dubbed as Kubernetes. It's a geneticized version of that, right? So folks who do tech deep down, they know that, right? So I think it's easier to stand with a solid project, but when the newer projects come in, then your metal will get tested at CNCF, right? And CNCF, I mean, they've got over 140 projects right now. So there's definitely much beyond Kubernetes. Yeah, so they, I haven't seen numbers yet. 18 graduated, right? 37 in incubation, and then 81 in sandbox stage. They have three stages, right? So they have a lot to chew on, and the more they take on, the less quality goes into it. Who's putting the money behind it, which vendors are sponsoring like CNCF, like how they are getting funded up? I think it's great. Something I pay attention to as well. Yeah, at least I know you've got some insight. Those are the things I was thinking about today. I got to ask you, what's your take on what Keith said? Are you also seeing the maturation of the enterprise here at coupon? Yes, I am actually, when you say enterprise versus what's the other side? Startups, right? So startups start using open source a lot more earlier or a lot more than enterprises. Enterprises is what they need. Number one thing is that for the production workloads, they want a vendor sporting them. I said that yesterday as well, right? So depending on the size of the enterprise, if you're a big shop, definitely if you have one of the Fortune 500s and you're a tech savvy shop, then you can absorb the open source directly coming from the open source sort of universe, right? Coming to you. But if you are the second tier of enterprise, you want to go to a provider, which is managed services provider, or it can be cloud service provider in this case. Most of the cloud service providers have multiple versions of Kubernetes, for example. I'm not talking about Kubernetes only, but that is one example, right? So at Amazon, you can get five different flavors of Kubernetes, right? Fully managed, half managed, all kind of stuff. So people don't have bandwidth to manage that stuff locally. You have to patch it, you have to roll in the new updates and all that stuff. It's a lot of work for many. So CNCF actually is formed for that reason. Like the charter is to bring the quality to open source. Like in other companies, they have the release process and they're stringent guidelines and QA and all that stuff. So is something ready for production? That's the question when it comes to any software, right? So they do that kind of work and they have these buckets defined at high level, but it needs more work. So one of the things that kind of stood out to me, I have a good friend in the community, Alex Ellis, who does OpenFaz. It's a serverless platform, great platform. Two years ago or in 2019, there was a serverless day. And in serverless day, you had Knative, you had OpenFaz, you had WISP, which is supported by IBM, completely not CNCF platforms. Knative came into the CNCF fold when Google donated the project a few months ago or a couple of years ago. Now all of a sudden there's a Knative day. It's not a serverless day, it's a Knative day. And I asked the CNCF event folks like, what happened to serverless day? I missed having OpenFaz at serverless day. And they came out and said, you know what? Knative got big enough, they came in and I think Red Hat and Google wanted to sponsor a Knative day, so serverless day went away. So I think what I'm interested in over the next couple of years is they're going to be pushed back from the CNC, against the CNCF. Is the CNCF now too big? Is it now the gatekeeper for, do I have to be one of those 147 projects in order enough to get my project noticed? The OpenFaz great project. I don't think Alex has any desire to have his project hosted by CNCF, but it probably deserves shoulder left recognition with that. So I'm pushing to happen to say, okay, if this is open community, this is open source. If CNCF is the place to have the cloud native conversation, what about the projects that's not CNCF like? How do we have that conversation when we don't have the power of a Google or a Linux, et cetera? Or a Linux foundation. What are your thoughts on that? Is CNCF too big? I don't think it's too big. I think it's too small to handle what we are doing in open source, right? So it can become a bottleneck. I think too big in a way that it has power. From that point of view, it has that cloud fuel that people listen to. If it's CNCF project, oh, this must be good. It's like in incubators, like if you are a Y Combinator company, it must be good, you know, it may not be too good. I think there's a bold assumption there though. I mean, I think everyone's just trying to do the best they can. And when we're evaluating projects of very different origin and background, it's incredibly hard. CNCF is a staff of 30 people. They've got 180,000 people that are contributing to these projects and 1,000 maintainers that they're trying to uphold. I think the challenge is actually really great. And to me, I actually look at events as an illustration of what's the culture and the health of an organization. If I were to evaluate CNCF based on that, I'd say we're very healthy right now. I would say that we're in a good spot. There's a lot of momentum. Yeah, I think CNCF is very healthy. I'm appreciative for it being here. I love KubeCon. It's becoming the de facto conference to have this conversation. It has a totally different vibe. It is a totally different vibe. There needs to be a conduit. And truth be told, enterprise buyers, to Sarbjeet's point, with this something that we do absolutely agree on, enterprise buyers, we want someone to pick winners and losers. We do. We don't want a box of Lego dumped on the middle of our table. We want somebody to have sorted that out. So while there may be five or six different service mesh solutions, at least the CNCF, I can go there and say, oh, I'll pick between the three or four that are most popular and it's a place to curate. But I think with that curation comes the other side of it, of how do we, without the big corporate sponsor, how do I get my project pushed up? Right, elevated. Elevated and put onto the show floor. Another way that projects get notices that startups will adopt them, push them. They may not even be, I don't, my CNCF project may not, my product may not even be based on the CNCF product, but the Newstack has a booth, the Ford has a booth. Nothing to do with a individual product, but promoting open source. What happens when you're not sponsored? I got to ask you guys, what do you disagree on? Oh, so what, what do we disagree on? So I'm of the mindset, I can say this, I believe hybrid infrastructure is the future of IT, bar none. If I built my infrastructure, if I built my application in the cloud 10 years ago, and I'm still building that new applications, I have stuff that I built 10 years ago that looks a lot like on-prem. What do I do with it? I can't modernize it because I don't have the developers to do it. I need to stick that somewhere. And where I'm going to stick that at is probably a hybrid infrastructure. So Colo, I'm not going to go back to the data center, but I'm going to pick up something that looks very much like the data center. And I'm saying embrace that, it's the future. And if you're Boeing, and you have, and Boeing is a member of CNCF, that's a whole other topic. If you have AS400s, HPUX, et cetera, stick that stuff in the Colo, build new stuff, but and continue to support OpenStack, et cetera, because that's the future. Hybrid is the future. And Subjee, agree, disagree? Aye? Okay. On hybrid, nobody can deny that. The hybrid is the reality, not the future. It's a reality right now. It's a necessity right now. You can't do without it, right? And okay, hybrid is very relative term. You can be like 10% here, 90% still hybrid, right? So the data center is shrinking, it will keep shrinking, right? Is the data center shrinking? This is one quick engine collection. The cloud is growing by a clip. But there's no data supporting, David Litchom just came out with a report, I think last year that showed that the data center is holding steady, holding steady, not growing, but not shrinking. Who sponsored that study, hold on. So that's the question, right? So more than one million data centers have been closed, I can dig that number through somebody, like some organization, we publish that, maybe they are cloud people only. So when you get these kind of statements, it can be very skewed statements, right? But if you have seen the scene out there, which you have, I know, but I have also seen a lot of data centers walk the floor of 100,000 servers in a data center. I cannot imagine us consuming the infrastructure the way we were going into the future, okay, with one caveat. Actually, I'm not a big fan of broad strokes, like make a blank statement, oh no, data center is dead, or if you are- That's how you get those zesty headlines. Yeah, yeah, I know. I'm all about the zesty headlines. What is taking the ground? Actually, I think that you get more intelligence intelligence from the nuance, right? Small little details, if you will. If you're Goldman Sachs or Bank of America, you have so many data centers and you will still have data centers because performance matters to you, right? Your latency matters for applications, but if you are, even a Fortune 500 company on the lower end and or healthcare vertical, right? That, your situation is different. If you are, it's high growth startup, your situation is different, right? You will be 100% cloud. So, cloud gives you velocity, the pace of change, the pace of experimentation. That, actually, you're buying innovation through cloud. It's proxy for innovation, and that's how I see it. But if you have, if you're stuck with older applications, I totally understand. You need that on-prem or some time. I think the brilliant field, so what we agree is that cloud is the place where innovation happens. At some point, innovation becomes legacy debt and you have, thus, hybrid. You are not going to keep your old applications up to date forever. The math just doesn't add up. And where I differ in opinion is that not everyone needs innovation to keep moving. They need innovation for a period of time and then they need steady state. So, Sarbjeev, we're going to argue about this. I have a comment, too. I love this debate, though. I got to say it's the efficiency and stability also plays an important role. I see exactly what you're talking about. It's great. I have a comment to do that. Let me tell you. Why? Let's hear it. Because if you look at the storage only, I just took storage from a computer and network for a minute. There are three constructs and infrastructure, right? So, storage, early on, there was one tier of storage. You say, pay the same price. Then now there are like five storage tiers, right? What I'm trying to say is the market sets the price. The market will tell you where this whole thing will go. But I know that margins are high in cloud, 20 plus percent. And margin will shrink as we go forward. That means the cloud will become cheaper relative to on-prem. In some cases, it's already cheaper. But even if it's a stable workload, even in that case, we will have a lower tier of service. I mean, you can't argue with me that the cloud versus your data center, they are on the same tier of services. Like cloud is a better product than your data center. Hands down. I love it. We are going to relish in the debates between the two of you. We're mic dry. The energy is great. I love it perspective. It's not like any of us can quite see through the crystal wall that we have very informed opinions, which is super exciting. Lisa, any last thoughts today? Just love, I love the debate as well. And that's part of what being in this community is all about. So sharing, about sharing opinions, expressing opinions. That's how it grows. That's how we innovate. Obviously we need the cloud, but that's how we innovate. That's how we grow. And we've seen that demonstrated the last couple of days and your takes here on theCUBE and on Twitter. Brilliant. I absolutely love it. I'm going to close this out with a really important analysis on the swag of the show. As you know yesterday, we were looking at what is the weirdest swag or most unique swag. We had that bucket hat that took the grand prize. Today we're going to focus on something that's actually quite cool. A lot of the vendors here have really dedicated their swag to being local. To Detroit, very specific in their sourcing. Sonotype here has koozies. They're beautiful. You can't quite feel this flannel, but it's very legit. Hand sound here in Michigan. I can't say that I've been to too many conferences, if any, where there was this kind of commitment to localizing and sourcing swag from around the corner. We also see this with the Intel booth. They've got screen printers out here doing custom hoodies on spot. They're even like appropriately sized. They had local artists do these designs. And if you're like me and you care about what's on your wrist, you're familiar with Shinola. This is one of my favorite swags that's available. There is a contest going on here. Hello. Yeah, so if you are at KubeCon, make sure that you go and check this out. I talked about this on the show. We've had the founder on the show, or the CEO. And yeah, I mean, Shinola is just full of class. Since we are in Detroit as well, one of the fun things is cars. Yes. And StormForge, who are also on the show, is actually giving away an Aston Martin, which is very exciting, not exactly manufactured in Detroit. However, still very cool on the car front. And the 007 version, the best. I know. From the 60s. Love it. It's very cool. Two quick last things. We talk about it a lot on the show. Every company now wants to be a software company on that vein and keeping up with my hat theme. The Home Depot is here because they want everybody to know that they in fact are a technology company, which is very cool. They have over 500,000 employees. You can imagine there's a lot of technology that has to go into keeping that up wild to think about. And then last but not least, very quick rapid fire, best t-shirt contest. If you've ever been to one of these events, there are a ton of t-shirts out there. I rate them on two things, wittiest line and softness. If you combine the two, you'll really be our grand champion for the year. I'm just going to hold these up and set them down for your laughs. Not afraid to commit, which is pretty great. This is another one designed by locals here. Detroit Code City. Oh, love it. This one made me chuckle the most. Kiss my cash. Oh, that's good. These are also really nice and soft, which is fantastic. Also high on the softness category is this Opsara one. I also like their bird logo. These guys are just, you know, just real nice touch. So unfortunately if you have the FOMO, you're not here with us live in Detroit. At least you're going to get a taste of the swag, taste of the stories and some smiles here from those of us on theCUBE. Thank you both so much for being here with us. Lisa, thanks for another fabulous day. My name's Savannah Peterson. Thank you for joining us from Detroit. We're theCUBE and we can't wait to see you tomorrow.