>> From around the globe, it's theCUBE! With coverage of KubeCon and CloudNativeCon Europe 2020, virtual. Brought to you by Red Hat, the Cloud Native Computing Foundation, and ecosystem partners. >> Hi, I'm Stu Miniman and welcome to theCUBE's coverage of KubeCon CloudNativeCon 2020 in Europe. Of course the event this year was supposed to be in the Netherlands, I know I was very much looking forward to going to Amsterdam. This year of course it's going to be virtual, I'm really excited theCUBE's coverage, we've got some great members of the CNCF, we've got a bunch of end users, we've got some good thought leaders, and I'm also bringing a little bit of the Netherlands to help me bring in and start this keynote analysis, happy to welcome back to the program my cohost for the show, Joep Piscaer, who is an industry analyst with TLA. Thank you, Joep, so much for joining us, and we wish we could be with you in person, and check out your beautiful country. >> Absolutely, thanks for having me Stu, and I'm still a little disappointed we cannot eat the (indistinct foreign term) rijsttafel together this year. >> Oh, yeah, can we just have a segment to explain to people the wonder that is the fusion of Indonesian food and the display that you get only in the Netherlands? Rijsttafel, I seriously had checked all over the US and Canada, when I was younger, to find an equivalent, but one of my favorite culinary delights in the world, but we'll have to put a pin in that. You've had some warm weather in the Netherlands recently, and so many of the Europeans take quite a lot of time off in July and August, but we're going to talk about some hardcore tech, KubeCon, a show we love doing, the European show brings good diversity of experiences and customers from across the globe. So, let's start, the keynote, Priyanka Sharma, the new general manager of the CNCF, of course, just some really smart people that come out and talk about a lot of things. And since it's a foundation show, there's some news in there, but it's more about how they're helping corral all of these projects, of course, a theme we've talked about for a while is KubeCon was the big discussion for many years about Kubernetes, still important, and we'll talk about that, but so many different projects and everything from the sandbox, their incubation, through when they become fully, generally available, so, I guess I'll let you start and step back and say when you look at this broad ecosystem, you work with vendors, you've been from the customer side, what's top of mind for you, what's catching your attention? >> So, I guess from a cloud-native perspective, looking at the CNCF, I think you hit the nail on the head. This is not about any individual technology, isn't about just Kubernetes or just Prometheus, or just service mesh. I think the added value of the CNCF, and the way I look at it at least, looking back at my customer perspective, I would've loved to have a organization curate the technology world around me, for me. To help me out with the decisions on a technology perspective that I needed to make to kind of move forward with my IT stack, and with the requirements my customer had, or my organization had, to kind of move that into the next phase. That is where I see the CNCF come in and do their job really well, to help organizations, both on the vendor side as well as on the customer side, take that next step, see around the corner, what's new, what's coming, and also make sure that between different, maybe even competing standards, the right ones surface up and become the de facto standard for organizations to use. >> Yeah, a lot of good thoughts there, Joep, I want to walk through that stack a little bit, but before we do, big statement that Priyanka made, I thought it was a nice umbrella for her keynote, it's a foundation of doers powering end user driven open-source, so as I mentioned, you worked at a service provider, you've done strategies for some other large organizations, what's your thought on the role of how the end users engage with and contribute to open-source? One of the great findings I saw a couple years ago, as you said, it went from open-source being something that people did on the weekend to the sides, to many end users, and of course lots of vendors, have full-time people that their jobs are to contribute and participate in the open-source communities. >> Yeah, I guess that kind of signals a maturity in the market to me, where organizations are investing in open-source because they know they're going to get something out of it. So back in the day, it was not necessarily certain that if you put a lot of effort into an open-source project, for your own gain, for your own purposes, that that would work out, and that with the backing of the CNCF, as well as so many member organizations and end user organizations, I think participating in open-source becomes easier, because there's more of a guarantee that what you put in will kind of circulate, and come out and have value for you, in a different way. Because if you're working on a service mesh, some other organization might be working on Prometheus, or Kubernetes, or another project, and some organizations are now kind of helping each other with the CNCF as the gatekeeper, to move all of those technology stacks forward, instead of everyone doing it for themselves. Maybe even being forced to reinvent the wheel for some of those technology components. >> So let's walk through the stack a little bit, and the layers that are out there, so let's start with Kubernetes, the discussion has been Kubernetes won the container orchestration battles, but whose Kubernetes am I going to use? For a while it was would it be distributions, we've seen every platform basically has at least one Kubernetes option built into it, so doesn't mean you're necessarily using this, before AWS had their own flavor of Kubernetes, there was at least 15 different ways that you could run Kubernetes on top of it, but now they have ECS, they have EKS, even things like Fargate now work with EKS, so interesting innovation and adoption there. But VMware baked Kubernetes into vSphere 7. Red Hat of course, with OpenShift, has thousands of customers and has great momentum, we saw SUSE buy Rancher to help them move along and make sure that they get embedded there. One of the startups you've worked with, Spectro Cloud, helps play into the mix there, so there is no shortage of options, and then from a management standpoint, companies like Microsoft, Google, VMware, Red Hat, all, how do I manage across clusters, because it's not going to just be one Kubernetes that you're going to use, we're expecting that you're going to have multiple options out there, so it sure doesn't sound boring to me yet, or reached full maturity, Joep. What's your take, what advice do you give to people out there when they say "Hey, okay, I'm going to use Kubernetes," I've got hybrid cloud, or I probably have a couple things, how should they be approaching that and thinking about how they engage with Kubernetes? >> So that's a difficult one, because it can go so many different ways, just because, like you said, the market is maturing. Which means, we're kind of back at where we left off virtualization a couple years ago, where we had managers of managers, managing across different data centers, doing the multicloud thing before it was a cloud thing. We have automation doing day two operations, I saw one of the announcements for this week will be a vendor coming out with day two operations automation, to kind of help simplify that stack of Kubernetes in production. And so the best advice I think I have is, don't try to do it all yourself, right, so Kubernetes is still maturing, it is still fairly open, in a sense that you can change everything, which makes it fairly complex to use and configure. So don't try and do that part yourself, necessarily, either use a managed service, which there are a bunch of, Spectro Cloud, for example, as well as Platform9, even the bigger players are now having those platforms. Because in the end, Kubernetes is kind of the foundation of what you're going to do on top of it. Kubernetes itself doesn't have business value in that sense, so spending a lot of time, especially at the beginning of a project, figuring that part out, I don't think makes sense, especially if the risk and the impact of making mistakes is fairly large. Like, make a mistake in a monitoring product, and you'll be able to fix that problem more easily. But make a mistake in a Kubernetes platform, and that's much more difficult, especially because I see organizations build one cluster to rule them all, instead of leveraging what the cloud offers, which is just spin up another cluster. Even spin it up somewhere else, because we can now do the multicloud thing, we can now manage applications across Kubernetes clusters, we can manage many different clusters from a single pane of glass, so there's really no reason anymore to see that Kubernetes thing as something really difficult that you have to do yourself, hence just do it once. Instead, my recommendation would be to look at your processes and figure out, how can I figure out how to have a Kubernetes cluster for everything I do, maybe that's per team, maybe that's per application or per environment, per cloud, and they kind of work from that, because, again, Kubernetes is not the holy grail, it's not the end state, it is a means to an end, to get where we're going with applications, with developing new functionality for customers. >> Well, I think you hit on a really important point, if you look out in the social discussion, sometimes Kubernetes and multicloud get attacked, because when I talk to customers, they shouldn't have a Kubernetes strategy. They have their business strategy, and there are certain things that they're trying to, "How do I make sure everything's secure," and I'm looking at DevSecOps, I need to really have an edge computing strategy because that's going to help my business objectives, and when I look at some of the tools that are going to help and get me there, well, Kubernetes, the service meshes, some of the other tools in the CNCF are going to help me get there, and as you said, I've got managed services, cloud providers, integrators are going to help me build those solutions without me having to spend years to understand how to do that. So yeah, I'd love to hear any interesting projects you're hearing about, edge computing, the security space has gone from super important to even more important if that's possible in 2020. What are you hearing? >> Yeah, so the most interesting part for me is definitely the DevSecOps movement, where we're basically not even allowed to call it DevOps anymore. Security has finally gained a foothold, they're finally able to shift lift the security practices into the realm of developers, simplifying it in a way, and automating it in a way that, it's no longer a trivial task to integrate security. And there's a lot of companies supporting that, even from a Kubernetes perspective, integrating with Kubernetes or integrating with networking products on top of Kubernetes. And I think we finally have reached a moment in time where security is no longer something that we really need to think about. Again, because CNCF is kind of helping us select the right projects, helping us in the right direction, so that making choices in the security realm becomes easier, and becomes a no-brainer for teams, special security teams, as well as the application development teams, to integrate security. >> Well, Joep, I'm glad to hear we've solved security, we can all go home now. That's awesome. But no, in all seriousness, such an important piece, lots of companies spending time on there, and it does feel that we are starting to get the process and organization around, so that we can attack these challenges a little bit more head-on. How 'about service mesh, it's one of those things that's been a little bit contentious the last couple of years, of course ahead of the show, Google is not donating Istio to the foundation, instead, the trademark's open. I'm going to have an interview with Liz Rice to dig into that piece, in the chess moves, Microsoft is now putting out a service mesh, so as Corey Quinn says, the plural of service mesh must be service meeshes, so, it feels like Mr. Meeseeks, for any Rick and Morty fans, we just keep pressing the button and more of them appear, which may cause us more trouble, but, what's your take, do you have a service mesh coming out, Kelsey Hightower had a fun little thing on Twitter about it, what's the state of the state? >> Yeah, so I won't be publishing a service mesh, maybe I'll try and rickroll someone, but we'll see what happens. But service meshes are, they're still a hot topic, it's still one of the spaces where most discussion is kind of geared towards. There is yet to form a single standard, there is yet a single block of companies creating a front to solve that service mesh issue, and I think that's because in the end, service meshes are, from a complexity perspective, they're not mature enough to be able to commoditize into a standard. I think we still need a little while, and maybe ask me this question next year again, and we'll see what happens. But we'll still need a little while to kind of let this market shift and let this market innovate, because I don't think we've reached the end state with service meshes. Also kind of gauging from customer interest and actual production implementations, I don't think this has trickled down from the largest companies that have the most requirements into the smaller companies, the smaller markets, which is something that we do usually see, now Kubernetes is definitely doing that. So in terms of service meshes, I don't think the innovation has reached that endpoint yet, and I think we'll still need a little while, which will mean for the upcoming period, that we'll kind of see this head to head from different companies, trying to gain a foothold, trying to lead a market, introduce their own products. And I think that's okay, and I think the CNCF will continue to kind of curate that experience, up to a point where maybe somewhere in the future we will have a noncompeting standard to finally have something that's commoditized and easy to implement. >> Yeah, it's an interesting piece, one of the things I've always enjoyed when I go to the show is just wander, and the things you bump into are like "Oh my gosh, wow, look at all of these cool little projects." I don't think we are going to stop that Cambrian explosion of innovation and ideas. When you go walk around there's usually over 200 vendors there, and a lot of them are opensource projects. I would say many of them, when you have a discussion with them, I'm not sure that there's necessarily a business behind that project, and that's where you also see maturity in spaces. A year or so ago, in the observability space, open tracing helped pull together a couple of pieces. Storage is starting to mature. Doesn't mean we're going to get down to one standard, there's still a couple of storage engines out there, I have some really good discussions this week to go into that, but it goes from, "Boy, storage is a mess," to "Oh, okay, we have a couple of uses," and just like storage in the data center, there's not a box or a protocol to do anything, it's what's your use case, what performance, what clouds, what environments are you living on, and therefore you can do that. So it's good to see lots of new things added, but then they mature out and they consolidate, and as you said, the CNCF is help giving those roadmaps, those maps, the landscapes, which boy, if you go online, they have some really good tools. Go to CNCF, the website, and you can look through, Cheryl Hung put one, I'm trying to remember which, it's basically a bullseye of the ones that, here's the one that's fully baked, and here's the ones that are making its way through, and the customer feedback, and they're going to do more of those to help give guidance, because no one solution is going to fit everybody's needs, and you have these spectrums of offerings. Wild card for you, are there any interesting projects out there, new things that you're hearing about, what areas should people be poking around that might not be the top level big things? >> So, I guess for me, that's really personal because I'm still kind of an infrastructure geek in that sense. So one of the things that really surprised me was a more traditional vendor, Zerto in this case, with a fantastic solution, finally, they're doing data protection for Kubernetes. And my recommendation would be to look at companies like Zerto in the data protection space, finally making that move into containers, because even though we've completed the discussion, stateful versus stateless, there's still a lot to be said for thinking about data protection, if you're going to go all-in into containers and into Kubernetes, so that was one that really provoked my thoughts, I really was interested in seeing, "Okay, what's Zerto doing in this list of CNCF members?" And for that matter, I think other vendors like VMware, like Red Hat, like other companies that are moving into this space, with a regained trust in their solutions, is something that I think is really interesting, and absolutely worth exploring during the event, to see what those more traditional companies, to use the term, are doing to innovate with their solutions, and kind of helping the CNCF and the cloud data world, become more enterprise-ready, and that's kind of the point I'm trying to make, where for the longest time, we've had this cloud-native versus traditional, but I always thought of it like cloud-native versus enterprise-ready, or proven technology. This is kind of for the developers doing a new thing, this is for the IT operations teams, and we're kind of seeing those two groups, at least from a technology perspective, being fused into one new blood group, making their way forward and innovating with those technologies. So, I think it's interesting to look at the existing vendors and the CNCF members to see where they're innovating. >> Well, Joep, you connected a dotted line between the cloud-native insights program that I've been doing, you were actually my first guest on that. We've got a couple of months worth of episodes out there, and it is closing that gap between what the developers are doing and what the enterprise was, so absolutely, there's architectural pieces, Joep, like you, I'm an infrastructure geek, so I come from those pieces, and there was that gap between, I'm going to use VMs, and now I'm using containers, and I'm looking at things like serverless too, how do we built applications, and is it that bottom-up versus top-down, and what a company's needs, they need to be able to react fast, they need to be able to change along the way, they need to be able to take advantage of the innovation that ecosystems like this have, so, I love the emphasis CNCF has, making sure that the end users are going to have a strong voice, because as you said, the big companies have come in, not just VMware and Red Hat, but, IBM and Dell are behind those two companies, and HPE, Cisco, many others out there that the behemoths out there, not to mention of course the big hyperscale clouds that helped start this, we wouldn't have a lot of this without Google kicking off with Kubernetes, AWS front and center, and an active participant here, and if you talk to the customers, they're all leveraging it, and of course Microsoft, so it is a robust, big ecosystem, Joep, thank you so much for helping us dig into it, definitely hope we can have events back in the Netherlands in the near future, and great to see you as always. >> Thanks for having me. >> All right, stay tuned, we have, as I said, full spectrum of interviews from theCUBE, they'll be broadcasting during the three days, and of course go to theCUBE.net to catch all of what we've done this year at the show, as well as all the back history. Feel free to reach out to me, I'm @Stu on Twitter, and thank you, as always, for watching theCUBE. (calm music)