 Welcome to Amsterdam and KubeCon CloudNativeCon 2023. Join John Furrier, Savannah Peterson, Rob Streche, and UPSCOD. As the Kube covers the largest conference on Kubernetes, CloudNative, and open source technologies together with developers, engineers, and IT leaders from around the globe. Live coverage of KubeCon CloudNativeCon 2023 is made possible by the support of Red Hat, the CNCF, and its ecosystem partners. Hello and welcome back to the Kube this afternoon here in Amsterdam from KubeCon and CloudNativeCon. I want to thank you for joining us. We have an exciting 4.30 in the afternoon bringing the cupcake energy type of interview going on. I'm Rob Streche and I'm here with Joe and we have two fantastic Red Haters with us, Christian and Harriet. So nice to meet you and thank you for coming on board. Thank you, thanks. What's been the most exciting thing you guys have seen? I mean, I saw that it was packed. You were saying when we were catching up earlier that when you had the hats, the fedoras, I should say fedoras and not hats, we had one here and it magically disappeared. Somebody must have co-opted it. They didn't have to do that. Yeah, they co-opted it. You have to hold on to them when you get them. I know, I know. The valuable commodity, you said it was completely packed. What are some of the conversations that you're both having with these customers who are coming by? Yeah, everyone wants to talk about get-ups, which is really exciting. I think it's a wonderful technology and I'm so glad that so many people are interested in it. Yeah, yeah, and getting, over the years, getting people coming by the booth, the questions they're getting more and more advanced. So, you know, kind of speaks to a little bit about the evolution of the technology. So it's very, it was very great to be able to talk a little bit more advanced topics now that it's been around for a little bit. So help me out. What is, where was it and where did it go to? Because what makes it more advanced the discussions that you're having with them? Yeah, I think the adoption is growing. So before it was a buzzword, right? It was buzzwordy, right? On purpose, right? It's kind of earwormy, get-ups, like what is that? But as folks are starting to adopt it and are starting to actually putting in production, that the situations and the questions are starting to get a little bit more advanced. And so what's interesting to me is, back in the day, we used to focus on the technology aspect of this. So we would ask questions like, what is get-ups? What products do I use? How do I install them? The questions I've been seeing is, so we've got get-ups now. How do I make my company successful with it? And so instead of focusing on the technical solutions, we're looking at the organizational problems. And that makes me wonder, what are those organizational problems that we solve by using get-ups? Yeah, absolutely. And this is something that our customers are coming to us with all the time. Especially now that, like Christian said, get-ups has been around for a while, it's matured enough that the larger organizations are interested and they're at the point where they feel comfortable adopting it. So they've got problems like, all right, we've got all these, we've got thousands of edge devices, how do we start deploying out to those? And we've got all of these different teams, how do we do multi-tenancy? Like, how do we make our access control work for us with get-ups? That makes a lot of sense. I think we've been hearing it as well. And I think that as Kubernetes, I mean, we're almost 10 years in now. So it's crazy. That overlap back with OpenStack, when I was doing stuff in OpenStack, is kind of a funny time, because now I'm hearing we were on with the Kuvert guys earlier, and how some of that is actually part of the stuff that I did when I was back in OpenStack. So it's fun to see how it is. What I have seen is that the energy here has just been unbelievable. And I think part of that energy has been the movement from DevOps, SRE, all different groups, all highly fragmented to your organizational. Now it's more platform engineering. Are you seeing get-ups as being kind of the foundation to platform engineering and how people are looking at that as well? Yeah, I've seen it as a cornerstone, really. Get-ups is really, I always say it's a cornerstone to a lot of different practices. And only because it's taking advantage of the technology, the cloud-native technology, right? Of Kubernetes and just in general cloud-native. And so it's really the cornerstone for things like platform engineering. It's kind of like on the tool belt of platform engineering. And also things like DevSecOps and just DevOps in general, CICD. It's just been really kind of a foundational thing now more than anything. Yeah, but at the same time, if we're talking about get-ups, I mean, the definition evolves. And that's a natural state of things. But get-ups is not a single thing. Get-ups is not one piece of technology. So help the audience kind of understand what get-ups is and by definition, what kind of technologies do we need to be successful? Yeah, so Red Hat is part of the get-ups working group in the CNCF, right? So Sandbox Project, but it's really a group of SMEs from both our Argo world and the Flux world. And also from folks like from the Spinnaker background and the Carvel people, just kind of like practitioners that are getting together and trying to define what get-ups is and kind of just like, let's bring all our best practices, all the knowledge that we've known and lessons that we learn and let's define what get-ups is. And that's kind of what one of the things that the get-ups working group, which I'm involved with, which I'm a maintainer of, is that, okay, let's define what get-ups is because we don't want to turn into something like cloud, right? Like what does cloud mean? It could mean anything, right? Nowadays, right? So I encourage everyone to go to openget-ups.dev. We have a set of four principles that kind of just a guide you to what like, okay, if I want to do get-ups, what are like kind of the things I need to, kind of the practices that I need to do in order to really take advantage of the get-ups practices. Yeah, I think exactly building off what Joe was saying is the fact that people, like the first day on the keynote, I went in there and I started looking at the readouts from this, from all of the different successes that have been built over in the different, why am I blanking on the name of the things, but different working groups that are going about it. The one thing that was, I guess you could say, very telling to me, and I know with your guys' background and how you just described get-ups and that, I think it was one of these, and this is building off of that, is that having just a one minute, hey, this is what Flux does, this is what does Argo does. I think that helps bring those new people because 58% of the people here are new to this, right? I mean, that was one of the stats that you were saying, if you have 10,000 people and 58%, it's their first coupon, it's crazy. That's wonderful. Yeah, it's fantastic, and I think that brings the new energy and helps build that. So what is Flux for those at home that don't know what Flux does? Yeah, yeah, so Flux is a get-ups controller that continuously syncs your desired state, which is get to your running state, which is in Kubernetes, right? And so Flux does that, Argos CD does that, which is what Red has, a big contributor to Argos CD as well, and really any other get-ups controller is really just like from a baseline, kind of like you were saying, like what's the one sentence? You have a desired state somewhere in Git, usually, and you want that matched into your running state. And so that is like really the ground, you know, like the main idea of get-ups and any tool that employs the get-ups principles. And is it looking for drift and things like that? Yeah, drift detection, drift correction, and you know, things like that. Yeah, and you get the like self-healing right out of the box. So it doesn't just have to be application deployment, which I feel like a lot of people coming to get-ups, they think it's just CD, it's just that you're deploying your applications, but it's not, it's infrastructure as well and configuration management. So you can set your desired state and then you've got your controller that's always polling and checking, like has someone gone into my cluster and made silly mistakes? And it can go and self-heal that for you. So you don't even have to think about it. After you set up your desired state in Git, it's set and forgotten, it fixes it for you. So I wonder how this works in larger organizations and maybe you can speak to that a little bit because if I'm a single developer or a small team, this is not a hard problem to solve. It becomes very, very complex once you start to scale this across the organization. So I wonder what are kind of the best practices that you need to implement as a get-ups, ready organization to be able to scale it across thousands of developers? I think my favorite topic there, it's always going to be access control. So when you're looking at, okay, so we've got all these teams, we're taking our first steps into get-ups. How do we make it go? How do we scale it up? You can't just kind of launch yourself into that. You've got to like take a look at your organization and be like, all right, what teams do I have? What are our business needs? What are our regulatory requirements? Can we just give everyone access to all this stuff? Is that all right? Do we need our teams who deal with billing to be hived off this side and no other development teams can push to those clusters, or you've got to have a really solid understanding of what your business needs and what your threat model is and what your risk tolerance is before you kind of come in and we can't say, this is how you do it because it's so based on everyone's individual organizational needs. Yeah, and you bring up a good point where your risk tolerance is, I think, a very good way of capturing it because it's not a all-or-nothing situation. It's always a gray area, especially in larger organizations. And so giving people the guardrails of, this is where we start. It's guard of the heavily, it's a small team. The freedom is fairly small, but then as you grow, as you adopt, the guardrails come out, basically. Yeah, and I always say to what Harriet, to build off what she was saying, I always say that get-ups almost reveals some of the gaps you have in your organization because you are, when I talk to customers, when I talk to people about get-ups, they're like, well, what do you mean it's automatically synced? And then my first question is like, why does that make you nervous? Because if you have good practices in place, to Harriet's point, those aren't as scary, right? If it's scary to you, then it's more of an organizational people problem, as I like to say. It's like, okay, what's wrong with our structure and how can we best change in order to implement some of these things? Which almost, I don't want to over exaggerate, but it almost goes back to lean practices. Yes. But applied in a very technical sense, down to earth, in a way that actually makes sense to technologists in that space, in the operational space. And again, I think that makes sense because get-ups is not one technology, it is a way to solve your organizational issues. And I think that you're right on the money there. That being said, I think there's a bright future ahead of us with get-ups because I think it makes a ton of sense. It solves a lot of operational problems, but we have this AI thing looming over us. Because we're basically talking about, policy is code, compliance is code. We're talking about drift detection and remediation. That is something we have to do fairly manually right now, defining those policies. What's your take on AI and kind of taking over that manual work? Yeah, I think, well, for me, I think it's a very important step, especially in the sustainability aspect of it. I think AI is going to play a big part in the sustainability with respect to get-ups. Because as I was saying, it's like the cornerstone to a lot of these things. And being able to automatically shift workloads, get-ups is going to play a big part there. Just by, but when you say sustainability, your meaning, sustainability of the infrastructure? Well, I mean, environmental sustainability is what I meant. Yeah, thank you, yeah. Clarify a little bit is, yeah, environmental sustainability. AI being able to say, oh, it's better to run this workload over here because it's better environmentally than over here, right? It's going to be less, cost is always going to be an aspect of it. It's going to be less about cost and more about environmental sustainability. And get-ups is going to be a cornerstone of that because it's going to need to automatically shift those workloads. And as you were saying, a lot of that stuff is manual now and it's AI becoming more and more advanced. We'll be able to do more of these things. And I see it at least from the environmental sustainability. AI and get-ups is going to play a big role there. And so this, you know, this all solves a lot of operational toil. It removes a lot of these manual work. I think that's great. But then we have this, you know, this pool of talented people that have automated away a big part of their job, which kudos, that's the goal of your job. But what's kind of next for this group? And by extension, what's next for get-ups? What's next for Argo specifically? That's an exciting question. Well, like I don't think AI is going to take away, like it's never going to take away everything. There is always stuff that needs human oversight and human inputs. Like who is going to have that business knowledge if not the people who are working there? The AI can't understand that itself. So I think the folks who have automated away a bunch of their jobs, they now have the time to explore more into like, okay, so we've got these business needs, how can we meet them more efficiently and how can we meet them better? Yeah, I think it's interesting that you brought up access control earlier because I think that's one place where actually you need more people involved because of AI and what AI could do and potential leaks or confidentiality or what have you changing that types of stuff. What else do you see on the horizon for this? Is there other stuff that you're looking at and going, hey, these are what I want to, we're here next year and sitting here and you hope that, hey, we've gotten through acts to get there. So for me, at least from the get-off space, I've been working a lot with me specifically with our Ansible team at Red Hat because get-offs was born and grown in the cloud native space. You need almost Kubernetes, it's almost like the default, you have to use it, but a lot of these practices extend beyond just like Kubernetes, right? There's always going to be a need for controlling infrastructure outside of Kubernetes, not just like bare metal deployments but also things like switches and things like firewalls, those kinds of things and Ansible is, at least for Red Hat, it's going to play a big role there and I've been working with our Ansible team too, it was okay, how do we define get-offs in a non-cloud native way, right? How does that look like? And because there's going to be a lot of gaps, there's the immutability aspect of Kubernetes that doesn't exist on bare metal or on a switch or a network equipment or something like that. I think evolving get-offs past cloud native where it's all encompassing, I think for me, that's hopefully next year, you sit me down here, I'll be talking about that because that's something that, I think, get-offs needs to extend just beyond cloud native. That's awesome, I think that's really what I think the people at home are going to hope and they're going to hold you to it, so. Yes, yes, yes, exactly, yes. Let's get there. And a long camera's saying it, so yeah. Exactly, it's on the internet, so it's forever now, immutable. So I want to thank you both for coming here, I want to thank Joe, I want to thank you for co-hosting, this has been fun, it's our first time together. Absolutely. Yeah, so it was a good time, really appreciate it, and I want to thank you all for watching us again on 420 from Amsterdam, really love it, and I want you to have a great rest of your day wherever you are in the world, and thank you for watching V-Cube where, oh, why did I screw that up already? But the leader in high tech coverage, thank you for coming and we'll see you again.