KubeCon + CloudNativeCon NA 2025 | Dan Ciruli, Nutanix & Sudeep Goswami, Traefik Labs

Clips
More from KubeCon + CloudNativeCon NA 2025

Dan Ciruli

Vice President and General Manager, Cloud Native

Nutanix

Sudeep Goswami

CEO

Traefik Labs

play_circle_outline Infrastructure consolidation is reshaping the cloud native landscape and workload management.

play_circle_outline Streamlining Management: Unified Application Intelligence for VMs, Containers, and Serverless Environments to Reduce Infrastructure Fragmentation

play_circle_outline Complexity and skill gap in app development and infrastructure highlighted as major challenges.

play_circle_outline Unlocking On-Premises Serverless Solutions: The Rise of Cloud Alternatives and the Power of Strategic Partnerships for Organizational Sovereignty

Info
Transcript

Dan Ciruli, Nutanix & Sudeep Goswami, Traefik Labs

Dan Ciruli

Vice President and General Manager, Cloud Native Nutanix

Sudeep Goswami

CEO Traefik Labs

In this KubeCon + CloudNativeCon North America 2025 segment, Nutanix general manager of cloud native Dan Ciruli and Traefik Labs chief executive officer Sudeep Goswami join theCUBE’s Rob Strechay and theCUBE Research’s Paul Nashawaty to unpack how infrastructure is evolving beyond “just Kubernetes.” They explain how enterprises are moving from experimentation to operationalization, with VMs, containers and serverless now coexisting across on-prem, hybrid and multi-cloud environments. The discussion highlights the Nutanix–Traefik Labs partnership, including Nu... Read more

explore Keep Exploring

What trends and changes are currently occurring in the infrastructure business, particularly in relation to Kubernetes and CloudNativeCon? add

What solution was discussed for addressing the fragmentation in application management? add

What are the challenges faced in managing virtual machines and containers within enterprise environments? add

What options do enterprises have for deploying serverless solutions besides public cloud services like AWS Lambda and Google Cloud Run? add

bolt Powered by CUBE AI

Dan Ciruli, Nutanix & Sudeep Goswami, Traefik Labs

search

>> Hello, and we're winding down on day two here live from KubeCon + CloudNativeCon, North America 2025 from Warmlanta, and we're having a lot more fun. We're continuing to talk through how infrastructure is really changing to support these new workloads and really understanding what it means to be cloud native. Really excited to be joined by two different companies who are here to help us understand how they're working together and how they're working through to help organizations. First, we have Dan Ciruli, who's the GM and cloud native at Nutanix. Welcome on board.

Dan Ciruli

>> Thanks for having me.

Sudeep Goswami

>> And Sudeep Goswami.

Sudeep Goswami

>> You got it.

Sudeep Goswami

>> Almost screwed it up. CEO of Traefik Labs. Thanks for coming on board. And of course, Paul Nashawaty back for another segment. Welcome on board, our analyst.

Paul Nashawaty

>> Thanks, Rob, to be here.

Dan Ciruli

>> And practice lead.

Sudeep Goswami

>> Good to see you. Guys, I think when you look at what's going on, there's really a big consolidation going on in the infrastructure business. I mean, it's funny, we were talking about this earlier in the week that it's sexy to be an infrastructure again, which again, people are having their moments. Yeah, it's different, but sexy again. I mean, all the way down to the hardware for that matter. I mean, again, so when you look at what's going on with VMs and containers and the fragmentation that's existed, where do you see this going into the future? Why don't we start with you, Dan?

Dan Ciruli

>> So I think I've been around coming to KubeCons for close to a decade now. And in the beginning, KubeCon was hyper-focused, really hyper-focused on Kubernetes. I'm glad you call it CloudNativeCon, because it's more than just Kubernetes, but it really was hyper-focused on the cutting edge of how do we get this stuff into production. And what's changed now 10 years in is that now, we've got a lot of stuff in production, and it is starting to become the job of the infrastructure team to say, "Okay, how do we operate this at scale in a big enterprise? How do we think about the fact that most of our stuff is still in VMs? We do have this new stuff that's in containers growing every day, AI is happening. But don't think of the containers as this isolated workload. How do we view it in the whole landscape of the enterprise?" And that is a change from what we saw here several years ago.

Sudeep Goswami

>> Yeah, I would agree. I mean, how do you see it from the traffic perspective?

Sudeep Goswami

>> Yeah, like Dan said, it is about operationalizing all of these different technologies. So the learning period is over, the experimentation is over. Now, people are thinking seriously about optimizing workloads for different tasks, and therefore this coexistence of VMs and containers are here. But then we're also seeing serverless come in. You have to think about, and this is on-prem we're talking about, not just cloud, but everywhere. You got to think about these co-existence. And this requires a really, you kind of take a step back and you think about what is the right architectural decisions that you have to make to enable this at velocity now.

Paul Nashawaty

>> Yeah. So Dan, Sudeep, great having you on here today. When we look at what you just talked about, I know you've both been around this ecosystem for quite some time, the co-existence of VMs and containers. You talked about how this infrastructure is kind of evolving and changing, but it also creates fragmentation in the market and in the infrastructure. How are you dealing with that? And I guess Sudeep, let's start with you on that.

Sudeep Goswami

>> Yeah. We have done a great integration with Nutanix. The way we kind of look at this is the way to solve the fragmentation is through abstraction, and you create the right abstraction layer. And what we launched this week is this concept of a unified application intelligence layer that operates at a higher level closer to the applications. So as the traffic is coming in to go to different places, the application layer logic decides whether that's going to a VM, whether that's going to a container or to a serverless. So you get away from this isolated management of you control the VMs one way, you expose the VMs a different way than containers and serverless. You have to expose, secure, manage and govern them through a consistent policy layer, consistent lens. And that's what we're bringing and we're excited to partner with Nutanix on this effort.

Paul Nashawaty

>> Yeah, I mean it makes a lot of sense. I mean, the two things that we see in our research, in our app dev research is complexity and skill gap issues that are major challenges with overcoming these things. So Dan, when we look at this, what's your response to this from your perspective?

Dan Ciruli

>> So one of the silos that we see happening frequently is that people are running their, especially in enterprises, running their VMs in one place made literally some set of clusters that are virtualized and running their containers in another place. Running Kubernetes on bare metal. And it leads to big issues, how you direct the traffic, how you do the networking between them, how you do the RBAC between them? Because you have entirely different workflows, entirely different ways to handle all of that. And so what we are doing is we're helping customers to say, "Hey, we can help you run everything that you need to run, your virtualized workloads, your containerized workloads along with your storage on one stack." Traffic has done a fantastic job of saying, "We will be able to ..." Traffic is coming in to an application, and I love the way you think of it in an app-centric way. We will handle that. You're not going to have to set up separate networking, separate ingress for that VM cluster versus that Knative cluster maybe versus that Kubernetes cluster.

Paul Nashawaty

>> Hence reducing complexity.

Dan Ciruli

>> So breaking down the silos is then of course ... Yeah, reducing the complexity. You're not deploying a lot of different pieces.

Paul Nashawaty

>> Yeah, interesting.

Sudeep Goswami

>> Yeah, I mean, I think again, Nutanix has always been about reducing complexity. I mean, basically, starting hyper-converged back in the day and now going to cloud and everywhere else. How do you see this from a perspective? Because there are a lot of different services. You mentioned serverless for instance, and bringing serverless into that fold and saying, "Hey, we can help make it more accessible in private deployments as well as those public cloud deployments." How are what you guys are working on together really play in that?

Sudeep Goswami

>> Yeah. So the foundation was already set what Nutanix did by introducing NKP, which sets the foundation of Kubernetes-driven deployments. And what we have done is created or opened up the routing layer to deploy serverless inside of NKP. And now for the first time, I believe for enterprises, they have an option other than going to the public cloud for deploying serverless. The other two options would've been kind of go to AWS Lambda or go to Google Cloud Run. But now, as hybrid cloud and on-prem cloud is becoming more common patterns, customers want that flexibility, not just VMs, not just containers, but serverless. So we bring the application layer, Nutanix brings the rest of the infrastructure, and that's why it's a beautiful partnership.

Dan Ciruli

>> We were happy to see Knative graduate this year. I've been watching the Knative project. I was at Google when it started working on Istio. They were a sister project and been watching it, following it for a long time. And I think that there has been this gap on-prem that people will, if they want that serverless experience, they've had to go to the public cloud. There are strong reasons to want to do that. Scale to zero, event-based routing is useful even on-prem. And so we've supported Knative for quite a while. Again, now that it's graduated, now that we've got tools like what traffic has built to help route, I think it will really start to be the era of serverless on-prem.

Paul Nashawaty

>> So this is also really emphasizing the deploy anywhere kind of message, right? We're really talking about whether you're deploying in a public cloud, private cloud, an air-gapped environment, whatever it may be, you have the ability to do this deployment anywhere. Is that accurate?

Dan Ciruli

>> Yeah, definitely. We have always had customers who deploy us in all of those things. Sometimes because we can deploy in all of those areas. Almost everybody today is whether they want to be or not, they're hybrid multi-cloud, right? At least hybrid. At least the data centers haven't dried up. Very few customers have moved a hundred percent of the cloud. I know customers who are moving stuff back and forth. The cloud is fantastic, but the ability to deploy this whole situation, traffic included on those air-gapped environments is really useful.

Paul Nashawaty

>> I mean, not too long ago, it's probably a couple of years ago, we did a presentation together on air-gapped environments. That was well received that obviously continues to grow as a need in the market. So from a traffic perspective, how does that deploy anywhere model work?

Sudeep Goswami

>> Yeah, it's extremely seamless. It's the same binary that you deploy whether on a public cloud, private cloud or an air gap. And from a installation perspective, it's a Helm chart installation process. And if you're doing an offline air gap, installation is one extra line, a parameter that you set to true and that's it. Everything else works exactly the same. The same operating model, the same configuration set, the same way you would kind of do day two ops, whether you're observing the traffic through OpenTelemetry, nothing changes. You just have it on-prem, and that's the beautiful part about this.

Sudeep Goswami

>> Yeah. I mean to everybody's point, I think even in our data, we see that people may have been multi-cloud by accident probably 5, 10 years ago, but now they're multi-cloud on purpose and with a purpose. And I think to that point, they want to have all the features that they have everywhere because cloud is not a place anymore. It's an operating model. And I think that really gets at it. When you look at this and you look at why companies would go down this path, that app-centric view. I mean, again, I'm an old networking guy. I started out and load balancing and when all of that was in ethernet and token ring. And even before that with that name frame.

Dan Ciruli

>> I could go on back.

Sudeep Goswami

>> I know, I could go way back, and then routing and then we went to these switched routers and things like that. So to your point about routing at that kind of Layer 7 type of atmosphere, what is really the leverage that organizations are going to get out of this? And what control and architectural control can they get out of this? One of the things also and kind of ties into what you guys were talking about with air gap is also more generally sovereignty as well. So how do those three things really play with your approach to this?

Dan Ciruli

>> Why don't you take on the Layer 7 stuff? Hashtag life at Layer 7.

Sudeep Goswami

>> Yeah. I think the way I look at this is in three different ways. You get a sense of operational leverage. I mean, that's the first thing. If you are an enterprise and you want to have this hybrid, multi-cloud, offline strategy, you need to have an architectural leverage point. And by decoupling the application layer from the underlying infrastructure, that's the first step in getting that operational leverage. Because once you have the operational leverage, then you decide when traffic is coming in, how you authenticate, how you route. Your infrastructure is not dictating that. The vendors that you have underneath are not dictating how you do that. You as a company get to define that as a policy. And then you go from there. Once you have that, then you get to decide where you route. You're routing to a cloud. You're routing, let's say you don't want that cloud today, you want to leave that cloud or you want to migrate away to something else, you can just shift the routing layer. So that's the leverage point that you want. And then on top of that, the sovereignty discussion is becoming more hot these days. So one of the defining things about being sovereign is you have the option to leave. If you cannot leave a particular environment, it doesn't matter what you're doing, you're just not sovereign.

Sudeep Goswami

>> Right. That's the whole Dora thing as well that really started in France. And you have to be able to move between clouds back on-prem, wherever if there's a failure out of a major cloud or something like that.

Sudeep Goswami

>> So portability key is a key benefit that comes out of having this operational leverage.

Paul Nashawaty

>> Oh, we're actually seeing that in our app dev research. We're seeing that application portability, 20% of respondents indicated that it's critical. 67% of respondents say that it's very important for their environment. So that's a very big point.

Dan Ciruli

>> It is all my favorite thing about doing something with Paul is he's got all these numbers at the top of his head that you never get through a conversation without. They just start coming out. I think it's fantastic. I want to jump on his L7 stuff because when I was at Google, my team built and ran Google's API infrastructure and service mesh and all we thought, we thought about Layer 7 all the time. And the reason I like it is that when you think about your application health, you think about your application as how it's performing. And you think at Layer 7, how long are requests taking, how many of those requests are failing? Are we doing a beta test? We need to route 10% of our requests. All of that is stuff that has to happen at Layer 7. It has to be application aware. It's going to be different for different applications. And so the freedom it gives you to focus on the app rather than on the infrastructure, the networking really gets you results in a better customer experience. And I don't know how many people make that leap, but that's what I love about what these guys do.

Paul Nashawaty

>> I do want to touch quickly on this because I think this is a definite segue into one of the last points I want to make. But when we look at our research, we see that 68% of organizations are looking at employing infrastructure as code as part of their platform engineering and DevOps teams. But this is also employing a modern CI/CD pipeline as well as the GitOps practices. Do you want to touch on that a little bit? Because I think that's a big part of this as well.

Dan Ciruli

>> I am a huge believer in declarative APIs. I think one of the reasons ... There's a lot of reasons that Kubernetes won. We can do a half an hour on why Kubernetes won. Part of it would be on declarative APIs. We are big believers in that. And our customers are deploying not just their applications and the configuration for their applications, but the infrastructure itself via declarative APIs. We are big believers in the Kubernetes cluster API. And one of the reasons is that it allows you to declaratively state, "Hey, here's how I want my cluster to be." And that API, it's a Kubernetes API. It's the same API, whether you're deploying on-prem, deploying in Amazon, deploying in Google, that's a really big deal. Many companies write tens of thousands of lines of Terraform that is different every environment they deploy in. And you move to those declarative APIs like I need six nodes, I need them to shape like this. And it works wherever you go. That's a huge, huge benefit. So I think that eventually as an industry, everything will move to declarative. You're already there and I think everything is going to get there.

Sudeep Goswami

>> I agree. I think that's a fantastic place for us to leave it because I think GitOps and all of that and infrastructure as code is a big theme. And we've been seeing it here that to get to that is going to help platform engineering really wrestle this to the ground and have a better, get less toil out of it as well. So hey, thanks for coming on board. Really appreciate it.

Paul Nashawaty

>> Great to be here.

Sudeep Goswami

>> I can't wait to do it again.

Dan Ciruli

>> Thanks so much for having us.

Sudeep Goswami

>> Thank you.

Sudeep Goswami

>> And thank you, Paul, as always, Mr. Numbers. Really good to see you too.

Paul Nashawaty

>> Of course, Rob.

Sudeep Goswami

>> And thank you for watching this episode of KubeCon + CloudNativeCon. And I also agree, they should be swapped, CloudNativeCon + KubeCon, but I'm not part of the Linux Foundation, but here we are. KubeCon + CloudNativeCon North America 2025 from Warmlanta. We'll see you right back here on theCUBE, the leader in tech news and analysis. Stay tuned.