Cloud Native ComputingData Protection/HA/DRDevelopersDevOpsFeaturedKubernetesLet's TalkVideo

Kasten Helps Simplify The Complexities Of Data Protection For Kubernetes | Tom Leyden


In this episode of Let’s Talk recorded at KubeCon + CloudNativeCon EU, Swapnil Bhartiya sat down with Tom Leyden, VP of Marketing at Kasten by Veeam, to talk about Kasten’s presence at the event. Although the pandemic has thrown the world into turmoil, the tech sector has continued to see strong growth. It has also acted as a catalyst for the increasing adoption of open source technologies such as Kubernetes. Leyden discusses the increasing interest in people wanting to learn about Kubernetes and how Kasten is helping them in their journey.

Leyden goes into detail about the difference in data protection in the cloud compared to traditional IT, and some of the misconceptions their customers can have. He explains the evolution they are seeing with awareness of data protection and the cultural shift they are seeing in organizations where the responsibility lies.

Key highlights of this video interview are:

  • Leyden explains the level of interest they are seeing in people wanting to learn about Kubernetes, which led them to the launch of their hands-on learning experience
  • Leyden discusses the difference in data protection in the cloud compared to the traditional IT world. He explains that customers can sometimes be confused with whether they are running stateless or stateful applications, and do not always realize that they need to backup the data.
  • Some people presume that Kubernetes has data protection built-in. Leyden explains some of the misconceptions people can have about their data and protecting the entire architecture. He explains how Kubernetes-native backup solutions like Kasten K10 by Veeam can help.
  • Leyden goes into detail about how much awareness there is about data protection. He details some of the trends they are seeing with backup and explains that when building new applications, backup is not always seen as a priority, instead it is on getting it running. He explains the evolution they are seeing with Kubernetes’ awareness though.
  • While the shift-left movement has made security top priority, Leyden feels there will be a similar shift seen with data protection. Leyden explains that while we are seeing a redefining of personas, backup concerns and responsibilities are being embraced by the people designing, building, and running the application.
  • Leyden explains some of the verticals that are seeing strong adoption of Kubernetes. He goes into detail on the success of one of their use cases and how Kasten helped them shift part of their infrastructure to Kubernetes.

Connect with Tom Leyden (LinkedIn, Twitter)

The summary of the show is written by Emily Nicholls.


Here is the automated and unedited transcript of the recording. Please note that the transcript has not been edited or reviewed. 

Swapnil Bhartiya: Hi, this is your host, Swapnil Bhartiya, and welcome to another episode of TFiR: Let’s Talk here Kubecon and CloudNativeCon in Valencia, Spain. And today we have with us Tom Leyden, VP of Marketing at Kasten by Veeam. Tom, it’s great to have you on the show.

Tom Leyden: Thank you for having me.

Swapnil Bhartiya: We have been covering Kasten for a very long time before the whole acquisition happened. But the market has also changed and things are changing when it comes to security, data protection, high availability, and all those things. And that’s what we are going to talk about. But before, of course we are here at Kubecon. It’s day two, day three, I have no idea which day we are in. Only I know is that tomorrow will be the last day when we wrap and leave. So share with us what kind of energy, what has been your experience so far?

Tom Leyden: So let’s call it Thursday. Then we don’t have to be confused with day one or two. And, and typically Kubecon, the show starts a little bit later in the week. So we’re officially on day two of the show right now, but there’s been a lot of meetings. We also run CNDM day with Kasten our community event.

And that was before the actual show started. I think they call it day zero to make things even more complicated. So my experience is that it’s finally growing again. We were all super excited to go to Kubecon in Los Angeles, and we were all very happy to meet people again. But quite frankly, to use it with maybe not such a nice word, we called it a vendor fest.

It was vendors amongst each other. Customers were not showing up all that much, but everybody was excited after lockdowns and whatnot to back be face to face. What we’re experiencing right now is customers are on the floor and that’s awesome. We are seeing customers come up to us, asking us for solutions to their Kubernetes challenges.

So that’s really great. We’ve been having great conversations, our sales team, we have a huge sales team here. They’re super excited, getting great feedback about the level of conversations that are getting. So I think gradually we’ll be getting back to businesses as usual. And, that’s very good for the whole industry I think.

Swapnil Bhartiya: Interestingly, I think even during pandemic, the businesses they grew because the adoption of cloud did not shrink. It’s only that in-person interaction and that also leads to a lot of business as well.

Tom Leyden: Absolutely. I mean, Kasten got acquired during the pandemic. I mean, kudos to the guys to make that happen without being able to meet anyone face to face or practically not. So it is actually surprising how, how well the market kept growing specifically tech market and maybe also thanks to the pandemic, right? I mean, think of companies like Zoom that did amazing back in those days, but we’re still looking at an emerging market, right? So the market was growing during pandemic. Obviously Kubernetes has been growing for the past four, five years gradually.

I always like to do a Google trends search and compare Kubernetes to VMware back in the years. And when you do that, you should do it back home when you’re watching this video. When you do that search and you go back in time, as long as you can, when the service started, you will clearly see that the VMware adoption curve, it’s really the traditional shape.

And that is just based on how much search there was online. And then if you compare that to the one from Kubernetes, you clearly see that a graph is going up, but it’s no way up there, like where we expect it to get, right? So there is growth. There is good growth. There is, I mean, all indicators are there to say the market, I mean, people are adopting it. But the way how I see it is, and I always like to compare how I use technology, because I’m just a marketing guy. But back in the days when VMware was coming up, I was switching from Windows to Mac. I needed to run something on Windows, in my Mac. So I was like, okay, let’s run it in a VM. I downloaded, I think it was virtual box back then, downloaded it, installed Windows.

And I was running. I didn’t even have to look at a YouTube or anything. I was just all straightforward. That’s how easy it was to adopt VMs. Learning Kubernetes is a different game, right? To the point where I’m like, maybe it’s not for marketing people. You need to be a little bit more advanced than where I am at. And, but that is also for engineers and DevOps people and everybody who sees the vision of, okay, this is where the market’s going. That is where the more innovative and scalable applications are being built. So it’s just taken a little bit longer than the shift from a physical machine to a virtual machine, which creates a lot of opportunities because when, after my first Kubecon, which was a virtual one a year ago, my team and I, we got together and we were like, okay, what did we learn?

Right? And we’re like, okay, a lot of people are interested in what we’re doing. A lot of people wanted to talk to us, but a lot of people are still learning what Kubernetes is. That was one thing. The other thing that we saw is to get used to our product we have a free version because the that’s standard in the industry right now. But we also have a lab-based version where people do not have to build their own infrastructure. We do it for them and it’s a controlled environment. And what we learned was that it’s still a lot easier to make people use the lab, then download the free version to run it on their own cluster.

Why is that? Because they don’t have their clusters yet. They’re still playing with it. And that is when we decided, okay, a lot of people, the market is growing. A lot of people learning, people don’t have their own clusters yet. Let’s build a hands-on learning experience for those folks. And then, in just a few months, my amazing team, they put together what’s called We launched it at the next Kubecon, which was a previous one, the one in LA. And I think today we are beyond 10,000 users.

I was actually pinged this morning by my lead from the digital team. And she was like, “My God, the labs are going crazy. So many people are taking the labs. It’s crazy. Should we do something about it?” I was like, “No, keep it going.” It’s costing us some money, but really helping the community to learn Kubernetes eventually will pay back because once everybody’s on the level, they will start building their applications. And then they will realize that they need backup solutions that were built for in a Kubernetes native way. And that’s what we at Kasten do.

Swapnil Bhartiya: Yeah and that is a perfect segue to my next question, which is you talking about the adoption of community is early days, communities was more stateless. Now it is stateful. And that’s where we start talking about data, data production. So first of all, can you tell me now how different is data production in the communities clouded you versus the traditional IT world? Because the way we deploy and we do things is quite different or it could be similar as well from traditional IT, which also means that there are a lot of vendors who come from a legacy world. Yeah. And then there are a lot of vendors who were born in cloud IT world.

Tom Leyden: Yeah. It’s a very good question. And maybe first to the stateless, stateful thing, when I first joined Kasten I was talking to some guy to get educated what this market was about. And, and he told me doesn’t make sense what, what you guys are doing because Kubernetes is not meant for, for stateful applications. You’re meant to be able to lose the data. And I got a little bit worried, talked to the founders of Kasten and no, he’s got it wrong because Kubernetes is being used and will more and more be used by people with stateful applications. Actually what I’m hearing from our sales people, especially pre-sales people is that a lot of companies don’t even realize that they’re not stateless. That they’re often like, oh, I don’t need backup because I’m stateless. And then once they start asking questions, yeah but you run PostgreSQL or you run Cassandra or you run this, right?

That’s not, oh, okay. So I guess I’m not stateless. And then they realize that they actually need to back up the data that is in those databases. Even if that is just data to run the application, they still need to back it up because they risk losing it. Right?

And then the other misperception and that probably gets in, into the second part of your question is a lot of folks think we’re using Kubernetes, so Kubernetes provides data protection and everything. We’re cool, we’re covered. But you’re not, it’s really, you need to make sure that your entire architecture, including your application, all of the comp all of the containers, all of the secrets, the user management and your application data, all of that needs to be protected. And I’m not the most technical guys I told you.

But in the old days it was fairly simple. You had a few servers that you needed to back up. There was a conversation between agent or agent-less and Veeam can say a lot of smart things about that because they got that right back in the days. But now with containers, it’s really, you have pretty complex, because of the scalability that we need to see these days, you have pretty complex architectures, very often with Kubernetes applications that need to be backed up A to Z, including databases, including secrets, including all of that application data. And that is where a Kubernetes native backup solution like Kasten K10 by Veeam comes in really handy because what the solution does is it discovers your application. You don’t have to manually say include this and that. And that no, our solution will say, this is what your application consists of.

We’ll back it up. And then it’s just up to you to decide what kind of policies you want, whether you just want to be able to restore a backup, whether you want it to be offsite, whether you want it, whether you want to use the functionality to actually move from sandbox into production or from one cloud to another. So the essence is that Kasten K10 is really data management solution. But because people kind of get a little bit, they get the deer in the headlight a little bit. When you talk about data management for Kubernetes, we try to keep it simple and focus on the key use case where most of our customers are using is for in these early days. And that is just the backup. This is your application, we store a backup, a snapshot. And, and then if you were to need it, we can restore it. But then of course there’s all the other use cases, mobility, migration, disaster recovery. So we support all of that as well.

Swapnil Bhartiya: Right. And you, once again, as just said when people move to Kubernetes, they think just go a bit on top of that when people move to cloud, they think that cloud will solve all their problem. You move to cloud. Yes, cloud offer is scalability allows the flexibility access, but it doesn’t solve all the problem you still have to. And people just think, no, I moved to cloud and it’s all taken care of. And you said, oh, I’m a stateless. I don’t have to worry about then. You’re like, no, you do have exactly. So from the awareness point of view, from, as you also mentioned the lab, the kind of interest that is going, which also means that there is an awareness about data prediction in the cloud because the virtual loads are maturing.

People are running protection. So open source solve day one problem, day two problem is where you have to actually manage, update, maintain and data production comes after that. Because you don’t know, when something goes wrong, it’s not about if it was when, and it’s not, when it went wrong. When you got to know something went wrong because it already happened. So first of all, talk about how much awareness you have seen about which will all school help. And then we’ll also talk about what is Kasten doing? Because you are seeing the trends are changing the adoption. You did mention that you’re looking only at a specific use case, but I’m pretty sure in six months from now, one year from now, you folks will evolve as well there.

Tom Leyden: Yeah, absolutely. And I like to compare selling and backup solution or data management solution, but with a focus on backup to selling insurance policies. Like when you build a house, you’re not going to get an insurance in case it burns down before you start building it, right? Your first concern is I need to buy land. And then I need to find a contractor to actually build the house. And I need bricks and doors and windows and everything. And it’s when you’re about ready to move in that you’re going to think, oh, but what if it burns down? Right? And then you talk, call an insurance company and you say, Hey, I have a house here. I’m going to move in. I want to be insured in case it burns down. That’s how it works with, with backup as well. When, when folks are building new applications, backup is not the first thing on their mind.

The first thing on their mind is where are we going to run it? Right? Or who’s going to build it? And, and then they start building the application. And then typically at the point where they’re in a sandbox and they have a pretty good idea of how it runs and when they’re going to be deploying it, then someone raises the flag and says, oh, but we’re probably going to have to back up the data as well. That is what we’re seeing with Kubernetes. But on top of that, they first need to learn it. So a lot of folks are even the pre-sandbox period, but what we’re really seeing, and that is probably circular that goes back to being here at the show back in November. And definitely last year for the virtual one, we had a lot of conversations that were just, what is Kubernetes?

We had to tell people what Kubernetes is. Just this morning. I had a couple of gentlemen from Israel, walk onto the booth. And their question was not, what is Kubernetes? Their question was, how do you guys compare it to an open source solution that they were looking at back then? So people really know what Kubernetes is a lot of them. I mean, I like to look at it as a pyramid and there is a lot of folks still learning, but it’s evolving. And there is a lot of sandbox people right now. And there’s a whole lot of people that are actually running Kubernetes and production as well. And that need a proper solution for that.

Swapnil Bhartiya: Will it be wrong? Like if you look at the cloud native word at one point, security was someone else’s problem. But now with the shift let movement security is something which is priority. So do you think the same thing will also happen with data protection that when I go and buy a car, I don’t install air bag after I bought the car. The air bag, it comes with that. Or if you’re flying a plane, you don’t get a parachute once you have taken off. Yeah. So do you think that the trend, the whole culture will shift where the data protection will be baked in just the way security is being baked in for us? So we will see similar movement there as well?

Tom Leyden: Yeah. And, and a shift is probably the right word because some folks tend to think like, there is the legacy folks, and then there is the DevOps folks, but it’s not like the DevOps folks are falling from the sky or anything. There is a lot of legacy folks that are transitioning into DevOps, right? That is really what we’re seeing. But what we’re also seeing is that whereas in more traditional environments, and we’re seeing that we get an example very well with Veeam where the Veeam sales people, they have very specific personas who they like to talk to. There is the server administrators, there’s the backup administrators, there’s the storage administrators. And they know what conversations to have with each of those folks, because they understand very well what each of those folks’ challenges are. In our case, we are starting to define those new personas, right?

Like the DevOps. It’s not just like one DevOps persona it’s still folks with more of a focus on actual engineering. And then folks that are a little bit more on the operations side. And I think we’re going to see that evolve moving forward still a little bit. But to your point, the element of data protection, data management is part of that it’s no longer, they’re not shifting it off like, okay, that’s for the backup guys. No, the people that are designing, building, running the application, those are also the people that are concerned about not losing data or making sure that users of the application are not using the data. So that is really a shift that we’re seeing exactly.

Swapnil Bhartiya: And earlier we you’re talking about the adoption is growing, but is there any specific vertical that you are seeing adoption or you’re focusing on?

Tom Leyden: That’s a really good question. And I’m getting that question a lot. Also from sales people, you always need enough data to be able to make those conclusions, I think.` And at this point, because of how early we are in the adoption cycle, it’s a bit early to say more this vertical or that I can tell you that we’ve got successes in financials. We have successes in automotive and gaming and a lot of application development, obviously. But I think it’s too soon to say those are the specific ones. For me, the one common denominator, is that how we say it, is innovative applications. That is where we’re seeing adoption folks are not at this point, moving existing applications to Kubernetes.

I mean, if something runs, if it’s not broken, don’t fix it. Right? So existing applications, and I think for quite some time, are going to continue to run on VMs. What we’re seeing is people in AI, people who really need to scale out from a user’s point of view, from a data capture, point of view, scalability is key, but innovation, that is where we really see the adoption. And that was in the end, what Kubernetes was made for. And then maybe finally, so we’ve got a customer success story that we’re going to announce next week. Unfortunately, I cannot share it here yet, but they are in the automotive industry. They actually do enable self-driving car technology.

And they reached out to Veeam because they’re a big user of Veeam VBR. And they said, look, we’re shifting part of our infrastructure to Kubernetes. And they said, at this point, we’re 80-20. And we need a solution for the Kubernetes infrastructure. And Veeam said, well, glad you knocked on our door. We just acquired Kasten. So they put the customer in touch with Kasten and they just loved the fact that it was all coming from the same brand. And they loved the fact how the collaboration between the sales teams from Veeam and Kasten happened and very short sales cycle, they tested the technology, they adopted it. And now it’s really interesting to see how that customer is shifting their infrastructures. They told us clearly they’re never going to move away from VMs. They will always have certain workloads that will be running on VMs, because it is good enough. It’s been working for years and it will continue to be working. But really the innovative applications that are building, that is what are shifting to Kubernetes; and then what they’re talking to the Kasten branch of Veeam for.

Swapnil Bhartiya: Tom, thank you so much for taking time out today. And of course, share with us the whole evolution. And of course the problem that Kasten is trying to solve, which is a unique problem, but it’ll become as we said, it become a kind of the shift left happening; a norm. So thank for sharing those insights. And as usual, I would love to have you back on the show and also to discuss all those new success stories that you folks are working on.

Tom Leyden: Thank you for having me. This was a great conversation.