Matt Klein, Aparna Sinha, Michelle Noorali – Panel: The Future of Cloud Native

Hi, everyone, I'm Alan Shimel, CEO of MediaOps, and welcome to our panel on the future of Cloud Native. I'm going to introduce our panel members here in just a moment. But this panel on the future of Cloud Native, I guess we couldn't really discuss it without discussing what impact events around COVID and the pandemic have had on our community, on the cloud native community, as well as on the adoption curve and pace of innovation in in cloud native.

But it's not the only thing that is happening around the cloud native world. You know, time stops for no one. And innovation and progress stop for no one or no virus either.

And there's certainly been plenty of that innovation and progress over the last months. But first, let me introduce you to I think what I think is an amazing panel of of cloud native community leaders. And I'm going to introduce them by name and let them give you a little background.

We're going to start with Michelle Noorali. And hi, Michelle. Welcome.

Would you mind just giving a little background to our audience? Hey everyone, thank you so much for having me on this panel. My name is Michelle Noorali, I'm a software engineer at Microsoft.

I have been working on Kubernetes native tools for about I guess like 2015. So pretty early on in Kubernetes, I had the pleasure of working on some open source projects called Home Dropped Zeynab, and now I work on open service mesh and service mesh interface, so I'm excited to be here. I also serve on a TOC and I also serve on the governing board as a developer representative.

And that's for CNCF. That's correct. Yeah.

Yep fantastic. Michelle, welcome. Second, I'd like to introduce our partner, Sinha, a partner.

Welcome to our panel. And if you wouldn't mind giving us a little background or giving our audience a little background. Yes.

Hello, everyone, and thanks for having me, Alan. My name is Aparna Sinha and I am a director of product management in Google Cloud. I've been involved with the Kubernetes project since the very early days and I'm also involved with the KNative and Tekton in a number of other open source projects.

I lead the developer platforms team within Google and I'm also I've been involved with the Cloud Native Computing Foundation for a while and I'm currently the governing board chair. Fantastic. And then last but certainly not least, Matt Klein.

Matt, welcome. If you wouldn't mind giving us a little background. Hey Everyone.

Thanks for having me. I'm Matt Klein. I'm a software engineer at Lyft.

I spend about fifty percent of my time at Lyft helping lead our infrastructure team. So our core and for around things like Kubernetes envoy, service smash deploys, things like that. And then obviously from an industry perspective, I'm on the CNCF TOC, I'm also on the GB and I helped lead the envoy project, which I've been working on for the last four or five years.

Thank you and thank you for all that you've been doing. So I want to mention this panel is being pre-recorded. I don't know which of our panel members may be able to join us when it's presented live at Operationalizing Cloud Native with Kubernetes on that on October 1st.

But we will be offering live Q&A chat. I, at the very least, will be on live and do our best to answer questions. But if we can't answer them live, if you do have any questions, you can use the chat window of your interface and we'll try to get your answers from our panel members.

With that being said, guys, let's dive right in here. So, as I said at the outset, no discussion about the future of cloud native, whether we're talking short term or long term, we can't but help but look through the lens of COVID-19 and what in effect, this is that on our on our everything, on our daily lives. Right on how we do how we do things.

But specifically on cloud native and Kubernetes. And you know what's happened. I don't know if we've quite reached the quote unquote new normal, but whatever it is that we're in now and we have been, you know, how how do you think the cloud, specifically the cloud native community has adapted or has been affected by by what's gone on around COVID and related, you know, everything that goes with it?

Anybody want to start off? Sure, I can start off COVID has obviously been an unexpected phenomenon by most customers, and what we've seen is that there's been unprecedented interest in using cloud. And I think it's a natural thing because there's a lot more folks working from home.

And so remote work has become the norm as long as they can ensure business continuity of their services. That has been the top priority, especially in the early days of the pandemic. And then over the last few months, what we've seen is that they're looking for infrastructure and services that are extremely responsive to uncertain demand.

So demand patterns aren't known. Things may go up dramatically or they may go down at any time. And that's what I mean by cloud.

We're finding that developers and customers at large are looking for responsiveness to demand. They're also very, very sensitive to cost. They want to make sure that they're optimizing efficiency and optimizing cost.

And then lastly, I would say developer productivity and time to value, as well as the ability to experiment because no one really knows what the market needs at this time. So developers and companies are really experimenting, agreed. That are mature.

Yeah, I we're experiencing the same, same pattern. There's definitely increased cloud consumption and I think all the clouds are seeing that and really all the software companies are saying that. I mean, like a Zoom, for example.

But, yeah, I've seen that there is this shift. And, you know, in practicality, it's like people are looking for practical solutions. And at a time when there is just so much going on, there is a focus on making money and making sure, you know, as Aparna was saying to play off that, there is an increase in wanting developer productivity and making sure that people can get to that as quick as possible.

I'm also seeing a lot more conversations around security and making sure that people are doing things securely. And there's just lots of various topics there. But that's definitely something I hear a little a little bit more about.

Mostly seeing a lot of the conversation around people starting to burn out and not really knowing how to deal with work life balance. And then on top of that, you know, for those of us who have kids at home, you know, it's not uncommon anymore for me to see people's kids in the background and just have them be a part of our meetings. And that's cool.

Like for me, that works. We're all just dealing with uncertain times. But, yeah, I mean.

My experience are the kids are usually better behaved than the cats. Cats, cats have a way of just. Very True.

Working on keyboards. So but but you know, yes, there has been this move to the cloud. Right.

We've seen six years of digital transformation, cloud transformation take place in six months. But what I think is really relevant, especially for our audience and for you all, is that it's not just a cloud transformation, it's a cloud need of transformation. I've seen surveys that something like 70 or 75 percent of all new applications being deployed.

These are new applications granted, not legacy, but all new applications being deployed to the cloud are using a Kubernetes cloud made of stack, if you will, an infrastructure based on cloud native. That's phenomenal. I mean, Matt, what do you think about that type of those numbers are just hard to argue.

You know, I don't think I've seen honestly that much change in the tech from COVID. But I think it's pretty much what you said where I think we're seeing 5 to 10 years of change in terms of remote work, in terms of acceleration towards apps and cloud technology. I think we're seeing that accelerate into 6 months, 12 months, you know, so that that is a that is a super exciting thing to do to see the focus in this area.

But I think for me, what's more interesting is, is just all of the stuff that's happening around it. Right. Which is what was said before, burnout out, just trying to make make all of this stuff continue to happen while, you know, we have such crazy things happening in our lives.

And that's been the most difficult part for me, which is, again, to be honest, I don't know that my job has changed that much pre-COVID to post-COVID. Pre-COVID, my job was already primarily remote. You know, it's working a lot in open source, working with all of these companies, doing all of these things.

But the the move towards cloud and just the relentless pace and then just all of the stuff that's happening in the world. It's been it's been a pretty intense six months. It certainly has.

But could we have done this without the fact that cloud native is predominantly open source, that, you know, has open source been a big enabler for this rush to to transform to the cloud? Right. It seems to me.

Right. Where would we've been able to accomplish this? But for things like Kubernetes and Service Smashing, Envoy, and a lot of the projects that we see, it's the NCF in that we think of when we think of cloud native, right.

That, you know, it's great to say I want to move to the cloud. But without that stack, without that infrastructure in place, would it have been possible at this pace? At this scale?

Right. I think probably not, but, you know, I mean, it's not it's not just about open source. I mean, that's certainly a big part of it.

But I think if you look at the larger cloud native community, it's a set of cloud vendors, product providers, open source. I mean, it's entire it's an entire ecosystem. And we can have people we can have a larger conversation about all of these pieces.

But I don't I don't think it's any one thing. And I think we're still very early in the cloud native journey. I think there's lots of improvements that we can make.

And I certainly do think that open source and CNCF have been a core part of it. But I think it's a very complex topic that involves lots of lots of different entities. One thing I would add, Alan, is that it's not only cloud native.

I mean, I think just realistically, we are seeing even traditional workloads moving to the cloud at an accelerated pace. But you are right that new workloads are disproportionately cloud native. And those workloads that our cloud native and teams that have invested in cloud native beforehand have a significant advantage.

And those that haven't are now wanting to move to cloud native at a much higher pace. So cloud nativist certainly accelerated. And then on open source, I would say open source is a very important part of every developer's toolkit.

It's part of so much of what they use that it's sort of ingrained in there. And it's not even necessarily a conscious choice. But what we're seeing is that many more enterprises are more comfortable with open source.

It's more managed open source generally in the enterprise context, but they're more comfortable. It's become more ingrained. And I'm not sure that it's necessarily because of COVID.

I think that was a trend that was already there. I think maybe it's been an accelerant. I'm sorry, go ahead, Michelle.

Open source has definitely made things, I think, more accessible and just having that source code there for folks at smaller companies where they understand how to make something modular, how to make sure it scales, and how to build this large scale system. Those examples in those patterns are there because the source code is open, the docs are there. But what I also appreciate is, you know, CNCF making that that source code vendor neutral, putting that source code in that IP neutral space so that even folks from larger companies with more rigorous contribution guidelines can also participate in the in the open source community.

And then having like having that foundation there also gives us this framework to get feedback from and users. And that, I think, is the bigger point. I think it's not just that this stuff is open source, but that we can see what the needs are using this project, on using certain projects and what way and how it's actually going for them.

I think that also helps the transition, too, and it helps people feel more confident in what they're what they're using to move to the cloud. Agreed, you know, that you know the saying you can't make wine before its time, right? And if one of us mentioned, you know, relatively relative immaturity of some of the cloud native technologies and tools and and, you know, in normal times, whatever normal means, software, even especially open source projects, have a maturity model and maturity pattern where you'll see, you know, general release and then updates and regular updates and maybe there's a long term release or some short term stuff like this.

I think with COVID, not only have we seen an acceleration to the cloud, but have we seen an acceleration of the maturity of these tools. Right. Or or do you think it's just the normal course of of, you know, this is the normal pattern they would follow anyway?

I you know, I just see us. I think I'm seeing more innovation, more with more usage comes more operationalization, if you will, of some of these tools. And so we see, you know, and beyond Kubernetes, when you look at things like Prometheus and and the telemetry now stuff in the observability space.

I mean, all of these kind of cloud native technologies that are hanging out there, right? They seem to be. Like me, it's almost like, you know, the old Japanese monster movies, when they're exposed to radiation, these animals, these these projects seem to be hyper maturing, really, you know, gaining users at an incredible rate.

The community growing at an incredible rate. And that's because it is being driven by cloud usage, but, you know, a byproduct is are these are these projects and tools getting better, faster than they would without COVID? Any thoughts on that?

I mean, Matt you're involved in envoy, for instance, right? Yeah, you know, to be to be honest, I, I have not really seen that and it's a.. Well.

It at least from where I'm sitting. Right. Because I think there's different technologies in this space.

Right. There's as I was saying before, there's hundreds of open source projects probably or there's 40 or 50 within the now, but there's hundreds in the overall ecosystem and there's tons of products and cloud products. And I think the reality is that these these projects sit at different layers of the stack and some of them are harder to adopt than others.

So if you look at base level technologies, things like Kubernetes, things like envoy, the honest reality is that you can't adopt Kubernetes within a week. You can't adopt envoy within a week. It's just impossible.

I mean, the technologies are too complicated. They require they require wrapping. They require abstractions.

So to be perfectly honest with you, at least speaking from the envoy perspective, which is where I have most of the experience, I really haven't seen much change due to code. Like for me, again, there's the whole other life aspect that's that's made it very difficult. But from a project perspective, I just really haven't seen that much difference.

I mean, we continue to march forward and do things and try to make the project better. I suspect that where you're seeing more adoption or more more need to make things more robust. Faster is actually on some of the vendor products that are probably under massive load that they weren't previously.

So if you go to the cloud providers, the caching providers, the API gateway providers, the database providers, they're the ones seeing the load and they probably have to get operationally better. But they're on the front lines. I mean, all of the open source projects that are the base, there's kind of a there's kind of a leading edge, you know, and then the products are up front and then there's all the technology behind.

So at least that's my own hypothesis, is that I think you're seeing more pressure on the service providers right now. I would agree with that. I think the major trend is that customers are looking for simplicity.

They are looking for things that are turnkey, where they can focus on writing their applications and getting the new experiment out to see if the customer demand is going to meet the, you know, what they have developed. So they're doing a lot of experiments. And I'll give you an example.

Like we work with an educational institution which is trying to try out different schooling methodologies for children. They're putting out different types, of course, plans for for studying at home because they don't know what exactly is going to work for different age groups. They're not that concerned about.

I mean, they want to use the most efficient technology, but they don't really necessarily have the wherewithal to use all the open source and kind of all the details of it. What they want to do is focus on doing their experiments with the coursework, and that is that needs to happen more quickly because of COVID. So Matt is right.

There's a lot of demand and usage of these managed technologies. And I think the cloud native technologies are disproportionately growing. That is very much true because they're more efficient, because they're more scalable.

But as far as the timeline of when these projects mature, that's a fairly long timeline. And I think that the work that we do in the community we're still doing that was always remote work. And if anything, it's only impacted by the the work life balance.

But on the whole, it is continuing. Excellent. So I have one more topic on this COVID thing, and then I want to put COVID in our rearview mirror.

I wish we could tell you the truth in real life, but all all of you mention the toll it's taking on people potential burn out. We've seen studies where, look, we may be working something like, believe it or not. Twenty, twenty five percent more hours during COVID than we were, you know, per week than we were before.

Some of it is because we're not commuting to other parts of it is people who are not used to working from home. They don't have boundaries. They don't know when to hit the on off switch.

It's great as an employer, I love it. Right. But I also know it's not sustainable.

People will burn out, people will you can't go with that pace forever. So let's talk about what can we do in the cloud native community. You're all on various boards, part of CNCF.

What can CNCF can do to help? I mean, we can offer counseling, but what can we do to make people more aware of it, make people more sensitive to it, try to make this situation better? Can I please jump in on this one first go?

I have been so obsessed with my new policy of not doing any meetings on Fridays, I think it's every other Friday now. No meeting Fridays are my favorite. And they just allow me to recite and to like focus.

I think when just we were always online on slack, doing zoo meetings before, but now it's just even more so because we're not having off sites or not having in-person meetings. So there are just more meetings and I, I can't get work done efficiently that way. So and context switching even between even within the day, every few hours is very hard for me.

So I wish that we would still like a no meeting day for the cloud. Native community would be really nice. That's that's my hope and wish.

I think it's just ah how about how about a no GitHub day. I mean that would seriously that would improve my mental state by about five hundred percent. Really.

Oh really. Ok, well you know, I think one thing like Michelle was saying, you know, people have pets and people have kids. And I think one of the things that helps is being very welcoming to the environment and the context that people have.

And it's actually a joy to be able to connect with with people's lives. So that's maybe a benefit that we have are moving to Covid. But I would say the other thing is acknowledging people's context and making room for it.

And that's sometimes hard to do in a company context and in a community context. But that's what is helpful. Absolutely, yeah, the only thing that I would add is that I think from a corporate perspective, companies are obviously doing what they can with the meeting days and with, again, trying to be understanding of people's personal situations.

And I think that's important and that's fantastic. Know, I would point out that even pre-COVID from an open source maintain our perspective. It was already, I would say, a very difficult situation in terms of splitting time between open source maintenance and, you know, between probably corporate obligations and those types of things.

And, you know, I don't think we had done a good job of solving that pre-COVID in terms of work life balance for open source maintainers. And I think it's frankly, it's just gotten worse during COVID just like speaking for myself. I have a job at Lyft.

I have a two and a half year old kid. There's COVID. There's a bunch of other stuff going on.

And I just find that in the current state of the world and news and everything else, I'm just like mentally at my limit. I'm like barely holding it together. And when it comes to especially the opensource maintenance side, where there's very little boundaries, GitHub is nonstop.

You know, it is mentally taxing. And I don't have any answers here. But I think that this continual conversation of how do you take some of these maintainers who are really the crux and they're holding together some of these projects that are the linchpin of the entire ecosystem, and they're all always on the verge of burnout.

I think it's a it's a conversation that's worth having and it's one that we need to have more, especially during COVID. No doubt, no, no doubt about it. Thank all three of you for for the background on that, let's put COVID.

I wish I really hope you pray we do put COVID in our rearview mirror. Let's put it behind us on this panel. Let's talk a little bit about the future of operationalizing cloud native.

A little tough time there. So COVID or not, Matt you touched on it. Kubernetes is hard, envoy is hard.

Mesh, some of the service mesh stuff is hard. A lot of the cloud native stuff is not for the weak of heart. It's been a little, you know, and as a result, we've seen the rise of managed services for a lot of cloud native.

Right. A lot of these vendors, basically, they built the business around managed service of an open source project, whatever it may be. Looking into your crystal balls now post-COVID, things are back to whatever new normal is.

Where do we see improvements in the operationalizing of cloud native? Anybody want to start off there? Aparna, you look like you're ready.

Yeah, I can start off. You know, I think one of the things that at Google, we are very much maintainers. We're very much starters of open source projects.

And I think that we blend the two things of contribution, community involvement and mentoring in the community, along with serving users with something that is simpler. And as I mentioned, in the current environment, simplicity is extremely important to our users. So I found that that there's a good, virtuous cycle between providing a simple experience of an open source nature to a customer, to a user who doesn't have to do all of who doesn't have to deal with the operational complexity and getting that feedback of, well, what are the things that don't work?

What are the bugs, what are the issues with the current code and then contributing that into the open source, contributing those fixes into the open source? There's a big, virtuous cycle there because a lot of developers can get started with the open source. D.

or that require a lot of understanding of the details, especially for I think for developers. They want to focus on writing their applications. They don't necessarily want to understand all the operational difficulties of putting together an open source project.

And so I think one of the things we do well at Google is we bridge that gap and we make we make it a closed loop, working very actively in the open source, but also working very actively with the mass of users. Michelle. You know.

I think. So the original question, Alan remind me. I'm sorry, so what what can we or what do you see that can be done or what may happen to kind of just make it easier, make it easier to adopt these technologies and use them?

Yeah, I mean, like all credit goes to Matt and I said that's like in a few different places, but he introduced me to this idea of operational dashboards and like Lyft come out with one and Backstage has come out with one. And I get so excited about these pluggable dashboards because I think it's just much easier to understand things when you can see them. And you're just like, for me, visual is easier for different types of people learning as people have their own learning patterns and and all that.

But for me, visual is easier when I can see everything in one place and that workflow is custom to me. I think we've been thinking about and building custom workflows for a while and we've seen these like an increase in projects that are not so broad and big and all encompassing. But instead they tackle this like one small problem, for example, like a testing.

There have been a few sandbox projects that have come into the CNCF app around chaos testing and making that easier on cloud native workloads. So I like this idea of custom workflows, then visualizing those custom workflows with pluggable dashboards and just this all around patterning towards making these simpler tools like Aparna also mentioned simple, simple tools, simple, simple and easy to understand workflows and those I especially love when those come from end users, because a lot of this like the foundational work around building modular components that fit in with each other. For example, Containers, Kubernetes is ongoing service.

These are broader things that have laid some some of the groundwork. Prometheus, if there's anything like quite a few but databases, there's like all these foundational ones and they're being worked on. And in fact, in these areas and these broader areas, we are seeing more consolidation.

But what I'm more excited about as end users coming back and saying, hey, we built this small thing on top of this broader thing and it works for us and it'll probably work for everybody. I've read a lot of other people in our situation looking forward to that. So, Matt?

Yeah, I think that this is a topic that I think about a lot. And I think it's interesting. And the way that I usually talk about is I think we're in a really interesting time right now where, as others have said, we have a lot of these foundational technologies that are opensource but taken on their own.

They're incredibly difficult to put together in a way that makes sense for most users. And that's the key, is that you have leading edge users like companies like Google and Lyft and folks like that. But then you have the massive tail of users that want something that's simpler to use.

And I think what's fascinating about this is that there's a push towards open source. There's people want the comfort of being able to look at the source code of the things that they're running. But the reality is that we're trying to get back to something like Heroku.

Right? So it's like there's there's the usability of Heroku. That was from 10 years ago.

And what we have today is vastly worse than the experience that what people still have with with something like a like a built in really opinionated platform. So, you know, I think the massive amounts of Yamal and the complexity of setting up things like communities and envoy and all these technologies, it's just too much it's too much for a average company to take on. It just doesn't make any sense.

So I think we have a lot of things going on where we have people that are probably adopting technologies that they may not need to actually solve their business case. So usually I counsel people. Let's start with the problem that you're trying to solve and let's find the right technologies to solve that problem.

But I also think that over time, a bunch of these foundational opensource technologies, again, like Kubernetes, Envoy, et cetera, they're going to be hidden for most end users. Most end users don't care about container scheduling. They don't care about networking.

You know, they want to run basically functional code or run end points. And they want to talk to a cache, talk to a database. Read and write from queues.

Right. And as Michelle was saying, you know, we can have better dashboards to understand what's going on and integrated deploys. So we're going back to Heroku and.

That's right. And I think you see this like you have each of the cloud providers are building things like Fargate and LAMBDA and Google functions and, you know, all of these things. And I think that's going to be the trend.

Right. The trend is that open source is going to move back towards a fast type, quote, serverless situation. I think the vendors are going to continue to provide these integrated solutions.

And ultimately, in 5 or 10 years, we're going to get back to to a more basic environment, but we're just not there yet. Right? And this is a very tumultuous time where we have all these technologies and there's just too much Yamal.

There's the cognitive load is too high and there's a bunch of work that's needed to make it more consumable to most users. I just want to jump on this just to echo Matt's point even more. It's like I've been working with these technologies for so long.

But on the on the weekends, I help with this. I volunteer with this nonprofit group and they have they have like a backend service that they're running. And at first they kind of brought me in because they're like, oh, we're using Kubernetes and Hellum and these other technologies.

And I was like, that's cool. I like work on these things. I could help.

And then I just actually migrated us to Azure services because it's like the Heroku for Azure and it's just like give it a give it a container, make it right. And I'm using energy identies. I'm using the MySQL service and it's like I'm just going completely 180.

And as someone who's like ingrained in the community, it's easier for me like to to go do these things on on some other platform that may run. Kubernetes under the hood, but that's exposed to the end users and it's easier to echo what Matt said. I don't disagree.

Aparna, do you want to add anything I left before I was going to add my two cents to it, but sure, I mean, I'll add two things, actually. One. So there are two sides of the coin there.

The experts that are that are dedicated engineers. And for them, at least in the sense that we've been talking about providing a lot more best practices, providing a lot more of user-oriented forums, because you can always learn from other users. And actually that's one of the best ways of learning, is to see what did X, Y, Z implement, how did they do this?

And that's a good way of figuring out what how do you architect your solution? And then the other piece is on the on the novice side or the I think the more important side, which is the non-expert users, even people like Michelle assaying, saying even expert engineers, when they need to build something, they go for the thing that's the easiest, fastest, simplest thing. So I'll give you an example from our side.

We use a service called Cloud Run to do the work that we needed to do for COVID is not because we aren't experts at the other pieces. But, you know, the best way to get started, the fastest way to get started with containers is with a serverless container offering. And that means that you don't have to know much about containers.

You just you just write your application, which is what a service like that allows you to do. So even we do that internally. And of course, for external users, that makes sense as well.

So I would say this is normal, believe it or not. I think what you guys are describing is, is a relatively normal adoption pattern in technology, right. Where you're going to get about 15 percent of the market, who are the quote unquote, early adopters, the people who do it because it's cool.

B, it gives them a chance to stretch their their boundaries a little. It has great potential. And they like to be builders.

They like to be innovators. They like to kind of set that course, if you will. And but beyond that 15, maybe 20 percent, there's this huge mainstream and that mainstream's divide crossing the chasm.

Right. And it's divided into two segments. The first part of that mainstream, maybe as much as 30 percent of the market of the entire market are people who they'll adopt a solution if it solves a problem for them.

They don't care whether that solutions is new, old. If it solves the problem for them, they'll give it a whirl, right beyond them is a second mainstreamer later, later adopting mainstream who they'll only adopt a solution not only if it will solve their problem, but they know someone else who solved that problem by using the solution. Right.

So it's not just that, you know, Kubernetes is a great way to get better distributed, you know, application out there. But one of my peers has used it and they they've been richly rewarded for it or whatever. Right.

And so when we look at cloud, the cloud native adoption, where is it on this cycle? I think we're coming into mainstream. And when you get to mainstream, you're getting the kind of people you're talking about who don't want to get bogged down in YAML.

Right. And they Michelle, you know, Kubernetes is great. They don't know Kubernetes from you know, I don't even know what rhymes with Kubernetes.

I can't make up a rhyme, but they don't. You know what I mean? They don't care.

They just want to know it works. They want to know what works. They want to make it easy.

And that that's a typical mainstream adoption type of use case. Right. And then, of course, they'll be laggards 10 years from now.

You know, they'll they'll figure it out. But, you know. But that's healthy.

I think I think that's a healthy regular adoption cycle. And it's what we should see. Guys, we're just about out of time, which is one I want to give each of you a chance to kind of sum up, if you will, and maybe any advice for our audience on what they can do to make their cloud native Germany better over the next 6 months to a year, 18 months.

I don't know if we really anyone can see beyond that. Right. But for the next 6 months to a year, 18 months, what what could what could our audience do to make their cloud native journey better?

Matt, you went last to start, so I'm going to ask you to go first to end, if you don't mind. Sure. I think my advice here is always the same and is actually fairly simple, which is I think in this space there's a lot of shiny technology and I think there can be some FOMO and some feelings of, you know, if you don't adopt the latest and the greatest that you are behind.

And my advice to people on their cloud native journey is don't do the journey if you don't have to. And by that, I mean look at the problems that your organization actually needs to solve, like approach it from a customer perspective, approach it from a business problem perspective, and then find the simplest technology that will satisfy that use case. And that may be a managed service offering.

It may be an open source solution and maybe a combination of those. But I just I see too much of folks tend to get bogged down in all of the possibilities and all the different technologies and the feeling that we have to use all these different things. And that's when I think things can go badly wrong just from taking on too much complexity perspective.

So my advice is always just start, start with it, use case, start with the business requirements and then try to find the simplest thing that will satisfy that. And typically it's simpler than most people think. Fair.

Aparna, you're kind of in the middle, if you don't mind, I'm going to ask you to go next. Yeah, sure. My advice, I think for the coming years, it is really important to prioritize development, software development and software developers looking after their health, looking after their career, making sure that they are well-skilled and focused on adding value to the business.

Whatever company you're in is more and more a software business. Every business. And I think with the pandemic, we're seeing everything move digital.

So it's incredibly important for for users to invest in developer efficiency. And I think they're already doing that. And in order to do that, yes, you want to use the simplest tools that you also want to be able to bring more and more people into that fold of development.

And to personally, like you said, I think that's a very exciting place for cloud native to be. And I think as a society, it's a it's a very exciting place for us to be where we can bring more of the public into this value added work of developing software. Thank you.

Michelle, You've got the last word so, well, Math and Aparna summed it up quite well. I think those are two very good pieces of advice. I think I'll just add on a small thing, which is ask questions and make sure you understand the thing that you're adopting.

I don't take that for granted and make sure you thoroughly, thoroughly understand how it works, what the security aspect of it is, what what is the onboarding process. And don't be afraid to jump ship if things are too hard, because evaluating early is better than suffering in the long term. So thank you.

Thank you. Michelle, Aparna, Matt. Thank you all for being our guest here on this panel for operationalizing cloud native with Kubernetes, continued success.

And also, guys, thank you so much for your volunteering on boards and TOCs and everything that you do, because you know what? It takes involvement to keep those wheels turning. And if it wasn't for your involvement, the wheels don't turn.

So, thank you all for that as well. This is Alan Shimel from MediaOps,. Thank you.

And enjoy the rest of our event today.