DevOps Evolution with The Futurum Group’s Paul Nashawaty at KubeCon Paris 2024

This is Textron tv. Hey, everyone. We're back here.

Live in Paris. Day three coverage. I feel like day three DevOps.

It's day two. DevOps. No, but it's day three of CubeCon.

We're in Paris. We're coming to you live on Textron TV and all the whole tech drunk network. Let me introduce you to our next guest.

This is a real life DevOps analyst right here, Paul Nati. Paul is with the Futurum Group. If you were tuning in yesterday, we, we had another Futurum analyst on.

We're happy to have Paul on though, because he's their DevOps guy. Well, Paul, I hope I didn't steal your whole story. No, not at all.

Please, Welcome, welcome to techron. Thank you, Alan. Thank you.

It's great to be here. It's great to be here. You know, I really, I really love the warm intro you just did, uh, love DevOps.

It's really my passion. It's where I live. Uh, app, app dev, app modernization.

It's kinda like that space that I'm in. So you're fairly new to Futurum, but you're not new to DevOps, as you mentioned. Let's hear a little bit of the pulse story.

Sure, sure. Yeah, no, thanks for that. So, uh, definitely a little bit new to Futurum.

Um, I, I built a practice, uh, an analyst practice about four years ago. I've, so in, in the app dev space, app modernization, app development, uh, growth in that space, focused on the CICD pipeline, uh, really about cloud native, past, present, and future states. Uh, prior to being an analyst, I was in the vendor world for about 25 years.

Both large and small companies took four companies to events. You know, so, so really, uh, understand that kind of startup mentality. That's, but also the, uh, the big company in politics as well.

Yep. What kind of, it's very hard to do something for a job if you're not passionate about it. Right?

Where's your passion in DevOps? Oh, wow. You know, so I have a passion.

I mean, I wrote a book on, on rapid application development, so, you know, for Dummies book. So, I mean, I really, really, that's my passion is like, I really love it. So when I look at DevOps, I look at the day zero, day one, day two, right?

The build, release and operation side of it. I look at like what it means to take, to build those applications, get 'em out the door, what does it mean from a perspective of today's technology versus what, what's happening in current stance, but also what's happening in the future, right? I'm looking at, you know, uh, monolithic applications, moving to microservices and containerization, but I'm also looking at like serverless and wasm and web assembly and what that, where that's going.

So it's really kind of an evolving state, really exciting time. High been, well, you know, so March, uh, 14th, yeah. com.

Very publishing DevOp. Very good. Congratulations com.

Thank you. So we've been, you know, in this DevOps space a while, and, you know, I've certainly seen changes evolution over the 10 years. Uh, but you know what I just realized before we get into that, there are people out here who maybe aren't familiar with Futura.

Oh, okay. So, as I mentioned, you're, you're, you're the DevOps guy there, right? I Am.

But There's more to Futur than DevOps. If you would, I, I don't want to just take us off track, but take a minute. Explain to our audience how is Futurum organized?

Absolutely. So Futurum has a broad set of offerings that, um, allow, uh, organizations to really help their, uh, go to market strategies, understand how they're going to, uh, broadcast, who they, uh, who that company is, what their message is, where they're trying to get out to the market. So, uh, whether it's, uh, video assets or written assets or content anywhere from top of funnel, mid funnel to conversion assets, have really helping that go to market and drive through.

So, but first and foremost, I look at it in the context of being a primary research firm. Uh, that's what I focus on is a research element. Mm-Hmm.

But there's a media element as well, and there's, and there's, uh, intelligence portal. And, but with the practices, there's 14 practices throughout, uh, fu um, the FU group. Everything from cybersecurity to infrastructure to storage, to networking to all the, all the things you can imagine.

Sure. And of course, application development, that's where I, that's where I fit in. Absolutely.

And I, and that's, uh, that's where I, I focus in on. Excellent. Alright.

We got that out of the way. Let Daniel, you're welcome. Um, but let's now focus back in on DevOps.

'cause both of our passions lie there. Absolutely. As I said, 10 years on, I've seen a lot of water under this DevOps bridge.

Right? I mentioned to you off camera, we're doing this whole report on what we call DevOps next. Where, where do you see us on this journey, on this DevOps journey, you know, as an industry, as an ecosystem, as a community.

Yeah, absolutely. So, I mean, I'm looking forward to your, to your DevOps next report. I, I, that also aligns nicely to this observability report I'm putting out as well.

So I'm looking forward to collaborating you with you on that. But let's talk about the evolution, right? Um, you know, it, it really is still immaturity, right?

I mean, there's a, there's, there's organizations that are out there that have it and lines of business. Then there's organizations that have it, DevOps and lines of business organizations that have it, DevOps, SREs and lines of business, and then platform engineering. Of course, that comes into the play as well, right?

Right. So when you look at this kind of evolution, it really depends on where the organization is in a maturity cycle. Yeah.

Um, I was talking about was, here we are live on the show of, of K coupon. I would, I'll give you a nice shiny quarter. If you find me somebody that identifies as a, a server or a storage admin, they're not, right.

They're, who are they? They're platform engineers or engineers. Right.

So, and this is, I'm glad you brought this up. 'cause here's my take on platform engineering, you know, to quote Shakespeare a rose by any other name would not, would still smell as sweet. We've been doing things that are platform engineering, we just didn't call it that.

Right. But the, the kinds of things that platform engineers do are, are the kinds of things that have needed to get done. Sure.

I think more than anything, platform engineering as a term has arisen because I was never a fan of the term DevOps engineer. Right. I I think DevOps is more about multifunction teams.

Yeah. But saying someone's a DevOps engineer, I understood why a lot of engineers, a lot of people who are today calling themselves platform engineers said, no, I, what, what the, what the heck's a DevOps engineer. Right?

Right. Right. I I do CICD, I do ops, or I do, you know, so I, I get that, but I also think there's a continuum that's even bigger than the ones you spoke about.

I think it starts way here to the left with Agile. Yeah. Right.

Agile is 25 years old. Absolutely. It's pretty well defined where Agile leaves off sort of DevOps picks up.

And, and when I say DevOps, I mean many flavors including SRE, including platform engineering, including that whole software, uh, uh, s software supply chain and, and SDLC. But then it goes further right? To ITSF to service management.

That's right. This is the world we live in, in the software factory. Yeah.

Right. And, and I think the sooner we realize that, like how you said, how organizations are kind of defining themselves, I think those are artificial silos. Well, I mean, that eventually gotta come together a bit.

Yeah. A I mean, I think it comes down to what are the business KPIs? What are businesses trying to release?

Yeah. So when I look at my recent research and I see of a, of a, just a sub 400 respondent survey, I found that 24% of those organizations that surveyed, that responded to my survey wanna release code on an hourly basis. But yet only 8% are able to do so.

But the 8% that are able to do so touched on what you're talking about. They, they, their commonalities between it was agile software development, they could use DevOps practices. They use infrastructure as code, and they, and basically they kind of, they push it out that way.

Now, the reason why that's important to note is because when you look at the release cycles and you look at pushing code out the door, DevOps KPIs, get the code out the door, push it out, right? Yeah. But they'll also need to worry about things like quality.

Quality, right. And when, when QA is going, Hey, I got a problem. That's, you know, DevOps, like, I need to push it out the door.

Right? Well, what I found in my trending survey is from 2022, I found that only 29% of organizations are doing continuous testing. Really only 20 in the CICD pipeline.

Right. But in 2023, it jumped up to 66%. That's more what I see.

Okay. Right. And, and, and the reason behind that in my mind while I was kind of talking through this, was these organizations were using their, their end user clients as the test bid As the beta Yeah.

The beta testers. Right. Right.

And their justification was, well, we use software, a agile methodologies, what pushing code on as Scrum I love every two weeks. Right. Not a good model, right?

No, because you lose the, you Stick, well, you're making your customers Guinea pigs. Exactly. Not a, not a good place to be.

No. Right. So let me, I, I'll interject one thing 'cause we are here at CubeCon.

Yeah. I think one of the reasons for that jump though too, Paul, is the whole GI ops Argo, which is a huge CNCF project here, Argo, that allows us to do our CICD within a cloud native stack within a cloud native system. And there's a place for testing in there to the left of deployment, right?

Yeah. Yeah. And, and then I I, so one of the things we do at Techstrong right?

Is we have this video series called DevOps Unbound sponsored by tricentis. And you know, they're, they're kind of the leaders in continuous testing. That's their tagline.

Um, they've really made so much progress because of this cloud native stack cloud as cloud native has become more dominant, continuous testing is part of your release cycle Yeah. Has become easier, more organic. And, and I think that's a, a, a reason you see that jump too.

Yeah, I agree. I mean, I think part of the, uh, continuous testing, uh, continuum there Yeah. Um, fits into the, uh, the, the, the adoption of using open source as well.

Right? So if you start using open source, we find that in our research that, uh, 87% of organizations who responded, they want to work with vendors that sponsor open source projects. Yes.

But they also don't want to go at it alone. So they want those vendors to provide enterprise level support. So, but when I bring up open source in, in relation to testing, the reason why these vendors wanna, or I'm sorry, these organizations wanna work with the vendors that it sponsor these projects, is because open source hardens the code.

It, it, people can bang on it, make sure it works, and then it comes out the door and says, okay, now it's proven. It works, it's tested in, it works the ecosystem, and plus you have community support. So if something goes wrong, you can lift it from there.

Well, that, you know, that's one of the reasons why this place is so damn crowded and they do so well. Right. It's this foundational open source model where that rising tide lifts all the boats.

Absolutely. And then vendors get to kind of riff off of that. Right.

But, you know, you bring it up. But now we're, we're, we're starting to see a little bit of a backlash around open source vendors and changing licensing and saying, wait a second, we got a supporting this open source project or training for it is not enough for our business model. I can't make a business where converting 4% of my users is my success formula.

Right? Right. I gotta do more, I gotta figure out how to ring out more revenue from that model.

Yeah. And I think we're just entering that kind of phase here. It's gonna have repercussions.

Oh, for sure. Right? Yeah.

And, and the way I would see it is, uh, you know, there are many vendors that's that use op, like work with open source models, right. If you don't have strategic differentiation, and this is something that I work with a lot of my clients with Mm-Hmm. It's like, what are your three to five strategic differentiation points?

If you don't have those points and truly understand why those points are differentiators, then it's going to be very difficult to cut through the noise. There's a lot of noise. Well, no.

Take observability, which is what you're working on now, right? Right. Look, at the end of the day, how many of 'em are just open telemetry and Prometheus?

Maybe a little Grafana right? Under the covers. Right.

So where's the table stakes, right? If if, if you Have, that's the table stakes, you better do something more than that. You want my money.

That's Right. If you have 5, 6, 7 companies doing the same thing using open telemetry, where are the three to five things that make it work? That, and I think, I think, look, there are gonna be companies that struggle with that, and quite frankly may not be around.

Well, I mean, look, look what's happening in the industry, right? You have the Cisco Splunk acquisition, right? You have, uh, the merging of, of different companies that are going under pe, you know, and really getting investment from those.

And they're consolidating, they're coming Together. You're gonna see a lot of this. You're gonna want to see a lot of it.

And because of that differentiation. So I would strongly recommend vendors that are looking at these using open source to put their products forward, but validate their differentiation against market data. That market data is really gonna provide that solid point to say, this is why it matters.

And find that little, little spot that makes it work. I, and, and by the way, this is not just a problem for the smaller guys. No.

I think even some of the companies you mentioned need to figure out what their sauce is, what's their special sauce here? Those three to five points? It'll be interesting.

I I think we're, we're, we're certainly in for consolidation. I think observability in particular Yeah. Is gonna change drastically The consolidation and rapid expansion of the ecosystem.

Absolutely. You know, we we're probably talking, I don't know, I don't know how I forgot, I didn't time it, but we haven't even mentioned AI yet. We gotta talk.

Well, we have to talk about ai. I mean, if we didn't, we'd miss it. Yeah.

So, ai, DevOps, AI and DevOps. What do you see there? Oh, well, you know, AI and DevOps, you know, I just, I did a study around AI impacts to the developers, right?

And we, and, and what we find is 18% of of organizations are using AI in their production code. But 27% of organizations are in consideration that they're thinking about it. They're trying to understand what's going on with it.

So why is that? Well, there's a little bit of uncertainty. I mean, we saw with, with a recent airline that they put out some stuff that didn't actually help them, right?

Because they say they try to blame it on the ai, but it's like, it's their business, right? So if you, but the common use cases, like for DevOps and developers, I mean, you know, rapid code creation, uh, you know, documentation, uh, customer satisfaction, knowledge base, all is the real, like those are the real use cases. But what I'm finding in our research, excuse me, what I'm finding out research is, uh, it varies based on geographical locations.

The, the adoption is really Yeah, very much so. Uh, there's some, some, uh, org uh, parts of the world that are much more, uh, willing to adopt faster than other parts. So for example, we're in, which we're in AMEA right now, we're in Paris right now.

Uh, far more, much more conservative in AMEA than it is in say, apac. What about North America? North America's kind more in the middle.

Really? Yeah. Somewhere in the middle.

Yeah. Guess What, what you think about it? That's kind of, I could see the reasoning behind it.

Yeah. We could share those data points. But, but the thing is, is, uh, when you look at it, it really does come down to look at what organizations are doing.

I mentioned earlier about releasing code on an hourly basis, right? Right. Applications are being developed very, very rapidly.

Now, if you're not using some type of AI to take actionable insights and do some automation, you're competitively going to be behind you. There's, there's not a firm and organization out there that can put enough people resources behind creation of applications moving forward. It has to be generated with autonomous ways of doing it.

Because if it's not, um, you're just gonna be falling behind. And then if you're, if, if, again, if you're not using that, uh, uh, um, automation to do it, the impacts of security, the impacts of, of, of bug is the impacts of anything will fall off and you'll, and your product will be impacted by that. To me, it's a speed of business issue.

Absolutely. If you're not automating you, you're, you're, you're in that right lane, and you're gonna have people blowing their horn at you all day, right? Yeah.

Correct. That's the bottom line. Um, you know, I sit here listening to you talk about hourly, you know, that's the goal.

Hourly. So you're talking to someone in my last venture backed company that I co-founded, you know, we were really proud of ourselves when we went from twice a year to quarterly. I know.

Yeah. We thought, my God, how am I gonna finish all the testing Right? In a quarter, right?

Right. Now we talk about hourly. Well, Waterfall versus agile, right?

Well, we were, we were, it wasn't that long ago. We were, we thought we were doing agile, right? Maybe what we were really doing was fast waterfall, Fast waterfall.

But um, that being said, that's the world we live in today. And that's why when we talk about things like automation and we talk about things like ai, you're not going to hit those KPIs without those. Not at all technologies.

Not at all. It's just Not at all. It's just gonna accelerate.

I don't how big a team you have. That's right. 'cause at some level, the more, the more people you add, the more time it's going to take for people to do, you know, to coordinate.

Right. Um, good stuff, Paul. Where can people get access to some of your reports and stuff?

Oh, I appreciate that. com is where my, my page is. My, my information is, uh, most of my research, uh, from the briefs and such are, they're ungated.

So you can go there right now and pull it right off the site. Um, I do have some contact information there. If you wanna reach out to me directly and talk to me, That would be great.

Yeah. Hey, man, thank you for coming on. Thank You.

It's been great. Futurum group. I don't wanna mess up your last name.

Paul Nacho. Wadi. You get it?

Got it. Paul nti here on Textron tv. We're coming back to you live from Paris in just another minute.

Stay tuned as we wrap up day three.