How CIOs are Managing the Cost of DevOps | DevOps Dialogues: Insights & Innovations – EP12

Hello, and welcome to this episode of DevOps Dialogues. My name is Paul nti and I'm the practice lead for the app dev practice at the Futureum Group. Today I'm joined by Alan and Steven Allen, would you like to introduce yourself?

Sure. com, security Boulevard, cloud native. Now, a bunch of other places, techron TV as well.

And hopefully, uh, well joining up with the RUM group shortly. Well, great to have you here. Great to have you on the session today.

And Steven. Yep. So I'm the founder of, uh, the tech Field Day event, including app Dev Field Day, which we are enjoying here, uh, as of this recording.

And, uh, so I've been focusing on a lot of areas of, uh, it more coming from the operations and infrastructure side of things, but of course, DevOps is real. And, uh, I can't wait to talk about how we make all this real. So it really is an exciting time in the industry, right?

We see that there's a lot of change happening, a lot of initiatives happening, and money is being spent, right? I mean, we see this, uh, it's, it's, it's about the evolution of, of new initiatives happening. Um, when I talk to folks in my practice, I often talk about the context of past, present, and future applications that, that are heritage applications that are looking to be modernized into maybe containerization and microservices, and maybe even thinking about technologies like serverless or, uh, uh, web assembly.

But all of this comes in with, uh, the context of somebody has to have a budget that has to buy this, whether it's the tech stack. Uh, in my, in a previous episode, we talked about people, process and technology, and that has a lot to play with it as well. Um, historically, we, we thought about, uh, at least I thought about the, uh, DevOps persona and the senior IT leaderships as the maybe the user and the buyer, right?

Historically. Um, but what we're seeing in our research we're seeing that recently did a persona study that showed that there's a pivot in the spending of, of it senior leadership's money. So the money's still with the IT senior leadership, but the spending of that money is occurring at the, what I would call the OT level, the, the, the DevOps teams, the, the SREs, the, uh, platform engineering teams.

Alan, let's start with you. What are your thoughts on this? So, I, I think that was true until about a year and a half ago.

Within the last year, year and a half, I think we're seeing a return to sanity. Normalcy or before this, right? I, I think what we went through with the rise of DevOps and, and, and DevOps maybe was the tip of the spear, but we also saw the rise of DevOps, the foundational era of open source software and open source becoming, you know, dominant, 75% of all the code and apps being part of open source.

And, you know, and Covid actually played a role into this, right? Because we had all of these remote teams and, and how we were developing our software, the rise of the software factory. And an interesting thing happened with this rise of the software factory tool vendors, IT tool vendors, DevOps tool vendors, got into the business of what I call delivering delight.

They just wanted to get their software into the hands of the people who were going to use it. Mm-Hmm. Because those people were in such high demand that they, they were getting as much money ver in, in essence, as they wanted.

So money no longer was the key ingredient in why I went to work somewhere. I want to know who am I working with, what am my work conditions? What tools am I gonna use?

I want to have a say in what tools I'm going to use. Yeah. So now that swarthy enterprise sales guy who carries a bag and takes the CIO out for three martinis, he doesn't do so well with this developer.

Hmm. Software. Two vendors realize this.

They, they hired dere people, developer relations people. They had, they wore funny hats or had purple hair and all kinds, you know, different things to stand out. And they befriended the developer 'cause they talked that language and they made sure the developer can get the software in their hands.

And then they basically helped that the software delivered delight. Yeah. And the buying cycle became very much a bottom up.

That developer would then go to his manager who would then go to the VP or director who would then go eventually all the way up to the CIO. This tool really works. It's what we need.

Yeah. It's what I want to work on. Yeah.

That makes, that makes sense. And I think to back that up with some research, what we see in a recent study of 870 respondents global study that we did at the future, I, uh, intelligence, uh, team, we did this study that showed that there was, um, uh, organizations have a desire, 24% of respondents actually have a desire to release code on an hourly basis, but yet only 8% are able to do so. And I That sounds aggressive, but when you think about, uh, you know, aggressive, Aggressive, let's, Steven, when was the last time you worked at an organization that was releasing code hourly?

Uh, let's say never. Never. Yeah, But, but here's the thing though.

You have a cyber attack, what do you do? Yeah. Well, you've gotta respond right away.

Yes. But, but let's, you know, so when we talk about John Spars, you know, seminal presentation, 10 x sort, 10 XA day software, I mean, that was aspirational. When you are a company that released code once or twice a year, and you move to monthly releases.

Yep. That's huge. That's a revolution.

Yeah, it's A revolution. Yep. Yeah.

Monthly to weekly. Crazy. Yeah.

Yep. You're talking about hourly. Yeah.

We saw, we've seen this trend. I've been doing this trending report for since 22, 23. We've seen the trend moving closer and closer to that hourly release.

Yep. So, So, but you just said it, 8% of organizations able to Do so, You got a long way to move. Absolutely.

And it's not all applications should do that. Right? But it, but I guess Steven, when we think about it in the context of the spend, you know, what the, the Alan was talking about the tool sets that are being used, right?

And why it's important. Um, so senior IT leaders are, are going back to their organization going, Hey, we need to get this done. Say we do wanna release code on an hourly basis.

How do we get it done? Are they the ones that senior IT leaders in there rolling up their sleeves, figuring it out? Or are they going to their, their experts on their teams going, Hey, this is what we need to do.

What are your thoughts? Yeah, I, I definitely think that there's a normalization happening here. And to Alan's point, I think that there was definitely too far of a pivot, uh, too much of an embrace of these ideas.

And I think that a lot of the folks that are involved, even those people that are kind of pushing, you know, for their favorites and everything, I think a lot of those people would agree as well that that things just got a little too crazy. And what we're seeing now is sort of a fallback to earth from that. And so, to your point, you said like, bottom up sales.

Mm-Hmm. I, I have been talking about a middle up sale where, you know, and I think this is reflected in this trend toward platform engineering, for example, that we talked about on the, uh, the tech field day podcast. Um, platform engineering to me is a signal that everyone realizes that it's completely untenable for every developer to use whatever the heck he wants to make his own thing work, because there's no chance that he'll ever be able to put together a, um, a uniform supportable, scalable stack if, if, if that developer's expected to know everything, they have to have kind of a return to standards or return to experts.

And I think that that's what's happening on the, uh, application tool set side as well. I think it's happening everywhere where there's this idea that, wait a second, we pivoted too far toward sort of the cult of making the developers happy, and now we have to make This the also predator more. That was the cult, the cult of the developer Alpha Predator.

Yeah. And now we have to make things more stable, more predictable, more repeatable. Gotta get some maturity in here.

Yeah. And, and so, so there is that, uh, this ebb and flow, right? So if you look at the last 10, 15 years, maybe even more, uh, there was, uh, it owned everything.

They stood up their own infrastructure. They built, built everything. Mm-Hmm.

Then all the budget was pulled out of it and said, okay, we're gonna push that into lines of businesses, which we realized that was a mistake, right? And then there was all the shadow IT that was occurring and all these things that were happening, and there was no governance. It was kind of the wild west.

So then there was this re internalization of it, which is kind of where you're going, right? Internalization. But it morphed and changed to be more of a, a business manager understanding the relationship between these SaaS vendors Had to, Had to, right.

'cause it 'cause times have changed. Well, no, the problem was 15 years ago, most ex non-IT executives thought of it as a cost center. Now they think of it as a profit generator, right?

Because good. It makes you more profitable. It's not a cost center.

So it's, it's a, it's a different mindset. Now. You're a money maker, you're not a sinkhole.

And Yeah. That, and that really changed the game Here. And, and it was a sinkhole because, so that's, you know, 20 years ago, I was one of the people in enterprise it, and I'll tell you, we were entirely, it, it, the, the stories are all true.

We were disconnected completely from the developers. We were disconnected from the lines of business. We didn't care.

We're gonna build our infrastructure the way we want it because that's how we roll. And obviously that was wrong too. Right?

And what we're seeing actually, this to both your points, uh, we've learned, right? And what we're, what we're seeing in our research is we see that, um, that this observability study, I just, I just fielded came back, uh, we see that observability practices are young and, and their, uh, maturity within organizations, but they're starting with internal applications first to, to make sure that things are working. So we see that actually 25% of of these applications are being checked with observability practices internally before they rule them out externally.

Yeah. But my question is, what about the other 75%? Correct.

Right. Well, that's maturity, right? So it's, it's Like this, when did, when does it go like that?

I think the other thing we're seeing, and this is what I, right. I said in the beginning, it was true up until maybe a year, a year and a half ago, is we've seen the pendulum swing back and the rabbit now has the gun again. Mm-Hmm.

The developers are not quite the alpha predators. They, they're not invincible. There are plenty of developers out there.

I could hire more. And if you don't want to come work in my office and you wanna work at home, that's fine. I'll find someone who will.

And so, but the leverage is changing and the tool vendors are seeing this. They're getting rid of the Dev re people, they're cutting back on Dev Re and they're hiring salespeople again, enterprise salespeople who are selling to IT, leadership. And I, so, you know, you live long enough, it goes round and round and round this circle.

But I think there's something else happening here too that's really interesting on the tool and software side. And that is, I'm seeing a lot more tools and software that have real value, like real business value to the customers, and are really targeted at challenging problems that need solving and can be solved and can make money. And there again, I think that we saw an over shift on the vendor side to vendors who are just embracing open source, and we're not gonna worry about making money, and we're not gonna worry about solving problems.

We're gonna do cool things. We're gonna do big things, exciting things, you know, observability and, and telemetry doesn't sound all that sexy, but it's, it's required a big necessary thing that customers can get value from. Yeah.

Why do you think there's so many companies that are making really interesting products in this space? That's why. Yeah, absolutely.

And you know, I, Alan, I do want to touch a little bit actually. I've been, uh, I've been attending Cobe Con for, for many years, and actually we are looking at doing our app Dev Field Day and at Cobe Con in, uh, in, in the fall, salt Lake City, in Salt Lake City. So we're really excited about that.

But one of the things I've noticed in a pivot with the, with the audience is the sports coats versus hoodies. And there's, uh, far more sports coats now. Well, the three of us, I don't wear sports coats that often.

I I, I, I know. I mean, do I, but, but, but about 25% of the audience in Paris was wearing sports coats. Um, and Detroit was the same way.

Just not, I, I'll tell you, Chicago was after Detroit, wasn't it? No, Chicago, I'm sorry. Right?

It was Chicago and Paris more than how they were dressed is, what were they talking about? Right? Yep.

They were talking about ops. Mm-Hmm. When I first used, and I've been going to K Con since there was a K Con and it was primarily a developer audience initially.

Yep. It was. Now, I would say at best it's 50 50 if not more ops to developers.

C**n Cloud native is about ops, it's about platform engineering, it's about SRE, it's about ops. com, right? I'm a DevOps fan, but for too many people in DevOps Shift Left became shifted onto the developer.

And there's only so many straws you could put on that camel's back until it breaks. And, and so again, a little bit of a return to normalcy, like Steven said, with Platform Engineering, the SREs, it's not that what they're doing is so radically revolutionary. No, it's the same ops stuff we were doing.

Right. And hopefully what they're doing is learning that ops and devs really do need to get along the, the, the thing about the DevOps concept that I love is because, is that ultimately it could mean that we could finally stop the battle, stop the war, and start working together. Absolutely.

Including security. Let me just make a quick, absolute sec. Absolutely, absolutely.

Security. But, but thank you both. This has been great.

I know we only touched on, you know, a couple of the topics here. It was really meant to kinda understand where the money's coming from and where the, how do you, how to overcome in your own organizations, how you might think about spending those dollars that need to address those business initiatives. I wanna thank you both for your time and, and your perspectives.

Thank you. I think it's been great. And, uh, I wanna thank the audience for attending our session today.

com and find out more about our research there. Thank you very much. Have a good day.