Platform Engineering in 2024, CI/CD Frameworks and Securing Source to Production – DevOps Chats Podcast EP6
Alan and Mitch discuss whether platform engineering will be “the game changer” some predict will happen in 2024, the evolution of CI/CD frameworks (ArgoCD, Tekton), and Mitch’s interview with Cole Kennedy, TestifySec CEO, about securing software from source to production.
Transcript
Hi, everyone. I'm Alan Shimmel, and this is Mitch Ashley, and you're listening to DevOps Chats. All right.
To dumb. Hey, Mitch. Welcome.
How are you, man? Man, it, it is, it is hopping. I'm, I'm doing real well.
It's, we're, you know, getting into conference season, it's no longer the new year. Um, the good news is, you know, some companies are reporting public companies anyway in tech have reported strong fourth quarters Mm-Hmm. All the doom and gloom and the layoffs and everything.
You know, I I, I don't want to be a conspiracy theorist who thinks this is all jacked up to get to have these companies to do 25% more with 25% less people. But the numbers tell a different story a little bit. Well, that means people are people Sure.
Creating software, still building apps, need to get things done for their companies. So, you know, that, that fuels our industry. So, good news is there's a lot of work to be done and probably more coming.
So, Mitchell, for today's DevOps chat, I, I picked DevOps Chats. I picked a few stories and interviews, but there's a theme behind it. And the theme behind it is something you are heading up a tech strong research, which is our DevOps Next Mm-Hmm.
Uh, report, which will be out probably i, I guess by RSA timeframe may. Um, and what it really looks at is, you know, so a lot of the different facets of what's driving the DevOps movement today, not from 12 years ago, but today and tomorrow. And so a lot of these topics I, that I selected for today are around these, you know, fast movers within the DevOps movement itself.
com, and it was a good article. It actually got a lot of views as things as a, it's a pretty trendy, uh, subject and it's platform engineering, the 2024 game changer in tech. Now, I, I, and by the way, that, that article is by Yossi Carmen, um, I don't know if it's a game changer in tech, but certainly platform engineering as a viable, uh, player within the software development, life, life cycle and software, uh, you know, development pipelines is, is real and growing and becoming ever more important.
It, it is, and it follows with kind of the platform theme you and I have kind of seen and know about, you've heard it from a lot of our customers and executives that you've talked, talked about, of thinking more of DevOps as a platform, not DevOps as a tool chain. Right. As a series of, you know, independent, unique standalone tools.
Mm-Hmm. But what it, it's interesting because it seems like you can't throw a stone without either hitting something that says AI or platform engineering. Those tend to be pretty hot topics, which I suppose is, you know, why, why claiming 2020 or is the year of platform engineering?
I, I think what it's about is an extension is at least in part, an extension of the whole developer productivity wave that kicked off with the early kind of tough times starting in, in the tech industry a couple years ago, a year and a half ago, and a big focus on developer productivity. Well, guess what, uh, platform engineering added to its kind of mantra or its, its discipline, uh, in internal developer portals, things like that. Now, more platforms than standardizing.
So it, it is, I dunno if it's 25% less, get 25% more like you mentioned earlier, but, but it is getting more from the same or less resources. Yep. I, I think also specific, first of all, look, I wouldn't put platform engineering anywhere near ai.
No, no, no. Right? I mean, AI didn't either.
Yeah. ai, civilization changing. Yeah.
Platform engineer is another incremental thing. But, um, in my mind, I think one of the reasons we've seen platform engineering gain the popularity it is, is because no one was really familiar, comfortable with the idea of a DevOps engineer. What exactly was it?
DevOps engineer. Good point. Yeah.
And, you know, it was kind of a nebulous, but yet, you know, there were plenty of jobs for DevOps engineers. Yep. com, a DevOps engineer, the requirements were you had to know Chef Puppet and Ansible.
Mm-Hmm. Maybe a little Jenkins that hardly qualifies as a DevOps engineer today. Yeah.
Nothing easy. Um, that being said, a lot of people always said that wasn't a real title, it wasn't a real job. And I think platform engineer is a better handle to at least express some of which, what the DevOps engineers were doing.
Make sense? Yeah. It, it does.
And I think it's got a little more structure around it kind of being an engineering discipline is about creating standard platforms and tools and processes and, you know, maybe subsuming some of the DevOps engineer role, maybe not, just depends on the organization infrastructure's code. This article that you, you referenced does a nice job of kind of categorizing what it is in, in the focus of platform engineering. So I, to be honest with you, I never, I never understood the DevOps is not a job title.
Anything's a job title in my world. I mean, you know, what do you want your title? If someone's willing to pay you, it's a job.
Yes. Yeah. I'll, I'll, I'll wear whatever label that helps us.
Right. And so it, it, that was one of those philosophical arguments, which I didn't have a dog in that hunt, so didn't really matter to me. Okay.
com article if we can. But it, it's actually, again, it fits this theme I've been working on, right. Which is the part of DevOps Next.
com and it's called Driving the DevOps Evolution, Argo cd, teon and Seamless Migrations. And it's by Siva Guhan. Right.
com, we'll have the link in the notes. But, you know, Mitchell, I, I think what, to me, what this was about is clearly, you know, when we talk about CICD mm-Hmm. We, we, we've seen an evolution of the CIC d tools and the platforms and everything else, right?
com, Jenkins was probably 90 plus percent. I bet you of the CICD uh, market. And, you know, then with the rise of Cloud native, I think that's had a huge effect on CICD.
And we saw, well, we saw Jenkins X, which I don't believe is still being maintained or is still a viable, well, viability is a whole different word. I don't think it's officially maintained anymore. Mm.
Um, but, you know, we saw a lot of commercial CICD products, probably one of the most successful being like harness Mm-Hmm. But then specifically we, we saw CICD in the age of GI ops. Right, right, right.
Because, you know, be, be, and that Argo CD seems to be the, the big GI ops CICD tool. Mm-Hmm. It's an open source tool run by Linux, uh, CNCF.
Well, you know, the people at Codefresh had a lot to do with it, my friend Dan. Absolutely. Um, but Codefresh just this week was announced, right?
Was acquired by Octopus Deploy. Mm-Hmm mm-Hmm. Which leads me to believe maybe they weren't monetizing the open source Argo CD project project as much as they had hoped or what have you.
Or could a lot of companies have struggled with that model. Right. It's not, yeah.
It's not an automatic, especially, especially times. Yeah. You know, in these economic times, VCs are a little impatient.
Mm-Hmm. Um, but in any event, you know, there's also Spinnaker has, has come along, or Spinnaker was always there, but it's gotten much better, much more wieldy and usable. You don't have to be Netflix and Google to use it.
Right. Um, so you've, you've got spiny, you've got Harness, you've got Argo, you've got Teton. Mm-Hmm.
Right. And a lot of people are saying, ultimately that's where you should move. And Kubernetes, there's a lot in this article about, you know, moving or adopting Teton and Mm-Hmm.
Good for them. Um, but what it really points to me is DevOps next Mm-Hmm. Right.
DevOps isn't going anywhere. CICD is certainly not going anywhere, but the tools we're using it, the platforms we deploy and develop on, they're changing. They are.
And, and so the, the CICD tools are keeping up and, you know, it's all good. We, we of course, do the CD Foundation show with the folks at the Continuous Delivery Foundation. Mm-Hmm.
We talk a lot about these things, but, um, again, it's just, you know, DevOps isn't going anywhere in terms of leaving the scene. It is just, you know, some of the names and faces are being changed to protect the innocent or guilty. It's the evolution of DevOps.
Right? Yeah. The, the good news is, one of the pieces of good news is it's not static.
It's not stuck in what we thought of DevOps five years ago. And, you know, five years ago we kind of matured the things that, what we were doing. Yes.
CICD got better and we extended into more parts of the, the workflow, the, the DevOps pipeline. And now we seem to see sort of a, I i, I don't want to label it next gen, but it is kind of that thinking about it, well, how do we build our process around, you know, a Kubernetes like kind of platform cloud native kind of platform? How do we think about this as multiple pipelines and workflow, not just code arriving in the repository and spin up a, a build process and toss it into an environment.
I, I don't mean to minimize what it does, but a lot more sophisticated things are going on now where multiple applications have multiple threads and pipelines of code and coordination and elements that are dependent upon each other. Our world's more complex. So I think the, the GI ops and the, the, uh, the CICD tools are getting more sophisticated.
They have to Yeah. Keep pace. Absolutely.
But good. I, it's not a bad thing. I, I not at all.
I encourage it and welcome it. Yeah. Absolutely.
Um, so again, so those are my two kind of DevOps next themes, Mitch. Mm-Hmm. The next thing I wanted to bring, actually, the next two things really are you're closely involved with.
One is a a, a text drunk TV interview you did. Right. And, um, let me just make sure we get it here.
Securing source to production, and you spoke to with testified sex, Cole Kennedy. Yeah. Coles, the co-founder in CEO there.
Mm-Hmm. And one of the, one of the most fun interviews to do, I know you enjoy this too, is talking to entrepreneurs. Yeah.
And kinda knowing where they are in that cycle of starting building, growing their company, maturing. And, uh, Cole's another great example of, well, I was doing this job and I really had a tough time of getting all the information together, all the attestations together to, to validate that where components came from and our things, really the software we think it is that's going into our supply chain. And so he started, uh, testify SEC with another co-founder.
And, uh, really thinking about, he, he calls it an observer role. You kind of need this observer role, not that it's a person, it would be in, in, you know, physical world, but this observer w world that's kind of collecting and managing as you're going through your, your workflows, the, the, the artifacts, the assets, the attestations that will help you in the end validate here's what our software is and here's how we can prove it. Rather than here's a, here's a mound of data from all of our DevOps tools and build tools and everything else.
Go extract outta that, you know, after the fact, some semblance of demonstrating artifacts and, and provenance and et cetera of, of, we did build it the way we said we were gonna build it. So it's, it's, which was a great, it's a great solution. We've always talked about kind of the data exhaust coming out of DevOps and CICD, let's use that data.
Now. He's, he's taking the testify organization, their team into the next, that next level. And they have some announcements coming up at Kon.
They're gonna be there. Is that a great time talking? It's, it's just fun to talk to fellow entrepreneurs.
Yeah. I, yeah. I I love the enthusiasm too, man, because, you know, they all, I just wanna tell you their story, right?
Yep. Um, yeah. But nevertheless, so that, that one's on Tech Drunk tv.
And again, we'll put the link to it in, in our notes. com around a report you did with Redis. I, I guess that came out late last year.
Yeah, just, uh, right towards the end of the year. Um, we were working with Redis on, you know, I, I, as a generalist, you kind of do a lot of different things. You know, I, I've done a little bit of security, well, a lot of security, some networking, some da I used to be a DBA database person and all that stuff.
And of course, you know, caching was kind of this one thing you did. You enabled caching like you do on a web server, right? You might, you know, flush the cache or whatever.
It's so much more sophisticated. There are different, uh, patterns of, uh, transactions and data, uh, acquisition coming out of not just databases, but other sources. Could be applications, could be a cloud native applications.
And it was a lot of fun working with the Reds people on this and kind of really seeing where the state of the art of it is and what's being applied in web applications, mobile applications, cloud native applications. So we wrote this report called the pulse meter report on caching transforming data application performance, and really dove into, so what are still are the challenges, because it isn't just a feature you turn on now, you need to use the right pattern based on the characteristics of that kind of flow or transaction in your application. And the other challenge is, you know, oftentimes you're using in, in a cache, you need to transform the data in a way that multiple sources kinda look similar.
So you can use that cache data, not one, one source looks this way and another source totally foreign way of representing the data. And that, you know, that's a data management problem, not A DBA or a developer problem, but they've got some really nice tools to help with that transformation. So the, the report, the articles about the report and some of the things that we learned, the report is free.
You can download it, uh, easy to get access to. Um, it's on devs, I think you download it. Um, there's a link in the article, and we'll include a link in the, in our or, uh, description to for both the article and how to download the report.
Beautiful. It was a nice piece of work on that one too. Mitch.
Thank that report. Thank you. I remember it.
Good stuff. Great to work with. And, uh, you know, it was some, some good work and you helped on that one too.
I remember you were, I remember I did some writing on it. Some feedback on it. Yeah.
Writing. Yep. Cool stuff.
So, Mitch, that's what we had, you know, in terms of, uh, references for today's DevOps. Chad, anything fun you wanna throw out there? What'd you think of Dune two?
Dude, dune two? You know what? I was hoping you'd bring it up.
Look, I would, first of all, I saw an interesting article. It's had the highest rating of any film in like, I forget how long. Really?
Like 94% of people thought it was an exceptionally great movie. Really? That's amazing.
That's hard to accomplish these days. But you gotta remember, I'm thinking the people who go to see this movie right away are Doon Headss. Oh, absolutely.
We've been waiting, you know, a long time for this fanatics. Uh, let's see how it plays in the gen pop. Right.
Um, but that being said, I mean, kudos, kudos to Dennis Villa and, and the rest of the team. Mm-Hmm. Because this got done during the writer's and actor strike and all that too.
Right? It delayed was delayed a little bit, delayed the fact it was supposed to be in November last year. Yeah.
Uh, but man, they, they killed it for me when he got up there in front of all the, and said to follow him, you know, in essence to Paradise. And he is Paul Mo aches. Mm-Hmm.
I was ready to throw on my still suit and go You were signing up. Yeah, I'm signing up. I'll kill those.
You you getting in your orna and flying over there. Go. The theater I was in those Oro were so loud, I felt like they were on top of me.
They were loud. Loud. Yeah.
Um, but yeah. What a great movie, man. I I enjoyed the hell out of that movie.
I cannot wait for, uh, uh, dune Messiah, I guess will be the third movie. Yep. Yep.
It, it's one of the, I'm trying to remember back Dune Messiah. Paul has disappeared. He's wandering, no one knows what happened to him.
Yeah. It's before the children of Dune. Right?
This is between Right. Then there's Children of Dune and then there's Dune Emperor is Mm-Hmm. I think the grandson is Lido.
Yeah, that sounds right. Maybe the son is Lido. I don't remember.
I'll have to go back and reread. It's a good r reason to make you go reread it, but man, that was, that was an amazing couple of things though. Number one, I think you and I both said that in the book, you know, where it diverged from the book, in the book, Alia, uh, Leo, whatever her, the younger, his younger sister was born.
Mm-Hmm. Spoiling blur. Yes.
By the, by the revolution here. And then secondly, we didn't really get a lot into like the Spacing Guild. Mm-Hmm.
And they definitely showed Benny Jesuit. Mm-Hmm mm-Hmm. And Reman and the, you know, the, the LANsat houses, but the Spacing Guild and how these, and the role spice plays in that holding of space for being able to do space, interplanetary, interal space, travel, light speed kind of space.
Yeah. So it's, um, so there's even more, if they wanted to peel back some more, there's more. I, I loved it though, man.
Highly, highly, highly recommend. You know, I, I felt like a little kid 'cause I've just kind of like shaken or jiggling my feet, or my wife is like, are you okay? Would you settle down?
It's my wife's like, I can't believe you're making me watch this. Because she's not a dune freak though. I've made her watch every Dune movie and show and everything ever.
My family is into it. So that's nice. Yeah.
But it brings up an interesting point. If you haven't read the books, is it worthwhile to go see the movie? Well, you know, my wife's not read the book.
She's read a little bit of it. Um, and she really enjoyed it. She, she thought it was a lot of fun.
She, she's not a sci-fi nut like me. Yeah. But she does enjoy, you know, sci-fi and uh, and fantasy and other genres.
Um, I think you could follow it along. I think you'd need to watch the first movie for the second one to really make sense. You definitely need to watch the first for the second.
Yes. Yeah. That might be tough.
Stepping into Dune two and going like, so what now? What happened? Yeah.
How did we get here? What is this? Right.
Be the origin story. Right. And this is the kinda second half of it.
No doubt. No doubt. It was good.
Um, so Mitch, I'm just disappointed we didn't get to see it together. We've seen so many other Well, that's why on the, I was trying to, I was trying to make that work, but, uh, couldn't make it happen. Um, anyway, Mitch, let's, let's call it a wrap on DevOps chats.
You bet. Thumbs up for Dune two thumbs up and, uh, absolutely. So I hope you've enjoyed this.
This is Alan Shimel. And Mitch Ashley. You've just listened to DevOps Chats.