Exploring nOps: Revolutionizing Cloud Cost Management | KubeCon SLC 2024

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

We're here at Q Con in Salt Lake City. Uh, wrapping up, we we're not wrapping up. We still have a few hours left here on day two.

I wanna introduce you to our next guest. His name is James Wilson. Hi.

Okay. Oh, everybody out there have To make sure I got the name right. I'm only kidding.

You did get it right. It's a very haunted name. Was it could have been John Smith.

Right? It's one of The easier names I've had to pronounce this week. But anyway, James is with a company called an ops.

You may not have heard of them, but we're gonna hopefully get you wise to them in a minute. First James, welcome. Thanks for coming in.

Before we get onto N Ops, let's hear a little bit about Common. James Wilson. Common.

James Wilson. Well, yeah. Common name.

Um, let's see. So I've been in this tech industry for about 25 years now, um, and sort of followed the path in, into the, the cloud native and, and public cloud world. Uh, let's see.

Most of it, I've been in four startups. This is the fourth one. This is the fourth.

Uh, yeah. So I, I started out building private clouds for the telcos over in Seattle for, for many years, uh, for a company called New Bay, which Sure. Which ultimately was acquired by rim.

Yeah. Yeah. So worked for Blackberry for a while.

Mm. Um, and then started riding on the wall there. So moved to the Bay Area, joined a startup called Aspera.

Okay. Wanted to get deep into r and d, and, you know, through those experiences helped companies like Netflix move into, uh, AWS really. So, yeah.

So then we, we were firmly in the public cloud. Um, you know, we got involved with Cloud Native and, and really Docker and Kubernetes and Docker compose, like really just so that we, we could start testing our software. We, it became the best way for us to manage configurations and set up all of these different environments still to deliver on-prem software.

And somewhere along the way we realized that we could start using this technology to build an as a service model. Um, so over there we built a, a product called desper on Cloud, and then one day woke up and, and were, uh, working for IBM. Um, and so I was with that, that technology for about 10 years and decided that, that it was time for me to get on to, to the next challenge, which, which led me to, to N Ops.

Excellent. Let's hear the N Ops, how you got it. Yeah.

Well, you know what really compelled me about N Ops was that, uh, we had a lot, we made a lot of mistakes. Building in the public cloud, who does it right? Security and compliance mistakes, like going back and like figuring out, there's something called SOC two mm-hmm.

Um, and then once we were acquired by, by a larger company, um, all of a sudden ran into a finance team who had questions, uh, about our bill. And so what, what I learned is that for most engineers, finops is punitive, right? Yeah, it is.

It's like real reliability. That's a KPI, um, performance. That's a KPI, but, but we're only dealing with, with cost and efficiency when somebody in the finance team is mad at us and coming after us with a big hammer.

Um, and so n os uh, you know, I, I met our founder, jt, and, and I really felt like there was an opportunity to get out there and build capabilities that would've been really meaningful to my own team in, in our early experiences in building in the public cloud. Sure. So, uh, N Ops is a fully automated finops platform.

Uh, we started really just in the cloud cost management game, and I think we always knew that that was gonna lead us into the Kubernetes world. Sure. Um, because today, and I have a lot of data, data Point, it's the compute stack.

It is the compute stack. And, and in fact, I, we have a tremendous amount of data on this. I, I think any day it could be the inflection point where Kubernetes is driving more compute than traditional Autoscaling group.

If it's not already, it's close, It's close. Uh, you know, we manage about $2 billion worth of a AWS spend. And what I can tell you is just from our data, it's very close.

And so I'm gonna talk a little about ops Yeah. Because I have my own history here. Right?

Right. And so I first met JT and at a company called nClouds. NClouds.

Right. And Randy, my friend Randy was CMO there, We worked at IBM together. Right.

Okay. So, you know, Randy, Randy was, Randy was, yeah. sli or anyway, um, so they, and, and nClouds was one of, was like the go-to partner, I think in northern California.

Right. For everyone who was migrating onto AWS Onto AWS Yeah. And so, a tremendous amount of know-how knowledge and experience real world, right.

And moving to the cloud led to n Ops almost being like spun out of Yes. Yeah. Yeah.

So I met JT when he made the decision that he wanted to build a product company. And, you know, based on my experience in his experience, I think I was the employee number one. Really?

That didn't work for, for N Clouds cloud. Yeah. You're the first true ops.

The True for the first Yeah. The employee number one. Yeah.

There you go. Alright. Now, at some point though, you know, you guys embrace the open source world and the cloud native world because you see it, you know, it, it's, it's growing and it's there.

But particularly Carpenter Particularly, right? So, you know, o obviously we we're in both the price optimization game and the, the cloud efficiency game. And so when we started looking at auto scaling, um, and, and talking about how we were gonna build a product that, that intelligently provisioned compute for workloads that, that did things like make sure that, that your RIS and savings plans are, are fully utilized, but also, uh, de divert excess capacity over to, to spot capacity.

Um, carpenter was just early in Alpha and, and we got together and we said, we think that there, there's a lot of promise and, and the promise is really because Carpenter, it challenges conventional wisdom. Right. And, and, you know, we always think about Autoscaling in terms of a bin packing problem.

And I guess Amazon would, you know, uh, that they were the, the leaders in the idea that maybe bin selection is more important than bin packing. Right. And so when we saw that, that that conventional wisdom was being disrupted, uh, we decided we wanted to get involved in the community.

And so we built our integration, our first compute co-pilot integration on top of Carpenter and have been, I, I guess just carpenter fans and part of the Carpenter community ever since. Very cool, man. So we're here at CubeCon.

Yeah. What are you guys doing with Carpenter? With finops?

What do you see around The so forth? Yeah, so, um, one thing that we're doing, because, you know, we've helped hundreds of companies now adopt Carpenter and integrate with our platform. And so, uh, we we're launching like the first fully hosted carpenter lab series, um, which is, uh, free and available to, to everyone in the CNCF community.

Uh, and, and it's cool. We, we partner with, with a technology who's actually here at the show called Instruct. Sure.

Uh, really cool lab technology. Mm-Hmm. And we got together with them and said, what we want is a really dynamic environment where engineers can get their hands on, like just instantly be in a carpenter environment and start doing experiments.

You know what? It's funny you say that. So we, we do about, I don't know, 300 webinars a year.

Right. And what we have found from our webinar attendees is they don't want a talking head and slides. They like a round table to hear the discussion, but what they really like is they wanna get their hands on software.

They, they like product demos. Not show me a slide of what product Looks like. No, no, no.

I, I agree. I mean, real hands on dirt under the fingernails. Can I get in a sandbox and actually use this product and see how it works, right.

So that when I come outta there, I know how it works and is it good for me? Is it not, you Can veer from the path. That's what I really like about this is, is, you know, you get shell access into the container and you, I mean, if you wanna go off script and try a little bit of a different experiment or screw something up or, or just go a little further, it, it's fully open for, for you to be able to, to do that and, and test out your theories.

So, but I, I wanna plug them, instruct it's a cool technology and, and we're gonna be now launching three more in the, in the series, so, so that we get you to the, the carpenter ninja level. Got it. Um, doing badges and, and all of that stuff.

So, so people can come out of it and really feel like they, they got something out of it. And what's good for the community is good for us. Absolutely.

io or, or that's N-O-P-S-N-O-P-S Or, or you can swing by my talk tomorrow. We're doing a panel on the state of Kubernetes optimization and the role of ai. Oh, that's cool.

And we've got, I think, some of the coolest thought leaders on, on the panel. We've got Katie Gaji. Mm-Hmm.

Got Jasmine James. We've got I love Jasmine. Yeah.

Tell her I said hi. I I will, uh, Chu, uh, who's, uh, just graduated with his PhD now working over at Microsoft. And then we've got Josh Cipher, who's, who's a senior engineer over at Sonos.

Very cool. I love we're gonna talking Sonos, Talking to people who are I know, I know. Full brand.

A lot of Sonos In the house. Yeah, me too. Um, and so, you know, we're gonna be talking to people who are really out there doing it and talking about the tools that they're using and really kind of talking about some advanced topics, multidimensional autoscaling and, and where, at what point do we have an inflection point where AI is taking control and, and making the decisions about scaling.

So swing by it 11 o'clock tomorrow. Uh, and we're also just having a great time. Absolutely.

James, thanks for coming over. Hey. Yeah, thanks for having, that was easy.

You were worried. What were you gonna talk about? Well, Lack I, I know.

Well, I never am gonna laugh for words. Yeah. So, alright.

io. Check it out. We are live here in Salt Lake City.

We're gonna be back. We've got, we've got a lot coming up, so stay tuned.