Jesse Robbins interview “On the Money”: Opscode, DevOps, & more (transcript)

A few weeks ago I was invited to appear on Mike Adams’ “On The Money”, a business radio show on KKNW 1150.  It was a fun conversation.  A transcript of my segment is below:

Mike Adams: Welcome back to “On the Money.” You’re listening to Mike Adams of Adams Financial Concepts. I’m here with you Friday afternoons from 4:00 until 5:00 here on KKNW Alternative Talk Radio. Yes, yes, get off of my cloud. Or get into my cloud. Depends upon computing networks, doesn’t it? There’s a lot of talk about the cloud. And one of the focuses of this program is to bring you, the listener, greater insight and understanding of what’s happening in the economy.

And one of the big things that’s talked about is, in fact, the cloud. Almost every week I have a guest, the CEO of a private company, and not only is picking stocks an art instead of a science, and not only is it learning from the past, but to some extent, that’s the story of my guest today as well.

On the forefront of technology, and yet applying the lessons of the past, my guest today is the CEO of a company called Opscode: Jesse Robbins. Welcome to the program, Jesse.

Jesse Robbins: Thanks for having me.

Mike: So Jesse, why don’t we start with your background, before you got into Opscode?

Jesse: Sure. So I like to say that I got a start in high school, actually, running Internet service providers, back when that wasn’t a ubiquitous thing. And then, in the 2000s, or as the dot-com boom began, I got fed up, and I actually left and became a firefighter. An actual firefighter. People often ask, “You mean like a real one, or like a computer one?” An actual firefighter. And then I moved to Seattle to actually join the Seattle Fire Department, or their very long testing process, and I needed to take a job along the way. Amazon ended up calling me back first, and so I became an engineer for Amazon. And ultimately I ended up owning website availability for every property that bore the Amazon name. Sort of a funny title, “Master of Disaster,” which I guess I was infamous for more than anything else.

Mike: [laughs]

Jesse: I left Amazon. I founded a conference called the Velocity Web Performance and Operations Conference, which I’m still the co-chair of, which is actually coming up in June. And then I and several others co-founded Opscode to bring something we call infrastructure automation to the masses.

Mike: So infrastructure automation. What is infrastructure automation?

Jesse: The simplest answer is there are common tasks required to run and operate IT, ranging from small to medium-size businesses, all the way up to big websites like Amazon and Facebook and Google. And the common tasks that you have to do in order to run your business, you want to automate away. And so infrastructure automation is what the developers and systems engineers do to allow them to focus on actually creating business value and not just repeating the same things over and over and over again.

Mike: And that’s what you do at Opscode.

Jesse: Yeah. Opscode has a product called Chef, which is an open-source framework that software developers and systems engineers and administrators use to automate away tasks that would take them hours and make it take seconds, with no interruption at all.

Mike: So when you say “open source,” can you explain that?

Jesse: Yeah. Open source, well, it’s been around for quite some time, but began to take off in the mid-’90s. If you’ve heard of the Linux operating system, which is a very common operating system used for websites and lots of different systems, it is a concept where people come together and share software code. So people don’t specifically own it, and you don’t charge for it. It’s free software in both it is not something that you typically pay directly for, and it is free as in you’re able to modify it and improve it over time.Chef is a project like that. There are currently 350 individuals and 70 companies, including some pretty big names, that contribute to make Chef better and stronger every day.

Mike: And this is an ongoing process, then. It’s an evolving process.

Jesse: Yes.

Mike: It’s not something that you say, “OK, we have this program,” and then two years later you do an upgrade, and two years after that you do another upgrade, and then…

Jesse: Yeah, it’s continuous improvement. And usually, it’s developers scratching an itch. They say, “I wish that this thing did that.” And then they write the code required to do it, they push it into the project, and then Chef does that for everybody.

Mike: That explains the open source. But explain what Chef actually does.

Jesse: Let me give an example. First of all, it’s a framework, and so by itself it doesn’t really do anything. It requires a software developer to actually use it to solve a problem. As an example that we actually spoke about earlier, one group, a major biotech, used Chef, along with some other software, to build a supercomputer that they assembled out of 1,250 servers, 10,000 cores, and configure and run it for eight hours and then turn it back off, and all those machines went back and did other things.

Mike: When you say they were using 10,000 cores, or 1,200 servers, what you’re saying is they were using computers that did not belong to them.

Jesse: Correct. They used Amazon’s EC2 service, a service that allows you to rent machines by the hour. For eight hours, they were renting those servers from Amazon. And rather than having to, when you have to upgrade a machine, you put in a CD and you’re spending all that time installing it and all that kind of frustrating configuration and you’re putting your applications on, well, they do all of that stuff with Chef. So they don’t have to spend any time at all doing that manually, it just happens for them in a minute. And so they could do that across thousands of machines if they chose to. And they were able to run a very complex protein-folding solution, which they wouldn’t have been able to do otherwise, for a couple of thousand dollars in eight hours, as if it were nothing.

Mike: That’s fascinating. And when we talk about that, we’re talking about something in the cloud.

Jesse: Mm-hmm.

Mike: So let’s talk a little bit about the cloud.

Jesse: Sure. What is it?

Mike: Yeah.

Jesse: So, cloud computing is, really, an industry shift, and it is a shift from running things that are customized and local to your business, to running things that are commodity and centralized, typically running somewhere else over the Internet. That might be a server, like Amazon’s EC2 or Rackspace’s cloud-computing project. It might be a service, like Salesforce, which many people use. And it might be another kind of thing, like a place to store files, like or Dropbox. And the difference is that rather than doing all of that configuration, that management, paying for those resources for your business directly and keeping them locally, you’re instead achieving an economy of scale, running somewhere else, and typically with far less customization available.So you gain benefits of speed and price, hopefully, with standardization, but you lose the ability to customize it, unless you’re writing your own service and using something like our tool set.

Mike: OK. What kind of tools, then, are you providing for the IT people?

Jesse: We provide a variety of tools. Talking about my background as a systems engineer, when I ran lots of servers, you get the typical tasks: “Please install this application. Please configure this application. Please install a new server. Please get rid of this old one.” People usually don’t say, “Please get rid of this old one,” but you end up running out of space. So with Opscode, we provide something called “cookbooks” that work with Chef. We have well over 200 of them now, which are, again, either generated by us and shared with the community or generated by other individuals, who share them because we don’t all want to have to solve the same problem over and over and over and over again.So IT professionals are able to take our cookbooks, from, and apply them using Chef to their infrastructure to solve those common problems, so that they can actually do the more important parts of their job, which are creating business value or doing what their boss actually wants them to get done, or building new kinds of applications like the protein-folding clusters that couldn’t exist before.

Mike: So this is really a gain in productivity.

Jesse: Yeah. Productivity and agility. And if you think about the history of these sorts of changes, we’ve seen them going all the way back to electrification in the 1900s, where first you had to build your own power stations, and then you ended up connecting to a utility grid. And there were trade-offs. One is centralization: when the power goes out, everybody’s powers go out. But on the other hand, you’re not having to maintain your own generating facility. That’s sort of a typical benefit of cloud computing and a typical trade-off. You’re not having to install a server, by hand, racking and stacking in a data center. Which is a terrible task if you’ve ever had to do it, and it’s one that I’m glad we don’t have to do anymore. I’d rather run thousands and thousands of them.

Mike: [laughs] So this has had a lot of parallels to what’s happened in the past is what you’re saying.

Jesse: It has. Many people in cloud computing love to think that they are inventing an entirely new world. My friend, and prolific blogger, Simon Wardley, has a great presentation that I encourage your listeners to search for, which is “Situation Normal, Everything Must Change.” And he actually goes back to the 1800s, to authors that have described this process occurring over and over and over again in industry. So it’s certainly nothing new. We’re just following a predictable chain of business evolution.

Mike: But it also sounds, if you use the example of the electrification, for example, in the beginning, where everyone had their own generating facility, they could customize it, they could build it to what they needed, and things operated as they wanted it to.

Jesse: Yeah.

Mike: But there was a capital expense and a manpower expense that was involved. It sounds like, as we move forward in the next years, those people that hang back and don’t do the cloud computing are going to have excessive cost compared to those who actually move forward. They gain some productivity or those things which will lower the cost to the company and lower the cost, probably, to the consumer. Am I getting that right?

Jesse: That’s exactly right. Salesforce is probably the best example. When you follow the evolution of CRM software, which is probably something that every business uses a version of, even if it’s just a spreadsheet or a hand file that they’re writing down, it used to be that most environments couldn’t afford it in the first place. The few that could derived specialized business value out of investing heavily in it. And then, over time, what’s happened over the past few years, with appearances like companies like Salesforce, is anyone can have it for 100 bucks per person a month, connecting over the Internet, getting it at scale.

And so what we’ve seen is something that only large companies that could afford the massive capital investment could have these great tools, and it provided strategic value, to now it’s a commodity that everyone has. In Opscode’s case, it’s very much the same. The tools that we provide give small organizations, single-developer shops, small IT teams, the same power and resources that the big companies have and had previously to invest heavily in, and employ large workforces to maintain.

That is now becoming a commodity that everyone can have. People are sharing. The business value shifts from this being something that provides competitive advantage to something that you have to have in order to compete. So then the focus is sort of “up the chain” from there.

Mike: It’s a fascinating company, fascinating what you’re doing.

Jesse: Thank you.

Mike: So if somebody wanted to get a hold of you and use your services, how would they do that?

Jesse: They can reach our website at If they’re software developers or they manage software developers, they should certainly encourage them to come check us out. We’re active in lots of communities, and we have very active mailing lists. But please come, and there’s a prominent “contact us” button that they can do, or they can just sign up for a free trial with the platform.

Mike: Sounds wonderful, and, certainly, one of the necessary things to gain productivity, gain cost benefit. And that’s the name of the game in today’s world. And that’s been the name of the game since we’ve been around, huh?

Jesse: Absolutely.

Mike: So you’re listening to Mike Adams of Adams Financial Concepts. I’m here with you Friday afternoons from 4:00 until 5:00. We’re coming up to another commercial break. And after we come back from the commercial break, I want to talk about “Dow 100,000.” Join me right after the break. You’re listening to Mike Adams, Adams Financial Concepts. I’m here with you Fridays from 4:00 until 5:00 here on KKNW Alternative Talk Radio. [music]

Jesse Robbins

Chef Co-Founder & Advisor