You are here

You are here

Gene Kim on DOES16 London: Who's talking, what's trending

Mike Perrow Technology Evangelist, Vertica

Since DevOps enthusiast and IT Revolution founder Gene Kim launched the DevOps Enterprise Summit (DOES) three years ago, it has become a seminal annual event for DevOps practitioners. Co-produced by IT Revolution and founding partner Electric Cloud, DOES is described on the DOES16 website as “a conference for the leaders of large, complex organizations implementing DevOps principles and practices." Attendees can check in with enterprise IT leaders in DevOps, who describe the technical, architectural, and management best practices they have used to lead change in their organizations. "The goal is to give leaders the tools and practices they need to develop and deploy software faster and to win in the marketplace.”

In advance of the London DOES event, to be held on June 30 and July 1, editor Mike Perrow asked Kim, author of The Phoenix Project and tireless enthusiast for the principles and practices of DevOps, about this year's featured speakers, the topics they will be discussing, and the key takeaways from those presentations.

What can attendees look forward to at DOES16? What stories are going to be coming out in London?  

Gene Kim: We’re hearing something similar from all of the announced speakers, across different industries. For example, Thomas Cook Group is one of the most well-known travel companies in the U.K., offering prepackaged vacations, hotels, and travel. I can’t imagine how much they’ve been disrupted by e-commerce. They’re going to tell this amazing story of building up capability within the organization. Their head of DevOps, Lee Sexton, is going to tell how he’s made a material contribution to his company’s competitiveness with DevOps practices.

Chris Jackson, the director of cloud product engineering at the international media company Pearson, is going to talk about building internal platforms so that any development team can deploy internally to a cloud, or to a public cloud using the same practices that you would see at Google or Netflix. They’re often dealing with the personal information of learners inside of higher education, so they want that protected, not only in speaking to market but protecting student information. This is a publications company that has been around for over a century and building an online presence and iPad application capabilities, very much like what we saw with The New York Times. One could say that the survival of the organization depended on creating a modern, digital platform that could reach consumers the way we read news. 

Thomas Cook's head of DevOps, Lee Sexton, is going to tell how he’s made a material contribution to his company’s competitiveness with DevOps practices.

Hiscox is a multibillion (British sterling) insurance business, with certain segments growing more than 30% a year—and the channel for that is primarily online. The notion of an insurance company growing at 30% a year, regardless of what category, is just amazing! Speaking for one of the largest insurance companies in the U.K., Jonathan Fletcher—he’s their enterprise architect lead—will be talking about modernizing development practices for one of the largest parts of their business portfolio.

I also want to mention ING. Ron Van Kemanade has been CIO as ING transformed into a world-class technology organization since 2011, and one of the most admired organizations in the DevOps community. I’m just super-excited that, here’s a CIO, someone who will give us his view from the top of technology leadership right now. It’s not just spreadsheets and budgets. This is really about an epic transformation.

Pearson is going to talk about building internal platforms so that any development team can deploy internally to a cloud using the same practices you see at Google or Netflix. 

What kinds of things have these traditional companies been able to accomplish with their DevOps and IT transformations? Can you really connect their results to DevOps practices? 

Gene Kim: Totally! The greatest hits have included a major U.S.-based department store, whose CTO (who’s been promoted twice by the way) explained how her team embarked on everything from the e-commerce sites to the in-store cafeteria systems, to the iPad application, to the on-floor department managers, just to see kind of all the places where they might make an improvement, make a material contribution to the organization. Another major retailer talked about specific things they did with APIs to enable 53 different business initiatives, including Instagram integration, and a mobile app for ship to store—which is one of the most strategic business capabilities for a retailer.

A major investment bank described how they’re rolling out DevOps across the organization, allowing them to do hundreds of releases per day—not in months but in minutes. Household names (companies I can’t disclose here, because I’m not at liberty) in everything from family entertainment to national defense have these incredibly heroic tales that have proved how wildly successful their DevOps transition has been. They’ve typically created billions of dollars of revenue. They’re bringing these new practices and principles, and they’re working in ways that are very different from before.

In a recent survey you asked IT organizations what their biggest issues were, what IT professionals are continuing to struggle with. Can you can give us a sneak peek on that? Maybe the top three takeaways?  

Gene Kim: There are four takeaways I can describe. We went through more than 300 responses, both from the speakers as well as the attendees, and put them all on the board and organized them into themes.

The first one I would call “leading change.” For example, we got question such as, “I see the light, but how do I get my leadership on board, especially kind of the more stodgy and conservative parts of the organization?” Or, “I’m already buried, so how do I create my first win?” Or, “We’ve achieved some DevOps successes, but how do I accelerate its adoption across the organization?”

A major investment bank described how they’re rolling out DevOps across the organization, allowing hundreds of releases per day—not in months but in minutes. 

Culture I think also fits in here. “How do I move from an organization that has optimized technology cost, but now it’s all about optimizing for speed? If I can’t get that mind-set changed, we have nothing to sell. We can’t survive in that environment.”

The second one is about the organization. “What does the org chart look like in this new world?” Today’s typical structure shows a functional organization, where devs live in one place, ops live in another. All the DBAs live in one place. All the networking people in another place. In the worst case, this can lead to a “ticketing culture” where to move a small change into a production requires 300 tickets and no one actually succeeds at the greater goal.

In DevOps, we need a very fast flow between these functional groups. So what does the org chart look like? Instead of focusing on the org chart, we’re actually cataloging the organizational archetypes for this. It’s about incentive structures and shared goals, but still, I think technology leaders have to be able to have good answers to questions around org design.

People want to know what needs to be automated. What are the basic skills that every dev and ops leader has to have? 

The third one is about security and compliance in the context of DevOps. “How do we do DevOps without making the security people freak out, especially in regulated environments? We know they can do it, but how do we really bridge that gap so we can get security and compliance onboard?”

The fourth one has to do with technology leadership. Some people asked, “What does every technology leader need to know in terms of technical practices and architecture, because I think it looks very different from how it did 30 years ago.” People want to know what needs to be automated. What are the basic skills that every dev and ops leader has to have? What do they need to know to be able to contribute to this new world? Underneath this, what re-scaling needs to happen, especially for ops and security people who have worked the same way for 30 years? What things do they need to know in order to be able to contribute?

What else would you like to mention regarding the DOES16 event? 

Gene Kim: Maybe I can just re-validate why I love what I’m doing so much right now. For many years, I’ve been benchmarking high performers—researching 20,000 organizations with Puppet Labs; working with Jez Humble, who wrote the book Continuous Delivery; and working with Dr. Nicole Forsgren at Chef. So, we know what high performers look like. We’re going into our fourth year of that study, and that’s really exciting.

But for me, our DevOps Enterprise Summit is equally interesting. It takes us from the theory of what high performance looks like, to the who and the how, to the people actually creating high performance in large, complex organizations. That’s where I believe the majority of the economic value is going to get created. There’s something really epic about these stories. Every one of these leaders in San Francisco and London have this utter sense of clarity and certainty and conviction that what they’re doing is genuinely needed by their organizations to survive and win in the marketplace.

The elements of their heroic struggles are what create these amazing stories. They’re winning, they’re getting promoted, they’re creating value. That’s personally very rewarding for me.

Keep learning

Read more articles about: App Dev & TestingDevOps