In 2012, software intelligence company Dynatrace set out to achieve seamless automation from development to production. Its other goal was to move code from development to production in an hour versus 24 hours. However, at the time, development and operations teams weren’t collaborating well.

DevOps seemed like the right answer. Although operations had its own idea about it: Operations would build APIs developers could use. However, things were moving too slowly.

Next, Dynatrace built a new cloud based on its SaaS offering. It was completely self-healing, alleviating the need for operations people in production.

RELATED CONTENT:
Creating a DevOps culture
5 DevOps myths
Still testing like it’s 1999?

“We actually arrived at NoOps in 2014. We have developed it into a practice called ‘autonomous cloud enablement,” said Dynatrace CTO and co-founder Bernd Greifeneder. “DevOps has evolved into NoOps and autonomous cloud with self-healing approaches, automating runbooks down to zero and establishing a NoOps/autonomous cloud culture. Everyone wants to deliver business-relevant competitive services, not nurse applications and watch charts on dashboards to then execute tedious, repetitive tasks.”

Achieving one-hour code to production wasn’t easy

Dynatrace spent three years shrinking its 24-hour code to production window to one hour. At times, people thought it wasn’t even practical. Back then, 200 engineers worked on a single product. Now, there are 470. 

Operations proved to be the biggest bottleneck which is why the company moved to NoOps in the first place. In fact, the 10-person R&D team is now the “autonomous cloud enablement team.” Their job is to help people understand the concept of self-healing.

Meanwhile, one of the architects built a visualization device called the “Dynatrace UFO” that advertises software errors in the cafeteria and other prominent places. Its purpose is to give engineers immediate feedback they can see and feel. Now when someone’s error appears, they fix it, pronto.

“Feedback is one of the most important cultural items. This is why we say, ‘It’s not about continuous delivery, it’s about continuous delivery and feedback,” Said Greifeneder.

Customers freaked out

Dynatrace was so proud of its NoOps practice, it told customers about it. Apparently, they freaked out.

“So often [I’m told], you’re alienating customers when you’re talking about NoOps because you sell operations. The reality is no because every CIO knows they need to do more with the same people and build more modern services and digital transformation,” said Greifeneder. “They love the idea. They just don’t know how to get there.”

Eventually, customers started asking how it works. In response, Greifeneder and his team built a practice and blueprints. 

“If you want a DevOps or NoOps culture, you need to break [down] the silos from the top down,” he said. “Successful implementation will require moving people from operations to development. It is [also] important to remember you can’t create a successful DevOps practice or autonomous cloud enablement practice without executive buy-in.”