Autonomous Cloud Enablement aka Scaling NoOps via Self-Service

Autonomous Cloud is not another lofty marketing term. Autonomous Cloud is what enables our globally distributed development teams at Dynatrace to deliver better software faster following our NoOps approach: Fully Autonomous and as a Self-Service!

Three waves of DevOps leading to Autonomous Cloud

At Dynatrace, we have been very proud and vocal about our own DevOps transformation story that we started when we incubated what is now known as the Dynatrace Software Intelligence Platform (formerly Ruxit). Over the years we have seen three major waves of evolution for us: Speed, Stability and Scale. The illustration below gives you an overview of this evaluation, the focus, the mindset and how we measured success:

3 waves of DevOps evolution led us to Autonomous Cloud which allows us to Scale NoOps across the enterprise
3 waves of DevOps evolution led us to Autonomous Cloud which allows us to Scale NoOps across the enterprise

In order to better explain why we are talking about Autonomous Cloud, what it means and how you can apply our lessons learned, let me explain the three waves of transformation in more detail:

Wave one: DevOps to increase speed

We first brought our DevOps story to life at Perform 2017 when our CTO & Founder Bernd Greifeneder, DevOps Lead Anita Engleder and I went on stage. We talked about the cultural and process changes in order to follow through on Bernd’s requirement of “1 hour Code to Production!”

DevOps Transformation at Dynatrace enacted live on stage at Perform 2017!
DevOps Transformation at Dynatrace enacted live on stage at Perform 2017!

I think the story not only impressed the audience because of our outfits (traditional Austrian Dirndl & Lederhosen), but because we were one of the few software companies that showed actionable results of a DevOps transformation that has been discussed in our industry since 2013 when Gene Kim released his famous The Phoenix Project.

A key to success from the start was that not only we did build Dynatrace, but Anita’s team was also always “Customer 0” of Dynatrace because clearly we were in need of a world class monitoring platform that gave us visibility into our deployments in dev, staging and production.

Wave two: NoOps to ensure stability!

With a growing product, more new end user features than ever and an expanding user base, came the need to focus even more on automating quality and stability as we see these attributes as clear competitive differentiators.

To do that, Anita’s team drove innovation around a common delivery pipeline to enable developers automating operational tasks such as runbook execution to solve infrastructure problems. As “Customer 0”, her team requested several new core product capabilities such as log analytics, configuration APIs or better integration with incident notification services. They also saw that auto-remediation only works if the software has self-healing capabilities for situations that can’t and shouldn’t be fixed from outside the platform. The engineering team therefore started building self-healing mechanisms into Dynatrace, allowing it to auto-remediate issues related to peak load or that came in through side-effects of new features.

This was the time when we started using the term NoOps because everything that was built had to be built with a mindset that no traditional operational tasks were necessary to deploy and operate our software. And the numbers confirmed the success of this NoOps strategy! Thanks to proactive problem detection and auto-remediation we reduced the number of production problems impacting our end users by 93% while increasing the features being delivered,

Wave three: Autonomous Cloud Enablement to scale

Just as in any enterprise that has a software engineering organization, the Dynatrace Software Intelligence Platform is not the only software that gets developed by 800+ engineers. We have teams building cloud-native services for our billing, licensing, customer experience, proactive customer support, marketplace or Davis® Assistant. We have other teams focused on developer productivity services, and others who build external extensions to our Dynatrace Platform supporting new use cases for our customers and internal demand.

At one time, we saw that some of these teams were lacking proper NoOps approaches and experimented with their own tooling. And then, we realized that this won’t scale for long. We needed to take what we have built to support NoOps for our Dynatrace Platform, make it extendable to support other target platforms used by our other teams and make it so easy that every engineering team wants to use this platform vs building their own. This eliminates duplicated work and makes our teams more efficient as they can build value vs supporting tools and therefore allows us to scale.

We’re now referring to it as Autonomous Cloud, with Anita’s team enabling the rest of the organization to become fully autonomous with the tooling and integrations they have built around the Dynatrace Software Intelligence Platform. Hence the new name of her team: ACE – Autonomous Cloud Enablement.

Autonomous Cloud with Dynatrace

At Dynatrace our ACE team is enabling our engineers to deliver better software faster. They came up with a set of tools that are centered around the Dynatrace Software Intelligence Platform, and that provides core capabilities you expect from a modern software platform. The following illustration shows many of these capabilities such as Self-Healing, ChatOps, Progressive Delivery, SRE or Security. Some of these capabilities are handled by external tools that are integrated with Dynatrace. Some capabilities have already made it into the core Dynatrace platform or are being offered through projects such as Keptn, our event-driven control plane for continuous delivery and automated operations:

Autonomous Cloud Enablement: Empowering engineers with NoOps as a Self-Service
Autonomous Cloud Enablement: Empowering engineers with NoOps as a Self-Service

While our internal development tools and agile processes may differ from those you have in your organization, I believe we’re all facing the same challenges as we try to scale our organizations without losing agility.

If you want to learn how Autonomous Cloud Enablement can look in your organization, which tools you can integrate or even replace with Dynatrace and how you can enable a NoOps culture and mindset, feel free to reach out to us. Here are a couple of options

  1. Assess yourself through our Autonomous Cloud Survey
  2. Have a look at our Autonomous Cloud Enablement Service offerings
  3. Do a Quickstart with Keptn and see how Autonomous Cloud feels like end-2-end
  4. Have a look at my other blogs and YouTube tutorials and learn how to integrate and automate Dynatrace into your existing processes
  5. Revisit our story around the Dynatrace UFO and how it changed our culture