Feature management is becoming a critical control point for software releases. As teams race to ship faster without letting operational complexity spiral, the ability to fine-tune rollouts is taking on new urgency.

Separating code releases from feature releases gives teams a way to ship faster without exposing every customer to the full risk of a big-bang rollout, according to Jonathan Norris (pictured, right), director of software engineering at Dynatrace Inc. That is the thinking behind Dynatrace’s push to bring DevCycle — a feature management and feature flagging platform it acquired in January — into its broader observability stack, letting teams use software toggles, or “flags,” to turn features on gradually while watching their impact unfold in real time.

“One of the core features of the DevCycle platform [is] where when you look at our dashboard, you look at the DevCycle platform, you’re not looking at a long list of feature flags that some developer made that you have no context [for],” Norris explained. “You’re looking at, ‘Oh, I’m shipping the new checkout flow, the new onboarding flow, the new whatever feature.'”

Norris and Michael Beemer (left), principal product manager at Dynatrace, spoke with theCUBE’s Rob Strechay and Paul Nashawaty at KubeCon + CloudNativeCon EU for an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how unified feature management and observability are transforming DevSecOps workflows and expediting release cycles in Kubernetes environments. (* Disclosure below.)

Feature management expands from developer tool to enterprise release strategy

Dynatrace’s push into feature management through DevCycle is unfolding alongside the rise of OpenFeature, a vendor-neutral open standard. Contributors such as Google LLC and Datadog Inc. have since joined the project, Norris noted. At the same time, Dynatrace is also working to make feature management more meaningful at the business level, not just the technical one, according to Beemer.

“One of the concepts we’re going to include in our offering is … wrapping multiple feature flags together under a [single] feature. The feature’s meant to encapsulate the business context around the release,” Beemer said. “If you could bring in performance and business telemetry there, you could actually see what the impact of a feature was. A lot of times it’s like [project managers] may have good intentions around collecting KPIs on a particular feature, but if we could do that automatically, then it’s something you could just get in the habit of tracking.”

More broadly, the rise of feature management is also forcing observability companies to think beyond telemetry and deeper about how software actually gets released. Adoption will depend on making feature controls valuable not only to developers, but also to operations teams that need visibility into how those releases behave in production, according to Beemer. In that sense, the long-term opportunity may be less about flags themselves than about making release management a shared discipline across the enterprise.

“You have to have a great developer experience, because developers are the ones that add feature flags to your code, so feature flags start and end with developers,” Beemer explained. “But really, the operations team needs to be on board as well. In a large enterprise, that can start with one team, prove successful and then spread throughout the organization.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of KubeCon + CloudNativeCon EU:

(* Disclosure: Dynatrace sponsored this segment of theCUBE. Neither Dynatrace nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE