Case Study: Spotify achieves faster scaling and improved efficiency with Kubernetes

An Early Adopter of Containers, Spotify Is Migrating from Homegrown Orchestration to Kubernetes

Spotify, a global audio-streaming platform with more than 200 million monthly active users, needed a more scalable way to support its autonomous engineering squads and growing microservices environment. The company had been running containerized services on VMs with its homegrown orchestration system, Helios, but by late 2017 it was clear that a small internal team could not keep pace with the broader ecosystem. Spotify turned to Kubernetes as a more feature-rich, community-supported alternative.

Kubernetes enabled Spotify to migrate gradually alongside Helios, using its APIs and extensibility to integrate with legacy infrastructure and reduce migration risk. The results included faster service creation—cutting provisioning from about an hour to seconds or minutes—autoscaling for a service handling over 10 million requests per second, and average CPU utilization improvements of two- to threefold through bin-packing and multi-tenancy. Spotify also reported reduced manual capacity work and better focus for teams on delivering features, with Kubernetes positively affecting lead time, deployment frequency, time to resolution, and operational load.

Spotify

Dave Zolotusky

Software Engineer, Infrastructure and Operations