Self-Service Infrastructure
summary
Spotify runs a self-service provisioning system that lets teams scale microservices from a capacity tab in the company dashboard rather than through ops. It auto-selects instance types, supports Google Cloud (<1 minute) and on‑prem (parallel provisioning so seven servers take ~25 minutes), and groups machines into reusable instance pools. Lifecycle hooks detect deprecated services and can auto‑clean them; basic monitoring (traffic/metrics) flags unused instances. The tooling and CLI are intentionally minimal and will be open‑sourced after removing Spotify‑specific bits.
AI-written summary · about
chapters (7)
related talks
01h 02mAutomating the component maintenance - Fleet management at Spotify
20mSpotify's Growth Exposed the Abstraction Paradox
34mThe Spotify Story: Moving from NIH to an Active CNCF Community Member
40mScaling Startup for Success: Building modern Architectures and Developer Acceleration
26mFull Life-Cycle Infrastructure Orchestration
28mScale CI/CD in Your Enterprise Through An Internal Spinnaker as-a-Service Model
31mKeeping Your Cloud Footprint in Check
29mThe Platform Engineering Revolution Panel: Empowering Developers Through Self-Service