GitLab Shared Runners: Fair CI Capacity, and the Cache That Lies
How GitLab.com scaled CI with shared runners, isolation, and caching — and why a cache hit that skips a compile can still ship the wrong artifact.
3 min read
Engineering at scale
3 articles · 9 min of reading
How GitLab.com scaled CI with shared runners, isolation, and caching — and why a cache hit that skips a compile can still ship the wrong artifact.
How GitLab built feature flags on an Unleash-compatible model so code can ship dark and roll out by percentage, actor, or environment.
How GitLab extracted Git disk operations into Gitaly so the Rails app no longer spoke filesystem paths to a shared NFS server.