Sponsored Keynote: What I Learned Running 10,000+ Kubernetes Clusters
summary
IBM reports running hundreds of thousands of created clusters and tens of thousands active daily, managed by a 25-person SRE team; Jenkins CI/CD proved too slow, so the team builds Slack-integrated chatops and a poll-based self-updating model plus dynamic inventory, rule-and-label targeting, and feature flags via LaunchDarkly to control rollouts. The system exposes per-cluster and per-service views, lets fleets pull updates with fine-grained control, and is released as the open source Razee/Open Razee project that the company says scales reliably.
AI-written summary · about
chapters (8)
- 0:00Intro
- 0:23Context
- 1:21Problems
- 1:52Team Management
- 4:52Cluster Overview
- 5:05Cluster Updates
- 5:21Rules
- 5:53Razzie
related talks
13mSponsored Keynote: Multi What???? Multi-cloud, Multi-cluster, Multi-tenanted K8s
16mKeynote: 7 Years of Running Kubernetes for Mercedes-Benz
52mOperating OpenSearch at Scale: Lessons From Managing 10,000+ Clusters Across...
32mSurviving From Endless Issues Coming From 7K+ Kubernetes Clusters
08mSponsored Keynote: Kubernetes Deployment Made Easy for Complex Applications
20mKeynote: Kubernetes at GitHub
37mOne Cluster to Rule Them All - Multi-Tenancy for 1k Users
14mKeynote: The Big Leap: Conquering the Legacy System and Building a Scalable Cloud Infra...