Lightning Talk: Evolving the Job API to Become Defacto Standard for Batch Worklo...
summary
Kubernetes Job API adds concrete features: indexed Jobs give each pod a numeric index accessible via env var and scale to 5,000+ pods; suspended jobs let external queue controllers pause and resume work for capacity-aware scheduling; tracking with finalizers fixes lost-completion counters after a controller rewrite that had to be disabled until 1.25; failure policies let jobs react to pod disruption reasons and exit codes. Completion vs parallelism controls rate, and future work targets per-index backoff and selective retries.
AI-written summary · about
chapters (5)
- 0:00Start
- 1:32Background
- 5:29Job suspension and mutable scheduling directives
- 11:25Pod Failure Policies (1/3)
- 17:15How to get involved
related talks
37mIntroduction to the Kubernetes WG Batch
05mLightning Talk: Kubernetes Jobs and the Sidecar Problem
30mWG-Batch Updates: What’s New and What Is Next?
06mLightning Talk: OpenKruise: Advancing Kubernetes Workload Auto...
27mKubernetes SIG Apps Updates
29mBuilding a Batch System for the Cloud with Kueue
16mKubernetes SIG Apps Updates
34mWG-Batch Updates: What’s New and What Is Next?