Kube-burner: one benchmark run, start to verdict

Scroll or pinch to zoom, drag to move, double-click to zoom in 100% SVG PNG Back to the post
Kube-burner run flow: a run starts with the configuration and an optional cluster health check. Jobs create objects in iterations, limited by queries per second and burst. The run waits for objects to be ready and can churn them. Measurements such as pod latency are collected, then Prometheus metrics and alerts are evaluated and indexed. Error or critical alerts and latency threshold breaches make the exit code non-zero, and created namespaces can be garbage collected.