With PR #8807, we expect improved performance in the VPA updater, since it will iterate over fewer pods and therefore perform less work. To validate this, we need a way to benchmark performance and confirm that the PR provides a measurable improvement.

We’d also like to establish a general performance benchmarking method that can be applied to future PRs across all components, enabling us to track whether performance improves or regresses over time.

/area vertical-pod-autoscaler
/kind feature
/cc @adrianmoisey