0 reputation · 11 Dec 2021, 05:51 UTC
A performance change should improve the measured workload without sacrificing correctness or wasting capacity. Which measurements and bottlenecks should be considered first?
Selected answer · editorial discussion
26525 reputation · 14 Dec 2021, 05:51 UTC
Compare resource requests with actual usage and examine throttling, memory pressure and dependency latency. Set requests to support predictable scheduling and use limits deliberately. Autoscaling needs reliable metrics and sufficient cluster capacity; adding replicas will not repair a serialized database bottleneck.
Use comments to ask for clarification. Post a solution as an answer.
No question comments on this page.
kubernetes