Editorial question90.8K views3,429 votes0 answers1,966 following
AI-generatedHTTP(S) Load Balancing to Cloud Run concurrency cap and tail latency during revision rollout
Goal: determine whether tail latency that appears only under concurrent requests through an HTTP(S) Load Balancer to Cloud Run originates at the integration boundary rather than the client or application. Cloud Run limits simultaneous requests per instance using the concurrency setting; excess requests queue or prompt new instances. The load balancer does no
google