hard20 pts · first answer countslatencydistributed-systems
Tail latency with fan-out
A frontend request fans out to 100 backend servers in parallel and must wait for all of them. Each backend independently has a 1% chance of a slow response. Roughly what fraction of frontend requests will be slow?
Discussion 0
Join free to post in the discussion.
Related questions
- Why add jitter to retries?Site Reliability Engineer
- Percentiles vs average latencySite Reliability Engineer
- Circuit breaker half-open stateSite Reliability Engineer
- p99 latency across all instancesSite Reliability Engineer
- Combining percentiles from many serversSite Reliability Engineer
- Coordinated omissionSite Reliability Engineer
- Misleading histogram quantilesSite Reliability Engineer
- Outage that outlives its triggerSite Reliability Engineer