Multiple ChoiceBeginner 45s
p99 vs mean latency
Why do experienced teams monitor p99 latency rather than only the mean?
Options
- 1The mean hides the tail; a small fraction of very slow requests is what users actually notice and what often stalls upstream callers
- 2The mean is statistically fragile, whereas p99 is cheaper to compute since it needs no full-sample sort
- 3The mean understates load badly, whereas p99 is always equal to the maximum of the observed latencies
- 4The mean is misleading for skewed traffic because it cannot be computed from histograms, only percentiles can
Think you can answer under pressure?
Get matched and answer live questions in 3-minute 1v1 battles. Elo ranking, badge tiers and bragging rights on the line.