Definition
Plain language
The slowest one percent of cases — used to track worst-case latency.
As stated in the literature
The 99th-percentile value of a latency distribution, a standard tail-latency metric for serving systems.
Also called: P99 latency, P99 tail
Why it matters: Tail latency, not average latency, is what users actually feel during a bad moment, so P99 is the metric serving teams optimize against.
For example, if P99 latency is 2 seconds, then 1 percent of user requests are taking longer than that.
Heard on the show
“The paper measures time-to-first-token for each round of an agent loop, and reports that baseline systems consistently exceed a thousand seconds at the P99 tail.”Episode 016 — Why Your Coding Agent Stalls While the GPU Runs Hot