Skip to content
DKKB

Latency

Section titled “Latency”

Latency is the elapsed time from the start of an operation until the relevant result is observed.

Latency is usually a distribution rather than one stable number. Percentiles such as p50, p95, and p99 show how typical and slow requests differ.

Latency is different from throughput. A system can finish individual requests quickly while completing little total work.

Used in DKKB

Latency vs throughputDistinguish how long one operation takes from how much work a system completes over time, because optimizing one can worsen the other.
Latency vs throughputDistinguish how long one operation takes from how much work a system completes over time, because optimizing one can worsen the other.