tail-latency · research · also latency tail

Tail latency

Tail latency describes the high-latency part of an operation response-time distribution, usually summarized using specified high percentiles.

State the operation population, timing boundaries, percentile estimator and observation window. An hourly mean cannot reconstruct request p99, and a wait-for-all fan-out request follows the slowest required response rather than an average leaf. [1] The linked synthetic examples isolate these measurement differences.

Relations

related

latency-storage, tail-hour, coordinated-omission, fail-slow

Citations

1. Dean and Barroso, The Tail at Scale, CACM 56(2), 2013; DOI 10.1145/2408776.2408794 (accessed 2026-10-15)
2. Hesela: Slow RAID hours are not request tail latency; source audit and reproducible counterexamples (accessed 2026-10-15)

JSON: /data/terms/tail-latency.json