This article explains why the 99th percentile (P99) is a critical performance metric, how to compute it efficiently with histogram, HDR Histogram and T‑Digest techniques, compares P90 vs P99, and shows how to answer interview questions about latency monitoring and related metrics.
This article explains how Citus extends PostgreSQL for distributed execution, detailing supported aggregation functions, count‑distinct handling, HyperLogLog columns, Top‑N estimation, percentile sketches with t‑digest, limit push‑down, view support, and various join strategies, while providing concrete SQL examples and configuration tips.
Baidu’s advertising platform computes high‑precision response‑time quantiles at massive scale by intercepting each API call, locally summarizing data with mergeable T‑Digest histograms, periodically uploading compressed, Base64‑encoded summaries to a warehouse where they are merged on demand, enabling low‑latency, cost‑effective percentile analysis with sub‑0.1% error.
