Why use latency heatmaps instead of averages?
mediumAnswer
- Average / median hides bimodal or long-tail distributions.
- Heatmap: histogram over time (Y = latency bucket, X = time).
- Reveals: (1) tail spikes not visible in p50.
- (2) bimodality (fast path + slow path).
- (3) periodic patterns (batch job every 5 min).
- (4) shift over time.
- Standard in Prometheus / Grafana.
- Complementary metrics: p99, p99.9.
- Rule: latency has weird distribution — always histogram.
Check yourself — multiple choice
- Random
- Average hides bimodal + long tail; heatmap = histogram over time (Y buckets, X time) reveals spikes / bimodality / periodic / shift; complement p99, p99.9; latency always histogram
- Not useful
- Just mean
Latency heatmap: reveals tail + bimodality + periodic patterns.
#observability#monitoring
Practise MLOps & Data Quality
215 interview questions in this topic.