advanced

The Tail Doubled, the Median Did Not

Read the evidence before you read the options. The signals are presented the way a dashboard would present them — nothing is labelled with the answer.

The report

A weekly performance review flagged that p99 on the search API has gone from 380 ms to 810 ms over the past two weeks. p50 has not moved. Nobody reported an incident and error rates are unchanged. The team is not sure whether this is worth investigating.

Search API — two week comparisonILLUSTRATIVE
SignalValueWhat it tells you
p50 latency74 ms → 76 msThe typical request is essentially unchanged.
p95 latency210 ms → 244 msThe 95th percentile has risen modestly.
p99 latency380 ms → 810 msThe 99th percentile has more than doubled.
Request rate3,200 req/s → 3,310 req/sTraffic is up 3%.
Error rate0.04% → 0.04%Unchanged.
CPU (fleet average)41% → 43%Compute usage is broadly unchanged.
GC pause time (p99)12 ms → 13 msCollector pauses are stable.
Deploys in window6Six releases went out over the two weeks.
Slowest 1% of requests, sliced by dimensionILLUSTRATIVE
SignalValueWhat it tells you
By instanceevenly spread across all 24No single host accounts for a disproportionate share.
By time of dayevenly spreadSlow requests are not clustered in any particular hour.
By tenant78% from tenant `acme-corp`Four in five slow requests come from a single customer.
Tenant `acme-corp` share of traffic2.4%That customer accounts for roughly one request in forty overall.
Tenant `acme-corp` p50620 msFor this customer, the typical request is already slow.
`acme-corp` document count890k → 4.1 M over 3 weeksTheir indexed dataset grew roughly fivefold.
What is the constraint?