Distributed tracing
P37.metrics-traces.02 · Audience: guest, it-ml, language-pro · Prerequisites: Metrics & the three-pillars model
A metric tells you a request was slow; a trace tells you which stage was slow. A trace is a tree of spans — one per unit of work, a root span for the whole request and a child span per stage — and reading that tree is how you find the bottleneck inside a single request. This module has you instrument a request pipeline and read its spans.
ⓘ Concept: A tree of timed work
Why it matters — Metrics say the request was slow; only the span tree says which stage ate the time — that is what makes a trace worth the instrumentation.
Ask the mentor about this module
Ask a question about this content. The mentor explains and grounds its answer in what you are studying; asking is recorded as a learning signal, not a grade.
🎓 Practice ladder
1 graded rung · ~20 minInstrument the provided request-pipeline simulator: read its span tree — the waterfall, the per-stage durations, and the slowest stage.
Rung 1 — read a trace: instrument the request pipeline
Loading exercise…
That is the three pillars end to end: structured logs to follow a request, metrics to alert on aggregates, and traces to find where the time went.