Observability & Reliability (SRE)
5 clusters, spanning half-life layers 0–3: Metrics & dashboards, Logging, Distributed tracing, Error & incident management, SRE practice.
Clusters
Section titled “Clusters”Metrics & dashboards
The four golden signals, cardinality discipline and the counter/gauge/histogram distinction transfer everywhere. PromQL transfers to anything that…
Logging
Structured logging, correlation ids and retention/cost trade-offs transfer to every backend. Query languages transfer to nothing.
Distributed tracing
OpenTelemetry is the durable member — spans, context propagation and the semantic conventions are now a standard rather than a product, so…
Error & incident management
Grouping, fingerprinting and escalation policy transfer. Incident command and the blameless postmortem are Layer 0 and belong to the next cluster.
SRE practice
Layer 0 and the reason to spend real time here. Error budgets, blameless postmortems and capacity reasoning are organisational skills that…