Three services for the three pillars of observability — logs, metrics, traces. Together they form Google Cloud Observability, the way production systems on GCP let you know what they are doing.
3 services
Logs tell you what happened. Metrics tell you how often, how fast, and with what shape. Traces tell you which call led to which call, and where the latency went. Each pillar by itself answers a slice of "is the system healthy"; together, they answer the whole question. The three GCP services in this chapter — Cloud Logging, Cloud Monitoring, Cloud Trace — were once branded as Stackdriver (then the Operations Suite), are now collectively called Google Cloud Observability, and remain integrated enough that a production setup uses all three rather than picking one.
Cloud LoggingLogs
What happened. Discrete events with full context — requests, errors, audit records. Searchable, routable to sinks.
Cloud MonitoringMetrics
How often, how fast. Numeric time series for dashboards, SLOs, and alerts — including burn-rate alerting.
Cloud TraceTraces
Which call led where. Request paths across services, with the latency of each span on the critical path.
Three pillars. Each answers a slice of "is the system healthy"; a production setup uses all three together.
Cloud LoggingLogs
Cloud MonitoringMetrics
Cloud TraceTraces