What is the difference between monitoring and tracing?

Monitoring is the collection and analysis of system metrics (CPU, RAM, network, disks, errors, delays) to identify problems in real-time or retrospectively.

Tracing (distributed tracing) is the recording and visualization of the request (transaction) path through various services and components of a distributed system, allowing understanding of how the request is processed at each stage, where delays and errors occur.

Attribute	Monitoring	Tracing
Purpose	Assess the overall state of the system/service	Understand the path of a specific request/transaction
Data	Aggregated metrics (counters, averages, percentiles)	Data about each segment of the request (spans)
Application	Detecting general problems, trends	Debugging specific issues, flow optimization
Examples	Prometheus, Zabbix, Grafana	Jaeger, Zipkin, OpenTelemetry

Tracing often complements monitoring by providing detail for metrics indicating problems. For example, a high delay metric (monitoring) can be investigated with tracing to understand which specific service or component caused this delay.