APM and distributed tracing

Traces, the service map, profiling and how data flows.

apmtracingoverview

What APM gives you

Application Performance Monitoring auto-instruments your services to capture distributed traces, latency, error rates, and method-level detail. A single user request is followed across every service it touches, so you see exactly where time was spent and where it failed.

The building blocks

  • Span — one unit of work (an HTTP handler, a DB call) with timing, attributes and status.
  • Trace — the tree of spans for one request across services.
  • Service map — the live graph of services and their call dependencies.
  • Profile — method/endpoint hot-spots behind a slow span.

How data flows

Your service's agent emits spans over OTLP to the VerOps ingest endpoint; they are stored in a columnar backend for sub-second queries at scale, and isolated per tenant. See Instrumenting your services to send your first trace.