Distributed tracing is how you follow one request across many services and find the exact hop that was slow or failed.
Distributed tracing is a technique that follows a single request as it travels across multiple services, recording each step as a span and linking them with a shared trace ID so you can see the request's full path and where its time was spent.
As a request enters the system it is assigned a trace ID. Each service it touches records a span, a timed operation, and passes the trace context downstream so the next service's spans attach to the same trace. The result is a tree of spans that reconstructs the request's journey with timing at every step.
Logs tell you what happened in one service; a trace tells you where a request spent its time across all of them. When an end-to-end request is slow, the trace shows which single hop, a database call, a downstream API, accounts for the delay, which no per-service metric can do on its own.
Ops Singularity uses trace data as one of the signals Sentinel AI correlates during investigation, so a latency incident is tied to the exact service and span responsible before any action is taken.
A span is one timed operation; a trace is the collection of spans, linked by a shared trace ID, that represents one request across services.
Not strictly, but OpenTelemetry is the open standard for it, providing the instrumentation and context propagation that make traces portable across tools.
Ops Singularity turns open telemetry into autonomous, governed resolution. See it on your own stack.