Comparison
Distributed tracingvsTracing / spans
Distributed tracing
you follow one slow request across six services and see exactly which hop ate eight hundred milliseconds.
Recording causally linked timing spans for a single request as it crosses process boundaries. It is the only tool that answers "where did the time go" in a distributed system. It requires context propagation through every hop, including queues, and one un-instrumented service breaks the chain for everything behind it.
Full entry →Tracing / spans
you record every prompt, tool call and response in one linked timeline, so you can replay exactly what the run did when it went wrong.
Recording each step of a multi-step run (model calls, tool executions, retrievals) as spans in one trace, exactly as in distributed tracing. For agents, the trace is the trajectory, and it is the primary debugging artefact.
Full entry →