Nsight Compute
Nsight Compute is a kernel profiler that collects supported GPU counters and relates them to source, native instructions and resource limits.
How it works
A capture may replay a kernel or application to collect incompatible counter sets. That affects duration and cache state, so target representative launches and understand the replay mode. Interpret metric names with their units, aggregation and normalization.
Work through an example
A report showing low DRAM throughput and many long-scoreboard waits suggests inspecting memory dependencies and available independent work. It does not establish that the DRAM interface is saturated.
Where the term can mislead
Profiled duration is not automatically a production latency measurement. Validate an optimization using an uninstrumented workload with correct outputs.
Connect the idea
Take it into practice
Continue with the existing concept, apply it to an interview problem, or follow a course lesson. Access requirements are shown before you open a page.
New to GPUs? Start with the free execution-model concept.
Sources and scope
Examples here are teaching calculations, not measured GPU performance. Consult the linked documentation for the exact device, instruction and tool version. Source review: 25 October 2026.