Nsight Systems
Nsight Systems is a timeline profiler for investigating how CPU activity, GPU work and transfers interact across an application.
How it works
Begin with an application interval that matters to the user, then locate long kernels, idle gaps, synchronization and communication. The timeline helps select a region for more expensive kernel-level inspection.
Work through an example
If individual kernels are short but the request is slow, inspect the spaces between them. Host scheduling, launch submission, synchronization or data preparation can explain time that kernel arithmetic counters cannot.
Where the term can mislead
A trace is affected by the enabled collection options. Keep a matching uninstrumented timing and do not treat tracing overhead as application work.
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.