CUDA event
A CUDA event marks progress in a stream and can be used for supported synchronization or device timing.
How it works
Recording an event queues the marker after earlier work in its stream. Another stream can wait for that completion, or the host can query it. Timing a pair of events measures the interval between the markers under the applicable execution and timing rules.
Work through an example
Place start and stop events around the device work you intend to measure, wait for completion, and repeat after warmup. Use a host clock separately for complete request latency.
Where the term can mislead
A device interval can omit CPU preparation, queuing and result handling. Do not report it as end-to-end service latency.
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.