Latency hiding
Latency hiding keeps an execution resource productive by doing independent work while another operation waits for its result.
How it works
A scheduler can issue from another eligible warp, and a thread may expose independent instructions before consuming a delayed result. Asynchronous copies can also overlap with compute. Each method needs resources to hold the additional in-flight state.
Work through an example
If every resident warp immediately consumes its own pending load, increasing residency may help only until other resources fill. Reordering independent work can improve eligibility without changing occupancy.
Where the term can mislead
Hiding a wait does not make the waited-on operation finish sooner. It reduces the exposed wait on the overall execution path.
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.