Compute-bound
A compute-bound workload is limited by the rate at which the relevant arithmetic work can execute, under the current input and implementation.
How it works
The relevant ceiling might be a matrix pipeline, ordinary floating-point arithmetic or a special-function path. Reducing memory traffic helps little if that traffic already overlaps and arithmetic remains on the critical path.
Work through an example
A large matrix multiplication with enough input reuse may approach its applicable matrix-compute ceiling. Switching to a smaller batch can reduce reuse and change the limiting resource.
Where the term can mislead
High nvidia-smi GPU utilization does not prove compute saturation. Inspect the instruction path, its throughput and the workload’s measured time.
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.