Roofline model
The roofline model bounds arithmetic throughput using the smaller of a compute ceiling and a bandwidth ceiling multiplied by arithmetic intensity.
How it works
For compute rate P, bandwidth B and intensity I, the idealized limit is min(P, B × I). The ridge point P/B marks where these ceilings intersect. Match precision, sparsity and the memory boundary before comparing values.
Work through an example
With illustrative ceilings of 100 TFLOP/s and 2 TB/s, the ridge is 50 FLOP/byte. A kernel at 10 FLOP/byte has a bandwidth-derived ceiling of 20 TFLOP/s, not a predicted measured speed.
Where the term can mislead
Latency, instruction mix, communication and insufficient parallel work can put performance far below both ceilings.
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.