ROCm perf metrics

Loading…

What the perf nightly measured, run by run: the HLO modules of real models replayed one at a time on a gfx950, on every ROCm stack in the matrix, against the same pinned openxla/xla commit. Each point is one nightly, not one day. The sibling view of presubmit CI health is ci-metrics.

ROCm stack
Data

Execution time index

The geometric mean of every module’s time against its own first measurement, so 100 is where the window started and 104 is four percent slower. The typical module: alphafold3 at 1.6 s counts exactly as much as detr at 1.4 ms. Lower is faster.

Total module time

The same runs, summed instead of averaged — what the whole suite costs, and therefore mostly what the three heavy modules cost (alphafold3, stable diffusion 1.5, sdxl). Read beside the index, not instead of it: a run can move here and not there, which says one big module moved and nothing else did. A run that did not measure the whole suite reports no total at all rather than a smaller one.

Where the time goes

Total milliseconds per run, split by model category. One panel per stack, all on one scale so the panels can be read against each other.