Poisson solver, hybrid MPI + OpenMP
MPI ranks and OpenMP threads scaled together; the scaling charts can be drawn over ranks, nodes, threads or cores.
Tags: cpu, hybrid
| run | file | recorded | ranks | threads | wall time | vs baseline | page |
|---|---|---|---|---|---|---|---|
| 1r-1t | 1r-1t.h5 | 2026-10-17 12:00 UTC | 1 | 1 | 822 ms | baseline | run page |
| 2r-1t | 2r-1t.h5 | 2026-10-17 12:00 UTC | 2 | 1 | 420 ms | −48.9% | run page |
| 2r-2t | 2r-2t.h5 | 2026-10-17 12:00 UTC | 2 | 2 | 254 ms | −69.1% | run page |
| 4r-2t | 4r-2t.h5 | 2026-10-17 12:00 UTC | 4 | 2 | 140 ms | −82.9% | run page |
| 4r-4t | 4r-4t.h5 | 2026-10-17 12:00 UTC | 4 | 4 | 83.6 ms | −89.8% | run page |
| 8r-4t | 8r-4t.h5 | 2026-10-17 12:00 UTC | 8 | 4 | 60.1 ms | −92.7% | run page |
Each run's page is linked on the right.
What changed
Section titled “What changed”Durations
Section titled “Durations”Charts
Section titled “Charts”Region durations
Grouped bars compare each region's mean call duration across the profiled runs.
Speedup
Each line is one region's speedup: its mean call duration on the run with the fewest MPI ranks × OpenMP threads divided by its mean call duration on each run. The dashed line is ideal scaling; a region below it gains less than its extra MPI ranks × OpenMP threads would allow. The 8 regions with the most time on that first run are shown. Valid for a strong-scaling study: the same total problem size on every run.
Rank imbalance
Each line compares a region's total duration by rank; its dashed line marks the mean across ranks. Points far from the mean identify stragglers.
Rank heatmap
This heatmap uses exclusive timings. Exclusive duration is the time spent in a region itself, excluding time spent in nested child regions; this prevents the enclosing session region from dominating the heatmap.
Metadata
Section titled “Metadata”| commit | 0000000 |
|---|---|
| system | synthetic |