Turbulence, single run: baseline
Summary
Section titled “Summary”solve(in simulation) is the largest hotspot: 911 ms over 10 calls, 99.1% of the session.
Hotspots
Section titled “Hotspots”The leaves of the call tree with the most time: regions without nested regions, and the own time parents spend outside theirs. Shares are of the session. Click one to find it in the table.
Region statistics
Section titled “Region statistics”| region | % session | total [s] | trend | ||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| scope_profiler.session | 100.00% | 0.919314 | |||||||||||||||||||||||||||||||||||||||||||||||||
| └─ (own) | 0.44% | 0.004000 | |||||||||||||||||||||||||||||||||||||||||||||||||
Per rank
| |||||||||||||||||||||||||||||||||||||||||||||||||||
| └─ simulation | 99.56% | 0.915314 | |||||||||||||||||||||||||||||||||||||||||||||||||
| │ └─ (own) | 0.00% | 0.000000 | |||||||||||||||||||||||||||||||||||||||||||||||||
Per rank
| |||||||||||||||||||||||||||||||||||||||||||||||||||
| │ └─ setup | 0.44% | 0.004000 | |||||||||||||||||||||||||||||||||||||||||||||||||
Per rank
| |||||||||||||||||||||||||||||||||||||||||||||||||||
| │ └─ solve (10x) | 99.13% | 0.911314 | |||||||||||||||||||||||||||||||||||||||||||||||||
physicshot kernels/solver.py:12 Per rank
Line profile solvesolver.py:12911 ms
| |||||||||||||||||||||||||||||||||||||||||||||||||||
| No regions match the filter. | |||||||||||||||||||||||||||||||||||||||||||||||||||
Durations are in seconds. (own) rows show a region's time excluding its children. % session uses wall-clock coverage; overlapping recursive calls count once. Click a row for its call site, line profile and per-rank breakdown.
Hardware counters
Section titled “Hardware counters”Counters are shown only for runs and ranks that recorded them. LIKWID groups include their raw events and derived metrics; Linux perf-event values are totals across each region's calls.
LIKWID: baseline, rank 0, group FLOPS_DP
| counter | solve |
|---|---|
| call count | 10 |
| runtime [s] | 0.911314 |
Events
| counter | solve |
|---|---|
| INSTR_RETIRED_ANY | 4200000000 |
Metrics
| counter | solve |
|---|---|
| CPI | 0.74 |
Linux perf events: baseline, rank 0
| region | calls | cache-misses | cycles | instructions |
|---|---|---|---|---|
| solve | 10 | 1800000 | 3100000000 | 4200000000 |
Charts
Section titled “Charts”Region durations
Each bar shows a region's total recorded duration, summed over the selected ranks, longest first.
Timeline: baseline
Each bar is one recorded region call on rank 0. Its position and width show when the call started and how long it ran; colors identify regions.
LIKWID: CPI
Bars compare the selected LIKWID metric across regions and ranks. The hardware-counter tables above retain every recorded event and metric.
Metadata
Section titled “Metadata”| chip_information | synthetic CPU |
|---|---|
| mpi_size | 1 |
| num_nodes | 1 |
| omp_num_threads | 1 |
| platform | synthetic |
| scope_profiler_version | sample |
| timestamp | 2026-10-17T12:00:00Z |
| total_cores | 1 |