Poisson solver, strong scaling: np8

Filter regionsRegionsall 5Depthall levels

np8.h52026-10-17 12:00 UTCsynthetic CPU8 ranks5 regions0.0747872 s profiled

← Poisson solver, strong scaling · ‹ np4

Summary

Wall time74.8 mssetup to finalize

Ranks8

Regions5

In regions94%of the session

Top hotspotsolve67.7% of the session

Load imbalance+8%solve, slowest on rank 7

solve (in simulation) is the largest hotspot: 385 ms over 80 calls, 67.7% of the session.
The three largest hotspots account for 94% of the session.

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

region

% session

total [s]

trend

scope_profiler.session (8x)

100.00%

0.569294

└─ (own)

5.62%

0.032000

Per rank

rank	calls	total [s]	avg [s]	min [s]	max [s]
0	1	0.0695235	0.0695235	0.0695235	0.0695235
1	1	0.0703267	0.0703267	0.0703267	0.0703267
2	1	0.0702778	0.0702778	0.0702778	0.0702778
3	1	0.0722968	0.0722968	0.0722968	0.0722968
4	1	0.0683748	0.0683748	0.0683748	0.0683748
5	1	0.0726222	0.0726222	0.0726222	0.0726222
6	1	0.0710855	0.0710855	0.0710855	0.0710855
7	1	0.0747872	0.0747872	0.0747872	0.0747872

└─ simulation (8x)

94.38%

0.537294

│ └─ (own)

0.00%

0.000000

Per rank

rank	calls	total [s]	avg [s]	min [s]	max [s]
0	1	0.0655235	0.0655235	0.0655235	0.0655235
1	1	0.0663267	0.0663267	0.0663267	0.0663267
2	1	0.0662778	0.0662778	0.0662778	0.0662778
3	1	0.0682968	0.0682968	0.0682968	0.0682968
4	1	0.0643748	0.0643748	0.0643748	0.0643748
5	1	0.0686222	0.0686222	0.0686222	0.0686222
6	1	0.0670855	0.0670855	0.0670855	0.0670855
7	1	0.0707872	0.0707872	0.0707872	0.0707872

│ └─ setup (8x)

5.62%

0.032000

Per rank

rank	calls	total [s]	avg [s]	min [s]	max [s]
0	1	0.004	0.004	0.004	0.004
1	1	0.004	0.004	0.004	0.004
2	1	0.004	0.004	0.004	0.004
3	1	0.004	0.004	0.004	0.004
4	1	0.004	0.004	0.004	0.004
5	1	0.004	0.004	0.004	0.004
6	1	0.004	0.004	0.004	0.004
7	1	0.004	0.004	0.004	0.004

│ └─ solve (80x)

67.68%

0.385294

Per rank

rank	calls	total [s]	avg [s]	min [s]	max [s]
0	10	0.0465235	0.00465235	0.00465235	0.00465235
1	10	0.0473267	0.00473267	0.00473267	0.00473267
2	10	0.0472778	0.00472778	0.00472778	0.00472778
3	10	0.0492968	0.00492968	0.00492968	0.00492968
4	10	0.0453748	0.00453748	0.00453748	0.00453748
5	10	0.0496222	0.00496222	0.00496222	0.00496222
6	10	0.0480855	0.00480855	0.00480855	0.00480855
7	10	0.0517872	0.00517872	0.00517872	0.00517872

│ └─ exchange (80x)

21.08%

0.120000

Per rank

rank	calls	total [s]	avg [s]	min [s]	max [s]
0	10	0.015	0.0015	0.0015	0.0015
1	10	0.015	0.0015	0.0015	0.0015
2	10	0.015	0.0015	0.0015	0.0015
3	10	0.015	0.0015	0.0015	0.0015
4	10	0.015	0.0015	0.0015	0.0015
5	10	0.015	0.0015	0.0015	0.0015
6	10	0.015	0.0015	0.0015	0.0015
7	10	0.015	0.0015	0.0015	0.0015

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 and per-rank breakdown.

Load balance

Each region's own time per rank, excluding nested regions, so a parent does not repeat its children's imbalance. Excess is how much longer the slowest rank spends than the average rank: the most balancing that region could save. Regions are ordered by it.

region	mean	min	max	slowest rank	imbalance	excess
solve	48.2 ms	45.4 ms	51.8 ms	7	+8%	3.63 ms
setup	4 ms	4 ms	4 ms	0	+0%	0 s
exchange	15 ms	15 ms	15 ms	0	+0%	0 s

Rank comparison

Each cell compares one rank's time in a region with the region's mean over the ranks. Rank 7, in red in the header, spends the most time in regions overall. Hover a cell for its time.

faster than the meanslower

region	mean	0	1	2	3	4	5	6	7
solve	48.2 ms	−3.4%	−1.7%	−1.8%	+2.4%	−5.8%	+3.0%	−0.2%	+7.5%
exchange	15 ms	0%	0%	0%	0%	0%	0%	0%	0%
setup	4 ms	0%	0%	0%	0%	0%	0%	0%	0%
all regions	67.2 ms	−2.4%	−1.2%	−1.3%	+1.7%	−4.1%	+2.2%	−0.1%	+5.4%

Charts

Region durations

Each bar shows a region's total recorded duration, summed over the selected ranks, longest first.

stack children

Timeline: np8 (rank 0)

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.

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

commit	`0000000`
system	`synthetic`
chip_information	`synthetic CPU`
mpi_size	`8`
num_nodes	`2`
omp_num_threads	`1`
platform	`synthetic`
scope_profiler_version	`sample`
timestamp	`2026-10-17T12:00:00Z`
total_cores	`8`