perf(detray): Avoid recomputation of tangential ray

[stephenswat]

The detray navigation functions currently recompute the tangential ray multiple times, even if this is not necessary. This commit changes the behaviour to compute the tangential once and then thread it through the visitor pattern.

[stephenswat]

Performance summary

Here is a summary of the performance effects of this PR:

Graphical

Tabular

Kernel	Reciprocal Throughput			Parallelism
	8fe5044	954ba3a	Delta	8fe5044	954ba3a
propagate_to_next_surface	6.72 ms	6.54 ms	-2.6%	3.61	3.62
find_tracks	1.19 ms	1.19 ms	-0.1%	1.84	1.83
count_doublets	818.27 μs	818.61 μs	0.0%	1.61	1.61
count_triplets	566.37 μs	566.93 μs	0.1%	1.02	1.02
find_doublets	542.44 μs	534.47 μs	-1.5%	3.08	3.08
ccl_kernel	434.06 μs	435.39 μs	0.3%	1.71	1.71
Thrust::sort	384.77 μs	384.87 μs	0.0%	7.28	7.28
condense_tracks	250.89 μs	250.28 μs	-0.2%	5.75	5.76
find_triplets	169.09 μs	170.06 μs	0.6%	1.31	1.32
build_tracks	124.61 μs	124.98 μs	0.3%	3.71	3.71
select_seeds	59.22 μs	58.94 μs	-0.5%	1.34	1.34
estimate_track_params	26.46 μs	26.43 μs	-0.1%	2.69	2.68
populate_grid	23.93 μs	23.92 μs	-0.0%	1.22	1.22
count_grid_capacities	22.04 μs	22.03 μs	-0.1%	1.22	1.22
remove_duplicates	19.75 μs	19.97 μs	1.1%	25.81	25.64
fill_sorted_measurements	16.15 μs	16.20 μs	0.3%	1.13	1.13
update_triplet_weights	14.79 μs	14.89 μs	0.7%	1.27	1.27
fill_finding_propagation_sort_keys	8.78 μs	8.76 μs	-0.2%	7.66	7.66
form_spacepoints	8.29 μs	8.26 μs	-0.4%	1.49	1.49
reduce_triplet_counts	5.59 μs	5.65 μs	1.1%	3.09	3.08
unknown	5.04 μs	5.05 μs	0.1%	4.28	4.28
fill_finding_duplicate_removal_sort_keys	1.56 μs	1.56 μs	-0.0%	38.12	38.13
Thrust::scan	1.05 μs	1.05 μs	-0.0%	147.82	147.81
Total	11.41 ms	11.23 ms	-1.6%	3.22	3.22

Important

All metrics in this report are given as reciprocal throughput, not as wallclock runtime.

Note

This is an automated message produced upon the explicit request of a human being.

[github-actions]

📊: Physics performance monitoring for aaa035f

Full contents

physmon summary

• ✅ Particles fatras
• ✅ Particles geant4
• ✅ Particles ttbar
• ✅ Vertices ttbar
• ✅ Truth tracking (KF)
• ✅ Truth tracking (GSF)
• ✅ Truth tracking (GX2F)
• ✅ Truth tracking (KF refit)
• ✅ Truth tracking (GSF refit)
• ✅ CKF finding performance | trackfinding | single muon | truth smeared seeding
• ✅ CKF fitting performance | trackfinding | single muon | truth smeared seeding
• ✅ CKF track summary | trackfinding | single muon | truth smeared seeding
• ✅ Seeding trackfinding | single muon | truth estimated seeding
• ✅ CKF finding performance | trackfinding | single muon | truth estimated seeding
• ✅ CKF fitting performance | trackfinding | single muon | truth estimated seeding
• ✅ CKF track summary | trackfinding | single muon | truth estimated seeding
• ✅ Seeding trackfinding | single muon | default seeding
• ✅ CKF finding performance | trackfinding | single muon | default seeding
• ✅ CKF fitting performance | trackfinding | single muon | default seeding
• ✅ CKF track summary | trackfinding | single muon | default seeding
• ✅ Seeding trackfinding | single muon | orthogonal seeding
• ✅ CKF finding performance | trackfinding | single muon | orthogonal seeding
• ✅ CKF fitting performance | trackfinding | single muon | orthogonal seeding
• ✅ CKF track summary | trackfinding | single muon | orthogonal seeding
• ✅ Seeding trackfinding | 4 muon x 50 vertices | default seeding
• ✅ CKF finding performance | trackfinding | 4 muon x 50 vertices | default seeding
• ✅ CKF fitting performance | trackfinding | 4 muon x 50 vertices | default seeding
• ✅ CKF track summary | trackfinding | 4 muon x 50 vertices | default seeding
• ✅ Ambisolver finding performance | trackfinding | 4 muon x 50 vertices | default seeding
• ✅ IVF notime | trackfinding | 4 muon x 50 vertices | default seeding
• ✅ AMVF gauss notime | trackfinding | 4 muon x 50 vertices | default seeding
• ✅ AMVF grid time | trackfinding | 4 muon x 50 vertices | default seeding
• ✅ Seeding trackfinding | ttbar with 200 pileup | default seeding
• ✅ CKF finding performance | trackfinding | ttbar with 200 pileup | default seeding
• ✅ CKF fitting performance | trackfinding | ttbar with 200 pileup | default seeding
• ✅ CKF track summary | trackfinding | ttbar with 200 pileup | default seeding
• ✅ Ambisolver finding performance | trackfinding | ttbar with 200 pileup | default seeding
• ✅ ML Ambisolver | trackfinding | ttbar with 200 pileup | default seeding
• ✅ AMVF gauss notime | trackfinding | ttbar with 200 pileup | default seeding
• ✅ AMVF grid time | trackfinding | ttbar with 200 pileup | default seeding
• 🔵️ Comparison - Truth tracking (GX2F vs KF)

❗️: Downstream build failure

• eic-shell EICrecon (cc @acts-project/epic-contacts)

[stephenswat]

Performance summary

Here is a summary of the performance effects of this PR:

Graphical

Tabular

Kernel	Reciprocal Throughput			Parallelism
	c6fbf62	dc8d410	Delta	c6fbf62	dc8d410
propagate_to_next_surface	5.21 ms	5.13 ms	-1.6%	4.38	4.39
find_tracks	1.19 ms	1.18 ms	-0.3%	1.84	1.84
count_doublets	811.35 μs	812.61 μs	0.2%	1.61	1.61
count_triplets	567.83 μs	566.39 μs	-0.3%	1.02	1.02
find_doublets	528.02 μs	533.10 μs	1.0%	3.08	3.08
ccl_kernel	433.83 μs	434.89 μs	0.2%	1.71	1.71
Thrust::sort	384.66 μs	384.78 μs	0.0%	7.28	7.28
condense_tracks	250.33 μs	250.61 μs	0.1%	5.76	5.76
find_triplets	168.30 μs	168.58 μs	0.2%	1.32	1.32
build_tracks	124.66 μs	125.00 μs	0.3%	3.71	3.71
select_seeds	59.42 μs	58.57 μs	-1.4%	1.34	1.34
estimate_track_params	26.51 μs	26.48 μs	-0.1%	2.69	2.68
populate_grid	23.94 μs	23.92 μs	-0.1%	1.22	1.22
count_grid_capacities	22.03 μs	22.03 μs	0.0%	1.22	1.22
remove_duplicates	19.89 μs	19.86 μs	-0.1%	25.73	25.71
fill_sorted_measurements	16.12 μs	16.24 μs	0.7%	1.13	1.13
update_triplet_weights	14.70 μs	14.79 μs	0.6%	1.27	1.27
fill_finding_propagation_sort_keys	8.77 μs	8.79 μs	0.2%	7.65	7.65
form_spacepoints	8.37 μs	8.31 μs	-0.7%	1.49	1.49
reduce_triplet_counts	5.60 μs	5.60 μs	-0.1%	3.08	3.08
unknown	5.05 μs	5.04 μs	-0.2%	4.27	4.28
fill_finding_duplicate_removal_sort_keys	1.56 μs	1.56 μs	-0.2%	38.09	38.15
Thrust::scan	1.06 μs	1.06 μs	0.1%	147.81	147.70
Total	9.88 ms	9.80 ms	-0.8%	3.57	3.57

Important

All metrics in this report are given as reciprocal throughput, not as wallclock runtime.

Note

This is an automated message produced upon the explicit request of a human being.

[github-actions]

Public API surface diff

No change to the public API surface. ✅

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
33.3% Duplication on New Code

See analysis details on SonarQube Cloud