Repository navigation
Conversation
cima22
requested review from
f3sch,
fprino,
mconcas,
mcoquet642 and
shahor02
as code owners
October 9, 2026 08:39
…and propate cells towards beamline
cima22
force-pushed
the
parallel-vertexer
branch
from
October 9, 2026 09:39
2ca6451 to
4c28bdc
Compare
Collaborator
|
Error while checking build/O2/fullCI_slc9 for 2ca6451 at 2026-10-09 12:31: Full log here. |
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR touches only the new seeding vertexer:
Clamp on the adaptive low-multiplicity debris threshold. Previously the threshold was free to move. It is now clamped to
[2, suppressLowMultDebris](suppressLowMultDebris = 16by default).Overpopulated-ROF pruning. The new vertexer is more sensitive to low-multiplicity vertices, so a noisy ROF can produce more fake vertices. If a ROF has more vertices than
mean + nSigma · sqrt(mean + 1), where mean is the per-ROF vertex count of the TF, the adaptive debris threshold is replaced by the standardsuppressLowMultDebris = 16for that ROF only.Motivation: in real data (run
568600), 7 ROFs in 4 of 1351 TFs had a burst of noisy hits (Claude says that most likely is beam-induced background). This caused an explosion of fake vertices, 33 to 81 in a single ROF, which the old vertexer suppressed with its hard threshold. In the worst TF, tracking time rose from 15 s to 226 s. With the pruning it is back to 17 s.Cells propagated to the beamline DCA. Following Ruben's advice, cells are now propagated to their DCA to the beamline, instead of being linearised to the beamline from the innermost cluster. This improves the z residuals: on 50 Pb-Pb TFs at 50 kHz, the RMS drops from 16.6 µm to 14.7 µm. Also slightly restricted some cuts, which for the same 50-TF comparison gives fakes 150 → 125 and splits 109 → 85, at efficiency 98.91% → 98.84%.
Updated settings for running the new vertexer on Pb-Pb TFs: