Navigation Menu

Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Disable looper track reconstruction for >= Run2_2016 #35721

Merged
merged 2 commits into from Oct 21, 2021

Conversation

mmusich
Copy link
Contributor

@mmusich mmusich commented Oct 18, 2021

PR description:

It has been suggested that disabling the "looper" 1 track reconstruction would save some reco computing time while not hindering any specific physics signal reconstruction.
A possible exception was identified with the B-Physics signals.
For this purpose a dedicated set of samples has been requested in this JIRA ticket PDMVRELVALS-128 to be studied by the BPH group.
The outcome of the study has been presented at the Tracking POG meeting of 18.08.2021.
Based on the conclusions that

"disabling loopers seems not to affect the topologies we are interested into. Some increase in the # of candidates is seen in line with expectations (@ lowp T & low eta). For some cases the performances are even better."

we proceed with the disabling for all eras >=Run2_2016, excluding the dedicated Heavy Ion ones.

PR validation:

Technically validated by running:

 runTheMatrix.py -l limited -j 4 --ibeos

if this PR is a backport please specify the original PR and why you need to backport that PR:

N/A

cc: @vmariani @AdrianoDee

Footnotes

  1. trajectories of low-energy particles (either from PU or from the underlying event of the main scattering) which are spiralling in the magnetic field within the Tracker radius. For comparison pT=0.63 GeV is the maximum pT for a charged particle to loop within the 1.1m radius of the outermost Tracker barrel layer (with B=3.8T).

@cmsbuild
Copy link
Contributor

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-35721/26042

@cmsbuild
Copy link
Contributor

A new Pull Request was created by @mmusich (Marco Musich) for master.

It involves the following packages:

  • Configuration/Eras (operations)
  • Configuration/ProcessModifiers (operations)
  • RecoTracker/IterativeTracking (reconstruction)

@perrotta, @cmsbuild, @slava77, @jpata, @qliphy, @fabiocos, @davidlange6 can you please review it and eventually sign? Thanks.
@fabiocos, @makortel, @felicepantaleo, @GiacomoSguazzoni, @JanFSchulte, @rovere, @VinInn, @Martin-Grunewald, @missirol, @ebrondol, @gpetruc, @mmusich, @mtosi, @dgulhan this is something you requested to watch as well.
@perrotta, @dpiparo, @qliphy you are the release manager for this.

cms-bot commands are listed here

@mmusich mmusich changed the title [RFC] Disable looper track reconstruction for Run3 [RFC] Disable looper track reconstruction for >= Run2_2016 Oct 19, 2021
@mmusich mmusich changed the title [RFC] Disable looper track reconstruction for >= Run2_2016 Disable looper track reconstruction for >= Run2_2016 Oct 19, 2021
@cmsbuild
Copy link
Contributor

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-35721/26074

@cmsbuild
Copy link
Contributor

Pull request #35721 was updated. @perrotta, @cmsbuild, @slava77, @jpata, @qliphy, @fabiocos, @davidlange6 can you please check and sign again.


Run2_2018_pp_on_AA = cms.ModifierChain(Run2_2018.copyAndExclude([trackingMkFitProd]), pp_on_AA, pp_on_AA_2018)
Run2_2018_pp_on_AA = cms.ModifierChain(Run2_2018.copyAndExclude([trackingMkFitProd, trackingNoLoopers]), pp_on_AA, pp_on_AA_2018)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@abaty @mandrenguyen
do you think looping track reconstruction is needed in the HI events?
I'd guess that they are even more costly in central events for CPU than for pp.

@slava77
Copy link
Contributor

slava77 commented Oct 19, 2021

@cmsbuild please test

@cmsbuild
Copy link
Contributor

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-24b217/19752/summary.html
COMMIT: 97af9c9
CMSSW: CMSSW_12_1_X_2021-10-19-1100/slc7_amd64_gcc900
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/35721/19752/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 17204 differences found in the comparisons
  • DQMHistoTests: Total files compared: 40
  • DQMHistoTests: Total histograms compared: 2751113
  • DQMHistoTests: Total failures: 65750
  • DQMHistoTests: Total nulls: 14
  • DQMHistoTests: Total successes: 2685327
  • DQMHistoTests: Total skipped: 22
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.02 KiB( 39 files compared)
  • DQMHistoSizes: changed ( 136.731 ): 0.008 KiB JetMET/SUSYDQM
  • DQMHistoSizes: changed ( 136.793 ): 0.016 KiB JetMET/SUSYDQM
  • DQMHistoSizes: changed ( 136.874 ): -0.004 KiB JetMET/SUSYDQM
  • Checked 170 log files, 37 edm output root files, 40 DQM output files
  • TriggerResults: no differences found

@slava77
Copy link
Contributor

slava77 commented Oct 20, 2021

Reco comparison results: 17204 differences found in the comparisons

As expected, changes show up starting from Run2 pp workflows, with differences starting from tracking distributions at low pt central eta region, propagating downstream

There is a small increase in the number of PF charged low-pt candidates in the central region
e.g. 2018C EGamma wf 136.874
all_OldVSNew_RunEGamma2018Cwf136p874c_log10recoPFCandidates_particleFlow__reRECO_obj_pt52

The distributions for 2016 data look somewhat similar.

in 2026 D76 PU(35?) wf 34834.999
there is some visible addition of tracks with hits with MTD
all_OldVSNew_TTbar14TeV2026D76PUwPRMXwf34834p999c_recoTracks_trackExtenderWithMTD__RECO_obj_found

The changes seem fairly roughly consistent with the primary validation done in Run3 MC

@slava77
Copy link
Contributor

slava77 commented Oct 20, 2021

+reconstruction

for #35721 97af9c9

@perrotta @qliphy
please consider this for pre5 as well.
Thank you.

@slava77
Copy link
Contributor

slava77 commented Oct 20, 2021

@perrotta @qliphy
please consider this for pre5 as well.

I checked in a local cmsrel that merges of #35721 and #35686 commute. (no conflict)

@qliphy
Copy link
Contributor

qliphy commented Oct 21, 2021

+1

@cmsbuild
Copy link
Contributor

This pull request is fully signed and it will be integrated in one of the next master IBs (tests are also fine). This pull request will be automatically merged.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants