MTD reconstruction validation: tracks and vertices

[fabiocos]

PR description:

This PR extends the previous work integrated in #35482 in several ways:

• cleaning of existing plots for 4D vertex validation;
• move into vertex class plots for vertex validation previously in track validation;
• addition in track validation class of plots based on a reco-sim truth matching using GenParticles associated to the true PV, as done for the selection sued to compute the MVA MTD quality flag.

An harvester has been introduced also for MtdTracksValidation class

PR validation:

Test wf 34634.0 and 34834.0 (PU privately built) runs and produces expected plots.

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-35989/26419

• This PR adds an extra 48KB to repository

[cmsbuild]

A new Pull Request was created by @fabiocos (Fabio Cossutti) for master.

It involves the following packages:

• Validation/MtdValidation (dqm)

@emanueleusai, @ahmad3213, @cmsbuild, @jfernan2, @pmandrik, @pbo0, @rvenditti can you please review it and eventually sign? Thanks.
@perrotta, @dpiparo, @qliphy you are the release manager for this.

cms-bot commands are listed here

[fabiocos]

please test

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-2ba253/20258/summary.html
COMMIT: 22b242c
CMSSW: CMSSW_12_2_X_2021-11-04-1100/slc7_amd64_gcc900
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/35989/20258/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: 4 differences found in the comparisons
• DQMHistoTests: Total files compared: 42
• DQMHistoTests: Total histograms compared: 2901845
• DQMHistoTests: Total failures: 15
• DQMHistoTests: Total nulls: 33
• DQMHistoTests: Total successes: 2901775
• DQMHistoTests: Total skipped: 22
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 43.731 KiB( 41 files compared)
• DQMHistoSizes: changed ( 23234.0,... ): 5.390 KiB MTD/Vertices
• DQMHistoSizes: changed ( 23234.0,... ): 3.357 KiB MTD/Tracks
• DQMHistoSizes: changed ( 312.0 ): -0.004 KiB MessageLogger/Warnings
• Checked 177 log files, 37 edm output root files, 42 DQM output files
• TriggerResults: no differences found

[fabiocos]

@gsorrentino18 FYI

[fabiocos]

Any comment? Otherwise can we move forward?

[emanueleusai]

@fabiocos can you please greenlight the new MEs here?
https://cmssdt.cern.ch/SDT/jenkins-artifacts/baseLineComparisons/CMSSW_12_2_X_2021-11-04-1100+2ba253/46670/dqm-histo-comparison-summary.html
especially:
https://tinyurl.com/yzesl6lu and https://tinyurl.com/ykxfjbw9

Some of the new ME appear to be empty, is this because of the small statistics of the unit tests?

[fabiocos]

@emanueleusai I did extensive checks of the histograms, and they look reasonable to me. Besides the newly added histograms, those removed or whose definition has been changed, all the tests involving multiple reconstructed vertices are empty if there is just a single reconstructed vertex (for no PU samples).

For samples with PU, some histograms (DeltaZ and DeltaT between vertices) may anyway have a limited number of entries for low PU. In particular DeltaT histograms are filled only for vertices close in Z, see for instance https://github.com/cms-sw/cmssw/pull/35989/files#diff-ad794caa0fa59947d715316dd1617c612e41544f6f43043f13b5a28321e81d1fR1406 .This could explain why even for PU samples some of these histograms are not filled. The idea is to monitor to which extent the time information may separate vertices very close in Z.

To give you an idea, in a private test with PU = 200, for 5 events I find just 1 entry for DeltaTfakefake, and 71 for DeltaTfakereal.

[fabiocos]

DeltaTfakefake.pdf

100 events with PU = 200 (comparing this PR and an additional update to the time reconstruction to be proposed soon)

[emanueleusai]

Thank you very much for the feedback!

[emanueleusai]

+1

[cmsbuild]

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 now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @qliphy (and backports should be raised in the release meeting by the corresponding L2)

[qliphy]

+1