Prepare ECAL multifit reconstruction to use different time calibration records - 132x

[thomreis]

PR description:

Adding two ESInputTag parameters to be able to select the record that the timing calibrations and timing offsets are taken from by the multifit algorithm.
This will be needed to support two records of those conditions within a GT.
Having two records in a GT will be required in some cases (e.g. MC GTs) because two different timing algorithms, using two different sets of conditions, run at the HLT and in offline reconstruction.

This PR just adds the capability to specify the record label but no changes to configuration are made and no changes in existing workflows are expected.

PR validation:

Passes limited matrix tests.

Backport of #42817

This backport is needed because the online DQM runs the CC timing algorithm in reconstruction with the HLT GT, which currently contains the ratio timing calibrations for the ratio timing running at the HLT. In the current situation the onine DQM shows bad ECAL timing distributions due to this.
With an additional EcalTimeCalibConstants record added to the HLT GT both algorithms can be given the correct calibrations with the parameters added in this PR.

[cmsbuild]

A new Pull Request was created by @thomreis (Thomas Reis) for CMSSW_13_2_X.

It involves the following packages:

• RecoLocalCalo/EcalRecProducers (reconstruction)

@cmsbuild, @jfernan2, @mandrenguyen can you please review it and eventually sign? Thanks.
@youyingli, @apsallid, @rchatter, @argiro, @missirol, @thomreis, @wang0jin this is something you requested to watch as well.
@rappoccio, @antoniovilela, @sextonkennedy you are the release manager for this.

cms-bot commands are listed here

• Backported from Prepare ECAL multifit reconstruction to use different time calibration records #42817

[thomreis]

type ecal

[thomreis]

backport of #42817

[jfernan2]

please test

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-db0601/34871/summary.html
COMMIT: 6076a12
CMSSW: CMSSW_13_2_X_2023-09-22-1100/el8_amd64_gcc11
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/42845/34871/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: 10 differences found in the comparisons
• DQMHistoTests: Total files compared: 48
• DQMHistoTests: Total histograms compared: 3198243
• DQMHistoTests: Total failures: 6
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3198215
• DQMHistoTests: Total skipped: 22
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 47 files compared)
• Checked 207 log files, 159 edm output root files, 48 DQM output files
• TriggerResults: no differences found

[jfernan2]

+1

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_13_2_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_13_3_X is complete. This pull request will now be reviewed by the release team before it's merged. @rappoccio, @antoniovilela, @sextonkennedy (and backports should be raised in the release meeting by the corresponding L2)

[mmusich]

This backport is needed because the online DQM runs the CC timing algorithm in reconstruction with the HLT GT, which currently contains the ratio timing calibrations for the ratio timing running at the HLT. In the current situation the onine DQM shows bad ECAL timing distributions due to this.

I am curious. This implies a change in the HLT reconstruction for the HI data-taking. Is this what the ECAL DPG is foreseeing?

[mmusich]

assign hlt

• pending clarification on the above

[cmsbuild]

New categories assigned: hlt

@mmusich,@missirol,@Martin-Grunewald you have been requested to review this Pull request/Issue and eventually sign? Thanks

[thomreis]

Yes you are right. If possible this is what we would like to have. But apart from an HLT update and a new release it also requires a new matching GT deployed at the same time. So we are aware that this could be difficult to get at this stage.
In any case we would like to be prepared if it is possible.
If not we have to live with the bad timing distributions in the online DQM for the HI run.

[mmusich]

Yes you are right. If possible this is what we would like to have. But apart from an HLT update and a new release it also requires a new matching GT deployed at the same time. So we are aware that this could be difficult to get at this stage.

we would need some clarifications to understand how to be prepared for this during data-taking. Would you be able to attend today's joint operations meeting at 13.00?

[thomreis]

Yes I will be there.

[mmusich]

+hlt

• this backport can go for the next data-taking release, BUT (as discussed during the Joint Ops meeting of 25th Sept 2023) before changing the default calibration in the HLT Global Tags, we need a couple of points to be addressed (ONLY in >=13.3.X):
  • the GPU uncalibrated rechit producer EcalUncalibRecHitProducerGPU (default reconstruction at HLT) needs to be updated in order to fetch conditions form a labeled payload here
  • a new release with this change + the above needs to become available in order to parse the new parameters in confDB. Subsequently a new HLT combined table can be produced (in parallel a customization to make the "old" menus to read from the unlabeled tags will be needed).
• in the meanwhile, for the online DQM use case during the 2023 HI run, the release used online for the ECAL client can be patched to use this PR + a customization of the conditions to use the desired "offline" conditions.

@cms-sw/alca-l2 please take note of the above.

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_13_2_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_13_3_X is complete. This pull request will now be reviewed by the release team before it's merged. @rappoccio, @antoniovilela, @sextonkennedy (and backports should be raised in the release meeting by the corresponding L2)

[thomreis]

For the online DQM during the I run I have created PR #42855 . It should be fine to use the default 13_2_4 release with that PR on top to use the correct tags for the CC timing algorithm using in the online DQM.

[rappoccio]

+1