Implement MIP Timing Detector reconstruction sequence for Phase-2 HLT by mmusich · Pull Request #51170 · cms-sw/cmssw

mmusich · 2026-06-09T17:28:16Z

PR description:

The goal of this PR is provide the configuration to run the MTD reconstruction (from local reconstruction down to 4D vertexing with mass hypothesis) in the Phase-2 HLT. The configuration parameters are taken from the respective offline sequences.
Such setup can be enabled by using the novel mtd_at_hlt process modifier and it is not active by default.
This development is purely meant to help POG developers when willing to use MTD information at HLT.
To this effect I provide a new runTheMatrix workflow (with offset .7522) in order to be able to test it. The workflow is for convenience included in the ph2_hlt relval matrix.
Together with the HLT configuration I also enable basic 4D vertexing validation (together the minimal changes needed to make it run together with the HLT step). Some configuration is provided to be able to run also the MTDTrackValidation but more work is needed to enable in workflows it in order to protect failures upon missing input (that might happen in the HLT as we don't run the same reconstruction on all the events).

PR validation:

I've run successfully the new workflow runTheMatrix.py -l 34434.7522 -t 4 -j 8 and observed in the resulting DQM file the expected plots for the 4D vertices.
I also benchmarked this configuration using a NGT node equipped with 2x AMD EPYC 9534 64-Core Processor + 4x NVIDIA L40S GPUs using 16 jobs with 16 threads/streams each by using the following script:

#! /bin/bash                                                                                                                                                                                               
LOCALPATH='/cms-hlt-nfs/data/relValTTbar_200PU/CMSSW_16_0_0_pre1'
echo "Input source: |${LOCALPATH}|"
LOCALFILES=$(ls -1 ${LOCALPATH})
ALL_FILES=""
for f in ${LOCALFILES[@]}; do
    ALL_FILES+="file:${LOCALPATH}/${f},"
done
# Remove the last character                                                                                                                                               
ALL_FILES="${ALL_FILES%?}"
echo "Discovered files: $ALL_FILES"

### Timing menu, current status
cmsDriver.py Phase2 -s L1P2GT,HLT:75e33_timing \
             --processName=HLTX \
             --conditions auto:phase2_realistic_T35 \
             --geometry ExtendedRun4D110 \
             --era Phase2C17I13M9 \
             --customise SLHCUpgradeSimulations/Configuration/aging.customise_aging_1000 \
             --eventcontent FEVTDEBUGHLT \
             --filein=$ALL_FILES \
             --mc \
             --inputCommands="keep *, drop *_hlt*_*_HLT, drop triggerTriggerFilterObjectWithRefs_l1t*_*_HLT" \
             -n -1 \
             --no_exec \
             --output={} \
             --python_filename HLTTimingCurrent_config.py

### Timing menu with MTD
cmsDriver.py Phase2 -s L1P2GT,HLT:75e33_timing \
             --processName=HLTX \
             --conditions auto:phase2_realistic_T35 \
             --geometry ExtendedRun4D110 \
             --era Phase2C17I13M9 \
             --procModifiers mtd_at_hlt  \
             --customise SLHCUpgradeSimulations/Configuration/aging.customise_aging_1000 \
             --eventcontent FEVTDEBUGHLT \
             --filein=$ALL_FILES \
             --mc \
             --inputCommands="keep *, drop *_hlt*_*_HLT, drop triggerTriggerFilterObjectWithRefs_l1t*_*_HLT" \
             -n -1 \
             --no_exec \
             --output={} \
             --python_filename HLTTimingMTDReco_config.py

The resulting pies are available here:

Vanilla Ph2 Timing menu link to pie	Ph2 Timing menu w/ MTD link to pie

The overall cost of the addition is definitely not negligible +536 ms, out of which:

~315 ms are coming from the MTD reconstruction sequence (unassigned in the plots above), with hltTrackExtenderWithMTD being the largest contributor.
~153 ms are coming from the vertexing (131ms of hltUnsortedOfflinePrimaryVertices4D)

If this PR is a backport please specify the original PR and why you need to backport that PR. If this PR will be backported please specify to which release cycle the backport is meant for:

Not a backport, it might be backported to CMSSW_20_0_X for MC production purposes.

Cc: @cms-sw/mtd-dpg-l2

cmsbuild · 2026-06-09T17:28:46Z

cms-bot internal usage

cmsbuild · 2026-06-09T17:30:09Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51170/49689

There are other open Pull requests which might conflict with changes you have proposed:
- File Configuration/PyReleaseValidation/python/relval_Run4.py modified in PR(s): Phase2-hgx367 Try to create a scenario with T37, C27, N17, I22, F9 #51080
- File Configuration/PyReleaseValidation/python/upgradeWorkflowComponents.py modified in PR(s): Rpc digi dev v9 #47447, [WIP] Prototype TruthInfo graph for truth history navigation and association studies #50913, Displaced Particle Gun #51006, Phase2-hgx367 Try to create a scenario with T37, C27, N17, I22, F9 #51080, Simplification in the definition of Phase 2 HLT workflows #51142
- File Configuration/PyReleaseValidation/scripts/runTheMatrix.py modified in PR(s): [Do not merge] For cms-bot PR testing #48637
- File HLTrigger/Configuration/python/HLT_75e33/modules/hltInitialStepTrackSelectionHighPurity_cfi.py modified in PR(s): Track Purity DNN for Phase-2 HLT #51084
- File HLTrigger/Configuration/python/HLT_75e33/modules/hltPfTICL_cfi.py modified in PR(s): ME0 integration in PFMuon algorithm #51053
- File HLTrigger/Configuration/python/HLT_75e33_cff.py modified in PR(s): Phase2 Single_Tau_Trigger Path Added #49637, LST: add LSTGeometry package and associated ESProducer #50679, [NGT] DNN for HP Pixel track selection #51042, Track Purity DNN for Phase-2 HLT #51084
- File HLTrigger/Configuration/python/HLT_75e33_timing_cff.py modified in PR(s): Phase2 Single_Tau_Trigger Path Added #49637, LST: add LSTGeometry package and associated ESProducer #50679, [NGT] DNN for HP Pixel track selection #51042, Track Purity DNN for Phase-2 HLT #51084
- File HLTrigger/Configuration/python/HLT_NGTScouting_cff.py modified in PR(s): LST: add LSTGeometry package and associated ESProducer #50679, [NGT] DNN for HP Pixel track selection #51042
- File HLTriggerOffline/Common/python/HLTValidationHarvest_cff.py modified in PR(s): Particle Flow cluster validation in ECAL #50292, [NGT] Implement new strategy for Tau Validation at HLT and RECO in Phase-2 #51092
- File HLTriggerOffline/Common/python/HLTValidation_cff.py modified in PR(s): Particle Flow cluster validation in ECAL #50292, [NGT] Implement new strategy for Tau Validation at HLT and RECO in Phase-2 #51092

cmsbuild · 2026-06-09T17:30:37Z

A new Pull Request was created by @mmusich for master.

It involves the following packages:

Configuration/ProcessModifiers (operations)
Configuration/PyReleaseValidation (pdmv)
HLTrigger/Configuration (hlt)
HLTriggerOffline/Common (dqm)
Validation/MtdValidation (dqm)

@AdrianoDee, @DickyChant, @Martin-Grunewald, @antoniovagnerini, @cmsbuild, @ctarricone, @davidlange6, @fabiocos, @ftenchini, @gabrielmscampos, @kfjack, @mandrenguyen, @miquork, @mmusich, @rseidita, @sroychow can you please review it and eventually sign? Thanks.
@Martin-Grunewald, @ReyerBand, @SohamBhattacharya, @VourMa, @argiro, @fabiocos, @makortel, @martinamalberti, @missirol, @mtosi, @rchatter, @rovere, @slomeo, @thomreis, @wang0jin this is something you requested to watch as well.
@ftenchini, @mandrenguyen, @sextonkennedy you are the release manager for this.

cms-bot commands are listed here

mmusich · 2026-06-09T17:31:05Z

test parameters:

workflows = ph2_hlt

mmusich · 2026-06-09T17:31:12Z

@cmsbuild, please test

cmsbuild · 2026-06-09T19:19:00Z

+1

Size: This PR adds an extra 252KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-b399f2/53790/summary.html
COMMIT: f28e835
CMSSW: CMSSW_20_1_X_2026-06-09-1100/el9_amd64_gcc13
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/51170/53790/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

You potentially added 2 lines to the logs
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 57
DQMHistoTests: Total histograms compared: 3647546
DQMHistoTests: Total failures: 0
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3647528
DQMHistoTests: Total skipped: 18
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 56 files compared)
Checked 242 log files, 197 edm output root files, 57 DQM output files

mmusich · 2026-06-10T15:33:57Z

test parameters:

enable = hlt_p2_integration
workflows = ph2_hlt

mmusich · 2026-06-10T15:34:04Z

@cmsbuild, please test

…ction

… HLT menu

…FTICLProducer

cmsbuild · 2026-06-11T07:36:42Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51170/49721

There are other open Pull requests which might conflict with changes you have proposed:
- File Configuration/PyReleaseValidation/python/relval_Run4.py modified in PR(s): Phase2-hgx367 Try to create a scenario with T37, C27, N17, I22, F9 #51080
- File Configuration/PyReleaseValidation/python/upgradeWorkflowComponents.py modified in PR(s): Rpc digi dev v9 #47447, [WIP] Prototype TruthInfo graph for truth history navigation and association studies #50913, Displaced Particle Gun #51006, Phase2-hgx367 Try to create a scenario with T37, C27, N17, I22, F9 #51080, Simplification in the definition of Phase 2 HLT workflows #51142
- File Configuration/PyReleaseValidation/scripts/runTheMatrix.py modified in PR(s): [Do not merge] For cms-bot PR testing #48637
- File HLTrigger/Configuration/python/HLT_75e33/modules/hltInitialStepTrackSelectionHighPurity_cfi.py modified in PR(s): Track Purity DNN for Phase-2 HLT #51084
- File HLTrigger/Configuration/python/HLT_75e33/modules/hltPfTICL_cfi.py modified in PR(s): ME0 integration in PFMuon algorithm #51053
- File HLTrigger/Configuration/python/HLT_75e33_cff.py modified in PR(s): Phase2 Single_Tau_Trigger Path Added #49637, LST: add LSTGeometry package and associated ESProducer #50679, [NGT] DNN for HP Pixel track selection #51042, Track Purity DNN for Phase-2 HLT #51084
- File HLTrigger/Configuration/python/HLT_75e33_timing_cff.py modified in PR(s): Phase2 Single_Tau_Trigger Path Added #49637, LST: add LSTGeometry package and associated ESProducer #50679, [NGT] DNN for HP Pixel track selection #51042, Track Purity DNN for Phase-2 HLT #51084
- File HLTrigger/Configuration/python/HLT_NGTScouting_cff.py modified in PR(s): LST: add LSTGeometry package and associated ESProducer #50679, [NGT] DNN for HP Pixel track selection #51042
- File HLTriggerOffline/Common/python/HLTValidationHarvest_cff.py modified in PR(s): Particle Flow cluster validation in ECAL #50292
- File HLTriggerOffline/Common/python/HLTValidation_cff.py modified in PR(s): Particle Flow cluster validation in ECAL #50292

cmsbuild · 2026-06-11T07:37:06Z

Pull request #51170 was updated. @AdrianoDee, @DickyChant, @Martin-Grunewald, @antoniovagnerini, @cmsbuild, @ctarricone, @davidlange6, @fabiocos, @ftenchini, @gabrielmscampos, @kfjack, @mandrenguyen, @miquork, @mmusich, @rseidita, @sroychow can you please check and sign again.

mmusich · 2026-06-11T07:37:53Z

@cmsbuild, please test

cmsbuild · 2026-06-11T09:32:24Z

+1

Size: This PR adds an extra 172KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-b399f2/53855/summary.html
COMMIT: 3a2677e
CMSSW: CMSSW_20_1_X_2026-06-10-2300/el9_amd64_gcc13
Additional Tests: HLT_P2_INTEGRATION
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/51170/53855/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: 11 differences found in the comparisons
DQMHistoTests: Total files compared: 57
DQMHistoTests: Total histograms compared: 3688981
DQMHistoTests: Total failures: 408
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3688555
DQMHistoTests: Total skipped: 18
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 56 files compared)
Checked 242 log files, 197 edm output root files, 57 DQM output files

mmusich · 2026-06-11T09:37:19Z

+hlt

cmsbuild added this to the CMSSW_20_1_X milestone Jun 9, 2026

cmsbuild added dqm-pending hlt-pending operations-pending pending-signatures tests-pending orp-pending pdmv-pending code-checks-pending labels Jun 9, 2026

cmsbuild added code-checks-approved and removed code-checks-pending labels Jun 9, 2026

cmsbuild added tests-started and removed tests-pending labels Jun 9, 2026

cmsbuild added operations-approved tests-approved and removed operations-pending tests-started labels Jun 9, 2026

cmsbuild mentioned this pull request Jun 10, 2026

Simplification in the definition of Phase 2 HLT workflows #51142

Open

cmsbuild mentioned this pull request Jun 11, 2026

Phase2 Single_Tau_Trigger Path Added #49637

Open

mmusich added 3 commits June 11, 2026 09:27

port of MTD local reconstruction in the Ph2 HLT menu

e5ac1e6

add MTD global reco at HLT (TrackExtenderWithMTD)

b4427d5

introduce mtd_at_hlt_cff and use it for running fast timing reconstru…

9470255

…ction

mmusich added 4 commits June 11, 2026 09:27

add offline-like 4D vertexing reco at HLT via hlt_at_mtd modifier

02aba8c

introduce a RelVal workflow to exercise the mtd_at_hlt in the Phase-2…

cdc1717

… HLT menu

run hltTofPID in hltTofPID in HLTVertex4DRecoSequence and use it in P…

9722ec2

…FTICLProducer

add a stub of MTD Validation at HLT

3a2677e

mmusich force-pushed the mm_mtd_at_HLT branch from f28e835 to 3a2677e Compare June 11, 2026 07:34

cmsbuild added operations-pending tests-pending code-checks-pending and removed operations-approved tests-approved code-checks-approved labels Jun 11, 2026

cmsbuild added code-checks-approved and removed code-checks-pending labels Jun 11, 2026

cmsbuild added tests-started and removed tests-pending labels Jun 11, 2026

cmsbuild mentioned this pull request Jun 11, 2026

[NGT] Bug fix for tau prevalidation sequence in Run2 workflow #51198

Open

cmsbuild added operations-approved tests-approved and removed operations-pending tests-started labels Jun 11, 2026

cmsbuild added hlt-approved and removed hlt-pending labels Jun 11, 2026

Conversation

mmusich commented Jun 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR description:

PR validation:

If this PR is a backport please specify the original PR and why you need to backport that PR. If this PR will be backported please specify to which release cycle the backport is meant for:

Uh oh!

cmsbuild commented Jun 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cmsbuild commented Jun 9, 2026

Uh oh!

cmsbuild commented Jun 9, 2026

Uh oh!

mmusich commented Jun 9, 2026

Uh oh!

mmusich commented Jun 9, 2026

Uh oh!

cmsbuild commented Jun 9, 2026

Comparison Summary

Uh oh!

mmusich commented Jun 10, 2026

Uh oh!

mmusich commented Jun 10, 2026

Uh oh!

cmsbuild commented Jun 11, 2026

Uh oh!

cmsbuild commented Jun 11, 2026

Uh oh!

mmusich commented Jun 11, 2026

Uh oh!

cmsbuild commented Jun 11, 2026

Comparison Summary

Uh oh!

mmusich commented Jun 11, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

mmusich commented Jun 9, 2026 •

edited

Loading

cmsbuild commented Jun 9, 2026 •

edited

Loading