Add combined premixing workflows for Run 3 #27753

makortel · 2019-08-12T19:17:50Z

PR description:

On an e-mail thread on #27672 it was noticed that there are no premixing workflows in the matrix for Run 3. This PR adds them for the upgrade matrix in a generic way, and brings three of them to the standard matrix (for years 2021, 2023, 2024).

~~I have two questions for which this PR is an RFC for now.~~

PR validation:

New workflows run

NANO is configured to use step3_inMINIAODSIM.root input file, but in combined premixing workflow the file is step4_inMINIAODSIM.root. Simplest option is to just drop the NANO step from these workflows.

cmsbuild · 2019-08-12T19:18:19Z

The code-checks are being triggered in jenkins.

cmsbuild · 2019-08-12T19:25:42Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-27753/11412

This PR adds an extra 16KB to repository

cmsbuild · 2019-08-12T19:26:09Z

A new Pull Request was created by @makortel (Matti Kortelainen) for master.

It involves the following packages:

Configuration/PyReleaseValidation

@cmsbuild, @prebello, @chayanit, @zhenhu, @kpedro88, @pgunnell can you please review it and eventually sign? Thanks.
@Martin-Grunewald this is something you requested to watch as well.
@davidlange6, @slava77, @fabiocos you are the release manager for this.

cms-bot commands are listed here

Configuration/PyReleaseValidation/python/relval_upgrade.py

makortel · 2019-08-12T19:34:18Z

@cmsbuild, please test workflow 11834.99,12634.99,13034.99

cmsbuild · 2019-08-12T19:34:50Z

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/1981/console Started: 2019/08/12 21:35

cmsbuild · 2019-08-12T21:12:46Z

+1
Tested at: 9e47bb1
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-9ec1f4/1981/summary.html

cmsbuild · 2019-08-12T21:12:51Z

Comparison job queued.

cmsbuild · 2019-08-12T23:34:51Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-9ec1f4/1981/summary.html

@slava77 comparisons for the following workflows were not done due to missing matrix map:

/build/cmsbld/jenkins/workspace/compare-root-files-short-matrix/results/JR-comparison/PR-9ec1f4/11834.99_TTbar_14TeV+TTbar_14TeV_TuneCUETP8M1_2021PU_GenSimFull+SingleNuE10_cf_2021PU_PremixFullPU_Premix+DigiFullPUPRMXCombined_2021PU+RecoFullPUPRMX_2021PU+HARVESTFullPU_2021PU
/build/cmsbld/jenkins/workspace/compare-root-files-short-matrix/results/JR-comparison/PR-9ec1f4/12634.99_TTbar_14TeV+TTbar_14TeV_TuneCUETP8M1_2023PU_GenSimFull+SingleNuE10_cf_2023PU_PremixFullPU_Premix+DigiFullPUPRMXCombined_2023PU+RecoFullPUPRMX_2023PU+HARVESTFullPU_2023PU
/build/cmsbld/jenkins/workspace/compare-root-files-short-matrix/results/JR-comparison/PR-9ec1f4/13034.99_TTbar_14TeV+TTbar_14TeV_TuneCUETP8M1_2024PU_GenSimFull+SingleNuE10_cf_2024PU_PremixFullPU_Premix+DigiFullPUPRMXCombined_2024PU+RecoFullPUPRMX_2024PU+HARVESTFullPU_2024PU

Comparison Summary:

No significant changes to the logs found
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 34
DQMHistoTests: Total histograms compared: 2939508
DQMHistoTests: Total failures: 1
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 2939166
DQMHistoTests: Total skipped: 341
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 33 files compared)
Checked 142 log files, 14 edm output root files, 34 DQM output files

makortel · 2019-08-13T13:19:26Z

An additional question that came to my mind: would it make sense to replace the combined premixing workflow in the limited matrix with the one for 2021?

fabiocos · 2019-08-13T14:40:33Z

@makortel in future perspective yes IMO, I would say that for the UL reprocessing that part is now frozen (@civanch @mdhildreth @srimanob may comment further)
From the RECO point of view I do not think this should matter @slava77 @perrotta please comment in case

cmsbuild · 2019-08-17T17:42:50Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-9ec1f4/2065/summary.html

Comparison Summary:

No significant changes to the logs found
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 34
DQMHistoTests: Total histograms compared: 2939508
DQMHistoTests: Total failures: 1
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 2939166
DQMHistoTests: Total skipped: 341
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 33 files compared)
Checked 142 log files, 14 edm output root files, 34 DQM output files

chayanit · 2019-08-20T04:55:45Z

Done. Now .98 and .99 are defined for all PU workflows for the years 2021, 2023, 2024, and 2026 for the "upgrade" matrix (but only TTbar .99 are still brought in "2017" matrix).

There are still couple of issues to be addressed before .98 (and .97) become generally usable

The --pileup_input is not currently specified for .98 workflows. I suppose something similar to the MinBias-for-pileup sample treatment for "upgrade" matrix https://github.com/cms-sw/cmssw/blob/4a44bfa59ce3f947e226cb6ee398d350e6b5e271/Configuration/PyReleaseValidation/python/relval_steps.py#L3011-#L3054

Right. I have noticed this as well. will look how to implement it in.

would be needed (or the mechanism extended to cover premixed pileup library as well)

The "GEN fragment" name in .97 is "SingleNuE10", whereas in "premix" matrix the PREMIXUP18_PU25 (etc) notation is used. This string ends up in the dataset name in DBS, right? Should the PREMIX* be kept there (or at least attempted to)? (I find some value from the pileup library dataset name being different from the usual datasets)

Yes. I found this part is a bit tricky as .97 seems to take GEN fragment as you pointed. Any suggestion?

chayanit · 2019-08-20T04:57:00Z

+1

makortel · 2019-08-20T15:23:07Z

The "GEN fragment" name in .97 is "SingleNuE10", whereas in "premix" matrix the PREMIXUP18_PU25 (etc) notation is used. This string ends up in the dataset name in DBS, right? Should the PREMIX* be kept there (or at least attempted to)? (I find some value from the pileup library dataset name being different from the usual datasets)

Yes. I found this part is a bit tricky as .97 seems to take GEN fragment as you pointed. Any suggestion?

I might have found a way to replace "SingleNuE10" with "PREMIX" in the workflow name, but let's leave that to a subsequent PR.

From where exactly is the dataset name picked up? E.g. the directory created by runTheMatrix has the form <num>_<name>+<step>[+<step2>], is it the <name> part? Or from one of the steps?

kpedro88 · 2019-08-22T16:57:11Z

+upgrade

cmsbuild · 2019-08-22T16:57:42Z

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. @davidlange6, @slava77, @smuzaffar, @fabiocos (and backports should be raised in the release meeting by the corresponding L2)

kpedro88 · 2019-08-22T22:42:10Z

@makortel is this ready to be merged?

makortel · 2019-08-23T13:28:49Z

@kpedro88 Yes.

(the only follow up that I recall would be #27753 (comment), and that I'd really leave for a subsequent PR)

kpedro88 · 2019-08-23T13:47:59Z

+1

chayanit · 2019-11-23T13:27:39Z

The "GEN fragment" name in .97 is "SingleNuE10", whereas in "premix" matrix the PREMIXUP18_PU25 (etc) notation is used. This string ends up in the dataset name in DBS, right? Should the PREMIX* be kept there (or at least attempted to)? (I find some value from the pileup library dataset name being different from the usual datasets)

Yes. I found this part is a bit tricky as .97 seems to take GEN fragment as you pointed. Any suggestion?

I might have found a way to replace "SingleNuE10" with "PREMIX" in the workflow name, but let's leave that to a subsequent PR.

From where exactly is the dataset name picked up? E.g. the directory created by runTheMatrix has the form <num>_<name>+<step>[+<step2>], is it the <name> part? Or from one of the steps?

@makortel I would like to come back to this after we had been busy on UL campaigns for several months. Could you point me how to replace "SingleNuE10" with "PREMIX" in the workflow name and change the dataset name to have "PREMIX" instead?

Now Phase2 (2026) workflows also start testing premixing workflows but as we noticed that *.98 still need to include --pileup_input.

makortel · 2019-11-25T16:54:57Z

I would like to come back to this after we had been busy on UL campaigns for several months. Could you point me how to replace "SingleNuE10" with "PREMIX" in the workflow name and change the dataset name to have "PREMIX" instead?

Let me try to find out what I did back then...

makortel · 2019-11-25T22:23:37Z

@chayanit
Let's move the discussion to #28470.

makortel added 2 commits August 12, 2019 19:49

Add (combined) premixing workflows for Run 3

a51708e

Drop NANO step from upgrade combined premixing workflow

9e47bb1

NANO is configured to use step3_inMINIAODSIM.root input file, but in combined premixing workflow the file is step4_inMINIAODSIM.root. Simplest option is to just drop the NANO step from these workflows.

cmsbuild added this to the CMSSW_11_0_X milestone Aug 12, 2019

cmsbuild added code-checks-pending comparison-pending orp-pending pdmv-pending pending-signatures tests-pending upgrade-pending labels Aug 12, 2019

cmsbuild added code-checks-approved and removed code-checks-pending labels Aug 12, 2019

makortel commented Aug 12, 2019

View reviewed changes

Configuration/PyReleaseValidation/python/relval_upgrade.py Show resolved Hide resolved

makortel commented Aug 12, 2019

View reviewed changes

Configuration/PyReleaseValidation/python/relval_upgrade.py Outdated Show resolved Hide resolved

cmsbuild added tests-started and removed tests-pending labels Aug 12, 2019

cmsbuild added tests-approved and removed tests-started labels Aug 12, 2019

cmsbuild added comparison-available and removed comparison-pending labels Aug 12, 2019

cmsbuild added comparison-available and removed comparison-pending labels Aug 17, 2019

cmsbuild mentioned this pull request Aug 19, 2019

Toplhefilter #27795

Closed

cmsbuild added pdmv-approved and removed pdmv-pending labels Aug 20, 2019

cmsbuild mentioned this pull request Aug 20, 2019

NanoAOD step added in upgrade 2017 samples #27806

Merged

cmsbuild mentioned this pull request Aug 22, 2019

Add Skewed Phase 2 Inner Tracker geometry (T16). #27824

Merged

cmsbuild added fully-signed upgrade-approved and removed pending-signatures upgrade-pending labels Aug 22, 2019

cmsbuild mentioned this pull request Aug 22, 2019

Phase2-hcx214 Add a new scenario and workflow with latest Phase2 configuration including HFNose #27665

Merged

cmsbuild added orp-approved and removed orp-pending labels Aug 23, 2019

cmsbuild merged commit dc71c16 into cms-sw:master Aug 23, 2019

This was referenced Aug 23, 2019

python 3 fix for GlobalTag.py #27849

Merged

Replace inconsistent tabbing with spaces in MatrixInector.py #27859

Merged

makortel mentioned this pull request Nov 25, 2019

Use PREMIX in the name of the .97 workflow #28470

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add combined premixing workflows for Run 3 #27753

Add combined premixing workflows for Run 3 #27753

makortel commented Aug 12, 2019 •

edited

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

makortel commented Aug 12, 2019

cmsbuild commented Aug 12, 2019 •

edited

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

makortel commented Aug 13, 2019

fabiocos commented Aug 13, 2019

cmsbuild commented Aug 17, 2019

chayanit commented Aug 20, 2019

chayanit commented Aug 20, 2019

makortel commented Aug 20, 2019

kpedro88 commented Aug 22, 2019

cmsbuild commented Aug 22, 2019

kpedro88 commented Aug 22, 2019

makortel commented Aug 23, 2019

kpedro88 commented Aug 23, 2019

chayanit commented Nov 23, 2019

makortel commented Nov 25, 2019

makortel commented Nov 25, 2019

Add combined premixing workflows for Run 3 #27753

Add combined premixing workflows for Run 3 #27753

Conversation

makortel commented Aug 12, 2019 • edited

PR description:

PR validation:

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

makortel commented Aug 12, 2019

cmsbuild commented Aug 12, 2019 • edited

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

cmsbuild commented Aug 12, 2019

makortel commented Aug 13, 2019

fabiocos commented Aug 13, 2019

cmsbuild commented Aug 17, 2019

chayanit commented Aug 20, 2019

chayanit commented Aug 20, 2019

makortel commented Aug 20, 2019

kpedro88 commented Aug 22, 2019

cmsbuild commented Aug 22, 2019

kpedro88 commented Aug 22, 2019

makortel commented Aug 23, 2019

kpedro88 commented Aug 23, 2019

chayanit commented Nov 23, 2019

makortel commented Nov 25, 2019

makortel commented Nov 25, 2019

makortel commented Aug 12, 2019 •

edited

cmsbuild commented Aug 12, 2019 •

edited