CMS Driver Based Workflows for the dev-recipes branch

[mulhearn]

Eventually this will go in a Twiki, but during development this is an issue.

The dev-recipes branch now supports various configurations via cmsDriver. Some important examples are listed here:

1. Test of Stage-1 / Stage-2 unpacking: unpacks RAW from run with both Stage-1 and Stage-2 included, dumps summary of each event to screen, and summarizes total number of each type:
   cmsDriver.py L1UNPACKTEST -s RAW2DIGI,L1Reco --era=Run2_2016 --conditions=auto:run2_data -n 100 --data --filein=/store/data/Run2015D/ZeroBias/RAW/v1/000/260/627/00000/00A76FFA-0C82-E511-B441-02163E01450F.root --customise=L1Trigger/Configuration/customiseUtils.L1TStage1DigisSummary --customise=L1Trigger/Configuration/customiseUtils.L1TStage2DigisSummary --no_output

...
INFO:  L1T Summary for Stage 1 Unpacked Digis
INFO: count of non-zero candidates for each type follows:
eg:    348
tau:   61
jet:   90
sum:   331
INFO:  L1T Summary for Stage 2 Unpacked Digis
INFO: count of non-zero candidates for each type follows:
eg:    1200
tau:   1200
jet:   1190
sum:   400
muon:  0

No stage 1 muons in this test until a legacy-to-upgrade format converter is written.
No stage 2 muons in this test because the upgrade Muon system was not included in this run.

2. Ntuple Stage-1 / Stage-2 unpacked digis: unpack Stage1 and Stage2 and store in L1Extra tree:

cmsDriver.py l1Ntuple -s RAW2DIGI,L1Reco --era=Run2_2016 --customise=L1Trigger/L1TNtuples/customiseL1Ntuple.L1NtupleRAW --conditions=auto:run2_data -n 100 --data --filein=/store/data/Run2015D/ZeroBias/RAW/v1/000/260/627/00000/00A76FFA-0C82-E511-B441-02163E01450F.root --no_output

You get an ntuple "L1Ntuple.root" containing the Stage-1 output in L1ExtraTree and the Stage-2 output in L1UpgradeTree. No stage 2 muons in this test because the upgrade Muon system was not included in this run. Stage 1 muons are present because they are filled in the ntuple via legacy l1extra.

3. Re-Emulate Stage-2 Calo/Global/Muon from RAW, and dump simulated Stage-2 digis to screen:

cmsDriver.py L1UNPACKTEST -s RAW2DIGI,L1Reco --era=Run2_2016 --conditions=auto:run2_data -n 10 --data --filein=/store/data/Run2015D/ZeroBias/RAW/v1/000/260/627/00000/00A76FFA-0C82-E511-B441-02163E01450F.root --customise=L1Trigger/Configuration/customiseReEmul.L1TReEmulFromRAW --customise=L1Trigger/Configuration/customiseUtils.L1TStage2SimDigisSummary --no_output --customise=L1Trigger/Configuration/customiseUtils.L1TTurnOffUnpackStage2GtAndGmt

4. Re-Emulate Stage-2 Calo/Global/Muon from RAW, and fill an ntuple with emulated Stage-2.
   cmsDriver.py L1UNPACKTEST -s RAW2DIGI,L1Reco --era=Run2_2016 --conditions=auto:run2_data -n 100 --data --filein=/store/data/Run2015D/ZeroBias/RAW/v1/000/260/627/00000/00A76FFA-0C82-E511-B441-02163E01450F.root --customise=L1Trigger/Configuration/customiseReEmul.L1TReEmulFromRAW --customise=L1Trigger/L1TNtuples/customiseL1Ntuple.L1NtupleEMU --no_output --customise=L1Trigger/Configuration/customiseUtils.L1TTurnOffUnpackStage2GtAndGmt

[mulhearn]

These workflows are starting to look decent...

[mulhearn]

I think next is to test that work flow (4) is bitwise equivalent to reEmul.py.

[mulhearn]

I have a proposal for warnings that I'm happy to implement eventually if @jimbrooke and @apana agree:

• if an input label is set to "none", a warning at start will be issued that input will not be processed.
• module will not attempt to recover from event anything with label "none" (no per event warnings in this case)

[jimbrooke]

Sounds good to me. This was always the plan, it just never quite happened...