AK5 to AK4 migration with new Global Tags

[diguida]

Merge of #5588 and #5662

[cmsbuild]

A new Pull Request was created by @diguida (Salvatore Di Guida) for CMSSW_7_2_X.

AK5 to AK4 migration with new Global Tags

It involves the following packages:

CommonTools/ParticleFlow
Configuration/AlCa
DQM/Physics
DQMOffline/JetMET
DQMOffline/RecoB
FastSimulation/ParticleFlow
JetMETCorrections/Configuration
JetMETCorrections/Modules
JetMETCorrections/Type1MET
PhysicsTools/PatAlgos
RecoJets/JetProducers
RecoTauTag/Configuration
TopQuarkAnalysis/Configuration
TopQuarkAnalysis/TopJetCombination
TopQuarkAnalysis/TopObjectResolutions
Validation/RecoJets
Validation/RecoParticleFlow

@civanch, @diguida, @StoyanStoynev, @lveldere, @danduggan, @mdhildreth, @monttj, @cmsbuild, @cerminar, @nclopezo, @rcastello, @deguio, @slava77, @vadler, @ojeda, @rovere can you please review it and eventually sign? Thanks.
@ghellwig, @mmarionncern, @imarches, @yslai, @acaudron, @jpavel, @TaiSakuma, @nhanvtran, @schoef, @ferencek, @pvmulder, @rociovilar this is something you requested to watch as well.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.
@nclopezo you are the release manager for this.
You can merge this pull request by typing 'merge' in the first line of your comment.

[StoyanStoynev]

@cmsbuild please test

[cmsbuild]

+1
Tested at: a9c40a3
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-5666/1110/summary.html

[slava77]

the GT changes show up only in Run1 MC: everything significant seems to start from RPC hits, the effects then propagate downstream to muons (mostly feature-less change in chi2 and number of hits), with negligible changes further down in the PF candidates and later. Apparently, muons with most notable variations don't make it through PF selections.

[slava77]

+1

for #5666 a9c40a3

tested in CMSSW_7_2_X_2014-10-03-0200 (test are sign436).
I tested separately the effects of the GT (affect physics in run1 MC) and the rest of the AK5/AK4 code (doesn't affect reco outputs, affects validation).

[diguida]

@slava77
thanks for reporting: we are investigating.

[diguida]

@slava77
The changes in the RPC hits are expected: as a follow-up of the data-MC workshop, we were requested to update the RPC efficiency and noise with values in agreement with 2012 conditions.
https://cms-conddb.cern.ch/gtList/GTdiff.html?GlobalTag=PRE_STA72_V4&GlobalTag2=PRE_STA72_V6
I am adding @pietverwilligen who can tell us more

[slava77]

@monttj @vadler
Volker and Tae Jeong we will need your signature here, if you don't have any further comments.

[diguida]

@slava77 I cannot see the changes you describe in the comparison plots attached to this PR (namely 5.1, 25.0). Is there a way to reproduce them?
@abbiendi @battibass @trocino as you can see in this discussion, RPC DPG requested a change of the RPC efficiencies and noise with values in agreement with 2012 conditions, as a follow-up of the data-MC workshop. This is affecting physics performance on Run1 simulations.

[slava77]

5.1 has them:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/baseLineComparisons/CMSSW_7_2_X_2014-10-04-1400+5666/4126/validateJR/all_OldVSNew_TTbarFSwf5p1/

I don't see diffs in 25.0 in the same place, but I have them in my local tests.
This seems inconsistent within jenkins.

Strange.

My tests for the complete PR were run on 03-Oct-2014 around 9pm.
Did the content of the database change, as of 05-Oct-2014 10AM when jenkins jobs were running?

@cmsbuild
could you please rerun the tests just to double-check.

[slava77]

in 25.0 (TTbar run 1) I see

• RPC hit cluster size
  
• globalMuons chi2
  

Same with higher stats in 20.0 (single mu pt 10, 2K events)

[diguida]

@slava77
Thanks for pointing me to the correct plots. I was looking at the RelMon summary for wf 5.1, where no differences for Muons are found.

Did the content of the database change, as of 05-Oct-2014 10AM when jenkins jobs were running?

No, the tag is there since 2012 and the Global Tag mapping is immutable.