Switch Voronoi subtraction from equalization based recovery of negative fluctuation to ghost-particle based clustering and energy resumming

[yslai]

This patch addresses the Voronoi subtraction performance issues. Previously, the negative fluctuation are recovered by equalization. This pull request switches to a postive energy-only clustering, where the negative fluctuation are propagated through the clustering as ghost particles, and then resummed.

The expected performance difference is a deterioration of 0.5 GeV/c or 1% energy scale difference for 50 GeV jets in the 2.76 TeV PbPb system, due to the effect of ECAL zero suppression on the underlying event - which is not handled in this implementation, as opposed to the previous equalization method.

[cmsbuild]

A new Pull Request was created by @yslai for CMSSW_7_5_X.

Switch Voronoi subtraction from equalization based recovery of negative fluctuation to ghost-particle based clustering and energy resumming

It involves the following packages:

RecoHI/HiJetAlgos

@cmsbuild, @cvuosalo, @nclopezo, @slava77 can you please review it and eventually sign? Thanks.
@jazzitup, @dgulhan, @appeltel, @yenjie, @kurtejung, @mandrenguyen, @richard-cms, @yetkinyilmaz 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.
If you are a L2 or a release manager you can ask for tests by saying 'please test' in the first line of a comment.
@nclopezo you are the release manager for this.
You can merge this pull request by typing 'merge' in the first line of your comment.

[slava77]

@cmsbuild please test

[cmsbuild]

The tests are being triggered in jenkins.

[cmsbuild]

+1
Tested at: f5bf763
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-9039/4631/summary.html

[slava77]

would a check with L2DoubleMu3Skim be more meaningful?
E.g. file
/store/results/hin/StoreResults/HIDiMuon/USER/L2DoubleMu3Skim_v10_38dff9fa051006d6e895e3c1df676d76-v1/40000/001FA005-07D0-E411-A6C2-0026B95A4D42.root

On 5/19/15 5:19 AM, yslai wrote:

@cvuosalo https://github.com/cvuosalo Since the RelVal sample has
virtually no hard processes, this behavior is expected. This is the pT
of the "eq" = equalization vs. "neg" = negative recovery using the
140.53, and you can see mostly low pT UE fluctuation with weak jet
contribution.

relval-pt
https://cloud.githubusercontent.com/assets/1032767/7700738/118ac674-fe21-11e4-827a-1f2c3f73b0c4.png

I am trying to identify CERN user-accessible triggered RAW to perform a
better check.

—
Reply to this email directly or view it on GitHub
#9039 (comment).

---

Vyacheslav (Slava) Krutelyov
TAMU: Physics Dept Texas A&M MS4242, College Station, TX 77843-4242
CERN: 42-R-027
AIM/Skype: siava16 googleTalk: slava77@gmail.com
(630) 291-5128 Cell (US) +41 76 275 7116 Cell (CERN)

---

[mandrenguyen]

better would be, e.g.,
/store/hidata/HIRun2011/HIHighPt/RAW/v1/000/181/688/02ACCF7A-380F-E111-B93E-BCAEC518FF54.root

[yslai]

This plot is made with /store/hidata/HIRun2011/HIHighPt/RAW/v1/000/181/777/069253FF-EA0F-E111-8F9A-003048F024FE.root and a tightened Delta R < 0.2 matching to reduce the jets from combinatorical UE:

The subtraction or accounting for the negative particles appears to be working properly, but the ability of the UE to produce jets is noticeably different.

[cvuosalo]

@ yslai: Is the conclusion now that this PR is performing correctly and that all the changes observed are acceptable and appropriate?

[yslai]

@cvuosalo Yes, the performance is as intended, and acceptable.

[cvuosalo]

+1

For #9039 f5bf763

For Heavy Ions, changing Voronoi subtraction related to negative fluctuations.

The code changes are satisfactory, as are Jenkins results. Additional tests showed significant differences, as seen above, but, these differences are the intended results of this PR.

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_7_5_X IBs unless changes (tests are also fine). This pull request requires discussion in the ORP meeting before it's merged. @davidlange6, @nclopezo, @smuzaffar

[davidlange6]

+1