Particle Flow Validation: ElectronBenchmarkGeneric

Overview

This page presents the results of the standard particle flow validation, performed with the ElectronBenchmarkGeneric package

General recipe for event production, merging, analysing and plotting

The process has been automatized as much as possible; all together, starting and running the validation should take less than 15min. It takes, of course, much longer to run. I recommend to do everything within a "screen" session. It is done in two steps: first install the release and launch the validation jobs. It will produce the validation plots and publish them. Then, run the comparisons.

Area installation

Just do
setenv SCRAM_ARCH slc5_amd64_gcc434
cmsrel CMSSW_X_Y_Z
cd CMSSW_X_Y_Z/src
cmsenv
addpkg Validation/RecoParticleFlow
cd Validation/RecoParticleFlow/Benchmarks/ElectronBenchmarkGeneric

Job configuration

All the changes to do are in the OvalFile.
  • on line 12, edit the CMSSW release
  • on line 15, edit the expression as to match the name of the CMS.RelVal samples. It is generally present (or there is a link) in the mail announcing the availability of the samples
  • on 109, make sure that the DBS_COMPARE_RELEASE is commented, i.e there is a ! just before "var"

Launch the jobs

oval run %Pt35 %ZEE %TTbarb %TTbarW %QCD80-120

Produce the comparisons plots

In general (especially when working from home), it is better to run root in batch mode. To do so, edit ../Tools/indexGenCompare.py and replace root -l with root -b This step should be done carefully, as it is quite painful to clean the webpage from the invalid links if something goes wrong.
  • Edit the OvalFile, and line 109, uncomment the DBS_COMPARE_RELEASE and put the value of a release for which the validation has been run (one of those below).
  • run with oval run comparison.sh%. It takes only a few seconds

Put the links in the twiki

Just type
oval run prepare-twiki.sh
and then copy and paste the text in the twiki.

Validation results

CMSSW_4_4_X

CMSSW_4_2_X

CMSSW_3_11_X

CMSSW_3_10_X

CMSSW_3_9_0_pre7

3_8_X

3_7_X

3_6_0_X

Note: for some time, a 15% loss of efficiency in the endcaps was observed. It was actually a problem in the validation: a mixture of Gflash/standard samples had been used (it probably means that there is a problem with GFlash, but it is an other story)

3_5_0_X

In CMS_3_5_3, a problem is observed with the single particle samples and is understood.

3_4_0_preX

3_3_X

-- FlorianBeaudette - 18-Nov-2009

Responsible: FlorianBeaudette

Topic attachments
I Attachment History Action Size Date Who Comment
PNGpng TrackEta.png r1 manage 6.9 K 2010-03-31 - 11:50 FlorianBeaudette track eta 360pre4 vs 355
PNGpng track-multi-355.png r1 manage 9.9 K 2010-03-28 - 23:10 FlorianBeaudette track multiplicity as a function eta in 355
PNGpng track-multi-360pre4.png r1 manage 9.8 K 2010-03-28 - 23:12 FlorianBeaudette track multiplicity in 360pre4
Edit | Attach | Watch | Print version | History: r36 < r35 < r34 < r33 < r32 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r36 - 2011-09-19 - FlorianBeaudette



 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    CMSPublic All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback