Difference: METRelValNotesForME (1 vs. 7)

Revision 72012-04-05 - SamanthaHewamanage

Line: 1 to 1
 
META TOPICPARENT name="SamanthaHewamanage"

MET Release Validation Notes For ME

  • Samples to check: QCDFlatPt_15_3000, QCD_Pt_80_120, TTbar, LMsfts, TTbar/QCD FastSim and FastVsFull for the same release.
Line: 17 to 17
 

VIPs to Contact in Desperation

Added:
>
>
  • Questions about 6XX campaign (tag issues etc), Shahzad Malik MUZAFFAR. (April 5th, 2012)
  -- SamanthaHewamanage - 14-Oct-2011

Revision 62012-01-14 - SamanthaHewamanage

Line: 1 to 1
 
META TOPICPARENT name="SamanthaHewamanage"

MET Release Validation Notes For ME

  • Samples to check: QCDFlatPt_15_3000, QCD_Pt_80_120, TTbar, LMsfts, TTbar/QCD FastSim and FastVsFull for the same release.
  • KS Test limit = 0.000001
  • Every new RelVal sample is generated from scratch. Meaning, new events are generated and ran through GEANT.
  • It seems the using the absolute release is more critical than the global tag used to process the given RelVal sample. I had trouble running a 42X DY sample which had the tag START44_V5. But it ran fine with START42_V5 tag. And Artur said, it should be fine. Need to double check with him.
Added:
>
>
    • There can be issues by not using the exact CMSSW version. The tag may not exists for other than a particular CMSSW version and the jobs would crash.!!! So better use the same CMSSW version and the Global Tag as the dataset was generated. (Jan, 2012)
 
  • Who does the data validation for a release?

Different RelVal Requests

Revision 52011-12-16 - SamanthaHewamanage

Line: 1 to 1
 
META TOPICPARENT name="SamanthaHewamanage"

MET Release Validation Notes For ME

  • Samples to check: QCDFlatPt_15_3000, QCD_Pt_80_120, TTbar, LMsfts, TTbar/QCD FastSim and FastVsFull for the same release.
Line: 13 to 13
 
  1. special HCAL ZS requests which need redigi-rereco to perform the validations. It may be necessary to do the redigi-rereco because the changes made are at reco level and some sometimes at digitization level, so you need re-reco or re-digi+re-reco for them to be checked by MET monitorables.
  2. Some validation requests comes via Physics Validation. Need to pay attention to those and do the requests ASAP. Report this first to JetMET conveners. Coordinate with them on reporting this to Physics Validation group.
Added:
>
>

VIPs to Contact in Desperation

 -- SamanthaHewamanage - 14-Oct-2011

Revision 42011-11-13 - SamanthaHewamanage

Line: 1 to 1
 
META TOPICPARENT name="SamanthaHewamanage"

MET Release Validation Notes For ME

Added:
>
>
  • Samples to check: QCDFlatPt_15_3000, QCD_Pt_80_120, TTbar, LMsfts, TTbar/QCD FastSim and FastVsFull for the same release.
 
  • KS Test limit = 0.000001
Changed:
<
<
  • Every new RelVal sample is generated from scratch. Meaning, new events are generated and ran through GEANT.
>
>
  • Every new RelVal sample is generated from scratch. Meaning, new events are generated and ran through GEANT.
 
  • It seems the using the absolute release is more critical than the global tag used to process the given RelVal sample. I had trouble running a 42X DY sample which had the tag START44_V5. But it ran fine with START42_V5 tag. And Artur said, it should be fine. Need to double check with him.
  • Who does the data validation for a release?

Revision 32011-10-18 - SamanthaHewamanage

Line: 1 to 1
 
META TOPICPARENT name="SamanthaHewamanage"

MET Release Validation Notes For ME

  • KS Test limit = 0.000001
  • Every new RelVal sample is generated from scratch. Meaning, new events are generated and ran through GEANT.
  • It seems the using the absolute release is more critical than the global tag used to process the given RelVal sample. I had trouble running a 42X DY sample which had the tag START44_V5. But it ran fine with START42_V5 tag. And Artur said, it should be fine. Need to double check with him.
Added:
>
>
  • Who does the data validation for a release?
 

Different RelVal Requests

  1. regular validations (uses DQM GUI and standalone to confirm)

Revision 22011-10-18 - SamanthaHewamanage

Line: 1 to 1
 
META TOPICPARENT name="SamanthaHewamanage"

MET Release Validation Notes For ME

Changed:
<
<
  • Every new RelVal sample is generated from scratch. Meaning, new events are generated and ran through GEANT.
>
>
  • KS Test limit = 0.000001
  • Every new RelVal sample is generated from scratch. Meaning, new events are generated and ran through GEANT.
  • It seems the using the absolute release is more critical than the global tag used to process the given RelVal sample. I had trouble running a 42X DY sample which had the tag START44_V5. But it ran fine with START42_V5 tag. And Artur said, it should be fine. Need to double check with him.
 
Added:
>
>

Different RelVal Requests

  1. regular validations (uses DQM GUI and standalone to confirm)
  2. special requests which uses standalone
  3. special HCAL ZS requests which need redigi-rereco to perform the validations. It may be necessary to do the redigi-rereco because the changes made are at reco level and some sometimes at digitization level, so you need re-reco or re-digi+re-reco for them to be checked by MET monitorables.
  4. Some validation requests comes via Physics Validation. Need to pay attention to those and do the requests ASAP. Report this first to JetMET conveners. Coordinate with them on reporting this to Physics Validation group.
  -- SamanthaHewamanage - 14-Oct-2011

Revision 12011-10-14 - SamanthaHewamanage

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="SamanthaHewamanage"

MET Release Validation Notes For ME

  • Every new RelVal sample is generated from scratch. Meaning, new events are generated and ran through GEANT.

-- SamanthaHewamanage - 14-Oct-2011

 
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