Difference between revisions of "Calibration Train"

From GlueXWiki
Jump to: navigation, search
(→‎Calibration Challenge)
Line 1: Line 1:
 
__TOC__
 
__TOC__
  
 +
=
 +
 +
= Job Structure =
 +
 +
Currently 3 "passes" through the data are performed. The first pass tries to do as many calibrations as possible with one file of data.  The second pass tries to do calibrations that need a larger data set.  The final pass runs through a full run to use the full statistics in a run and to generate outputs for other calibration procedures that can't be done automatically yet (e.g. pi0 calibrations).
 +
 +
The plugins that are run on each pass are:
 +
* '''Pass 1'''
 +
** Step 1 - RF_online
 +
** Step 2 - HLDetectorTiming (coarse timing+ADC/TDC alignment),TOF_TDC_shift
 +
** Step 3 - HLDetectorTiming (track-based timing),BCAL_TDC_Timing (timewalks)
 +
* '''Pass 2'''
 +
** Step 1 - PS_timing,TAGH_timewalk,BCAL_attenlength_gainratio,BCAL_TDC_Timing (full)
 +
** Step 2 - st_tw_corr_auto
 +
* '''Pass 3'''
 +
** HLDetectorTiming,PSC_TW,BCAL_gainmatrix,FCALgains,FCALpedestals,ST_Tresolution,ST_Propagation_Time,p2gamma_hists,imaging,pedestal_online,BCAL_LEDonline,PS_timing
 +
 +
= Output =
  
 
= Calibration Challenge =  
 
= Calibration Challenge =  
Line 6: Line 24:
 
* [[Calibration Challenge|Calibration Challenge 1]]
 
* [[Calibration Challenge|Calibration Challenge 1]]
  
−
= Organization =
+
= Old Planning =
 +
 
 +
== Organization ==
  
 
* The jobs will be submitted every Tuesday at noon, JLab time.
 
* The jobs will be submitted every Tuesday at noon, JLab time.
Line 13: Line 33:
 
* The output of the jobs will be stored in ...
 
* The output of the jobs will be stored in ...
  
−
= Run Ranges =  
+
== Run Ranges ==  
  
 
The following runs will be processed:
 
The following runs will be processed:
Line 20: Line 40:
 
* 2931, 3079, 3179, 3180, 3183, 3185
 
* 2931, 3079, 3179, 3180, 3183, 3185
  
−
= Calibrations =
+
== Calibrations ==
  
 
=== Job Requirements ===
 
=== Job Requirements ===
Line 46: Line 66:
 
* TOF calibrations - https://halldsvn.jlab.org/repos/trunk/home/zihlmann/TOF_calib/
 
* TOF calibrations - https://halldsvn.jlab.org/repos/trunk/home/zihlmann/TOF_calib/
  
−
= Code =
+
== Code ==
  
 
The current code can be obtained at
 
The current code can be obtained at
Line 55: Line 75:
 
The ROOT library directory can be found at:
 
The ROOT library directory can be found at:
 
...
 
...
−
 
−
= Older Information =
 
−
 
−
== Proposal ==
 
−
 
−
* Use case: Streamline tape library usage, provide common environment for production and development
 
−
** Example: Users develop plugins on their favorite files/run, use this for running over larger data
 
−
* Run every week (Wednesday to avoid conflict with monitoring?)
 
−
* Uses subset of runs
 
−
* Users provide:
 
−
** DANA plugin
 
−
** Optional post-processing scripts to be run after every/all runs
 
−
*** Curated in SVN
 
−
* Results stored in standard location(s)
 
−
** Possible results: ROOT files, images, calibration constants, web pages
 
−
* Uses SWIF? (buzzword compliance)
 
−
* Uses GlueX project account (gxproj3? 4? 5?)
 
−
 
−
== Runs to Use ==
 
−
 
−
Several possibilities:
 
−
 
−
# Large/Popular Runs
 
−
#* Fall 2014: 1514, 1769, 1777, 1787, 1803, 1807, 1810, 1825, 1847, 1852, 1854, 1871, 1872, 2138, 2206, 2207, 2209, 2223, 2397
 
−
#* Spring 2015: 2931, 3079, 3179, 3180, 3183, 3185
 
−
# Other big sets of runs:
 
−
#* Fall 2014: 1516-1520, 1769-1777, 1900, 2179-2184, 2205, 2228, 2407-2409, 2416-2420
 
−
#* Look for groups of runs with similar conditions
 
−
# Others
 
−
 
−
 
−
 
−
This also raises the question: what are the good runs?
 
−
 
−
One proposal:
 
−
* Tag each run in monitoring database with one of three values:
 
−
*# 0 - non-production quality
 
−
*# 1 - production quality
 
−
*# 2 - production quality, used for calibrations
 
−
* Idea is that > 1 means that the run is good to use for physics
 
−
** Finer grained information can be stored in RCDB
 
−
* To determine production quality, develop quality metrics for each subdetector, use combination of quality metrics and eye test
 

Revision as of 00:29, 9 March 2016

=

Job Structure

Currently 3 "passes" through the data are performed. The first pass tries to do as many calibrations as possible with one file of data. The second pass tries to do calibrations that need a larger data set. The final pass runs through a full run to use the full statistics in a run and to generate outputs for other calibration procedures that can't be done automatically yet (e.g. pi0 calibrations).

The plugins that are run on each pass are:

  • Pass 1
    • Step 1 - RF_online
    • Step 2 - HLDetectorTiming (coarse timing+ADC/TDC alignment),TOF_TDC_shift
    • Step 3 - HLDetectorTiming (track-based timing),BCAL_TDC_Timing (timewalks)
  • Pass 2
    • Step 1 - PS_timing,TAGH_timewalk,BCAL_attenlength_gainratio,BCAL_TDC_Timing (full)
    • Step 2 - st_tw_corr_auto
  • Pass 3
    • HLDetectorTiming,PSC_TW,BCAL_gainmatrix,FCALgains,FCALpedestals,ST_Tresolution,ST_Propagation_Time,p2gamma_hists,imaging,pedestal_online,BCAL_LEDonline,PS_timing

Output

Calibration Challenge

Old Planning

Organization

  • The jobs will be submitted every Tuesday at noon, JLab time.
  • The jobs will be run from the gxproj3 account [parallel use with EventStore jobs]
  • The output of the jobs will be stored in ...

Run Ranges

The following runs will be processed:

RunPeriod-2015-03

  • 2931, 3079, 3179, 3180, 3183, 3185

Calibrations

Job Requirements

Each calibration process should include the following:

What is Being Run

The following plugins are currently being run:

  • RF_online (RF signal)
  • BCAL_TDC_Timing
  • HLDetectorTiming
  • PSC_TW

Working on adding:

Code

The current code can be obtained at 

svn co https://halldsvn.jlab.org/repos/trunk/home/sdobbs/calibration_train/

The ROOT library directory can be found at: ...

Retrieved from "https://halldweb.jlab.org/wiki/index.php?title=Calibration_Train&oldid=73581"