P115 Reprocessing
status: Complete
last update: 15 July 2010
This page is a record of the configuration for the P115 reprocessing project, motivated by the Pass 7.2 event classification. This project involves reprocessing with Pass7 classification trees and (ultimately) new IRFs. The name "P115" derives from the word "processing" and the initial file version to be used for the output data products, e.g., r0123456789_v115_merit.root.
- P115-LEO-MERIT - Reclassify data from P106-LEO (full reprocessing of earth limb data from L&EO period) rather than Level 1, producing MERIT, FILTEREDMERIT and ELECTRONMERIT
- P115-LEO-FT1 - From P115-LEO-MERIT, generate FT1, LS1, LS3, ELECTRONFT1 products, including the calculation of the diffuse response.
Datafile names, versions and locations
Data file version numbers for this reprocessing will begin with v115.
XROOT location and file naming
Location template:
/glast/Data/Flight/Reprocess/<reprocessName>/<dataType>
Locations for P110:
/glast/Data/Flight/Reprocess/P115-LEO/merit /glast/Data/Flight/Reprocess/P115-LEO/filteredmerit /glast/Data/Flight/Reprocess/P115-LEO/electronmerit /glast/Data/Flight/Reprocess/P115-LEO/ft1 /glast/Data/Flight/Reprocess/P115-LEO/electronft1 /glast/Data/Flight/Reprocess/P115-LEO/ls1 /glast/Data/Flight/Reprocess/P115-LEO/ls3
File naming:
Data Type |
Send to FSSC |
Naming template |
|---|---|---|
MERIT |
No |
r<run#>_<version>_<dataType>.root |
FILTEREDMERIT |
No |
r<run#>_<version>_<dataType>.root |
ELECTRONMERIT |
No |
r<run#>_<version>_<dataType>.root |
ELECTRONFT1 |
No |
r<run#>_<version>_<dataType>.fit |
FT1 |
No |
gll_ph_r<run#>_<version>.fit |
LS1 |
No |
gll_ev_r<run#>_<version>.fit |
LS3 |
No |
gll_lt_r<run#>_<version>.fit |
DataCatalog location and naming
Logical directory and group template:
Data/Flight/Reprocess/<reprocessName>:<dataType>
Note that the <dataType> field (following the colon) is a DataCatalog 'group' name. All file names are of the form r<run#> as indicated by the examples below.
Logical directories for P115:
Data/Flight/Reprocess/P115-LEO:MERIT r0239557414 Data/Flight/Reprocess/P115-LEO:FILTEREDMERIT r0239557414 Data/Flight/Reprocess/P115-LEO:ELECTRONMERIT r0239557414 Data/Flight/Reprocess/P115-LEO:FT1 r0239557414 Data/Flight/Reprocess/P115-LEO:ELECTRONFT1 r0239557414 Data/Flight/Reprocess/P115-LEO:LS1 r0239557414 Data/Flight/Reprocess/P115-LEO:LS3 r0239557414
Data Sample
The currently defined data sample for P110 and P110-LEO reprocessing includes:
P115-LEO |
(MET) |
(UTC) |
|---|---|---|
First run |
237783740 |
2008-07-15 03:02:20 |
Last run |
244406327 |
2008-09-29 18:38:47 |
Total runs |
200 |
Total MERIT events |
488,288,751 |
Total photon events |
139,475,840 |
Total electron events |
736,950 |
Note that the L&EO data represent a discontiguous set of runs.
Bookkeeping
- (This page): Define ingredients of reprocessing (processing code/configuration changes)
- Processing History database: http://glast-ground.slac.stanford.edu/HistoryProcessing/HProcessingRuns.jsp?processingname=P115-LEO
- List of all reprocessings
- List of all data runs reprocessed
- Pointers to all input data files (-> dataCatalog)
- Pointers to associated task processes (-> Pipeline II status)
- Data Catalog database: http://glast-ground.slac.stanford.edu/DataCatalog/folder.jsp
- Lists of and pointers to all output data files
- Meta data associated with each output data product
P115-LEO-MERIT
Status chronology
- 8 Jul 2010 - Construct task
- 9 Jul 2010 - Task complete
Configuration
Task Location |
/nfs/farm/g/glast/u38/Reprocess-tasks/P115-LEO-MERIT |
Task Status |
http://glast-ground.slac.stanford.edu/Pipeline-II/index.jsp |
GlastRelease |
v17r35p1gr02 |
Skimmer |
07-07-00 |
Input Data Selection |
Set of 200 runs selected by Anders Borgland |
Input Run List |
ftp://ftp-glast.slac.stanford.edu/glast.u38/Reprocess-tasks/P115-LEO-MERIT/config/runFile.txt |
photonFilter |
evtClassDefs v0r6p1 CTBParticleType==0 && CTBClassLevel>0 |
electronFilter |
CTBParticleType==1 |
jobOpts |
ftp://ftp-glast.slac.stanford.edu/glast.u38/Reprocess-tasks/P115-LEO-MERIT/config/reClassify.txt |
Output Data Products |
Timing and Scaling
The processClump job step averaged 20-30 hequ-min or 70-80 fell-min.
The mergeClumps job step averaged 5-15 hequ-min or 8-20 fell-min.
It was observed that xroot became very stressed when the number of processClump jobs reached ~1000. The stress increased when mergeClumps jobs became part of the mix. Pipeline throttle needed.
P115-LEO-FT1
Status chronology
- 9 Jul 2010 - Assemble task
- 12 Jul 2010 - After a 20-run test, begin production processing
- 15 Jul 2010 - Task complete
Configuration
Task Location |
/nfs/farm/g/glast/u38/Reprocess-tasks/P115-LEO-FT1 |
Task Status |
http://glast-ground.slac.stanford.edu/Pipeline-II/index.jsp |
Input Data Selection |
MERIT (from P106-LEO-MERIT), FT2 (from P110-FT2) |
Input Run List |
ftp://ftp-glast.slac.stanford.edu/glast.u38/Reprocess-tasks/P115-LEO-FT1/config/runFile.txt |
evtClassDefs |
00-16-00 |
meritFilter |
pass7_FSW_cuts, |
eventClassifier |
Pass7_Classifier.py |
ScienceTools |
09-17-00 (SCons build) |
Code Variant |
redhat4-i686-32bit-gcc34 or redhat5-i686-32bit-gcc41 |
Diffuse Model |
/afs/slac.stanford.edu/g/glast/ground/releases/analysisFiles/diffuse/v2/source_model_v02.xml [Ref] |
Diffuse Response IRFs |
P7_v2_diff, P7_v2_extrad, P7_v2_datac |
IRFs |
implemented as 'custom irf', files in /afs/slac.stanford.edu/g/glast/ground/PipelineConfig/IRFS/Pass7.2 |
Output Data Products |
Processing chain for FITS data products
Data Product |
makeFT1 |
gtdiffrsp |
gtmktime |
gtltcube |
|---|---|---|---|---|
FT1 |
true |
true for |
true |
false |
LS1 |
true |
false |
true |
false |
LS3 |
false |
false |
false |
true |
ELECTRONFT1 |
true |
false |
true |
false |
Note on diffuse response calculation: gtdiffrsp is called three times in succession. The first time with IRF P7_v2_diff and evclsmin==8, followed by IRF P7_v2_extrad and evclsmin==9, and finally IRF P7_v2_datac and evclsmin==10. The resulting FT1 file has six columns of diffuse response, two columns (galactic and extragalactic response) for each of the three IRFs. This creates a non-standard FT1 file by FSSC standards as they expect only five diffuse response columns.
Timing and Scaling
Timing is dominated by the gtdiffrsp steps in the mergeClumps job step. There is a wide range of processing times due to: different classes of batch machines; different numbers of events to process; batch jobs being temporarily put into system suspend (SSUSP) states [due to pre-emptive queues]; or, possibly, due to processing dependencies in the data. Elapsed processing time for a single run ranges from 4 to 12+ hours (assuming no shortage of batch machines). The average CPU time per clump is 360 min +/- 160 min.
Thirteen of 200 jobs required xxl batch queue and then took >30 hours to complete.