P130 Reprocessing

status: In Development Complete
last update: 30 January 6 Aug 2012

Wiki MarkupThis page is a record of the configuration for the P130 reprocessing project, recalculating the diffuse response (see [here|https://confluence.slac.stanford.edu/download/attachments/112105781/DiffResp_Fegan_Nov2011.pdf?version=1&modificationDate=1320928938000] for a description of the change). This task is identical with P120-FT1 \, v2.1 [please see for details\], run in July 2011, except for the following details.

1. New version of ScienceTools (09-26-9202) which includes new integration scheme for gtdiffrsp
2. New run list (updated for data processed since the end of P120)
3. Name of Pipeline task (P120->P130) as well as in file naming

=============== WHAT FOLLOWS NEEDS TO BE UPDATED FOR P130 ====================

Data Sample

The currently defined data sample for P120 P130 reprocessing includes:

Summary from DataCatalog as of 4/17/2012 and updated 8/2/20116/2012 (after new evt selection filter).

=============== THE FOLLOWING HAS BEEN UPDATED FOR P130 ====================

Bookkeeping

1. (This page): Define ingredients of reprocessing (processing code/configuration changes)
2. Processing History database: http://glast-ground.slac.stanford.edu/HistoryProcessing/HProcessingRuns.jsp?processingname=P130^{Image Removed}
   1. List of all reprocessings
   2. List of all data runs reprocessed
   3. Pointers to all input data files (-> dataCatalog)
   4. Pointers to associated task processes (-> Pipeline II status)
3. Data Catalog database: http://glast-ground.slac.stanford.edu/DataCatalog/folder.jsp^{Image Removed}
   1. Lists of and pointers to all output data files
   2. Meta data associated with each output data product

...

This task generates all desired FITS data products. An example of the code processing chain appears on a child page.

Status chronology

• 1/26/2012 - Initial task configured, 1000 runs reprocessed. On hold for validation.
• 2/1/2012 - Okay to continue. Begin block 1 reprocessing, containing 19133 runs:

• 2/3/2012 - Block 1 reprocessing complete
• 4/2/2012 - Begin block 2, raising the total # runs to 20126, through run 354926369 2012-03-31 22:39:29 (UTC)
• 4/18/2012 - Final reprocessing block: additional 259 runs through 356429251, 2012-04-18 08:07:29 (UTC)
• 4/19/2012 - Complete.
• 5/1/2012 - Reconfigure start time - reducing by 1.9 seconds - for run 245403855 due to old leap second problem - to regain 27 events at the start of the run. This is stream 1019.
• 8/3/2012 - A request to update the event selection filter on data in the period 1 Aug 2011 and 19 Apr 2012 was made to bring EXTENDED FITS and FILTEREDMERIT files into synch. See this page for details. The P130-FT1 task has been modified to run a new event filter; mergeClumps job steps for the following runs rolled back:

  	Run	Task Stream
  Start	333886531	16460
  End	356429251	20384

  This action will recreate only the EXTENDEDFT1 and EXTENDEDLS1 data products. These will automatically be ingested by the astroserver. Note that the file original-config.py contains the former configuration for this task, while config.py contains the changes discussed above.
• 8/6/2012 - Re-reprocessing complete

Configuration

Task Location	/nfs/farm/g/glast/u38/Reprocess-tasks/P130-FT1
Task Status	http://glast-ground.slac.stanford.edu/Pipeline-II/task.jsp?task=74765664Image Removed
Input Data	MERIT ( from P120-MERIT (runs 239557414 - 333880535) and Level 1 (runs 333886531 - present)
spacecraft data	FT2 from P105 (runs 239557414 - 271844560), then from current Level 1 production
Input Run List	ftp://ftp-glast.slac.stanford.edu/glast.u38/Reprocess-tasks/P130-FT1/config/runFile.txtImage Removed
Reprocessing Mode	reFT1
meritFilter	FT1EventClass!=0 for P120-reprocessing, and (FT1EventClass& 0x00003EFF)!=0 for subsequent data
evtClassDefs	00-19-01
eventClassMap	EvtClassDefs_P7V6.xml
ScienceTools	09-26-02
Code Variants	redhat5-i686-32bit-gcc41 (Optimized)
Diffuse Model	based on contents of /afs/slac.stanford.edu/g/glast/ground/GLAST_EXT/diffuseModels/v2r0 (see https://confluence.slac.stanford.edu/display/SCIGRPS/Quick+Start+with+Pass+7Image Removed )
Diffuse Response	'source' using P7SOURCE_V6 IRF 'clean' using P7CLEAN_V6 IRF
IRFs	P6V7P7V6, contained within ScienceTools release
Output Data Products	FT1, LS1, EXTENDEDFT1, EXTENDEDLS1, ELECTRONFT1

...

Note on 'Code Variant': The SLAC batch farm contains a mixture of architectures , both hardware (Intel/AMD 64-bit) and software (RHEL5-64, gcc v4.1, etc.).

Timing and Performance

With a sample of the first 8128 runs, the CPU-intensive part of the task (mergeClumps) is consuming ~42 CPU-min/job (differences in machine class are small). TrickleStream configured to allow 1500 simultaneous mergeClumps jobs, and up to 150 job starts every 200 second cycle.