Page History

...

1. cd to the work area, and source ./setup.sh
2. Download the AthenaUser template package, e.g.,
   wget http://mgwilson.web.cern.ch/mgwilson/Software/AthenaUser-00-00-00.tgz
3. tar -xzvf AthenaUser-00-00-00.tgz
4. Edit the top of AthenaUser/scripts/renamePackage.sh with a new package name and namespace for the code in
   your package; this will be denoted <package> below.
5. AthenaUser/scripts/renamePackage.sh
6. cd PersonalAthenaPackage/cmt ; echo "PersonalAthenaPackage-00-00-00" > version.cmt; cmt config; gmake; cd ../../

Package overview

This package contains two Algorithms, one Tool, and three classes providing some utilities. The two Algorithms are for filtering events (FilterEvent) and for processing events that have passed the filter (ProcessEvent). The runtime configuration for these algorithms is in the file share/configAOD_jobOptions.py. The Tool and utilities are self explanatory.

The main application is scripts/processAOD_trf.py, which is configured in the file share/runAOD_jobOptions.py. Running processAOD_trf.py without arguments will give brief information about which arguments are expected.

How to run on a local file

If you are running on lxplus, first cd to a directory with a lot of space

• cd /tmp/<username>

Download an example AOD file from the Grid, and if running on simulated data, also download a conditions database file:

• dq2-get -f AOD.129495._000001.pool.root.1 mc09_7TeV.105861.TTbar_PowHeg_Pythia.merge.AOD.e521_s765_s767_r1250_r1260/
• dq2-get -f DBRelease-10.3.1.tar.gz ddo.000001.Atlas.Ideal.DBRelease.v100301

Then, to process all events in a simulated-data file, run a command like this:

• processAOD_trf.py AOD.129495._000001.pool.root.1 True ntuple.root 0 DBRelease-10.3.1.tar.gz -1

To process all events in a recorded-data file, download a file:

• dq2-get -f data10_7TeV.00159086.physics_MuonswBeam.merge.AOD.f275_m548._lb0002-lb0021.1 data10_7TeV.00159086.physics_MuonswBeam.merge.AOD.f275_m548

and run a command like this

• processAOD_trf.py data10_7TeV.00159086.physics_MuonswBeam.merge.AOD.f275_m548._lb0002-lb0021.1 False ntuple.root 0 NONE -1

How to submit jobs to the Grid

In the scripts directory, there are two files for submitting Grid jobs using pathena: submitAODRec.sh and submitAODSim.sh. The general configuration is at the top; the commands to run on specific datasets can be added at the bottom. The N_FILES_TO_MERGE setting at the top refers to the number of files that can be merged into a single file later on; see below. This merging can be useful because running jobs on the Grid may produce many output files; however, merging them all into a single file at the end may not be possible if the resulting file would be too large.

How to download and merge output files from the Grid

If you submitted the Grid jobs using the scripts above, then the file download_recorded.sh or download_simulated.sh should have been produced. Edit the scripts/downloadDS.sh to configure the download parameters, then run the download_recorded.sh or download_simulated.sh script to download and merge the output files.

General C++ questions and answers

...