THIS INFO IS OUTDATED

InDetBtag

Short description

InDetBTag is a prototype b-tagger which runs within Athena. It is located in: offline/InnerDetector/InDetRecAlgs/InDetBTag You can get it via cvs typing the following:

cvs co <nop>InnerDetector/InDetRecAlgs/InDetBTag

Running the b-tagger

There are two possible ways to run the b-tagger:

• on a lxplus machine (to run on few events, to test things)
• on a batch machine (to run overnight on lots of events)

Run on a lxplus machine

Since the b-tagger only needs inner detector information it can be run with the InDetExample package which provides a "run skeleton" to run InnerDetector software.

Check out this packages with typing:

cvs co <nop>InnerDetector/InDetExample/InDetRecExample

and follow the lines in the README file, except for the running part (do not type athena blabla as suggested).

To have the full information available from the b-tagger, the CBNT algorithms have to run. The basic InDetExample jobOptions file does not support this. However, you can use the jobOptions files provided here:

• jobOptions_DC1.py: to run on DC1 data (runs on data ZEBRA.P points to)
• jobOptions_G4.py: to run on G4 simulated data (runs on data specified in the jobOptions_G4 file itself)
• InDetRec_jobOptions.py: includes InnerDetector related stuff (like the b-tagger)

Notice that InDetExample provides a file InDetRec_jobOptions.py but it DOES NOT include the b-tagging software. So please use the one given here.

To run, copy the jobOptions_X.py (X = DC1|G4 depending on what you want to run) into the run directory of InDetExample. Copy InDetRec_jobOptions.py into the share directory (you might want to make a backup of the original one).

And finally, in the run directory type:

athena.py jobOptions_DC1.py (or G4) and it should run.

Do not forget to copy the PoolFileCatalog.xml file for the pool file you are using. If you for example want to run on preproduction 803 HW->munubb you will have to copy

/afs/cern.ch/atlas/project/dc2/preprod803/PoolFileCatalog.xml

into your run directory.

Run on a batch machine

To run on a batch machine do the following:

• set your machine up so that it can run on a lxplus machine (see above)
• somewhere (not in the InDetExample directory) create a new directory, say batch
• copy the following files into that directory (depending on what you want to run)

DC1	G4
jobstarter_DC1.sh	jobstarter_G4.sh
athena_batch_DC1.sh	athena_batch_G4.sh
InDetExample_jobOptions_batch_DC1.py	InDetExample_jobOptions_batch_G4.py

• open the files jobstarter_X.py and athena_batch_X.sh and edit them to your needs
  • don't worry, only few things need to be changed, e.g. you don't want to run in my home directory
  • you don't want your results written to my castor space, aso.
  • just look through and follow the instruction (I hope there are some ;-))
• the last thing you need to do is specify on how many events you want to run, on which samples, aso. Look into jobstarter_X.py and adjust the lines to your needs
• finally type (on a lxplus machine): ./jobstarter_X.sh and press enter to start the batch job

-- DerSchrecklicheSven - 28 May 2004