Grid

Running on the grid is as easy as having a grid certificate and using two command line arguments.

When running locally, we define an input dataset list file and an output directory; for the grid we'll define an input dataset name and an output dataset name (we still define a local output as well).

If your input dataset is:

group.det-indet.361107.PowhegPythia8EvtGen_AZNLOCTEQ6L1_Zmumu.simul.TRTxAOD.e3601_s2876_r7886_trt099-02_EXT0

Then we can process that sample on the grid using the --gridDS argument and --outDS argument. Your --outDS should always start with user.<your user name>

$ runAnaProjectLooperAlg -c default.cfg --gridDS group.det-indet.361107.PowhegPythia8EvtGen_AZNLOCTEQ6L1_Zmumu.simul.TRTxAOD.e3601_s2876_r7886_trt099-02_EXT0 --outDS user.<your user name>.xyz -o grid_run

You should make your output dataset name descriptive!

Of course, in order to be able to submit to the grid you need a grid certificate and you must setup panda and run voms-proxy-init:

$ lsetup panda
$ voms-proxy-init -voms atlas --valid 96:00