# mount lustre
mnt=/tmp/lustre
mkdir $mnt
sshfs lxpool-lustre:/lustre/panda/aakram/ $mnt
# unmount lustre
fusermount -u $mnt
# connect to virgo
ssh lxpool-virgo-debian10
# one job (single node)
./jobsim_complete.sh 100 llbar llbar_fwp.dec
# for job array
./submit.jobs
To run simulation one needs to run either runall.sh
(local) or runsif.sh
(singularity) scripts e.g.
# run on local machine
./runall 100 box DBGEN
Similary if one have singularity container for PandaRoot, then just run runsif.sh
.
# run on container
./runsif 100 box DBGEN
All bash scripts should run following simulation chain:
root -l -b -q sim_complete.C\($nevt,\"$outprefix\",\"$gen\",$pBeam,$seed\) > $outprefix"_sim.log" 2>&1
root -l -b -q digi_complete.C\($nevt,\"$outprefix\"\) > $outprefix"_digi.log" 2>&1
root -l -b -q skew_complete.C\($nevt,\"$outprefix\"\) > $outprefix"_skew.log" 2>&1
root -l -b -q recoideal_complete.C\($nevt,\"$outprefix\"\) > $outprefix"_reco.log" 2>&1
root -l -b -q data_complete.C\($nevt,\"$outprefix\",\"$_target\",\"$flag\"\) > $outprefix"_data.log" 2>&1