[GE users] Script problem on SGE.

bizon bizoniusz at o2.pl
Tue Mar 31 19:28:12 BST 2009


I changed the PE configuration for "fill_up" and now queue system allocated 2 nodes per 8 cores. But job still working on only one node generated 16 threads. This is the job script:

#!/bin/bash
#$ -cwd
#$ -N cpmd
#$ -j y
#$ -notify
#$ -v w
#$ -pe lam 16
echo "Got $NSLOTS processors."
echo "Machines:"
cat $TMPDIR/hostfile
export LAM_MPI_SOCKET_SUFFIX=$JOB_ID.$JOB_NAME
lamboot
/usr/bin/mpirun C -v -ssi rpi lamd -np $NSLOTS cpmd.x ala.inp >&ala.out
lamhalt
 
end hostfile file does not exist in /scratch directory.

So, what should I do??

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=150915

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list