[torqueusers] specifying nodes and ppn: what is expected?

Sean Davis sdavis2 at mail.nih.gov
Fri Nov 30 10:40:02 MST 2007


I have a simple cluster (two nodes for now) set up and running using
pbs_sched.  I submit the following using qsub as a script:

#!/bin/bash
# --- send the job to 4 nodes, with 2 processors per node
# PBS -l nodes=2:ppn=4
# --- send the output to the test.out file
#     the default is .o<jobid>
#PBS -o test.out
# --- send the error output to the test.err file
#     the default is .e<jobid>
#PBS -e test.err

# --- print out the list of nodes this job is running upon
/bin/cat $PBS_NODEFILE

echo "Print out the hostname and date"
/bin/hostname
/bin/date
exit 0

The output that I get for $PBS_NODEFILE is a single line representing a
single node.  However, if I replace the node specification with:

-l nodes=node1:ppn=4+node2:ppn=4

I get eight lines in PBS_NODEFILE, four for each node.  I would have
expected the output of both nodes specification to be the same.  The
documentation and examples I have seen suggest this, also.  Have I missed
some configuration flag or the like to get -l nodes=2:ppn=4 to get me four
processors on each of two nodes?

Thanks,
Sean
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20071130/5593c510/attachment.html


More information about the torqueusers mailing list