Dear all,<br><br>I am suffering the problem of network caused by transferring data for parallel runnings on PC clusters. It seems the pw.x occupies a lot of network from nodes to the server. If I run several pw.x jobs, the network of cluster got stuck. I noted that this is due to the pw.x write data (such as wave functions) to my HOME directory, the server, but not the temporary directory. My job file like<br>
<br>#$ -N SCF<br>#$ -q quad0<br>#$ -pe orte 16<br>#$ -j n<br><br>INDIR=`pwd`<br>echo $TMP > $INDIR/tmpdir<br>cd $TMP<br>cp $INDIR/<a href="http://JOB.scf.in">JOB.scf.in</a> .<br>mpirun -np 16 $PATH_TO_PW/pw.x -npool 4 < <a href="http://JOB.scf.in">JOB.scf.in</a> > scf.out<br>
cd $INDIR<br>cp $TMP/* . <br clear="all"><br>Then I copied the pw.x to the temporary directory, and run it locally. However, the pw.x still write files to my home directory. In this step my job file is :<br><br>#$ -N SCF<br>
#$ -q quad0<br>
#$ -pe orte 16<br>
#$ -j n<br>
<br>
INDIR=`pwd`<br>
echo $TMP > $INDIR/tmpdir<br>
cd $TMP<br>
cp $INDIR/<a href="http://JOB.scf.in">JOB.scf.in</a> .<br><span style="color: rgb(255, 0, 0);">cp $INDIR/pw.x .</span><br>
mpirun -np 16 pw.x -npool 4 < <a href="http://JOB.scf.in">JOB.scf.in</a> > scf.out<br>
cd $INDIR<br>
cp $TMP/* . <br><br>How can I make the pw.x just writes data to the $TMP and after its running copies the resulting files to my home directory, i.e. the server?<br>Any suggestion would be appreciated. Thanks a lot.<br><br>
Mithra Chan<br> <br>-- <br>Dr. Mithra Chan,<br>Department of Physics, National University of Singapore, Singapore<br><a href="mailto:chan.mithra@gmail.com">chan.mithra@gmail.com</a><br><br>