Dear Prof. Marks,<div><br></div><div> Thank you again !!</div><div> I will try it.</div><div> All the best,</div><div> Luis Ogando</div><div><br></div><div><br><br><div class="gmail_quote">2013/2/20 Laurence Marks <span dir="ltr"><<a href="mailto:L-marks@northwestern.edu" target="_blank">L-marks@northwestern.edu</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hmmm.<br>
<br>
This is my parallel_options on a machine with openmpi:<br>
<br>
setenv USE_REMOTE 1<br>
setenv MPI_REMOTE 0<br>
setenv WIEN_GRANULARITY 1<br>
setenv WIEN_MPIRUN "mpirun -x LD_LIBRARY_PATH -x PATH -np _NP_<br>
-machinefile _HOSTS_ _EXEC_"<br>
set a=`grep -e "1:" .machines | grep -v lapw0 | head -1 | cut -f 3 -d:<br>
| cut -c 1-2`<br>
#echo $a<br>
setenv MKL_NUM_THREADS $a<br>
setenv OMP_NUM_THREADS $a<br>
setenv MKL_DYNAMIC FALSE<br>
#echo MKL $MKL_NUM_THREADS $OMP_NUM_THREADS $MKL_DYNAMIC<br>
if (-e local_options ) source local_options<br>
set remote = "/bin/csh $WIENROOT/pbsh"<br>
<br>
The command pbsh should be in SRC_mpiutil at<br>
<a href="http://www.wien2k.at/reg_user/unsupported/" target="_blank">http://www.wien2k.at/reg_user/unsupported/</a> which has a few other<br>
things, let me know if I forgot to include it. It is<br>
<br>
mpirun -x LD_LIBRARY_PATH -x PATH -np 1 --host $1 /bin/csh -c " $2 "<br>
<br>
What this does is use mpirun locally for everything rather than<br>
communicating via ssh and spawning processes which can sometimes be a<br>
problem with openmpi.<br>
<div class="im HOEnZb"><br>
<br>
On Wed, Feb 20, 2013 at 10:45 AM, Luis Ogando <<a href="mailto:lcodacal@gmail.com">lcodacal@gmail.com</a>> wrote:<br>
> Dear Prof. Marks,<br>
><br>
> Thank you very much for your prompt answer.<br>
> I am using openmpi, but I believe that I am facing some of the tricky<br>
> issues you mentioned. I work in a SMP machine and the calculation starts<br>
> fine. After some tens of iterations, MPI suddenly asks for a password and<br>
> everything goes down to the drain.<br>
> I am using open mpi 1.6. Do you recommend any older version ??<br>
</div><div class="im HOEnZb">> All the best,<br>
> Luis Ogando<br>
><br>
><br>
><br>
><br>
</div><div class="HOEnZb"><div class="h5">> 2013/2/20 Laurence Marks <<a href="mailto:L-marks@northwestern.edu">L-marks@northwestern.edu</a>><br>
>><br>
>> One that works.<br>
>><br>
>> Some versions of openmpi have problems although that is probably the<br>
>> best option for the future. There are some tricky issues with openmpi<br>
>> related to how your flavor of ssh works, there is no standard and some<br>
>> do not propogate kill commands which means that they can leave<br>
>> orphans.<br>
>><br>
>> An alternative is mvapich. In a benchmark that I did a few months ago<br>
>> the Intel mpi was much better for AVX instructions, but that may have<br>
>> changed.<br>
>><br>
>> Openmpi is easy to compile; mvapich can be a little trickier.<br>
>><br>
>> N.B., if you have fast connections, e.g. infiniband, they are more<br>
>> than fast enough and I have never seen this as rate limiting with<br>
>> Wien2k. With ethernet it matters.<br>
>><br>
>> On Wed, Feb 20, 2013 at 10:29 AM, Luis Ogando <<a href="mailto:lcodacal@gmail.com">lcodacal@gmail.com</a>> wrote:<br>
>> > Dear Wien2k community,<br>
>> ><br>
>> > Is there any recommended flavor and version of an MPI compiler to use<br>
>> > with " Intel(R) Fortran Intel(R) 64 Compiler XE for applications running<br>
>> > on<br>
>> > Intel(R) 64, Version 12.0.3.174 Build 20110309 " ?<br>
>> > All the best,<br>
>> > Luis Ogando<br>
>><br>
>><br>
>><br>
>> --<br>
>> Professor Laurence Marks<br>
>> Department of Materials Science and Engineering<br>
>> Northwestern University<br>
>> <a href="http://www.numis.northwestern.edu" target="_blank">www.numis.northwestern.edu</a> <a href="tel:1-847-491-3996" value="+18474913996">1-847-491-3996</a><br>
>> "Research is to see what everybody else has seen, and to think what<br>
>> nobody else has thought"<br>
>> Albert Szent-Gyorgi<br>
>> _______________________________________________<br>
>> Wien mailing list<br>
>> <a href="mailto:Wien@zeus.theochem.tuwien.ac.at">Wien@zeus.theochem.tuwien.ac.at</a><br>
>> <a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien</a><br>
><br>
><br>
<br>
<br>
<br>
--<br>
Professor Laurence Marks<br>
Department of Materials Science and Engineering<br>
Northwestern University<br>
<a href="http://www.numis.northwestern.edu" target="_blank">www.numis.northwestern.edu</a> <a href="tel:1-847-491-3996" value="+18474913996">1-847-491-3996</a><br>
"Research is to see what everybody else has seen, and to think what<br>
nobody else has thought"<br>
Albert Szent-Gyorgi<br>
_______________________________________________<br>
Wien mailing list<br>
<a href="mailto:Wien@zeus.theochem.tuwien.ac.at">Wien@zeus.theochem.tuwien.ac.at</a><br>
<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien</a><br>
</div></div></blockquote></div><br></div>