<div dir="ltr">Dear Prof. Blaha,<div><br></div><div style> Many thanks for your comments !</div><div style> I have suspected that without the .machines file I would have problems to mix k-points and mpi parallelization.</div>
<div style> The reason to use mpiexec_mpt instead of mpirun I will have to ask to the guys from the computing center (I can not imagine a reasonable one).</div><div style> Anyway, I will also ask them about the -host and -hostfile possibilities.</div>
<div style> All the best,</div><div style> Luis</div><div style><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/4/8 Peter Blaha <span dir="ltr"><<a href="mailto:pblaha@theochem.tuwien.ac.at" target="_blank">pblaha@theochem.tuwien.ac.at</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">If you do not mix k-parallel and mpi-parallel runs, it is probably save. (i.e. for VERY big cases)<br>
<br>
The question is: why are you using mpiexec_mpt ??<br>
<br>
Usually it is better to use the more general mpirun (or the less general mpiexec) instead<br>
of the "special mpiexec_mpt comand.<br>
<br>
Besides that, for some mpi-versions/commands it might be -host or -hostfile instead of -machinefile<br>
<br>
checkout, which mpiXXX commands do you have, and with man/help pages, what are the available options.<br>
<br>
To my understanding PBS_NODEFILE is a variable of some queuing-systems (like PBS, but maybe some others<br>
too. It is not really "mpi-specific".<br>
<br>
PS: In "mixed" parallelizations you want to request eg. 256 cores, but you may have 4 k-points, thus doing<br>
4 k-parallel lapw1-jobs on 64 cores each. In such a situation you need to tell mpi, on which nodes it should run<br>
a particular process. And I'm pretty sure, that EVERY (meaningful) installation can do this.<br>
<br>
Am 08.04.2013 19:43, schrieb Luis Ogando:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">
Dear all,<br>
<br>
Dr. Gavin Abo pointed me out that I should have mentioned that I am using mpiexec_mpt instead of mpiexec, so, in my case, the mpi execution is controlled by the<br>
PBS_NODEFILE variable and not by a machines file (thanks Dr. Abo for this).<br>
Anyway, I would like to know if it is safe to run MPI Wien2k with mpiexec_mpt.<br>
All the best,<br>
Luis<br>
<br>
PS: I have no problem on generating the .machines file "on the fly" (queuing system). Despite the fact that it will not be used by mpiexec_mpt, I know that it is required<br>
by Wien2k.<br>
<br>
<br>
<br>
<br>
---------- Forwarded message ----------<br></div><div class="im">
From: *Gavin Abo* <<a href="mailto:gsabo@crimson.ua.edu" target="_blank">gsabo@crimson.ua.edu</a> <mailto:<a href="mailto:gsabo@crimson.ua.edu" target="_blank">gsabo@crimson.ua.edu</a>>><br>
Date: 2013/4/8<br>
Subject: Re: [Wien] -machinefile _HOSTS_<br></div><div class="im">
To: Luis Ogando <<a href="mailto:lcodacal@gmail.com" target="_blank">lcodacal@gmail.com</a> <mailto:<a href="mailto:lcodacal@gmail.com" target="_blank">lcodacal@gmail.com</a>>><br>
<br>
<br>
Dear Luis,<br>
<br>
You probably should have mentioned that you are using 'mpiexec_mpt' not 'mpiexec'.<br>
<br>
-machinefile is an option for mpiexec [<a href="http://linux.die.net/man/1/mpiexec" target="_blank">http://linux.die.net/man/1/<u></u>mpiexec</a>], but it doesn't seem to be a option for mpiexec_mpt<br>
[<a href="http://techpubs.sgi.com/library/tpl/cgi-bin/getdoc.cgi?coll=linux&db=man&fname=/usr/share/catman/man1/mpiexec_mpt.1.html" target="_blank">http://techpubs.sgi.com/<u></u>library/tpl/cgi-bin/getdoc.<u></u>cgi?coll=linux&db=man&fname=/<u></u>usr/share/catman/man1/mpiexec_<u></u>mpt.1.html</a>].<br>
<br>
mpiexec_mpt seems to use the PBS_NODEFILE variable instead of -machinefile [<a href="http://www.arl.hpc.mil/docs/pbsUserGuide.html" target="_blank">http://www.arl.hpc.mil/docs/<u></u>pbsUserGuide.html</a>]. So your parallel_options are probably fine as<br>
long as the PBS_NODEFILE variable is set automatically by your system or you.<br>
<br>
Kind Regards,<br>
<br>
Gavin<br>
<br>
<br>
On 4/8/2013 10:18 AM, Luis Ogando wrote:<br>
</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">
Hi Gavin,<br>
<br>
Thank you for your answer.<br>
Actually, I generated the .machines files for the queuing system without problems. The issue is that the " setenv WIEN_MPIRUN " in the "parallel_options" file has to<br>
be " setenv WIEN_MPIRUN "mpiexec_mpt -np _NP_ _EXEC_" ", without the "-machinefile" option because it is not defined in the system.<br>
All the best,<br>
Luis<br>
<br>
<br>
<br>
<br></div>
2013/4/8 Gavin Abo <<a href="mailto:gsabo@crimson.ua.edu" target="_blank">gsabo@crimson.ua.edu</a> <mailto:<a href="mailto:gsabo@crimson.ua.edu" target="_blank">gsabo@crimson.ua.edu</a>>><div class="im"><br>
<br>
Dear Luis,<br>
<br>
I'm sending this email off the mailing list as Prof. Marks or Blaha might respond with a better answer.<br>
<br>
I think the answer is yes, it can impact the Wien2k performance. This is because the machinefile variable contains the list of hostnames for 'multiple' nodes.<br>
Without it, the calculation will likely run on only 'one' node.<br>
<br>
Even if you don't have admin privileges, you can likely still define the machinefile variable as a user by creating a .machines file in your case directory. However,<br>
the creation of the .machines file may depend on whether or not you are required to use a queuing system. If you are not required to use a queuing system, you should<br>
be able to copy the .machines file in SRC_templates to your case directory and then edit it a text editor (note: you might not see the .machines file unless you do a<br>
directory listing that includes hidden files). The .machines file is should be described in the Wien2k userguide or you can search the internet for some examples<br>
[<a href="https://www.xsede.org/documents/10157/305826/ecss_hliu_051012.pdf" target="_blank">https://www.xsede.org/<u></u>documents/10157/305826/ecss_<u></u>hliu_051012.pdf</a>]. If you are required to use a queuing system, you likely need to setup a script that will create<br>
the .machines file as described at the link:<br>
<br>
<a href="http://www.wien2k.at/reg_user/faq/pbs.html" target="_blank">http://www.wien2k.at/reg_user/<u></u>faq/pbs.html</a><br>
<br>
If you have problems creating the script, your administrator(s) or support person(s) should know the most about your computer system, so they can likely help you<br>
create a script that will work on your system.<br>
<br>
Kind Regards,<br>
<br>
Gavin<br>
<br>
<br>
On 4/8/2013 6:19 AM, Luis Ogando wrote:<br>
<br>
Dear Prof. Marks, Blaha and Wien2k community,<br>
<br>
I want to do calculations in a computer where the machinefile variable is not defined for the mpiexec (I am not the administrator). I would like to know if<br>
this will have some impact on the MPI Wien2k performance.<br>
Thanks in advance,<br>
Luis<br>
<br>
<br>
<br>
</div></blockquote>
<br>
<br>
<br>
<br>
______________________________<u></u>_________________<br>
Wien mailing list<br>
<a href="mailto:Wien@zeus.theochem.tuwien.ac.at" target="_blank">Wien@zeus.theochem.tuwien.ac.<u></u>at</a><br>
<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.<u></u>ac.at/mailman/listinfo/wien</a><br>
SEARCH the MAILING-LIST at: <a href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html" target="_blank">http://www.mail-archive.com/<u></u>wien@zeus.theochem.tuwien.ac.<u></u>at/index.html</a><br>
<br>
</blockquote>
<br>
-- <br>
------------------------------<u></u>-----------<br>
Peter Blaha<br>
Inst. Materials Chemistry, TU Vienna<br>
Getreidemarkt 9, A-1060 Vienna, Austria<br>
Tel: <a href="tel:%2B43-1-5880115671" value="+4315880115671" target="_blank">+43-1-5880115671</a><br>
Fax: <a href="tel:%2B43-1-5880115698" value="+4315880115698" target="_blank">+43-1-5880115698</a><br>
email: <a href="mailto:pblaha@theochem.tuwien.ac.at" target="_blank">pblaha@theochem.tuwien.ac.at</a><br>
------------------------------<u></u>-----------<br>
______________________________<u></u>_________________<br>
Wien mailing list<br>
<a href="mailto:Wien@zeus.theochem.tuwien.ac.at" target="_blank">Wien@zeus.theochem.tuwien.ac.<u></u>at</a><br>
<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.<u></u>ac.at/mailman/listinfo/wien</a><br>
SEARCH the MAILING-LIST at: <a href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html" target="_blank">http://www.mail-archive.com/<u></u>wien@zeus.theochem.tuwien.ac.<u></u>at/index.html</a><br>
</blockquote></div><br></div>