[Wien] w2web -> cluster

[Florent Boucher]

Dear Marc,
we have in Nantes a configuration with a cluster of different node and 
LSF. We have a front end and are not allowed to run directly wien on the 
cluster.
So, it seems very similar to your configuration.
In order to have Wien running, I had to adapt a little the scripts and 
define specific commands but it is not too difficult.
First, I had to define the remote command in all the *para_lapw in order 
to use lsrun
set remote = "/var/lsf/bin/lsrun -m"

I had to change also the line to execute by removing the cd $PWD, but I 
do not remember why !

            if($number_per_job[$p] == 1 ) then
              if ($useremote == 1) then
                    ($remote  $machine[$p] $t $exe ${def}_$loop.def;rm 
-f .lock_$lockfile[$p]) >>.time1_$loop &     <--- this one
              else
                  (cd $PWD;$t $exe ${def}_$loop.def;rm -f 
.lock_$lockfile[$p]) >>.time1_$loop &
              endif
            else
              set ttt=(`echo $mpirun | sed -e 
"s^_NP_^$number_per_job[$p]^" -e "s^_EXEC_^$WIENROOT/${exe}_mpi 
${def}_$loop.def^" -e "s^_HOSTS_^.mach
ine[$p]^"`)
              (cd $PWD;$t $ttt;rm -f .lock_$lockfile[$p]) >>.time1_$loop &
            endif

I have created a small script the can extract the name of the node to 
submit wien in order to create automatically the .machines file
boucher at orchidee - 79% more createmachine
#!/bin/csh -f
#
rm -f .machines
foreach host ($LSB_HOSTS)
    echo "1:"$host >> .machines
end
echo 'granularity:1' >> .machines
echo 'extrafine'     >> .machines
set HOST = `echo "$LSB_HOSTS " | sed -e 's^ ^:1 ^g'`
echo "lapw0:$HOST" >> .machines

Then I define in w2web the specific command I need:
it is in the file execution.conf
/home/csesmao/boucher/.w2web/phoenix/conf
boucher at orchidee - 84% more execution.conf
single_1H=  /usr/local/lsf/bin/bsub -q developpement -n 1
single_72H= /usr/local/lsf/bin/bsub -q 2-T72         -n 1
2sur_72H=   /usr/local/lsf/bin/bsub -q 2-T72         -n 2  -E createmachine
2sur_1H=    /usr/local/lsf/bin/bsub -q developpement -n 2  -E createmachine
3sur_1H=    /usr/local/lsf/bin/bsub -q developpement -n 3  -E createmachine
4sur_1H=    /usr/local/lsf/bin/bsub -q developpement -n 4  -E createmachine
3sur9-T12=  /usr/local/lsf/bin/bsub -q 9-T12         -n 3  -E createmachine
4sur9-T12=  /usr/local/lsf/bin/bsub -q 9-T12         -n 4  -E createmachine
5sur9-T12=  /usr/local/lsf/bin/bsub -q 9-T12         -n 5  -E createmachine
6sur9-T12=  /usr/local/lsf/bin/bsub -q 9-T12         -n 6  -E createmachine
8sur9-T12=  /usr/local/lsf/bin/bsub -q 9-T12         -n 8  -E createmachine
9sur9-T12=  /usr/local/lsf/bin/bsub -q 9-T12         -n 9  -E createmachine
8sur16-T24= /usr/local/lsf/bin/bsub -q 16-T24        -n 8  -E createmachine
9sur16-T24= /usr/local/lsf/bin/bsub -q 16-T24        -n 9  -E createmachine
10sur16-T24= /usr/local/lsf/bin/bsub -q 16-T24        -n 10  -E 
createmachine
16sur16-T24=/usr/local/lsf/bin/bsub -q 16-T24        -n 16 -E createmachine
3sur9-T72=  /usr/local/lsf/bin/bsub -q 9-T72         -n 3  -E createmachine
4sur9-T72=  /usr/local/lsf/bin/bsub -q 9-T72         -n 4  -E createmachine
6sur9-T72=  /usr/local/lsf/bin/bsub -q 9-T72         -n 6  -E createmachine
8sur9-T72=  /usr/local/lsf/bin/bsub -q 9-T72         -n 8  -E createmachine
9sur9-T72=  /usr/local/lsf/bin/bsub -q 9-T72         -n 9  -E createmachine

If you don't use w2web for the submission, you just have to use the bsub 
command with the -E createmachine option
I hope it helps
Regards
Florent


Marc Petitmermet a écrit :
> Dear List Members
>
> Imaging that you have users who only know how to use wien2k using the  
> w2web gui. They are using it on there local Linux clients.
>
> I'm now challenged to configure their setup so that they can submit  
> parallel jobs to a cluster (288 dual-processor compute nodes based on  
> AMD Opteron 250 processors, 8 GB memory connected by means of a  
> Quadrics QsNet II interconnect). Installation of wien2k on the  
> cluster and password-less ssh connection from client to cluster are  
> no problem. But how do I get the job correctly submitted to the  
> queueing system (LSF, bsub [options] -n N prun ./program) on the  
> cluster from the w2web gui on the client (we are not allowed to run  
> w2web on the cluster itself)? Furthermore, I don't know the node  
> names which will be used in advance. I have found some information on  
> http://www.wien2k.at/reg_user/faq/pbs.html but I still don't know yet  
> how to "glue" all this information together. Is this possible at all  
> using the w2web gui? Thanks in advance for any pointers.
>
> Regards,
> Marc
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>   


-- 
 -------------------------------------------------------------------------
| Florent BOUCHER                    |                                    |
| Institut des Matériaux Jean Rouxel | Mailto:Florent.Boucher at cnrs-imn.fr |
| 2, rue de la Houssinière           | Phone: (33) 2 40 37 39 24          |
| BP 32229                           | Fax:   (33) 2 40 37 39 95          |
| 44322 NANTES CEDEX 3 (FRANCE)      | http://www.cnrs-imn.fr             |
 -------------------------------------------------------------------------