[gmx-users] problem with cluster

[Pietro Lopriore]

Dear Nguyen,
we had a similar problem in our cluster (powerPC based).
a) the following is an example

grompp –f grompp.mdp –c conf.gro –p topol.top –po mdout.mdp –shuffle –sort
–np <nr. processors>

mpirun c0, 2, 7 –ssi rpi gm
/usr/local/gromacs/powerpc64-unknown-linux-gnu/bin/mdrun –s topol.tpr –o
traj.trr –c confout.gro –e ener.edr –g md.log

In this case, specifying 0,2 and 7 means that the mdrun runs on processors
0, 2 and 7 as specified in the mpi machine file (We have a two-ways
cluster; node 1:  processors 0 and 1; node 2: processors 2 and 3,
etc.....).

b) We had a similar problem but it could be due to too many things. We had
problems with the CSM software launching ghost processes around. Swithcin
it off everything went fine (check the SYS usage of the memory). But it
could be also due to other things like networking or any other application
running on the cluster.

Hope this helps,

Pietro
_________________________________________________

Pietro Lopriore
EMEA South Region Technical Sales Support
IBM Healthcare and Life Sciences Industry
IBM Life Sciences Internet site: http://www.ibm.com/solutions/lifesciences
IBM Computational Biology Center:
http://www.research.ibm.com/compsci/compbio



                                                                           
             Nguyen Hoang                                                  
             Phuong                                                        
             <phuong at theochem.                                          To 
             uni-frankfurt.de>         gmx-users at gromacs.org               
             Sent by:                                                   cc 
             gmx-users-bounces                                             
             @gromacs.org                                          Subject 
                                       [gmx-users] problem with cluster    
                                                                           
             04/13/2005 03:09                                              
             PM                                                            
                                                                           
                                                                           
             Please respond to                                             
              Discussion list                                              
             for GROMACS users                                             
                                                                           
                                                                           





Dear All,

we have a new cluster consists of 11 Intel cpus with hyperthreading. I
have some problems with running gromacs on this cluster:

a) When I submit a job, it is distributed at random on 2 cpus (I asked for
2 cpus). How can I run the job on specific cpus?

b) The cpus which run the mdrun program alway work with about 40% of
power. How can I increase the power of the cpus?

Thank you very much in advance for any hint.

Phuong

_______________________________________________
gmx-users mailing list
gmx-users at gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-request at gromacs.org.