[GE users] Yet anaother MPICH tight-integration problem

David S. dgs at gs.washington.edu
Tue Sep 7 16:11:02 BST 2004


> 
> can you please post the configuration of your PE? The cpi demo of mpich is 
> working fine for me.

Here's the output of 'qconf -sp mpich':

	pe_name           mpich
	slots             78
	user_lists        NONE
	xuser_lists       NONE
	start_proc_args   /mnt/local/sge/mpi/startmpi.sh -catch_rsh $pe_hostfile
	stop_proc_args    /mnt/local/sge/mpi/stopmpi.sh
	allocation_rule   $round_robin
	control_slaves    TRUE
	job_is_first_task TRUE
	urgency_slots     min

> 
> >	setenv P4_RSHCOMMAND $TMPDIR/rsh
> >	#setenv P4_RSHCOMMAND /usr/bin/rsh
> 
> One note: setting it to rsh should be sufficient.
> 
> Do you have two network cards in the machines?

All the nodes a dual-homed.   We are using IBM's General Parallel File
System, and have one network dedicated to it.  The queue master and 
submission host actually has three network cards.

David S.

> 
> Reuti
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list