[GE users] Short medium and long queue setup

Reuti reuti at staff.uni-marburg.de
Wed Oct 18 10:42:39 BST 2006


Am 18.10.2006 um 03:03 schrieb Jerome:

> Jeff Block wrote:
>> Unfortunately, I didn't receive any responses on this with great  
>> ideas on
>> how to make this work.  The subordinate queues probably won't work  
>> for us
>> because of the MPI job suspension issue stated below.
>> Currently, I have just setup nice values on each of the queues  
>> (short-1,
>> medium-10, long-19) and we are in the testing phase to see if that  
>> is good
>> enough for us.
>> The other issue is that I'd like to be able to weight the short  
>> queue higher
>> with regards to fair share tickets used.  This would help to  
>> insure that
>> people don't just always submit jobs to the short queue unless  
>> they really
>> need the job to be processed quickly.
>> Regards.
>> Jeff
> Hi Jeff,
> thank's for your response.
> I used the comments you send me to try on our own cluster.
> But i notice that the nice value with the mpi jobs just affect the  
> master node of the job. All the other mpich program on the "slaves"  
> node run with a ice value of 0. Do you have the same behavior?
> Best regards.

did you set up the PE with Tight Integration for the parallel jobs?  
All processes on the slaves a children of the sge_shepherd?

-- Reuti

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