Eygene Ryabinkin wrote: > May be you should try to add 'np=X' to the end of the node description > in <PBS>/server_priv/nodes, telling Torque that this particular > node has X CPUs. Next thing to do is to play with $ideal_load and > $max_load in the MOM's configuration file to tell Torque what uptime > should be treated as the "Busy" ones.
We did try this in the past, however it became a problem because more than one of the niced 19 jobs would be running on the same node and this caused timing issue with our grid MPI jobs which are our high priority jobs. > May be you'll need to adjust Torque's server and queue variable > resources_available.nodect to make Torque believe that you have > large number of CPUs. But this will be needed if users will submit > jobs for more CPUs that the number of nodes in your cluster. If > your users run serial jobs, this shouldn't be needed. The niced 19 jobs are serial whereas the grid jobs are MPI based. I've nearly got the problem solved, I just need to somehow disable the checksum checking the Maui binaries do when trying to schedule jobs on my 2nd Maui installation. -- Jeremy Mann [EMAIL PROTECTED] University of Texas Health Science Center Bioinformatics Core Facility http://www.bioinformatics.uthscsa.edu Phone: (210) 567-2672 _______________________________________________ mauiusers mailing list [email protected] http://www.supercluster.org/mailman/listinfo/mauiusers
