[torqueusers] job allocation

Marco De Simone marco at cbm.fvg.it
Thu Oct 26 07:45:55 MDT 2006


Hi all,

this is the first time I use Torque as resource manager,

i have a cluster made of 30 nodes,
what i want is max 4 jobs running on each node,

and what I have now with the following setup is that more then  half are
at full load
while the rest are free or with 1,2 jobs running and in the queue there
are still other 10000 jobs...
in other words I don't see a uniform job allocation among the nodes..


this is my setup :

create queue batch
set queue batch queue_type = Execution
set queue batch max_running = 120
set queue batch resources_default.nodes = 1:ppn=1
set queue batch enabled = True
set queue batch started = True
#
# Set server attributes.
#
set server scheduling = True
set server max_running = 120
set server acl_host_enable = False
set server acl_hosts = *
set server default_queue = batch at masternode
set server log_events = 511
set server mail_from = adm
set server query_other_jobs = True
set server scheduler_iteration = 600
set server node_check_rate = 150
set server tcp_timeout = 6
set server default_node = 1
set server pbs_version = 2.1.2



what can i do to improve my throughput ?

thanks a lot

Marco






More information about the torqueusers mailing list