There seems to be a mistake either in the documentation or the Star Cluster
configuration. The documentation currently states:
In
http://star.mit.edu/cluster/docs/0.94/plugins/sge.html, under "Parallel
Environment Allocation Rule", it states:
By default StarCluster uses the *fill_up* allocation rule.
However, when I look at the queue configuration, I see:
ec2-user_at_master:~$ qconf -sp orte
pe_name orte
slots 4
user_lists NONE
xuser_lists NONE
start_proc_args /bin/true
stop_proc_args /bin/true
allocation_rule $round_robin
control_slaves TRUE
job_is_first_task FALSE
urgency_slots min
accounting_summary FALSE
The allocation rule shows round_robin. I would prefer to have fill_up
myself, so I'm not sure if the docs are incorrect, or the configuration of
StarCluster is incorrect. And, how do I fix this in my config script?
Received on Tue Sep 17 2013 - 11:11:47 EDT
This archive was generated by
hypermail 2.3.0.