HP-UX Workload Manager Toolkits User's Guide

HP-UX WLM Oracle Database Toolkit: Providing Database Metrics to WLM
How do I use the metrics?
Chapter 2 43
# Have HP-UX WLM give workload ora2_grp three CPU shares for each user
# connected to the associated Oracle instance. See the tune structure below
# for details on how the user counts are collected.
#
slo ora2_slo {
pri = 2;
mincpu = 5;
maxcpu = 90;
entity = PRM group ora2_grp;
cpushares = 3 total per metric oracle.instance2.user_cnt;
}
#
# Have HP-UX WLM give workload long_batch a fixed allocation of 20 CPU
# shares. The priority of the workload is set very low, thus giving
# preference to the Oracle workloads over the long_batch jobs.
#
slo long_batch {
pri = 10;
cpushares = 20 total;
entity = PRM group long_batch;
}
#
# Any CPU that remains after satisfying the above SLOs is given to the
# OTHERS group by default. You can change this default using the
# distribute_excess keyword. For more information on this keyword, see
# the wlmconf(4) manpage.
#
#
# Set the basic interval on which wlmd allocation changes
# are made to 30 seconds. For optimal performance with Oracle, do not use
# an interval of less than 10 seconds.
#
tune {
wlm_interval = 30;
#
# Collect any datacol stderr and send it to syslog.
# This is handy for catching errors a missing
# /opt/perl/bin/perl when using wlmoradc.
#
coll_stderr = syslog;
}