HP-UX Workload Manager Toolkits User's Guide

HP-UX WLM Duration Management Toolkit and HP-UX WLM Toolkit for Base SAS Software
How do I use DMTK / SASTK?
Chapter 6122
tune duration_1_procs_active {
coll_argv = wlmrcvdc glance_prm -i 2 APP_ACTIVE_PROC duration_1;
}
# Configure wlmdurdc to launch the example discovery script
# (/opt/wlm/toolkits/duration/examples/duration.sh). The duration.sh script
# takes two arguments: a workload group name (duration_1) and a file
# (dur_1_rate) containing application names, profile values, and desired
# durations. This file is created by the user based on values found in
# /var/adm/syslog/syslog.log when using “wlmdurdc -P”.
tune dur_1_rate {
coll_argv = wlmrcvdc /opt/wlm/toolkits/duration/bin/wlmdurdc
/opt/wlm/toolkits/duration/examples/duration.sh
duration_1 dur_1_rate;
}
# Next, create the SLO slo_dur_1. This SLO has a priority of 1. It has a
# CPU request range of 1 to 50 (mincpu=1, maxcpu=50). A priority 1 SLO has
# its shares allocated before a lower priority SLO. Thus, WLM will satisfy
# slo_dur_1 at the expense of the group OTHERS. The SLO defines a goal
# that allows its target application’s run-rate to be managed by WLM. The
# goal statement is typical for SLOs pertaining to duration management,
# with a goal that the metric be greater than 0.
#
# In addition, slo_dur_1 has a condition metric,
# duration_1_procs_active. This condition metric specifies that at least
# one process must be executing in the group before the SLO becomes active.
#
# This SLO is inactive when there are no processes in its associated
# workload group. Otherwise, the SLO provides enough shares to satisfy the
# desired duration requirements.
slo slo_dur_1 {
pri = 1;
entity = PRM group duration_1;
mincpu = 1;
maxcpu = 50;
goal = metric dur_1_rate > 0;
condition = metric duration_1_procs_active > 0;
}