HP-UX Workload Manager User's Guide

Example configuration files
distribute_excess.wlm
Chapter 9 287
# prm structure
#
# In this example we have two groups using the machine, Orders
# and Sales. In the event there are additional resources available,
# we’d like for Sales to get three shares for every share that Orders
# receives.
#
# See wlmconf(4) for complete HP-UX WLM configuration information.
#
prm {
groups= OTHERS : 1,
Orders : 2,
Sales : 3;
weight= Orders : 1,
Sales : 3;
}
#
# slo structure
#
# In this example we have two SLOs, one for actual order processing
# and another for sales’ product availability requests. These are
# deemed to be equal in priority and have goals based on the
# response times of their applications.
#
# In the event that both goals are met, we want to distribute the
# excess CPU resources according to the weight specification above.
# See the tune structure explanation for details on what the impact
# might be in this situation.
#
slo order_processing {
pri = 1;
mincpu = 0;
maxcpu = 100;
entity = PRM group Orders;
goal = metric order_transaction_time < 10.0;
}
slo product_availability {
pri = 1;
mincpu = 0;
maxcpu = 100;
entity = PRM group Sales;
goal = metric sales_request_time < 5.0;
}