HP-UX Workload Manager User's Guide

Understanding how PRM manages resources
Management of CPU resources
Appendix F448
Management of CPU resources
When you configure WLM, you specify minimum and maximum requests
for CPU shares and/or a shares-per-metric request for each SLO. Based
on the SLO’s priority, the workload’s performance, and the system’s
available CPU resources, WLM manages each workload group’s number
of CPU shares, increasing or decreasing that number automatically.
WLM drives each workload group’s CPU resources to a certain number of
shares, resulting in a new PRM configuration. PRM then enables the
new configuration, giving groups with more shares more opportunities to
use CPU time. (WLM partition management is performed without the
PRM configuration.)
For an overview of the method WLM uses to allocate CPU resources, see
“Allocating CPU resources: The rising tide model” on page 122.
Management of CPU resources for real-time processes
Although PRM is designed to treat processes fairly based upon their
assigned shares, PRM does not restrict real-time processes. Real-time
processes using either the POSIX.4 real-time scheduler (rtsched) or the
HP-UX real-time scheduler (rtprio) keep their assigned priorities
because timely scheduling is crucial to their operation. Hence, they are
permitted to exceed their group’s number of CPU shares and its cap.
However, the CPU resources they use are still counted as part of the
CPU allocated to their groups. Thus, they can prevent other processes in
their groups from running.