Automated Capacity Management with sub-capacity pricing (R4HA)

The R4HA Limit Type is used in conjunction with IBM's sub-capacity pricing to reduce software costs. When this limit type is enabled, Automated Capacity Management (ACM) monitors the rolling 4-hour average (R4HA) of CPU usage, the basis for sub-capacity software charges, and controls batch workload in order to reduce that average. ACM builds on the TM AE Policy by specifying constraints that are applied as the R4HA approaches the capacity limit.

This chart shows that, while the interval CPU usage may fluctuate significantly, the R4HA changes slowly. The peaks in the interval usage do not necessarily coincide with the peaks in the R4HA. Sub-capacity pricing is based on the highest monthly peak in the R4HA, not the interval peak.

When used together with soft capping, ACM minimizes the effects of reaching the capacity limit or avoids capping altogether, and enables your datacenter to lower your R4HA limit without impacting your online and critical batch systems.

The scope of ACM functionality is the LPAR. Since the scope of a ThruPut Manager AE Policy is a JESplex, ACM settings within the Policy are applicable to each LPAR within the JESplex.

Though not mandatory, results are optimized when ACM is enabled on all the LPARs within a CPC.