the local Adabas address space. Now it is possible to shut down the other Adabas address space to apply maintenance. If you are finished with one LPAR and Adabas has been restarted, you can do the same with the other LPAR. No more single point of failure As mentioned in 6.6.1, "Disadvantages" on page 131, the IZB Adabas Cluster Services has no single point of failure. If there is a problem with the Coupling Facility, it is possible to change the Adabas operation from cluster mode to non-cluster mode. Workload balancing with WLM The definitions in the WLM policy determine the workload balancing and priority within one LPAR or over many LPARs. Using the velocity statistics, Adabas Cluster nuclei get a dispatching priority to allow them to obtain their goal. See Chapter 9, "System" on page 177 for more information about this topic. Maintenance without outages Today it is possible to apply Adabas software updates in the evening and not in the maintenance window during the weekend. Because there is no outage in applications, availability increases and the risk of too many changes at the same time decreases. Scalability By changing the scheduling environment for batch jobs or changing the policy of WLM for CICS or other applications, it is possible to control the distribution of workload in the sysplex environment. This function also helps to increase application availability. Multi-processor usage Although Adabas is a single-processor DBMS, with Cluster Services this limitation is