GDPS performance and parallelism

What is GDPS performance and parallelism?

GDPS® performance and parallelism increases the speed with which GDPS drives hardware actions, along with enhancing the speed of the underlying hardware services.

The GDPS enhancements are a combination of changes to GDPS processing/scripting itself, and GDPS taking advantage of new function offered by the z15 Support Element (SE). APIs used by GDPS on the SE are streamlined and improved, while GDPS does more in parallel when possible. In addition, enhancements to the GDPS scripting language allow multiple systems to be addressed in parallel by some commands and improve the efficiency of some control flows.

How do the GDPS enhancements work?

The GDPS enhancements are available in GDPS 4.2 and above running on a z15 or higher CPC.

With GDPS 4.2 the system infrastructure management has been improved, allowing a reduction in the observed recovery time objective (RTO) when there is a need for the IT infrastructure to be reconfigured, such as with on-site system restart, planned site or region switch, or a disaster recovery process. This improvement is achieved by optimizing GDPS and CPC interactions and by providing the possibility for customers to perform actions against a group of system/LPARs rather than against each one individually.

Before this change, all actions were done sequentially, repeating in each step the required handshaking between GDPS and the CPCs hosting the systems. Grouping the actions eliminates the duplicate handshaking processes and reduces substantially the overall time needed to manage multiple systems.