The Proxy Coupling Technology solution offloads CPU resources to lower-cost hardware and software, reducing your Total Cost of Ownership (TCO) by up to 80% with no loss in Reliability, Availability, and Serviceability (RAS) and extending the value of the z System Mainframe for your business.
Primarily, it is envisaged that the majority of z System Mainframe users will utilize their investment in IFL specialty engines as an available resource to offload MSUs. Some z System Mainframe users seeking a further cost reduction might choose to offload a modicum of their z System Mainframe CPU workload to Distributed Systems technology, assisting them to maintain the cost viability of their z System Mainframe environment in the long-term. In essence, this releases generic General Purpose Processor CPU resources for other native z System mainframe activity. Either way, all z System Mainframe users will continue to enjoy the RAS attributes of the “zero downtime” Mainframe server, with the automatic “failback” mechanism of Proxy Coupling Technology.
PSCsort™ and proxCF™ are priced on a per-MIPS basis, resulting in a cost savings of up to 80% vs. running the same workload(s) in the GPP environment. A 100 MIPS minimum applies to a proxCF™ license, and a lower, entry-level threshold of 10 MIPS applies to a PSCsort™ license. PSCsort™ may be bundled with proxCF™ at no extra charge. The per-MIPS rate decreases with offload capacity, thus achieving greater economy of scale as adoption increases.
Essentially, your choice of processing task may be offloaded from one system to another by means of a proxy task, which has the identical interface as the task to be offloaded, delegating the majority of CPU activity to an offload task on another system.
The Proxy Coupling technology deployed by proxCF™ safeguards that no application or job changes are required, delivering an easy-to-use minimal impact solution. Furthermore, proxCF™ delivers a solution that defaults back to the incumbent z System mainframe processing platform should the target offload CPU platform be unavailable.
The above diagram illustrates the path from the mainframe to the cloud as supported by proxCF™. But it is entirely up to the customer to decide how fast and how far to pursue it – viable and cost-effective solutions exist at every point along the way:
Transparently offload z System MSU demand for data sorting from General Purpose Processors (GPP) to the Integrated Facility for Linux (IFL) specialty engine or Distributed Systems CPU platforms.
The sort function is widely used within many IBM z System Mainframe applications, sometimes being the major CPU resource user, primarily due to the type of application being processed. PSCsort™ provides an easy-to-install and easy-to-use option to offload some or all of this sort function processing, releasing GPP MSUs for other mission critical business application processing.