Since the early 1970s mainframes have been designed as multiprocessor systems, even when only a single processor is installed. All but the smallest mainframe installations typically use clustering techniques, although they do not normally use the terms cluster or clustering.

A clustering technique can be as simple as a shared DASD configuration where manual control or planning is needed to prevent unwanted data overlap. More common today are configurations that allow sharing of locking and enqueueing controls among all systems. Among other benefits, this automatically manages access to data sets so that unwanted concurrent usage does not occur.

The most sophisticated of the clustering techniques is a Parallel Sysplex®. This technology allows the linking of up to 32 servers with nearly linear scalability to create a powerful commercial processing clustered system. Every server in a Parallel Sysplex cluster has access to all data resources, and every "cloned" application can run on every server. When used with coupling technology, Parallel Sysplex provides a "shared data" clustering technique that permits multisystem data sharing with high performance read/write integrity.

Sysplex design characteristics help businesses to run continuously, even during periods of dramatic change. Sysplex sites can dynamically add and change systems in a sysplex, and configure the systems for no single points of failure. Through this state-of-the-art cluster technology, multiple z/OS® systems can be made to work in concert to more efficiently process the largest commercial workloads.