METHOD AND APPARATUS FOR A UNIFIED PARALLEL PROCESSING ARCHITECTURE.

摘要

A unified parallel processing architecture connects together an extendible number of clusters (100a, 100b, 100c, 100d) of multiple numbers of processors (102a, 102b, 102c, 102d) to create a high performance parallel processing computer system. Multiple processors are grouped together into four or more physically separable clusters (100a, 100b, 100c, 100d) each cluster having a common cluster shared memory (104a, 104b, 104c, 104d) that is symmetrically accessible by all of the processors in that cluster; however, only some of the clusters are adjacently interconnected. Clusters are adjacently interconnected (100a, 100c) to form a floating shared memory (104a, 104c) if certain memory access conditions relating to relative memory latency and relative data locality can create an effective shared memory parallel programming environment. A shared memory model can be used with programs that can be executed in the cluster shared memory of a single cluster, or in a floating shared memory that is defined across an extended shared memory space comprised of the cluster shared memories of any set of adjacently interconnected clusters. The adjacent interconnections of multiple clusters of processors to create a floating shared memory effectively combines all three types of memory models, pure shared memory (20, 22, 24), extended shared memory (30, 34, 32) and distributed shared memory (14, 12, 10) into a unified parallel processing architecture.