IXM2: A Parallel Associative Processor for Knowledge

This paper describes a parallel associative pro-cessor, IXM2, developed mainly for semantic network processing. IXM2 consists of 64 as-sociative processors and 9 network processors, having a total of 256K words of associative memory. The large associative memory en-ables 65,536 semantic network nodes to be pro-cessed in parallel and reduces the order of al-gorithmic complexity to O(1) in basic semantic net operations. We claim that intensive use of associative memory provides far superior per-formance in carrying out the basic operations necessary for semantic network processing: in-tersection, marker-propagation, and arithmetic operations.