A virtual memory system for a relational associative processor

The Relational Associative Processor (RAP) is an experimental "backend" cellular processor for implementing a data base management system which supports Codd's relational model of data. The capacity of a RAP device implemented with current IC and memory technology can be estimated to contain 10⁸ to 10⁹ bits of associatively processable data. Because of this limitation, a virtual memory environment for RAP has been proposed and its performance simulated. The environment incorporates conventional memories for bulk storage and a single RAP processor both controlled by a front-end computer. An entire relational data base is divided into pages of size equal to one RAP cell memory. A buffer memory is added to RAP to permit the overlap of paging with processing. It has been found that environments which process user queries that exhibit long processing times relative to paging time or some "locality", that is, a better than random probability of referencing the same relations over some interval of time, can efficiently page data from large data bases without significant losses in performance.