RAP—a Relational Associative Processor—is a back-end or peripheral device to augment a general purpose computer for implementing a database management system (DBMS) Its architecture is based on the fact that database operations are inherently set-oriented and that data base addressing is best accomplished through associative reference to achieve high data independence. RAP utilizes these characteristics by combining the features of associative and array processors. Previous publications on RAP have dealt separately with the details of the first version of its architecture [1]-[4] language interface [5],[6] and performance evaluation [7]-[9]. This paper provides details on a recently evolved, faster, and more flexible architecture for RAP called RAP. 2 [17].

RAP - a Relational Associative Processor - is a back-end or peripheral device to augment a general purpose computer for implementing a data base management system (DBMS). Its architecture is based on the fact that data base operations are inherently set-oriented and that data base addressing is best accomplished through associative reference to achieve high data independence. RAP utilizes these characteristics by combining the features of associative and array processors. Previous publications on RAP have dealt separately with the details of the first version of its architecture [1,2,3,4] language interface [5,6] and performance evaluation [7,8,9,]. This paper provides details on a recently evolved, faster, and more flexible architecture for RAP.

An associative processor called RAP has been designed to provide hardware support for the use and manipulation of databases. RAP is particularly suited for supporting relational databases. In this paper, the relational operations provided by the RAP hardware are described, and a representative approach to providing the same relational operations with conventional software and hardware is devised. Analytic models are constructed for RAP and the conventional system. The execution times of several of the operations are shown to be vastly improved with RAP for large relations.

An algorithm to evaluate primitive Boolean selections over single relations is presented. It will be argued that the algorithm is efficient with respect to the number of relational accesses and with respect to the merging of inverted lists. The algorithm's unique quality is its efficiency in evaluating selections over partially inverted relations. A simple cost function is used to drive the algorithm along the most efficient access paths. The cost function can also be used to predict its response time which then forms the basis of a procedure to suboptimize the selection of the domains to be inverted. The domains to be inverted are selected by analyzing, with respect to their costs, a sample of queries. Such a method does away with usual methods of updating usage counters for every domain and relation in the system. In this approach, the selection of a good set of inverted lists is based on the algorithm which uses those lists.