A Framework for Cooperative Deductive Database Systems

Abstract

In this paper, we address the problems of design, management, and integration of deductive database systems in a loosely coupled architecture, which constitute a cooperative deductive database system. We next address one important aspect of the problem of designing a cooperative deductive database system, namely, allocation of rules across the deductive database systems. We identify communication cost as the primary consideration in allocation of rules. The problem of optimal allocation of rules has been shown NP-complete, which has prohibitive execution times for large knowledge bases.  We propose a naďve algorithm for rule allocation and study its performance experimentally. We also show that this naďve algorithm can be used for reallocation of rules after rulebase gets updated.