Decentralized Dynamic Query Optimization based on Mobiles Agents for Large Scale Data Integration Systems

Abstract

The query processing in large scale distributed mediations systems raises new problems and presents real challenges: efficiency of access, communication, confidentiality of access, availability of data, memory allocation. In this paper, we propose an execution model based on mobile agents for the distributed dynamic query optimization. In this model, each relational operator of an execution plan is executed by a mobile agent. Also, we embed into agent a migration policy allowing agent to choose an execution site among execution sites of the considered system. The performance evaluation shows that the proposed model improves the response time whatever the variation of estimations errors.