Query optimization in multidatabase systems

Abstract

A multidatabase system (MDBS) integrates information from autonomous local databases managed by heterogeneous database management systems (DBMS) in a distributed environment. For a query involving more than one database, global query optimization should be performed to achieve good overall system performance. The significant differences between an MDBS and a traditional distributed database system (DDBS) make query optimization in the former more challenging than in the latter. Challenges for query optimization in an MDBS are discussed in this paper. A two-phase optimization approach for processing a query in an MDBS is proposed. Several global query optimization techniques suitable for an MDBS, such as semantic query optimization, query optimization via probing queries, parametric query optimization and adaptive query optimization, are suggested. The architecture of a global query optimizer incorporating these techniques is designed.