Proc. of 9th ISCA Int'l Conf. on Parall. and Distr. Comp. Syst., pp 640 - 46,
Sept. 25 - 27, 1996
------------------------------ 


 Global Query Processing and Optimization in the CORDS Multidatabase System

                               Qiang Zhu
             Department of Computer and Information Science
                  The University of Michigan - Dearborn
                          Dearborn, MI 48128, USA
 
                             Per-Ake Larson*
                    Department of Computer Science
                         University of Waterloo
                   Waterloo, Ontario N2L 3G1, Canada
               * Current address: Microsoft Corporation
                          One Microsoft Way
                     Redmond, WA 98052-6399, USA
 
 
                               ABSTRACT  
    
A multidatabase system (MDBS) integrates information from autonomous
pre-existing local databases managed by heterogeneous local database 
management systems in a distributed environment. To achieve good overall 
system performance, efficient global query processing and optimization 
techniques are required in the MDBS. In this paper, several new 
techniques such as query sampling, query probing, and piggybacking are 
introduced. These techniques were designed for an MDBS called the CORDS 
Multidatabase System. The paper gives an overview of the global query 
optimizer in the system, including its architecture and some critical 
implementation techniques.


KEYWORDS:   multidatabase system, distributed query processing, global 
            query optimization, cost estimation, database system architecture