Title :
Mediator join indices
Author :
Yan, Ling Ling ; Özsu, M. Tamer ; Liu, Ling
Author_Institution :
Dept. of Comput. Sci., Alberta Univ., Edmonton, Alta., Canada
Abstract :
A mediator join index (MJI) is proposed to speed up N-way inter-database joins by reducing the amount of data transfer during evaluation. A family of algorithms, the query scrubbing algorithms (QSA), are developed to maintain MJI and to evaluate queries using MJI. QSA algorithms use query scrubbing to cope with update and query anomalies related to materialized views in the mediator context. Compared with existing algorithms, QSA algorithms incur less overhead in handling the anomalies and makes MJI a promising technique for efficient mediator query processing
Keywords :
distributed algorithms; distributed databases; query processing; N-way inter-database joins; anomaly handling overhead; data transfer; efficient mediator query processing; materialized views; mediator join indices; query anomalies; query evaluation; query scrubbing algorithms; update anomalies; Database systems; Indexes; Indexing; Query processing;
Conference_Titel :
Research Issues in Data Engineering, 1997. Proceedings. Seventh International Workshop on
Conference_Location :
Birmingham
Print_ISBN :
0-8186-7849-6
DOI :
10.1109/RIDE.1997.583698
