PARALLEL PROCESSING OF OLAP QUERIES USING A CLUSTER OF WORKSTATIONS

Abstract

Online analytical processing (OLAP) queries normally incur enormous processing overheads due to the huge size of data warehouses. This results in unacceptable response times. Parallel processing using a cluster of workstations has of late emerged as a practical solution to many compute and data intensive problems. In this article, we present parallel algorithms for some of the OLAP operators. We have implemented these parallel solutions for a data warehouse implemented on Oracle hosted in a cluster of workstations. Our performance studies show that encouraging speedups are achieved.