Multidimensional Data Partitioning for Parallel Data Processing in Large Data Warehouses

Parallel data processing techniques become more and more prevalent because the rapid growing sizes of both databases and data warehouses. Their related queries tremendously increase the complexity of data and query processing and slow down the query processing. Both data partitioning and load balancing are very critical issues in parallel data processing environment. Query processing for data cube in star schema involves a fact table joining with multiple dimension tables. And these star queries are often found in computing aggregate functions for the applications in large databases and data warehouses. In this paper, we will present a load balance multidimensional data partitioning approach for parallel star query processing in large databases and/or data warehouses.