Placement of Objects in Parallel Object-Based Systems

Parallelism is a viable solution to constructing high performance object-oriented database systems. This paper analyzes the role of parallelism in such systems. In parallel systems based on a shared-nothing architecture, the database is horizontally declustered across multiple processors, enabling the system to employ multiple processors to speedup the execution time of queries and improve throughput. The placement of objects across the processors has a significant impact on the performance of queries that traverse a few objects. This paper describes and evaluates a greedy algorithm for the placement of objects across the processors of a system. Moreover, it describes three alternative availability strategies that maintain backup copies of objects to enable the system to continue operation in the presence of disk failures. It describes how the backup copies can be used to enhance the performance of the system during normal mode of operation. This study quantifies the performance tradeo..