Parallelism in relational database management systems

In order to provide real-time responses to complex queries involving large volumes of data, it has become necessary to exploit parallelism in query processing. This paper addresses the issues and solutions relating to intraquery parallelism in a relational database management system (DBMS). We provide a broad framework for the study of the numerous issues that need to be addressed in supporting parallelism efficiently and flexibly. The alternatives for a parallel architecture system are discussed, followed by the focus on how a query can be parallelized and how that affects load balancing of the different tasks created. The final part of the paper contains information about how the IBM DATABASE 2™ (DB2®) Version 3 product provides support for I/O parallelism to reduce response time for data-intensive queries.