On Resource Scheduling of Multi-Join Queries in Parallel Database Systems

Today’s database management systems (DBMS) must handle two “obstacles” in order to satisfy the increasing performance demands of their applications: (1) the large volume of data and (2) the complexity of queries. Parallelism represents the most feasible and economical solution. However, although parallel database systems have been successfully built (both academically and commercially), the development of effective and efficient query processing strategies to exploit the full potential of such architectures remains an issue of concern. In particular, the class of multi-join queries have received much attention. This is so because join is the most expensive relational operation and many applications (like decision support systems, AI applications and graphical modeling) require access to multiple relations. Thus, any performance improvement that can be made in processing multijoin queries will be beneficial.