Parallel processing for multirelational queries

This paper presents parallel algorithms for three type of relational queries having special query structures such as chain, ring, and star. The main object of each algorithm is to reduce the number of disk fetches required to compute the join of multirelations in order to respond the query. For evaluating the performance of proposed algorithms, the paged memory environment is assumed. The performance of these algorithms has been compared with the existing nested loops method. Significant improvements in the performance is observed