Performance analysis of a parallel distributive join algorithm on the Intel paragon

Author

Chung, Soon M. ; Chatterjee, Arindam

Author_Institution

Dept. of Comput. Sci., Wright State Univ., Dayton, OH, USA

fYear

1997

fDate

10-13 Dec 1997

Firstpage

714

Lastpage

721

Abstract

In this paper, we analyze the performance of the parallel distributive join algorithm that we proposed previously (1996). We implemented the algorithm on an Intel Paragon machine and analyzed the effect of the number of processors and the join selectivity on the performance of the algorithm. We also compared the performance of the distributed join (DJ) algorithm with that of the hybrid-hash (HH) join algorithm. Our results show that the DJ performs comparably with the HH over the entire range of number of processors used and different join selectivities. A big advantage of the parallel DJ algorithm over the RH join algorithm is that it can easily support non-equijoin operations. The results can also be used to estimate the performance of file I/O intensive applications to be implemented on the Intel Paragon machine

Keywords

file organisation; parallel algorithms; relational databases; software performance evaluation; Intel paragon; distributed join algorithm; file I/O intensive applications; hybrid-hash join algorithm; parallel distributive join algorithm; performance analysis; Algorithm design and analysis; Computer science; Delay; Hardware; Partitioning algorithms; Performance analysis; Poles and towers;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel and Distributed Systems, 1997. Proceedings., 1997 International Conference on

Conference_Location

Seoul

Print_ISBN

0-8186-8227-2

Type

conf

DOI

10.1109/ICPADS.1997.652621

Filename

652621