A Randomized Parallel Sorting Algorithm with an Experimental Study

Loading...
Thumbnail Image

Files

CS-TR-3669.ps (4.19 MB)
No. of downloads: 132
CS-TR-3669.pdf (329.56 KB)
No. of downloads: 701

Publication or External Link

Date

1998-10-15

Advisor

Citation

DRUM DOI

Abstract

Previous achemes for sorting on general-purpose parallel machines have had to choose betwen poor load balancing and irregular communication or multiple rounds of all-to-all personalized communication. In this paper, we introduce a novel variation on sample sort which uses only two rounds of regular all-to-all personalized communication in a scheme that yields very good load balancing with virtually no overhard. Moeover, unlike precious variations, our algorithm efficiently handles the presence of duplicate values without the overhead of tagging each element with a unique identifier. The algorithm was implemented in SPLIT-C and run on a variety of platforms, including the Thinking Machines CM-5, the IBM SP-2, and the Cray Research T3D. We ran our code useing widely different benchmarks to examine the dependence of our algorithm on the input distribution. Our experimental results illustrate the efficiency and scalability of our algorithm across different platforms. In fact, it seems to outperform all similar algorithms known to the authors on these platforms, and its performance is invariant over the set of input distributions unlike previous efficient algorithms. Our results also compare facorably with those reported for the simpler ranking problem posed by the NAS Integer Sorting (IS) Benchmark. (Also cross-referenced as UMIACS-TR-96-53)

Notes

Rights