Parallelizing merge sort onto distributed memory parallel computers

Minsoo Jeon, Dong Seung Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Merge sort is useful in sorting a great number of data progressively, especially when they can be partitioned and easily collected to a few processors. Merge sort can be parallelized, however, conventional algorithms using distributed memory computers have poor performance due to the successive reduction of the number of participating processors by a half, up to one in the last merging stage. This paper presents load-balanced parallel merge sort where all processors do the merging throughout the computation. Data are evenly distributed to all processors, and every processor is forced to work in all merging phases. An analysis shows the upper bound of the speedup of the merge time as (P- 1)/log P where P is the number of processors. We have reached a speedup of 8.2 (upper bound is 10.5) on 32-processor Cray T3E in sorting of 4M 32-bit integers.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages25-34
Number of pages10
Volume2327 LNCS
DOIs
Publication statusPublished - 2002 Dec 1
Event4th International Symposium on High Performance Computing, ISHPC 2002 - Kansai Science City, Japan
Duration: 2002 May 152002 May 17

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2327 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other4th International Symposium on High Performance Computing, ISHPC 2002
CountryJapan
CityKansai Science City
Period02/5/1502/5/17

Fingerprint

Distributed Memory
Parallel Computers
Merging
Sort
Sorting
Data storage equipment
Speedup
Upper bound
Parallel algorithms
Integer

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Jeon, M., & Kim, D. S. (2002). Parallelizing merge sort onto distributed memory parallel computers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2327 LNCS, pp. 25-34). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2327 LNCS). https://doi.org/10.1007/3-540-47847-7_5

Parallelizing merge sort onto distributed memory parallel computers. / Jeon, Minsoo; Kim, Dong Seung.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2327 LNCS 2002. p. 25-34 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2327 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jeon, M & Kim, DS 2002, Parallelizing merge sort onto distributed memory parallel computers. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 2327 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2327 LNCS, pp. 25-34, 4th International Symposium on High Performance Computing, ISHPC 2002, Kansai Science City, Japan, 02/5/15. https://doi.org/10.1007/3-540-47847-7_5
Jeon M, Kim DS. Parallelizing merge sort onto distributed memory parallel computers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2327 LNCS. 2002. p. 25-34. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/3-540-47847-7_5
Jeon, Minsoo ; Kim, Dong Seung. / Parallelizing merge sort onto distributed memory parallel computers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2327 LNCS 2002. pp. 25-34 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{780e6adff5c7457086140eaaccddeab9,
title = "Parallelizing merge sort onto distributed memory parallel computers",
abstract = "Merge sort is useful in sorting a great number of data progressively, especially when they can be partitioned and easily collected to a few processors. Merge sort can be parallelized, however, conventional algorithms using distributed memory computers have poor performance due to the successive reduction of the number of participating processors by a half, up to one in the last merging stage. This paper presents load-balanced parallel merge sort where all processors do the merging throughout the computation. Data are evenly distributed to all processors, and every processor is forced to work in all merging phases. An analysis shows the upper bound of the speedup of the merge time as (P- 1)/log P where P is the number of processors. We have reached a speedup of 8.2 (upper bound is 10.5) on 32-processor Cray T3E in sorting of 4M 32-bit integers.",
author = "Minsoo Jeon and Kim, {Dong Seung}",
year = "2002",
month = "12",
day = "1",
doi = "10.1007/3-540-47847-7_5",
language = "English",
isbn = "354043674X",
volume = "2327 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "25--34",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Parallelizing merge sort onto distributed memory parallel computers

AU - Jeon, Minsoo

AU - Kim, Dong Seung

PY - 2002/12/1

Y1 - 2002/12/1

N2 - Merge sort is useful in sorting a great number of data progressively, especially when they can be partitioned and easily collected to a few processors. Merge sort can be parallelized, however, conventional algorithms using distributed memory computers have poor performance due to the successive reduction of the number of participating processors by a half, up to one in the last merging stage. This paper presents load-balanced parallel merge sort where all processors do the merging throughout the computation. Data are evenly distributed to all processors, and every processor is forced to work in all merging phases. An analysis shows the upper bound of the speedup of the merge time as (P- 1)/log P where P is the number of processors. We have reached a speedup of 8.2 (upper bound is 10.5) on 32-processor Cray T3E in sorting of 4M 32-bit integers.

AB - Merge sort is useful in sorting a great number of data progressively, especially when they can be partitioned and easily collected to a few processors. Merge sort can be parallelized, however, conventional algorithms using distributed memory computers have poor performance due to the successive reduction of the number of participating processors by a half, up to one in the last merging stage. This paper presents load-balanced parallel merge sort where all processors do the merging throughout the computation. Data are evenly distributed to all processors, and every processor is forced to work in all merging phases. An analysis shows the upper bound of the speedup of the merge time as (P- 1)/log P where P is the number of processors. We have reached a speedup of 8.2 (upper bound is 10.5) on 32-processor Cray T3E in sorting of 4M 32-bit integers.

UR - http://www.scopus.com/inward/record.url?scp=68749103290&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=68749103290&partnerID=8YFLogxK

U2 - 10.1007/3-540-47847-7_5

DO - 10.1007/3-540-47847-7_5

M3 - Conference contribution

AN - SCOPUS:68749103290

SN - 354043674X

SN - 9783540436744

VL - 2327 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 25

EP - 34

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -