Scheduling GEN_BLOCK array redistribution

H. G. Yook, Myong Soon Park

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

This article is devoted to the redistribution of arrays that are distributed in a GEN_BLOCK fashion over a processor grid. While GEN_BLOCK redistribution is essential for load balancing, prior research about redistribution has been focused on block-cyclic redistribution. The proposed scheduling algorithm exploits a spatial locality in message passing from a seemingly irregular array redistribution. The algorithm attempts to obtain near optimal scheduling by trying to minimize communication step size and the number of steps. According to experiments on CRAY T3E and IBM SP2, the algorithm shows good performance in typical distributed memory machines.

Original languageEnglish
Pages (from-to)251-267
Number of pages17
JournalJournal of Supercomputing
Volume22
Issue number3
DOIs
Publication statusPublished - 2002 Jul 1

Fingerprint

Redistribution
Scheduling
Message passing
Scheduling algorithms
Resource allocation
Data storage equipment
Communication
Optimal Scheduling
Distributed Memory
Message Passing
Load Balancing
Scheduling Algorithm
Locality
Experiments
Irregular
Grid
Minimise
Experiment

Keywords

  • Array redistribution
  • GEN_BLOCK
  • HPF
  • Parallelizing compiler

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Hardware and Architecture
  • Theoretical Computer Science
  • Computational Theory and Mathematics

Cite this

Scheduling GEN_BLOCK array redistribution. / Yook, H. G.; Park, Myong Soon.

In: Journal of Supercomputing, Vol. 22, No. 3, 01.07.2002, p. 251-267.

Research output: Contribution to journalArticle

Yook, H. G. ; Park, Myong Soon. / Scheduling GEN_BLOCK array redistribution. In: Journal of Supercomputing. 2002 ; Vol. 22, No. 3. pp. 251-267.
@article{16b820e3d7cc4c8e8ad6e0ab0ccb628b,
title = "Scheduling GEN_BLOCK array redistribution",
abstract = "This article is devoted to the redistribution of arrays that are distributed in a GEN_BLOCK fashion over a processor grid. While GEN_BLOCK redistribution is essential for load balancing, prior research about redistribution has been focused on block-cyclic redistribution. The proposed scheduling algorithm exploits a spatial locality in message passing from a seemingly irregular array redistribution. The algorithm attempts to obtain near optimal scheduling by trying to minimize communication step size and the number of steps. According to experiments on CRAY T3E and IBM SP2, the algorithm shows good performance in typical distributed memory machines.",
keywords = "Array redistribution, GEN_BLOCK, HPF, Parallelizing compiler",
author = "Yook, {H. G.} and Park, {Myong Soon}",
year = "2002",
month = "7",
day = "1",
doi = "10.1023/A:1015313917431",
language = "English",
volume = "22",
pages = "251--267",
journal = "The Journal of Supercomputing",
issn = "0920-8542",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Scheduling GEN_BLOCK array redistribution

AU - Yook, H. G.

AU - Park, Myong Soon

PY - 2002/7/1

Y1 - 2002/7/1

N2 - This article is devoted to the redistribution of arrays that are distributed in a GEN_BLOCK fashion over a processor grid. While GEN_BLOCK redistribution is essential for load balancing, prior research about redistribution has been focused on block-cyclic redistribution. The proposed scheduling algorithm exploits a spatial locality in message passing from a seemingly irregular array redistribution. The algorithm attempts to obtain near optimal scheduling by trying to minimize communication step size and the number of steps. According to experiments on CRAY T3E and IBM SP2, the algorithm shows good performance in typical distributed memory machines.

AB - This article is devoted to the redistribution of arrays that are distributed in a GEN_BLOCK fashion over a processor grid. While GEN_BLOCK redistribution is essential for load balancing, prior research about redistribution has been focused on block-cyclic redistribution. The proposed scheduling algorithm exploits a spatial locality in message passing from a seemingly irregular array redistribution. The algorithm attempts to obtain near optimal scheduling by trying to minimize communication step size and the number of steps. According to experiments on CRAY T3E and IBM SP2, the algorithm shows good performance in typical distributed memory machines.

KW - Array redistribution

KW - GEN_BLOCK

KW - HPF

KW - Parallelizing compiler

UR - http://www.scopus.com/inward/record.url?scp=0036642231&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036642231&partnerID=8YFLogxK

U2 - 10.1023/A:1015313917431

DO - 10.1023/A:1015313917431

M3 - Article

AN - SCOPUS:0036642231

VL - 22

SP - 251

EP - 267

JO - The Journal of Supercomputing

JF - The Journal of Supercomputing

SN - 0920-8542

IS - 3

ER -