Load-balanced parallel banded-system solvers

Kuo Liang Chung; Wen Ming Yan; Jung Gen Wu

doi:10.1016/S0304-3975(01)00284-5

Load-balanced parallel banded-system solvers

Kuo Liang Chung^*, Wen Ming Yan, Jung Gen Wu

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract

Solving banded systems is important in the applications of science and engineering. This paper presents a load-balancing strategy for solving banded systems in parallel when the number of processors used is small. An optimization-based load-balancing analysis is given to determine how many loads should be assigned to each processor in order to minimize the time requirement. Some experimentations are carried out on the nCUBE 2E multiprocessor to demonstrate the speedup advantage of the proposed load-balancing strategy. The speedup improvement ratio ranges from 47% to 66% (from 12% to 24%) when using 4 (8) processors.

Original language	English
Pages (from-to)	313-334
Number of pages	22
Journal	Theoretical Computer Science
Volume	289
Issue number	1
DOIs	https://doi.org/10.1016/S0304-3975(01)00284-5
Publication status	Published - 2002 Oct 23
Externally published	Yes

Keywords

Banded systems
Load-balancing analysis
Parallel algorithms
nCUBE 2E multiprocessor

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1016/S0304-3975(01)00284-5

Cite this

@article{cc47501f14b245dabffc629e1d042a3b,

title = "Load-balanced parallel banded-system solvers",

abstract = "Solving banded systems is important in the applications of science and engineering. This paper presents a load-balancing strategy for solving banded systems in parallel when the number of processors used is small. An optimization-based load-balancing analysis is given to determine how many loads should be assigned to each processor in order to minimize the time requirement. Some experimentations are carried out on the nCUBE 2E multiprocessor to demonstrate the speedup advantage of the proposed load-balancing strategy. The speedup improvement ratio ranges from 47% to 66% (from 12% to 24%) when using 4 (8) processors.",

keywords = "Banded systems, Load-balancing analysis, Parallel algorithms, nCUBE 2E multiprocessor",

author = "Chung, {Kuo Liang} and Yan, {Wen Ming} and Wu, {Jung Gen}",

note = "Funding Information: E-mail addresses: klchung@cs.ntust.edu.tw (K.-L. Chung), ganboon@csie.ntu.edu.tw (W.-M. Yan), jgwu@ice.ntnu.edu.tw (J.-G. Wu). 1Supported by the National Science Council of R.O.C. under contract NSC88-2213-E011-005. 2Supported by the National Science Council of R.O.C. under contract NSC87-2119-M002-006. 3Supported by the National Science Council of R.O.C. under contract NSC85-2213-E003-003.",

year = "2002",

month = oct,

day = "23",

doi = "10.1016/S0304-3975(01)00284-5",

language = "English",

volume = "289",

pages = "313--334",

journal = "Theoretical Computer Science",

issn = "0304-3975",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - Load-balanced parallel banded-system solvers

AU - Chung, Kuo Liang

AU - Yan, Wen Ming

AU - Wu, Jung Gen

N1 - Funding Information: E-mail addresses: klchung@cs.ntust.edu.tw (K.-L. Chung), ganboon@csie.ntu.edu.tw (W.-M. Yan), jgwu@ice.ntnu.edu.tw (J.-G. Wu). 1Supported by the National Science Council of R.O.C. under contract NSC88-2213-E011-005. 2Supported by the National Science Council of R.O.C. under contract NSC87-2119-M002-006. 3Supported by the National Science Council of R.O.C. under contract NSC85-2213-E003-003.

PY - 2002/10/23

Y1 - 2002/10/23

N2 - Solving banded systems is important in the applications of science and engineering. This paper presents a load-balancing strategy for solving banded systems in parallel when the number of processors used is small. An optimization-based load-balancing analysis is given to determine how many loads should be assigned to each processor in order to minimize the time requirement. Some experimentations are carried out on the nCUBE 2E multiprocessor to demonstrate the speedup advantage of the proposed load-balancing strategy. The speedup improvement ratio ranges from 47% to 66% (from 12% to 24%) when using 4 (8) processors.

AB - Solving banded systems is important in the applications of science and engineering. This paper presents a load-balancing strategy for solving banded systems in parallel when the number of processors used is small. An optimization-based load-balancing analysis is given to determine how many loads should be assigned to each processor in order to minimize the time requirement. Some experimentations are carried out on the nCUBE 2E multiprocessor to demonstrate the speedup advantage of the proposed load-balancing strategy. The speedup improvement ratio ranges from 47% to 66% (from 12% to 24%) when using 4 (8) processors.

KW - Banded systems

KW - Load-balancing analysis

KW - Parallel algorithms

KW - nCUBE 2E multiprocessor

UR - http://www.scopus.com/inward/record.url?scp=0037163959&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037163959&partnerID=8YFLogxK

U2 - 10.1016/S0304-3975(01)00284-5

DO - 10.1016/S0304-3975(01)00284-5

M3 - Article

AN - SCOPUS:0037163959

SN - 0304-3975

VL - 289

SP - 313

EP - 334

JO - Theoretical Computer Science

JF - Theoretical Computer Science

IS - 1

ER -

Load-balanced parallel banded-system solvers

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this