Numerical algorithms for the hypercube concurrent processor

Jean E. Patterson, Farzin Manshadi, Ruel H. Calalo, Paulett C. Liewer, William A. Imbriale, James Lyons

Research output: Contribution to journalArticle

Abstract

With the development of concurrent computing architectures which promise cost-effective means of obtaining supercomputing performance, there is much interest in applying and in evaluating the actual performance on large, computationally-intensive problems. Of particular interest is the concurrent performance of large scale electromagnetic scattering problems. Two electromagnetic codes with differing underlying algorithms have been converted to run on the Mark III Hypercube. One is a time domain finite difference solution of Maxwell's equations to solve for scattered fields and the other is a frequency domain moment method solution. Important measures for demonstrating the utility of the parallel architecture are the size of the problem that can be solved and the efficiency by which the paralleling can increase the speed of execution.

Original languageEnglish (US)
Pages (from-to)55-57
Number of pages3
JournalMathematical and Computer Modelling
Volume11
Issue numberC
DOIs
StatePublished - Jan 1 1988
Externally publishedYes

Fingerprint

Parallel architectures
Maxwell equations
Method of moments
Hypercube
Numerical Algorithms
Concurrent
Scattering
Frequency Domain Method
Costs
Moment Method
Electromagnetic Scattering
Finite-difference Time-domain (FDTD)
Parallel Architectures
Scattering Problems
Maxwell's equations
Computing

Keywords

  • Electromagnetic scattering
  • hypercubes
  • parallel algorithms
  • parallel processing

ASJC Scopus subject areas

  • Modeling and Simulation
  • Computer Science Applications

Cite this

Patterson, J. E., Manshadi, F., Calalo, R. H., Liewer, P. C., Imbriale, W. A., & Lyons, J. (1988). Numerical algorithms for the hypercube concurrent processor. Mathematical and Computer Modelling, 11(C), 55-57. https://doi.org/10.1016/0895-7177(88)90453-0

Numerical algorithms for the hypercube concurrent processor. / Patterson, Jean E.; Manshadi, Farzin; Calalo, Ruel H.; Liewer, Paulett C.; Imbriale, William A.; Lyons, James.

In: Mathematical and Computer Modelling, Vol. 11, No. C, 01.01.1988, p. 55-57.

Research output: Contribution to journalArticle

Patterson, JE, Manshadi, F, Calalo, RH, Liewer, PC, Imbriale, WA & Lyons, J 1988, 'Numerical algorithms for the hypercube concurrent processor', Mathematical and Computer Modelling, vol. 11, no. C, pp. 55-57. https://doi.org/10.1016/0895-7177(88)90453-0
Patterson, Jean E. ; Manshadi, Farzin ; Calalo, Ruel H. ; Liewer, Paulett C. ; Imbriale, William A. ; Lyons, James. / Numerical algorithms for the hypercube concurrent processor. In: Mathematical and Computer Modelling. 1988 ; Vol. 11, No. C. pp. 55-57.
@article{8dafe39d7bfc48d7ab8d7fb1c7cba3e1,
title = "Numerical algorithms for the hypercube concurrent processor",
abstract = "With the development of concurrent computing architectures which promise cost-effective means of obtaining supercomputing performance, there is much interest in applying and in evaluating the actual performance on large, computationally-intensive problems. Of particular interest is the concurrent performance of large scale electromagnetic scattering problems. Two electromagnetic codes with differing underlying algorithms have been converted to run on the Mark III Hypercube. One is a time domain finite difference solution of Maxwell's equations to solve for scattered fields and the other is a frequency domain moment method solution. Important measures for demonstrating the utility of the parallel architecture are the size of the problem that can be solved and the efficiency by which the paralleling can increase the speed of execution.",
keywords = "Electromagnetic scattering, hypercubes, parallel algorithms, parallel processing",
author = "Patterson, {Jean E.} and Farzin Manshadi and Calalo, {Ruel H.} and Liewer, {Paulett C.} and Imbriale, {William A.} and James Lyons",
year = "1988",
month = "1",
day = "1",
doi = "10.1016/0895-7177(88)90453-0",
language = "English (US)",
volume = "11",
pages = "55--57",
journal = "Mathematical and Computer Modelling",
issn = "0895-7177",
publisher = "Elsevier Limited",
number = "C",

}

TY - JOUR

T1 - Numerical algorithms for the hypercube concurrent processor

AU - Patterson, Jean E.

AU - Manshadi, Farzin

AU - Calalo, Ruel H.

AU - Liewer, Paulett C.

AU - Imbriale, William A.

AU - Lyons, James

PY - 1988/1/1

Y1 - 1988/1/1

N2 - With the development of concurrent computing architectures which promise cost-effective means of obtaining supercomputing performance, there is much interest in applying and in evaluating the actual performance on large, computationally-intensive problems. Of particular interest is the concurrent performance of large scale electromagnetic scattering problems. Two electromagnetic codes with differing underlying algorithms have been converted to run on the Mark III Hypercube. One is a time domain finite difference solution of Maxwell's equations to solve for scattered fields and the other is a frequency domain moment method solution. Important measures for demonstrating the utility of the parallel architecture are the size of the problem that can be solved and the efficiency by which the paralleling can increase the speed of execution.

AB - With the development of concurrent computing architectures which promise cost-effective means of obtaining supercomputing performance, there is much interest in applying and in evaluating the actual performance on large, computationally-intensive problems. Of particular interest is the concurrent performance of large scale electromagnetic scattering problems. Two electromagnetic codes with differing underlying algorithms have been converted to run on the Mark III Hypercube. One is a time domain finite difference solution of Maxwell's equations to solve for scattered fields and the other is a frequency domain moment method solution. Important measures for demonstrating the utility of the parallel architecture are the size of the problem that can be solved and the efficiency by which the paralleling can increase the speed of execution.

KW - Electromagnetic scattering

KW - hypercubes

KW - parallel algorithms

KW - parallel processing

UR - http://www.scopus.com/inward/record.url?scp=45549111663&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=45549111663&partnerID=8YFLogxK

U2 - 10.1016/0895-7177(88)90453-0

DO - 10.1016/0895-7177(88)90453-0

M3 - Article

AN - SCOPUS:45549111663

VL - 11

SP - 55

EP - 57

JO - Mathematical and Computer Modelling

JF - Mathematical and Computer Modelling

SN - 0895-7177

IS - C

ER -