Standard

Parallelization Methods for Implementation of Magnetic Induction Tomography Forward Models in Symmetric Multiprocessor Systems. / Watson, Stuart; Roula, Mohammed; Patz, Ralf; Maimaitijiang, Y.; Williams, R.J.; Griffiths, H.

In: Journal of Parallel Computing, Vol. 34, No. 9, 01.09.2008, p. 497 - 507.

Research output: Contribution to journalArticlepeer-review

Harvard

Watson, S, Roula, M, Patz, R, Maimaitijiang, Y, Williams, RJ & Griffiths, H 2008, 'Parallelization Methods for Implementation of Magnetic Induction Tomography Forward Models in Symmetric Multiprocessor Systems', Journal of Parallel Computing, vol. 34, no. 9, pp. 497 - 507. https://doi.org/10.1016/j.parco.2008.03.008

APA

Watson, S., Roula, M., Patz, R., Maimaitijiang, Y., Williams, R. J., & Griffiths, H. (2008). Parallelization Methods for Implementation of Magnetic Induction Tomography Forward Models in Symmetric Multiprocessor Systems. Journal of Parallel Computing, 34(9), 497 - 507. https://doi.org/10.1016/j.parco.2008.03.008

Vancouver

Watson S, Roula M, Patz R, Maimaitijiang Y, Williams RJ, Griffiths H. Parallelization Methods for Implementation of Magnetic Induction Tomography Forward Models in Symmetric Multiprocessor Systems. Journal of Parallel Computing. 2008 Sep 1;34(9):497 - 507. https://doi.org/10.1016/j.parco.2008.03.008

Author

Watson, Stuart ; Roula, Mohammed ; Patz, Ralf ; Maimaitijiang, Y. ; Williams, R.J. ; Griffiths, H. / Parallelization Methods for Implementation of Magnetic Induction Tomography Forward Models in Symmetric Multiprocessor Systems. In: Journal of Parallel Computing. 2008 ; Vol. 34, No. 9. pp. 497 - 507.

BibTeX

@article{329e3647aba44782964e2ecff0286c2d,
title = "Parallelization Methods for Implementation of Magnetic Induction Tomography Forward Models in Symmetric Multiprocessor Systems",
abstract = "This paper describes four parallelization approaches used in a finite-difference-based electromagnetic modeller for application in magnetic induction tomography (MIT) and suitable for implementation on computer systems with symmetric multiprocessor (SMP) architecture. The approaches include: (i) splitting by coils using a distributed memory approach, (ii) splitting by physical domain using a distributed memory approach, (iii) splitting by physical domain using hybrid distributed/shared memory approach and (iv) splitting by both coils and physical domain using multi-level distributed and shared memory approaches respectively. All four approaches were implemented and tested on an IBM SP supercomputer. Coil parallelization was the most efficient method due to low inter-processor communication requirements but was limited by the number of coils in the MIT system. Approaches (ii) and (iii) allowed a larger number of processors to be employed but the efficiency versus number of processors was found to drop at a faster rate in comparison to (i). The fourth approach both allowed a larger number of processors to be employed and was found to provide higher efficiency than the parallelization by physical domain only. This multi-level hybrid approach therefore appears to offer an effective parallelization method for implementation of the MIT forward model on SMP clusters.",
keywords = "Magnetic induction tomography, SMP cluster, finite-difference algorithm, Red–black SOR",
author = "Stuart Watson and Mohammed Roula and Ralf Patz and Y. Maimaitijiang and R.J. Williams and H. Griffiths",
year = "2008",
month = sep,
day = "1",
doi = "10.1016/j.parco.2008.03.008",
language = "English",
volume = "34",
pages = "497 -- 507",
journal = "Journal of Parallel Computing",
number = "9",

}

RIS

TY - JOUR

T1 - Parallelization Methods for Implementation of Magnetic Induction Tomography Forward Models in Symmetric Multiprocessor Systems

AU - Watson, Stuart

AU - Roula, Mohammed

AU - Patz, Ralf

AU - Maimaitijiang, Y.

AU - Williams, R.J.

AU - Griffiths, H.

PY - 2008/9/1

Y1 - 2008/9/1

N2 - This paper describes four parallelization approaches used in a finite-difference-based electromagnetic modeller for application in magnetic induction tomography (MIT) and suitable for implementation on computer systems with symmetric multiprocessor (SMP) architecture. The approaches include: (i) splitting by coils using a distributed memory approach, (ii) splitting by physical domain using a distributed memory approach, (iii) splitting by physical domain using hybrid distributed/shared memory approach and (iv) splitting by both coils and physical domain using multi-level distributed and shared memory approaches respectively. All four approaches were implemented and tested on an IBM SP supercomputer. Coil parallelization was the most efficient method due to low inter-processor communication requirements but was limited by the number of coils in the MIT system. Approaches (ii) and (iii) allowed a larger number of processors to be employed but the efficiency versus number of processors was found to drop at a faster rate in comparison to (i). The fourth approach both allowed a larger number of processors to be employed and was found to provide higher efficiency than the parallelization by physical domain only. This multi-level hybrid approach therefore appears to offer an effective parallelization method for implementation of the MIT forward model on SMP clusters.

AB - This paper describes four parallelization approaches used in a finite-difference-based electromagnetic modeller for application in magnetic induction tomography (MIT) and suitable for implementation on computer systems with symmetric multiprocessor (SMP) architecture. The approaches include: (i) splitting by coils using a distributed memory approach, (ii) splitting by physical domain using a distributed memory approach, (iii) splitting by physical domain using hybrid distributed/shared memory approach and (iv) splitting by both coils and physical domain using multi-level distributed and shared memory approaches respectively. All four approaches were implemented and tested on an IBM SP supercomputer. Coil parallelization was the most efficient method due to low inter-processor communication requirements but was limited by the number of coils in the MIT system. Approaches (ii) and (iii) allowed a larger number of processors to be employed but the efficiency versus number of processors was found to drop at a faster rate in comparison to (i). The fourth approach both allowed a larger number of processors to be employed and was found to provide higher efficiency than the parallelization by physical domain only. This multi-level hybrid approach therefore appears to offer an effective parallelization method for implementation of the MIT forward model on SMP clusters.

KW - Magnetic induction tomography

KW - SMP cluster

KW - finite-difference algorithm

KW - Red–black SOR

U2 - 10.1016/j.parco.2008.03.008

DO - 10.1016/j.parco.2008.03.008

M3 - Article

VL - 34

SP - 497

EP - 507

JO - Journal of Parallel Computing

JF - Journal of Parallel Computing

IS - 9

ER -

ID: 75228