Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 20
ADVANCES IN PARALLEL AND VECTOR PROCESSING FOR STRUCTURAL MECHANICS
Edited by: B.H.V. Topping and M. Papadrakakis
Paper IV.3

A Comparison of Problem Partitioning Algorithms for the Intel Paragon

B. Hendrickson and R. Leland

Sandia National Laboratories, Albuquerque, United States of America

Full Bibliographic Reference for this paper
B. Hendrickson, R. Leland, "A Comparison of Problem Partitioning Algorithms for the Intel Paragon", in B.H.V. Topping, M. Papadrakakis, (Editors), "Advances in Parallel and Vector Processing for Structural Mechanics", Civil-Comp Press, Edinburgh, UK, pp 133-137, 1994. doi:10.4203/ccp.20.4.3
Abstract
To solve finite element or finite difference problems on parallel computers the data structures must be decomposed and partitioned among the processors. Efficient execution requires that this decomposition balance the computational load while minimizing the interprocessor communication. We empirically compare a variety of algorithms for this problem using Chaco, a publically available partitioning tool written by the authors. Algorithms are evaluated in terms of their computational cost and the quality of the partition as judged by the execution time of the application on the Intel Paragon parallel super computer.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description