Abstract
Many mathematical models of physical phenomena lead to solving dense systems of linear equations. As the models are refined, the order of these problems increases, usually beyond the capacity of the computer to contain the problem in central memory. This paper reviews block Gaussian elimination, which can be used to solve these problems efficiently. An implementation that achieves the maximum sustainable computational rate on a wide range of computers is given. The question of how large of a problem is currently feasible is addressed.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Dongarra, J. J., Bunch, J. R., Moler, C. B., and Stewart, G. W. 1979. LINPACK Users' Guide. SIAM.
Dongarra, J. J., DuCroz, J. J., Hammarling, S. J., and Hanson, R. J. 1987. An extended set of FORTRAN basic linear algebra subprograms. To appear in Trans. Math. Software.
Stewart, G. W. 1973. Introduction to Mathematical Computations. Academic Press, New York.
Vector Pak Subroutine Library Users' Manual. 1987. Boeing Computer Services, Seattle, Washington.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Grimes, R.G. Solving systems of large dense linear equations. J Supercomput 1, 291–299 (1988). https://doi.org/10.1007/BF00154340
Issue date:
DOI: https://doi.org/10.1007/BF00154340