Self-Adapting Linear Algebra Algorithms and Software
Top Cited Papers
- 27 June 2005
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Proceedings of the IEEE
- Vol. 93 (2) , 293-312
- https://doi.org/10.1109/jproc.2004.840848
Abstract
One of the main obstacles to the efficient solution of scientific problems is the problem of tuning software, both to the available architecture and to the user problem at hand. We describe approaches for obtaining tuned high-performance kernels and for automatically choosing suitable algorithms. Specifically, we describe the generation of dense and sparse Basic Linear Algebra Subprograms (BLAS) kernels, and the selection of linear solver algorithms. However, the ideas presented here extend beyond these areas, which can be considered proof of concept.Keywords
This publication has 44 references indexed in Scilit:
- A general parallel sparse-blocked matrix multiply for linear scaling SCF theoryComputer Physics Communications, 2000
- Cache miss equationsACM Transactions on Programming Languages and Systems, 1999
- Superscalar GEMM-based level 3 BLAS—The on-going evolution of a portable and high-performance libraryPublished by Springer Nature ,1998
- Recursive blocked data formats and BLAS’s for dense linear algebra algorithmsPublished by Springer Nature ,1998
- Sparse matrix multiplicationACM SIGPLAN Notices, 1996
- A parallel block implementation of Level-3 BLAS for MIMD vector processorsACM Transactions on Mathematical Software, 1994
- Algorithm 656: an extended set of basic linear algebra subprograms: model implementation and test programsACM Transactions on Mathematical Software, 1988
- Distribution of mathematical software via electronic mailCommunications of the ACM, 1987
- Two Fast Algorithms for Sparse Matrices: Multiplication and Permuted TranspositionACM Transactions on Mathematical Software, 1978
- Improving the efficiency of portable software for linear algebraACM SIGNUM Newsletter, 1973