"Wide or tall" and "sparse matrix dense matrix" multiplications

Authors:
Gary W. Howell
Affiliations:
North Carolina State University Raleigh, North Carolina
Venue:
Proceedings of the 19th High Performance Computing Symposia
Year:
2011

Citing 3
Cited 0

Improving the memory-system performance of sparse-matrix vector multiplication

IBM Journal of Research and Development
Optimizing the performance of sparse matrix-vector multiplication

Optimizing the performance of sparse matrix-vector multiplication
When cache blocking of sparse matrix vector multiply works and why

Applicable Algebra in Engineering, Communication and Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Sparse matrix dense matrix (SMDM) multiplications are useful in block Krylov or block Lanczos methods. SMDM computations are AU, and VA, multiplication of a large sparse m x n matrix A by a matrix V of k rows of length m or a matrix U of k columns of length n, k m, k n. In a block Lanczos or Krylov algorithm, matrix matrix multiplications with the "tall" U and "wide" V are also needed. This note relates some experience in efficient SMDM and "Wide or Tall" computations on multi-core architectures.