GEMMW: A portable level 3 BLAS Winograd variant of Strassen’s matrix- matrix multiply algorithm. The authors describe a set of routines called GEMMW for implementing a Winograd variant of Strasser’s matrix-matrix multiply algorithm. The aim is the multiplication of large, full, explicitly stored matrices. Both real and complex matrices are considered. Parallel environments with distributed memory are studied carefully. All matrix sizes are allowed and the source codes are available from LINALG/GEMMW. SHAR in NETLIB.
Keywords for this software
References in zbMATH (referenced in 6 articles , 1 standard article )
Showing results 1 to 6 of 6.
- Boyer, Brice; Pernet, Clément; Zhou, Wei: Memory efficient scheduling of Strassen-Winograd’s matrix multiplication algorithm (2009)
- Elmroth, Erik; Gustavson, Fred; Jonsson, Isak; Kågström, Bo: Recursive blocked algorithms and hybrid data structures for dense matrix library software (2004)
- Valsalam, Vinod; Skjellum, Anthony: A framework for high-performance matrix multiplication based on hierarchical abstractions, algorithms and optimized low-level kernels (2002)
- Kågström, Bo; Ling, Per; Van Loan, Charles: GEMM-based level 3 BLAS: High-performance model implementations and performance evaluation benchmark (1998)
- Toledo, Sivan: Locality of reference in LU decomposition with partial pivoting (1997)
- Douglas, Craig C.; Heroux, Michael; Slishman, Gordon; Smith, Roger M.: GEMMW: A portable level 3 BLAS Winograd variant of Strassen’s matrix- matrix multiply algorithm (1994)