论文信息 - Data parallel programming and basic linear algebra subroutines

Data parallel programming and basic linear algebra subroutines

The Data Parallel programming model is conceptually simple and provides powerful programming primitives AS in shared memory models of computation. With an appropriate underlying architecture, primitives requiring global memory accesses do not require significantly longer execution times than primitives only requiring local accesses. In the data parallel programming model primitives are also available for expressing simple forms of local data interaction in relative coordinates, as for instance required in relaxation on multidimensional lattices.

S. Lennart Johnsson

[1] William Jalby,et al. Optimizing matrix operations on a parallel multiprocessor with a memory hierarchy , 1986 .

[2] Lennart Johnsson. HIGHLY CONCURRENT ALGORITHMS FOR SOLVING LINEAR SYSTEMS OF EQUATIONS , 1984 .

[3] Sartaj Sahni,et al. Parallel Matrix and Graph Algorithms , 1981, SIAM J. Comput..

[4] Geoffrey C. Fox,et al. Matrix algorithms on a hypercube I: Matrix multiplication , 1987, Parallel Comput..

[5] W. Daniel Hillis,et al. The connection machine , 1985 .

[6] Lennart Johnsson. Matrix Multiplication on Boolean Cubes using Generic Communication Primitives , 1989 .

[7] Charles L. Lawson,et al. Basic Linear Algebra Subprograms for Fortran Usage , 1979, TOMS.

[8] Arthur Wouk,et al. Parallel processing and medium-scale multiprocessors , 1989 .

[9] Lynn Elliot Cannon,et al. A cellular computer to implement the kalman filter algorithm , 1969 .

[10] Lennart Johnsson. VLSI algorithms for Doolittle's, Crout's, and Cholesky's methods , 1982 .

[11] Jack J. Dongarra,et al. An extended set of FORTRAN basic linear algebra subprograms , 1988, TOMS.

[12] S. Lennart Johnsson,et al. Algorithms for Matrix Transposition on Boolean n-Cube Configured Ensemble Architectures , 1988, ICPP.

[13] Lennart Johnsson. Computational Arrays for Band Matrix Equations , 1981 .