This paper presents numerical experiments with assorted versions of parallellu matrix decomposition algorithms (Gauss and Crout algorithm). The tests have been carried out on the hardware platform with four-core Skyl...
详细信息
This paper presents numerical experiments with assorted versions of parallellu matrix decomposition algorithms (Gauss and Crout algorithm). The tests have been carried out on the hardware platform with four-core Skylake processor featuring hyperthreading technology doubling virtually core number. parallelization algorithms have been implemented with the aid of classic POSIX threads library. Experiments have shown that basic 4-thread acceleration of all parallel implementations is almost equal to the number of threads/processors. Both algorithms are worth considering in real-world applications (Florida University collection). Gauss algorithm is a better performer, with respect to timing, in the case of matrices with lower density of nonzeros, as opposed to higher density matrices. The latter are processed more efficiently with the aid of Crout algorithm implementation.
CARLOS-3D is a three-dimensional general purpose computer code for calculating the scattering from complex three-dimensional bodies. The code uses the method of moments technique with Galerkin testing, to solve the St...
详细信息
CARLOS-3D is a three-dimensional general purpose computer code for calculating the scattering from complex three-dimensional bodies. The code uses the method of moments technique with Galerkin testing, to solve the Statton-Chu integral equations for the specified geometry. All of the surfaces describing the scatterer are replaced by equivalent electric and magnetic currents which are then determined via the solution of a matrix equation. The parallel implementation of this code is discussed together with the parallel in-core solver developed at Sandia National Laboratories. Results for the parallel performance of the code are presented. (C) 1997 Elsevier Science Ltd.
The independent set ordering algorithm is a heuristic algorithm based on finding maximal independent sets of vertices in the matrix adjacency graph, which is commonly used for parallel matrix factorization. Howeve...
详细信息
The independent set ordering algorithm is a heuristic algorithm based on finding maximal independent sets of vertices in the matrix adjacency graph, which is commonly used for parallel matrix factorization. However, Disadvantages appear when it is applied to large-scale sparse linear systems. In this paper, we propose an improved algorithm by finding an optimal size of independent set in each elimination step rather than find a maximal independent set, which is proved to be effective by both theoretical analysis and parallel implementation.
暂无评论