In this paper, a new asynchronous iterative sub-structuring method is presented. this method is based on a classical sub-structuring approach, but during the iterations of the algorithm, at the end of each iteration t...
详细信息
In this paper, a new asynchronous iterative sub-structuring method is presented. this method is based on a classical sub-structuring approach, but during the iterations of the algorithm, at the end of each iteration the synchronisation between the processors is here removed leading to totally asynchronous iterations. the mathematical proof of the convergence of this new asynchronous iterative sub-structuring method is first introduced, followed by an example of the parallel implementation. Numerical results performed on a three dimensional test case illustrate the robustness, performance and efficiency of the asynchronous version over its synchronous counterpart. (C) 2016 international Association for mathematics and Computers in Simulation (IMACS). Published by Elsevier B.V. All rights reserved.
Genetic clustering consists in performing the analysis genetic optimization results using a clustering technique to get approximations of central parts of attractor of a multimodal objective. this work presents how ou...
详细信息
ISBN:
(纸本)3540219463
Genetic clustering consists in performing the analysis genetic optimization results using a clustering technique to get approximations of central parts of attractor of a multimodal objective. this work presents how outputs of Hierarchical Genetic Strategy can be clustered with EM algorithm. the approach gives an opportunity of theoretical analysis aimed on evaluating of approximation accuracy. In considered case genetic clustering can be easily implemented in parallel.
In this paper an O(n) parallel algorithm is presented for fast unranking t-ary trees with n internal nodes in Zaks' representation. A sequential O(nt) algorithm is derived on the basis of dynamic programming parad...
详细信息
ISBN:
(纸本)3540219463
In this paper an O(n) parallel algorithm is presented for fast unranking t-ary trees with n internal nodes in Zaks' representation. A sequential O(nt) algorithm is derived on the basis of dynamic programming paradigm. In the parallel version of the algorithm processing is performed in a dedicated parallel architecture containing certain systolic and associative features. At first a coefficient table is created by systolic computations. then, n subsequent elements of a tree codeword is computed in O(1) time through associative search operations.
Effective exploitation of a parallel computer system is a result of cooperation between the communication and the computing parts of a parallel computer system, and the application. these elements are described by var...
详细信息
ISBN:
(纸本)3540219463
Effective exploitation of a parallel computer system is a result of cooperation between the communication and the computing parts of a parallel computer system, and the application. these elements are described by various parameters. It is not easy to grasp the connection between the values of particular parameters and the efficiency of parallel computations. In this paper we propose a new way of representing the relations between the parameters of a parallel computer system, and its performance. Results of simulations are presented and discussed.
Early suggested parallel "ring" algorithm for solving of the spatially one-dimensional initial-boundary-value problem (IBVP) for a parabolic equation using an explicit difference method is shortly described....
详细信息
ISBN:
(纸本)3540219463
Early suggested parallel "ring" algorithm for solving of the spatially one-dimensional initial-boundary-value problem (IBVP) for a parabolic equation using an explicit difference method is shortly described. Asymptotical behaviour of the communication complexity of this parallel algorithm is studied. Communication complexity is determined as a ratio between the number of interchanges and the number of arithmetical operations. It is proved that the coefficient of the communication complexity for spatially m-dimensional IBVP tends in general to 3/4.
In this paper we present novel approach to the parallel computing in the distributed environment. Described solution utilizes new communication algorithm, which adjusts communication pattern to the grid environment. P...
详细信息
ISBN:
(纸本)3540219463
In this paper we present novel approach to the parallel computing in the distributed environment. Described solution utilizes new communication algorithm, which adjusts communication pattern to the grid environment. Presented results show, that this approach reduces significantly communication latency and allows to speedup data transfer up to several times. In result, one can run parallel applications in the distributed environment.
We present the results of parallel 3D simulation of Czochralski crystal growth. the description of the program is given and its parallel performance is estimated. the code verification is made by comparing the compute...
详细信息
ISBN:
(纸本)3540219463
We present the results of parallel 3D simulation of Czochralski crystal growth. the description of the program is given and its parallel performance is estimated. the code verification is made by comparing the computed melt-crystal interface shape withthe experiment.
this paper is devoted to applications of evolutionary algorithms into optimal design of nonlinear structures and identification of holes. the parallel and the distributed evolutionary algorithms are considered. the op...
详细信息
ISBN:
(纸本)3540219463
this paper is devoted to applications of evolutionary algorithms into optimal design of nonlinear structures and identification of holes. the parallel and the distributed evolutionary algorithms are considered. the optimum criterion is to minimize the plastic strain areas and stress values or an identification functional. the fitness functions are computed using the finite element method or the coupled finite and boundary element method.
In this paper, we propose an implementation method with high throughput for a single-chip 4096 complex point FFT. In order to increase transform speed, a parallel FFT architecture has been used. there are eight parall...
详细信息
ISBN:
(纸本)078037889X
In this paper, we propose an implementation method with high throughput for a single-chip 4096 complex point FFT. In order to increase transform speed, a parallel FFT architecture has been used. there are eight parallel basic processing modules in the entire FFT chip, which can work at the same time independently. the proposed structure can compute 4096 complex point forward or inverse FFT in real time with up to 320MHZ sampling frequency. and will be applied widely in high-speed signal processing.
the vertex ranking problem is closely related to the problem of finding the elimination tree of minimum height for a given graph. this implies that the problem has applications in the parallel Cholesky factorization o...
详细信息
ISBN:
(纸本)3540219463
the vertex ranking problem is closely related to the problem of finding the elimination tree of minimum height for a given graph. this implies that the problem has applications in the parallel Cholesky factorization of matrices. We describe the connection between this model of graph coloring and the matrix factorization. We also present a polynomial time algorithm for finding edge ranking of complete bipartite graphs. We use it to design an O(m(2+d)) algorithm for edge ranking of graphs obtained by removing 0(log m) edges from a complete bipartite graph, where d is a fixed number. then we extend our results to complete k-partite graphs for any fixed k > 2. In this way we give a new class of matrix factorization instances that can be optimally solved in polynomial time.
暂无评论