the computing velocity and memory storage of Single PC are often limited in large-scale electromagnetic simulation by finite element method (FEM), parallelprocessing is an important means to overcome such problems. T...
详细信息
ISBN:
(纸本)9781467371063
the computing velocity and memory storage of Single PC are often limited in large-scale electromagnetic simulation by finite element method (FEM), parallelprocessing is an important means to overcome such problems. the domain decomposition method (DDM) which decomposes the domain by nodes dominating and suits for parallel computing was illustrated first in this paper;A 2D electrostatic model was built and decomposed by the DDM;And the FEM linear system of equations was solved by using parallel CG method on the distributed parallel system composed of 6 PCs, the effective speed up reaching 97.5% was satisfying. Especially for large-scale simulation which consists of more than millions of freedoms, the parallelprocessing reduces computing time and increases the computing velocity greatly, it's the base on which large-scale 3D electromagnetic parallel computing.
parallel computing is a simultaneous use of multiple compute resources, for example, processors to solve complex computational problems. It has been used in high-end computing areas such as pattern recognition, medica...
详细信息
ISBN:
(纸本)9781479982523
parallel computing is a simultaneous use of multiple compute resources, for example, processors to solve complex computational problems. It has been used in high-end computing areas such as pattern recognition, medical diagnosis, national defense, and web search engine. this paper focuses on the implementation of pattern classification technique, Support Vector Machine (SVM) using vector processor approach. We have carried out a performance analysis to benchmark the sequential SVM program against the Graphics processing Units (GPUs) optimization. the result shows that the parallelization of SVM training duration achieves a better performance than the sequential code speedups by 6.40.
A parallel algorithm cannot be evaluated apart from the architecture it is implemented on. So, we define a parallel system as the combination of a parallel algorithm and a parallel architecture. the paper is devoted t...
详细信息
ISBN:
(纸本)3540341412
A parallel algorithm cannot be evaluated apart from the architecture it is implemented on. So, we define a parallel system as the combination of a parallel algorithm and a parallel architecture. the paper is devoted to the extension of well-known isoefficiency scalability metrics to heterogeneous parallel systems. Based on this extension the scalability of SUMMA (Scalable Universal Matrix Multiplication Algorithm) on parallel architecture with homogeneous communication system supporting simultaneous point-to-point communications is evaluated. Two strategies of data distribution are considered (i) homogeneous - data are distributed between processors evenly;i:(ii) data are distributed between processors according to their performance. It, is shown that under some assumption both strategies ensure the same scalability of heterogeneous parallel system. this theoretical results corroborated with experiment.
parallel reducts in the information view and an algorithm withthe matrix of attribute significance in the information view are introduced in this paper. Furthermore, F -attribute significance in the information view ...
详细信息
the intention of this paper is to provide an overview of the IA-64 Explicitly parallel Instruction Computing (EPIC) architecture. this quick overview of EPIC computer architecture evolution is provided to highlight so...
详细信息
ISBN:
(纸本)3540437924
the intention of this paper is to provide an overview of the IA-64 Explicitly parallel Instruction Computing (EPIC) architecture. this quick overview of EPIC computer architecture evolution is provided to highlight some of the motivating factors for developing IA-64 architecture as well as showing the most important areas where the architecture has overcome traditional limitations in processor architecture. Before describing the important IA-64 architecture features I will outline the goals and strategy of IA-64 architecture.
A parallelization scheme, which drives processing in simulations of the Monte Carlo type, suitable in highly heterogeneous computer system of a general purpose, is proposed. the message passing is applied and the MPI ...
详细信息
ISBN:
(纸本)9783540681052
A parallelization scheme, which drives processing in simulations of the Monte Carlo type, suitable in highly heterogeneous computer system of a general purpose, is proposed. the message passing is applied and the MPI library is exploited. For testing, the 2D Ising model in a magnetic field is taken. the dependence of speedup on the number of parallel processes is studied, showing that the scheme works well in different parallel computer systems. the condition for the best speedup in these simulations is explained. the possibility of parallel use of any available computing power from the surrounding is also indicated.
Due to very complex coastlines and shallow depths along the west coast of Korea it should be appliedthat very fine boundary fitted meshes in modeling Also it is necessary that fine meshes should be applied to generat...
详细信息
ISBN:
(纸本)9789814287982
Due to very complex coastlines and shallow depths along the west coast of Korea it should be appliedthat very fine boundary fitted meshes in modeling Also it is necessary that fine meshes should be applied to generate nonlinear hydrodynamics with reasonable accuracy In this study a parallel Linux cluster system is designed and applied for the evaluation of computational efficiency and reliability of the Yellow Sea tidal hydrodynamics Computational efficiency reaches up to 7 times according to NPB bench-marking test when 8 nodes cluster are used Model results by pADCIRC model on reproduction of the Yellow Sea resemble well with previous studies Computed results show remarkable shallow tidal amplification along west coast of Korea Tidal residuals with coastal currents and eddies arc also well generated owing to Fine grid system which can resolve complex coasts effectively
In the paper we consider strongly NP-hard flow shop problem withthe criterion of minimization of the sum of job's finishing times. We present the parallel algorithm based on the scatter search method. Obtained re...
详细信息
ISBN:
(纸本)9783540681052
In the paper we consider strongly NP-hard flow shop problem withthe criterion of minimization of the sum of job's finishing times. We present the parallel algorithm based on the scatter search method. Obtained results are compared to the best known from the literature. Superlinear speedup has been observed in the parallel calculations.
parallel two-step W-methods (shortly PTSW-methods) use s linearly-implicit external stages which may be processed in parallel. We discuss convergence properties of these methods on singularly perturbed problems and gi...
详细信息
ISBN:
(纸本)3540437924
parallel two-step W-methods (shortly PTSW-methods) use s linearly-implicit external stages which may be processed in parallel. We discuss convergence properties of these methods on singularly perturbed problems and give estimates for the global error for non-constant stepsizes. Due to the high stage order of the method no order reduction occurs.
Image Registration is the key step of Image processing as it is the process to locate most accurate relative orientation among two or more images, captured at the same or different times by distinguishable or indistin...
详细信息
暂无评论