John L. Volakis¹, Kubilay Sertel¹, Erik Jørgensen², Rick W. Kindt¹

CMES-Computer Modeling in Engineering & Sciences, Vol.5, No.5, pp. 463-476, 2004, DOI:10.3970/cmes.2004.005.463

Abstract In this paper we address several topics relating to the development and implementation of volume integral and hybrid finite element methods for electromagnetic modeling. Comparisons of volume integral equation formulations with the finite element-boundary integral method are given in terms of accuracy and computing resources. We also discuss preconditioning and parallelization of the multilevel fast multipole method, and propose higher-order basis functions for curvilinear quadrilaterals and volumetric basis functions for curvilinear hexahedra. The latter have the desirable property of vanishing divergence within the element but non-zero curl. In addition, a new domain decomposition is introduced for solving array problems involving… More >

Raju R. Namburu¹, Eric R. Mark, Jerry A. Clarke

CMES-Computer Modeling in Engineering & Sciences, Vol.5, No.5, pp. 443-454, 2004, DOI:10.3970/cmes.2004.005.443

Abstract Computational electromagnetic (CEM) simulations of full-range military vehicles play a critical role in enhancing the survivability and target recognition of combat systems. Modeling of full-range military systems subjected to high frequencies may involve generating large-scale meshes, solving equations, visualization, and analysis of results in the range of billions of unknowns or grid points. Hence, the overall objective of this research is to develop and demonstrate a scalable CEM software environment to address accurate prediction of radar cross sections (RCS) for full- range armored vehicles with realistic material treatments and complex geometric configurations. A software environment consisting of scalable preprocessing, postprocessing,… More >

J. S. Hesthaven¹, T. Warburton²

CMES-Computer Modeling in Engineering & Sciences, Vol.5, No.5, pp. 395-408, 2004, DOI:10.3970/cmes.2004.005.395

Abstract We discuss the formulation, validation, and parallel performance of a high-order accurate method for the time-domain solution of the three-dimensional Maxwell's equations on general unstructured grids. Attention is paid to the development of a general discontinuous element/penalty approximation to Maxwell's equations and a locally divergence free form of this. We further discuss the motivation for using a nodal Lagrangian basis for the accurate and efficient representation of solutions and operators. The performance of the scheme is illustrated by solving benchmark problems as well as large scale scattering applications. More >

O. Hassan¹, K. Morgan, J. Jones, B. Larwood, N. P. Weatherill

CMES-Computer Modeling in Engineering & Sciences, Vol.5, No.5, pp. 383-394, 2004, DOI:10.3970/cmes.2004.005.383

Abstract A numerical procedure for the simulation of 3D problems involving the scattering of electromagnetic waves is presented. As practical problems of interest in this area often involve domains of complex geometrical shape, an unstructured mesh based method is adopted. The solution algorithm employs an explicit finite element procedure for the solution of Maxwell's curl equations in the time domain using unstructured tetrahedral meshes. A PML absorbing layer is added at the artificial far field boundary that is created by the truncation of the physical domain prior to the numerical solution. The complete solution procedure is parallelised and several large scale… More >

Wanil Byun¹, Seung Jo Kim²

CMES-Computer Modeling in Engineering & Sciences, Vol.86, No.6, pp. 551-576, 2012, DOI:10.3970/cmes.2012.086.551

Abstract A structural eigensolver for large-scale finite element analysis is developed. The algorithms and data structures implemented in this paper are well suited for a distributed memory environment. As an eigenvalue extracting algorithm, the well-known M orthogonal block Lanczos iteration incorporated with a parallel multifrontal solver (PMFS) was chosen. Basically, for the better performance of this algorithm in parallel computation, Lanczos vector allocation, mass matrix multiplication, and M inner product procedures were efficiently implemented. And the PMFS for a linear equation which is the most time-consuming part during Lanczos iterations was improved. The idea was to optimize network topologies of parallel… More >

Yixiong Wei¹, Qifu Wang^1,2, Yingjun Wang¹, Yunbao Huang¹

CMES-Computer Modeling in Engineering & Sciences, Vol.86, No.3, pp. 241-274, 2012, DOI:10.3970/cmes.2012.086.241

Abstract In this study, we propose a novel method to accelerate the process of elastodynamic analysis in 3D problems with BEM (boundary element method). With applying the DRBEM (dual reciprocity boundary element method) to form new integral equations for reducing complexity;the modified FMM (fast multipole method)is introduced to simplify the computation process and save storage space by avoiding intermediate coefficientmatrices. At the same time, FMM-DRBEM is reprogrammed in parallel byapplying GPU with CUDA (Compute Unified Device Architecture)for improving efficiency further.The main features in this paper are: ( 1 )with respect to defects of classical method for elastodynamic, modified FMM-DRBEM algorithm is… More >

Min Ki Kim¹, Seung Jo Kim²

CMES-Computer Modeling in Engineering & Sciences, Vol.84, No.4, pp. 297-332, 2012, DOI:10.3970/cmes.2012.084.297

Abstract Hybrid parallelization of multifrontal solution method and its parallel performances in a multicore distributed parallel computing architecture are represented in this paper. To utilize a state-of-the-art multicore computing architecture, parallelization of the multifrontal method for a symmetric multiprocessor machine is required. Multifrontal method is easier to parallelize than other direct solution methods because the solution procedure implies that the elimination of unknowns can be executed simultaneously. This paper focuses on the multithreaded parallelism and mixing distributed algorithm and multithreaded algorithm together in a unified software. To implement the hybrid parallelized algorithm in a distributed shared memory environment, two innovative ideas… More >

M. Koyuncu¹, F. Y. Ikikat¹, G. C. Icoz², B. Baranoglu³, A. Yazici²

CMES-Computer Modeling in Engineering & Sciences, Vol.84, No.1, pp. 13-26, 2012, DOI:10.3970/cmes.2012.084.013

Abstract A parallel dual reciprocity boundary element method solution to thermoelasticity and thermoviscoelasticity problems is proposed. The DR-BEM formulation is given in Fourier Transform Space where the Time Space solutions are obtained through inverse Fourier Transform. The parallellization of the code is achieved through solving each frequency at a distinct computational node. The implemented parallel code is tested on 64-core IBM blade servers and it is seen that a linear speed-up is achieved. More >

Y.F. Nie, Y.Q. Li, L. Wang

CMES-Computer Modeling in Engineering & Sciences, Vol.83, No.6, pp. 575-598, 2012, DOI:10.3970/cmes.2012.083.575

Abstract In this paper, we introduce the parallel Node-based Local FEM (NLFEM)-a novel version of FEM with natural parallel performances and adaptivity to various complex parallel environments. As a crucial part of 3D NLFEM, the Node-based Local Tetrahedron Mesh Generation (NLTMG) algorithm is presented, where guaranteeing the global conformity of local meshes is a strategy of ensuring the validity of the NLFEM and decoupling the NLFEM. In order to generate conforming local mesh, we search the neighbor nodes by the neighborhood of a corresponding multilevel bucket and then those nodes are Delaunay tetrahedralized by the TetGen. Accordingly, we prove that the… More >

G.A. Gravvanis¹, K.M. Giannoutakis²

CMES-Computer Modeling in Engineering & Sciences, Vol.71, No.4, pp. 305-330, 2011, DOI:10.3970/cmes.2011.071.305

Abstract In this paper we present parallel explicit approximate inverse matrix techniques for solving sparse linear systems on shared memory systems, which are derived using the finite element method for biharmonic equations in three space variables. Our approach for solving such equations is by considering the biharmonic equation as a coupled equation approach (pair of Poisson equation), using a FE approximation scheme, yielding an inner-outer iteration method. Additionally, parallel approximate inverse matrix algorithms are introduced for the efficient solution of sparse linear systems, based on an anti-diagonal computational approach that eliminates the data dependencies. Parallel explicit preconditioned conjugate gradient-type schemes in… More >