Alternate Parallel Processing Approach for FEM

被引:30
作者
Fernandez, David M. [1 ]
Dehnavi, Maryam Mehri [1 ]
Gross, Warren J. [1 ]
Giannacopoulos, Dennis [1 ]
机构
[1] McGill Univ, Dept Elect & Comp Engn, Montreal, PQ H3A 2A7, Canada
关键词
Acceleration; finite element method; graphic processing units (GPU); multicore; parallel processing;
D O I
10.1109/TMAG.2011.2173304
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
In this work we present a new alternate way to formulate the finite element method (FEM) for parallel processing based on the solution of single mesh elements called FEM-SES. The key idea is to decouple the solution of a single element from that of the whole mesh, thus exposing parallelism at the element level. Individual element solutions are then superimposed node-wise using a weighted sum over concurrent nodes. A classic 2-D electrostatic problem is used to validate the proposed method obtaining accurate results. Results show that the number of iterations of the proposed FEM-SES method scale sublinearly with the number of unknowns. Two generations of CUDA enabled NVIDIA GPUs were used to implement the FEM-SES method and the execution times were compared to the classic FEM showing important performance benefits.
引用
收藏
页码:399 / 402
页数:4
相关论文
共 10 条
[1]  
[Anonymous], 2010, CUBL US GUID E MAN
[2]   A New Method to Solve 3-D Magnetodynamic Problems Without Assembling an Ax = b System [J].
Bastos, Joao Pedro A. ;
Sadowski, Nelson .
IEEE TRANSACTIONS ON MAGNETICS, 2010, 46 (08) :3365-3368
[3]   Enhancing the Performance of Conjugate Gradient Solvers on Graphic Processing Units [J].
Dehnavi, Maryam Mehri ;
Fernandez, David M. ;
Giannacopoulos, Dennis .
IEEE TRANSACTIONS ON MAGNETICS, 2011, 47 (05) :1162-1165
[4]   Multicore Acceleration of CG Algorithms Using Blocked-Pipeline-Matching Techniques [J].
Fernandez, David M. ;
Giannacopoulos, Dennis ;
Gross, Warren J. .
IEEE TRANSACTIONS ON MAGNETICS, 2010, 46 (08) :3057-3060
[5]  
Itoh T., 1996, FINITE ELEMENT SOFTW, P385
[6]  
Jin J-M, 2015, The finite element method in electromagnetics
[7]   A finite element domain decomposition combined with algebraic multigrid method for large-scale electromagnetic field computation [J].
Liu, YQ ;
Yuan, JS .
IEEE TRANSACTIONS ON MAGNETICS, 2006, 42 (04) :655-658
[8]  
NVIDIA Corporation, 2010, CUDA PROGR GUID CUDA
[9]   Full Wave Analyses of Electromagnetic Fields With an Iterative Domain Decomposition Method [J].
Takei, Amane ;
Sugimoto, Shin-Ichiro ;
Ogino, Masao ;
Yoshimura, Shinobu ;
Kanayama, Hiroshi .
IEEE TRANSACTIONS ON MAGNETICS, 2010, 46 (08) :2860-2863
[10]  
TOSELLI A., 2005, SPR S COMP, V34