GPU Accelerated Finite-Element Computation for Electromagnetic Analysis

被引：19

作者：

Meng, Huan-Ting ^{[1
]}

Nie, Bao-Lin ^{[1
]}

Wong, Steven ^{[2
]}

Macon, Charles ^{[3
]}

Jin, Jian-Ming ^{[1
]}

机构：

[1] Univ Illinois, Dept Elect & Comp Engn, Ctr Computat Electromagnet, Urbana, IL 61801 USA

[2] Dynam Res Corp, Wright Patterson AFB, OH 45433 USA

[3] US Air Force Res Lab, Wright Patterson AFB, OH 45433 USA

来源：

IEEE ANTENNAS AND PROPAGATION MAGAZINE | 2014年 / 56卷 / 02期

关键词：

Computational electromagnetics; finite element analysis; frequency-domain analysis; high performance computing; graphics processing units; parallel programming; GALERKIN FEM COMPUTATIONS; FDTD METHOD; CONJUGATE GRADIENTS; CUDA; MOMENTS; IMPLEMENTATION; ALGORITHM; MODEL;

D O I：

10.1109/MAP.2014.6837065

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：

General-purpose computing on graphics processing units (GPGPU), with programming models such as the Compute Unified Device Architecture (CUDA) by NVIDIA, offers the capability for accelerating the solution process of computational electromagnetics analysis. However, due to the communication-intensive nature of the finite-element algorithm, both the assembly and the solution phases cannot be implemented via fine-grained many-core GPU processors in a straightforward manner. In this paper, we identify the bottlenecks in the GPU parallelization of the Finite-Element Method for electromagnetic analysis, and propose potential solutions to alleviate the bottlenecks. We first discuss efficient parallelization strategies for the finite-element matrix assembly on a single GPU and on multiple GPUs. We then explore parallelization strategies for the finite-element matrix solution, in conjunction with parallelizable preconditioners to reduce the total solution time. We show that with a proper parallelization and implementation, GPUs are able to achieve significant speedups over OpenMP-enabled multi-core CPUs.

引用

页码：39 / 62

页数：24

共 39 条

[1]

[Anonymous], CUDA C PROGR GUID

[2]

[Anonymous], 2003, ITERATIVE METHODS SP, DOI DOI 10.1137/1.9780898718003

[3] Assembly of finite element methods on graphics processors [J].

Cecka, Cris ;

Lew, Adrian J. ;

Darve, E. .

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 2011, 85 (05) :640-669

[4]

Cevahir A, 2009, LECT NOTES COMPUT SC, V5544, P893, DOI 10.1007/978-3-642-01970-8_90

[5] Sparse systems solving on GPUs with GMRES [J].

Couturier, Raphael ;

Domas, Stephane .

JOURNAL OF SUPERCOMPUTING, 2012, 59 (03) :1504-1516

[6] PARALLEL EFFICIENT METHOD OF MOMENTS EXPLOITING GRAPHICS PROCESSING UNITS [J].

De Donno, D. ;

Esposito, A. ;

Monti, G. ;

Tarricone, L. .

MICROWAVE AND OPTICAL TECHNOLOGY LETTERS, 2010, 52 (11) :2568-2572

[7] Introduction to GPU Computing and CUDA Programming: A Case Study on FDTD [J].

De Donno, Danilo ;

Esposito, Alessandra ;

Tarricone, Luciano ;

Catarinucci, Luca .

IEEE ANTENNAS AND PROPAGATION MAGAZINE, 2010, 52 (03) :116-122

[8]

Demir V, 2010, APPL COMPUT ELECTROM, V25, P323

[9]

Demir V, 2010, APPL COMPUT ELECTROM, V25, P303

[10]

Dosopoulos S., 2010, 2010 URSI International Symposium on Electromagnetic Theory (EMTS 2010), P989, DOI 10.1109/URSI-EMTS.2010.5637389

← 1 2 3 4 →