GPU Acceleration of Multilevel Solvers for Analysis of Microwave Components With Finite Element Method

被引：35

作者：

Dziekonski, Adam ^{[1
]}

Lamecki, Adam ^{[1
]}

Mrozowski, Michal ^{[1
]}

机构：

[1] Gdansk Univ Technol, Dept Elect Telecommun & Informat, Gdansk, Poland

来源：

IEEE MICROWAVE AND WIRELESS COMPONENTS LETTERS | 2011年 / 21卷 / 01期

关键词：

Finite elements; graphics processing unit (GPU); multilevel preconditioner; ALGORITHM; UNITS;

D O I：

10.1109/LMWC.2010.2089974

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

080906 [电磁信息功能材料与结构]; 082806 [农业信息与电气工程];

摘要：

The letter discusses a fast implementation of the conjugate gradient iterative method with E-field multilevel preconditioner applied to solving real symmetric and sparse systems obtained with vector finite element method. In order to accelerate computations, a graphics processing unit (GPU) was used and significant speed-up (2.61 fold) was achieved comparing to a central processing unit (CPU) based approach. These results indicate that performance of electromagnetic simulations can be significantly improved thereby enabling full wave optimization of microwave components in more manageable time.

引用

页码：1 / 3

页数：3

共 21 条

[1]

Adams S, 2007, PROCEEDINGS OF THE HPCMP USERS GROUP CONFERENCE 2007, P334

[2]

Amdahl G., 1967, Proceedings of AFIPS Spring Join Computer Conference, P483, DOI DOI 10.1145/1465482.1465560

[3]

[Anonymous], 2008, NVR2008004

[4]

[Anonymous], 2008, PROGR GUIDE2

[5]

Bai Z., 2000, TEMPLATES SOLUTION S

[6]

Carvalho R. F., 2010, P 14 BIENN IEEE C EL, P1

[7]

Comparison of hierarchical basis functions for efficient multilevel solvers [J].

Ingelstroem, P. ;

Hill, V. ;

Dyczij-Edlinger, R. .

IET SCIENCE MEASUREMENT & TECHNOLOGY, 2007, 1 (01) :48-52

[8]

KRAKIWSKY SE, 2004, IEEE MTT S INT MICR, P1033

[9]

Lezar E, 2009, ICEAA: 2009 INTERNATIONAL CONFERENCE ON ELECTROMAGNETICS IN ADVANCED APPLICATIONS, VOLS 1 AND 2, P380, DOI 10.1109/ICEAA.2009.5297413

[10]

NVIDIA, 2011, NEXT GEN CUDA ARCH

← 1 2 3 →