DocumentCode :
1836969
Title :
Accelerating FDTD algorithm using GPU computing
Author :
Bo, Zhang ; Zheng-hui, Xue ; Wu, Ren ; Wei-ming, Li ; Xin-qing, Sheng
Author_Institution :
Sch. of Inf. & Electron., Beijing Inst. of Technol., Beijing, China
fYear :
2011
fDate :
22-25 May 2011
Firstpage :
410
Lastpage :
413
Abstract :
Hardware acceleration of Finite-Difference Time-Domain (FDTD) algorithm has always been an important part of FDTD research. In this essay, we discussed the advantage and feasibility of accelerate FDTD algorithm using Graphics Processing Unit (GPU). With the implement of lattice-threads mapping and other techniques, we proposed a GPU FDTD programming model that is both efficient and accurate. Then we simulated two projects using proposed model, the result shows good agreement with CPU computing while simulation speedup is 20 at minimum, thus proves the efficiency and great potential of GPU accelerating in FDTD research.
Keywords :
coprocessors; finite difference time-domain analysis; mathematics computing; FDTD algorithm; FDTD research; GPU computing; finite-difference time-domain algorithm; hardware acceleration; lattice-threads mapping; Acceleration; Computer architecture; Finite difference methods; Graphics processing unit; Instruction sets; Lattices; Time domain analysis; FDTD; GPU; Hardware Accelerating;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Microwave Technology & Computational Electromagnetics (ICMTCE), 2011 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-8556-7
Type :
conf
DOI :
10.1109/ICMTCE.2011.5915546
Filename :
5915546
Link To Document :
بازگشت