article · 2011
GPU-based 3D-FDTD computation for electromagnetic field dosimetry
Tomoaki Nagaoka, Soichi Watanabe
Cited by 7Abstract and details only
Abstract
Numerical dosimetry with the computational human model using the finite-difference time-domain (FDTD) method has recently been used for a safety assessment of electromagnetic field applications. However, the FDTD calculation runs very slowly and requires a large amount of computational memory. We focus, therefore, on general purpose programming on the graphics processing unit (GPGPU). We implemented the three-dimensional (3D) FDTD method on GPUs using Compute Unified Device Architecture (CUDA). In this study, we used the NVIDIA Tesla C2070 as a GPGPU board and tested the performance of FDTD computation on GPUs. The results indicated that while the single GPU/CPU speed ratio varies depending on the calculation domain, 3D-FDTD computation using a GPU requires significantly less run time than that using a conventional CPU. We confirmed that the FDTD computation on a multi-GPU is much faster than that on a single GPU, and we also found that eight GPUs can compute faster than a vector supercomputer.
Cite this paper
Nagaoka, T., & Watanabe, S. (2011). GPU-based 3D-FDTD computation for electromagnetic field dosimetry. https://doi.org/10.1109/afrcon.2011.6072180
Nagaoka, Tomoaki, and Soichi Watanabe. “GPU-based 3D-FDTD computation for electromagnetic field dosimetry.” 2011, pp. 1–6, https://doi.org/10.1109/afrcon.2011.6072180.
Nagaoka, Tomoaki, and Soichi Watanabe. 2011. “GPU-based 3D-FDTD computation for electromagnetic field dosimetry.” https://doi.org/10.1109/afrcon.2011.6072180.
Nagaoka, T. and Watanabe, S. (2011) ‘GPU-based 3D-FDTD computation for electromagnetic field dosimetry’. Available at: https://doi.org/10.1109/afrcon.2011.6072180.
@article{nagaoka2011based,
title = {{GPU-based 3D-FDTD computation for electromagnetic field dosimetry}},
author = {Nagaoka, Tomoaki and Watanabe, Soichi},
pages = {1--6},
year = {2011},
doi = {10.1109/afrcon.2011.6072180}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.
Related papers
Parallel connected-component labeling algorithm for GPGPU applications
In-Yong Jung, Chang‐Sung Jeong · International Symposium on Communications and Information Technologies · 2010
Cited by 13
Introductory on GPGPU Programming Technique
Jinnan Zhang · Computer Programming Skills & Maintenance · 2010
Cited by 0
Parallel Programming For High-Performance Computing on CUDA
Zhang Min-fang · Microcomputer applications · 2009
Cited by 2
Новітні архітектури відеоадаптерів. Технологія GPGPU. Частина 2
Sergiy D. Pogorilyy, Dmytro Vitel, O. A. Vereshchynsky · Реєстрація зберігання і обробка даних · 2013
Cited by 0Open access
CUDA를 이용한 GPGPU 시스템 성능 분석
조명진, 이홍석, 김선욱 · 대한전자공학회 학술대회 · 2009
Cited by 0
Moving object detection strategy for augmented-reality applications in a GPGPU by using CUDA
Daniel Berjón, Carlos Cuevas, Francisco Morán et al. · International Conference on Consumer Electronics · 2012
Cited by 2Open access
Implementation and optimization of high speed AES algorithm based on GPGPU and CUDA
Zhen Bao · Journal of the Graduate School of the Chinese Academy of Sciences · 2011
Cited by 2