検索対象:     
報告書番号:
※ 半角英数字
 年 ~ 
 年

科学技術計算のためのGPU利用方法調査

Study on availability of GPU for scientific and engineering calculations

坂本 健作 ; 小林 清二*

Sakamoto, Kensaku; Kobayashi, Seiji*

近年、GPU(Graphic Processing Unit)を科学技術計算に用いる事例が増えている。GPUはCPU(Central Processing Units)と比べて、高い浮動小数点演算性能とメモリバンド幅を有していると言われている。そこで、開発環境CUDA(Compute Unified Device Architecture)を用いて、GPUによる科学技術計算の有用性を調査した。調査の結果、原子力研究でも用いられる行列,FFT(Fast Fourier Transform),流体力学等の計算で利用価値が高いことがわかった。一方、高性能な性能を引き出すためには高度なプログラミング技術が必要であり、また、科学技術計算に必要な倍精度演算性能やメモリ誤りの検出等の機能が不十分であることがわかった。

Recently, the number of scientific and engineering calculations is increasing. It is said that GPUs (Graphic Processing Units) have much higher peak floating-point processing power and memory bandwidth than CPUs (Central Processing Units). We have studied the effectiveness of GPUs by applying them to fundamental scientific and engineering calculations with CUDA (Compute Unified Device Architecture) development tools. The results have shown as follow; (1) GPGPU (General-Purpose computations on GPUs) is effective for such calculations as matrix operation, FFT (Fast Fourier Transform) and CFD (Computational Fluid Dynamics) in nuclear research region. (2) Highly-advanced programming is required for bringing out high performance of GPUs and double-precision. (3) ECC (Error Correction Code) in graphic memory systems supports are inadequate.

Access

:

- Accesses

InCites™

:

Altmetrics

:

[CLARIVATE ANALYTICS], [WEB OF SCIENCE], [HIGHLY CITED PAPER & CUP LOGO] and [HOT PAPER & FIRE LOGO] are trademarks of Clarivate Analytics, and/or its affiliated company or companies, and used herein by permission and/or license.