熔盐堆堆芯流体力学计算的GPU并行方法研究

使用计算流体力学（Computational Fluid Dynamics,CFD）数值方法对熔盐堆堆芯的流动和热传导等相关物理问题进行模拟求解,需要大量的计算时间。利用图形处理器（Graphics Processing Unit,GPU）加速技术对开源CFD软件Code_Saturne进行二次开发,研究求解熔盐堆堆芯流场的GPU并行算法。采用Open ACC语言在GPU上实现了向量运算、矩阵向量相乘等基本线性代数运算,从而实现预处理共轭梯度法（Preconditioned Conjugate Gradients,PCG）的GPU并行算法,并使用该算法求解压力状态方程。模拟了方腔驱动流模型及带...

Full description

Saved in:
Bibliographic Details
Published in核技术 Vol. 40; no. 11; pp. 57 - 63
Main Author 胡传伟 鄂彦志 邹杨 徐洪杰
Format Journal Article
LanguageChinese
Published 中国科学院上海应用物理研究所嘉定园区 上海 201800 2017
中国科学院钍基熔盐核能系统研究中心 上海 201800
中国科学院大学 北京 100049%中国科学院上海应用物理研究所嘉定园区 上海 201800
Subjects
ACC
Open
共轭梯度法
熔盐堆
计算流体力学
通用图形计算技术
计算流体力学
CFD
熔盐堆
通用图形计算技术
Molten Salt Reactor
Conjugate gradient (CG)
OpenACC
General-purpose graphic processing units (GPGPUs)
共轭梯度法
Online AccessGet full text
ISSN0253-3219
DOI10.11889/j.0253-3219.2017.hjs.40.110601

Cover

More Information
Summary:使用计算流体力学（Computational Fluid Dynamics,CFD）数值方法对熔盐堆堆芯的流动和热传导等相关物理问题进行模拟求解,需要大量的计算时间。利用图形处理器（Graphics Processing Unit,GPU）加速技术对开源CFD软件Code_Saturne进行二次开发,研究求解熔盐堆堆芯流场的GPU并行算法。采用Open ACC语言在GPU上实现了向量运算、矩阵向量相乘等基本线性代数运算,从而实现预处理共轭梯度法（Preconditioned Conjugate Gradients,PCG）的GPU并行算法,并使用该算法求解压力状态方程。模拟了方腔驱动流模型及带下降段的熔盐堆堆芯模型的流场分布。结果表明,GPU加速后的软件与原版软件的结果一致,但计算时间更少,证明了GPU算法的正确性及有效的加速性。
Bibliography:Molten Salt Reactor, CFD, Conjugate gradient （CG）, General-purpose graphic processing units （GPGPUs）, OpenACC
31-1342/TL
Background： The simulation of fluid dynamics process for molten salt reactor proposes a large compute complexity, which requires high performance computer systems to enhance speed and efficiency. Purpose： This study aims to achieve graphics processing unit （GPU） parallelization of fluid dynamics process of molten salt reactor core. Methods： OpenACC directives were used as the main programming model to speed up the vector and matrix linear operation. And the preconditioned conjugate gradients for solving linear equations were implemented on the GPU. Finally, the parallel implementation and general optimization strategies to the OpenACC version of Code_Saturne were tested and validated on a simplified molten salt reactor. Results： From the result of the implementation of the GPU-parallel code, it is manifested that the empirical tuning of OpenACC accelerated code sections are valid for obtainin
ISSN:0253-3219
DOI:10.11889/j.0253-3219.2017.hjs.40.110601