熔盐堆堆芯流体力学计算的GPU并行方法研究

胡传伟; 鄂彦志; 邹杨; 徐洪杰

doi:10.11889/j.0253-3219.2017.hjs.40.110601

您当前的位置：

首页 >

文章列表页 >

熔盐堆堆芯流体力学计算的GPU并行方法研究

核能科学与工程 | 更新时间：2021-05-19

- 熔盐堆堆芯流体力学计算的GPU并行方法研究
- Research on GPU parallelization of fluid dynamics process of molten salt reactor core
- 核技术 2017年40卷第11期文章编号：110601
- 作者机构：
  
  1.中国科学院上海应用物理研究所嘉定园区上海 201800
  2.中国科学院大学北京 100049
  3.中国科学院钍基熔盐核能系统研究中心上海 201800
- 作者简介：
  
  [ "胡传伟, 男, 1991年出生, 2014年毕业于哈尔滨工程大学, 现为硕士研究生, 研究方向为反应堆热工水力数值计算", "HU Chuanwei, male, born in 1991, graduated from Harbin Engineering University in 2014, master student, focusing on thermal-hydraulics numerical calculation of reactor" ]
  邹杨, E-mail:zouyang@sinap.ac.cn ZOU Yang, E-mail:zouyang@sinap.ac.cn
- 基金信息：
  
  若干关键问题研究项目(QYZDY-SSW-JSC016);中国科学院战略性先导科技专项(XDA02001002)
- DOI：10.11889/j.0253-3219.2017.hjs.40.110601
  中图分类号： TL334
- 收稿日期：2017-03-16，
  
  修回日期：2017-06-19，
  
  纸质出版日期：2017-11-10
- 稿件说明：
移动端阅览
胡传伟, 鄂彦志, 邹杨, 等. 熔盐堆堆芯流体力学计算的GPU并行方法研究[J]. 核技术, 2017,40(11):110601

Chuanwei HU, Yanzhi E, Yang ZOU, et al. Research on GPU parallelization of fluid dynamics process of molten salt reactor core[J]. Nuclear techniques, 2017, 40(11): 110601
胡传伟, 鄂彦志, 邹杨, 等. 熔盐堆堆芯流体力学计算的GPU并行方法研究[J]. 核技术, 2017,40(11):110601 DOI： 10.11889/j.0253-3219.2017.hjs.40.110601.

Chuanwei HU, Yanzhi E, Yang ZOU, et al. Research on GPU parallelization of fluid dynamics process of molten salt reactor core[J]. Nuclear techniques, 2017, 40(11): 110601 DOI： 10.11889/j.0253-3219.2017.hjs.40.110601.

摘要

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OpenACC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。

Abstract

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 obtaining correct results

and enhancing performance and portability.

Conclusion

With OpenACC

we find that the instance of fluid dynamics process for molten salt reactor is given out using the GPU version of Code_Saturne and the performance of the GPU version of Code_Saturne can be enhanced compared with that of the CPU version.

关键词

Keywords

references

应智. 基于GPU的OpenFOAM并行加速研究[D]. 上海: 上海交通大学, 2012. http://cdmd.cnki.com.cn/Article/CDMD-10248-1012018545.htm

YING Zhi. Research on acceleration of open FOAM based on GPU[D]. Shanghai:Shanghai Jiaotong University, 2012

Wienke S, Springer P, Terboven C, et al. OpenACC:first experiences with real-world applications[C]. Berlin:Euro-Par 2012 Parallel Processing, 2012:859-870

R Wyrzykowski , J Dongarra , K Karczewski , . Parallel processing and applied mathematics , : Berlin Springer Berlin Heidelberg , 2008 . 311 - 321 .

Kraus J, Schlottke M, Adinetz A, et al . Accelerating a C++ CFD code with OpenACC[C]. 2014 First Workshop on Accelerator Programming Using Directives (WACCPD), Los Alamitos, USA, 2014:47-54

张朝晖 , 刘俊起 , 徐勤建 . GPU并行计算技术分析与应用 . 信息技术 , 2009 . 11 86 - 89 . DOI: 10.3969/j.issn.1674-2117.2009.04.050 http://doi.org/10.3969/j.issn.1674-2117.2009.04.050 .

Zhaohui ZHANG , Junqi LIU , Qinjian XU . Analysis and application of the GPU parallel computing technology . Information Technology , 2009 . 11 86 - 89 . DOI: 10.3969/j.issn.1674-2117.2009.04.050 http://doi.org/10.3969/j.issn.1674-2117.2009.04.050 .

王福军 . 计算流体动力学分析:CFD软件原理与应用 , : 北京清华大学出版社 , 2004 . 54 - 57 .

Fujun WANG . Computational fluid dynamic analysis:theory and application of CFD , : Beijing Tsinghua University Press , 2004 . 54 - 57 .

宋士雄. 熔盐冷却球床堆热工水力特性研究[D]. 上海: 中国科学院上海应用物理研究所, 2014. http://cdmd.cnki.com.cn/Article/CDMD-80014-1015033051.htm

SONG Shixiong. Thermal-hydraulics characteristics research of molten salt cooled pebble bed reactor[D]. Shanghai:Shanghai Institute of Applied Physics, Chinese Academy of Sciences, 2014

Anzt H, Baboulin M, Dongarra J, et al . Accelerating the conjugate gradient algorithm with GPUs in CFD simulations[C]. High Performance Computing for Computational Science, VECPAR, Porto, Portugal, 2016:35-43

Rodriguez M R, Philip B, Wang Z, et al . Block-relaxation methods for 3D constant-coefficient stencils on GPUs and multicore CPUs[EB/OL]. 2017-01-05. https://arxiv.org/pdf/1208.1975v2.pdf https://arxiv.org/pdf/1208.1975v2.pdf

张志能. 基于GPU的稀疏线性方程组求解及其应用[D]. 南昌: 南昌大学, 2013. http://cdmd.cnki.com.cn/Article/CDMD-10403-1013344055.htm

ZHANG Zhineng. Solution and application of sparse linear system based on GPU[D]. Nanchang:Nanchang University, 2013

马超 , 韦刚 , 裴颂文 , . GPU上稀疏矩阵与矢量乘积运算的一种改进 . 计算机系统应用 , 2010 . 19 ( 5 ): 116 - 120 . http://www.cnki.com.cn/Article/CJFDTOTAL-XTYY201005028.htm .

Chao MA , Gang WEI , Songwen PEI , . Improvement of sparse matrix-vector multiplication on GPU . Computer Systems & Applications , 2010 . 19 ( 5 ): 116 - 120 . http://www.cnki.com.cn/Article/CJFDTOTAL-XTYY201005028.htm .

Shear driven cavity flow[EB/OL]. 2012-10-27. http://code-saturne.org/cms/sites/default/files/file_attach/Tutorial/version-3.0/Shear-driven-cavity-flow.pdf http://code-saturne.org/cms/sites/default/files/file_attach/Tutorial/version-3.0/Shear-driven-cavity-flow.pdf

Simplified nuclear vessel[EB/OL]. 2013-01-05. http://code-saturne.org/cms/sites/default/files/file_attach/Tutorial/version-3.0/Simplified-nuclear-vessel.pdf http://code-saturne.org/cms/sites/default/files/file_attach/Tutorial/version-3.0/Simplified-nuclear-vessel.pdf .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

新型搅混翼对燃料棒流致振动导致的微动磨蚀影响研究

熔盐堆临界与⁷Li丰度变化的核安全分析

150 MWt熔盐堆（SM-MSR）反应性引入事故后果的不确定性及参数敏感性分析

双面冷却熔盐堆组件热工分析程序开发与验证

熔盐堆流体动力学模型降阶方法适用性分析