一种可扩展异构并行核外高阶矩量法

doi:10.3969/j.issn.1001-2400.2017.01.026

西安电子科技大学学报

一种可扩展异构并行核外高阶矩量法

左胜;陈岩;张玉;赵勋旺;林中朝

(西安电子科技大学天线与微波技术重点实验室，陕西西安 710071)

收稿日期:2015-12-10 出版日期:2017-02-20 发布日期:2017-04-01
作者简介:左胜(1992-)，男，西安电子科技大学硕士研究生，E-mail：zuosheng0503@163.com
基金资助:
国家高技术研究发展计划(863计划)资助项目(2012AA01A308)；国家自然科学基金资助项目(61301069, 61072019)；教育部新世纪优秀人才支持计划资助项目(NCET-13-0949)；陕西省青年科技新星资助项目(2013KJXX-67)；中央高校基本科研业务费重点资助项目(JY10000902002)

Study of the scalable heterogeneous parallel out-of-core higher order method of moments

ZUO Sheng;CHEN Yan;ZHANG Yu;ZHAO Xunwang;LIN Zhongchao

(Science and Technology on Antennas and Microwave Lab., Xidian Univ., Xi'an 710071, China)

Received:2015-12-10 Online:2017-02-20 Published:2017-04-01

摘要/Abstract

摘要：

为了实现异构并行矩量法可跨节点，研究了CPU/GPU与CPU/MIC分布式异构平台上矩量法并行编程模型．利用图形处理器通用编程标准CUDA中提供的CONTEXT技术及集成众核中环境变量的概念，提出了一种适用于CPU/GPU与CPU/MIC的通用异构并行编程模型，满足了异构并行矩量法可跨节点对静态负载均衡的要求．数值结果表明，基于该并行编程模型设计的异构并行矩量法程序，可获得理想的加速比并具有良好的可扩展性．

关键词: 矩量法, 异构平台, 并行编程模型, 可扩展性

Abstract:

To achieve the across-nodes technology of the heterogeneous parallel method of moments (MoM), a parallel MoM programming model is studied on CPU/GPU and CPU/MIC heterogeneous platforms. By utilizing the CONTEXT technology in the CUDA which contains the GPU common programming criterions, and the concept of environment variables in MIC, a general heterogeneous parallel programming model for the CPU/GPU and CPU/MIC is proposed, which meets the static load-balancing for the across-nodes heterogeneous parallel MoM. Numerical results show that the heterogeneous parallel MoM codes based on the proposed parallel programming model can obtain ideal speedup and good scalability.

Key words: method of moments, heterogeneous platforms, parallel programming model, scalability

左胜;陈岩;张玉;赵勋旺;林中朝. 一种可扩展异构并行核外高阶矩量法[J]. 西安电子科技大学学报, 2017, 44(1): 146-151.

ZUO Sheng;CHEN Yan;ZHANG Yu;ZHAO Xunwang;LIN Zhongchao. Study of the scalable heterogeneous parallel out-of-core higher order method of moments[J]. Journal of Xidian University, 2017, 44(1): 146-151.

参考文献

［1］张舒, 褚艳利. GPU高性能运算之CUDA［M］. 北京:中国水利水电出版社, 2009.
［2］ LEZAR E, DAVIDSON D B. GPU-based LU Decomposition for Large Method of Moments Problems［J］. Electronics Letters, 2010, 46(17): 1194-1196.
［3］ D'AMBROSIO K, PIRICH R. MoM Software for GPU Hardware［C］//2011 IEEE Long Island Systems, Applications and Technology Conference. Piscataway: IEEE Computer Society, 2011: 1-4.
［4］ MU X, ZHOU H X, CHEN K, et al. Higher Order Method of Moments with a Parallel Out-of-core LU Solver on GPU/CPU Platform ［J］. IEEE Transactions on Antennas and Propagation, 2014, 62(11): 5634-5646.
［5］王恩东. MIC高性能计算编程指南［M］. 北京: 中国水利水电出版社, 2012.
［6］ ZHANG G H, CHEN Y, ZHANG Y, et al. MIC Accelerated LU Decomposition for Method of Moments［C］//Proceedings of 2015 IEEE Antennas and Propagation Society International Symposium: 2015. Piscataway: IEEE, 2015: 756-757.
［7］ TOP 10 Sites for November 2015 ［EB/OL］. ［2015-12-10］. http://www.top500.org/lists/2015/11/.
［8］ ZHANG Y, SARKAR T K, DONORO D G, et al. Parallel Solution of Integral Equation-based EM Problems in the Frequency Domain ［M］. Hoboken: Wiley-IEEE Press, 2009.
［9］ CHEN Y, LIN Z C, ZHANG Y, et al. Parallel Out-of-core Higher-order Method of Moments Accelerated by Graphics Processing Units［C］//Proceedings of 2015 IEEE Antennas and Propagation Society International Symposium: 2015. Piscataway: IEEE, 2015: 1674-1675.
［10］林中朝, 陈岩, 张玉, 等. 国产CPU平台中并行高阶矩量法研究［J］. 西安电子科技大学学报, 2015, 42(3): 43-47.
LIN Zhongchao, CHEN Yan, ZHANG Yu, et al. Study of the Parallel Higher-order MoM on a Domestically-made CPU Platform［J］. Journal of Xidian University, 2015, 42(3): 43-47.
［11］奎因. MPI与OpenMP并行程序设计［M］. 北京: 清华大学出版社, 2004.
［12］桑德斯. GPU高性能编程CUDA实战［M］. 北京: 机械工业出版社, 2011.
［13］杰弗斯. Intel Xeon Phi协处理器高性能编程指南［M］. 北京: 人民邮电出版社, 2014.
［14］ WANG E, ZHANG Q, SHEN B, et al. High-performance Computing on the Intel^ Xeon Phi^TM: How to Fully Exploit MIC Architectures［M］. Berlin: Springer International Publishing, 2014: 167-188.

[1]	李艳艳;左胜;王永;张玉. 高阶矩量法中几何结构及波端口对称性分析[J]. 西安电子科技大学学报, 2016, 43(6): 34-38+146.
[2]	方小星;朱志宇;张明新;李阳. 雷达波对孔缝腔体内复杂传输线耦合影响[J]. 西安电子科技大学学报, 2016, 43(6): 135-140.
[3]	陈岩;张玉;王永;赵勋旺;林中朝. 大规模并行RWG矩量法矩阵填充优化[J]. 西安电子科技大学学报, 2016, 43(5): 46-51.
[4]	陈文锋;龚书喜;董海林;张鹏飞;赵博. 雷达目标宽带散射特性的GGO快速分析方法[J]. J4, 2016, 43(1): 60-65.
[5]	王永;林中朝;张玉;梁昌洪. 并行矩量法分析波导缝隙天线的辐射特性[J]. J4, 2015, 42(5): 75-79+132.
[6]	林中朝;陈岩;张玉;江树刚;赵勋旺;吕兆峰. 国产CPU平台中并行高阶矩量法研究[J]. J4, 2015, 42(3): 43-47.
[7]	吴君辉;梁昌洪;袁浩波;曹祥玉. 自适应交叉近似算法的核外计算方法[J]. J4, 2014, 41(5): 161-165.
[8]	吕政良;龚书喜;张鹏飞;陈文锋;赵博. 改进型IE-FFT结合物理光学的宽带分析方法[J]. J4, 2014, 41(5): 74-78.
[9]	吕政良;龚书喜;张鹏飞;赵博;王夫蔚. 快速分析线面结构天线宽带特性的扫频方法[J]. J4, 2014, 41(1): 87-91.
[10]	张连波;郭立新;苟雪银;王安琪. 三层粗糙面电磁散射的矩量法研究[J]. J4, 2013, 40(6): 147-154.
[11]	王文涛;王夫蔚;龚书喜;王兴. 采用隐式空间映射算法的阵列方向图综合[J]. J4, 2012, 39(2): 159-163+199.
[12]	党晓杰;袁浩波;王楠;梁昌洪. MOM-PO混合方法中一种遮挡消影的新算法[J]. J4, 2012, 39(1): 38-41.
[13]	贺之莉;黄锴;蒋晓博;梁昌洪. 一种稳定高效的NURBS建模MOM-PO法[J]. J4, 2012, 39(1): 56-61.
[14]	王兴;龚书喜;王文涛;张玉洁. 一种快速分析面面结构电磁辐射特性算法[J]. J4, 2011, 38(4): 106-111.
[15]	黄锴;贺之莉;张红伟;梁昌洪;苏航. NURBS建模电大平台附近天线受扰方向图的分析[J]. J4, 2011, 38(4): 124-128.

一种可扩展异构并行核外高阶矩量法

Study of the scalable heterogeneous parallel out-of-core higher order method of moments

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10