基于JNI和C+〖KG-*3〗+的Intel集成众核并行方法

doi:10.3969/j.issn.10062475.2018.04.007

计算机与现代化

基于JNI和C+〖KG-*3〗+的Intel集成众核并行方法

(1.重庆邮电大学生物信息学院，重庆400065; 2.重庆邮电大学软件工程学院，重庆400065)

出版日期:2018-04-28 发布日期:2018-05-02
作者简介:桑喆(1995)，男，河南新乡人，重庆邮电大学生物信息学院本科生，研究方向：高性能计算; 邓川(1994)，男，四川达州人，本科生，研究方向：生物信息学; 苟聪(1996)，男，本科生，研究方向：高性能服务器; 刘开兴(1997)，男，重庆酉阳人，重庆邮电大学软件工程学院本科生，研究方向：高性能计算; 白明泽(1982)，男(土家族)，重庆酉阳人，副教授，博士，研究方向：生物信息学。

A Parallelization Method with JNI and C+〖KG-*3〗+ for Many Integrated Core Application

(1. School of Bioinformation, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
2. School of Software Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China)

Online:2018-04-28 Published:2018-05-02

摘要/Abstract

摘要： 针对当前Intel集成众核协处理器（MIC）只能使用C/C+〖KG-*3〗+/Fortran编程语言进行并行计算，不能对已有的Java程序提供高性能计算支持的问题，提出基于Java Native Interface(JNI)技术和C+〖KG-*3〗+的MIC混合并行计算方法。该方法基于JNI设计Java代码与C+〖KG-*3〗+代码的数据交换机制，使MIC协处理器强大的浮点计算能力加速Java应用程序成为可能。通过实验测试分析基于MIC多线程并行的Java程序计算性能效果，结果表明该方法能有效利用MIC协处理器，对Java程序的计算性能提升显著。

关键词: 集成众核, 并行计算, Java native interface

Abstract: Currently, the Intel many integrated core (MIC) coprocessor can be used for parallel computing only through C/C+〖KG-*3〗+ and Fortran programming language, lacking the support for existed Java program. In this passage, we propose to exploit the powerful computing resources of MIC coprocessor to boost the Java application basing on the hybrid parallel computing strategy of Java native interface (JNI) technique and C+〖KG-*3〗+, the interface achieves the data exchange between C+〖KG-*3〗+ and Java program. We design an experiment to test and analyze MICbased Java multithreads parallel computing program. The results show the performance improvement of Java program brought by Phi coprocessor Java program.

Key words: many integrated core, parallel computing, Java native interface

中图分类号:

TP311

桑喆1，邓川1，苟聪1，刘开兴2，白明泽1. 基于JNI和C+〖KG-*3〗+的Intel集成众核并行方法[J]. 计算机与现代化, doi: 10.3969/j.issn.10062475.2018.04.007.

SANG Zhe1, DENG Chuan1, GOU Cong1, LIU Kaixing2, BAI Mingze1. A Parallelization Method with JNI and C+〖KG-*3〗+ for Many Integrated Core Application[J]. Computer and Modernization, doi: 10.3969/j.issn.10062475.2018.04.007.

参考文献

［1］〖JP3〗国际TOP500组织. 全球超级计算机排行榜TOP500［DB/OL］. https://www.top500.org/lists/2017/06/, 20170630. 
［2］〖JP3〗王涛. “天河二号”超级计算机［J］. 科学, 2013,65(4):52. 
［3］ White T. Hadoop: The Definitive Guide［M］. Yahoo Press, 2011.
［4］ Zaharia M, Chowdhury M, Franklin M J, et al. Spark: Cluster computing with working sets［C］// Proceedings of the 2nd USENIX Conference on Hot Topics in Cloud Computing. 2010: Article No. 10.
［5］ Zhu Jie, Li Juanjuan, Hardesty E, et al. GPUinHadoop: Enabling MapReduce across distributed heterogeneous platforms［C］// Proceedings of the 13th IEEE/ACIS International Conference on Computer and Information Science. 2014:321326.
［6］任俊伟,林东岱. JNI技术实现跨平台开发的研究［J］. 计算机应用研究, 2005,22(7):180184.
［7］ Fast T, Wall T, Chen Liang. Java Native Access (JNA)［EB/OL］. https://github.com/twall/jna, 20070509.
［8］ Okur S, Radoi C, Lin Yu. Hadoop+Aparapi: Making Heterogenous MapReduce Programming Easier［DB/OL］. http://www.semihokur.com/docs/okur2012hadoop_aparapi.pdf, 20120410.
［9］ Yan Yonghong, Grossman M, Sarkar V. JCUDA: A programmerfriendly interface for accelerating Java programs with CUDA［C］// Proceedings of the 15th International EuroPar Conference on Parallel Processing. 2009:887899.
［10］Chandra R, Dagum L, Kohr D, et al. Parallel Programming in OpenMP［M］. Morgan Kaufmann, 2001.
［11］〖JP3〗Chrysos G. Intel Xeon PhiTM coprocessor (codename knights corner)［C］// Proceedings of the 24th Hot Chips Symposium. 2012, doi: 10.1109/HOTCHIPS.2012.7476487. 
［12］Wang Endong, Zhang Qing, Shen Bo, et al. Highperformance Computing on the Intel Xeon PhiTM: How to Fully Exploit MIC Architectures［M］. Springer, 2014.
［13］Clarke L, Glendinning I, Hempel R. The MPI message passing interface standard［M］// Programming Environments for Massively Parallel Distributed Systems. Birkhauser, 1994:213218.
［14］〖JP3〗Reinders J. Intel Threading Building Blocks: Outfitting C+〖KG-*3〗+ for Multicore Processor Parallelism［M］. O’Reilly Media, 2007.
［15］Randall K. Cilk: Efficient Multithreaded Computing［R］. Massachusetts Institute of Technology, 1998.
［16］Stone J E, Gohara D, Shi Guochun. OpenCL: A parallel programming standard for heterogeneous computing systems［J］. Computing in Science & Engineering, 2010,12(3):6673.
［17］Dongarra J J. The Linpack benchmark: An explanation［C］// Proceedings of the 1st International Conference on Supercomputing. 1987:456474.
［18］Tian Xinmin, Chen Yenkuang, Girkar M, et al. Exploring the use of hyperthreading technology for multimedia applications with Intel OpenMP* compiler［C］// Proceedings of the 17th International Symposium on Parallel and Distributed Processing. 2003, doi:10.1109/IPDPS.2003.1213118.

[1]	王鑫, 吴俊辉, . 基于FPGA的分子动力学模拟交互控制系统[J]. 计算机与现代化, 2022, 0(09): 25-31.
[2]	项武铭，李雪巍. 基于CUDA的梯级泵站调度算法实现[J]. 计算机与现代化, 2018, 0(11): 60-.
[3]	赵博颖，肖鹏，张力. 基于Docker的MPI和OpenMP混合编程[J]. 计算机与现代化, 2018, 0(05): 60-.
[4]	邬可可,黄国伟,孔令晶. 一种灵活的椭圆曲线密码并行化方法[J]. 计算机与现代化, 2018, 0(02): 71-.
[5]	王龙，姚文明. 基于Spark的并行遗传算法在物流配送问题中的应用[J]. 计算机与现代化, 2018, 0(01): 19-22.
[6]	栗家棋，郑昊. 基于矢量地图数据的异构混合快速绘制框架[J]. 计算机与现代化, 2018, 0(01): 56-61.
[7]	孙立华1，胡牧1，孟庆强1，钱亚康1，王松2. 配网线损大数据高性能计算解决方案[J]. 计算机与现代化, 2016, 0(12): 42-46,50.
[8]	汪方良，施慧彬. 基于OpenCL的人脸识别与隐藏系统[J]. 计算机与现代化, 2016, 0(1): 16-19,58.
[9]	赵成龙1，施慧彬1，俞忻峰2. 基于OpenCL的双GPU基数排序算法[J]. 计算机与现代化, 2015, 0(1): 27-30.
[10]	贺维维，吴京，曾瑶源. 基于MPI+OpenMP的红外弱小目标检测并行计算[J]. 计算机与现代化, 2014, 0(7): 53-57.
[11]	王兴，王介君，孙宁，汪瑶. 基于GPU的交叉相关外推并行化算法[J]. 计算机与现代化, 2014, 0(2): 213-218.
[12]	刘晴和，杨云，贺兴亚，徐文春，周媛媛. 基于业务分割的并行式大数据迁移策略研究[J]. 计算机与现代化, 2014, 0(11): 86-89.
[13]	张唯唯，张玉洁. 基于GPU的并行报文分类方法[J]. 计算机与现代化, 2014, 0(11): 9-14.
[14]	邓世垠. 基于GPU的图像处理并行算法研究[J]. 计算机与现代化, 2013, 1(7): 142-145.
[15]	郝晓飞;谭跃生;王静宇. Hadoop平台上Apriori算法并行化研究与实现[J]. 计算机与现代化, 2013, 1(3): 1-4,8.

基于JNI和C+〖KG-*3〗+的Intel集成众核并行方法

A Parallelization Method with JNI and C+〖KG-*3〗+ for Many Integrated Core Application

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价