? 一种面向数据流架构的流水循环优化方法
Journal of Computer Science and Technology
Quick Search in JCST
 Advanced Search 
      Home | PrePrint | SiteMap | Contact Us | Help
 
Indexed by   SCIE, EI ...
Bimonthly    Since 1986
Journal of Computer Science and Technology 2018, Vol. 33 Issue (1) :116-130    DOI: 10.1007/s11390-017-1748-5
Computer Architecture and Systems << Previous Articles | Next Articles >>
一种面向数据流架构的流水循环优化方法
Xu Tan1,2, Student Member, CCF, Xiao-Chun Ye1,3, Member, CCF, Xiao-Wei Shen1,2, Yuan-Chao Xu1,4,*, Member, CCF, Da Wang1, Member, CCF, Lunkai Zhang5, Wen-Ming Li1, Member, CCF, Dong-Rui Fan1,2, Senior Member, CCF, Zhi-Min Tang1, Distinguished Member, CCF
1 State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences Beijing 100190, China;
2 School of Computer and Control Engineering, University of Chinese Academy of Sciences, Beijing 100049, China;
3 State Key Laboratory of Mathematical Engineering and Advanced Computing, Wuxi 214125, China;
4 College of Information Engineering, Capital Normal University, Beijing 100048, China;
5 Department of Computer Science, The University of Chicago, Chicago, IL 60637, U.S.A
A Pipelining Loop Optimization Method for Dataflow Architecture
Xu Tan1,2, Student Member, CCF, Xiao-Chun Ye1,3, Member, CCF, Xiao-Wei Shen1,2, Yuan-Chao Xu1,4,*, Member, CCF, Da Wang1, Member, CCF, Lunkai Zhang5, Wen-Ming Li1, Member, CCF, Dong-Rui Fan1,2, Senior Member, CCF, Zhi-Min Tang1, Distinguished Member, CCF
1 State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences Beijing 100190, China;
2 School of Computer and Control Engineering, University of Chinese Academy of Sciences, Beijing 100049, China;
3 State Key Laboratory of Mathematical Engineering and Advanced Computing, Wuxi 214125, China;
4 College of Information Engineering, Capital Normal University, Beijing 100048, China;
5 Department of Computer Science, The University of Chicago, Chicago, IL 60637, U.S.A

摘要
参考文献
相关文章
Download: [PDF 754KB]  
摘要 在未来计算场景下,能效将成为构建E级计算系统最大的障碍。数据流体系结构在处理科学应用时具备天然的能效优势,然而目前的数据流处理结构不能充分挖掘循环中的并行性。为了解决这一问题,我们提出了一种流水循环优化方法(PLO),它让不同迭代在处理阵列上同时流动,这种方法包含两种技术:硬件支持的硬迭代技术和指令支持的软迭代技术。在硬迭代执行模型中,片上循环控制器负责产生循环索引,简化了数据流图的复杂性同时为流水执行提供了良好的基础;在软迭代执行模型中,本文设计了循环指令来解决循环之间的依赖问题。通过这两种技术,处理阵列上同一时刻可被执行的指令数大大增加,使得浮点单元保持运转。模拟结果显示本文提出的方法达到的浮点效率比静态和动态执行模型分别高2.45倍和1.1倍,同时本文方法的硬件开销非常有限。
关键词数据流模型   控制流模型   循环优化   E级计算   科学应用     
Abstract: With the coming of exascale supercomputing era, power efficiency has become the most important obstacle to build an exascale system. Dataflow architecture has native advantage in achieving high power efficiency for scientific applications. However, the state-of-the-art dataflow architectures fail to exploit high parallelism for loop processing. To address this issue, we propose a pipelining loop optimization method (PLO), which makes iterations in loops flow in the processing element (PE) array of dataflow accelerator. This method consists of two techniques, architecture-assisted hardware iteration and instruction-assisted software iteration. In hardware iteration execution model, an on-chip loop controller is designed to generate loop indexes, reducing the complexity of computing kernel and laying a good foundation for pipelining execution. In software iteration execution model, additional loop instructions are presented to solve the iteration dependency problem. Via these two techniques, the average number of instructions ready to execute per cycle is increased to keep floating-point unit busy. Simulation results show that our proposed method outperforms static and dynamic loop execution model in floating-point efficiency by 2.45x and 1.1x on average, respectively, while the hardware cost of these two techniques is acceptable.
Keywordsdataflow model   control-flow model   loop optimization   exascale computing   scientific application     
Received 2016-09-04;
本文基金:

This work was supported by the National Key Research and Development Program of China under Grant No. 2016YFB0200501, the National Natural Science Foundation of China under Grant Nos. 61332009 and 61521092, the Open Project Program of State Key Laboratory of Mathematical Engineering and Advanced Computing under Grant No. 2016A04 and the Beijing Municipal Science and Technology Commission under Grant No. Z15010101009, the Open Project Program of State Key Laboratory of Computer Architecture under Grant No. CARCH201503, China Scholarship Council, and Beijing Advanced Innovation Center for Imaging Technology.

通讯作者: Yuan-Chao Xu     Email: xuyuanchao@cnu.edu.cn
About author: Xu Tan received his Bachelor's degree in computer science and technology from Capital Normal University, Beijing, in 2012. He is currently a Ph.D. candidate in Institute of Computing Technology, Chinese Academy of Sciences, Beijing. His main research interests include dataflow architecture and high-performance computer systems.
引用本文:   
Xu Tan, Xiao-Chun Ye, Xiao-Wei Shen, Yuan-Chao Xu, Da Wang, Lunkai Zhang, Wen-Mi.一种面向数据流架构的流水循环优化方法[J]  Journal of Computer Science and Technology , 2018,V33(1): 116-130
Xu Tan, Xiao-Chun Ye, Xiao-Wei Shen, Yuan-Chao Xu, Da Wang, Lunkai Zhang, Wen-Ming Li, Dong-Rui Fan, Zhi-Min Tang.A Pipelining Loop Optimization Method for Dataflow Architecture[J]  Journal of Computer Science and Technology, 2018,V33(1): 116-130
链接本文:  
http://jcst.ict.ac.cn:8080/jcst/CN/10.1007/s11390-017-1748-5
Copyright 2010 by Journal of Computer Science and Technology