中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Pragma Directed Shared Memory Centric Optimizations on GPUs

文献类型：期刊论文

作者Li, Jing1,2; Liu, Lei1; Wu, Yuan3; Liu, Xiang-Hua3; Gao, Yi3; Feng, Xiao-Bing1; Wu, Cheng-Yong1
刊名JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY
出版日期2016-03-01
卷号31期号:2页码:235-252
关键词GPU shared memory pragma directed data centric
ISSN号1000-9000
DOI10.1007/s11390-016-1624-8
英文摘要GPUs become a ubiquitous choice as coprocessors since they have excellent ability in concurrent processing. In GPU architecture, shared memory plays a very important role in system performance as it can largely improve bandwidth utilization and accelerate memory operations. However, even for affine GPU applications that contain regular access patterns, optimizing for shared memory is not an easy work. It often requires programmer expertise and nontrivial parameter selection. Improper shared memory usage might even underutilize GPU resource. Even using state-of-the-art high level programming models (e.g., OpenACC and OpenHMPP), it is still hard to utilize shared memory since they lack inherent support in describing shared memory optimization and selecting suitable parameters, let alone maintaining high resource utilization. Targeting higher productivity for affine applications, we propose a data centric way to shared memory optimization on GPU. We design a pragma extension on OpenACC so as to convey data management hints of programmers to compiler. Meanwhile, we devise a compiler framework to automatically select optimal parameters for shared arrays, using the polyhedral model. We further propose optimization techniques to expose higher memory and instruction level parallelism. The experimental results show that our shared memory centric approaches effectively improve the performance of five typical GPU applications across four widely used platforms by 3.7x on average, and do not burden programmers with lots of pragmas.
资助项目National High Technology Research and Development 863 Program of China[2012AA010902] ; National Natural Science Foundation of China (NSFC)[61432018] ; Innovation Research Group of NSFC[61221062]
WOS研究方向Computer Science
语种英语
WOS记录号WOS:000372154100002
出版者SCIENCE PRESS
源URL[http://119.78.100.204/handle/2XEOYT63/8637]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Li, Jing; Liu, Lei; Wu, Yuan; Liu, Xiang-Hua; Gao, Yi; Feng, Xiao-Bing; Wu, Cheng-Yong
作者单位1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
3.Beijing Samsung Telecom Res & Dev Ctr, Beijing 100028, Peoples R China
推荐引用方式
GB/T 7714		 Li, Jing,Liu, Lei,Wu, Yuan,et al. Pragma Directed Shared Memory Centric Optimizations on GPUs[J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,2016,31(2):235-252.
APA Li, Jing.,Liu, Lei.,Wu, Yuan.,Liu, Xiang-Hua.,Gao, Yi.,...&Wu, Cheng-Yong.(2016).Pragma Directed Shared Memory Centric Optimizations on GPUs.JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,31(2),235-252.
MLA Li, Jing,et al."Pragma Directed Shared Memory Centric Optimizations on GPUs".JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 31.2(2016):235-252.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。