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
Bench IP: Benchmarking Intelligence Processors

文献类型：期刊论文

作者Tao, Jin-Hua2,3,4; Du, Zi-Dong2,4,5; Guo, Qi2,4,5; Lan, Hui-Ying2,4; Zhang, Lei2,4; Zhou, Sheng-Yuan2,4; Xu, Ling-Jie6; Liu, Cong7; Liu, Hai-Feng8; Tang, Shan9
刊名JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY
出版日期2018
卷号33期号:1页码:1-23
关键词deep learning intelligence processor benchmark
ISSN号1000-9000
DOI10.1007/s11390-018-1805-8
英文摘要The increasing attention on deep learning has tremendously spurred the design of intelligence processing hardware. The variety of emerging intelligence processors requires standard benchmarks for fair comparison and system optimization (in both software and hardware). However, existing benchmarks are unsuitable for benchmarking intelligence processors due to their non-diversity and nonrepresentativeness. Also, the lack of a standard benchmarking methodology further exacerbates this problem. In this paper, we propose BenchIP, a benchmark suite and benchmarking methodology for intelligence processors. The benchmark suite in BenchIP consists of two sets of benchmarks: microbenchmarks and macrobenchmarks. The microbenchmarks consist of single-layer networks. They are mainly designed for bottleneck analysis and system optimization. The macrobenchmarks contain state-of-the-art industrial networks, so as to offer a realistic comparison of different platforms. We also propose a standard benchmarking methodology built upon an industrial software stack and evaluation metrics that comprehensively reflect various characteristics of the evaluated intelligence processors. BenchIP is utilized for evaluating various hardware platforms, including CPUs, GPUs, and accelerators. BenchIP will be open-sourced soon.
资助项目National Key Research and Development Program of China[2017YFB1003101] ; National Natural Science Foundation of China[61472396] ; National Natural Science Foundation of China[61432016] ; National Natural Science Foundation of China[61473275] ; National Natural Science Foundation of China[61522211] ; National Natural Science Foundation of China[61532016] ; National Natural Science Foundation of China[61521092] ; National Natural Science Foundation of China[61502446] ; National Natural Science Foundation of China[61672491] ; National Natural Science Foundation of China[61602441] ; National Natural Science Foundation of China[61602446] ; National Natural Science Foundation of China[61732002] ; National Natural Science Foundation of China[61702478] ; Beijing Science and Technology Projects[Z151100000915072] ; Science and Technology Service Network Initiative (STS) Projects of Chinese Academy of Sciences ; National Basic Research 973 Program of China[2015CB358800]
WOS研究方向Computer Science
语种英语
WOS记录号WOS:000423587100001
出版者SCIENCE PRESS
源URL[http://119.78.100.204/handle/2XEOYT63/5599]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Tao, Jin-Hua
作者单位1.Adv Micro Devices Inc, Sunnyvale, CA 94085 USA
2.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China
3.Univ Chinese Acad Sci, Sch Comp & Control Engn, Beijing 100049, Peoples R China
4.Chinese Acad Sci, Inst Comp Technol, Intelligent Processor Res Ctr, Beijing 100190, Peoples R China
5.Cambricon Ltd, Beijing 100190, Peoples R China
6.Alibaba Grp, Alibaba Infrastruct Serv, Hangzhou 311121, Zhejiang, Peoples R China
7.Iflytek Co Ltd, Hefei 230088, Anhui, Peoples R China
8.Beijing Jingdong Century Trading Co Ltd, Beijing 100176, Peoples R China
9.RDA Microelect Inc, Shanghai 201203, Peoples R China
推荐引用方式
GB/T 7714		 Tao, Jin-Hua,Du, Zi-Dong,Guo, Qi,et al. Bench IP: Benchmarking Intelligence Processors[J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,2018,33(1):1-23.
APA Tao, Jin-Hua.,Du, Zi-Dong.,Guo, Qi.,Lan, Hui-Ying.,Zhang, Lei.,...&Chen, Tian-Shi.(2018).Bench IP: Benchmarking Intelligence Processors.JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,33(1),1-23.
MLA Tao, Jin-Hua,et al."Bench IP: Benchmarking Intelligence Processors".JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 33.1(2018):1-23.

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。