FlexPDA: A Flexible Programming Framework for Deep Learning Accelerators
文献类型:期刊论文
作者 | Liu, Lei3,4; Ma, Xiu3,4; Liu, Hua-Xiao3,4; Li, Guang-Li1,2; Liu, Lei2 |
刊名 | JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY |
出版日期 | 2022-10-01 |
卷号 | 37期号:5页码:1200-1220 |
ISSN号 | 1000-9000 |
关键词 | deep learning accelerator programming framework domain-specific language |
DOI | 10.1007/s11390-021-1406-9 |
英文摘要 | There are a wide variety of intelligence accelerators with promising performance and energy efficiency, deployed in a broad range of applications such as computer vision and speech recognition. However, programming productivity hinders the deployment of deep learning accelerators. The low-level library invoked in the high-level deep learning framework which supports the end-to-end execution with a given model, is designed to reduce the programming burden on the intelligence accelerators. Unfortunately, it is inflexible for developers to build a network model for every deep learning application, which probably brings unnecessary repetitive implementation. In this paper, a flexible and efficient programming framework for deep learning accelerators, FlexPDA, is proposed, which provides more optimization opportunities than the low-level library and realizes quick transplantation of applications to intelligence accelerators for fast upgrades. We evaluate FlexPDA by using 10 representative operators selected from deep learning algorithms and an end-to-end network. The experimental results validate the effectiveness of FlexPDA, which achieves an end-to-end performance improvement of 1.620x over the low-level library. |
资助项目 | National Key Research and Development Program of China[2017YFB1003103] ; Natural Science Research Foundation of Jilin Province of China[20190201193JC] ; Fundamental Research Funds for the Central Universities |
WOS研究方向 | Computer Science |
语种 | 英语 |
出版者 | SCIENCE PRESS |
WOS记录号 | WOS:000870734200013 |
源URL | [http://119.78.100.204/handle/2XEOYT63/19772] |
专题 | 中国科学院计算技术研究所期刊论文 |
通讯作者 | Liu, Hua-Xiao |
作者单位 | 1.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China 3.Jilin Univ, Minist Educ, Key Lab Symbol Computat & Knowledge Engn, Changchun 130012, Peoples R China 4.Jilin Univ, Coll Comp Sci & Technol, Changchun 130012, Peoples R China |
推荐引用方式 GB/T 7714 | Liu, Lei,Ma, Xiu,Liu, Hua-Xiao,et al. FlexPDA: A Flexible Programming Framework for Deep Learning Accelerators[J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,2022,37(5):1200-1220. |
APA | Liu, Lei,Ma, Xiu,Liu, Hua-Xiao,Li, Guang-Li,&Liu, Lei.(2022).FlexPDA: A Flexible Programming Framework for Deep Learning Accelerators.JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,37(5),1200-1220. |
MLA | Liu, Lei,et al."FlexPDA: A Flexible Programming Framework for Deep Learning Accelerators".JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 37.5(2022):1200-1220. |
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。