Institute of Computing Technology, Chinese Academy IR
| DLPlib: A Library for Deep Learning Processor | |
| Lan, Hui-Ying1,2,3; Wu, Lin-Yang1,2,3; Zhang, Xiao1,2; Tao, Jin-Hua1,2; Chen, Xun-Yu1,2; Wang, Bing-Rui1,2,4; Wang, Yu-Qing1,2,4; Guo, Qi1,2; Chen, Yun-Ji1,2 | |
| 2017-03-01 | |
| 发表期刊 | JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY
 |
| ISSN | 1000-9000 |
| 卷号 | 32期号:2页码:286-296 |
| 摘要 | Recently, deep learning processors have become one of the most promising solutions of accelerating deep learning algorithms. Currently, the only method of programming the deep learning processors is through writing assembly instructions by bare hands, which costs a lot of programming efforts and causes very low efficiency. One solution is to integrate the deep learning processors as a new back-end into one prevalent high-level deep learning framework (e.g., TPU (tensor processing unit) is integrated into Tensorflow directly). However, this will obstruct other frameworks to profit from the programming interface. The alternative approach is to design a framework-independent low-level library for deep learning processors (e.g., the deep learning library for GPU, cuDNN). In this fashion, the library could be conveniently invoked in high-level programming frameworks and provides more generality. In order to allow more deep learning frameworks to gain benefits from this environment, we envision it as a low-level library which could be easily embedded into current high-level frameworks and provide high performance. Three major issues of designing such a library are discussed. The first one is the design of data structures. Data structures should be as few as possible while being able to support all possible operations. This will allow us to optimize the data structures easier without compromising the generality. The second one is the selection of operations, which should provide a rather wide range of operations to support various types of networks with high efficiency. The third is the design of the API, which should provide a flexible and user-friendly programming model and should be easy to be embedded into existing deep learning frameworks. Considering all the above issues, we propose DLPlib, a tensor-filter based library designed specific for deep learning processors. It contains two major data structures, tensor and filter, and a set of operators including basic neural network primitives and matrix/vector operations. It provides a descriptor-based API exposed as a C++ interface. The library achieves a speedup of 0.79x compared with the performance of hand-written assembly instructions. |
| 关键词 | deep learning processor API library DLPlib |
| DOI | 10.1007/s11390-017-1722-2 |
| 收录类别 | SCI |
| 语种 | 英语 |
| 资助项目 | National Natural Science Foundation of China[61432016] ; National Natural Science Foundation of China[61472396] ; 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 Basic Research 973 Program of China[2015CB358800] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDB02040009] |
| WOS研究方向 | Computer Science |
| WOS类目 | Computer Science, Hardware & Architecture ; Computer Science, Software Engineering |
| WOS记录号 | WOS:000397835500008 |
| 出版者 | SCIENCE PRESS |
| 引用统计 | |
| 文献类型 | 期刊论文 |
| 条目标识符 | http://119.78.100.204/handle/2XEOYT63/7330 |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Lan, Hui-Ying |
| 作者单位 | 1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Microprocessor Res Ctr, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 4.Univ Sci & Technol China, Dept Comp Sci, Hefei 230026, Peoples R China |
| 推荐引用方式 GB/T 7714 | Lan, Hui-Ying,Wu, Lin-Yang,Zhang, Xiao,et al. DLPlib: A Library for Deep Learning Processor[J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,2017,32(2):286-296. |
| APA | Lan, Hui-Ying.,Wu, Lin-Yang.,Zhang, Xiao.,Tao, Jin-Hua.,Chen, Xun-Yu.,...&Chen, Yun-Ji.(2017).DLPlib: A Library for Deep Learning Processor.JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,32(2),286-296. |
| MLA | Lan, Hui-Ying,et al."DLPlib: A Library for Deep Learning Processor".JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 32.2(2017):286-296. |
| 条目包含的文件 | 条目无相关文件。 | |||||
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。
修改评论