Hardware-software co-exploration with racetrack memory based in-memory computing for CNN inference in embedded systems
文献类型:期刊论文
作者 | Choong, Benjamin Chen Ming2; Luo, Tao3; Liu, Cheng4; He, Bingsheng5; Zhang, Wei6; Zhou, Joey Tianyi1 |
刊名 | JOURNAL OF SYSTEMS ARCHITECTURE |
出版日期 | 2022-07-01 |
卷号 | 128页码:20 |
ISSN号 | 1383-7621 |
关键词 | Artificial intelligence Hardware-software co-design Deep learning Embedded systems Emerging memory |
DOI | 10.1016/j.sysarc.2022.102507 |
英文摘要 | Deep neural networks generate and process large volumes of data, posing challenges for low-resource embedded systems. In-memory computing has been demonstrated as an efficient computing infrastructure and shows promise for embedded AI applications. Among newly-researched memory technologies, racetrack memory is a non-volatile technology that allows high data density fabrication, making it a good fit for in memory computing. However, integrating in-memory arithmetic circuits with memory cells affects both the memory density and power efficiency. It remains challenging to build efficient in-memory arithmetic circuits on racetrack memory within area and energy constraints. To this end, we present an efficient in-memory convolutional neural network (CNN) accelerator optimized for use with racetrack memory. We design a series of fundamental arithmetic circuits as in-memory computing cells suited for multiply-and-accumulate operations. Moreover, we explore the design space of racetrack memory based systems and CNN model architectures, employing co-design to improve the efficiency and performance of performing CNN inference in racetrack memory while maintaining model accuracy. Our designed circuits and model-system co-optimization strategies achieve a small memory bank area with significant improvements in energy and performance for racetrack memory based embedded systems. |
资助项目 | Joey Tianyi Zhou's SERC Central Research Fund (Use-inspired Basic Research) ; Singapore Government's Research, Innovation and Enterprise 2020 Plan (Advanced Manufacturing and Engineering domain)[A18A1b0045] |
WOS研究方向 | Computer Science |
语种 | 英语 |
出版者 | ELSEVIER |
WOS记录号 | WOS:000802886800003 |
源URL | [http://119.78.100.204/handle/2XEOYT63/19592] |
专题 | 中国科学院计算技术研究所期刊论文_英文 |
通讯作者 | Luo, Tao |
作者单位 | 1.ASTAR, Ctr Frontier AI Res, 1 Fusionopolis Way,16-16 Connexis, Singapore 138632, Singapore 2.Natl Univ Singapore, Dept Elect & Comp Engn, 4 Engn Dr 3, Singapore 117583, Singapore 3.Agcy Sci Technol & Res, Inst High Performance Comp, 1 Fusionopolis Way,16-16 Connexis, Singapore 138632, Singapore 4.Chinese Acad Sci, Inst Comp Technol, 6 Kexueyuan South Rd, Beijing 100190, Peoples R China 5.Natl Univ Singapore, Sch Comp, COM1,13 Comp Dr, Singapore 117417, Singapore 6.Hong Kong Univ Sci & Technol, Kowloon, Clear Water Bay, Hong Kong, Peoples R China |
推荐引用方式 GB/T 7714 | Choong, Benjamin Chen Ming,Luo, Tao,Liu, Cheng,et al. Hardware-software co-exploration with racetrack memory based in-memory computing for CNN inference in embedded systems[J]. JOURNAL OF SYSTEMS ARCHITECTURE,2022,128:20. |
APA | Choong, Benjamin Chen Ming,Luo, Tao,Liu, Cheng,He, Bingsheng,Zhang, Wei,&Zhou, Joey Tianyi.(2022).Hardware-software co-exploration with racetrack memory based in-memory computing for CNN inference in embedded systems.JOURNAL OF SYSTEMS ARCHITECTURE,128,20. |
MLA | Choong, Benjamin Chen Ming,et al."Hardware-software co-exploration with racetrack memory based in-memory computing for CNN inference in embedded systems".JOURNAL OF SYSTEMS ARCHITECTURE 128(2022):20. |
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。