HyCA: A Hybrid Computing Architecture for Fault-Tolerant Deep Learning

doi:10.1109/TCAD.2021.3124763

	HyCA: A Hybrid Computing Architecture for Fault-Tolerant Deep Learning
	Liu, Cheng 1; Chu, Cheng 1,2; Xu, Dawen 1,2; Wang, Ying 1; Wang, Qianlong 1,2; Li, Huawei 1; Li, Xiaowei 1; Cheng, Kwang-Ting 3
	2022-10-01
发表期刊	IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
ISSN	0278-0070
卷号	41 期号:10 页码:3400-3413
摘要	Hardware faults on the regular 2-D computing array of a typical deep learning accelerator (DLA) can lead to dramatic prediction accuracy loss. Prior redundancy design approaches typically have each homogeneous redundant processing element (PE) to mitigate faulty PEs for a limited region of the 2-D computing array rather than the entire computing array to avoid the excessive hardware overhead. However, they fail to recover the computing array when the number of faulty PEs in any region exceeds the number of redundant PEs in the same region. The mismatch problem deteriorates when the fault injection rate rises and the faults are unevenly distributed. To address the problem, we propose a hybrid computing architecture (HyCA) for fault-tolerant DLAs. It has a set of dot-production processing units (DPPUs) to recompute all the operations that are mapped to the faulty PEs despite the faulty PE locations. According to our experiments, HyCA shows significantly higher reliability, scalability, and performance with less chip area penalty when compared to the conventional redundancy approaches. Moreover, by taking advantage of the flexible recomputing, HyCA can also be utilized to scan the entire 2-D computing array and detect the faulty PEs effectively at runtime.
关键词	Circuit faults Computational modeling Deep learning Hardware Redundancy Neural networks Computer architecture Deep learning accelerator (DLA) fault detection fault tolerance hybrid computing architecture (HyCA)
DOI	10.1109/TCAD.2021.3124763
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2020YFB1600201] ; National Natural Science Foundation of China[62174162] ; National Natural Science Foundation of China[61902375] ; National Natural Science Foundation of China[61834006]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Interdisciplinary Applications ; Engineering, Electrical & Electronic
WOS记录号	WOS:000856129900022
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：11[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/19415
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Xu, Dawen
作者单位	1.Chinese Acad Sci, Inst Comp Technol, SKLCA, Beijing 100080, Peoples R China 2.Hefei Univ Technol, Sch Microelect, Hefei 230009, Peoples R China 3.Hong Kong Univ Sci & Technol, Dept Comp Sci & Engn, Hong Kong, Peoples R China
推荐引用方式 GB/T 7714	Liu, Cheng,Chu, Cheng,Xu, Dawen,et al. HyCA: A Hybrid Computing Architecture for Fault-Tolerant Deep Learning[J]. IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,2022,41(10):3400-3413.
APA	Liu, Cheng.,Chu, Cheng.,Xu, Dawen.,Wang, Ying.,Wang, Qianlong.,...&Cheng, Kwang-Ting.(2022).HyCA: A Hybrid Computing Architecture for Fault-Tolerant Deep Learning.IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,41(10),3400-3413.
MLA	Liu, Cheng,et al."HyCA: A Hybrid Computing Architecture for Fault-Tolerant Deep Learning".IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS 41.10(2022):3400-3413.