Long Live TIME: Improving Lifetime and Security for NVM-Based Training-in-Memory Systems

doi:10.1109/TCAD.2020.2977079

	Long Live TIME: Improving Lifetime and Security for NVM-Based Training-in-Memory Systems
	Cai, Yi 1; Lin, Yujun 2; Xia, Lixue 3; Chen, Xiaoming 4; Han, Song 2; Wang, Yu 1; Yang, Huazhong 1
	2020-12-01
发表期刊	IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
ISSN	0278-0070
卷号	39 期号:12 页码:4707-4720
摘要	Nonvolatile memory (NVM)-based training-inmemory (TIME) systems have emerged that can process the neural network (NN) training in an energy-efficient manner. However, the endurance of NVM cells is disappointing, rendering concerns about the lifetime of TIME systems, because the weights of NN models always need to be updated for thousands to millions of times during training. Gradient sparsification (GS) can alleviate this problem by preserving only a small portion of the gradients to update the weights. However, conventional GS will introduce nonuniform writes on different cells across the whole NVM crossbars, which significantly reduces the excepted available lifetime. Moreover, an adversary can easily launch malicious training tasks to exactly wear-out the target cells and fast break down the system. In this article, we propose an efficient and effective framework, referred as SGS-ARS, to improve the lifetime and security of TIME systems. The framework mainly contains a structured GS (SGS) scheme for reducing the write frequency, and an aging-aware row swapping (ARS) scheme to make the writes uniform. Meanwhile, we show that the backpropagation mechanism allows the attacker to localize and update fixed memory locations and wear them out. Therefore, we introduce Random-ARS and Refresh techniques to thwart adversarial training attacks, preventing the systems from being fast broken in an extremely short time. Our experiments show that when TIME is programmed to train ResNet-50 on ImageNet dataset, 356x lifetime extension can be achieved without sacrificing the accuracy much or incurring much hardware overhead. Under the adversarial environment, the available lifetime of TIME systems can still be improved by 84x.
关键词	Gradient sparsification lifetime neural networks training-in-memory wear-leveling
DOI	10.1109/TCAD.2020.2977079
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2017YFA0207600] ; National Natural Science Foundation of China[61832007] ; National Natural Science Foundation of China[61622403] ; National Natural Science Foundation of China[61621091] ; Beijing National Research Center for Information Science and Technology ; Beijing Innovation Center for Future Chips ; Beijing Academy of Artificial Intelligence
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:000592111400032
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：8[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/2XEOYT63/16086
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Wang, Yu
作者单位	1.Tsinghua Univ, Dept Elect Engn, Beijing Natl Res Ctr Informat Sci & Technol, Beijing 100084, Peoples R China 2.Dept EECS, 77 Massachusetts Ave, Cambridge, MA 02139 USA 3.Alibaba Grp, Dept Cloud Intelligence, Beijing 100022, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Cai, Yi,Lin, Yujun,Xia, Lixue,et al. Long Live TIME: Improving Lifetime and Security for NVM-Based Training-in-Memory Systems[J]. IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,2020,39(12):4707-4720.
APA	Cai, Yi.,Lin, Yujun.,Xia, Lixue.,Chen, Xiaoming.,Han, Song.,...&Yang, Huazhong.(2020).Long Live TIME: Improving Lifetime and Security for NVM-Based Training-in-Memory Systems.IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,39(12),4707-4720.
MLA	Cai, Yi,et al."Long Live TIME: Improving Lifetime and Security for NVM-Based Training-in-Memory Systems".IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS 39.12(2020):4707-4720.