Degradation analysis and optimization of temperature effect on MEMRISTOR-based Neural Network Accelerators by electro-thermal simulation

doi:10.1088/1742-6596/1812/1/012025

	Degradation analysis and optimization of temperature effect on MEMRISTOR-based Neural Network Accelerators by electro-thermal simulation
	Shang,Mengjun 1; Yin,Longxiang 2; Xu,Ning 1
刊名	Journal of Physics: Conference Series
	2021-02-01
卷号	1812 期号:1
ISSN号	1742-6588
DOI	10.1088/1742-6596/1812/1/012025
英文摘要	Abstract Nowadays, memristor-based neural network accelerators have been widely studied due to their outstanding performance in massive parallel vector matrix multiplication. However, the memristor is sensitive to temperature and its on/off state operation window can be seriously degraded by the increasing temperature, which may lead to computation failures in memristor -based NN accelerators. In this work, we establish an electro-thermal simulation platform to evaluate the temperature impact on memristor -based NN accelerators. With this platform, we first investigate the impact on computation accuracy with the temperature increase in different NN layers in the accelerators. We then apply a temperature-aware NN weight mapping scheme to the most temperature-sensitive layer and achieve 28.89% improvement in computation accuracy, which only has 0.06% difference with the improvement achieved by applied the mapping scheme to the whole NN model. This finding can help to simplify the temperature-aware hardware optimization design in memristor-based neural network accelerators and reduce the power consumption.
语种	英语
出版者	IOP Publishing
WOS记录号	IOP:1742-6588-1812-1-012025
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/16247]
专题	中国科学院计算技术研究所
作者单位	1.School of Information Engineering, Wuhan University of Technology, Wuhan, China 2.State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
推荐引用方式 GB/T 7714	Shang,Mengjun,Yin,Longxiang,Xu,Ning. Degradation analysis and optimization of temperature effect on MEMRISTOR-based Neural Network Accelerators by electro-thermal simulation[J]. Journal of Physics: Conference Series,2021,1812(1).
APA	Shang,Mengjun,Yin,Longxiang,&Xu,Ning.(2021).Degradation analysis and optimization of temperature effect on MEMRISTOR-based Neural Network Accelerators by electro-thermal simulation.Journal of Physics: Conference Series,1812(1).
MLA	Shang,Mengjun,et al."Degradation analysis and optimization of temperature effect on MEMRISTOR-based Neural Network Accelerators by electro-thermal simulation".Journal of Physics: Conference Series 1812.1(2021).