ReNEW: Enhancing lifetime for ReRAM crossbar based neural network accelerators

W Wen, Y Zhang, J Yang - 2019 IEEE 37th International …, 2019 - ieeexplore.ieee.org
2019 IEEE 37th International Conference on Computer Design (ICCD), 2019ieeexplore.ieee.org
With analog current accumulation feature, resistive memory (ReRAM) crossbars are widely
studied to accelerate neural network applications. The ReRAM crossbar based accelerators
have many advantages over conventional CMOS-based accelerators, such as high
performance and energy efficiency. However, due to the limited cell endurance, these
accelerators suffer from short programming cycles when weights that stored in ReRAM cells
are frequently updated during the neural network training phase. In this paper, by exploiting …
With analog current accumulation feature, resistive memory (ReRAM) crossbars are widely studied to accelerate neural network applications. The ReRAM crossbar based accelerators have many advantages over conventional CMOS-based accelerators, such as high performance and energy efficiency. However, due to the limited cell endurance, these accelerators suffer from short programming cycles when weights that stored in ReRAM cells are frequently updated during the neural network training phase. In this paper, by exploiting the wearing out mechanism of ReRAM cell, we propose a novel comprehensive framework, ReNEW, to enhance the lifetime of the ReRAM crossbar based accelerators, particularly for neural network training. Evaluation results show that, our proposed schemes reduce the total effective writes to ReRAM crossbar based accelerators by up to 500.3×, 50.0×, 2.83× and 1.60× over two MLC ReRAM crossbar baselines, one SLC ReRAM crossbar baseline and an SLC ReRAM crossbar design with optimal timing, respectively.
ieeexplore.ieee.org
以上显示的是最相近的搜索结果。 查看全部搜索结果