Insights and Optimizations on IR-drop Induced Sneak-Path for RRAM Crossbar-based Convolutions

RRAM crossbar structure has been proposed to accelerate the convolution computation neural networks because its current-mode weighted summation operation intrinsically matches the dominant multiplication-and-accumulation (MAC) operations. However, there is an inevitable IR-drop problem with the RRAM crossbar, which may introduce sneak-path and thus reduce the accuracy of neural network algorithms and the system reliability. This work addresses the sneak-path problem caused by the IR-drop in a RRAM crossbar. We first present the characteristics of variation distribution of the sneak-path through numerous experiments, taking into account RRAM cell resistance, input voltage, and cell location in a crossbar. Then we propose optimization strategies from the hardware and software perspectives respectively to mitigate the variations resulting from sneak-path. The experimental results show that the proposed methods can compensate the accuracy of algorithms.

[1]  Yu Wang,et al.  A peripheral circuit reuse structure integrated with a retimed data flow for low power RRAM crossbar-based CNN , 2018, 2018 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[2]  Jingtong Hu,et al.  Design Exploration for Multiple Level Cell Based Non-Volatile FPGAs , 2017, 2017 IEEE International Conference on Computer Design (ICCD).

[3]  Qing Wu,et al.  Hardware realization of BSB recall function using memristor crossbar arrays , 2012, DAC Design Automation Conference 2012.

[4]  Jiale Liang,et al.  Cross-Point Memory Array Without Cell Selectors—Device Characteristics and Data Storage Pattern Dependencies , 2010, IEEE Transactions on Electron Devices.

[5]  Tao Zhang,et al.  Overcoming the challenges of crossbar resistive memory architectures , 2015, 2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA).

[6]  Rainer Waser,et al.  Complementary resistive switches for passive nanocrossbar memories. , 2010, Nature materials.

[7]  Yiran Chen,et al.  Vortex: Variation-aware training for memristor X-bar , 2015, 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC).

[8]  Miao Hu,et al.  ISAAC: A Convolutional Neural Network Accelerator with In-Situ Analog Arithmetic in Crossbars , 2016, 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA).

[9]  Xuefei Ning,et al.  Fault-Tolerant Training Enabled by On-Line Fault Detection for RRAM-Based Neural Computing Systems , 2019, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[10]  Huazhong Yang,et al.  TIME: A Training-in-Memory Architecture for RRAM-Based Deep Neural Networks , 2019, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[11]  Yu Wang,et al.  Switched by input: Power efficient structure for RRAM-based convolutional neural network , 2016, 2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC).

[12]  Kyeong-Sik Min,et al.  Two-Step Write Scheme for Reducing Sneak-Path Leakage in Complementary Memristor Array , 2012, IEEE Transactions on Nanotechnology.

[13]  Borivoje Nikolić,et al.  A Variation-Tolerant, Sneak-Current-Compensated Readout Scheme for Cross-Point Memory Based on Two-Port Sensing Technique , 2018, IEEE Transactions on Circuits and Systems II: Express Briefs.

[14]  P. Bafna,et al.  Punchthrough-Diode-Based Bipolar RRAM Selector by Si Epitaxy , 2012, IEEE Electron Device Letters.

[15]  Yiran Chen,et al.  Reduction and IR-drop compensations techniques for reliable neuromorphic computing systems , 2014, 2014 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).