Technology-Array-Algorithm Co-Optimization of RRAM for Storage and Neuromorphic Computing: Device Non-idealities and Thermal Cross-talk
Yimao Cai, Zongwei Wang, Zhizhen Yu, Yaotian Ling, Qingyu Chen, Yunfan Yang, Shengyu Bao, Lindong Wu, Lin Bao, Runsheng Wang, Ru Huang
Abstract
The device non-idealities and array thermal cross-talk will inevitably impair the performance of RRAM-based storage and neuromorphic systems. In this paper, we present the modeling of device non-idealities, including parameter variations and time-dependent fluctuation (TDV), and the simulation of array thermal crosstalk. We also investigate the impact of non-idealities and thermal issues on storage and neuromorphic computing from the device, array, and algorithm perspectives, providing guidelines for the technology-array-algorithm co-optimization.
Topics & Concepts
Neuromorphic engineeringCrosstalkComputer scienceElectronic engineeringThermalResistive random-access memoryComputer architectureComputational scienceComputer engineeringEngineeringElectrical engineeringArtificial neural networkArtificial intelligencePhysicsVoltageMeteorologyAdvanced Memory and Neural ComputingFerroelectric and Negative Capacitance DevicesSemiconductor materials and devices