Abstract:In recent years, memristive crossbar-based neuromorphic computing systems (NCS) have provided a promising solution to the acceleration of neural networks. However, stuck-at faults (SAFs) in the memristor devices significantly degrade the computing accuracy of NCS. Besides, the memristor suffers from the process variations, causing deviation of the actual programming resistance from its target resistance. In this paper, we propose a reliability-driven design framework for a memristive crossbar-based NCS in combination with general and chip-specific design optimizations. First, we design a general reliability-aware training scheme to enhance the robustness of NCS to SAFs and device variations; a dropconnect-inspired approach is developed to alleviate the impact of SAFs; a new weighted error function, including cross-entropy error (CEE), the $l_{2}$ -norm of weights, and the sum of squares of first-order derivatives of CEE with respect to weights, is proposed to obtain a smooth error curve, where the effects of variations are suppressed. Second, given the neural network model generated by the reliability-aware training scheme, we exploit chip-specific mapping and re- training to further improve computation accuracy loss incurred by SAFs. Experimental results show that the proposed method can boost the computation accuracy of NCS and improve the NCS robustness. Note to Practitioners—This work is motivated by the manufacturing reliability problem in a memristive crossbar-based NCS. To enhance the robustness of an NCS to SAFs and device variations, this paper presents a reliability-driven design framework with taking account of both general and chip-specific design optimizations. The experimental results have demonstrated that the proposed framework is superior to the prior arts, and can be easily integrated with existing industrial hardware-based fault tolerance solutions for higher accuracy at lower overhead. Memristive crossbar-based computing system gives hope for the anticipated efficient implementation of artificial neuromorphic networks. With the help of the reliability-driven designs, the computation accuracy is restored, and hence we can expect the wide use of memristive crossbar-based computing system in neuromorphic computing applications.

An Adaptive Cross-Layer Fault Recovery Solution for Reconfigurable SoCs.

A Cross-Layer Framework for Early-Stage Full Chip Oxide Breakdown Reliability Analysis

Design of a Reconfigurable Cryptographic Engine

The Research of FPGA Reliability Based on Redundancy Methods

A Flexible Soft Error Mitigation Framework Leveraging Dynamic Partial Reconfiguration Technology

Ada-FA: A Comprehensive Framework for Adaptive Fault Tolerance and Ageing Mitigation in FPGAs

Improving FPGA resilience through Partial Dynamic Reconfiguration

A Novel Method for Soft Error Mitigation in FPGA using Adaptive Cross Parity Code

M2STaR: A Multimode Spatio-Temporal Redundancy Design for Fault-Tolerant Coarse-Grained Reconfigurable Architectures

Fault-tolerant Technology of SRAM-based FPGA

CLEAR: Cross-Layer Exploration for Architecting Resilience - Combining Hardware and Software Techniques to Tolerate Soft Errors in Processor Cores

System on Chip Rejuvenation in the Wake of Persistent Attacks

Cross-Layer Design for Reliability in Advanced Technology Nodes: an EDA Perspective

A New Approach To Detect-Mitigate-Correct Radiation-Induced Faults For Sram-Based Fpgas In Aerospace Application

Reconfigurable Designs for Networking Silicon.

Implementation of Highly Reliable Convolutional Neural Network with Low Overhead on Field-Programmable Gate Array

Implementation of a Reconfigurable Computing System for Space Applications

Configuration Redundancy for Enhanced Reliability in SRAM-based FPGAs

A Fault-tolerant Structure for Reliable Multi-core Systems Based on Hardware-Software Co-design

Improved Design of Reconfigurable Computing Hardware Platform

Reliability-Driven Memristive Crossbar Design in Neuromorphic Computing Systems