发表机构
Amirkabir University of Technology (Tehran Polytechnic)（阿米尔卡比尔理工大学（德黑兰理工大学））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文开发故障注入框架，探索基于ReRAM的存内计算CNN加速器在推理阶段的故障脆弱性，发现脆弱性受层类型、深度、参数位置及故障类型影响，且SaL比SaH更降低分类精度。
AI 中文摘要
基于阻变随机存取存储器（ReRAM）的存内计算（PIM）加速器因其具备模拟计算能力、超高密度、近零漏电流和非易失性，成为在并行域中处理神经网络中大规模内存密集型矩阵-向量乘法的有前景平台。尽管具有诸多优势，但由于技术制造工艺的限制导致工艺偏差和缺陷，基于ReRAM的加速器极易出错。这些限制降低了在PIM加速器上运行的深度卷积神经网络（Deep CNNs）的精度。虽然这些CNN加速器被广泛部署在安全关键系统中，但其对故障的脆弱性尚未得到充分探索。在本文中，我们开发了一个故障注入框架，以在推理阶段的软件层和硬件层两个层面研究大规模CNN的脆弱性。故障ReRAM器件是另一个可靠性挑战，因为当CNN参数映射到加速器时，会导致分类精度显著下降。为了研究这一挑战，我们将CNN学习参数映射到ReRAM交叉阵列中，并向交叉阵列注入故障。所提出的框架分析了高阻态（SaH）和低阻态（SaL）故障模型对CNN学习参数的不同层和位置的影响。通过执行大量故障注入，我们表明基于ReRAM的PIM加速器对CNN的脆弱性行为极大地受层类型和深度、每层中学习参数的位置以及故障值和类型的影响。我们的观察表明，不同模型对故障具有不同的脆弱性。具体而言，我们表明SaL比SaH进一步降低分类精度。
英文摘要
Resistive random-access memory (ReRAM)-based Processing-in-Memory (PIM) accelerator is a promising platform for processing massively memory intensive matrix-vector multiplications of neural networks in parallel domain, due to its capability of analog computation, ultra-high density, near-zero leakage current, and non-volatility. Despite many advantages, ReRAM-based accelerators are highly error-prone due to limitations of technology fabrication that lead to process variations and defects. These limitations degrade the accuracy of Deep Convolutional Neural Networks (Deep CNNs) running on PIM accelerators. While these CNNs accelerators are widely deployed in safety-critical systems, their vulnerability to fault is not well explored. In this paper, we have developed a fault injection framework to investigate the vulnerability of large-scale CNNs at both software- and hardware-level of inference phases. Faulty ReRAM devices are another reliability challenges due to significant degradation of classification accuracy when CNN parameters are mapped to the accelerators. To investigate this challenge, we map the CNN learning parameter to the ReRAM crossbar and inject faults into crossbar arrays. The proposed framework analyzes the impact of stuck-at high (SaH) and stuck-at low (SaL) fault models on different layers and locations of CNN learning parameters. By performing extensive fault injections, we illustrate that the vulnerability behavior of ReRAM-based PIM accelerator for CNNs is greatly impressible to the types and depth of layers, the location of the learning parameter in every layer, and the value and types of faults. Our observations show that different models have different vulnerabilities to faults. Specifically, we show that SaL further reduces classification accuracy than SaH.
Journal refA. Dorostkar, H. Farbeh and H. R. Zarandi, "An Empirical Fault Vulnerability Exploration of ReRAM-Based Process-in-Memory CNN Accelerators," in IEEE Transactions on Reliability, vol. 74, no. 1, pp. 2290-2304, March 2025
