STRIVE: Empowering a Low Power Tensor Processing Unit with Fault Detection and Error Resilience
Open Access
- 10 January 2025
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 30 (2) , 1-25
Abstract
Rapid growth in Deep Neural Network (DNN) workloads has increased the energy footprint of the Artificial Intelligence (AI) computing realm. For optimum energy efficiency, we propose operating a DNN hardware in the Low-Power Computing (LPC) region. However, operating at LPC causes increased delay sensitivity to Process Variation (PV). Delay faults are an intriguing consequence of PV. In this article, we demonstrate the vulnerability of DNNs to delay variations, substantially lowering the prediction accuracy. To overcome delay faults, we present STRIVE—a post-fabrication fault detection and reactive error reduction technique. We also introduce a time-borrow correction technique to ensure error-free DNN computation.Keywords
This publication has 44 references indexed in Scilit:
- TRe-Map: Towards Reducing the Overheads of Fault-Aware Retraining of Deep Neural Networks by Merging Fault Maps2021
- Test Architecture for Systolic Array of Edge-Based AI AcceleratorIEEE Access, 2021
- Improving DNN Fault Tolerance using Weight Pruning and Differential Crossbar Mapping for ReRAM-based Edge AI2021
- A Pipelined Multi-Level Fault Injector for Deep Neural Networks2020
- An Experimental Study of Reduced-Voltage Operation in Modern FPGAs for Neural Network Acceleration2020
- Optimizing energy efficiency of CNN-based object detection with dynamic voltage and frequency scalingJournal of Semiconductors, 2020
- A Low Power Digitally Assisted Analog Front End for Neural Interface with Optical Reception2017
- VARIUS-TC: A modular architecture-level model of parametric variation for thin-channel switches2016
- Man vs. computer: Benchmarking machine learning algorithms for traffic sign recognitionNeural Networks, 2012
- VARIUS-NTV: A microarchitectural model to capture the increased sensitivity of manycores to process variations at near-threshold voltages2012