The impact of the soft errors in convolutional neural network on GPUS: Alexnet as case study

Adam Ismail Hammad, Khalid and Mohamed Abdelaziz, Izzeldin Ibrahim and Younis, Younis M. (2021) The impact of the soft errors in convolutional neural network on GPUS: Alexnet as case study. Procedia Computer Science, 182. pp. 89-94. ISSN 1877-0509. (Published)

[img]
Preview
Pdf
The impact of the soft errors in convolutional neural network on GPUS_Alexnet as case study.pdf
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (753kB) | Preview

Abstract

Convolutional Neural Networks (CNNs) have been increasingly deployed in many applications, including safety critical system such as healthcare and autonomous vehicles. Meanwhile, the vulnerability of CNN model to soft errors (e.g., caused by radiation mduced) rapidly increases, thus reliability is crucial especially in real-tmie system. There are many traditional techniques for miprove the reliability of the system, e.g.. Triple Modular Redundancy, but these techniques incur high overheads, which makes them hard to deploy. In tins paper, we experimentally evaluate the vulnerable parts of Alexnet mode (e.g., fault mjector). Results show that FADD and LD are the top vulnerable mstructions against soft errors for Alexnet model, both mstruetions generate at least 84% of injected faults as SDC errors. Thus, these the only parts of the Alexnet model that need to be hardened mstead of usmg fully duplication solutions.

Item Type: Article
Additional Information: Indexed by Scopus
Uncontrolled Keywords: Convolutional neural networks; Fault injection; GPUS; Reliability; Soft errors
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
T Technology > T Technology (General)
T Technology > TA Engineering (General). Civil engineering (General)
T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division: Faculty of Electrical & Electronic Engineering
College of Engineering
Faculty of Computing
Institute of Postgraduate Studies
Depositing User: Mr Muhamad Firdaus Janih@Jaini
Date Deposited: 26 Oct 2022 03:31
Last Modified: 26 Oct 2022 03:31
URI: http://umpir.ump.edu.my/id/eprint/35434
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item