Optimizing energy efficiency of CNN-based object detection with dynamic voltage and frequency scaling

Jiang, Weixiong, Yu, Heng, Zhang, Jiale, Wu, Jiaxuan, Luo, Shaobo and Ha, Yajun (2020) Optimizing energy efficiency of CNN-based object detection with dynamic voltage and frequency scaling. Journal of Semiconductors, 41 (2). 022406. ISSN 1674-4926

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (6MB) | Preview

Abstract

On the one hand, accelerating convolution neural networks (CNNs) on FPGAs requires ever increasing high energy efficiency in the edge computing paradigm. On the other hand, unlike normal digital algorithms, CNNs maintain their high robustness even with limited timing errors. By taking advantage of this unique feature, we propose to use dynamic voltage and frequency scaling (DVFS) to further optimize the energy efficiency for CNNs. First, we have developed a DVFS framework on FPGAs. Second, we apply the DVFS to SkyNet, a state-of-the-art neural network targeting on object detection. Third, we analyze the impact of DVFS on CNNs in terms of performance, power, energy efficiency and accuracy. Compared to the state-of-the-art, experimental results show that we have achieved 38% improvement in energy efficiency without any loss in accuracy. Results also show that we can achieve 47% improvement in energy efficiency if we allow 0.11% relaxation in accuracy.

Item Type: Article
Keywords: CNN; FPGA; DVFS; object detection
Schools/Departments: University of Nottingham Ningbo China > Faculty of Science and Engineering > School of Computer Science
Identification Number: https://doi.org/10.1088/1674-4926/41/2/022406
Depositing User: Zhou, Elsie
Date Deposited: 22 Jun 2020 03:09
Last Modified: 22 Jun 2020 03:09
URI: https://eprints.nottingham.ac.uk/id/eprint/60946

Actions (Archive Staff Only)

Edit View Edit View