DNN performance optimization based on Gemmini neural network hardware accelerator

Denghui Zhao; Jianrui He; Xuyu Jing; Xibiao Hou

doi:10.1117/12.3049564

8 November 2024 DNN performance optimization based on Gemmini neural network hardware accelerator

Denghui Zhao, Jianrui He, Xuyu Jing, Xibiao Hou

Proceedings Volume 13416, Fourth International Conference on Advanced Algorithms and Neural Networks (AANN 2024); 1341611 (2024) https://doi.org/10.1117/12.3049564
Event: 2024 4th International Conference on Advanced Algorithms and Neural Networks, 2024, Qingdao, China

Abstract

Gemmini is an open-source, full-stack DNN accelerator generator. For an DNN algorithm running on Gemmini, traditional optimizations focus on changing Gemmini's hardware configuration (e.g., changing the size of internal storage and the size of the systolic array). This work proposes an an optimization-test iteration route combining Spike simulation and FPGA simulation under the constraints of unacceptable running speed of verilator simulator and limited FPGA resources. According to this route, this work starts from two aspects of hardware optimization and algorithm optimization. The internal storage size and calculation scale of Gemmini are changed in hardware, and the convolutional layer of Resnet50 neural network algorithm is optimized in algorithm. The experimental results show that for Resnet50 DNN algorithm, Gemmini with basic configuration provides 2000x speedup compared with running on CPU. On the platform of Gemmini accelerator, the optimization of hardware configuration provides 1.59x speedup for Resnet50. On the basis of the hardware optimization, the algorithm optimization provides 1.012x speedup.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Denghui Zhao, Jianrui He, Xuyu Jing, and Xibiao Hou "DNN performance optimization based on Gemmini neural network hardware accelerator", Proc. SPIE 13416, Fourth International Conference on Advanced Algorithms and Neural Networks (AANN 2024), 1341611 (8 November 2024); https://doi.org/10.1117/12.3049564

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available