A CNN Accelerator with Embedded Risc-V Controllers

Li Zhang; Xian Zhou; Chuliang Guo

doi:10.1109/CSTIC52283.2021.9461576

A CNN Accelerator with Embedded Risc-V Controllers

Li Zhang
, Xian Zhou
, Chuliang Guo

Zhejiang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Scopus citations

Abstract

Convolutional Neural Network is a promising technology in machine learning. Due to its vast computing and data requirements, it needs to be run with a specific accelerator to achieve reasonable energy efficiency. Improving the performance of accelerators has become the research hotspot. A mixed-precision structure can be used to improve hardware utilization, thus reducing area and power. However, the mixed-precision flow control is so complicated that it costs too much hardware resources. In this paper, a CNN accelerator with embedded RISC-V controllers is introduced to achieve flexible control at a very low cost. The ASIC synthesized results show that the proposed design area with two embedded cores is 5% less than the basic design.

Original language	English
Title of host publication	China Semiconductor Technology International Conference 2021, CSTIC 2021
Editors	Cor Claeys, Steve X. Liang, Qinghuang Lin, Ru Huang, Hanming Wu, Peilin Song, Linyong Pang, Ying Zhang, Beichao Zhang, Xinping Xinping Qu, Cheng Zhuo, Hsiang-Lan Lung
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665449458
DOIs	https://doi.org/10.1109/CSTIC52283.2021.9461576
State	Published - 14 Mar 2021
Externally published	Yes
Event	2021 China Semiconductor Technology International Conference, CSTIC 2021 - Shanghai, China Duration: 14 Mar 2021 → 15 Mar 2021

Publication series

Name	China Semiconductor Technology International Conference 2021, CSTIC 2021

Conference

Conference	2021 China Semiconductor Technology International Conference, CSTIC 2021
Country/Territory	China
City	Shanghai
Period	14/03/21 → 15/03/21

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Access to Document

10.1109/CSTIC52283.2021.9461576

Cite this

Zhang, L., Zhou, X., & Guo, C. (2021). A CNN Accelerator with Embedded Risc-V Controllers. In C. Claeys, S. X. Liang, Q. Lin, R. Huang, H. Wu, P. Song, L. Pang, Y. Zhang, B. Zhang, X. Xinping Qu, C. Zhuo, & H.-L. Lung (Eds.), China Semiconductor Technology International Conference 2021, CSTIC 2021 Article 9461576 (China Semiconductor Technology International Conference 2021, CSTIC 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CSTIC52283.2021.9461576

Zhang, Li ; Zhou, Xian ; Guo, Chuliang. / A CNN Accelerator with Embedded Risc-V Controllers. China Semiconductor Technology International Conference 2021, CSTIC 2021. editor / Cor Claeys ; Steve X. Liang ; Qinghuang Lin ; Ru Huang ; Hanming Wu ; Peilin Song ; Linyong Pang ; Ying Zhang ; Beichao Zhang ; Xinping Xinping Qu ; Cheng Zhuo ; Hsiang-Lan Lung. Institute of Electrical and Electronics Engineers Inc., 2021. (China Semiconductor Technology International Conference 2021, CSTIC 2021).

@inproceedings{ded58fbb4f274186b63f7a7a150b58f8,

title = "A CNN Accelerator with Embedded Risc-V Controllers",

abstract = "Convolutional Neural Network is a promising technology in machine learning. Due to its vast computing and data requirements, it needs to be run with a specific accelerator to achieve reasonable energy efficiency. Improving the performance of accelerators has become the research hotspot. A mixed-precision structure can be used to improve hardware utilization, thus reducing area and power. However, the mixed-precision flow control is so complicated that it costs too much hardware resources. In this paper, a CNN accelerator with embedded RISC-V controllers is introduced to achieve flexible control at a very low cost. The ASIC synthesized results show that the proposed design area with two embedded cores is 5\% less than the basic design.",

author = "Li Zhang and Xian Zhou and Chuliang Guo",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 China Semiconductor Technology International Conference, CSTIC 2021 ; Conference date: 14-03-2021 Through 15-03-2021",

year = "2021",

month = mar,

day = "14",

doi = "10.1109/CSTIC52283.2021.9461576",

language = "英语",

series = "China Semiconductor Technology International Conference 2021, CSTIC 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

editor = "Cor Claeys and Liang, \{Steve X.\} and Qinghuang Lin and Ru Huang and Hanming Wu and Peilin Song and Linyong Pang and Ying Zhang and Beichao Zhang and \{Xinping Qu\}, Xinping and Cheng Zhuo and Hsiang-Lan Lung",

booktitle = "China Semiconductor Technology International Conference 2021, CSTIC 2021",

address = "美国",

}

Zhang, L, Zhou, X & Guo, C 2021, A CNN Accelerator with Embedded Risc-V Controllers. in C Claeys, SX Liang, Q Lin, R Huang, H Wu, P Song, L Pang, Y Zhang, B Zhang, X Xinping Qu, C Zhuo & H-L Lung (eds), China Semiconductor Technology International Conference 2021, CSTIC 2021., 9461576, China Semiconductor Technology International Conference 2021, CSTIC 2021, Institute of Electrical and Electronics Engineers Inc., 2021 China Semiconductor Technology International Conference, CSTIC 2021, Shanghai, China, 14/03/21. https://doi.org/10.1109/CSTIC52283.2021.9461576

A CNN Accelerator with Embedded Risc-V Controllers. / Zhang, Li; Zhou, Xian; Guo, Chuliang.
China Semiconductor Technology International Conference 2021, CSTIC 2021. ed. / Cor Claeys; Steve X. Liang; Qinghuang Lin; Ru Huang; Hanming Wu; Peilin Song; Linyong Pang; Ying Zhang; Beichao Zhang; Xinping Xinping Qu; Cheng Zhuo; Hsiang-Lan Lung. Institute of Electrical and Electronics Engineers Inc., 2021. 9461576 (China Semiconductor Technology International Conference 2021, CSTIC 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A CNN Accelerator with Embedded Risc-V Controllers

AU - Zhang, Li

AU - Zhou, Xian

AU - Guo, Chuliang

PY - 2021/3/14

Y1 - 2021/3/14

N2 - Convolutional Neural Network is a promising technology in machine learning. Due to its vast computing and data requirements, it needs to be run with a specific accelerator to achieve reasonable energy efficiency. Improving the performance of accelerators has become the research hotspot. A mixed-precision structure can be used to improve hardware utilization, thus reducing area and power. However, the mixed-precision flow control is so complicated that it costs too much hardware resources. In this paper, a CNN accelerator with embedded RISC-V controllers is introduced to achieve flexible control at a very low cost. The ASIC synthesized results show that the proposed design area with two embedded cores is 5% less than the basic design.

AB - Convolutional Neural Network is a promising technology in machine learning. Due to its vast computing and data requirements, it needs to be run with a specific accelerator to achieve reasonable energy efficiency. Improving the performance of accelerators has become the research hotspot. A mixed-precision structure can be used to improve hardware utilization, thus reducing area and power. However, the mixed-precision flow control is so complicated that it costs too much hardware resources. In this paper, a CNN accelerator with embedded RISC-V controllers is introduced to achieve flexible control at a very low cost. The ASIC synthesized results show that the proposed design area with two embedded cores is 5% less than the basic design.

UR - https://www.scopus.com/pages/publications/85113609588

U2 - 10.1109/CSTIC52283.2021.9461576

DO - 10.1109/CSTIC52283.2021.9461576

M3 - 会议稿件

AN - SCOPUS:85113609588

T3 - China Semiconductor Technology International Conference 2021, CSTIC 2021

BT - China Semiconductor Technology International Conference 2021, CSTIC 2021

A2 - Claeys, Cor

A2 - Liang, Steve X.

A2 - Lin, Qinghuang

A2 - Huang, Ru

A2 - Wu, Hanming

A2 - Song, Peilin

A2 - Pang, Linyong

A2 - Zhang, Ying

A2 - Zhang, Beichao

A2 - Xinping Qu, Xinping

A2 - Zhuo, Cheng

A2 - Lung, Hsiang-Lan

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 China Semiconductor Technology International Conference, CSTIC 2021

Y2 - 14 March 2021 through 15 March 2021

ER -

Zhang L, Zhou X, Guo C. A CNN Accelerator with Embedded Risc-V Controllers. In Claeys C, Liang SX, Lin Q, Huang R, Wu H, Song P, Pang L, Zhang Y, Zhang B, Xinping Qu X, Zhuo C, Lung HL, editors, China Semiconductor Technology International Conference 2021, CSTIC 2021. Institute of Electrical and Electronics Engineers Inc. 2021. 9461576. (China Semiconductor Technology International Conference 2021, CSTIC 2021). doi: 10.1109/CSTIC52283.2021.9461576

A CNN Accelerator with Embedded Risc-V Controllers

Abstract

Publication series

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this