一种2D权值固定数据流架构的研究

程智; 杨靓; 王硕; 娄冕

Chinese

您当前的位置：

首页 >

文章列表页 >

一种2D权值固定数据流架构的研究

更新时间：2026-09-04

- 一种2D权值固定数据流架构的研究
- Research on a 2D Weight Stationary Dataflow Architecture
- 微电子学与计算机 2021年38卷第2期页码：30-33
- 作者机构：
  
  西安微电子技术研究所,陕西,西安,710054
- 作者简介：
- 基金信息：
- DOI：
  中图分类号：
- 纸质出版：2021
- 稿件说明：
移动端阅览
程智, 杨靓, 王硕, 娄冕. 一种2D权值固定数据流架构的研究[J]. 微电子学与计算机, 2021, 38(2): 30-33.

CHENG Zhi, YANG Liang, WANG Shuo, et al. Research on a 2D Weight Stationary Dataflow Architecture[J]. 2021, 38(2): 30-33.
程智, 杨靓, 王硕, 娄冕. 一种2D权值固定数据流架构的研究[J]. 微电子学与计算机, 2021, 38(2): 30-33. DOI：

CHENG Zhi, YANG Liang, WANG Shuo, et al. Research on a 2D Weight Stationary Dataflow Architecture[J]. 2021, 38(2): 30-33. DOI：

摘要

随着人工智能算法的发展，卷积神经网络(CNN)在图像、音频等方面的应用越来越广泛，CNN算法的计算量也越来越大.权值固定数据流(WS)将权值固定在寄存器中，是一种最大化利用卷积重用和filter重用的数据流.不过当前的权值固定数据流结构存在建立流水线时间过长的问题.本文研究了一种去除PE(Process Element)行之间的FIFO，用加法器连接PE行的2D权值固定数据流结构.这种2D权值固定的数据流结构计算AlexNet时减少了近2.7倍建立流水线时间，并且能够灵活地调整卷积步长.

Abstract

With the development of artificial intelligence algorithms

convolutional neural networks (CNN) is more and more widely used in image

audio and other aspects

and the amount of calculation of CNN algorithms is also increasing. Weight stationary (WS) is a dataflow that maximizes the use of convolutional reuse and filter reuse by fixing weights in registers. However

the current WS dataflow structure has the problem that the pipeline filling time is lager. This paper studies a 2D-WS dataflow structure that uses PE adders to remove the FIFO between PE lines. This kind of 2D-WS dataflow structure reduces the pipeline filling time by nearly 2.7 times when calculating AlexNet

and can flexibly adjust the stride size.

关键词

Keywords

references

LIN M, CHEN Q, YAN S C. Network in network[J]. arXiv: 1312.4400, 2013.

CHEN H Y, KRISHNA T, EMER J S, et al. Eyeriss: an energy-efficient reconfigurable accelerator for deep convolutional neural networks[J]. IEEE Journal of Solid-State Circuits, 2017, 52(1): 127-138. DOI: 10.1109/JSSC.2016.2616357.

KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification withdeep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. Lake Tahoe, Nevada: ACM, 2012: 1097-1105.

DENG L, LI J Y, HUANG J T, et al.Recent advances in deeplearning for speech research at Microsoft[C]//Proceedings of2013 IEEE International Conference on Acoustics, Speech and Signal Processing. Vancouver, BC, Canada: IEEE, 2013: 8604-8608. DOI: 10.1109/ICASSP.2013.6639345.

FARABET C, POULET C, HAN J Y, et al. CNP: an FPGA-based processor for convolutional networks[C]//Proceedings of2009 International Conference on Field Programmable Logic and Applications. Prague, Czech Republic: IEEE, 2009: 32-37. DOI: 10.1109/FPL.2009.5272559.

HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 770-778. DOI: 10.1109/CVPR.2016.90.

CHEN Y H, EMER J, SZE V. Eyeriss: a spatial architecture for energy-efficient dataflow for convolutional neural networks[C]//Proceedings of2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture. Seoul, South Korea: IEEE, 2016: 367-379. DOI: 10.1109/ISCA.2016.40.

LU W Y, YAN G H, LI J J, et al. FlexFlow: a flexible dataflow accelerator architecture forconvolutional neural networks[C]//Proceedings of2017 IEEE International Symposium on High Performance Computer Architecture. Austin, TX, USA: IEEE, 2017: 553-564. DOI: 10.1109/HPCA.2017.29.

SZE V, CHEN Y H, YANG T J, et al. Efficient processing of deep neural networks: a tutorial and survey[J]. Proceedings of the IEEE, 2017, 105(12): 2295-2329. DOI: 10.1109/JPROC.2017.2761740.

浏览量

204

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种混合型流水线架构的FFT加速引擎设计

抗相关性功耗分析的DES掩码方案

一种多数据集混合累加电路设计

一种16位定点式DSP核的设计及Modelsim仿真验证

基于Booth算法的32位流水线型乘法器设计