首页|Research and design of matrix operation accelerator based on reconfigurable array

Research and design of matrix operation accelerator based on reconfigurable array

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
In the case of massive data,matrix operations are very computationally intensive,and the mem-ory limitation in standalone mode leads to the system inefficiencies.At the same time,it is difficult for matrix operations to achieve flexible switching between different requirements when implemented in hardware.To address this problem,this paper proposes a matrix operation accelerator based on reconfigurable arrays in the context of the application of recommender systems(RS).Based on the reconfigurable array processor(APR-16)with reconfiguration,a parallelized design of matrix opera-tions on processing element(PE)array is realized with flexibility.The experimental results show that,compared with the proposed central processing unit(CPU)and graphics processing unit(GPU)hybrid implementation matrix multiplication framework,the energy efficiency ratio of the ac-celerator proposed in this paper is improved by about 35×.Compared with blocked alternating least squares(BALS),its the energy efficiency ratio has been accelerated by about 1×,and the switc-hing of matrix factorization(MF)schemes suitable for different sparsity can be realized.

matrix factorization(MF)recommender system(RS)array processorrecon-figurablematrix multiplication

邓军勇、ZHANG Pan、JIANG Lin、XIE Xiaoyan、DENG Jingwen

展开 >

School of Electronic Engineering,Xi'an University of Posts and Telecommunications,Xi'an 710121,P.R.China

School of Computer Science and Technology,Xi'an University of Science and Technology,Xi'an 710054,P.R.China

School of Computer,Xi'an University of Posts and Telecommunications,Xi'an 710121,P.R.China

国家重点研发计划国家自然科学基金Shaanxi Province Key Research and Development PlanKey Scientific Research Project of Shaanxi Department of EducationEducation Research Project of Xi'an University of Posts and TelecommunicationsGraduate Student Innovation Fund of Xi'an University of Posts and Telecommunications

2022ZD0119001618340052022GY-02722JY060JGA202108CXJJYL2022035

2024

10.3772/j.issn.1006-6748.2024.02.003

高技术通讯(英文版)
中国科学技术信息研究所（ISTIC）

高技术通讯(英文版)

影响因子：0.058
ISSN：1006-6748
年,卷(期)：2024.30(2)
  • 28