Convolutional neural network adaptation and optimization method in SIMT computing mode

作     者：Feng Zhenfu Zhang Yaying Yang Lele Xing Lidong 

作者机构：School of Electronic EngineeringXi'an University of Posts and TelecommunicationsXi'an 710121China 

出 版 物：《The Journal of China Universities of Posts and Telecommunications》 (中国邮电高校学报（英文版）)

年 卷 期：2024年第31卷第2期

页      面：105-112页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 081104[工学-模式识别与智能系统] 0835[工学-软件工程] 081201[工学-计算机系统结构] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：the Scientific Research Program Funded by Shaanxi Provincial Education Department(20JY058) 

主　　题：parallel computing single instruction multiple threads(SIMT) convolutional neural network(CNN) memory optimization 

摘      要：For studying and optimizing the performance of general-purpose computing on graphics processing units(GPGPU)based on single instruction multiple threads(SIMT)processor about the neural network application,this work contributes a self-developed SIMT processor named Pomelo and correlated assembly *** parallel mechanism of SIMT computing mode and self-developed Pomelo processor is briefly introduced.A common convolutional neural network(CNN)is built to verify the compatibility and functionality of the Pomelo *** computing flow with task level and hardware level optimization is adopted on the Pomelo processor.A specific algorithm for organizing a Z-shaped memory structure is developed,which addresses reducing memory access in mass data computing *** the above-combined adaptation and optimization strategy,the experimental result demonstrates that reducing memory access in SIMT computing mode plays a crucial role in improving performance.A 6.52 times performance is achieved on the 4 processing elements case.