PRF: a process-RAM-feedback performance model to reveal bottlenecks and propose optimizations

作     者：Xie Zhen Tan Guangming Liu Weifeng Sun Ninghui 谢震;Tan Guangming;Liu Weifeng;Sun Ninghui

作者机构：Key Laboratory of Computer ArchitectureInstitute of Computing TechnologyChinese Academy of SciencesBeijing 100190P.R.China University of Chinese Academy of SciencesBeijing 100190P.R.China College of Information Science and EngineeringChina University of PetroleumBeijing 102249P.R.China 

出 版 物：《High Technology Letters》 (高技术通讯（英文版）)

年 卷 期：2020年第26卷第3期

页      面：285-298页

核心收录：

学科分类：08[工学] 081201[工学-计算机系统结构] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Supported by the National Key Research and Development Program of China(No.2017YFB0202105,2016YFB0201305,2016YFB0200803,2016YFB0200300) the National Natural Science Foundation of China(No.61521092,91430218,31327901,61472395,61432018) 

主　　题：performance model feedback optimization convolution sparse matrix-vector multiplication sn-sweep 

摘      要：Performance models provide insightful perspectives to predict performance and to propose optimization *** there has been much researches,pinpointing bottlenecks of various memory access patterns and reaching high accurate prediction of both regular and irregular programs on various hardware configurations are still not *** work proposes a novel model called process-RAM-feedback(PRF)to quantify the overhead of computation and data transmission time on general-purpose multi-core *** PRF model predicts the cost of instruction for singlecore by a directed acyclic graph(DAG)and the transmission time of memory access between each memory hierarchy through a newly designed cache *** using performance modeling and feedback optimization method,this paper uses PRF model to analyze and optimize convolution,sparse matrix-vector multiplication and sn-sweep as case study for covering with typical regular kernel to irregular and data *** the PRF model,it obtains optimization guidance with various sparsity structures,algorithm designs,and instruction sets support on different data sizes.