Tetris:A Heuristic Static Memory Management Framework for Uniform Memory Multicore Neural Network Accelerators  

作　　者：Xiao-Bing Chen Hao Qi Shao-Hui Peng Yi-Min Zhuang Tian Zhi Yun-Ji Chen Distinguished Member,CCF 陈小兵;齐豪;彭少辉;庄毅敏;支天;陈云霁;Distinguished Member,CCF5(State Key Laboratory of Computer Architecture,Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China;University of Chinese Academy of Sciences,Beijing 100049,China;School of Computer Science and Technology,University of Science and Technology of China,HeFei,230026,China;Chinese Academy of Sciences Center for Excellence in Brain Science and Intelligence Technology,Shanghai,200031,China;不详)

机构地区：[1]State Key Laboratory of Computer Architecture,Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China [2]University of Chinese Academy of Sciences,Beijing 100049,China [3]School of Computer Science and Technology,University of Science and Technology of China,HeFei,230026,China [4]Chinese Academy of Sciences Center for Excellence in Brain Science and Intelligence Technology,Shanghai,200031,China [5]不详

出　　处：《Journal of Computer Science & Technology》2022年第6期1255-1270,共16页计算机科学技术学报（英文版）

基　　金：the Beijing Natural Science Foundation under Grant No.JQ18013;the National Natural Science Foundation of China under Grant Nos.61925208,61732007,61732002 and 61906179;the Strategic Priority Research Program of Chinese Academy of Sciences(CAS)under Grant No.XDB32050200;the Youth Innovation Promotion Association CAS,Beijing Academy of Artificial Intelligence(BAAI)and Xplore Prize.

摘　　要：Uniform memory multicore neural network accelerators(UNNAs)furnish huge computing power to emerging neural network applications.Meanwhile,with neural network architectures going deeper and wider,the limited memory capacity has become a constraint to deploy models on UNNA platforms.Therefore how to efficiently manage memory space and how to reduce workload footprints are urgently significant.In this paper,we propose Tetris:a heuristic static memory management framework for UNNA platforms.Tetris reconstructs execution flows and synchronization relationships among cores to analyze each tensor’s liveness interval.Then the memory management problem is converted to a sequence permutation problem.Tetris uses a genetic algorithm to explore the permutation space to optimize the memory management strategy and reduce memory footprints.We evaluate several typical neural networks and the experimental results demonstrate that Tetris outperforms the state-of-the-art memory allocation methods,and achieves an average memory reduction ratio of 91.9%and 87.9%for a quad-core and a 16-core Cambricon-X platform,respectively.

关 键 词：multicore neural network accelerators liveness analysis static memory management memory reuse genetic algorithm 

分 类 号：TP333[自动化与计算机技术—计算机系统结构] TP183[自动化与计算机技术—计算机科学与技术]