MMNNN: A tree-based Multicast Mechanism for NoC-based deep Neural Network accelerators

作     者：Ouyang, Yiming Tang, Feiyang Hu, Chunlei Zhou, Wu Wang, Qi 

作者机构：Hefei Univ Technol Sch Comp Sci & Informat Engn 485 Danxia Rd Hefei 230601 Peoples R China 

出 版 物：《MICROPROCESSORS AND MICROSYSTEMS》 (微处理机与微型系统)

年 卷 期：2021年第85卷

页      面：104242-104242页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National Natural Sci-ence Foundation of China (NSFC) Natural Science Foundation of Department of Education of Anhui Province China [KJ2017A477] 

主　　题：Network-on-Chip Deep Neural Network (DNN) accelerator Multicast routing algorithm Router architecture 

摘      要：Network-on-Chip (NoC) devices have been widely used in multiprocessor systems. In recent years, NoC-based Deep Neural Network (DNN) accelerators have been proposed to connect neural computing devices using NoCs. Such designs dramatically reduce off-chip memory accesses of these platforms. However, the large number of one-to-many packet transfers significantly degrade performance with traditional unicast channels. We propose a multicast mechanism for a NoC-based DNN accelerator called Multicast Mechanism for NoCbased Neural Network accelerator (MMNNN). To do so, we propose a tree-based multicast routing algorithm with excellent scalability and the ability to minimize the number of packets in the network. We also propose a router architecture for single-flit packets. Our proposed router transfers flits to multiple destinations in a single process and has no head-of-line blocking issue, offering higher throughput and lower latency than traditional wormhole router architectures. Simulation results show that our proposed multicast mechanism offers excellent performance in classification latency, average packet latency, and energy consumption.