摘要
当前人工智能对算力的需求以超摩尔定律的速度增长,算法并行性高、数据重用性强,为处理器体系结构设计带来了更大的设计空间.众核处理器以其强大的片上计算能力、灵活的片上体系结构、高效的片上通信、柔性优化的存储等特性,为人工智能提供了更广阔的发展空间.本文在介绍众核处理器发展历史的基础上梳理了主要技术路线,重点论述人工智能应用对国产众核处理器体系结构和关键特性的需求.
In recent times,the demand for the computational capability of artificial intelligence(AI)is increasing rapidly.It is well-known that high parallelism algorithm and strong reusability of data provide more design space for processor architecture design.The manycore processor has a huge development space of AI with its strong on-chip computing power,flexible on-chip architecture,efficient on-chip communication,and flexible optimized storage.Based on the history of the development of manycore processors,this paper summarizes the main technical routes and focuses on the requirements of AI applications for the architecture and critical features of domestic manycore processors.
作者
李宏亮
郑方
郝子宇
高红光
过锋
唐勇
吕晖
刘鑫
陈芳园
Hongliang LI;Fang ZHENG;Ziyu HAO;Hongguang GAO;Feng GUO;Yong TANG;Hui LV;Xin LIU;Fangyuan CHEN(Jiangnan Institute of Computing Technology,Wuxi 214083,China)
出处
《中国科学:信息科学》
CSCD
北大核心
2019年第3期247-255,共9页
Scientia Sinica(Informationis)
基金
核高基项目面向数据中心(云平台)与集群计算的智能计算单元(批准号:2018ZX01028-102)资助项目
关键词
众核处理器
智能计算
体系结构
通信机制
存储体系
manycore processor
intelligent computing
computer architecture
communication mechanism
memory system