存算一体技术研究现状
Research Status of Computing-in-Memory Technology
李嘉宁 1姚鹏 1揭路 1唐建石 1伍冬 1高滨 1钱鹤 1吴华强1
作者信息
- 1. 清华大学集成电路学院,北京 100084
- 折叠
摘要
冯诺依曼计算机体系结构面临着"存储墙"的瓶颈,阻碍AI(Artificial Intelligence)计算性能提升.存算一体硬件结构打破了"存储墙"的限制,大大提升了 AI计算的性能.目前存算一体计算方案已在多种存储介质上得到实现,根据计算信号类型,可以将存算一体计算方案分成数字存算一体方案和模拟存算一体方案.存算一体硬件结构使得AI计算的性能取得巨大提升,然而进一步发展仍面临重大挑战.本文对不同信号域的存算一体方案的进行了对比分析,指出了每一种方案的主要优缺点,也指明了存算一体技术面临的挑战.我们认为,随着工艺集成、器件、电路、架构,软件工具链的跨层次协同研究发展,存算一体技术将在边缘端和云端,为AI计算提供更加强大和高效的算力.
Abstract
Von Neumann computer architecture faces the bottleneck of"storage wall",which hindering the performance improvement of AI(Artificial Intelligence)computing.Computing-In-Memory(CIM)breaks the limitation of"storage wall"and greatly improves the performance of AI computing.At present,CIM schemes have been implemented in a variety of storage media.According to the type of calculation signal,CIM scheme can be divided into digital CIM and analog CIM scheme.CIM has greatly improved the performance of AI computing,but the further development still faces major challenges.This article provides a detailed comparative analysis of CIM schemes in different signal domains,pointing out the main advantages and disadvantages of each scheme,and also pointing out the challenges faced by CIM.We believe that with the cross level col-laborative research and development of process integration,devices,circuits,architecture,and software toolchains,CIM will provide more powerful and efficient computing power for AI computing at the edge and cloud ends.
关键词
人工智能/存算一体/存储介质/计算信号类型/评价指标Key words
artificial intelligence/computing-in-memory/storage media/calculate signal type/evaluation index引用本文复制引用
基金项目
国家自然科学基金(92164302)
国家自然科学基金(62025111)
出版年
2024
国家科技期刊平台
NSTL
万方数据