高迪1,
陈烁1,
卢旭东1,
庞展曦1,
陈闯涛1,
尹勋钊1,
卓成1, 2,,
1.浙江大学信息与电子工程学院 杭州 310007
2.浙江大学国际联创中心 海宁 314400
基金项目:国家自然科学基金(61974133, 62034007),浙江省重点研发计划(2020C01052)
详细信息
作者简介:张力:男,1989年生,博士后,研究方向为人工智能算法及硬件加速、集成电路设计
高迪:女,1995年生,博士生,研究方向为新型计算架构、人工智能算法及硬件加速
尹勋钊:男,1991年生,研究员,研究方向为新型器件、电路、架构跨层协同设计
卓成:男,1981年生,研究员,研究方向为低功耗芯片设计、人工智能算法及硬件加速、3D芯片设计及优化
通讯作者:卓成 czhuo@zju.edu.cn
中图分类号:TN432计量
文章访问数:291
HTML全文浏览量:118
PDF下载量:43
被引次数:0
出版历程
收稿日期:2020-11-18
修回日期:2021-04-19
网络出版日期:2021-05-07
刊出日期:2021-06-18
An Energy Efficient Floating Point Computing Infrastructure Embedding Ferroelectric Field Effect Transistor Based Ternary Content Addressable Memories
Li ZHANG1,Di GAO1,
Shuo CHEN1,
Xudong LU1,
Zhanxi PANG1,
Chuangtao CHEN1,
Xunzhao YIN1,
Cheng ZHUO1, 2,,
1. College of Information Science & Electronics Engineering, Zhejiang University, Hangzhou 310007, China
2. International Joint Innovation Center, Zhejiang University, Haining 314400, China
Funds:The National Natural Science Foundation of China(61974133, 62034007), Zhejiang Provincial Key R&D program(2020C01052)
摘要
摘要:随着数据密集型应用的日益增多,内存墙问题已成为制约计算效率的瓶颈。该文提出一种新型的浮点数(FP)运算结构,该结构嵌入了基于铁电场效应晶体管(FeFET)的三元内容寻址存储器(TCAM)以实现高效的计算。通过特定规则设计的超高密度TCAM结构,可以用能效更高的TCAM搜索操作代替部分传统浮点运算,从而节约整体能耗。仿真实验证明,该文所提结构和运算执行流程,与常规浮点运算单元(FPU)相比,可以降低多达33%的能耗。
关键词:铁电场效应晶体管/
三元内容寻址寄存器/
浮点运算/
能效
Abstract:With the growing abundance of data-intensive applications, memory wall has become a bottleneck to computing efficiency. A novel Floating Point (FP) computing infrastructure that embeds Ferroelectric Field Effect Transistor (FeFET) based Ternary Content Addressable Memories (TCAMs) for energy efficient computing is proposed. With an ultra-dense TCAM implementation following the designed guidelines, the infrastructure can replace unnecessary Float Point Unit (FPU) executions with more efficient TCAM searching, thereby saving the overall energy consumption. Thanks to the proposed execution flow, the infrastructure can achieve up to 33% energy saving compared to regular FPUs.
Key words:Ferroelectric Field Effect Transistor (FeFET)/
Ternary Content Addressable Memory(TCAM)/
Floating-Point(FP) computing/
Energy-efficiency
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=9b0f7541-51a4-41df-83b0-0e0e04f5ab3e