A 44.3 TOPS/W SRAM Compute-in-Memory With Near-CIM Analog Memory and Activation for DAC/ADC-Less Operations

IF 2.2 Q3 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE IEEE Solid-State Circuits Letters Pub Date : 2024-06-24 DOI:10.1109/LSSC.2024.3418099
Peiyu Chen;Meng Wu;Wentao Zhao;Yufei Ma;Tianyu Jia;Le Ye
{"title":"A 44.3 TOPS/W SRAM Compute-in-Memory With Near-CIM Analog Memory and Activation for DAC/ADC-Less Operations","authors":"Peiyu Chen;Meng Wu;Wentao Zhao;Yufei Ma;Tianyu Jia;Le Ye","doi":"10.1109/LSSC.2024.3418099","DOIUrl":null,"url":null,"abstract":"In this letter, we present an analog compute-in-memory (CIM) macro design which incorporates near-CIM analog memory and nonlinearity activation unit (NAU) to alleviate the DAC/ADC power bottleneck. Fully differential analog memory is designed with switched capacitor storage circuits. Activation function, e.g., rectified linear unit, is also performed in analog domain in NAU. The CIM macro is fabricated using TSMC 55-nm technology, with a peak macro-level efficiency of 44.3 TOPS/W and a system energy efficiency of 27.7 TOPS/W for analog input and output with 4-bit weight. The near-CIM analog memory and NAU solution brings 76.0% energy reduction compared with DAC/ADC solution, which contributes \n<inline-formula> <tex-math>$1.34\\times $ </tex-math></inline-formula>\n to \n<inline-formula> <tex-math>$2.37\\times $ </tex-math></inline-formula>\n energy efficiency improvement.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2000,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Solid-State Circuits Letters","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/10569024/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}
引用次数: 0

Abstract

In this letter, we present an analog compute-in-memory (CIM) macro design which incorporates near-CIM analog memory and nonlinearity activation unit (NAU) to alleviate the DAC/ADC power bottleneck. Fully differential analog memory is designed with switched capacitor storage circuits. Activation function, e.g., rectified linear unit, is also performed in analog domain in NAU. The CIM macro is fabricated using TSMC 55-nm technology, with a peak macro-level efficiency of 44.3 TOPS/W and a system energy efficiency of 27.7 TOPS/W for analog input and output with 4-bit weight. The near-CIM analog memory and NAU solution brings 76.0% energy reduction compared with DAC/ADC solution, which contributes $1.34\times $ to $2.37\times $ energy efficiency improvement.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
具有近 CIM 模拟存储器的 44.3 TOPS/W SRAM 存贮器计算功能,可激活无 DAC/ADC 操作
在这封信中,我们介绍了一种模拟内存计算 (CIM) 宏设计,它结合了近 CIM 模拟存储器和非线性激活单元 (NAU),以缓解 DAC/ADC 的功率瓶颈。全差分模拟存储器采用开关电容存储电路设计。激活功能(如整流线性单元)也在 NAU 的模拟域中执行。CIM 宏采用台积电 55 纳米技术制造,峰值宏级能效为 44.3 TOPS/W,模拟输入和输出的系统能效为 27.7 TOPS/W,权重为 4 位。与 DAC/ADC 解决方案相比,近 CIM 模拟存储器和 NAU 解决方案的能耗降低了 76.0%,能效提高了 1.34 美元至 2.37 美元。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
IEEE Solid-State Circuits Letters
IEEE Solid-State Circuits Letters Engineering-Electrical and Electronic Engineering
CiteScore
4.30
自引率
3.70%
发文量
52
期刊最新文献
0.6-V, μW-Power Four-Stage OTA With Minimal Components, and 100× Load Range Broadband GaN MMIC Doherty Power Amplifier Using Compact Short-Circuited Coupler A 12 V Compliant Multichannel Dual Mode Neural Stimulator With 0.004% Charge Mismatch and a 4×VDD Tolerant On-Chip Discharge Switch in Low-Voltage CMOS A 14 nm MRAM-Based Multi-bit Analog In-Memory Computing With Process-Variation Calibration for 72 Macros-Based Accelerator S2D-CIM: SRAM-Based Systolic Digital Compute-in-Memory Framework With Domino Data Path Supporting Flexible Vector Operation and 2-D Weight Update
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1