HERMES Core -基于14nm CMOS和pcm的内存计算核心,采用300ps/LSB线性化cco adc阵列和本地数字处理

R. Khaddam-Aljameh, M. Stanisavljevic, J. F. Mas, G. Karunaratne, M. Braendli, Femg Liu, Abhairaj Singh, S. M. Müller, U. Egger, A. Petropoulos, T. Antonakopoulos, K. Brew, Samuel Choi, I. Ok, F. Lie, N. Saulnier, V. Chan, I. Ahsan, V. Narayanan, S. Nandakumar, M. L. Gallo, P. Francese, A. Sebastian, E. Eleftheriou
{"title":"HERMES Core -基于14nm CMOS和pcm的内存计算核心,采用300ps/LSB线性化cco adc阵列和本地数字处理","authors":"R. Khaddam-Aljameh, M. Stanisavljevic, J. F. Mas, G. Karunaratne, M. Braendli, Femg Liu, Abhairaj Singh, S. M. Müller, U. Egger, A. Petropoulos, T. Antonakopoulos, K. Brew, Samuel Choi, I. Ok, F. Lie, N. Saulnier, V. Chan, I. Ahsan, V. Narayanan, S. Nandakumar, M. L. Gallo, P. Francese, A. Sebastian, E. Eleftheriou","doi":"10.23919/VLSICircuits52068.2021.9492362","DOIUrl":null,"url":null,"abstract":"We present a 256×256 in-memory compute (IMC) core designed and fabricated in 14nm CMOS with backend-integrated multi-level phase-change memory (PCM). It comprises 256 linearized current controlled oscillator (CCO)-based ADCs at a compact 4µm pitch and a local digital processing unit performing affine scaling and ReLU operations A novel frequency-linearization technique for CCOs is introduced, leading to accurate on-chip matrix-vector-multiply (MVM) when operating over 1 GHz. Measured classification accuracies on MNIST and CIFAR-10 datasets are presented when two cores are employed for deep learning (DL) inference The measured energy efficiency is 10.5 TOPS/W at a performance density of 1.59 TOPS/mm2.","PeriodicalId":106356,"journal":{"name":"2021 Symposium on VLSI Circuits","volume":"21 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"52","resultStr":"{\"title\":\"HERMES Core – A 14nm CMOS and PCM-based In-Memory Compute Core using an array of 300ps/LSB Linearized CCO-based ADCs and local digital processing\",\"authors\":\"R. Khaddam-Aljameh, M. Stanisavljevic, J. F. Mas, G. Karunaratne, M. Braendli, Femg Liu, Abhairaj Singh, S. M. Müller, U. Egger, A. Petropoulos, T. Antonakopoulos, K. Brew, Samuel Choi, I. Ok, F. Lie, N. Saulnier, V. Chan, I. Ahsan, V. Narayanan, S. Nandakumar, M. L. Gallo, P. Francese, A. Sebastian, E. Eleftheriou\",\"doi\":\"10.23919/VLSICircuits52068.2021.9492362\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We present a 256×256 in-memory compute (IMC) core designed and fabricated in 14nm CMOS with backend-integrated multi-level phase-change memory (PCM). It comprises 256 linearized current controlled oscillator (CCO)-based ADCs at a compact 4µm pitch and a local digital processing unit performing affine scaling and ReLU operations A novel frequency-linearization technique for CCOs is introduced, leading to accurate on-chip matrix-vector-multiply (MVM) when operating over 1 GHz. Measured classification accuracies on MNIST and CIFAR-10 datasets are presented when two cores are employed for deep learning (DL) inference The measured energy efficiency is 10.5 TOPS/W at a performance density of 1.59 TOPS/mm2.\",\"PeriodicalId\":106356,\"journal\":{\"name\":\"2021 Symposium on VLSI Circuits\",\"volume\":\"21 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-06-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"52\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 Symposium on VLSI Circuits\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.23919/VLSICircuits52068.2021.9492362\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 Symposium on VLSI Circuits","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/VLSICircuits52068.2021.9492362","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 52

摘要

我们提出了一种基于后端集成多级相变存储器(PCM)的14nm CMOS设计和制造的256×256内存计算(IMC)内核。它包括256个基于线性化电流控制振荡器(CCO)的adc,间距紧凑,为4µm,以及一个执行仿射缩放和ReLU操作的本地数字处理单元。介绍了用于CCO的新型频率线性化技术,当工作频率超过1 GHz时,可实现精确的片上矩阵矢量乘法(MVM)。在MNIST和CIFAR-10数据集上,当使用两个核心进行深度学习推理时,测量到的分类精度为10.5 TOPS/W,性能密度为1.59 TOPS/mm2。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
HERMES Core – A 14nm CMOS and PCM-based In-Memory Compute Core using an array of 300ps/LSB Linearized CCO-based ADCs and local digital processing
We present a 256×256 in-memory compute (IMC) core designed and fabricated in 14nm CMOS with backend-integrated multi-level phase-change memory (PCM). It comprises 256 linearized current controlled oscillator (CCO)-based ADCs at a compact 4µm pitch and a local digital processing unit performing affine scaling and ReLU operations A novel frequency-linearization technique for CCOs is introduced, leading to accurate on-chip matrix-vector-multiply (MVM) when operating over 1 GHz. Measured classification accuracies on MNIST and CIFAR-10 datasets are presented when two cores are employed for deep learning (DL) inference The measured energy efficiency is 10.5 TOPS/W at a performance density of 1.59 TOPS/mm2.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信