C3SRAM: In-Memory-Computing SRAM Macro Based on Capacitive-Coupling Computing

被引：40

作者：

Jiang, Zhewei ^{[1
]}

Yin, Shihui ^{[2
]}

Seo, Jae-Sun ^{[2
]}

Seok, Mingoo ^{[1
]}

机构：

[1] Columbia Univ, Elect Engn Dept, New York, NY 10027 USA

[2] Arizona State Univ, Sch Elect Comp & Energy Engn, Tempe, AZ 85287 USA

来源：

IEEE SOLID-STATE CIRCUITS LETTERS | 2019年 / 2卷 / 09期

关键词：

Capacitive coupling; in-memory-computing (IMC); machine learning accelerator; mixed-signal processing; neural network;

D O I：

10.1109/LSSC.2019.2934831

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This letter presents C3SRAM, an in-memory-computing SRAM macro, which utilizes analog-mixed-signal capacitive-coupling computing to perform XNOR-and-accumulate operations for binary deep neural networks. The 256 x 64 C3SRAM macro asserts all 256 rows simultaneously and equips one ADC per column, realizing fully parallel vector-matrix multiplication in one cycle. C3SRAM demonstrates 672 TOPS/W and 1638 GOPS, and achieves 98.3% accuracy for MNIST and 85.5% for CIFAR-10 dataset. It achieves 3975x smaller energy-delay product than conventional digital processors.

引用

页码：131 / 134

页数：4

共 50 条

[1] C3SRAM: In-Memory-Computing SRAM Macro Based on Capacitive-Coupling Computing
Jiang, Zhewei
Yin, Shihui
Seo, Jae-Sun
Seok, Mingoo
IEEE 45TH EUROPEAN SOLID STATE CIRCUITS CONFERENCE (ESSCIRC 2019), 2019, : 131 - +
[2] C3SRAM: An In-Memory-Computing SRAM Macro Based on Robust Capacitive Coupling Computing Mechanism
Jiang, Zhewei
Yin, Shihui
Seo, Jae-Sun
Seok, Mingoo
IEEE JOURNAL OF SOLID-STATE CIRCUITS, 2020, 55 (07) : 1888 - 1897
[3] SRAM-based In-Memory-Computing for AI edge devices
Xu, Weidong
Lou, Mian
Xie, Chengmin
Li, Li
Shi, Zhu
Gong, Longqing
THIRD INTERNATIONAL CONFERENCE ON ELECTRONICS AND COMMUNICATION; NETWORK AND COMPUTER TECHNOLOGY (ECNCT 2021), 2022, 12167
[4] An Energy Consumption Model for SRAM-Based In-Memory-Computing Architectures
Akgul, Berke
Karalar, Tufan Coskun
ELECTRONICS, 2024, 13 (06)
[5] A 16Kb Transpose 6T SRAM In-Memory-Computing Macro based on Robust Charge-Domain Computing
Song, Jiahao
Wang, Yuan
Tang, Xiyuan
Wang, Runsheng
Huang, Ru
IEEE ASIAN SOLID-STATE CIRCUITS CONFERENCE (A-SSCC 2021), 2021,
[6] XNOR-SRAM: In-Bitcell Computing SRAM Macro based on Resistive Computing Mechanism
Jiang, Zhewei
Yin, Shihui
Seo, Jae-sun
Seok, Mingoo
GLSVLSI '19 - PROCEEDINGS OF THE 2019 ON GREAT LAKES SYMPOSIUM ON VLSI, 2019, : 417 - 422
[7] Design of In-Memory Computing Enabled SRAM Macro
Monga, Kanika
Behera, Sunit
Chaturvedi, Nitin
Gurunarayanan, S.
2022 IEEE 19TH INDIA COUNCIL INTERNATIONAL CONFERENCE, INDICON, 2022,
[8] A Computing-in-Memory SRAM Macro Based on Fully-Capacitive-Coupling With Hierarchical Capacity Attenuator for 4-b MAC Operation
Xiao, Kanglin
Cui, Xiaoxin
Qiao, Xin
Pan, Nanbing
Wang, Xin'An
Wang, Yuan
2022 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS 22), 2022, : 2551 - 2555
[9] A Multi-Bit In-Memory-Computing SRAM Macro Using Column-Wise Charge Redistribution for DNN Inference in Edge Computing Devices
Chae, Changseon
Kim, Subin
Choi, Jonghang
Park, Jun-Eun
18TH INTERNATIONAL SOC DESIGN CONFERENCE 2021 (ISOCC 2021), 2021, : 421 - 422
[10] A Charge-domain 10T SRAM based In-Memory-Computing Macro for Low Energy and Highly Accurate DNN inference
Kim, Joonhyung
Park, Jongsun
18TH INTERNATIONAL SOC DESIGN CONFERENCE 2021 (ISOCC 2021), 2021, : 89 - 90

← 1 2 3 4 5 →