Benchmarking monolithic 3D integration for compute-in-memory accelerators: Overcoming ADC bottlenecks and maintaining scalability to 7nm or beyond

Xiaochen Peng, Wriddhi Chakraborty, Ankit Kaul, Wonbo Shim, Muhannad S. Bakir, Suman Datta, Shimeng Yu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

22 Scopus citations

Abstract

This paper presents 3D NeuroSim, a benchmark framework of monolithic 3D (M3D) integrated compute-in-memory (CIM) accelerators. To address the challenges of analog-to-digital converter (ADC) overhead and scaling limitations caused by high write voltage in emerging nonvolatile memory (eNVM), we propose partitioning the circuit modules in hybrid technology nodes across two stacked tiers with massive inter-tier vias. This framework features versatile back-end-of-line (BEOL)-compatible transistors, including laser-recrystallized silicon transistor and oxide transistor, and analyzes the thermal profile for M3D integration. Finally, we benchmark the CIM accelerators for VGG-8 on CIFAR-10 and reveal the substantial benefits in energy efficiency of a hybrid M3D architecture (45nm eNVM array+7nm ADC and logic).

Original languageEnglish
Title of host publication2020 IEEE International Electron Devices Meeting, IEDM 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages30.4.1-30.4.4
ISBN (Electronic)9781728188881
DOIs
StatePublished - 12 Dec 2020
Event66th Annual IEEE International Electron Devices Meeting, IEDM 2020 - Virtual, San Francisco, United States
Duration: 12 Dec 202018 Dec 2020

Publication series

NameTechnical Digest - International Electron Devices Meeting, IEDM
Volume2020-December
ISSN (Print)0163-1918

Conference

Conference66th Annual IEEE International Electron Devices Meeting, IEDM 2020
Country/TerritoryUnited States
CityVirtual, San Francisco
Period12/12/2018/12/20

Fingerprint

Dive into the research topics of 'Benchmarking monolithic 3D integration for compute-in-memory accelerators: Overcoming ADC bottlenecks and maintaining scalability to 7nm or beyond'. Together they form a unique fingerprint.

Cite this