A 256-kb Fully Row/Column-parallel 22nm MRAM In-Memory-Computing Macro with Differential Readout for Robust Parallelization and Scale-up

Peter Deaville, Bonan Zhang, Naveen Verma

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

This paper presents a 256-kb in-memory computing (IMC) macro implemented in 22nm CMOS with foundry MRAM. IMC leverages high levels of parallelism (both rows/columns within macro and multiple macros across chip), but where previous architectures have not addressed the data-dependent interference/noise between highly sensitive parallel analog circuits. Here, an IMC macro with differential readout architecture is demonstrated, which retains high energy efficiency, while overcoming the power-supply interference between the many parallel readout channels. High-sensitivity conductance-to-current readout is demonstrated, without relying on non-standard foundry cells, having atypical cell-state resistances or non-standard array configurations, and without relying on reduced row/column parallelism. The macro achieves state-of-the-art energy efficiency of 28.0-68.6 lb-TOPS/W and compute density of 5.43 1b-TOPS/mm2. CIFAR-10 classification is demonstrated, achieving software iso-accuracy of 90.25%, with all convolutional layers mapped to the chip.

Original languageEnglish (US)
Title of host publicationESSCIRC 2023 - IEEE 49th European Solid State Circuits Conference
PublisherIEEE Computer Society
Pages21-24
Number of pages4
ISBN (Electronic)9798350304206
DOIs
StatePublished - 2023
Event49th IEEE European Solid State Circuits Conference, ESSCIRC 2023 - Lisbon, Portugal
Duration: Sep 11 2023Sep 14 2023

Publication series

NameEuropean Solid-State Circuits Conference
Volume2023-September
ISSN (Print)1930-8833

Conference

Conference49th IEEE European Solid State Circuits Conference, ESSCIRC 2023
Country/TerritoryPortugal
CityLisbon
Period9/11/239/14/23

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'A 256-kb Fully Row/Column-parallel 22nm MRAM In-Memory-Computing Macro with Differential Readout for Robust Parallelization and Scale-up'. Together they form a unique fingerprint.

Cite this