Leveraging Voltage-Controlled Magnetic Anisotropy to Solve Sneak Path Issues in Crossbar Arrays

TL;DR

This paper introduces a spintronic device leveraging voltage-controlled magnetic anisotropy to address sneak path issues in crossbar arrays, enhancing energy efficiency and switching reliability for AI hardware.

Contribution

It proposes a novel MTJ-based device design utilizing VCMA effect to mitigate sneak path problems in crossbar arrays, with insights on optimal voltage conditions for robust switching.

Findings

VCMA-based MTJ device reduces sneak path errors

Optimal voltage conditions improve switching robustness

Enhanced energy efficiency in crossbar array programming

Abstract

In crossbar array structures, which serves as an "In-Memory" compute engine for Artificial Intelligence hardware, write sneak path problem causes undesired switching of devices that degrades network accuracy. While custom crossbar programming schemes have been proposed, device level innovations leveraging non-linear switching characteristics of the cross-point devices are still under exploration to improve the energy efficiency of the write process. In this work, a spintronic device design based on Magnetic Tunnel Junction (MTJ) exploiting the use of voltage-controlled magnetic anisotropy (VCMA) effect is proposed as a solution to the write sneak path problem. Additionally, insights are provided regarding appropriate operating voltage conditions to preserve the robustness of the magnetization trajectory during switching which is critical for proper switching probability manipulation.