Antiferromagnetic Tunnel Junctions (AFMTJs) for In-Memory Computing: Modeling and Case Study

TL;DR

This paper introduces a comprehensive simulation framework for antiferromagnetic tunnel junctions (AFMTJs), demonstrating their superior speed and energy efficiency in in-memory computing compared to traditional magnetic tunnel junctions (MTJs).

Contribution

It presents the first end-to-end modeling and case study of AFMTJs, highlighting their potential for ultrafast, low-power in-memory computing architectures.

Findings

AFMTJs achieve ~8x lower write latency than conventional MTJs.

AFMTJs reduce write energy by ~9x compared to MTJs.

In-memory computing with AFMTJs delivers 17.5x speedup and 20x energy savings over CPU baseline.

Abstract

Antiferromagnetic Tunnel Junctions (AFMTJs) enable picosecond switching and femtojoule writes through ultrafast sublattice dynamics. We present the first end-to-end AFMTJ simulation framework integrating multi-sublattice Landau-Lifshitz-Gilbert (LLG) dynamics with circuit-level modeling. SPICE-based simulations show that AFMTJs achieve ~8x lower write latency and ~9x lower write energy than conventional MTJs. When integrated into an in-memory computing architecture, AFMTJs deliver 17.5x average speedup and nearly 20x energy savings versus a CPU baseline-significantly outperforming MTJ-based IMC. These results establish AFMTJs as a compelling primitive for scalable, low-power computing.