Litcius/Paper detail

High RA Dual-MTJ SOT-MRAM devices for High Speed (10ns) Compute-in-Memory Applications

M. Y. Song, Kuo-Lung Chen, K. M. Chen, K. T. Chang, I. J. Wang, Yu-Chen Hsin, Chih‐Yang Lin, Elia Ambrosi, Win-San Khwa, Yu Lu, Che‐Yuan Hu, Shao‐Yu Yang, S. H. Li, J. H. Wei, T. Y. Lee, Y. J. Wang, Meng‐Fan Chang, Chi‐Feng Pai, X. Y. Bao

202316 citationsDOI

Abstract

The rapid development of artificial intelligence in recent decades has been continuously driving new software and hardware advancements. High-dimensional matrix-vector multiplication (MVM) is a crucial component in signal processing and machine learning computations. To achieve MVM, the 2D crossbar array of memristors has been widely discussed and studied. In this work, a novel SOT-MRAM device structure with 10ns write speed and >100x scalable resistance and read current are demonstrated to address the persistent problems of the traditional 2D crossbar array, leveraging its read-write path separation nature.

Topics & Concepts

Crossbar switchComputer scienceMagnetoresistive random-access memoryScalabilityMultiplication (music)Computer hardwarePath (computing)Component (thermodynamics)Parallel computingComputer architectureRandom access memoryPhysicsTelecommunicationsDatabaseProgramming languageAcousticsThermodynamicsAdvanced Memory and Neural ComputingFerroelectric and Negative Capacitance DevicesSemiconductor materials and devices
High RA Dual-MTJ SOT-MRAM devices for High Speed (10ns) Compute-in-Memory Applications | Litcius