Shaping the future of non-volatile memories with the MRAM Alliance
Siamak Salimy
CTO
Hprobe
List of Authors
J.-P. Nozières1, 2, S. Salimy3, J. Guedj4, D. Worledge5
1MRAM Alliance, SNIA, Santa Clara, United States, 2CNRS, Institut Néel, Grenoble, France, 3Hprobe, Eybens, France, 4Numem, Sunnyvale, United States, 5IBM Research, San Jose, United States
Abstract
Over the past several decades, the semiconductor sector has been driven by Moore’s Law, continuously increasing transistor density to improve chip performance. With computational power of AI chips having grown much faster than memory bandwidth and data transfer efficiency, data centers may soon be choked by the demand and the cost of providing high performance data access. At the same time, software-defined vehicles with centralized, high-compute chip architectures, need more bandwidth, faster boot times and persistent data for performance, safety, updateability, and user experience. Memory, once considered a commodity, is now left, right, and center.Jump in MRAM! Magnetic Random Access Memory (MRAM) has become a compelling and widely deployed non-volatile memory technology across a growing range of applications spanning standalone and embedded products. With built-in persistence, near-infinite endurance, intrinsic robustness, faster read speed than DRAM, and better density than SRAM, MRAM is now poised to expand into next-generation automotive, storage and AI systems.MRAM however stands at a critical juncture. Despite technical readiness and availability at tier-1 foundries (TSMC, Samsung and GF leading the pack), adoption still faces hesitation. Fragmented efforts, slow standardization, and inconsistent messaging hinder progress. This triggered stakeholders across the whole value chain, from foundries to chip makers, IP vendors and system companies, equipment suppliers and leading universities/research centers, to join forces. SNIA's MRAM Alliance Special Interest Group (SIG) is designed to align the industry on key priorities, address real-world challenges, improve education and awareness, create standards when applicable, facilitate product implementation, and ultimately maximize end-system product benefits.Whether you're ready to deploy today or want to help shape what comes next, come hear what’s not to love about MRAM, and how the MRAM Alliance can shape the future.
Biography
From permanent magnets to MRAM, Dr. Jean-Pierre Nozières has been involved in magnetics research and development since the late 80’s. After graduating with an Electrical Engineering Degree and a PhD in Physics, he spent his all career in between industry and research, from IBM (USA) where he pioneered the work on “spin valves” to CNRS, first at Laboratoire Louis Néel, then at Spintec, which he founded and headed for 9 years. After short spells at Applied Magnetics Corporation, a US disk-drive head company and PHS, a French start-up, he founded several MRAM start-ups, first Crocus Technology, where he acted as the CTO for 4 years, then eVaderis, Hprobe and Antaios. He is now back at CNRS and is in parallel spearheading within SNIA the MRAM Alliance Special Interest Group, a consortium of all MRAM industry stakeholder aiming at promoting and accelerating the rollout of the technology.
