In the first episode of Season 3 of IoT Unplugged, IoT Insider Editor Caitlin Gittins spoke to Sanjeev Aggarwal, President and CEO of Everspin Technologies on the topic of MRAM technology.
Aggrwal, whose own experience of non-volatile emerging memory dates back to 30 years in this industry, shared his insights into this cutting-edge technology. Talking about his appointment as CEO in 2022, he said: “That gave me an opportunity to learn about various aspects of our sprint, you know, what does it take to actually build this technology? What does it take to actually commercialise it? What are the challenges faced by each of the teams that work on this project?”
MRAM, or magnetic random access memory, stands out for its unique combination of persistence, data security, and low latency. Unlike traditional memory solutions, EMRAM retains information even when power is removed, ensuring data integrity.
Sanjeev highlighted two primary types of MRAM: toggle MRAM and spin transfer torque (STT) MRAM, with the latter gaining more attention in recent times. MRAM’s ability to combine persistence with high performance makes it a versatile technology with wide-reaching implications across various sectors.
The evolving demands on memory technology, particularly in the context of AI and data storage, position memory as a central component in computing. Traditional memory solutions like NAND flash, DRAM, and SRAM each serve specific roles, from storage to working memory and high-speed execution. MRAM, with its low latency and persistence, has the potential to replace these traditional memories, offering significant advantages in terms of performance and endurance.
“At the end of the day, it’s an emerging technology,” Aggarwal said in the episode on the subject of challenges. “People need to get used to it, or they need to feel it’s a reliable technology before they actually deploy it. If there is only one supplier of MRAM, for example Everspin, then that creates even more challenges.”
The potential for growth in MRAM technology is significant, particularly in AI inference and flash memory applications. Its advantages in power savings and performance make it a compelling alternative to traditional flash memory, offering a scalable solution for the future. Its low latency and endurance also make it suitable for AI applications, where it can store data on the Edge, reducing power consumption and enhancing performance.
To hear more from Everspin Technologies on MRAM technology, tune into the full podcast episode available on Spotify, Apple Podcasts, and at the link below.
How MRAM technology supports industrial IoT and more – IoT Unplugged
If the idea of appearing on the podcast to talk about IoT inspires you, feel free to reach out to us and pitch a topic you want to talk about and help us unplug the potential of IoT and explore the limitless opportunities it presents.
