Embedded-MRAM, what the future holds

Embedded MRAM solutions are now entering the market, as leading foundries Samsung, TSMC and GlobalFoundries are scaling-up and ramping up their solutions.

Embedded MRAM market, MRAM-Info poll results 2021-09

This has been a slow process, as getting next-generation technologies into large fab is never an easy task, but we are finally seeing these foundries start to offer these technologies to their customers.

Researchers suggest using stochastic MRAM elements to create highly efficient AI neural network devices

Researchers from Northwestern University developed a new method of building artificial neural networks using MRAM-based stochastic computing units. The researchers say that this design could enable AI devices that are highly energy efficient.

MTJ-based stochastic computing unit structure (Northwestern University)

Embedded MRAM technologies are being adopted at major foundries, which enable the use of these technologies for unconventional computing architectures that use the stochasticity of MRAM cells (rather than their nonvolatility), to perform energy-efficient computing operations. MRAM cells exhibit stochastic switching characteristics, which is a challenge for reliable memory devices. But for neural networks, this can be taken advantage of if the MTJs are appropriately designed.

TSMC shows its eMRAM technology roadmap

During a presentation during Persistent Memory Summit, a new slide from TSMC was shown that describes the company's eMRAM roadmap:

TSMC memory roadmap (Persistant Mmemory Summit 2021)

As we already know, TSMC is offering 22nm eMRAM option as an eFlash alternative. The company is also looking to develop 14/12 nm eMRAM option to replace SRAM memory (this is interesting as before it was revealed that the company is developing 16nm MRAM). Finally eMRAM is touted as a possible replacement for configuration memory (eFUSE / OTP / Flash).

Everspin reported its preliminary Q4 2020 and full-year 2020 financial results

Everspin Technologies announced its preliminary Q4 2020 financial results. Total revenues were $10 million, down from $10.1 million in Q3 and up from $9.7 million in Q4 2019. Net loss int he quarter was $1.6 million, but the company generated $0.6 million in cash flow - Everspin's second consecutive positive cash flow from operations quarter.

Everspin Technologies chip photo

Looking at FY2020, Everspin's revenues increased 12.1% to $42 million. At the end of the year, Everspin had $14.6 million in cash and equivalents. In Q4 2020, the company also received its first royalty revenue from GF for embedded 22FDX process MRAM.

Samsung improves its MRAM performance, will expand its target applications

In March 2019 Samsung Electronics announced that it has started to mass produce its first embedded MRAM, made using the company's 28nm FD-SOI process. The company now announced that it managed to improve the MTJ function of its MRAM, which makes it suitable for more applications.

Samsung eMRAM image

Samsung will now expand the application of its eMRAM solutions to more markets - specifically the automotive, wearable, graphic memory, low level cache, internet of things and edge artificial intelligence markets.

IBM to reveal the world's first 14nm STT-MRAM node

IBM announced that during the 2020 IEEE International Electron Devices Meeting (IEDM 2020), that is now being held virtually, its researchers will reveal the first 14 nm node STT-MRAM. IBM says that efficient and high-performance STT-MRAM systems will help to address memory-compute bottlenecks in hybrid cloud systems.

IBM says that the 14 nm node embedded MRAM which will be revealed is the most advanced MRAM demonstrated to date. It features circuit design and process technology that could soon enable system designers to replace SRAM with twice the amount of MRAM in last-level CPU cache.

Everspin and Globalfoundries extend their MRAM agreement to 12 nm processes

Everspin Technologies announced that it has amended its STT-MRAM joint development agreement (JDA) with GLOBALFOUNDRIES to set the terms for a future project on an advanced 12 nm FinFET MRAM solution. Everspin agreement included 40 nm, 28 nm and 22 nm processes, and now also include 12 nm.

Everspin 1Gb STT-MRAM chip photo

GF recently announced it has achieved initial production of embedded MRAM (eMRAM) on its 22FDX platform.

GlobalFoundries starts producing eMRAM solutions, first customer tape-outs by the end of 2020

GlobalFoundries announced that it has delivered the first production-ready eMRAM on its 22FDX platform for IoT and automotive applications. The company says that its advanced eMRAM provides a "cost-effective solution for low-power, non-volatile code and data storage applications".

22FDX eMRAM vs 28nm eFLASH Value Proposition (GlobalFoundries)

GF says that it has entered production and is working with several clients with multiple production tape-outs scheduled in 2020. GF's eMRAM is designed as a replacement for high-volume embedded NOR flash (eFLASH). GF says that its eMRAM has passed five rigorous real-world solder reflow tests, and has demonstrated 100,000-cycle endurance and 10-year data retention across the -40°C to 125°C temperature range. The FDX eMRAM solution supports AEC-Q100 quality grade 2 designs, with development in process to support an AEC-Q100 quality grade 1 solution next year.

Mentor to provide IC test solutions for Arm's eMRAM compiler IP

Mentor announced that it will provide a unique IC test solution for the Arm's eMRAM compiler IP which is built on Samsung Foundry’s 28nm FDSOI process technology.

Mentor says it is working with Arm to leverage industry-leading Tessent software Built-In Self-Test (BIST) Design-for-Testability (DFT) technologies for testing the next-generation of Arm's eMRAM compiler IP in development.