Business Wire

Samsung Demonstrates the World's First MRAM Based In-memory Computing

13.1.2022 12:00:00 EET | Business Wire | Press release

Share

Samsung Electronics Co., Ltd., a world leader in advanced semiconductor technology, today announced its demonstration of the world's first in-memory computing based on MRAM (Magnetoresistive Random Access Memory). The paper on this innovation was published online by Nature on January 12 (GMT), and is set to be published in the upcoming print edition of Nature. Titled ‘A crossbar array of magnetoresistive memory devices for in-memory computing’1, this paper showcases Samsung’s leadership in memory technology and its effort to merge memory and system semiconductors for next-generation artificial intelligence (AI) chips.

This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20220113005393/en/

To view this piece of content from mms.businesswire.com, please give your consent at the top of this page.

Dr. Seungchul Jung, Dr. Donhee Ham, Dr. Sang Joon Kim (Photo: Business Wire)

The research was led by Samsung Advanced Institute of Technology (SAIT) in close collaboration with Samsung Electronics Foundry Business and Semiconductor R&D Center. The first author of the paper, Dr. Seungchul Jung, Staff Researcher at SAIT, and the co-corresponding authors Dr. Donhee Ham, Fellow of SAIT and Professor of Harvard University, and Dr. Sang Joon Kim, Vice President of Technology at SAIT, spearheaded the research.

In the standard computer architecture, data is stored in memory chips and data computing is executed in separate processor chips.

In contrast, in-memory computing is a new computing paradigm that seeks to perform both data storage and data computing in a memory network. Since this scheme can process a large amount of data stored within the memory network itself without having to move the data, and also because the data processing in the memory network is executed in a highly parallel manner, power consumption is substantially reduced. In-memory computing has thus emerged as one of the promising technologies to realize next-generation low-power AI semiconductor chips.

For this reason, research on in-memory computing has been intensely pursued worldwide. Non-volatile memories, in particular RRAM (Resistive Random Access Memory) and PRAM (Phase-change Random Access Memory), have been actively used for demonstrating in-memory computing. By contrast, it has so far been difficult to use MRAM ─ another type of non-volatile memory ─ for in-memory computing despite MRAM’s merits such as operation speed, endurance, and large-scale production. This difficulty stems from the low resistance of MRAM, due to which MRAM cannot enjoy the power reduction advantage when used in the standard in-memory computing architecture.

The Samsung Electronics researchers have provided a solution to this issue by an architectural innovation. Concretely, they succeeded in developing an MRAM array chip that demonstrates in-memory computing, by replacing the standard, ‘current-sum’ in-memory computing architecture with a new, ‘resistance sum’ in-memory computing architecture, which addresses the problem of small resistances of individual MRAM devices.

Samsung’s research team subsequently tested the performance of this MRAM in-memory computing chip by running it to perform AI computing. The chip achieved an accuracy of 98% in classification of hand-written digits and a 93% accuracy in detecting faces from scenes.

By ushering MRAM ─ the memory which has already reached commercial-scale production embedded in the system semiconductor fabrication ─ into the realm of in-memory computing, this work expands the frontier of the next-generation low-power AI chip technologies.

The researchers have also suggested that not only can this new MRAM chip be used for in-memory computing, but it also can serve as a platform to download biological neuronal networks. This is along the line of the neuromorphic electronics vision that Samsung’s researchers recently put forward in a perspective paper published in the September 2021 issue of the journal Nature Electronics.

"In-memory computing draws similarity to the brain in the sense that in the brain, computing also occurs within the network of biological memories, or synapses, the points where neurons touch one another,” said Dr. Seungchul Jung, the first author of the paper. “In fact, while the computing performed by our MRAM network for now has a different purpose from the computing performed by the brain, such solid-state memory network may in the future be used as a platform to mimic the brain by modelling the brain’s synapse connectivity."

As highlighted in this work, by building on its leading memory technology and merging it with system semiconductor technology, Samsung plans to continue to expand its leadership in next-generation computing and AI semiconductors.

# # #

About Samsung Electronics Co., Ltd.

Samsung inspires the world and shapes the future with transformative ideas and technologies. The company is redefining the worlds of TVs, smartphones, wearable devices, tablets, digital appliances, network systems, and memory, system LSI, foundry and LED solutions. For the latest news, please visit the Samsung Newsroom at news.samsung.com.


1 https://www.nature.com/articles/s41586-021-04196-6

To view this piece of content from cts.businesswire.com, please give your consent at the top of this page.

Contact information

Ujeong Jahnke
Samsung Semiconductor Europe GmbH
Tel. +49(0)89-45578-1000
Email: sseg.comm@samsung.com

About Business Wire

For more than 50 years, Business Wire has been the global leader in press release distribution and regulatory disclosure.

Subscribe to releases from Business Wire

Subscribe to all the latest releases from Business Wire by registering your e-mail address below. You can unsubscribe at any time.

Latest releases from Business Wire

New Holiday Data From Xsolla Shows Players Are Shopping, Gifting, and Choosing Brands Inside Video Games16.11.2026 19:00:00 EET | Press release

Xsolla (USA), Inc., a global commerce company, today announced a new data and insights report on how players shop, spend, and respond to brands during the holiday season. Titled “Why Video Games Belong in Your Holiday Commerce Strategy,” the report draws on direct answers from more than 4,000 players, gathered through opt-in rewarded surveys on Xsolla Offerwall. Players chose to take part and received a reward for sharing, which makes the data first-party and consent-based. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20261111104582/en/ Graphic: Xsolla The research builds on the conversation Xsolla opened at Advertising Week New York 2026 about how brands can reach players in more ways through video games. Combined with Xsolla’s insights into in-game purchasing behavior across an audience of over 300 million players, the survey shows an engaged audience with established spending habits: 84% of respondents call themselves gam

IQM Supports NVIDIA’s Work to Make Quantum Computers Schedulable in Slurm16.11.2026 18:00:00 EET | Press release

IQM Quantum Computers (Nasdaq: IQMX), a global leader in superconducting quantum computers, today announced at SC26 in Chicago its support for NVIDIA’s investment in making quantum processing units (QPUs) a schedulable resource in the Slurm workload manager by integrating IQM quantum computers. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20261116814640/en/ NVIDIA is making quantum computers a schedulable resource in Slurm by integrating IQM quantum computers. NVIDIA, which has committed to keeping Slurm open source and vendor neutral, announced that Slurm 26.11 adds plugin support for Hierarchical Resources (HRES), allowing a supercomputing center to describe a QPU as a shared, cluster-level resource and connect it to external services, such as the software that manages the device. Until now, quantum computers have not fit the scheduling models Slurm uses for CPUs and GPUs, so integrations have often been built separately f

The Rock-It Company Opens New Regional Office in Nashville16.11.2026 18:00:00 EET | Press release

The Rock-It Company, a global leader in live events and luxury goods logistics, announced the opening of a new regional office in Nashville, Tennessee, reinforcing its presence in one of the world's most important live music and cultural markets. The Nashville office brings together the full Rock-It platform with dedicated offices for Rock-It Cargo, Rock-It Sports, CARS by Rock-It and DIETL by Rock-It, serving as a regional hub for the company's broader portfolio of specialty logistics offerings. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20261116286387/en/ The Rock-It Company office in Nashville, Tennessee “Nashville has been a central part of our business for decades, and we see tremendous opportunity to continue building on that foundation," said Daniel Rosenthal, President and CEO of The Rock-It Company. "Our longstanding partnership with the Rock organization, combined with the expertise and relationships our teams ha

What Our Pupils Reveal: MindMetrix Joins Wyss Geneva to Bring Eye-Based Biofeedback to Mental Health16.11.2026 16:45:00 EET | Press release

We are proud to welcome MindMetrix AG as one of the newest Wyss Geneva ventures. Born out of research at ETH Zurich’s Neural Control of Movement Lab, MindMetrix is a neurotechnology company that reads the brain’s arousal system through the eyes and turns it into live biofeedback, so people can see, and learn to regulate, their own mental state. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20261116303495/en/ MindMetrix’s myflow platform turns pupil-based measurements into live biofeedback, with each user’s training sessions and progress shown in a dashboard. Image: MindMetrix The mind is still assessed mostly through questionnaires and self-report, with no objective, real-time measure of brain activity. The size of our pupils changes with the brain’s arousal center, a tiny region deep in the brainstem that is otherwise hard to observe without a brain scan. MindMetrix uses pupillometry to read that signal and turn it into live

Factbird Launches Visual Intelligence and Factbird MCP: AI That Sees the Shop Floor and Connects to Any AI Assistant16.11.2026 16:39:00 EET | Press release

Today, Factbird announces the launch of two new AI capabilities. Visual Intelligence uses cameras and AI to measure what sensors and PLCs miss. Factbird MCP lets AI assistants such as Claude, ChatGPT, and Microsoft Copilot work directly with production data in Factbird. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20261113769016/en/ Factbird Visual Intelligence tracking a bottling line in real time, showing filler and shrink wrapper status, line speed of 72 bottles per minute, and operator activity around the line. (AI-generated illustration) The losses that are hardest to explain happen around the machine. A station waits for material. A manual task runs long on the night shift. Each one costs a few minutes, and over a year they add up to real capacity. Factbird is a System of Action, built for manufacturers who have plenty of data and struggle to act on it. It connects production data from every line, machine, and operator

In our pressroom you can read all our latest releases, find our press contacts, images, documents and other relevant information about us.

Visit our pressroom
World GlobeA line styled icon from Orion Icon Library.HiddenA line styled icon from Orion Icon Library.Eye