In-memory computing: Samsung Electronics releases study on next-generation AI semiconductors with MRAM technology

Microprocessor manufacturers have been interested in MRAM (Magnetoresistive Random Access Memory) since the 1990s. Researchers from Samsung Advanced Institute of Technology (SAIT) in collaboration with Samsung Electronics Foundry Business and Semiconductor R&D Center performed the first calculation in memory based on this technology and published the results of their research in the journal Nature titled: A crossbar array of magnetoresistive memory devices for in-memory computing ». This approach is seen as one of the most promising ways to produce ultra-efficient and sober artificial intelligence chips.

In standard computer architecture, data is stored in memory chips and data processing is performed in separate processor chips. In contrast, in-memory computing is a new computing paradigm that seeks to perform both data storage and data computation in a memory array while consuming less power. In-memory computing is therefore a promising technology for realizing next-generation low-power AI semiconductor chips.

Computer systems include two types of memory. DRAM is the main memory while NAND flash memory is used for storage memory. High-speed, high-capacity DRAM is essential to the proper functioning of processors and plays an important role in today’s Von Neumann computer architecture (a single structure to store both program and data).

DRAM memory has evolved in terms of density, bandwidth and low power consumption but being volatile, it loses its data without power and therefore needs to be constantly refreshed.

MRAM, on the other hand, which retains data for long periods of time without having to be refreshed, is a good alternative to DRAM and could become the memory of choice for new applications. It is also superior to NAND flash memory in terms of performance (it enables write speeds about 1,000 times faster) and durability. Its advantages are the speed, the flow, the capacity, the non-volatility, the robustness but it offers only a weak resistance.

The study by SAMSUNG researchers

The research was led by Samsung Advanced Institute of Technology (SAIT) researcher Jung Seung-chul, lead author of the paper, SAIT Fellow Ham Don-hee, and SAIT Vice President of Technology Kim Sang-joon. . Developing a spin transfer torque magnetoresistive random access memory (MRAM) crossbar array is very complicated because of the low resistance of MRAM, which would cause large power consumption in a conventional crossbar array that uses sum current to analog multiply-accumulate operations.

To circumvent this difficulty, the researchers developed a 64×64 crossbar network based on MRAM cells whose architecture uses resistor sum for analog multiply-accumulate operations. This network has been integrated with the readout electronics in complementary 28-nanometer metal-oxide-semiconductor technology.

They then implemented a two-layer perceptron to classify 10,000 digits from the Modified National Institute of Standards and Technology with an accuracy of 93.23%. For a deeper eight-layer Visual Geometry Group-8 neural network with measured errors, the classification accuracy reached 98.86%.
A ten-layer neural network achieved 93.4% accuracy for face detection. Dr Seungchul Jung, lead author of the study said:

“Memory computing is like the brain in that in the brain, computing also occurs in the network of biological memories, or synapses, the points where neurons touch each other. In fact, while the computation performed by our MRAM array currently serves a different purpose than that performed by the brain, such a semiconductor memory array could in the future be used as a platform to mimic the brain by modeling the brain synapse. »

The researchers said that while this new MRAM chip can be used for in-memory computing, it can also serve as a platform for downloading biological neural networks.

Samsung said it wants to continue expanding its leadership in next-generation computing and AI semiconductors.

Sources of the article:
“A crossbar array of magnetoresistive memory devices for in-memory computing”Nature.

We wish to thank the writer of this write-up for this awesome material

In-memory computing: Samsung Electronics releases study on next-generation AI semiconductors with MRAM technology


Find here our social media profiles , as well as other pages related to it.https://www.ai-magazine.com/related-pages/