industry news
Subscribe Now

CEA-Leti & Stanford Target Edge-AI Apps with Breakthrough Memory Cell

Paper at ISSCC 2019 Presents Proof-of-Concept Multi-Bit Chip That Overcomes NVM’s Read/Write, Latency and Integration Challenges

SAN FRANCISCO – Feb. 20, 2019 – Researchers at CEA-Leti and Stanford University have developed the world’s first circuit integrating multiple-bit non-volatile memory (NVM) technology called Resistive RAM (RRAM) with silicon computing units, as well as new memory resiliency features that provide 2.3-times the capacity of existing RRAM. Target applications include energy-efficient, smart-sensor nodes to support artificial intelligence on the Internet of Things, or “edge AI”.

The proof-of-concept chip has been validated for a wide variety of applications (machine learning, control, security). Designed by a Stanford team led by Professors Subhasish Mitra and H.-S. Philip Wong and realized in CEA-Leti’s cleanroom in Grenoble, France, the chip monolithically integrates two heterogeneous technologies: 18 kilobytes (KB) of on-chip RRAM on top of commercial 130nm silicon CMOS with a 16-bit general-purpose microcontroller core with 8KB of SRAM.

The new chip delivers 10-times better energy efficiency (at similar speed) versus standard embedded FLASH, thanks to its low operation energy, as well as ultra-fast and energy-efficient transitions from on mode to off mode and vice versa. To save energy, smart-sensor nodes must turn themselves off. Non-volatility, which enables memories to retain data when power is off, is thus becoming an essential on-chip memory characteristic for edge nodes. The design of 2.3 bits/cell RRAM enables higher memory density (NVM dense integration) yielding better application results: 2.3x better neural network inference accuracy, for example, compared to a 1-bit/cell equivalent memory.

The technology was presented on Feb. 19, at the International Solid-State Circuits Conference (ISSCC) 2019 in San Francisco in a paper titled, “A 43pJ/Cycle Non-Volatile Microcontroller with 4.7μs Shutdown/Wake-up Integrating 2.3-bit/Cell Resistive RAM and Resilience Techniques”.

But NVM technologies (RRAM and others) suffer from write failures. Such write failures have catastrophic impact at the application level and significantly diminish the usefulness of NVM such as RRAM. The CEA-Leti and Stanford team created a new technique called ENDURER that overcomes this major challenge. This gives the chip a 10-year functional lifetime when continuously running inference with the Modified National Institute of Standards and Technology (MNIST) database, for example.

“The Stanford/CEA-Leti team demonstrated a complete chip that stores multiple bits per on-chip RRAM cell. Stored information is correctly processed when compared with previous demonstrations using standalone RRAM or a few cells in a RAM array,” said Thomas Ernst, Leti’s chief scientist for silicon components and technologies. “This multi-bit storage improves the accuracy of neural network inference, a vital component of AI.”

Mitra said the chip demonstrates several industry firsts for RRAM technology. These include new algorithms that achieve multiple bits-per-cell RRAM at the full memory level, new techniques that exploit RRAM features as well as application characteristics to demonstrate the effectiveness of multiple bits-per-cell RRAM at the computing system level, and new resilience techniques that achieve a useful lifetime for RRAM-based computing systems.

“This is only possible with a unique team with end-to-end expertise across technology, circuits, architecture, and applications,” he said. “The Stanford SystemX Alliance and the Carnot Chair of Excellence in NanoSystems at CEA-Leti enabled such a unique collaboration.”

About CEA-Leti (France)

Leti, a technology research institute at CEA Tech, is a global leader in miniaturization technologies enabling smart, energy-efficient and secure solutions for industry. Founded in 1967, Leti pioneers micro-& nanotechnologies, tailoring differentiating applicative solutions for global companies, SMEs and startups. CEA-Leti tackles critical challenges in healthcare; Leti’s multidisciplinary teams deliver solid expertise, leveraging world-class pre-industrialization facilities. With a staff of more than 1,900, a portfolio of 2,700 patents, 91,500 sq. ft. of cleanroom space and a clear IP policy, the institute is based in Grenoble, France, and has offices in Silicon Valley and Tokyo. CEA-Leti has launched 60 startups and is a member of the Carnot Institutes network. This year, the institute celebrates its 50th anniversary. Follow us on www.leti-cea.com and @CEA_Leti.

CEA Tech is the technology research branch of the French Alternative Energies and Atomic Energy Commission (CEA), a key player in innovative R&D, defence & security, nuclear energy, technological research for industry and fundamental science, identified by Thomson Reuters as the second most innovative research organization in the world. CEA Tech leverages a unique innovation-driven culture and unrivalled expertise to develop and disseminate new technologies for industry, helping to create high-end products and provide a competitive edge.

Leave a Reply

featured blogs
Nov 13, 2019
At the third stroke of midnight on 30 September 2019, Australia's talking clock fell silent....
Nov 13, 2019
By Elven Huang – Mentor, A Siemens Business SRAM debugging at advanced nodes is challenging. With pattern matching and similarity checking, Calibre tools enable designers to more quickly and precisely locate SRAM modification errors and determine the correct fix. Static...
Nov 13, 2019
Decisions, Decisions … I may be in the market for a new car in the near future. Unless you'€™ve got a strong preference (and most car buyers DO have a strong preference, IMO), choosing a vehicle is a series of trade-offs.  Fuel efficiency vs. horsepower. Functionali...
Nov 13, 2019
One of the big trends that has been happening somewhat below the radar is the growth of various forms of 3D packaging. I noted this at HOT CHIPS in summer, when a big percentage of the designs were... [[ Click on the title to access the full blog on the Cadence Community sit...
Nov 8, 2019
[From the last episode: we looked at the differences between computing at the edge and in the cloud.] We'€™ve looked at the differences between MCUs and SoCs, but the one major thing that they have in common is that they have a CPU. Now'€¦ anyone can define their own CPU ...