100x Faster: Light-Powered Memory That’s Revolutionizing Computing

A new era in computing is emerging as researchers overcome the limitations of Moore’s Law through photonics.

This cutting-edge approach boosts processing speeds and slashes energy use, potentially revolutionizing AI and machine learning applications.

Photonics: A Potential Game-Changer

For decades, computer and smartphone circuits have steadily become smaller and more powerful, following the trend known as Moore’s Law. However, this era of consistent progress is nearing its end due to physical limits, such as the maximum number of transistors that can fit on a chip and the heat generated by densely packed components. As a result, the pace of performance improvements is slowing, even as the demand for computational power continues to grow with data-intensive technologies like artificial intelligence and machine learning.

To overcome these challenges, innovative solutions are required. One promising approach lies in photonics, which uses light instead of electricity to process information. Photonics offers significant advantages, including lower energy consumption and faster data transmission with reduced latency.

One of the most promising approaches is in-memory computing, which requires the use of photonic memories. Passing light signals through these memories makes it possible to perform operations nearly instantaneously. But solutions proposed for creating such memories have faced challenges such as low switching speeds and limited programmability.

Breakthrough in Photonic Technologies

Now, an international team of researchers has developed a groundbreaking photonic platform to overcome those limitations. Their findings were published in the journal Nature Photonics.

Working with UC Santa Barbara professor of electrical and computer engineering (ECE) John Bowers and ECE associate professor Galan Moody, project scientist Paolo Pintus, an assistant professor at the University of Cagliari, coordinated the project with University of Pittsburgh’s Nathan Youngblood, Institute of Science Tokyo professor Yuya Shoji, and Mario Dumont, who received his Ph.D. in the Bowers lab.

High-Speed, Efficient Photonic Memories

The researchers used a magneto-optical material, cerium-substituted yttrium iron garnet (YIG), the optical properties of which dynamically change in response to external magnetic fields. By employing tiny magnets to store data and control the propagation of light within the material, they pioneered a new class of magneto-optical memories. The innovative platform leverages light to perform calculations at significantly higher speeds and with much greater efficiency than can be achieved using traditional electronics.

This new type of memory has switching speeds 100 times faster than those of state-of-the-art photonic integrated technology. They consume about one-tenth the power, and they can be reprogrammed multiple times to perform different tasks. While current state-of-the-art optical memories have a limited lifespan and can be written up to 1,000 times, the team demonstrated that magneto-optical memories can be rewritten more than 2.3 billion times, equating to a potentially unlimited lifespan.

Revolutionizing Computing with Light

“These unique magneto-optical materials make it possible to use an external magnetic field to control the propagation of light through them,” Pintus said. “In this project, we use an electrical current to program micro-magnets and store data. The magnets control the propagation of light within the Ce:YIG material, allowing us to perform complex operations, such as matrix-vector multiplication, which lies at the core of any neural network.”

The authors believe that the findings could mark the beginning of a revolution in optical computing, paving the way for practical applications in the near future.

Reference: “Integrated non-reciprocal magneto-optics with ultra-high endurance for photonic in-memory computing” by Paolo Pintus, Mario Dumont, Vivswan Shah, Toshiya Murai, Yuya Shoji, Duanni Huang, Galan Moody, John E. Bowers and Nathan Youngblood, 23 October 2024, Nature Photonics.
DOI: 10.1038/s41566-024-01549-1

Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.