https://cdn.gophotonics.com/news/jj_638724407856941368.jpg712370
A team of international researchers from UC Santa Barbara, University of Cagliari, University of Pittsburgh, and Tokyo Institute of Science has unveiled a groundbreaking photonic memory platform designed to overcome key challenges in modern computing. Published in Nature Photonics, their innovation uses a magneto-optical material called cerium-substituted yttrium iron garnet (YIG), which changes its optical properties in response to magnetic fields. By integrating tiny magnets into the system, the team developed a new class of magneto-optical memories that utilize light for data storage and processing, achieving exceptional speed and efficiency.
This advancement comes as the limitations of Moore’s Law slow the progress of traditional electronic circuits. Computing systems are struggling to meet the demands of data-intensive applications like artificial intelligence and machine learning. Photonic computing, which processes information using light instead of electricity, offers a promising alternative with lower energy consumption and reduced latency.
The newly developed magneto-optical memories are a significant leap forward. They are 100 times faster than existing photonic technologies, consume only 10% of the energy required by conventional methods, and can be reprogrammed for multiple tasks. Unlike current optical memories, which have a limited lifespan, these memories can be rewritten more than 2.3 billion times, effectively giving them an unlimited operational life.
Led by Paolo Pintus of UC Santa Barbara and supported by collaborators from institutions worldwide, this breakthrough marks a major milestone in photonic computing. It opens the door to creating faster, more efficient computing systems capable of meeting the growing technological demands of the modern world.
“These unique magneto-optical materials make it possible to use an external magnetic field to control the propagation of light through them,” Pontus 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 their findings could mark the beginning of a revolution in optical computing, paving the way for practical applications in the near future.
Click here to read the article titled, "Integrated non-reciprocal magneto-optics with ultra-high endurance for photonic in-memory computing."