For decades, Moore’s Law has provided a roadmap for computing performance, but traditional transistor scaling is beginning to encounter the limits of physics. At the same time, artificial intelligence and other data-intensive workloads are driving unprecedented demand for greater computing power. In this episode of The Tech Between Us, Raymond Yin sits down with Mo Steinman, senior vice president and general manager at Lightelligence, to examine how photonics could help move computing beyond the limits of conventional electronic architectures.
The conversation explores how silicon photonics uses light rather than electrical signals to perform computation and transfer data. Mo explains how waveguides, modulators, phase shifters, and the properties of light can enable optical arithmetic processing, while also providing higher-bandwidth connections between CPUs, GPUs, ASICs, memory, and other devices.
The episode also examines the role photonics could play in artificial intelligence, where optical processors can deliver high compute bandwidth with extremely low latency. Raymond and Mo discuss the evolution of optical interconnects, including near-package and co-packaged optics, and why bringing optical connections closer to compute, memory, and switching devices could help eliminate electrical bottlenecks in increasingly complex systems.
Looking ahead, they consider how photonics could scale across chips, systems, data centers, and even multiple data centers as computing architectures continue to grow. The ultimate opportunity may be a future in which electronic and optical technologies work together, each operating where it delivers the greatest advantage.
Listen now and discover how photonics could help extend computing performance beyond the traditional limits of Moore’s Law.
Learn more at: https://www.mouser.com/en/empowering-innovation/future-computing
