Presentation Information
[Plenary_Talk_2]Photonic Switching and Computing in Future AI and Data Systems
○S. J. Ben Yoo (Distinguished Professor of Electrical and Computer Engineering, University of California, Davis)
Ubiquitous deployment of AI methods in every sector of our daily lives is now driving explosive demands for more data to be collected, processed, and analyzed by our data systems with unprecedented speed, efficiency, and intelligence. The advances in electronics alone, even with the extension of Moore’s Law, evidently cannot keep up with the demands that are increasing at a 10x pace every year. By harnessing the unique properties of light, photonics offers a route to overcome the bottlenecks faced by electronic circuits, paving the way for unprecedented advancements in information processing, switching, and routing.
Photonic switching achieves the reconfigurable flow of information exploiting photonics' massive parallelism and bandwidth without necessitating high-speed electronics. Thus, photonic switching can offer extremely high reconfigurable bandwidths without substantially costing latency and power.
Photonic computing extends beyond switching to encompass the use of light for performing information processing. In addition to the massive parallelism, the unique capability to achieve matrix multiplications by simply utilizing tunable optical couplers in a mesh offers significant advantages in energy efficiency, throughput, and real-time data processing.
With new advances in silicon photonics and 3D photonic-electronic integration technologies now becoming available at 300 mm wafer-scale and beyond, photonic switching and computing represent the next frontier in the evolution of AI and data systems. By harnessing the power of light, these technologies promise to deliver unprecedented speed, efficiency, and scalability, transforming how we process and analyze information. We will discuss this new frontier and the future potential.
Bio: S. J. Ben Yoo is a Distinguished Professor of Electrical and Computer Engineering at the University of California at Davis (UC Davis). He is currently leading the 3D EPIC AI Project and the UC Davis part of Northwest-AI Microelectronics Commons Hub under the US CHIPS and Science Act, the ExPlor Project under the US Air Force Office of Scientific Research, the NanoHybrid project Photonic Processing Activity under US Department of Energy’s Micro Microelectronics Science Research Center, and the NaPSAC project under DARPA support. His research at UC Davis includes 2D/3D photonic integration for future computing, cognitive networks, communication, imaging, and navigation systems, micro/nano systems integration, and the future Internet. Prior to joining UC Davis in 1999, he was a Senior Research Scientist at Bellcore, leading technical efforts in integrated photonics, optical networking, and systems integration. His research activities at Bellcore included the next-generation Internet, reconfigurable multiwavelength optical networks (MONET), wavelength interchanging cross connects, wavelength converters, vertical-cavity lasers, and high-speed modulators. He led the MONET testbed experimentation efforts, and participated in ATD/MONET systems integration and a number of standardization activities. Prior to joining Bellcore in 1991, he conducted research on nonlinear optical processes in quantum wells, a four-wave-mixing study of relaxation mechanisms in dye molecules, and ultrafast diffusion-driven photodetectors at Stanford University (BS’84, MS’86, PhD’91, Stanford University). Prof. Yoo is Fellow of IEEE, OSA, NIAC and a recipient of the DARPA Award for Sustained Excellence, the Bellcore CEO Award, the Mid-Career Research Faculty Award (UC Davis), the Senior Research Faculty Award (UC Davis), and numerous best paper awards from IEEE, ACM, and Optica conferences.
Photonic switching achieves the reconfigurable flow of information exploiting photonics' massive parallelism and bandwidth without necessitating high-speed electronics. Thus, photonic switching can offer extremely high reconfigurable bandwidths without substantially costing latency and power.
Photonic computing extends beyond switching to encompass the use of light for performing information processing. In addition to the massive parallelism, the unique capability to achieve matrix multiplications by simply utilizing tunable optical couplers in a mesh offers significant advantages in energy efficiency, throughput, and real-time data processing.
With new advances in silicon photonics and 3D photonic-electronic integration technologies now becoming available at 300 mm wafer-scale and beyond, photonic switching and computing represent the next frontier in the evolution of AI and data systems. By harnessing the power of light, these technologies promise to deliver unprecedented speed, efficiency, and scalability, transforming how we process and analyze information. We will discuss this new frontier and the future potential.
Bio: S. J. Ben Yoo is a Distinguished Professor of Electrical and Computer Engineering at the University of California at Davis (UC Davis). He is currently leading the 3D EPIC AI Project and the UC Davis part of Northwest-AI Microelectronics Commons Hub under the US CHIPS and Science Act, the ExPlor Project under the US Air Force Office of Scientific Research, the NanoHybrid project Photonic Processing Activity under US Department of Energy’s Micro Microelectronics Science Research Center, and the NaPSAC project under DARPA support. His research at UC Davis includes 2D/3D photonic integration for future computing, cognitive networks, communication, imaging, and navigation systems, micro/nano systems integration, and the future Internet. Prior to joining UC Davis in 1999, he was a Senior Research Scientist at Bellcore, leading technical efforts in integrated photonics, optical networking, and systems integration. His research activities at Bellcore included the next-generation Internet, reconfigurable multiwavelength optical networks (MONET), wavelength interchanging cross connects, wavelength converters, vertical-cavity lasers, and high-speed modulators. He led the MONET testbed experimentation efforts, and participated in ATD/MONET systems integration and a number of standardization activities. Prior to joining Bellcore in 1991, he conducted research on nonlinear optical processes in quantum wells, a four-wave-mixing study of relaxation mechanisms in dye molecules, and ultrafast diffusion-driven photodetectors at Stanford University (BS’84, MS’86, PhD’91, Stanford University). Prof. Yoo is Fellow of IEEE, OSA, NIAC and a recipient of the DARPA Award for Sustained Excellence, the Bellcore CEO Award, the Mid-Career Research Faculty Award (UC Davis), the Senior Research Faculty Award (UC Davis), and numerous best paper awards from IEEE, ACM, and Optica conferences.
Comment
To browse or post comments, you must log in.Log in