Tuesday, 01 April,
10:45
–
11:45
Theater II
Session Description:
The panel "Lighting the Path to Exascale AI: Photonics in High-Performance Clusters" explores the state of the art and the technical and market-wise progress that photonics must make in enabling the next generation of AI systems. As we approach the exascale era of computing, traditional electronic interconnects face significant challenges in meeting the bandwidth, latency, and energy efficiency demands of massive AI clusters. The Open Compute Project brings together a panel of leading experts in photonics and high-performance computing to discuss the ground truth of building AI clusters and how optical communications and integrated circuits can overcome these bottlenecks. Panelists will examine emerging developments in silicon photonics, optical interconnects, and photonic neural networks, highlighting their potential to solve complex data movement between chiplets and AI nodes. The discussion will also cover the integration challenges of photonic technologies in data centers, the roadmap to photonics-enabled exascale AI, and the implications for future machine learning architectures. By illuminating the convergence of photonics and AI, this panel aims to provide insights into how light-based technologies pave the way for unprecedented AI capabilities and efficiency at scale.
Organized by: 