Enabling AI at Scale with 1.6T and Beyond Energy Efficient DSP, LPO, LRO, and CPO Optics
Open Compute Project via YouTube
Power BI Fundamentals - Create visualizations and dashboards from scratch
All Coursera Certificates 40% Off
Overview
Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore cutting-edge optical solutions for scaling AI workloads in data centers through this 21-minute conference presentation. Examine the evolution of data center interconnects beyond 1.6T bandwidth, focusing on energy-efficient approaches including DSP-based pluggables, Linear Receive Optics (LRO), Linear Pluggable Optics (LPO), and Co-Packaged Optics (CPO). Compare power consumption, performance metrics, and integration complexity across different optical technologies. Analyze DR8 optics and their variants including 2xVR4, VR8, and 2xFR4 configurations for various deployment scenarios. Discover developments in co-packaged modules, external laser sources (ELS), and fiber shuffle solutions that enable scalable, low-power architectures. Learn how to select optimal optical solutions based on specific system requirements and understand the trade-offs between different approaches. Gain insights into building energy-efficient, high-performance infrastructure capable of supporting AI workloads at scale, with practical guidance for system designers and data center architects planning next-generation optical interconnect deployments.
Syllabus
Enabling AI at Scale with 1 6T and Beyond Energy Efficient DSP, LPO LRO, and CPO Optics prese
Taught by
Open Compute Project