Scaling AI Infrastructure with Optical Interconnects
Josef Berger AVP Marketing - Marvell As AI networks scale up and scale out, meeting the demands for higher bandwidth, lower latency, and improved power efficiency is pushing the limits of traditional interconnect architectures—at rack, row, data center and inter-data center scale. This session will examine the range of approaches being applied to meet next-generation bandwidth, performance and deployment requirements in every domain.

▶︎
The AI Bandwidth Wall & Co-Packaged Optics

▶︎
Optical Opportunities for the AI Scale up Fabric

▶︎
Panel Progress in Optical Interconnects for AI Clusters

▶︎
Above the Cloud: Building Data Centers in Space - Richard Campbell - NDC Copenhagen 2026

▶︎
A Silicon-Proven Unified Low-Latency CXL Controller and Port-Based Routing Switch (ISCA 2026)

▶︎
WHY CO-PACKAGED OPTICS FAILED IN 2011 - AND WHY IT WON'T THIS TIME : IBM Europe

▶︎
Helios Is AMD’s First AI System To Rival Nvidia Vera Rubin — We Got An Exclusive, First Look

▶︎
The final filmed interview with Sir Alex Younger, the late MI6 chief | The Economist

▶︎
Enabling AI at Scale with 1 6T and Beyond Energy Efficient DSP, LPO LRO, and CPO Optics prese

▶︎
Optical Interconnects in Data Centers with Rob Stone

▶︎
A CPU Made of Atoms: IBM's Breakthrough 0.7nm Transistors

▶︎
Co-Packaged Optics: Why Traditional Pluggable Transceivers Are Struggling to Keep Up With AI

▶︎
Blue gradient background - screensaver, mood lighting, ambiance, TV art, focus, study

▶︎
Leading in the Age of AI: A Conversation with NVIDIA CEO Jensen Huang | Global Conference 2026

▶︎
The DRAM Crisis: 600% Price Increases by Micron, SK Hynix, & Samsung

▶︎
health shouldn't be a competition

▶︎
Scaling AI The Case for 100T Switching

▶︎
Is Fine-Tuning Still Needed? LLMs, RAG, & LoRA

▶︎
Android 17 sucks. So I put Linux on a phone.

▶︎
