At the AI Infra Summit held at the Santa Clara Convention Center, Ian Buck, NVIDIA’s vice president of hyperscale and high-performance computing, delivered a talk on AI factory efficiency. He addressed advancements in the Vera Rubin and DSX platforms, emphasizing energy efficiency through optimizing tokens per watt. The event drew more than 8,000 attendees this year, up from 3,500 the previous year.
Why it matters
Energy efficiency is a growing concern for large-scale AI infrastructure. Framing performance in terms of tokens per watt reflects an industry focus on balancing compute output against power consumption in AI factories.
Who should care
Operators of AI infrastructure and organizations tracking NVIDIA’s hardware roadmap may find the Vera Rubin and DSX platform developments relevant.