Uploaded February 2026 | Updated November 2026, 1 day ago
Abstract
The rapid growth of AI workloads is driving a fundamental and non-negotiable shift in technology infrastructure, from facilities to networks. This presentation, led by experts from AHEAD, will explore how to meet these new demands. We'll begin by framing the unprecedented scale of AI/HPC environments, detailing the evolution away from traditional data center approaches. A core focus will be on the physical challenges: analyzing rising power consumption and the critical evolution of cooling technologies. Finally, we'll dive into the networking backbone, revealing the architectural changes taking place—not just in speed, but in scale, bandwidth, and controls—to successfully support the next generation of HPC/AI. Attend to understand the immediate infrastructure decisions you need to make to power the future of compute.
nanog.org/events/nanog-96/content/5612
Abstract
The rapid growth of AI workloads is driving a fundamental and non-negotiable shift in technology infrastructure, from facilities to networks. This presentation, led by experts from AHEAD, will explore how to meet these new demands. We'll begin by framing the unprecedented scale of AI/HPC environments, detailing the evolution away from traditional data center approaches. A core focus will be on the physical challenges: analyzing rising power consumption and the critical evolution of cooling technologies. Finally, we'll dive into the networking backbone, revealing the architectural changes taking place—not just in speed, but in scale, bandwidth, and controls—to successfully support the next generation of HPC/AI. Attend to understand the immediate infrastructure decisions you need to make to power the future of compute.
nanog.org/events/nanog-96/content/5612










