AI servers consume significantly more power than traditional IT equipment, primarily due to the use of GPUs and high-performance accelerators. Typical ranges include: • Traditional servers: 300–800 W per server • GPU servers: 2–10 kW per server • AI racks: 20–100+ kW per rackThe rack itself is deeper, typically 1200mm instead of the standard 1000mm, because GPU servers need more space for cooling hardware and power distribution. But the real difference isn't visible in the rack itself. It's in the liquid cooling manifolds running overhead, the coolant distribution. Where traditional server racks once operated at around 5–10 kW, modern AI environments are pushing far beyond that, often reaching 30 kW, 60 kW or even over 100 kW per rack. By 2028, racks are projected to reach 1 MW.
[PDF Version]