Modern AI GPUs consume 700W-1,100W each. An 8-GPU server can draw 10kW or more, creating facility challenges that traditional IT infrastructure never faced. Accurate planning prevents budget overruns and identifies. Most teams budgeting for AI inference focus on one number: the GPU hourly rate. It is clean, predictable, and easy to model. The electricity bill does not show up until the first month of on-premise or colocation operations, and by then the budget is already set. Data centres are facilities used to house servers, storage systems, networking equipment and associated components that are installed in racks and organised into rows. Today, a single NVIDIA GB200 NVL72 AI rack draws 132 kW — more than 16 times as much. Google's latest-generation TPU, Ironwood, is claimed to be 30× more energy-efficient than its first publicly available TPU.
[PDF Version]