AI computing power server
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 start-up SPAN wants to bundle AI computing power decentrally in private households. A piece of data center: The servers from SPAN are to be housed in a white box on the house wall, which – networked with other boxes – will. 2 AI data center racks draw 60+ kW each, compared to 5-10 kW for standard server racks. This 6-12x density difference is why AI facilities require entirely different power infrastructure, liquid cooling, and grid connections than conventional data centers. In collaboration with NVIDIA, Infineon will develop the next generation of power systems based on a new architecture with centralized power generation through 800V high-voltage direct current. Despite this, rack space and PSU form factors will remain unchanged, pressuring PSU vendors to achieve higher power density.
Read More