➀ Microsoft showcased its newest NVIDIA GB200 systems, highlighting the significant cooling requirements for these systems. The cooling solution, a large coolant distribution unit (CDU), occupies twice as much space as the GB200 compute rack. ➁ The systems are part of Microsoft Azure's next-gen NVIDIA training and inferencing platform, emphasizing the company's commitment to advanced AI models. ➂ The design of the cooling system is notably open, in contrast to other secretive organizations, showcasing Microsoft's confidence in its innovation.
Related Articles
- Frore's new LiquidJet coldplates are equipped to handle the spiralling power demands of future AI GPUs — built to handle up to 4.4Kw TDPs, solution could be deployed in power-hungry Feynman data centers19 days ago
- Microsoft deploys world's first 'supercomputer-scale' GB300 NVL72 Azure cluster — 4,608 GB300 GPUs linked together to form a single, unified accelerator capable of 92.1 exaFLOPS of FP4 inference24 days ago
- Microsoft Etches Complex Microfluidics Channels Directly into Silicon Chipsabout 1 month ago
- DIYer creates ‘RTX 5090 countermeasures’ with cabling spaghetti to mitigate 16-pin meltdowns — system automatically shuts down when the overcurrent protection is triggered3 months ago
- Future AI processors said to consume up to 15,360W — massive power draw will demand exotic immersion and embedded cooling tech5 months ago
- Crazed modder straps CPU cooler to Nvidia GTX 960 with a 3D-printed bracket, breaks 3DMark benchmark record5 months ago
- Prototype of a Particularly Sustainable and Energy-Autonomous E-Bike Terminal Developed at HKA7 months ago
- Enhancing Chitosan Films with Silanized Hexagonal Boron Nitride for Sustainable Applications7 months ago
- White Knight to save Shibaura7 months ago
- Ed Rides The Tariff Roller-Coaster7 months ago