xAI’s Colossus supercomputer cluster achieves massive scale using the NVIDIA Spectrum-X Ethernet networking platform to connect 100,000 NVIDIA Hopper Tensor Core GPUs.
The most powerful liquid-cooled AI SuperCluster is designed to take xAI’s Grok AI to another era. Supermicro accelerates the industry’s transition to liquid-cooled AI data centers to deliver a new paradigm of energy efficiency for the rapidly-heightened demand of energy and power requirements of today’s AI infrastructure. With extensive experience deploying large-scale direct-to-chip (DLC) liquid-cooled AI systems, Supermicro’s leading liquid cooling technology is powering the most ambitious AI infrastructure projects in the world.
Featuring ServeTheHome
One of the most powerful liquid-cooled AI SuperClusters is designed to take xAI’s Grok AI to the next level. Supermicro is developing leading liquid cooling rack technology powering the world's most ambitious AI infrastructure projects.
Inside the 100K GPU xAI Colossus that Supermicro Helped Build for Elon Musk
Authored by Patrick Kennedy, ServeTheHome
Leading Liquid-Cooled AI Cluster
With 32 NVIDIA HGX H100/H200 8-GPU, 4U Liquid-cooled Systems (256 GPUs) in 5 Racks
- Doubling compute density through Supermicro’s custom liquid-cooling solution with up to 40% reduction in electricity cost for data center
- 256 NVIDIA H100/H200 GPUs in one scalable unit
- 20TB of HBM3 with H100 or 36TB of HBM3e with H200 in one scalable unit
- 1:1 networking to each GPU to enable NVIDIA GPUDirect RDMA and Storage for training large language model with up to trillions of parameters
- Customizable AI data pipeline storage fabric with industry leading parallel file system options
- Supports NVIDIA Quantum-2 InfiniBand and Spectrum-X Ethernet platform
- Certified for NVIDIA AI Enterprise Platform including NVIDIA NIM microservices
Compute Node
