Leading Liquid-Cooled AI Cluster: Doubling Compute Density, 256 NVIDIA H100/H200 GPUs, HBM3 / HBM3e Memory, 1:1 Networking to Each GPU.
Engineered for professionals who demand the best in performance, security, and design.
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 Al 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
Full specifications for the model
| SKU | SRS-48UGPU-AI-LCSU |
|---|---|
| Scalable Unit | 32-Node Scalable Unit |
| Overview | Fully integrated liquid-cooled 32-node cluster with 256 NVIDIA H100/H200 GPUs |
| Compute Fabric Leaf | 8x SSE-MQM9700-NS2F, 64-port NVIDIA Quantum-2 InfiniBand 400G NDR, 32 OSFP ports switch |
| Compute Fabric Spine | 4x SSE-MQM9700-NS2F, 64-port NVIDIA Quantum-2 InfiniBand 400G NDR, 32 OSFP ports switch |
| In Band Management Switch | 3x SSE-MSN4600-CS2FC 64-port 100GbE QSFP28, 2U switch |
| Out Of Band Management Switch | 2x SSE-G3748R-SMIS, 48-port 1Gbps Ethernet ToR management switch 1x SSE-F3548SR, 48-port 10Gbps Ethernet ToR management switch |
| Rack And Pdu | 5x 48U 750mm x 1200mg PDU: 18x 415V 60A 3Ph |
| Liquid Cooling | 4x Supermicro 80kW capacity CDU with redundant PSU and dual hot-swap pumps |
Download product documentation and resources
Looking for more configuration options or custom setups? Our team can help you find the right solution based on your requirements.
Contact UsNeed multiple devices for your business or project? Contact us to discuss bulk pricing, availability, and delivery options.
Contact UsExplore other configurations and models that might suit your needs.
Recommended
#SYS-222GS-NB3OT-ALC
Liquid-Cooled 2 OU NVIDIA HGX™ B300 AI Cluster: 2-OU Liquid-cooled system with front I/O NICs, DPUs, storage, and management.
Recommended
#SuperCluster SRS-48UAC-B300SX
SuperCluster SRS-48UAC-B300SX: Full-Stack Reference Architecture, Up to 256x NVIDIA HGX™ B300 GPUs, NVIDIA Software Stack Compatibility, Plug-and-Play Integration.
Recommended
#SRS-48UDLC-4U8N-L1
Liquid-Cooled NVIDIA HGX B200 AI Cluster: Fully integrated liquid-cooled 32-node cluster with 256 NVIDIA B200v GPUs.
Our team of experts is ready to help you find the perfect solution for your business needs. Get personalized advice and competitive quotes.
We're here to help with any questions