Supermicro expands NVIDIA Blackwell system portfolio with new direct liquid-cooled (DLC-2) systems, enhanced air-cooled models, and front I/O options for AI factory deployments.
Supermicro, Inc. announced the expansion of its NVIDIA Blackwell system portfolio with a new 4U DLC-2 liquid-cooled NVIDIA HGX B200 system ready for volume shipment and an air-cooled 8U front I/O system. These systems are tailored to large-scale AI training and cloud-scale inference workloads, and are designed to support coming NVIDIA HGX B300 platforms while allowing easy front I/O access, simplifying cabling, improving thermal efficiency and compute density, and reducing operational expenses. According to CEO Charles Liang, "Supermicro's DLC-2 enabled NVIDIA HGX B200 system leads our portfolio to achieve greater power savings and faster time to online for AI Factory deployments. Our Building Block architecture enables us to quickly deliver solutions exactly as our customers request. Supermicro's extensive portfolio now can offer precisely optimized NVIDIA Blackwell solutions to a diverse range of AI infrastructure environments, whether deploying into an air- or liquid-cooled facility."
Supermicro now offers one of the broadest portfolios of NVIDIA HGX B200 solutions with two new front I/O systems and six rear I/O systems, allowing customers to choose their most optimized CPUs, memory, networking, storage, and cooling configuration. The systems feature 8 high-performance 400G NVIDIA ConnectX-7 NICs and 2 NVIDIA Bluefield-3 DPUs moved to the front, along with support for NVIDIA Quantum-2 InfiniBand and Spectrum-X Ethernet platform. Each NVIDIA HGX B200 8-GPU configuration is connected via 5th Generation NVLink at 1.8TB/s, providing a combined total of 1.4TB of HBM3e GPU memory per system. All systems include upgraded memory expansion with 32 DIMM slots, and dual-socket CPUs supporting up to 350W Intel Xeon 6 6700 Series processors. NVIDIA's Blackwell platform delivers up to 15x faster real-time inference performance and 3x faster training for LLMs compared to the Hopper generation of GPUs.
The newly introduced 4U front I/O liquid-cooled system features dual-socket Intel Xeon 6700 Series processors with P-cores up to 350W and NVIDIA HGX B200 8-GPU configuration with 180GB HBM3e per GPU. The system supports 32 DIMMs with up to 8TB capacity at 5200MT/s or up to 4TB at 6400MT/s DDR5 RDIMM, plus 8 hot-swap E1.S NVMe storage drive bays and 2 M.2 NVMe boot drives. Network connectivity includes 8 single-port NVIDIA ConnectX-7 NICs or NVIDIA BlueField-3 SuperNICs and two dual-port NVIDIA BlueField-3 DPUs. Supermicro designed this liquid-cooled system for densely populated AI factories that can reach cluster sizes beyond thousands of nodes, delivering up to 40% data center power savings compared to air cooling.
The 8U front I/O air-cooled system shares the same front-accessible architecture and core specifications while providing a streamlined solution for AI factories without liquid-cooling infrastructure. It features a compact 8U form factor (compared to Supermicro's 10U system) with a reduced-height CPU tray while maintaining the full 6U-height GPU tray to maximize air-cooling performance. NVIDIA vice president Kaustubh Sanghani stated, "Advanced infrastructure is accelerating the AI industrial revolution for every industry. Based on the latest NVIDIA Blackwell architecture, Supermicro's new front I/O B200 systems equip enterprises to deploy and scale AI at unprecedented speed-delivering breakthrough innovation, efficiency, and operational excellence."