SC25: Supermicro Showcases Future of HPC Clusters and AI Infrastructure

Summary:

• In-booth demos will feature new Data Center Building Block Solutions (DCBBS) incorporating Nvidia GB300 NVL72 and Nvidia HGX B300 Systems
• Future-ready data centers are designed to drive energy efficiency, scalability, performance, and reduce time-to-online
• Advanced cooling products, including Rear Door Heat Exchangers and Sidecar Cooling Distribution Units will be on display

Super Micro Computer, Inc. (SMCI) will showcase its latest AI Factory, HPC, and liquid-cooled data center innovations at Supercomputing 2025 (SC25) in St. Louis, MO.

This broad portfolio, ranging from desktop workstations to rack-scale solutions, demonstrates Supermicro’s commitment to powering the next-gen of high-performance computing, scientific research, and enterprise AI deployments.

“Supermicro continues to lead the industry in delivering complete, next-generation infrastructure solutions in close collaboration with our technology partners,” said Charles Liang, president and CEO, Supermicro. “At SC25, we are demonstrating our high-performance DCBBS architecture, direct liquid cooling, and rack-scale innovations, which empower customers to deploy AI and HPC workloads faster, more efficiently, and more sustainably.”

The company will showcase new platforms designed to improve performance for both CPU- and GPU-bound workloads in large-scale HPC and AI environments.

Key highlights include:

• Nvidia GB300 NVL72 with Liquid Cooling – Rack-scale solution with Nvidia GB300 Grace Blackwell Superchips providing 72 Nvidia Blackwell Ultra GPUs and 36 Grace CPUs per rack with 279GB HBM3e per GPU
• 4U HGX B300 Server Liquid Cooled Rack with In-Rack CDU
• 1U Nvidia GB200 NVL4 Server (ARS-121GL-NB2B-LCC) – A high-density, liquid-cooled compute node purpose built for large-scale HPC and AI training.
• Super AI Station Based on Nvidia GB300 (ARS-511GD-NB-LCC) – An AI and HPC development platform integrated into a desktop workstation form factor.
• Liquid-cooled 8U 20-node and 6U 10-node SuperBlade – An advanced liquid cooled platform delivering maximum CPU and GPU density, supporting Intel Xeon 6900, 6700, and 6500 Series processors up to 500W.
• Liquid-cooled 2U FlexTwin multi-node system – An advanced liquid cooled (up to 95% heat capture) platform delivering maximum CPU computing density with four independent nodes, each equipped with the highest-performance dual socket CPUs supporting either AMD EPYC 9005 processors or Intel Xeon 6900 Series processors up to 500W.

DCBBS and Direct Liquid Cooling Innovations
Supermicro’s DCBBS integrates compute, storage, networking, and thermal management to simplify the deployment of complex AI and HPC infrastructure.

Key highlights include:

• Rear Door Heat Exchangers – Supporting cooling capacities of 50kW or 80kW
• Liquid-to-Air Sidecar CDUs (cooling distribution units)- Supporting cooling capacities up to 200kW, with no external infrastructure needed.
• Water Cooling and Dry Towers – Energy-efficient external towers which cool the liquid, in a closed loop design.

Optimized Product Families for HPC Workloads and AI Infrastructure
The company’s high-density, liquid-cooled systems address use cases across financial services, manufacturing, climate and weather modeling, oil and gas, and scientific research. Each distinct product family is designed with an optimized combination of density, performance, and efficiency.

• SuperBlade – The SuperBlade systems have been winning HPC customers worldwide for over 18 years. The latest-gen X14 SuperBlade systems offer maximum performance and the highest density with both CPU and GPU for the most demanding HPC and AI workloads. Both air-cooling and direct-to-chip liquid cooling can be supported. With integrated InfiniBand and Ethernet switches, SuperBlade is for HPC and AI applications. 
• FlexTwin– The firm’s FlexTwin architecture is purpose-built for HPC and is cost-efficient, designed to deliver maximum compute power and density in a multi-node configuration, with up to 24,576 performance cores in a 48U rack. Optimized for HPC and other compute-intensive workloads, each node features direct-to-chip liquid cooling to maximize efficiency and reduce CPU thermal throttling, enabling low latency front and rear I/O with a range of flexible networking options up to 400G/node.
• BigTwin – Versatile BigTwin is available as a 2U –4-Node or 2U 2-Node system. It shares power supplies and fans, which reduces power consumption. The BigTwin is available with the Xeon 6 processor.
• MicroBlade  – The 6U 40-node and 6U 20-node MicroBlade systems offer customers the highest density and a cost-effective single-socket x86 server solution. They have been used by leading semiconductor companies to design and develop ICs for over 10 years. MicroBlade systems support a wide range of CPUs including Xeon 6300, Xeon D, and AMD EPYC 4005 Series. The latest-gen MicroBlade systems can support up to 20 AMD EPYC 4005 Series CPUs and 20 GPUs in a 6U enclosure.
• MicroCloud – Industry-proven design, scalable up to 10 CPU nodes or up to 5 CPU + GPU nodes per chassis. With up to 10 server nodes in only 3U of rack space, customers can increase their computing density by over 3.3x compared to industry-standard 1U rackmount servers.
• Petascale Storage – Density-maximized all-flash storage systems optimized for scale-out and scale-up software-defined storage, with easy-to-deploy 1U and 2U form factors supporting industry standard EDSFF media.
• Workstation – Workstation performance and flexibility in a rackmount form factor, offering increased density and security for organizations looking to utilize centralized resources.

Resource:
Supermicro SC25 event webpage