Supermicro, Inc. a Add up to IT Arrangement Supplier for AI, HPC, Cloud, Capacity, and 5G/Edge, nowadays declared the development of its NVIDIA Blackwell framework portfolio with the unused 4U DLC-2 liquid-cooled NVIDIA HGX B200 framework prepared for volume shipment and the presentation of an aircooled 8U front I/O framework. Custom-made to the most requesting large-scale AI preparing and cloud-scale induction workloads, Supermicro’s unused frameworks streamline the sending, administration, and upkeep of discuss- or liquid-cooled AI foundation and are planned to back coming NVIDIA HGX B300 stages, permitting simple front I/O get to, rearranging cabling, making strides warm effectiveness and compute thickness, and diminishing operational costs (OPEX).
“Supermicro’s DLC-2 empowered NVIDIA HGX B200 framework leads our portfolio to accomplish more noteworthy control reserve funds and quicker time to online for AI Manufacturing plant arrangements,” said Charles Liang, CEO and president, Supermicro. “Our Building Square design empowers us to rapidly convey arrangements precisely as our clients ask. Supermicro’s broad portfolio presently can offer absolutely optimized NVIDIA Blackwell arrangements to a assorted run of AI framework situations, whether conveying into an discuss- or liquid-cooled facility.”
For more data, if it’s not too much trouble visit https://www.supermicro.com/en/accelerators/nvidia
Supermicro’s DLC-2 speaks to the another era of Coordinate Fluid Cooling arrangements, designed to meet the raising requests of AI-optimized information centers. This comprehensive cooling design conveys noteworthy operational and fetched benefits for high-density computing environments.
Up to 40% information center control savings
Faster time-to-deployment and diminished time-to-online by giving an end-to-end information center scale fluid cooling solution
Up to 40% diminished water utilization with warm water cooling at an channel temperature of up to 45°C, lessening the need of chillers
Up to 98% framework warm capture by liquid-cooling CPUs, GPUs, DIMMs, PCIe switches, VRMs, control supplies, and more
Enabling calm information center operation at a commotion level as moo as 50dB
Supermicro presently offers one of the broadest portfolios of NVIDIA HGX B200 arrangements with the two unused front I/O frameworks and six raise I/O frameworks permitting clients to select their most optimized CPUs, memory, organizing, capacity, and cooling arrangement. The unused 4U and 8U front I/O NVIDIA HGX B200 frameworks are built upon demonstrated arrangements for large-scale AI preparing and deduction sending by tending to major torment focuses of arrangement, counting organizing, cabling, and thermals.
“Advanced foundation is quickening the AI mechanical insurgency for each industry,” said Kaustubh Sanghani, bad habit president of GPU item administration at NVIDIA. “Based on the most recent NVIDIA Blackwell engineering, Supermicro’s modern front I/O B200 frameworks prepare endeavors to send and scale AI at phenomenal speed—delivering breakthrough development, proficiency, and operational excellence.”
Modern AI information centers request tall adaptability that requires considerable node-to-node associations. The system’s 8 high-performance 400G NVIDIA ConnectX®-7 NICs and 2 NVIDIA Bluefield®-3 DPUs are moved to the front of the framework to permit for the arrangement of organizing cables, capacity drive narrows, and administration all from the cold path. The NVIDIA Quantum-2 InfiniBand and the Spectrum™-X Ethernet stage are completely backed to guarantee the highest-performing compute fabric.
In expansion to framework engineering changes, Supermicro has fine-tuned components to maximize proficiency, execution, and fetched reserve funds for AI information center workloads. Overhauled memory extension with 32 DIMM spaces conveys more noteworthy adaptability for framework memory arrangement, empowering large-capacity memory usage. Expansive framework memory complements the NVIDIA HGX B200’s HBM3e GPU memory by disposing of CPU-GPU bottlenecks, optimizing huge workload preparing, upgrading multi-job effectiveness in virtualized situations, and quickening information preprocessing.
All Supermicro 4U liquid-cooled frameworks and 8U or 10U air-cooled frameworks are optimized for NVIDIA HGX B200 8-GPU with each GPU associated by means of 5th Era NVLink® at 1.8TB/s, giving a combined add up to of 1.4TB of HBM3e GPU memory per framework. NVIDIA’s Blackwell stage conveys up to 15x speedier real-time induction execution and 3x quicker preparing for LLMs compared to the Container era of GPUs. The modern front I/O frameworks with dual-socket CPUs supporting up to 350W Intel® Xeon® 6 6700 Arrangement processors convey tall execution and productivity for a wide run of AI workloads.
The recently presented 4U front I/O liquid-cooled framework highlights front-accessible NICs, DPUs, capacity, and administration components. It utilizes dual-socket Intel® Xeon® 6700 Arrangement processors with P-cores up to 350W and NVIDIA HGX B200 8-GPU setup (180GB HBM3e per GPU). The framework bolsters 32 DIMMs with up to 8TB capacity at 5200MT/s or up to 4TB at 6400MT/s DDR5 RDIMM, furthermore 8 hot-swap E1.S NVMe capacity drive inlets and 2 M.2 NVMe boot drives. Organize network incorporates 8 single-port NVIDIA ConnectX®-7 NICs or NVIDIA BlueField®-3 SuperNICs and two dual-port NVIDIA BlueField®-3 DPUs.
Supermicro outlined this liquid-cooled framework as the building piece for thickly populated AI industrial facilities that can reach cluster sizes past thousands of hubs, conveying up to 40% information center control reserve funds compared to discuss cooling
The 8U front I/O air-cooled framework offers the same front-accessible engineering and center determinations whereas giving a streamlined arrangement for AI production lines without liquid-cooling framework. It highlights a compact 8U frame calculate (compared to Supermicro’s 10U framework) with a reduced-height CPU plate whereas keeping up the full 6U-height GPU plate to maximize air-cooling execution.
About Super Micro Computer, Inc.
Supermicro (NASDAQ: SMCI) is a global leader in Application-Optimized Total IT Solutions. Founded and operating in San Jose, California, Supermicro is committed to delivering first to market innovation for Enterprise, Cloud, AI, and 5G Telco/Edge IT Infrastructure. We are a Total IT Solutions provider with server, AI, storage, IoT, switch systems, software, and support services. Supermicro’s motherboard, power, and chassis design expertise further enable our development and production, enabling next generation innovation from cloud to edge for our global customers. Our products are designed and manufactured in-house (in the US, Taiwan, and the Netherlands), leveraging global operations for scale and efficiency and optimized to improve TCO and reduce environmental impact (Green Computing). The award-winning portfolio of Server Building Block Solutions® allows customers to optimize for their exact workload and application by selecting from a broad family of systems built from our flexible and reusable building blocks that support a comprehensive set of form factors, processors, memory, GPUs, storage, networking, power, and cooling solutions (air-conditioned, free air cooling or liquid cooling).
Supermicro, Server Building Block Solutions, and We Keep IT Green are trademarks and/or registered trademarks of Super Micro Computer, Inc.
About Super Micro Computer, Inc.
Supermicro (NASDAQ: SMCI) is a global leader in Application-Optimized Total IT Solutions. Founded and operating in San Jose, California, Supermicro is committed to delivering first to market innovation for Enterprise, Cloud, AI, and 5G Telco/Edge IT Infrastructure. We are a Total IT Solutions provider with server, AI, storage, IoT, switch systems, software, and support services. Supermicro’s motherboard, power, and chassis design expertise further enable our development and production, enabling next generation innovation from cloud to edge for our global customers. Our products are designed and manufactured in-house (in the US, Taiwan, and the Netherlands), leveraging global operations for scale and efficiency and optimized to improve TCO and reduce environmental impact (Green Computing). The award-winning portfolio of Server Building Block Solutions® allows customers to optimize for their exact workload and application by selecting from a broad family of systems built from our flexible and reusable building blocks that support a comprehensive set of form factors, processors, memory, GPUs, storage, networking, power, and cooling solutions (air-conditioned, free air cooling or liquid cooling)
Supermicro, Server Building Block Solutions, and We Keep IT Green are trademarks and/or registered trademarks of Super Micro Computer, Inc.
All other brands, names, and trademarks are the property of their respective owners.
