The broadest system portfolio, from cloud to edge, with support for NVIDIA HGX H100 systems, L40 and L4 GPUs, and OVX 3.0 systems
SAN JOSE, California, March 22, 2023 /PRNewswire/ — Supermicro, Inc. (NASDAQ:SMCI), a comprehensive provider of AI/ML, cloud, storage and 5G/edge computing solutions, today announced that it has started shipping its new high-end GPU servers with the latest NVIDIA HGX H100 8- system has started. Supermicro servers integrate the new NVIDIA L4 Tensor Core GPU into a wide range of application-optimized servers, from the edge to the data center.
“Supermicro offers the industry’s broadest portfolio of GPU systems, including 8U, 6U, 5U, 4U, 2U, and 1U servers, as well as SuperBlade workstations and systems supporting the full range of new NVIDIA H100 GPUs said Charles Liang, President and CEO of Supermicro. “With our new NVIDIA HGX H100 Delta-Next server, customers can expect a 9x performance increase over the previous generation for AI training applications. Our GPU servers feature an innovative airflow design that reduces fan speed, lowers noise levels, and consumes less power, reducing total cost of ownership (TCO). In addition, we offer complete rack-scale liquid cooling options for customers looking to future-proof their data centers. »
This new 8U GPU server, Supermicro’s most powerful, is now available in bulk. The new Supermicro 8U server is optimized for AI, DL, ML and HPC workloads and is powered by the NVIDIA HGX H100 8 GPU with the highest GPU-to-GPU communication with the most advanced NVIDIA NVLink® 4.0 technology and fast NVSwitch connections, and NVIDIA Quantum-2 InfiniBand and Spectrum-4 Ethernet networks to break down AI barriers at scale. In addition, Supermicro offers several performance-optimized GPU server configurations, including direct attach/single root/dual root CPUs to GPUs and front or rear I/O models with AC and DC power supplies in standard and OCP CC rack configurations. The Supermicro X13 SuperBlade® chassis accommodates 20 NVIDIA H100 Tensor Core PCIe GPUs or 40 NVIDIA L40 GPUs in an 8U chassis. Additionally, up to 10 NVIDIA H100 PCIe GPUs or 20 NVIDIA L4 Tensor Core GPUs can be used in a 6U chassis. These new systems offer optimized acceleration ideal for running NVIDIA AI Enterprise, the software layer of the NVIDIA AI platform.
Continue reading
Liquid cooling on these servers is also supported by many GPU servers. In addition, Supermicro announces a liquid-cooled AI development system (in tower or rack configuration) with two CPUs and four NVIDIA A100 Tensor Core GPUs that is ideal for office and home office environments and in departmental and corporate groups can be used.
Supermicro systems support the new NVIDIA L4 GPU, which multiplies acceleration and power efficiency over previous generations. The same applies to AI inference, video streaming, virtual desktops and graphics applications in enterprise, cloud and edge. Powered by NVIDIA’s AI platform and comprehensive approach, the L4 is optimized for large-scale inference across a wide range of AI applications such as recommendations, voice assistants, chatbots, research, and contact center automation to deliver the best personalized experiences. As NVIDIA’s most efficient accelerator for consumer servers, the L4 delivers up to 4x better AI performance, increased power efficiency, and more than 3x better video streaming capacity and efficiency with support for AV1 encode/decode. The L4 GPU’s versatility for inference and visualization, along with its small size, energy efficiency, one-slot, low profile, and low power consumption (72W) make it ideal for global deployments, including in peripheral areas.
“By arming Supermicro servers with the unprecedented performance of the new NVIDIA L4 Tensor Core GPU, customers can efficiently and sustainably accelerate their workloads,” said Dave Salvator, director of accelerated computing products at NVIDIA. “Optimized for consumer applications, the NVIDIA L4 features a low-profile form factor that operates in a low-power 72W envelope and enables new heights of AI performance and efficiency at the edge. »
Supermicro’s new accelerated PCIe solutions enable the creation of 3D worlds, digital twins, 3D simulation models and industrial metaverses. In addition to supporting previous generations of NVIDIA OVX™ systems, Supermicro offers an OVX 3.0 configuration with four NVIDIA L40 GPUs, two NVIDIA ConnectX®-7 SmartNICs, one NVIDIA BlueField®-3 DPU, and the latest NVIDIA Omniverse Enterprise™ software at.
To learn more about all of Supermicro’s new advanced GPU systems, visit: https://www.supermicro.com/en/accelerators/nvidia
Learn more about Supermicro at NVIDIA GTC 2023 – https://register.nvidia.com/events/widget/nvidia/gtcspring2023/sponsorcatalog/exhibitor/1564778120132001ghs2/?ncid=ref-spo-128510
About Super Micro Computers, Inc.
Supermicro (NASDAQ: SMCI) is a leading global provider of application-optimized computing solutions. Supermicro was founded and operates in San Jose, California. It strives to provide market-leading innovations for enterprise, cloud, AI and 5G telco/edge computing infrastructures. We are transforming into a comprehensive IT solution provider with server, AI, storage, IoT and switch systems, software and services while providing advanced motherboards, power supplies and high-volume chassis. Products are designed and manufactured in-house (in the United States, Taiwan and the Netherlands), leveraging and optimizing global operations for scale and efficiency to improve total cost of ownership and reduce environmental impact (green IT). Server Building Block Solutions® award-winning product portfolio enables customers to optimize their workloads and applications by choosing from a wide range of systems comprised of our flexible, reusable building blocks, which include a full range of form factors, processors, memory and Graphics processing support devices, storage, networking, power and cooling solutions (air conditioning, 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.
Other brands, names and trademarks are the property of their respective owners.
Photo – https://mma.prnewswire.com/media/2036923/Comprehensive_Portfolio_of_Systems_from_the_Edge_to_Cloud.jpg
Logo – https://mma.prnewswire.com/media/1443241/3947729/Supermicro_Logo.jpg