Supermicro Expands NVIDIA Ampere Architecture-based GPU Product Line for Enterprise AI Including an Industry-First 5 petaFLOPS in a 4U Tier 1 AI Platform

Featured at NVIDIA GTC21, the New Generation of Supermicro NVIDIA-Certified Systems Deliver up to 4.8 TB/sec GPU Interconnect, a 1:1 GPU to Network Card Ratio, and Support for Air or Liquid Cooling Solutions

SAN JOSE, Calif., April 12, 2021 /PRNewswire/ -- Super Micro Computer, Inc. (Nasdaq: SMCI), a global leader in enterprise computing, storage, networking, and green computing technology, extends its market leadership in the growing areas of Artificial Intelligence (AI), Machine Learning (ML), High-Performance Computing (HPC), 3D-Graphics Visualization, and Desktop Virtualization. Supermicro has the broadest Tier 1 portfolio of systems that integrate state-of-the-art capabilities achieving 5 petaFLOPS of AI performance in a 4U form factor with the latest NVIDIA A100, NVIDIA A40, NVIDIA RTX A6000, and the new NVIDIA A30, NVIDIA A10, and NVIDIA A16 GPUs.

"Our collaboration with NVIDIA enables us to design a complete portfolio of GPU systems and platforms empowering maximum customer choice. Customers can exactly match their workload requirements to system architecture to achieve superior scale, performance, efficiency, and cost," said Charles Liang, president and CEO, Supermicro. "Supermicro offers multi-GPU optimized systems that deliver advanced solutions to AI, Deep Learning, HPC, and video streaming applications. Leveraging our building block solutions and our time-to-market advantage, we continue developing and releasing application-optimized solutions. Our flexible 2U 2-node GPU system with PCIe 4.0 is tuned for cloud gaming and social media and leads the industry with support for up to three double-width or six single-width GPUs per node."  

The breadth of Supermicro's systems includes the flagship Intel-based SYS-420GP-TNAR or the AMD-based AS -4124GO-NART(+), featuring the  4U Server powered by the NVIDIA HGX A100 8-GPU board, with the option of either 40GB (HBM2) or 80GB (HBM2e) of memory on each GPU. These systems are currently shipping worldwide.

"From AI-powered drug discovery to designing advanced, efficient factories, accelerated computing is essential to powering the most innovative work in the world today," said Justin Boitano, vice president and general manager, Enterprise and Edge Computing at NVIDIA. "Supermicro's NVIDIA-Certified Systems provide customers with a broad range of servers built to deliver top performance for AI, graphics and simulation workloads across the data center, to the cloud, to the edge."