Supermicro, Inc., a leading IT solution provider for AI, Cloud, Storage, and 5G/Edge, has introduced a new addition to its SuperCluster lineup of plug-and-play AI infrastructure solutions for the NVIDIA Omniverse platform. This new SuperCluster is designed to deliver high-performance generative AI-enhanced 3D workflows at an enterprise scale, featuring the latest Supermicro NVIDIA OVX systems that allow for easy scalability as workloads increase.
"Supermicro has been at the forefront of developing GPU-optimized products for 3D graphics, application acceleration, and now AI," said Charles Liang, president and CEO of Supermicro. "With the growing demand for AI capabilities, enterprises are looking for computing infrastructure that can combine all these features into a single package. Our SuperCluster for NVIDIA Omniverse includes fully interconnected 4U PCIe GPU NVIDIA-Certified Systems, with up to 256 NVIDIA L40S PCIe GPUs per scalable unit, to deliver high performance across the Omniverse platform."
The SuperCluster for NVIDIA Omniverse expands Supermicro's range of application-optimized AI rack solutions, catering to professionals with compute-intensive 3D workflows in fields such as product design and industrial digital twins. Generative AI has revolutionized 3D workflows and is driving a new wave of applications, making SuperCluster for NVIDIA Omniverse a valuable tool for simplifying the deployment of scale-out infrastructure for multi-workload needs.
Supermicro's NVIDIA OVX systems form the foundation of the SuperCluster's compute power, hosting up to 8 of the latest NVIDIA PCIe GPUs per system node. These systems are powered by 4x 2700 W Titanium Level PSUs and feature high-airflow chassis for stability under high-utilization scenarios. The inclusion of NVIDIA BlueField-3 SuperNICs or NVIDIA ConnectX-7 NICs per system provides high-speed networking capabilities.
Supermicro's 4U PCIe GPU systems are certified by NVIDIA for use with NVIDIA Omniverse, ensuring performance, reliability, scalability, and security. The SuperCluster for NVIDIA Omniverse offers a fully interconnected infrastructure solution that provides access to GPU computing for developers, artists, engineers, and others, with seamless access to virtual GPUs or full system nodes.
Customers can choose from a range of deployment sizes and options, from a single rack to an enterprise scale, with the ability to customize the configuration based on their specific requirements. The SuperCluster for NVIDIA Omniverse Scalable Unit includes a variety of components such as system nodes, GPUs, control nodes, Ethernet fabric switches, and NICs, allowing for flexible scalability.
Supermicro's SuperCluster for NVIDIA Omniverse is a highly customizable solution that can be tailored to fit the needs of any organization, offering a range of deployment sizes and options to accommodate various workloads and applications.
