Supermicro, an IT solutions provider for AI, Cloud, Storage and 5G/Edge, has unveiled a plug-and-play SuperCluster for the NVIDIA Omniverse™ platform. This powerful rack solution offers up to 256 of the most advanced NVIDIA PCIe GPUs in one scalable unit, optimised for large-scale deployments. Designed to maximise performance for 3D & AI workloads, the SuperCluster simplifies the deployment of scale-out infrastructure for multi-workload needs. It’s equipped with Supermicro NVIDIA OVX™ systems, providing enterprises with a high-performance, flexible, and resource-optimised solution for generative AI-enhanced 3D workflows at an enterprise scale.
- Supermicro launches a new SuperCluster for NVIDIA Omniverse, offering scalable performance and flexibility for 3D & AI workloads.
- The SuperCluster features up to 256 NVIDIA PCIe GPUs per scalable unit, optimised for GenAI 3D workflows at enterprise scale.
- Supermicro’s solution provides fully interconnected infrastructure for developers, artists and engineers to access high-level GPU computing, customisable from a single rack to enterprise scale.
Unleashing the Power of AI with Supermicro’s SuperCluster for NVIDIA Omniverse™
Supermicro, a leading IT solution provider, has launched a new entrant in its SuperCluster suite of AI infrastructure solutions, specifically designed for the NVIDIA Omniverse™ platform. The aim? To bolster high-performance, AI-aided 3D workflows at an enterprise scale. With this new SuperCluster, Supermicro’s NVIDIA OVX™ systems are at the forefront, providing businesses with the ability to effortlessly scale as their workloads expand.
From 3D Graphics to AI – The Supermicro Evolution
Supermicro has been a prominent player in the development of GPU-optimised products, initially focusing on 3D graphics and application acceleration. Now, the company is turning its attention to AI.
“With the rise of AI, enterprises are seeking computing infrastructure that combines all these capabilities into a single package. Supermicro’s SuperCluster features fully interconnected 4U PCIe GPU NVIDIA-Certified Systems™ for NVIDIA Omniverse, with up to 256 NVIDIA L40S PCIe GPUs per scalable unit. The system helps deliver high performance across the Omniverse platform, including generative AI integrations. By developing this SuperCluster for Omniverse, we’re not just offering a product; we’re providing a gateway to the future of application development and innovation,” said Charles Liang, president and CEO of Supermicro.
A New Era of Applications with Generative AI
The SuperCluster for NVIDIA Omniverse is designed to simplify the deployment of scale-out infrastructure for the multi-workload needs of 3D and AI. Supermicro NVIDIA OVX systems serve as the fundamental building block of the cluster’s computing power.
Each system node hosts up to 8 of the latest NVIDIA PCIe GPUs that deliver the combination of highest 3D graphics and virtual reality performance, plus exceptional generative AI performance via Tensor Cores and Transformer Engine support.
Maximising Performance with Supermicro’s SuperCluster
SuperCluster for NVIDIA Omniverse is a fully interconnected infrastructure solution that seeks to ensure developers and other professionals can access the highest level of GPU computing when necessary, with seamless access to virtual GPUs or bare-metal access to full system nodes.
The inclusion of the NVIDIA Spectrum™-X Ethernet platform, providing a 400Gb/s high-performance network fabric, allows for the pooling of GPU memory across system nodes, essential for training large AI models.
Final Thoughts
Supermicro’s SuperCluster for NVIDIA Omniverse™ signifies a monumental step towards the future of application development and innovation. This solution promises high performance, scalability and resource optimisation, all while simplifying the deployment process.
It’s clear that the rise of AI is reshaping the way businesses operate, and with the introduction of SuperCluster for NVIDIA Omniverse™, Supermicro is set to be at the forefront of this transformation. This launch shows how Supermicro is not just adapting to the changing landscape, but is actively shaping the future of AI-enhanced 3D workflows.
FAQ
Q: What is the Supermicro SuperCluster for NVIDIA Omniverse designed for?
A: The Supermicro SuperCluster for NVIDIA Omniverse is designed to deliver high-performance generative AI-enhanced 3D workflows at enterprise scale.
Q: How many NVIDIA PCIe GPUs can the Supermicro Rack Solution feature?
A: The Supermicro Rack Solution can feature up to 256 of the most advanced NVIDIA PCIe GPUs in one scalable unit to maximize performance for 3D & AI workloads.
Q: What kind of professionals can benefit from the SuperCluster for NVIDIA Omniverse?
A: A wide range of professionals, including those in product design and industrial digital twins, who rely on compute-intensive 3D workflows can benefit from the SuperCluster for NVIDIA Omniverse.
Q: What is the compute power foundation of the SuperCluster for NVIDIA Omniverse?
A: The Supermicro NVIDIA OVX systems serve as the foundational building block of the cluster’s compute power, with each system node hosting up to 8 of the latest NVIDIA PCIe GPUs.
Q: How is network connectivity ensured in the SuperCluster for NVIDIA Omniverse?
A: The system features up to four NVIDIA BlueField®-3 SuperNICs or four NVIDIA ConnectX®-7 NICs per system, providing multiple ports of 400Gb/s networking with high scalability and security.
Q: How are Supermicro’s 4U PCIe GPU systems certified for use with NVIDIA Omniverse?
A: Supermicro’s 4U PCIe GPU systems are certified by NVIDIA for use with NVIDIA Omniverse through a rigorous validation process that tests for performance, reliability, scalability and security.
Q: Can the SuperCluster for NVIDIA Omniverse be customised according to customer requirements?
A: Yes, the Supermicro SuperCluster for NVIDIA Omniverse can be deployed in a range of available sizes and options, with system nodes equipped with either 4 GPUs per system or 8 GPUs per system, and deployments ranging from a single rack to a scalable unit with 32 systems in 5 racks. Large deployments can be further expanded to build clusters of virtually any size.
