#USA #Supermicro #San Jose #NVIDIA RTX #GPU Optimization

Supermicro Expands Enterprise AI Offerings

Supermicro, a leading IT solution provider, has made a significant announcement at the GTC 2025 conference, unveiling an expanded portfolio of over 100 GPU-optimized systems. These systems specifically support the newly anticipated NVIDIA RTX PRO 6000 Blackwell Server Edition as well as the NVIDIA H200 NVL platform. This expansion represents a crucial step towards enhancing capabilities in artificial intelligence (AI), machine learning (ML), high performance computing (HPC), cloud infrastructure, and 5G Edge technologies.

Driving Innovation in AI and HPC

With an extensive array of form factors, Supermicro's GPU systems are adaptable to the most demanding data center requirements. They can be configured to accommodate up to 10 double-width GPUs or tailored low-power Intelligent Edge systems that optimize AI large language model (LLM) inference workloads in enterprise environments. This flexibility ensures businesses can effectively utilize accelerated computing power for various tasks, including visualization, graphics rendering, and virtualization.

Charles Liang, the CEO and chairman of Supermicro, emphasized the company’s leadership in the sector, stating, "Supermicro leads the industry with our extensive portfolio of application-optimized GPU servers designed for diverse business environments with rapid turnaround times." The integration of NVIDIA’s GPU technology not only enhances performance but also provides the necessary flexibility for customers eager to harness cutting-edge technologies.

Featuring the Latest NVIDIA Innovations

The new NVIDIA RTX PRO 6000 Blackwell Server Edition acts as a multipurpose GPU optimized for AI and graphical workloads. It showcases significant improvements over its predecessor, the NVIDIA L40S, integrating advanced GDDR7 memory, double the memory capacity, and supporting PCIe 5.0 for quicker GPU to CPU communication. Moreover, the introduction of Multi-Instance GPU (MIG) functionality allows a single GPU to be partitioned into up to four fully isolated instances, maximizing operational efficiency.

Supermicro's systems are not just about raw computational power; they also support NVIDIA SuperNICs like NVIDIA BlueField®-3 and NVIDIA ConnectX®-8, facilitating the best infrastructure scaling and GPU clustering through NVIDIA Quantum InfiniBand and NVIDIA Spectrum Ethernet technologies. Bob Pette, NVIDIA’s VP of Enterprise Platforms, highlighted this collaboration by describing the RTX PRO 6000 as the ultimate datacenter GPU for AI and visual computing, prepared to accelerate the most demanding workloads.

A Range of Supported Systems

Alongside the RTX PRO 6000 Blackwell Server Edition, Supermicro plans to support various workstation versions of the new GPUs, including the NVIDIA RTX PRO 6000 Blackwell Workstation Edition and the more powerful Max-Q edition, aimed at professional workloads in AI processing, development, and content creation.

Among the systems supporting these GPUs include:

• - 5U PCIe GPU systems: Designed to support up to 10 GPUs in a single chassis with thermal optimization for diverse workloads including AI inference.
• - NVIDIA MGX™ systems: Modularity designed for industries requiring intensive computational capabilities, supporting up to four GPUs in 2U and eight in 4U.
• - 3U Edge-optimized systems: Compact enough for edge implementations, able to handle up to eight double-width GPUs.
• - SuperBlade® architecture: Focused on maximizing rack density, with configurations allowing for up to 120 GPUs per rack.

A Commitment to Sustainable Growth

Supermicro, headquartered in San Jose, California, is dedicated to providing comprehensive IT solutions optimal for a broad spectrum of applications. With in-house design and production capabilities in the US, Taiwan, and the Netherlands, the company is committed to reducing total cost of ownership (TCO) while minimizing environmental impacts through green computing initiatives. Supermicro’s award-winning Server Building Block Solutions® allow customers to tailor systems meticulously to their workloads by selecting from a rich variety of systems and components.

This significant expansion in the AI domain not only enhances Supermicro's product offerings but also enables businesses across industries to adopt and innovate in their AI capabilities, leading to smarter and more efficient technology solutions.

For more information, you can visit Supermicro's website.