#USA #Supermicro #San Jose #AI Systems #NVIDIA Vera

Supermicro's Game-Changing DCBBS® Systems

Supermicro has announced the launch of its latest lineup of advanced computing systems, which leverage the cutting-edge NVIDIA Vera Rubin platform. These new systems, including the NVIDIA Vera Rubin NVL72 and the HGX Rubin NVL8, are designed with the DCBBS (Data Center Building Block Solutions) liquid-cooling architecture, revolutionizing the way data centers can support AI-powered applications.

Targeting Industry Evolution

As the demands for AI processing escalate, organizations are facing a critical need for enhanced data center infrastructure capable of supporting extensive AI workloads. The NVIDIA Vera Rubin NVL72 system is positioned to deliver an astonishing throughput of up to 10 times the performance per watt, offering a significant advantage over previous NVIDIA Blackwell solutions. This leap in technology not only promises efficiency but also aims to minimize operational costs associated with high-performance computing.

Charles Liang, CEO of Supermicro, emphasized the pivotal moment we find ourselves in, stating, "We are entering a new era where every organization requires an AI factory to win in the marketplace, as the demand for inference workloads is reshaping what data center infrastructure must deliver."

Features of the NVIDIA Vera Rubin NVL72

The Vera Rubin NVL72 system operates as a robust accelerator at a rack-scale, integrating six co-designed components such as the Rubin GPU and Vera CPU. This cohesive design is targeted to provide up to 3.6 Exaflops of inference capabilities which is essential for handling vast amounts of data efficiently. In addition, organizations can now implement this system with considerably lower costs, operating at just one-tenth the token cost compared to its predecessor models.

The Versatile HGX Rubin NVL8 System

Not only does Supermicro aim for unmatched processing capabilities, but they also emphasize flexibility in their designs. The 2U HGX Rubin NVL8 system stands out as one of the most versatile platforms available, supporting varied CPU configurations. It allows the pairing of NVIDIA Vera CPUs with the latest AMD and Intel processors, enabling a tailored fit for diverse workloads, be it for AI training or high-performance computing tasks.

Cutting-Edge Cooling Technology

A key feature of the DCBBS refresh is the sophisticated liquid-cooling technology that is integral to the design of these upcoming systems. By fully liquid-cooling the NVIDIA Vera Rubin NVL72, Supermicro ensures that thermal management will not be a bottleneck, fostering reliability and performance. This innovative cooling approach is crafted to respond to the continuous growth in power and thermal requirements of massive AI deployments.

Future-Ready AI Infrastructure

Supermicro is also developing a new Context Memory Storage platform designed to complement these AI systems. This platform, powered by NVIDIA’s latest processing technologies, aims to extend the capacity of GPU key-value stores efficiently, optimizing performance for extensive AI inference applications.

Hands-On Experience at GTC San Jose 2026

For industry professionals looking to explore these advancements firsthand, Supermicro will showcase the Vera Rubin series alongside their existing Blackwell solutions during the upcoming GTC in San Jose. Attendees can meet with Supermicro experts to discuss procurement options and the future roadmap of these innovative solutions designed for AI infrastructure.

Supermicro's Commitment to Innovation

Founded in San Jose, California, Supermicro remains a strong advocate for pushing the boundaries of IT solutions that meet today’s needs while anticipating future demands. Their commitment to green computing and energy efficiency plays a vital role in their strategic development, providing customers with cutting-edge technology that is not just powerful, but also sustainable.

In summary, with the introduction of the NVIDIA Vera Rubin NVL72 and HGX Rubin NVL8 systems equipped with DCBBS technology, Supermicro is setting a new standard for AI infrastructure, preparing organizations to leap into the future of computing while managing costs and complexities effectively.