Supermicro Introduces Advanced AI and HPC Infrastructure Solutions

Supermicro Redefines AI and HPC Infrastructure

Super Micro Computer, Inc. has recently launched an advanced end-to-end solution designed for high-performance computing (HPC) and artificial intelligence (AI) infrastructures, showcasing a flawless convergence of traditional computing and modern AI methodologies. This innovative offering is centered around the NVIDIA Vera Rubin NVL4 platform, which was unveiled during the International Supercomputing Conference (ISC) 2026.

The versatility of Supermicro's new infrastructure lies in its capacity to support up to 1,152 NVIDIA Rubin GPUs and 576 NVIDIA Vera CPUs, all housed in state-of-the-art liquid-cooled racks. This scalable unit, performing at an impressive 3.2 MW, not only promises unparalleled processing power but also heralds the future of AI workloads and FP64 simulations, allowing researchers to harness its capabilities effectively across varied scales.

Advanced Cooling Solutions

One of the standout features of this solution is the Direct Liquid Cooling (DLC-2) technology, which efficiently manages up to 362 kW per rack with multiple in-row cooling distribution units included per scalable unit. The use of copper cold plates and a specially developed coolant, the SMC PG25-A, ensures thermal and chemical stability. Such robust cooling mechanisms are crucial in a world where computational requirements are continuously evolving, particularly in fields such as climate research and pharmaceutical development.

Accelerating AI Deployment

Supermicro's DCBBS (Data Center Building Block Solutions) framework targets reducing the time-to-deployment for organizations looking to modernize their AI and HPC infrastructures. With comprehensive site assessments and meticulous planning, Supermicro's team customizes each design proposal to fit the unique needs of its clients.

System integration begins long before the actual delivery, ensuring a seamless transition to operational status. This includes careful racking, stacking, and rigorous testing at multiple levels within Supermicro's global production facilities, preparing systems for installation in record time.

On-site integration services, often referred to as white-glove delivery, cover essential tasks such as rack positioning, power and cooling connections, as well as full system validation. The commitment to support extends post-installation, providing organizations with reliable response times for critical operations, ensuring maximum uptime.

Transforming Research Capabilities

As research institutions increasingly recognize the importance of converged computing approaches—where traditional simulations work alongside accelerated computing and AI technologies—Supermicro positions itself as a leader in this transformative movement. The NVIDIA Vera Rubin NVL4 platform is specifically tailored to accommodate this blend, helping scientists make significant breakthroughs across diverse domains. From energy solutions to medicine, the potential applications are vast and varied, enabling faster discoveries through enhanced computational power.

Charles Liang, President and CEO of Supermicro, emphasized the necessity for organizations to embrace cutting-edge infrastructure to foster future innovations: "The institutions that accelerate infrastructure deployment will drive the next generation of breakthroughs. Our DCBBS Blueprints for the NVIDIA Vera Rubin NVL4 empower research organizations to implement HPC and AI infrastructure at any scale with confidence."

In an era defined by technological advancements, Supermicro remains committed to offering adaptable AI infrastructures strategically constructed from validated components. This level of modularity ensures that organizations can implement custom solutions tailored to their precise needs, whether they require individual servers or comprehensive data center configurations.

Supermicro's prominence in this sector shines through their extensive portfolio, allowing organizations globally to deploy efficient, scalable, and eco-friendly AI data centers.

For further information on Supermicro’s DCBBS solutions and to explore their innovative offerings in AI and HPC infrastructure, interested parties are invited to visit Supermicro's official site.