MUNICH, Germany and SANTA CLARA, CA, May 22, 2025 – Infineon Technologies AG, in collaboration with NVIDIA, is developing a new power system architecture for data centers that uses centralized 800 V high-voltage direct current (HVDC) distribution. This architecture improves energy efficiency by enabling direct power conversion at the AI chip (Graphic Processing Unit, GPU) within the server board. Infineon is leveraging its capabilities in power conversion across silicon (Si), silicon carbide (SiC), and gallium nitride (GaN) technologies to support the transition from grid to core in HVDC-based systems.
The shift to advanced power architectures are essential as data centers scale beyond 100,000 GPUs. Power demands are expected to exceed one megawatt per IT rack by the end of the decade, highlighting the need for reliable distribution systems. The adoption of HVDC architecture, combined with high-density multiphase solutions, offers a path to meet these requirements. This approach is expected to influence the design of next-gen components and establish new benchmarks for power delivery in high-performance data centers.
“Infineon is driving innovation in artificial intelligence,” said Adam White, division president power & sensor systems at Infineon. “The combination of Infineon’s application and system know-how in powering AI from grid to core, combined with NVIDIA’s world-leading expertise in accelerated computing, paves the way for a new standard for power architecture in AI data centers to enable faster, more efficient and scalable AI infrastructure.”
“The new 800V HVDC system architecture delivers high reliability, energy-efficient power distribution across the data center,” said Gabriele Gorla, vice president of system engineering at NVIDIA. “Through this innovative approach, NVIDIA is able to optimize the energy consumption of our advanced AI infrastructure, which supports our commitment to sustainability while also delivering the performance and scalability required for the next generation of AI workloads.”
At present, AI data centers use a decentralized power system. In this setup, each AI chip gets power from different power supply units (PSU). Looking ahead, designs will shift to a centralized system where it will use the available space in a server rack effectively. It calls for power semiconductors that work with fewer power conversion steps and support higher distribution voltages.
Infineon believes centralized HVDC setups will need at least as many power semiconductors as today’s AC systems, if not more. As HVDC designs expand, Infineon will keep supporting popular DCDC multiphase systems and intermediate architectures. Their product range span all semiconductor materials for end-to-end power delivery in data center and AI workloads.
Source: Infineon
About Infineon
Infineon Technologies AG, established in 1999 as a spin-off from Siemens AG, is a German semiconductor manufacturer headquartered in Neubiberg, near Munich. The company specializes in designing, developing, and manufacturing a wide range of semiconductor products, including power semiconductors, microcontrollers, sensors, and application-specific integrated circuits (ASICs). These products serve various industries, such as automotive, industrial power control, power and sensor systems, and connected secure systems. Infineon’s solutions are integral to applications like automotive electronics, renewable energy systems, industrial drives, consumer electronics, and security systems. As of the fiscal year ending September 2023, Infineon reported revenues exceeding €16 billion and employed approximately 58,600 people worldwide. The company operates 69 research and development locations and 17 manufacturing sites globally, underscoring its commitment to innovation and technological advancement.
About NVIDIA
NVIDIA Corporation, based in Santa Clara, CA, is a U.S. technology company specializing in the design and production of graphics processing units (GPUs). Its hardware and software solutions support a range of applications and simulation. Operating for over 30 years, NVIDIA has seen strong financial growth, reporting $39.3 billion in revenue and $22.1 billion in net income for the fiscal quarter ending January 2025. Its headquarters are designed to promote a flat organizational structure that encourages open communication and collaboration between leadership and staff across industries. In gaming, its GPUs power high-performance visual rendering. In artificial intelligence and high-performance computing, NVIDIA provides the infrastructure needed for training and deploying large-scale models. The company also contributes to the automotive sector with systems for autonomous driving and supports robotics with tools for AI-based perception.