Nvidia Concedes China AI Chip Market to Huawei, AMD Warms Up to Beijing
The landscape of the AI chip market in China is undergoing a significant shift. Nvidia, a dominant player, has acknowledged Huawei's strong ...
Navitas' GaNFast™ and GeneSiC™ power semiconductors are selected to support NVIDIA’s ‘Kyber’ rack-scale systems.
The 800V HVDC architecture reduces copper wire thickness by 45% compared to traditional 54V systems.
End-to-end power efficiency is improved by up to 5%, and maintenance costs are reduced by 70%.
Navitas' technology enables the creation of highly efficient power supplies, including an 8.5 kW AI data center power supply with 98% efficiency.
Why this matters:: This collaboration addresses the growing power demands of AI data centers, making them more efficient and sustainable. The reduction in copper usage and improved efficiency contribute to significant cost savings and environmental benefits.
Modern AI data centers require massive amounts of power, often reaching gigawatts. Traditional 54V in-rack power distribution systems are limited in their ability to handle these demands efficiently. NVIDIA's 800V HVDC architecture directly converts grid power to 800V at the data center perimeter, eliminating multiple conversion steps and maximizing efficiency and reliability.
Navitas Semiconductor's GaNFast™ and GeneSiC™ technologies play a crucial role in this architecture. These technologies enable higher switching speeds, better thermal performance, and longer lifespans compared to traditional silicon solutions. Navitas' GaNSafe™ power ICs integrate control, drive, sensing, and protection features, ensuring reliability and robustness.
The collaboration has already produced significant results, including the development of an 8.5 kW AI data center power supply with 98% efficiency and a 12 kW PSU presented at Computex. These advancements demonstrate the potential for further innovation in AI data center power delivery.
This shift towards 800V HVDC architecture represents a significant advancement in data center technology, addressing the limitations of existing systems and paving the way for more efficient and sustainable AI infrastructure. To read NVIDIA’s technical blog, please click here?ref=yanuki.com. For more information on Navitas’ AI roadmap, please visit here?ref=yanuki.com.
Q: What are the benefits of NVIDIA's 800V HVDC architecture?
The architecture improves power efficiency by up to 5%, reduces maintenance costs by 70%, and lowers cooling costs by directly connecting HVDC to IT racks.
Q: How does Navitas' technology contribute to this architecture?
Navitas provides GaNFast™ and GeneSiC™ power semiconductors that enable higher efficiency, faster switching speeds, and better thermal performance.
Q: What are some of Navitas' recent achievements in AI data center power supplies?
Navitas has released an 8.5 kW AI data center power supply with 98% efficiency and a 12 kW PSU.
NVIDIA and Navitas are collaborating to create more efficient AI data centers.
The 800V HVDC architecture reduces copper usage, improves power efficiency, and lowers maintenance costs.
Navitas' GaN and SiC technologies are key components in this advanced power delivery system.
This collaboration marks a significant step towards sustainable and cost-effective AI infrastructure.
Do you think this trend will last? Let us know!
Share this article with others who need to stay ahead of this trend!
The landscape of the AI chip market in China is undergoing a significant shift. Nvidia, a dominant player, has acknowledged Huawei's strong ...
Micron (MU) is experiencing a surge in stock value as Wall Street assesses the impact of artificial intelligence (AI) demand on memory chips...
Marvell Technology (MRVL) is under scrutiny as its stock price fluctuates amid AI chip market shifts, inflation concerns, and rising oil pri...
Broadcom (NASDAQ: AVGO) is undergoing a significant transformation, shifting its focus to custom AI chips. This strategic move is projected ...
⚠ Disclaimer: Yanuki provides article summaries and links for reference only. Yanuki does not endorse, verify, or guarantee the accuracy of third-party sources. Please review original sources and verify information independently. Managed by the Yanuki Data Engine. Full Disclaimer