NVIDIA's new GB200 AI server cabinets: leaks in liquid cooling system, emergency work to fix it

As an Amazon Associate, we earn from qualifying purchases. TweakTown may also earn commissions from other affiliate partners at no extra cost to you.

NVIDIA was on the eve of unleashing its next-generation GB200 Superchip inside of new AI servers, until rumors of a "flooding" occurred.

2

UDN is reporting that the key water-cooling components from "major international manufacturers have been shipped out" causing water leakage inside of the GB200 AI server cabinet. The outlet reports that Taiwanese manufacturers like Shuanghong and Qihong are expected to come on-board for emergency rescue and receive transfer orders.

NVIDIA's issues with its new GB200-powered AI server cabinets seems to be coming from outsourcing some of the water cooling components, including the branch pipe, quick connector, and hose. If an issue happens through one of these components, as they're all related to the movement of coolant, a leakage can happen. That's not good for a top-of-the-line $3 million GB200 NVL72 server cabinet.

The leakage issues were discovered before mass production of the new NVL36 and NVL72 AI server cabinets begun, and before shipments to major AI customers. This has given manufacturers the time to fix the issues, with UDN reporting that the AI server cabinet leaking issues and their respective fixes, won't affect shipping times.

• Read more: NVIDIA places fresh new orders with TSMC for more Blackwell GB200, B100, B200 AI chips
• Read more: NVIDIA is using Foxconn as the sole supplier of NVLink switches for next-gen GB200 AI servers
• Read more: NVIDIA's new GB200 AI servers led by Foxconn with 40% and Quanta with 30%: ships in Q3 2024
• Read more: NVIDIA's next-gen GB200 AI server cabinets to ship in 'small quantities' in Q4 2024
• Read more: NVIDIA's new GB200 Superchip costs up to $70,000: full B200 NVL72 AI server costs $3 million

However, large cloud service providers (CSPs) are now 'nervous' with more Taiwanese factories stepping into the supply chain to provide more options for component replacement for NVIDIA's next-gen GB200 AI server cabinets moving forward.

The certification required for branch pipes, quick connectors, and hoses is not easy, with UDN reporting that it's "quite complicated," and since these components weren't what Taiwanese manufacturers were good at in the past, when NVIDIA said its next-gen AI GPUs would be water-cooled, it caused them to change it up.

• Read more: NVIDIA DGX GB200 AI servers expected to sell 40,000 servers in 2025
• Read more: NVIDIA's next-gen GB200 AI server chips go into mass production in September
• Read more: Quanta to make NVIDIA GB200-based AI servers for Google, Amazon, and Meta
• Read more: NVIDIA GB200 Grace Blackwell Superchip: 864GB HBM3E, 16TB/sec bandwidth
• Read more: NVIDIA's full-spec B200 AI GPU uses 1200W of power, Hopper H100 uses 700W

Taiwanese cooling-related manufacturers including Shuanghong and Qihong started from their original water-cooling plate that they make at their best, expanding into product research and development to branch pipes, quick connectors, and hoses. These companies are reportedly stepping in now, providing the required parts to stop the leakages from inside NVIDIA's new GB200 NVL36 and NVL72 server cabinets.