Presented with great fanfare by Jensen Huang, CEO of Nvidia, Blackwell chips are said to encounter problems in data centers, but the company remains confident.
We learn via Reuters that when Blackwell chips are installed in server racks designed to hold up to 72 chips, the chips overheat, causing delays in information processing.
A problem which seems to be known to Nvidia which indicates that it has repeatedly asked its suppliers to review the design of these racks to avoid these problems. A company spokesperson tells Reuters what “Nvidia works with leading cloud service providers as an integral part of our engineering team and processes. Engineering iterations are normal and expected ».
It remains to be seen whether the famous customers of these chips will have the time necessary to reinstall them correctly and get their data centers running. Designed for artificial intelligence, the chips will, for example, equip the Grok AI data center, Elon Musk's “irreverent” AI.
