(Reuters) -Nvidia’s new Blackwell AI chips, which have already faced delays, have encountered problems with accompanying servers overheating, leading some customers to fear they won’t have enough time to get the new centers up and running. of data, Information reported on Sunday. .
Blackwell graphics processing units overheat when connected to each other in custom server racks the company has designed, the report said, citing sources familiar with the problem.
The AI chip maker has asked its suppliers to change the design of the frames several times to resolve overheating issues, according to Nvidia (NASDAQ 🙂) employees who have been working on the issue, as well as customers and suppliers with knowledge of the issue, according to the report, he said without naming the suppliers.
Nvidia did not immediately respond to a request for comment outside of normal business hours.
Nvidia unveiled the Blackwell chips in March and had previously said they would ship in the second quarter before facing delays, which could impact customers such as Meta Platforms (NASDAQ 🙂), Alphabet’s Google (NASDAQ 🙂) and Microsoft (NASDAQ 🙂).
Nvidia’s Blackwell chip takes two squares of silicon the size of the company’s previous offering and stitches them together into a single component that’s 30 times faster at tasks like providing chatbot responses.