Nvidia’s latest Blackwell AI chips encounter overheating issues in servers.

Nvidia’s new Blackwell AI chips, which have already experienced delays, are now facing overheating issues in servers, raising concerns among customers about potential delays in setting up new data centers, The Information reported on Sunday.

According to the report, the Blackwell graphics processing units (GPUs) overheat when connected in server racks designed to accommodate up to 72 chips. Sources familiar with the matter revealed that Nvidia has repeatedly asked suppliers to redesign the racks to address the overheating problem. This information comes from Nvidia employees, as well as customers and suppliers involved, although the report did not name the suppliers.

“Nvidia is collaborating closely with leading cloud service providers as an integral part of our engineering process. Iterative engineering is both normal and expected,” a company spokesperson told Reuters.

In March, Nvidia introduced its Blackwell chips, initially scheduled for shipment in the second quarter. However, delays arose, potentially impacting major customers like Meta Platforms, Alphabet’s Google, and Microsoft.

The Blackwell chip features an innovative design that combines two silicon squares, each the size of the company’s previous chips, into a single component. This new architecture delivers performance up to 30 times faster for tasks such as generating chatbot responses.

FOR LATEST TREDING NEWS CLICK ON : https://inetfreak.com/admin