WORLD

More problems for Nvidia’s Blackwell AI chips with servers heating up

Nvidia’s new Blackwell AI chips, which have already faced delays, have encountered problems with accompanying servers that overheat, causing some customers to worry that they will not have enough time to get new data centres up and running, according to media reports.
The Blackwell graphics processing units overheat when connected together in server racks designed to hold up to 72 chips, the reports have said, citing sources familiar with the issue.
 The chipmaker has asked its suppliers to change the design of the racks several times to resolve overheating problems, according to Nvidia employees who have been working on the issue, as well as customers and suppliers with knowledge of the issue, the reports add, without naming the suppliers.
“Nvidia is working with leading cloud service providers as an integral part of our engineering team and process. The engineering iterations are normal and expected,” a company spokesperson has said in a statement to the Reuters.
In March, Nvidia had unveiled Blackwell chips and had earlier said that they would ship them in the second quarter before encountering delays, potentially affecting customers such as Meta Platforms, Alphabet’s Google and Microsoft.
Nvidia’s Blackwell chip takes two squares of silicon the size of the company’s previous offering and binds them into a single component that is 30 times speedier at tasks like providing responses from chatbots.

Report By