Qualcomm on Monday, May 19 announced that it will produce a dedicated central processing unit (CPU) for data centers that use Nvidia's technology to connect with Nvidia's artificial intelligence (AI) chips.
Nvidia chips are currently dominant in the AI market, but usually paired with CPUs, a market traditionally dominated by Intel and Advanced Micro Devices (AMD). Nvidia itself has entered the CPU market by designing chips using technology from Arm Holdings to develop its "Grace" CPU.
On Monday, Qualcomm announced that it would return to the data center CPU market. In the 2010s, Qualcomm had developed an Arm-based CPU that was tested with Meta Platforms, but stopped the effort due to cost cuts and legal challenges.
SEE ALSO:
However, after acquiring Apple's former chip design team in 2021, Qualcomm quietly revived the effort, and is currently in discussion with Meta about CPUs for data centers. Last week, Qualcomm also confirmed that it has a letter of understanding with Saudi Arabia's AI company, Humain, to develop a custom data center CPU.
Pada Senin, Qualcomm menyatakan bahwa chip masa depan mereka akan menggunakan teknologi dari Nvidia yang akan memungkinkan chip tersebut berkomunikasi dengan cepat dengan prosesor graphic (GPU) Nvidia, yang merupakan adalah dari portofolio chip AI mereka.
"With the ability to connect our custom processors with Nvidia's rack architecture, we are advancing a shared vision of energy-efficient high-performance computing for data centers," said Cristiano Amon, CEO of Qualcomm, on Monday, May 19, quoted by VOI from Reuters.
The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)
