Nvidia launches CPU, touts uptake of Clara AGX kit

By The Science Advisory Board staff writers

Graphics processing unit (GPU) technology developer Nvidia launched a central processing unit (CPU) specifically for giant-scale artificial intelligence (AI) and high-performance computing tasks at the Nvidia GPU Technology Conference (GTC) held online April 12-16. Nvidia also is highlighting the increasing uptake of its Clara AGX kit for the development of AI-enabled medical instruments

Called Grace after computer programming pioneer Rear Adm. Grace Hopper, the new CPU is a highly specialized processor targeting the most data-intensive applications in high-performance computing and AI. The company believes that when coupled with NVIDIA GPUs, a Grace-based computer will deliver 10 times faster performance than today's state-of-the-art Nvidia DGX-based systems, which run on x86 CPUs from Intel.

Nvidia's new AGX AI platform for medical devices. Image courtesy of Nvidia.

Market observers saw the introduction of Grace as a competitive move against Intel, which supplies most of the CPUs on the market. The new Grace CPUs will be based on technology from Arm Holdings, a chip architecture licensing company that Nvidia agreed to acquire last year.

Meanwhile, the Clara AGX kit helps researchers incorporate AI into software-based instruments like microscopes, the company said. Nvidia Clara AGX provides high-bandwidth signal processing, computing reconstruction, AI processing, and 3D visualization. Its design is based on the company's Jetson AGX Xavier developer kit, and it features Nvidia's RTX 6000 GPU and its Mellanox ConnectX-6 SmartNIC.

A number of companies are exploring the use of the new CPU unit, including Kaya Instruments who is developing computer vision for a number of applications including electron microscopes featuring Nvidia's GPUs for real-time processing and AI-accelerated analysis.

Copyright © 2021 scienceboard.net

To read this and get access to all of the exclusive content on The Science Advisory Board create a free account or sign-in now.

Member Sign In:
MemberID or email address:  
Do you have a Scienceboard.net password?
No, I want a free membership.
Yes, I have a password:  
Forgot your password?