Nvidia announces new ARM-based Grace CPU for data centers

Nvidia, the US-based multinational technology company, has announced its latest processor, Grace. This new ARM-based CPU is designed to be used in data centers, helping accelerate large computing workloads such as artificial intelligence (AI) and machine learning (ML).

Grace is named after Grace Hopper, the computer programming pioneer who was one of the first women to earn a PhD in mathematics. According to Nvidia, Grace is a highly specialized processor that can deliver 10 times the performance of today’s fastest servers on complex AI and high-performance computing workloads.

Grace is also Nvidia’s first data center CPU and is part of the company’s ongoing efforts to make AI more accessible. In addition to this, the new CPU offers a better energy efficiency ratio than traditional x86-based CPUs, allowing data centers to reduce their power consumption and carbon footprint.

According to Nvidia CEO Jensen Huang, the new CPU is expected to become the Swiss Army knife of the modern data center, offering a powerful computing platform for large-scale workloads. He also highlighted that the new CPU is part of Nvidia’s plan to create a data center infrastructure that is both secure and scalable.

With the release of the Grace CPU, Nvidia is entering a new market that is currently dominated by companies such as Intel, AMD, and Qualcomm. However, Nvidia believes that Grace’s specialized architecture and efficiency ratio will make it a popular choice among data center operators. The new CPU is also part of Nvidia’s ongoing partnership with ARM, which it acquired in 2020.

While no specific release date has been announced for the Grace CPU, Nvidia has said that it plans to start shipping the new processor in early 2023. The company has already received support from several major technology companies, including IBM and Fujitsu, which have both committed to building systems based on the new processor.