Nvidia Announces Grace CPU with Arm Cores for Data Centers and Supercomputers

Nvidia introduces its Grace CPU. This chip is based on the Arm architecture and is intended for data centers and high-performance computing. The CPU should be available in 2023, and will be used in a Swiss supercomputer that same year.

Nvidia will announce the Grace processor during its GTC 2021 presentation. The CPU will not be available to consumers, but will be used exclusively for data centers and HPC. For example, the chip is intended for running AI workloads. Nvidia claims that systems with the Grace processor in them “perform up to ten times better” than the current Nvidia DGX servers, which currently use AMD EPYC CPUs.

The Grace CPU will be equipped with future Arm Neoverse cores, Nvidia reports. The company also shares few concrete details about these cores and the chip itself. For example, it is still unknown how many of these cores the chip will receive. It is also unclear how many transistors are used in the CPU and in which process the chip is made. Nvidia also does not mention power consumption.

The Grace CPU. Image via Nvidia

The company does share details about Grace’s memory. The chip features fourth-generation Nvidia NVLink interconnects, which provide a maximum bandwidth of 900GB/s between the Grace CPU and Nvidia GPUs. The chips would also offer 600GB/s bandwidth from CPU to CPU. Grace will also get an Lpddr5x memory subsystem, which should provide more than 500GB/s of bandwidth and support for error-correcting code. This bandwidth would be twice that of DDR4, while this Lpddr5x memory would be up to ten times more energy efficient.

Nvidia also announces that it will ship an Alps supercomputer with Grace CPUs in 2023. This supercomputer is being built for the Swiss National Supercomputing Center in collaboration with Hewlett Packard Enterprise. Alps must have 20 Exaflops of computing power for AI workloads. Nvidia claims that Alps will be the fastest supercomputer for AI computing. In addition to the Grace CPU, the supercomputer will also have unannounced next-gen Nvidia GPUs.

Nvidia says it is now a company that is working on three types of chips; In addition to GPUs, the company will also release CPUs and so-called DPUs in the future. The latter is a data processing unit, which is used to accelerate data center infrastructure. Nvidia says it wants to update these different product groups every two years. The company says it releases at least one new product every year.

This isn’t the first time Nvidia has been working on Arm chips. The company announced Project Denver in 2011. The manufacturer then reported that it was working on Arm CPUs, which would be integrated into GPUs. The company has since released several Tegra SOCs, which combine Arm cores with a GPU. Such a chip is used in the Nintendo Switch, among other things. Nvidia plans to acquire Arm for $40 billion, although the acquisition has yet to be approved by competition authorities.

Nvidia wants to update its different product groups every two years