Nvidia will release its Grace processor for data centers and supercomputing in the first half of 2023. Server makers like ASUS, Foxconn, Gigabyte, and Supermicro, among others, will release servers based on the Arm chip during that time.

nvidia come first With different Grace reference platforms, which can be used in OEM servers and vary in the amount of CPUs, GPUs, and dpus Contain. The company comes with a CGX platform for “cloud graphics and gaming,” two different variants of HGX for AI and HPC, and an OVX platform for “digital twins and the global universe.” The company currently provides these server platforms, but it is based on x86 processors from AMD and Intel.

Nvidia’s Grace Superchips and its various reference platforms

These platforms will be based on the previously announced Grace CPU and Grace-Hopper ‘Superchips’. The former has two Grace Data Center Processors On a single unit, for a total of 144 Arm Neoverse cores. In addition, the unit is equipped with Lpddr5x memory with a bandwidth of up to 1 TB / s. The second Superchip contains a Grace processor with 72 cores and H100 GPU† To connect these chipsets, Nvidia uses . files Connecting its NVLink-C2C† In addition, systems can be equipped with CPU BlueField-3which aims to speed up network, storage and security services for data centers, among other things.

According to Nvidia, several Taiwanese OEMs have plans to release Grace systems in the first half of 2023. These include ASUS, Gigabyte, Foxconn, Qiwynn, QCT, and Supermicro. These OEMs make servers based on the reference platforms mentioned above, which include the “schemes and server baseboards” with which OEMs can design their own server. Nvidia says the chips will be available on 2U servers.