A new class of data center systems for digital twins, AI, high performance computing, cloud graphics and upcoming games from ASUS, Foxconn Industrial Internet, GIGABYTE, QCT, Supermicro, Wiwynn
COMPUTEX — NVIDIA today announced that major computer manufacturers in Taiwan are preparing to release the first wave of systems powered by the NVIDIA Grace™ CPU Superchip and Grace Hopper Superchip for a wide range of workloads spanning digital twins, AI, high-performance computing, cloud graphics, and gaming.
Dozens of server models AsusFoxconn Industrial Internet, gigabyte, QCT, Supermicro and Wiwynn are scheduled to begin in the first half of 2023. Grace-powered systems will join x86 and other Arm-based servers to offer customers a wide range of choices to achieve high performance and efficiency in their data centers.
“A new kind of data center is emerging – AI factories that process and refine mountains of data to produce intelligence – and NVIDIA is working closely with our Taiwanese partners to build the systems that will enable this transformation,” said Ian Buck, vice president of Hyperscale. and HPC at NVIDIA. “These new systems from our partners, powered by our Grace Superchips, will bring the power of accelerated computing to new markets and industries globally.”
The upcoming servers are based on four new system designs featuring the Grace CPU Superchip and Grace Hopper Superchip chips, announced by NVIDIA at its last two GTC conferences. 2U form factor designs provide the server base plans and boards for OEMs and OEMs to quickly bring systems to market for NVIDIA CGX™ cloud gaming, NVIDIA OVX™ digital twin and the NVIDIA HGX™ AI and HPC platforms.
Boost modern workloads
Both NVIDIA Grace Superchip technologies enable a wide range of compute-intensive workloads across a multitude of system architectures.
- The Grace CPU Superchip comprises two CPU chips, cohesively connected through an NVIDIA NVLink®-C2C Interconnect, featuring up to 144 high-performance Arm V9 cores with scalable vector extensions and a 1 terabyte per second memory subsystem. Breakthrough design delivers the highest performance and twice the memory bandwidth and power efficiency of today’s leading server processors to meet HPC, data analytics, digital twin, cloud gaming and computing applications. most demanding hyperscale computing.
- The Grace Hopper Superchip combines an NVIDIA Hopper™ GPU with a Grace processor on NVLink-C2C in an integrated module designed to handle large-scale HPC and AI applications. Thanks to the NVLink-C2C interconnect, the Grace processor transfers data to the GPU Hopper 15 times faster than traditional processors.
Large portfolio of Grace servers for AI, HPC, digital twins and cloud gaming
The Grace CPU Superchip and Grace Hopper Superchip server design portfolio includes systems available in single base boards with single, dual, and quad lane configurations available in four workload-specific designs that can be customized by manufacturers of servers according to customer needs:
- NVIDIA HGX Grace Hopper systems for AI training, inference, and HPC are available with the Grace Hopper chip and NVIDIA BlueField®-3 SFP.
- NVIDIA HGX Grace systems for HPC and supercomputers feature the CPU-only design with Grace CPU Superchip and BlueField-3.
- NVIDIA OVX systems for digital twins and collaborative workloads feature Grace CPU Superchip, BlueField-3, and NVIDIA graphics processors.
- NVIDIA CGX systems for cloud graphics and gaming feature the Grace CPU chip, BlueField-3 and NVIDIA A16 GPUs.
NVIDIA expands its NVIDIA Certified Systems™ to servers using NVIDIA Grace CPU Superchip and Grace Hopper Superchip processors, in addition to x86 processors. The first certifications of the OEM servers are expected shortly after the delivery of the partner systems.
The Grace server portfolio is optimized for NVIDIA’s rich computing software stacks, including NVIDIA HPC, NVIDIA AI, Omniverse™, and NVIDIA RTX™.
To learn more about NVIDIA Grace, watch the Computex Keynote on Tuesday, May 24 at 11 a.m. Taiwan time.