NVIDIA announced that a range of the world’s leading computer makers are adopting the new NVIDIA Grace superchips to create the next generation of servers turbocharging AI and HPC workloads for the exascale era. Atos, Dell Technologies, GIGABYTE, HPE, Inspur, Lenovo and Supermicro are planning to deploy servers built with the NVIDIA Grace CPU Superchip and NVIDIA Grace Hopper Superchip.
Grace CPU Superchip
All these new systems benefit from the just-announced Grace and Grace Hopper designs in the NVIDIA HGX platform, which provide manufacturers the blueprints needed to build systems that offer the highest performance and twice the memory bandwidth and energy efficiency of today’s leading data center CPU.
As supercomputing enters the era of exascale AI, NVIDIA is teaming up with our OEM partners to enable researchers to tackle massive challenges previously out of reach. Across climate science, energy research, space exploration, digital biology, quantum computing and more, the NVIDIA Grace CPU Superchip and Grace Hopper Superchip form the foundation of the world’s most advanced platform for HPC and AI.—Ian Buck, vice president of Hyperscale and HPC at NVIDIA
Leading supercomputing centers in the US and Europe will be among the first with systems featuring the superchips.
Los Alamos National Laboratory announced that Venado, its next-generation system, will be the first system in the US to be powered by NVIDIA Grace CPU technology. Built using the HPE Cray EX supercomputer, Venado is a heterogeneous system that will feature a mix of Grace CPU Superchip nodes and Grace Hopper Superchip nodes for a wide and emerging set of applications. When completed, the system is expected to exceed 10 exaflops of AI performance.
Alps, the Swiss National Computing Center’s new system, also to be built by HPE using the HPE Cray EX supercomputer, will use the Grace CPU Superchip to enable breakthrough research in a wide range of fields. It will serve as a general-purpose system open to the research community in Switzerland, as well as the rest of the world.
NVIDIA Grace speeds up compute-intensive workloads. The NVIDIA Grace CPU Superchip features two Arm-based CPUs, connected coherently through the high-bandwidth, low-latency, low-power NVIDIA NVLink-C2C interconnect. This design features up to 144 high-performance Arm Neoverse cores with scalable vector extensions and a 1 terabyte-per-second memory subsystem.
The Grace CPU Superchip interfaces with the latest PCIe Gen5 protocol to enable maximum connectivity with the highest-performing GPUs, as well as with NVIDIA ConnectX-7 smart network interface cards and NVIDIA BlueField-3 DPUs for secure HPC and AI workloads.
The Grace Hopper Superchip pairs an NVIDIA Hopper GPU with an NVIDIA Grace CPU in an integrated module connected with NVLink-C2C to address HPC and giant-scale AI applications.
The NVIDIA Grace-powered systems will run the portfolio of NVIDIA AI and NVIDIA HPC software for full-stack, integrated computing.