Nvidia announced that its Grace and Grace Hopper Superchips would be used in a new supercomputer being built by the Los Alamos National Laboratory.
The Nvidia Grace CPU Superchip contains two Arm-based CPUs that are linked coherently via the high-bandwidth, low-latency, low-power Nvidia NVLink-C2C connection. This concept includes up to 144 high-performance Arm Neoverse processors with scalable vector extensions and a memory subsystem capable of 1 terabyte per second.
The Grace CPU Superchip supports the newest PCIe Gen5 protocol for maximum communication with the most powerful GPUs, as well as Nvidia ConnectX-7 smart network interface cards and Nvidia BlueField-3 DPUs for safe HPC and AI applications.
To address HPC and large-scale AI applications, the Grace Hopper Superchip combines an Nvidia Hopper GPU with an Nvidia Grace CPU in an integrated module connected by NVLink-C2C.
The central processing units (CPUs) will be part of the Venado, an HPE Cray EX supercomputer. According to Nvidia, the chips are also being used by other computer manufacturers to develop the next generation of servers that will turbocharge AI and HPC workloads for the exascale era.
Ian Buck, who serves the role of Vice President of hyperscale and HPC at Nvidia, had the following to say:
As supercomputing enters the era of exascale AI, Nvidia is teaming up with our [hardware] 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.
Supercomputing facilities in the United States and Europe will be among the first to use systems using Superchips.
The Venado system at Los Alamos National Laboratory will be the first in the United States to use Nvidia Grace CPU technology. The Venado is a heterogeneous system that will have a combination of Grace CPU and Grace Hopper superchip nodes for a diverse and developing range of applications. When finished, the system is predicted to outperform ten exaflops in AI performance.
Irene Qualters, who serves the role of Associate Director for Simulation and Computation at LANL, had the following to say:
By equipping LANL’s researchers with the performance of Nvidia Grace Hopper, Venado will continue this laboratory’s commitment to pushing the boundaries of scientific breakthroughs. Nvidia’s accelerated computing platform and expansive ecosystem are removing performance barriers, allowing LANL to make new discoveries that will benefit the nation and society as a whole.
‘Venado’ another supercomputing CPU win for Nvidia, says Wells Fargo
After Los Alamos National Laboratory announced a next-generation supercomputer called “Venado” in conjunction with this week’s ISC22 conference, Wells Fargo analyst Aaron Rakers highligh…https://t.co/icDWUBd5kW— Hot Stocks NYC (@HotStocks_NYC) May 31, 2022
Servers developed with the Grace and Grace Hopper CPUs will be deployed by Atos, Dell Technologies, Gigabyte, HPE, Inspur, Lenovo, and Supermicro.
All of these new systems benefit from the recently released Grace and Grace Hopper designs in the Nvidia HGX platform, which give manufacturers the blueprints needed to construct systems with double the memory bandwidth and energy efficiency of today’s leading data center CPU.
Furthermore, the Grace CPU Superchip will be used in Alps, the Swiss National Computing Center’s new system, which will also be developed by HPE utilizing the HPE Cray EX supercomputer. This will enable breakthrough research in a variety of sectors. It will be a general-purpose system available to the Swiss scientific community as well as the rest of the globe.
