NVIDIA unveiled its first data center CPU, an Arm-based processor featuring a new memory architecture and designed for the most complex AI and high-performance computing workloads.
The forthcoming “Grace” CPU, which is named in honor of Grace Hopper, the U.S. computer-programming pioneer — is a highly specialized processor targeting workloads such as training next-generation NLP models that have more than 1 trillion parameters. When tightly coupled with NVIDIA GPUs, a Grace CPU-based system will deliver 10x faster performance than today’s state-of-the-art NVIDIA DGX-based systems, which run on x86 CPUs.
Underlying Grace’s performance is fourth-generation NVIDIA NVLink interconnect technology, which provides a record 900 GB/s connection between Grace and NVIDIA GPUs to enable 30x higher aggregate bandwidth compared to current generation servers. Grace will also utilize an innovative LPDDR5x memory subsystem that will deliver twice the bandwidth and 10x better energy efficiency compared with DDR4 memory. In addition, the new architecture provides unified cache coherence with a single memory address space, combining system and HBM GPU memory to simplify programmability.
“Leading-edge AI and data science are pushing today’s computer architecture beyond its limits – processing unthinkable amounts of data,” said Jensen Huang, founder and CEO of NVIDIA. “Using licensed Arm IP, NVIDIA has designed Grace as a CPU specifically for giant-scale AI and HPC. Coupled with the GPU and DPU, Grace gives us the third foundational technology for computing, and the ability to re-architect the data center to advance AI. NVIDIA is now a three-chip company.”
CSCS and Los Alamos National Laboratory both plan to bring Grace-powered supercomputers, built by Hewlett Packard Enterprise, online in 2023.
