NVIDIA’s Grace Hopper Superchip
NVIDIA today at the SIGGRAPH conference announced the NVIDIA GH200 Grace Hopper platform — based on a new Grace Hopper Superchip with the first HBM3e processor, according to NVIDIA — built for accelerated computing and generative AI.
Built for large language models, recommender systems and vector databases, the new platform will be available in a range of configurations, according to the company. The dual configuration, which delivers up to 3.5x more memory capacity and 3x more bandwidth than the current generation offering, comprises a single server with 144 Arm Neoverse cores, eight petaflops of AI performance and 282GB of the latest HBM3e memory technology.
“To meet surging demand for generative AI, data centers require accelerated computing platforms with specialized needs,” said Jensen Huang, founder and CEO of NVIDIA. “The new GH200 Grace Hopper Superchip platform delivers this with exceptional memory technology and bandwidth to improve throughput, the ability to connect GPUs to aggregate performance without compromise, and a server design that can be easily deployed across the entire data center.”
The new platform uses the Grace Hopper Superchip, which can be connected with additional superchips by NVIDIA NVLink, allowing them to work together to deploy the large models used for generative AI. This high-speed, coherent technology gives the GPU access to the CPU memory, providing a combined 1.2TB of fast memory when in dual configuration, NVIDIA said.
HBM3e memory, which the company said is 50 percent faster than current HBM3, delivers 10TB/sec of combined bandwidth.
NVIDIA said its new chip platform is compatible with the NVIDIA MGX server specification unveiled at COMPUTEX earlier this year. With MGX, system manufacturers add Grace Hopper into over 100 server variations.
System manufacturers are expected to deliver systems based on the platform in Q2 of 2024.
