NVIDIA has announced the next-generation NVIDIA GH200 Grace Hopper Superchip platform, a revolutionary new data center platform that is designed to meet the demands of the era of accelerated computing and generative AI.
The new platform is based on the Grace Hopper Superchip, which is the world’s first HBM3e processor. HBM3e is a new memory technology that offers up to 2x the bandwidth of HBM2e, which is the current state-of-the-art memory technology. This increased bandwidth will enable the Grace Hopper Superchip to deliver up to 10x the performance of previous-generation CPUs for applications running terabytes of data.
The GH200 Grace Hopper Superchip platform also includes a new interconnect technology called NVIDIA NVLink-C2C.NVLink-C2C is a high-speed interconnect that enables GPUs and CPUs to connect directly to each other, without going through the system’s memory controller. This will enable the platform to deliver up to 20x the bandwidth of PCI Express 5.0, which is the current state-of-the-art interconnect technology.
The new platform is expected to be available in a wide range of configurations, from single-node servers to large-scale clusters. It is expected to be available in Q2 of calendar year 2024.
Benefits of the GH200 Grace Hopper Superchip Platform
The GH200 Grace Hopper Superchip platform offers a number of benefits over previous-generation data center platforms,including:
- Up to 10x the performance of previous-generation CPUs for applications running terabytes of data
- Up to 20x the bandwidth of PCI Express 5.0
- A new server design that can be easily deployed across the entire data center
Applications of the GH200 Grace Hopper Superchip Platform
The GH200 Grace Hopper Superchip platform is designed for a wide range of applications, including:
- Generative AI
- High-performance computing (HPC)
- Data analytics
- Machine learning
Impact of the GH200 Grace Hopper Superchip Platform
The GH200 Grace Hopper Superchip platform is expected to have a significant impact on the data center industry. It is expected to enable a new era of accelerated computing and generative AI, and it is expected to be adopted by a wide range of organizations, including cloud providers, enterprises, and research institutions.
Conclusion
The NVIDIA GH200 Grace Hopper Superchip platform is a revolutionary new data center platform that is designed to meet the demands of the era of accelerated computing and generative AI. It is expected to have a significant impact on the data center industry and to enable a new era of innovation.
