NVIDIA MGX™ is the modular architecture with a variety of server configurations tailored for AI, HPC, and NVIDIA Omniverse applications efficiently.
NVIDIA MGX™ supports a range of GPUs, CPUs, data processing units (DPUs), and network adapters, including x86 and Arm processors.
e.g., NLP, DLRM
e.g., Hash Join
e.g., OpenFOAM, GROMACS
QuantaGrid S74G-2U is the NVIDIA Grace-Hopper Superchip server which combines ARM-based Grace CPU (72 cores) and Hopper GPU (H100) with NVLink chip-to-chip interconnection and supports 96GB of HBM3 and 480GB of LPDDR5 memory.
Powered by the revolutionary architecture of the NVIDIA Grace-Hopper, QCT QuantaGrid S74G-2U empowers the breakthrough computing performance to the AI and HPC applications.