NVIDIA Grace Hopper Superchips

URI: https://gptkb.org/entity/NVIDIA_Grace_Hopper_Superchips
GPTKB entity
Statements (52)
Predicate Object
gptkbp:instanceOf gptkb:personal_computer
gptkbp:announced 2022
gptkbp:architecture gptkb:ARM_(CPU)
gptkb:Hopper_(GPU)
gptkbp:connects gptkb:NVLink-C2C
gptkbp:contains gptkb:NVIDIA_Hopper_GPU
gptkb:NVIDIA_Grace_CPU
gptkbp:CPUModel gptkb:NVIDIA_Grace
gptkbp:CPUProcessNode gptkb:TSMC_4N
gptkbp:designedFor AI workloads
HPC workloads
gptkbp:feature Energy efficiency
High bandwidth memory
Low latency interconnect
Unified memory addressing
gptkbp:firstSystem NVIDIA DGX GH200
gptkbp:GPU gptkb:NVIDIA_Hopper
gptkbp:GPUArchitecture gptkb:Hopper
gptkbp:GPUProcessNode gptkb:TSMC_4N
https://www.w3.org/2000/01/rdf-schema#label NVIDIA Grace Hopper Superchips
gptkbp:manufacturer gptkb:NVIDIA
gptkbp:memoryType gptkb:HBM3
gptkb:LPDDR5x
gptkbp:microarchitecture gptkb:ARM_Neoverse
gptkbp:namedAfter gptkb:Grace_Hopper
gptkbp:RAM up to 480GB LPDDR5x
up to 96GB HBM3
gptkbp:supports gptkb:NVIDIA_AI_Enterprise
gptkb:NVIDIA_Omniverse
gptkb:NVLink
gptkb:NVIDIA_HPC_SDK
gptkb:PCIe_Gen5
gptkb:NVIDIA_Magnum_IO
gptkb:NVIDIA_CUDA
NVIDIA BlueField DPU
NVIDIA DOCA
NVIDIA NVSHMEM
gptkbp:targetMarket cloud computing
data centers
supercomputing
gptkbp:usedIn ASUS systems
Dell Technologies systems
GIGABYTE systems
Hewlett Packard Enterprise systems
Inspur systems
Lenovo systems
NVIDIA DGX GH200
QCT systems
Supermicro systems
WiWynn systems
gptkbp:bfsParent gptkb:JUPITER_supercomputer
gptkbp:bfsLayer 7