Statements (31)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:AI_supercomputer
|
| gptkbp:announced |
2023-05-28
|
| gptkbp:architecture |
Grace Hopper architecture
|
| gptkbp:connects |
gptkb:NVLink_Switch_System
|
| gptkbp:GPU |
gptkb:NVIDIA_H100_Tensor_Core_GPU
|
| gptkbp:location |
NVIDIA data centers
|
| gptkbp:manufacturer |
gptkb:NVIDIA
|
| gptkbp:memoryType |
gptkb:HBM3
|
| gptkbp:network |
gptkb:NVIDIA_Quantum-2_InfiniBand
|
| gptkbp:notableUser |
gptkb:Google
gptkb:Microsoft gptkb:Meta |
| gptkbp:operatingSystem |
gptkb:Linux
|
| gptkbp:processor |
gptkb:NVIDIA_Grace_CPU
gptkb:NVIDIA_Grace_Hopper_Superchip |
| gptkbp:purpose |
AI inference
AI training large language model development |
| gptkbp:RAM |
144 TB
|
| gptkbp:size |
multi-node cluster
|
| gptkbp:supports |
gptkb:CUDA
gptkb:NVIDIA_Base_Command NVIDIA AI Enterprise software |
| gptkbp:targetUser |
gptkb:researchers
enterprises |
| gptkbp:uses |
scientific computing
data analytics generative AI |
| gptkbp:bfsParent |
gptkb:NVIDIA_Grace_Hopper_Superchip
|
| gptkbp:bfsLayer |
7
|
| https://www.w3.org/2000/01/rdf-schema#label |
NVIDIA DGX GH200 system
|