Statements (17)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:personal_computer
|
| gptkbp:announced |
2022
|
| gptkbp:bandwidthPerLink |
50 GB/s
|
| gptkbp:developedBy |
gptkb:NVIDIA
|
| gptkbp:enables |
scalable GPU clusters
|
| gptkbp:generation |
4
|
| https://www.w3.org/2000/01/rdf-schema#label |
NVLink 4.0
|
| gptkbp:predecessor |
gptkb:NVLink_3.0
|
| gptkbp:purpose |
high-speed GPU interconnect
|
| gptkbp:successor |
gptkb:NVLink_3.0
|
| gptkbp:supports |
multi-GPU communication
|
| gptkbp:totalBandwidth |
900 GB/s
|
| gptkbp:usedIn |
gptkb:NVIDIA_H100_GPU
gptkb:NVIDIA_Hopper_architecture gptkb:NVIDIA_DGX_H100_system |
| gptkbp:bfsParent |
gptkb:Nvidia_Hopper
|
| gptkbp:bfsLayer |
6
|