Statements (24)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:personal_computer
|
| gptkbp:announced |
2014
|
| gptkbp:bandwidthPerLink |
up to 50 GB/s (per direction, per link, in NVLink 3.0)
up to 25 GB/s (per direction, per link, in NVLink 2.0) |
| gptkbp:competitor |
gptkb:PCI_Express
|
| gptkbp:developedBy |
gptkb:NVIDIA
|
| gptkbp:enables |
gptkb:NVSwitch
low latency communication high bandwidth communication |
| https://www.w3.org/2000/01/rdf-schema#label |
NVLink fabric
|
| gptkbp:introducedIn |
gptkb:NVIDIA_Pascal_architecture
|
| gptkbp:supports |
CPU-GPU communication
GPU-GPU communication multi-GPU configurations |
| gptkbp:usedFor |
high-speed interconnect
|
| gptkbp:usedIn |
gptkb:NVIDIA_DGX_systems
gptkb:NVIDIA_HGX_platform gptkb:NVIDIA_Tesla_GPUs gptkb:NVIDIA_A100_GPUs gptkb:NVIDIA_H100_GPUs GPU computing |
| gptkbp:bfsParent |
gptkb:NVIDIA_NVSwitch
gptkb:NVSwitch |
| gptkbp:bfsLayer |
7
|