Statements (23)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
GPU technology
|
| gptkbp:allows |
multiple users to share a single GPU
|
| gptkbp:developedBy |
gptkb:NVIDIA
|
| gptkbp:enables |
dedicated memory and compute resources per instance
improved GPU utilization isolation of GPU resources quality of service (QoS) for GPU users partitioning of a single GPU into multiple instances |
| gptkbp:features |
gptkb:NVIDIA_Ampere_architecture
gptkb:NVIDIA_Hopper_architecture |
| https://www.w3.org/2000/01/rdf-schema#label |
Multi-Instance GPU (MIG)
|
| gptkbp:introducedIn |
2020
|
| gptkbp:supportedBy |
gptkb:NVIDIA_A100
gptkb:NVIDIA_A10 gptkb:NVIDIA_H100 gptkb:NVIDIA_A30 gptkb:NVIDIA_A40 |
| gptkbp:usedIn |
gptkb:high_performance_computing_(HPC)
cloud computing data centers AI workloads |
| gptkbp:bfsParent |
gptkb:Nvidia_Hopper
|
| gptkbp:bfsLayer |
6
|