Statements (18)
Predicate | Object |
---|---|
gptkbp:instanceOf |
GPU virtualization technology
|
gptkbp:allows |
multiple users to share a single GPU
|
gptkbp:announced |
2020
|
gptkbp:developedBy |
gptkb:NVIDIA
|
gptkbp:enables |
dedicated memory and compute resources per instance
partitioning of a single GPU into multiple instances |
https://www.w3.org/2000/01/rdf-schema#label |
MIG (Multi-Instance GPU)
|
gptkbp:introducedIn |
gptkb:NVIDIA_A100_GPU
|
gptkbp:provides |
hardware-level isolation between GPU instances
|
gptkbp:supportedBy |
gptkb:NVIDIA_Ampere_architecture
gptkb:NVIDIA_Hopper_architecture gptkb:NVIDIA_GPU_drivers gptkb:NVIDIA_CUDA |
gptkbp:usedFor |
cloud computing
high-performance computing AI workloads |
gptkbp:bfsParent |
gptkb:NVIDIA_A100_GPUs
|
gptkbp:bfsLayer |
7
|