Query: gpu-operator supports DGX-1 ?

Open aii-shanker-jj opened this issue 2 years ago • 1 comments

1. Quick Debug Information

OS/Version: DGX OS 5.5
Container Runtime Type/Version: Containerd
K8s Flavor/Version(e.g. K8s, OCP, Rancher, GKE, EKS): RKE2 v1.24.13+rke2r1
GPU Operator Version: gpu-operator-v23.3.2
Hardware: DGX1-(GPU Model: V100-32GB)

2. Issue or feature description

We use DGX1-(GPU Model: V100-32GB) hardware. I would like to confirm whether gpu-operator-v23.3.2 supports driver installation, node discovery for DGX-1 ?

Nov 09 '23 11:11 aii-shanker-jj

@shivamerla I came across

https://github.com/NVIDIA/gpu-operator/issues/74

Looking forward to some input on this. I am planning to setup RKE2 cluster with

GPU only nodes
DGX-1 nodes

since DGX comes with gpu driver and toolkit, hope still I need to follow https://github.com/NVIDIA/gpu-operator/issues/256#issuecomment-917099415 right?

Nov 10 '23 15:11 aii-shanker-jj