Lab 06: NGC, private registries, Enroot and Pyxis
180 minutes suggested, up to 360 in one session. Price: 5 credits per minute.
Hosted, this lab covers Parts A, B, C, D. The other Parts need your own machine (Home Lab Kit) and are shown collapsed in the lab.
The GPU tier is not available on this site yet.
What you will be able to do
- Create an NGC personal API key, store it safely, and configure the NGC CLI with it.
- Choose an NGC tag that your driver supports, run it with NVIDIA's recommended Docker flags, and record its digest.
- Run a private registry, and pull from it into Kubernetes with an image pull secret, per pod or per namespace.
- Push an image to your NGC org's private registry and pull it into Kubernetes with an
ngc-pullsecret (where your org allows). - Import an NGC image with Enroot and run it unprivileged, with the GPU.
- Build Pyxis for a one-node Slurm and run GPU jobs in NGC containers with
srunandsbatch. - Diagnose credential, namespace,
#SBATCHand plug-in faults.