NVIDIA's NVKE lets you run a managed Kubernetes service on your own GPU hardware
Nvidia Kubernetes Engine (NVKE)
NVKE is a management plane that platform teams install to spin up and operate isolated Kubernetes workload clusters on their own NVIDIA GPU infrastructure, provisioning machines through NVIDIA Infra Controller and giving each cluster its own control plane, etcd, and node pools. It's a self-hosting factory for clusters, not a distribution or a Cluster API flavor. The preview ships with API authentication off and plain HTTP, and isn't recommended for production.
It is a self-hosting factory for Kubernetes clusters: a throwaway seed cluster stands up a management cluster, and the management plane on it then provisions and lifecycle-manages many isolated workload clusters on demand.
- QuentinM
This is for tenant-isolated k8s clusters using NVIDIA's DSX reference stack, relying on NICo / BlueField to set up isolated networking, and wipes/re-attests each host between tenants. Mostly useful if you're chasing NVIDIA Cloud Partner status or need NVIDIA support across the whole stack.
- random3
I just got flashbacks of stuff I forgot long ago like OpenStack and EMC Rex-ray and so many things like that
- Yasuraka
This seems to mix several (should-be) distinct concerns like CAPI, Kamaji and gpu-operator.
The only things I'd take from this is the kind-based bootstrapper for the management cluster and SPIFFE/TPM integration.