In this blog post I’m going to detail how deploy and configure a Nvidia GPU enabled Tanzu Kubernetes Grid cluster in AWS. The method will be similar for Azure, for vSphere there are a number of additional steps to prepare the system. I’m going to essentially follow the official documentation, then run some of the Nvidia tests. Like always, it’s good to get a visual reference and such for these kinds of deployments.
When deploying Tanzu Kubernetes Grid to AWS, the deployment was failing with the following output:
unable to set up management cluster, : unable to wait for cluster and get the cluster kubeconfig: error waiting for cluster to be provisioned (this may take a few minutes): cluster creation failed, reason:'InstanceProvisionFailed @ Machine/tkg-aws-mgmt-control-plane-dqb4v', message:'1 of 2 completed'
This walk-through will detail the technical configurations for using vRA Code Stream to deploy AWS EKS Clusters, register them as Kubernetes endpoints in vRA Cloud Assembly and Code Stream, and finally register the newly created cluster in Tanzu Mission Control.
Tanzu Mission Control has some fantastic capabilities, including the ability to deploy Tanzu Kubernetes Clusters to various platforms (vSphere, AWS, Azure). However today there is no support to provision native AWS EKS clusters, it can however manage most Kubernetes distributions.
Therefore, when I was asked about where VMware could provide such capabilities, my mind turned to the ability to deploy the clusters using vRA Code Stream, and provide additional functions on making these EKS clusters usable.
High Level Steps
Create a Code Stream Pipeline
Create a AWS EKS Cluster
Create EKS cluster as endpoint in both Code Stream and Cloud Assembly
Register EKS cluster in Tanzu Mission Control
vRA Cloud access
The pipeline can be changed easily for use with vRA on-prem