Deploy Kube AI Hub on GKE

This guide walks you through the steps of deploying Kube AI Hub on Google Kubernetes Engine.

Prepare a GKE Cluster

• A standard Kubernetes cluster in GKE is a prerequisite of installing Kube AI Hub. Go to the navigation menu and refer to the image below to create a cluster.

  

• In Cluster basics, select a Master version. The static version 1.15.12-gke.2 is used here as an example.

  

• In default-pool under Node Pools, define 3 nodes in this cluster.

  

• Go to Nodes, select the image type and set the Machine Configuration as below. When you finish, click Create.

  

• When the GKE cluster is ready, you can connect to the cluster with Cloud Shell.

  

Install Kube AI Hub on GKE

• Install Kube AI Hub using kubectl. The following commands are only for the default minimal installation.

  kubectl apply -f https://github.com/kubesphere/ks-installer/releases/download/v3.4.1/kubesphere-installer.yaml
  
  kubectl apply -f https://github.com/kubesphere/ks-installer/releases/download/v3.4.1/cluster-configuration.yaml
  

• Inspect the logs of installation:

  kubectl logs -n kubesphere-system $(kubectl get pod -n kubesphere-system -l 'app in (ks-install, ks-installer)' -o jsonpath='{.items[0].metadata.name}') -f
  

• When the installation finishes, you can see the following message:

  #####################################################
  ###              Welcome to Kube AI Hub!           ###
  #####################################################
  Console: http://10.128.0.44:30880
  Account: admin
  Password: P@88w0rd
  NOTES：
    1. After logging into the console, please check the
      monitoring status of service components in
      the "Cluster Management". If any service is not
      ready, please wait patiently until all components
      are ready.
    2. Please modify the default password after login.
  #####################################################
  https://kubesphere.io             2020-xx-xx xx:xx:xx
  

Access Kube AI Hub Console

Now that Kube AI Hub is installed, you can access the web console of Kube AI Hub by following the steps below.

• In Services & Ingress, select the service ks-console.

• In Service details, click Edit and change the type from NodePort to LoadBalancer. Save the file when you finish.

• Wait for GKE to assign an external endpoint, and then access the Kube AI Hub web console with that address.

  Tip

  Instead of changing the service type to LoadBalancer, you can also access Kube AI Hub console via NodeIP:NodePort (service type set to NodePort). You may need to open port 30880 in firewall rules.

• Log in to the console with the default account and password (admin/P@88w0rd). In the cluster overview page, you can see the dashboard.

Enable Pluggable Components (Optional)

The example above demonstrates the process of a default minimal installation. To enable other components in Kube AI Hub, see Enable Pluggable Components for more details.