Customizing Kubeflow on AWS

Tailoring a AWS deployment of Kubeflow

This guide describes how to customize your deployment of Kubeflow on Amazon EKS. Some of the steps can be done before you run the apply platform command, and some of them can be done before you run the apply k8s command. Please see the following sections for details. If you don’t understand the deployment process, please see deploy for details.

Customizing Kubeflow

Here are the optional configuration parameters for kfctl on the AWS platform.

Options Description Required
awsClusterName Name of your new or existing Amazon EKS cluster YES
awsRegion The AWS Region to launch in YES
awsNodegroupRoleNames The IAM role names for your worker nodes YES for existing clusters / No for new clusters

Customize your Amazon EKS cluster

Before you run ${KUBEFLOW_SRC}/scripts/ apply platform, you can edit the cluster configuration file to change cluster specification before you create the cluster.

Cluster configuration is stored in ${KUBEFLOW_SRC}/${KFAPP}/aws_config/cluster_config.yaml. Please see eksctl for configuration details.

For example, the following is a cluster manifest with one node group which has 2 p2.xlarge instances. You can easily enable SSH and configure a public key. All worker nodes will be in single Availability Zone.

kind: ClusterConfig
  # AWS_CLUSTER_NAME and AWS_REGION will override `name` and `region` here.
  name: kubeflow-example
  region: us-west-2
  version: '1.12'
# If your region has multiple availability zones, you can specify 3 of them.
#availabilityZones: ["us-west-2b", "us-west-2c", "us-west-2d"]

# NodeGroup holds all configuration attributes that are specific to a nodegroup
# You can have several node groups in your cluster.
  - name: eks-gpu
    instanceType: p2.xlarge
    availabilityZones: ["us-west-2b"]
    desiredCapacity: 2
    minSize: 0
    maxSize: 2
    volumeSize: 30
    allowSSH: true
    sshPublicKeyPath: '~/.ssh/'

  # Example of GPU node group
  # - name: Tesla-V100
  # Choose your Instance type for the node group.
  #   instanceType: p3.2xlarge
  # GPU cluster can use single availability zone to improve network performance
  #   availabilityZones: ["us-west-2b"]
  # Autoscaling Groups settings
  #   desiredCapacity: 0
  #   minSize: 0
  #   maxSize: 4
  # Node Root Disk
  #   volumeSize: 50
  # Enable SSH out side your VPC.
  #   allowSSH: true
  #   sshPublicKeyPath: '~/.ssh/'
  # Customize Labels
  #   labels:
  #     '': 'nvidia-tesla-k80'
  # Setup pre-defined iam roles to node group.
  #   iam:
  #     withAddonPolicies:
  #       autoScaler: true

Customize Private Access

Please see this section

Customize Logging

Please see this section

Customize Authentication

Please see this section