Set up a GCP Project

Creating a Google Cloud Platform (GCP) project for your Kubeflow deployment

Follow these steps to set up your GCP project:

• Select or create a project on theGCP Console.

• Make sure that you have theowner rolefor the project.The deployment process creates various service accounts withappropriate roles in order to enable seamless integration withGCP services. This process requires that you have theowner role for the project in order to deploy Kubeflow.

• Make sure that billing is enabled for your project. See the guide tomodifying a project’s billingsettings.

• Go to the following pages on the GCP Console and ensure that thespecified APIs are enabled:

  • Compute Engine API
  • Kubernetes Engine API
  • Identity and Access Management (IAM) API
  • Deployment Manager API
  • Cloud Resource Manager API
  • Cloud Filestore API
  • AI Platform Training & Prediction API
• If you are using theGCP Free Tier or the12-month trial period with $300 credit, note that you can’t run the defaultGCP installation of Kubeflow, because the free tier does not offer enoughresources. You need toupgrade to a paid account.

For more information, see the following issues:

• kubeflow/website #1065reports the problem.
• kubeflow/kubeflow #3936requests a Kubeflow configuration to work with a free trial project.
  • Read the GCP guide to resource quotasto understand the quotas on resource usage that Compute Engine enforces, andto learn how to check your quota and how to request an increase in quota.

You do not need a running GKE cluster. The deployment process creates acluster for you.

Next steps

• Set up an OAuth credential if you want to useCloud Identity-Aware Proxy (Cloud IAP).Cloud IAP is recommended for production deployments or deployments with accessto sensitive data. You can skip this step if you want to test Kubeflowin a non-production environment.

• Choose one of the following ways to deploy Kubeflow:

  • Using the UI. This option provides a simpleway to deploy Kubeflow.
  • Using the CLI. This option provides morecontrol over the deployment process.