Enable a GCP project

This is the first step to running pipelines in a Google Cloud project. It creates a Deployment in the Infrastructure Manager service — every resource GENI Workflows provisions in GCP is created through Terraform templates managed by Infra Manager.

What GENI uses inside a GCP project: Artifact Registry for engine images, Cloud Batch for jobs, Cloud Storage for the executions bucket, Compute Engine for Nextflow and its tasks, and Pub/Sub for task events.

Prerequisites

• The gcloud command installed.
• A GCP user with administrative permissions on the project.

Note

You do not need to be signed in to GENI for this step, only to GCP.

Run the setup

gcloud auth login
gcloud auth application-default login

geni setup enable-gcp \
  --project-id <project-id> \
  --region <region> \
  --credentials-output gcp-wif.json

Argument	Meaning
<project-id>	The project’s unique identifier, for example useful-loop-496215-i8
<region>	The region to use, for example us-central1
gcp-wif.json	Output file holding the generated credentials

The command provisions a deployment named geni-setup, containing IAM resources, and writes a JSON credentials file.

Caution

Keep gcp-wif.json. The next step — Create the environment — needs it, and it grants GENI access to the project.

Next

Set up the FOCUS dataset so cost data is available, then create the environment.