Skip to main content

Toy Data Portal for Hydrology

This tutorial shows how to stand up a basic hydrology stack on Google Cloud: a managed Kubernetes cluster, JupyterHub for notebooks, and a toy data portal for NetCDF uploads and metadata.

What you build​

  • A small GKE Autopilot cluster (Google handles node sizing).
  • JupyterHub so users can run notebooks in the browser.
  • A toy portal that uploads NetCDF files to Cloud Storage and exposes metadata/plots.

Steps​

  1. Start the cluster (1_access_gcp.md): Log in with the CloudBank account, enable Kubernetes Engine, create an Autopilot cluster, and verify it with a sample app.
  2. Deploy JupyterHub (2_deploy_jupyterhub.md): Install JupyterHub on the cluster for browser-based notebooks.
  3. Deploy the data portal (3_deploy_portal.md): Deploy the portal, connect it to a Cloud Storage bucket, and upload a NetCDF file to see extracted metadata.
  4. Use the notebook (4_analyze_portal_data.ipynb): Run inside JupyterHub to call the portal API, download a NetCDF file, and plot it (screenshots in img/).