JupyterHub Year 1 Development Phases

[kabilar]

• Phase 1
  • In collaboration with the DANDI Archive project, implement the AWS DoEKS JupyterHub deployment. Austin on the DANDI team is leading this development.
• Phase 2
  • Using the DANDI Archive DoEKS configuration (e.g. instance types, base images, example notebooks), launch a JupyterHub for the LINC project on the LINC AWS account. The JupyterHub is now available at https://hub.lincbrain.org.
• Phase 3
  • Clean up any AWS resources that were created during the initial deployment and are not currently needed.
  • Update Kabi to admin permissions
  • Shared directories for shared conda environments
    • One directory should have read/write access for all users
    • Another directory should only have read access for all users (and write access for select users?) (or we build Docker images)
  • Create Docker images for the LINC use cases.
  • Create an example notebook for streaming data from the LINC data sharing platform and saving back to the derivatives directory.
  • Create example to show how to download a single file (with the entire directory tree)
  • Save Terraform state to S3 bucket.

[kabilar]

Transferred this issue here since we are archiving the linc-hub repo.