How does JupyterHub work?

JupyterHub is the central access point and manages users and JupyterLab-servers. Let's take a closer look at the setup of JupyterHub:

  • Login: JupyterHub uses the central login service UHH Login of Universität Hamburg, to identify users and manage access permissions.
  • Spawning: A user receives a new instance of a JupyterLab-server on a dedicated server or cluster and can work with it. We use Docker herefor, creating thus a new container for each user.
  • Docker: A software solution to run applications isolated from others in containers. Each container contains the application and all its dependencies, so that every application functions equally on any system.
    A Docker Image is the foundation upon which a container runs and from which all other containers are generated. It contains all information required to start the system, such as filesystem, configurations, and other software packages.
  • JupyterLab: The web interface for Jupyter Notebooks and other projects that run in a browser. It offers a user interface with features like text editors, terminals, file and data management, tables, diagrams etc.



Related Questions