Understanding and using JupyterHub

Jupyter Notebook has become an extremely popular tool for writing code for ML projects. JupyterHub is a software that facilitates the self-service provisioning of computing environments that includes spinning up pre-configured Jupyter Notebook servers and provisioning the associated compute resources on the Kubernetes platform. On-demand end users such as data engineers and data scientists can provision their own instances of Jupyter Notebook dedicated only to them. If a requesting user already has his/her own running instance of Jupyter Notebook, the hub will just direct the user to the existing instance, avoiding duplicated environments. From the end user's perspective, the whole interaction is seamless. You will see this in the next section of this chapter.

When a user requests an environment in JupyterHub, they are also given the option to choose a pre-configured sizing of hardware resources such as CPU, memory, and storage. This allows...