Frequently Asked Questions (FAQ)
The following sections contain some frequently asked questions around deploying Data 8 with JupyterHub. If you’ve got a question that isn’t answered, feel free to open an issue on the z2d8 github repository.
How can I deploy a JupyterHub on hardware that isn’t listed in this guide?
These guides focus on deploying JupyterHub on large cloud providers such as Microsoft, Google, and Amazon. However, many organizations prefer to deploy on something other than this (e.g. their own local hardware).
You can deploy a JupyterHub on any hardware, though it must run Kubernetes. Setting up Kubernetes is not trivial, and so we recommend cloud providers because they simplify this process significantly. If there’s a specific provider or setup that you think should be officially documented, please reach out and we can discuss!
How can I adapt the Data 8 setup for a different course?
If you’d like to deploy a course other than Data 8 (or a small delta from it), you can use most of the same steps to do so. The JupyterHub deployment is the same, and the Docker image we use for user sessions should fit many data science course needs. If you’d like to adapt or modify parts of the textbook or labs, please reach out to the Data 8 team so that we can discuss how to do this most effectively.
For a more in-depth guide on customizing a JupyterHub (e.g. to install an environment that isn’t covered by the Data 8 Docker image), see the Zero to JupterHub guide. For a more generic jekyll-based textbook template, see the Textbooks with Jekyll template repository.
How can I contribute back to Data 8 materials?
If you feel that there are changes / improvements that should be made to the Data 8 material, feel free to open an issue on the Data 8 textbook page. Alternatively you are always welcome to fork the repository and make a pull-request!