Open to: Faculty, Graduate Students, and Staff
JupyterHub allows you to flexibly connect a custom computational environment with multiple users via the cloud. It is a rapidly-growing technology that has proven useful in pedagogy (currently being used for a class with more than 1000 students) as well as in research.
This three-hour hands-on workshop will cover how to set up your own JupyterHub in the cloud using Kubernetes, a popular technology for robustly managing cloud resources. We will begin by deploying a bare JupyterHub in Google Cloud that exposes a public IP address you can use to access the hub resources. We'll then cover some best-practices in debugging and managing your Kubernetes deployment, as well as how to extend your JupyterHub in order to tailor the user environment however you like.
This workshop will be useful for educators or researchers who are interested in deploying a JupyterHub for their own purposes (either in teaching or in research). This workshop doesn't require any specific technical skill - we step you through everything that is needed from start to finish. However, familiarity with the shell, editing text files, and a high-level knowledge of what we mean by "the cloud" will be helpful.
By the end of the workshop, each participant should have their own public JupyterHub running a custom Docker image. Note that while we are using resources on Google Cloud, Google provides $300 of free credits for new users, and this will only take up about $5 of those resources.