This guide provides a list of configurations used to set up the JupyterHub server for Data4All. It also contains guidelines on necessary steps such as adding/deleting users, installing packages, and backing up data.

Quick links below:

Add New Users

Install Packages

User gets stuck

Creating Notebook link

Setting Idle time

Allocating Resources

Server Setup Configuration:

The JupyterHub server was set up using TLJH (The Littlest JupyterHub)! We followed the official TLJH documentation, which provides clear guidelines on setting up a JupyterHub server. The following configurations were used to set up the server to meet the requirements for Data4All, following the TLJH documentation.

• The instance name is Data4All
• For Instance type, AWS r5.2xlarge instance was chosen. It provides 8 vCPUs and 64 GB memory. This is the smallest instance that allows us to reasonably (and efficiently) handle the load of Data4All 2023.
• A disk space of 1000 GiB of root volume was chose for storage.
• Used the following setup configuration:

#!/bin/bash
curl -L <https://tljh.jupyter.org/bootstrap.py> \
| sudo python3 - \
--admin <admin> --version 0.2.0

—-admin nickross --admin evelyncampbell --admin launa --show-progress-page

• The AWS region of the instance is us-east-1f
• An Elastic IP (52.4.135.109) was allocated and associated with the instance.
• The domain data4all-jupyterhub.com points to the Elastic IP above.
• JupyterLab interface was configured to be the default interface for the users. Instructions of configuring the default interface are provided in this documentation.
• HTTPS was enabled using Let’s Encrypt by following these instructions.

Creating Notebook link:

Simply follow this link on using NBgitpuller to create links that can be distributed to students and opened on JupyterHub