JupyterHub News Item

gfs250.png

jupyterlab.png

lorenz.png

unidata-jupyterhub.md

• A JupyterHub for the Unidata Community
  • Jupyter Notebooks
  • Unidata Promotes Jupyter
  • A JupyterHub for the Unidata Community
    • https://jupyterhub.unidata.ucar.edu
    • Additional Features
  • Give it a Try
  • A JupyterHub for Your Classroom or Workshop

A JupyterHub for the Unidata Community

Jupyter Notebooks

Jupyter is a web technology that enables running of code in an interactive manner in a web browser. Python was the first language available on Jupyter, but it has expanded to support many other programming languages such as R and Julia. With a mixture of explanatory text, interactive code blocks, equations and multimedia, Jupyter notebooks excel at scientific teaching and research objectives and are now visible in many scientific disciplines ¹. In the last few years, Jupyter computational notebooks have become an important tool for science professionals and students ².

Unidata Promotes Jupyter

For the last six years, Unidata has promoted Jupyter to our community for teaching and research in the atmospheric sciences. We have made extensive use of notebooks in the Unidata Python Workshop to teach analysis and visualization of geoscientific data. These instructional materials have expanded to two additional notebook projects: the Python Gallery which is a collection of atmospheric science notebooks and the Online Python Training which teaches Python basics with an atmospheric science emphasis.

A JupyterHub for the Unidata Community

https://jupyterhub.unidata.ucar.edu

As a next step in this advocacy of Jupyter technology and as part of Unidata's broader science gateway efforts, we would like to announce a new JupyterHub server for the Unidata community. https://jupyterhub.unidata.ucar.edu is a JupyterHub server tailored to atmospheric science users. Specifically, this server has three Unidata Python projects available and the kernel environments required to run them: Python Workshop, Python Gallery, Online Python Training.

The objective of this JupyterHub is to enable scientists and students to have immediate access to properly configured notebooks. Users will not have to configure difficult to install software on personal computing devices. The deployment of Jupyter and the configuration of the Python environments have been done on the user's behalf by Unidata. The end result is an acceleration of "time to science".

Additional Features

1. JupyterLab

   This JupyterHub server is equipped with JupyterLab, the next generation web user interface for Jupyter.

   

2. Terminal

   There is a Unix terminal which can be handy in any situation in which you need a command line interface. You can use git version control, for example or data fetching tools such as curl and wget.

3. nbgitpuller

   nbgitpuller ensures repositories remain in sync with their respective canonical repositories. nbgitpuller is in place for the three notebook projects described previously to make sure they stay up-to-date.

Give it a Try

We invite you to give the Unidata JupyterHub server a try. This JupyterHub leverages the GitHub login mechanism meaning that you will need to obtain a GitHub login, if you do not have one already. Once you are logged in, please peruse the README_FIRST.ipynb notebook to get oriented. Pay specific attention to the information about kernels to ensure notebooks run without errors.

For example, go to the notebook gallery folder, enter the notebook directory and select and open 250hPa_Hemispheric_Plot.ipynb notebook. Select the "notebook-gallery" kernel from the "Kernel, Change Kernel…" menu item. You can then run this notebook which will generate a plot of geopotential height and wind speed (knots) at 250 hectopascals derived from the GFS model.

A JupyterHub for Your Classroom or Workshop

Beyond the Unidata JupyterHub server discussed here, for the past few months, Unidata has also been deploying JupyterHub servers for semester-long classes and workshops. For those in the atmospheric sciences, please contact Unidata if you would like a pre-configured JupyterHub server for classroom or workshop settings.

Footnotes

¹ https://github.com/jupyter/jupyter/wiki/A-gallery-of-interesting-Jupyter-Notebooks

² https://www.nature.com/articles/d41586-018-07196-1