Getting Started with JupyterHub on Mistral

General Infos

Data analysis work will be done on the supercomputer Mistral at DKRZ. You find some information on high-performance computing platform here:

• https://www.dkrz.de/up/systems/mistral

The main part of the proposed data analysis will be done with Jupyter Notebooks using Python as programming language. For python, you will find a lot of information in the internet. As starting point, these tutorials are interesting:

• https://github.com/NCAR/ncar-python-tutorial/blob/master/notebooks/welcome.ipynb

• https://unidata.github.io/python-training/python/intro-to-python/

For an intro to Jupyter Notebooks, see here:

• https://www.dataquest.io/blog/jupyter-notebook-tutorial/

• https://jupyter.readthedocs.io/en/latest/index.html

Login

• Do you have an account for Mistral?

  • Your work is part of the Mistral traíning project bb1224. The first steps is to ask your supervisor for account details, i.e. username and password pair.

  • Student accounts are provided by DKRZ on a temporary basis.

• check JupyterHub Tutorial

  • You also find brief and concise information about jupyterhub here: https://www.dkrz.de/up/systems/jupyterhub-dkrz.de-1

  • This tutorial covers in more detail all aspects about jupyterhub at DKRZ: https://jupyterhub.gitlab-pages.dkrz.de/jupyterhub-docs/overview.html

• Load a Jupyter Spawner

  • type https://jupyterhub.dkrz.de

  • You will land on this page where you need to input your mentioned account details

    

  • At the next page, you need to select the “Preset” button below the Spawner Options. Click on the blue buttom: “start from preset profiles”

    

  • input the following options

    Category	Input
    profile	5GB, 1 core, prepost, 12:00
    account	bb1224
    qos	training

    which looks like this:

    

  • You will land in your mistral ${HOME} directory. Please, check the documentation linked above to know how much space is free for you (keyword: “quotas”).

• Create Folders: In your ${HOME} directory, you can created folders. Please do so, e.g. a “data” and a “notebooks” folder:

  

  Folders are created as “Untitled Folder”. Check the box and rename them:

  

Create your first Notebook:

• Click on “notebooks” folder

• Click on “New” and select the option “Python 3 bleeding edge (using the module anaconda3/bleeding_edge)”

• A new tab opens. It looks like:

  

• Rename the Notebook. Click on “Untitled” and type in e.g. “00-my_first_notebook” (don’t use space or special characters - it’s a linux system!)

Now, you are ready to get the course material!