Stakeholder request for documentation on using "dask" on CURC

[monaghaa]

A stakeholder has noted that dask has been a useful tool for them to parallelize python code on Alpine, however they find it challenging to configure for our system. They suggested that we add some modest documentation to help users better understand how to use dask on Alpine. This could potentially be added to the existing python documentation:

https://curc.readthedocs.io/en/latest/software/python.html

[b-reyes]

I think this is good documentation to add. We might consider putting this under the Programming, Profiling, and Parallelization section, as this is parallelization and might take a decent amount of space to explain. I would also suggest incorporating in the JupyterLab plugin for Dask. It is an extremely useful tool when trying to understand computational resource usage, see https://github.com/dask/dask-labextension