Skip to content

Latest commit

 

History

History
32 lines (26 loc) · 1.87 KB

README.md

File metadata and controls

32 lines (26 loc) · 1.87 KB
Raw

CANFAR Science Platform User Containers

This repository includes build recipes for containers to be launched on the CANFAR science-platform. It is currently a bit rough and work in progress. Please feel free to contribute.

CANFAR-maintained Containers

Curently the repository contains basic build system to build a hierarchy of default containers supported by CANFAR:

	                    Ubuntu LTS
	                    	|
		            base (headless)
	                    	|
		_____________________ _ _ _ _ _ _ _
		|                                  | 
astroml (headless)			Desktop Application (desktop-app)
		|
astroml-notebook (notebook) 
astroml-vscode   (contributed)
astroml-desktop  (desktop-app)
  • The base is a headless container are built from a vanilla Ubutu LTS, with extra operating system installed (compilers, development libraries), and conda install.
  • The astroml is another headless container and is built with a large set of astronomy, machine learning, deep learning, visualisations and data science libraries.
  • The astroml-* add visualisation and interactivity software. They can be launched as a notebook session, a contributed VSCode session, or a terminal through a desktop session.
  • The Desktop Application containers do not typically derive from astroml and sometimes not from base containers, as they are legacy and may rely on now unsupported OS.

There are CUDA-enabled versions of the containers, which include NVIDIA software and CUDA-powered libraries. They are built and named as *-gpu, i.e. astroml-gpu, astroml-gpu-notebook.

CANFAR User Customised Container

If you want to build your own containers, documentation can be found in the docs directory. This directory also contains the building of the CANFAR usage documentation in the readthedocs style.