tmsangohan/accelduck

Docker for accelerator physics

accelduck

**Docker container based on debian providing a debian linux environment for the developement and usage of code for accelerator pyhsics - start working on your actual problems in minutes

Getting Started

Prerequisites

• Install and setup Docker CE
• Install git

>>  apt-get install git

Installation

• get your local copy of this git repository

>> git clone 

• download required, prorietary software to the directory "3rdparty"
• comment other proprietary software in the Dockerfile
• build the docker image

>> docker build . -t debian:stretch_accelduck

Usage

• make runfile executable

>> chmod +x run 

• run the docker image

>> ./run

• activate the python 3 virtualenv

>> source /root/venv_python3/bin/activate 

• start JupyterLab

>> jupyter-lab --notebook-dir=/mnt/dockershare/ --no-browser --allow-root --NotebookApp.token='yourpassword' 

• access jupyter lab from the webbrowser of your host machine and start working...

>> localhost:8888

Distribution of the created Docker Image to colleagues (skip download heavy build process)

• Save the Docker image existing in your local Docker registry after the installation procedure.

>> docker save -o /PATHTOFILE/accelduck.tar debian:stretch_accelduck

• Copy the created tar file and the runfile to your colleagues machine
• Add the Docker image built on your machine to your colleagues local repository

>> docker load -i /PATHTOFILE/accelduck.tar

• Follow Usage Instructions

included opensource tools

accelerator specific

• accpy
• ozelot

generic

• jupyter-lab
• python 2.7
• python 3.5
• numpy
• scipy
• pandas
• tensorflow
• octave
• mayavi
• vtk
• blender
• julia

tested but proprietary tools

accelerator specific

• elegant
• mad-x
• femm
• pyepics
• epics

generic

• matlab
• mathematica

Authors

• Felix Armborst - Initial Work

License

This project is licensed under the copyleft GNU GENERAL PUBLIC LICENSE Version 3 - see the LICENSE.md file for details