How to install it?
The script only needs to be installed once on a cluster, check first that someone else hasn’t installed it yet!
- Clone the GitHub repository in a shared directory on your cluster:
cd the_shared_directory git clone https://github.com/Llannelongue/GreenAlgorithms4HPC.git
myCarbonFootprint.shto load the module enabling
python3 ...to load Python 3.7 or greater (the name of the module depends on your server).
- Make the bash script executable:
chmod +x the_shared_directory/GreenAlgorithms4HPC/myCarbonFootprint.sh
cluster_info.yamlto plug in the values corresponding to the hardware specs of your cluster. You can find a lot of useful values on the online calculator GitHub.
- Run the script a first time. It will check that the correct version of python is used
and will create the virtualenv with the required packages, based on
shell script the_shared_directory/GreenAlgorithms4HPC/myCarbonFootprint.sh
How to update the tool when a new version of the tool is released
More elegant solutions welcome! Discussion here.
⚠️ Make sure you have saved your custom version of
cluster_info.yaml and the module loading step of
git reset --hardTo remove local changes to files (hence the need for a backup!)
GreenAlgorithms_global.pywith the details needed for your cluster.
chmod +x myCarbonFootprint.shto make it executable again