The Miniconda3 environment module provides the Conda package and environment manager. Conda lets you install packages and their dependencies in dedicated environment, giving you more freedom to install software yourself at the expense of possibly less optimized packages and no curation by the NeSI team.


  • If you want a more reproducible and isolated environment, we recommend using the Singularity containers.
  • If you only need access to Python and standard numerical libraries (numpy, scipy, matplotlib, etc.), you can use the Python environment module.

Note about Māui Ancillary Nodes

The Miniconda3 module is not available on Māui Ancillary Nodes, use the Anaconda3 module instead.

Module loading and conda environments isolation

When using the Miniconda3 module, we recommend using the following snippet to ensure that your conda environments can be activated and are isolated as possible from the rest of the system:

module purge && module load Miniconda3
source $(conda info --base)/etc/profile.d/

Here are the explanations for each line of this snippet:

  • module purge && module load Miniconda3 ensures that no other environment module can affect your conda environments. In particular, the Python environment module change the PYTHONPATH variable, breaking the isolation of the conda environments. If you need other environment modules, make sure to load them after this line.
  • source $(conda info --base)/etc/profile.d/ ensures that you can use the conda activate command.
  • export PYTHONNOUSERSITE=1 makes sure that local packages installed in your home folder ~/.local/lib/pythonX.Y/site-packages/ (where X.Y is the Python version, e.g. 3.8) by pip install --user are included in your conda environments.

Prevent conda from using /home storage

Conda environments and the conda packages cache can take a lot of storage space. By default, Conda use /home storage, which is restricted to 20GB on NeSI. Here are some techniques to avoid running out of space when using Conda.

First, we recommend that you move the cache folder used for downloaded packages on the nobackup folder of your project:

conda config --add pkgs_dirs /nesi/nobackup/<project_code>/$USER/conda_pkgs

where <project_code> should be replace with your project code. This setting is saved in your ~/.condarc configuration file.

Next, we recommend using the -p or --prefix options when creating new conda environments, instead of -n or --name options. Using -p or --prefix, you can specify the conda environment folder location, ideally in your project folder. For example:

conda create --prefix /nesi/project/<project_code>/my_conda_env python=3.8

Then use the path of the conda environment to activate it:

conda activate /nesi/project/<project_code>/my_conda_env

Note that -p and --prefix options can also be used when creating an environment from an environment.yml file:

conda env create -f environment.yml -p /nesi/project/<project_code>/my_conda_env

Reduce prompt prefix

By default, when activating a conda environment created with -p or --prefix, the entire path of the environment is be added to the prompt. To remove this long prefix in your shell prompt, use the following configuration:

conda config --set env_prompt '({name})'
Was this article helpful?
0 out of 0 found this helpful