Skip to content

Quick Start

Users can connect to the Link HPC through two methods: SSH Clients or a Web Browser. In either case the connection URL is: link.phys.wvu.edu

Regardless of the methodology, users will have to use Duo 2-Factor Authentication to access the resources, as per WVU IT Security regulations.

SSH Connections

If you are physically on the WVU Networks (but not the WVU Guest Network), you can connect directly to Link HPC via your favorite SSH client or through your Unix Terminal (Linux / Mac):

ssh -X <WVU Username>@link.phys.wvu.edu

However, if you are connecting from off-campus, you must enter through the WVU SSH Gateway (ssh.wvu.edu). You can do this via a SSH Tunnel:

ssh -X <WVU Username>@ssh.wvu.edu
ssh -X link.phys.wvu.edu

Web Browser

Regardless of where in the world you are located, you can always access our Open OnDemand instance to connect to Link HPC by simply visiting https://link.phys.wvu.edu. You'll be directed to the SSO log-in for WVU. Once successfully authenticated, you'll be redirected to our OnDemand portal where you can launch Interactive Sessions such as a Virtual Desktop, Jupyter Notebook instance, and a terminal prompt.

Once connected to Link HPC using your WVUID, you will need to edit your ~/.bash_profile file.

However, to make things easy on our users, we have created a script which will automatically set up your default shell environment for pulsar-related science.

You can activate this script by running the following command (DO NOT ADD THIS TO YOUR ~/.bashrc or ~/.bash_profile; SIMPLY RUN THE COMMAND FROM TERMINAL):

source /scripts/SetUp_PulsarEnv.sh
Walkthrough

You can do this either through SSH using your favorite command-line text editor (vim, nanon, etc) -OR- via the text editor on the OnDemand Web Interface. Regardless of your editing preference, you need to add the following lines to your environment:

module load openmpi/4.1.4 python/3.9 pulsar

You then need to either restart your Shell session -OR- run the following via the command-line:

source ~/.bash_profile

After your environment has been modified correctly, you will then need to run:

conda init

This will load the pre-installed version of Anaconda located at /opt/miniconda3/ and allow you to start using conda controlled environments.

To use the pre-installed pulsar python packages, please run the following command:

conda activate pulsar

This will set your current conda environment from the (base) environment to (pulsar), which has several scientific packages pre-installed and optimized for our cluster's environment.

If you want to make this permanent, you need to add the above command to the end of your ~/.bash_profile. You can do this by running:

echo 'conda activate pulsar' >> ~/.bash_profile

Should you need to do any development work, you can create your own environment using the following commands:

conda deactivate
conda create --name ${USER} --clone pulsar
conda activate ${USER}

This will:

  • deactivate the pulsar environment

  • create your own personal conda environment (located in ~/.conda/envs/${USER}) which is a clone of pulsar

  • activate your personal conda environment.

The name of this environment will be your User Name.

Mamba vs Conda

We use mamba in place of conda here as it speeds up the process of installing Anaconda packages dramatically. Whenever you need to install packages, please use mamba install rather than conda install. You can find more information here: https://mamba.readthedocs.io/en/latest/

Last update: January 12, 2023
Created: August 3, 2022