Preparation

  1. To use NSF ACCESS resources, you need first to create an account HERE. Once you have created the account, please inform Prof. Shi or whoever in charge of the project, then they will add you to the project and allocation.
  2. After you are added to the project and allocation, you need to wait until the allocation resources provider (such as NCSA) creates an account for you. This may take few days. Once your login account is ready, you will receive an email from the resource provider. Follow instructions in the email to reset your password and set up the DUO authentication.

Login

Once everything is ready, we can login to the resource system. Here we use NCSA (National Center for Supercomputing Applications) as an example. To login to machine:

  1. Find your NCSA username is in your ACCESS Profile; once logged in, scroll to the bottom of the page to the “Resource Provider Site Usernames” table. If you don’t know your NCSA username, submit a support request for assistance.
  2. Open the Terminal (Linux/MacOS), or Command Prompt (Windows) on your laptop, or your favorite terminal software (Xshell/Xftp) to establish SSH connection. For example, in Command Prompt, type ssh [email protected]. Replace username with your actual NCSA username. And login.delta.ncsa.illinois.edu is the host address for the NCSA login node.
  3. Type in your password and get through the two-factor authentication, then you should be all set!
  4. (Optional but highly recommended!) If you want to make your command line results colorful, you can follow this small tip!

<aside> 💡 You can find more information on their Quick Start page.

</aside>

File Systems

<aside> ❗ No Delta file systems (internal or external) have backups or snapshots of any kind. You are responsible for backing up your files. There is no mechanism to retrieve a file if you have removed it, or to recover an older version of any file or data.

</aside>

Python & Anaconda

I recommend using Anaconda to manage Python libraries and versions. To use Anaconda, type module load anaconda3_gpu in your command line to load corresponding module. Then you can use typical commands to create, load, and delete Conda environments.

Run a gRASPA Simulation