Intro to Princeton's Research Computing Clusters

Important Background

To begin, read through the What is a cluster? page. This page covers not only the basic concept of a cluster, but also the basics of how Princeton's Research Computing clusters work, and why you might need to use them.

If you are unfamiliar with parallel programming, we provide a very basic introduction to parallel programming concepts.

If you are in the humanities or social sciences then be aware of these specialized training options.

Connecting to the Clusters

Before you can connect, ensure you've taken the necessary steps to get an account for one of Princeton's Research Computing clusters.

Once you have an account, you can connect by web or connect by ssh. You can access all clusters by ssh, but you can only connect to the Adroit, Della, or Stellar clusters by web.

How to Work with Files on the Clusters

Research Computing's clusters contain several folders (/home, /scratch, and /tigress or /projects) designed for specific uses. To understand how to properly work with and store different types of files on Research Computing's clusters, review our data storage page. 

To transfer files onto or off of the clusters, we generally recommend the 'scp' command, as explained in detail on our Transfer Files page. If you feel uneasy with Linux commands such as 'scp,' and you are working on the Adroit, Della, or the Stellar cluster, you also have the option of connecting through a web browser and using the upload/download buttons to transfer files (buttons are found in the pages under the "Files" menu in the OnDemand interface).

To work with visualization files, view our guide to working with visualizations and graphical user-interface (GUI) applications.