Introduction to research computing on the Palmetto cluster

Transferring files to/from the cluster

Learning objectives

  • Learn how to transfer files to or from the cluster
  • Learn how to download data from the web

Let’s start by logging in to the cluster and having a look around:

$ pwd

If this is your first time logging in to the cluster, this directory /home/username will be empty:

$ ls

Let’s populate the directory with some data. The data from Nelle’s experiments is available online at the following location:

To download this .zip file to your Desktop:

  1. Click on the “Raw” button - this downloads the .zip file (presumably to your downloads directory).
  2. Move the .zip file to the Desktop.

Now, we’d like to transfer this file to our home directory on the Palmetto cluster. How we do this depends on what operating system we run on our local machines.

Using scp (Mac OS X)

Mac OS X/Linux systems provide a command-line utility called scp that is used to transfer files to and from remote computers. The basic form of an scp command to transfer a file from the local machine to a remote computer is:

$ scp /path/to/source/file user@hostname:/path/to/destination/directory

Thus, to copy the file from your local machine’s Desktop to your home directory on the Palmetto, you can do the following:

$ scp ~/Desktop/

Using a file transfer client (Windows)

On Windows machines, a simple way to transfer files between your local machine and the Palmetto cluster is via the SSH client. Simply click on the file transfer button as shown below. This splits the screen in two halves, so that you can drag files in your local machines and copy them over to locations on the cluster:

Structure of the Palmetto cluster

Once you’ve downloaded the data, you can check the contents of your home directory:

$ ls

You can use unzip to extract the contents of this .zip file:

$ unzip
   creating: data-palmetto/
  inflating: data-palmetto/.bash_profile

If the data is successfully unzipped, you should see a new directory in your home directory:

$ ls