Log In to CHTC Resources

This guide assumes that you have already gotten a CHTC account for either our high throughput or high performance compute systems. If you haven't, see our getting started page.

1. Accessing the Submit Servers

You will need the following information to log into our CHTC submit servers or head nodes:

Username and Password

• UW-Madison NetID and password

Hostname

As of December 2022, we also require two-factor authentication with Duo to access CHTC resources.

Are you off-campus?
All of our CHTC submit servers and head nodes are firewalled to block log-ins from off-campus. If you are off-campus and want to log in, you can either:

• Activate the campus Virtual Private Network (VPN) (more details on how to set this up DoIT’s VPN webpage). This will allow you join the campus network when working off-campus.
• Log into another computer that is on campus (typically by SSH-ing into that computer) and then SSH to our submit server.

In either case, it will appear like you are on-campus, and you should then be able to log into CHTC as usual.

2. Logging In

Using the information described above, you can log in to servers two different ways -- from the command line or using an SSH program:

A. On the command line

On Mac, Linux, and modern Windows (10+) systems, you can use the "Terminal" application to log in. Open a terminal window and use the following command to connect to the appropriate server:

$ ssh username@hostname

You will be prompted for your password, and then for Duo authentication.

B. Using an SSH program (Windows)

There are multiple programs to connect to remote servers for Windows. We recommend "PuTTy", which can be downloaded here. To log in, click on the PuTTy executable (putty.exe). You should see a screen like this:

Fill in the hostname as described in part 1. You should use Port 22 and connect using "ssh" -- these are usually the defaults. After you click "connect" you will be prompted to fill in your username and password, and then to authenticate with Duo.

Note that once you have submitted jobs to the queue, you can leave your logged in session (by typing exit). Your jobs will run and return output without you needing to be connected.

C. Re-Using SSH Connections

To reduce the number of times it is necessary to enter your credentials, it’s possible to customize your SSH configuration in a way that allows you to “reuse” a connection for logging in again or moving files. More details are shown in this guide: Automating CHTC Log In

3. Learning About the Command Line

Why learn about the command line? If you haven't used the command line before, it might seem like a big challenge to get started, and easier to use other tools, especially if you have a Windows computer. However, we strongly recommend learning more about the command line for multiple reasons:

• You can do most of what you need to do in CHTC by learning a few basic commands.
• With a little practice, typing on the command line is significantly faster and much more powerful than using a point-and-click graphic interface.
• Command line skills are useful for more than just large-scale computing.

For a good overview of command line tools, see the Software Carpentry Unix Shell lesson. In particular, we recommend the sections on:

• understanding the filesystem and how to navigate it (Navigating Files and Directories)
• tab-completion (section entitled "Nelle's Pipeline, Organizing Files", in Navigating Files and Directories)
• creating files (Working With Files and Directories)
• using the star wildcard (first part of Pipes and Filters)
• writing shell scripts (Shell Scripts)