Remote Access

Notes on remote access via SSH and similar tools.

Image by Growtika on Unsplash.

Data science (and other work) often requires us to work on remote computers. There are several reasons for this, including access to resources (such as GPUs or higher-performance CPUs) that are not available on our local machines, or managing long-running jobs that we don’t want to tie up our laptops. It can also be useful to work from minimal devices such as lightweight laptops or tablets, and remote access gives us a way to do this.

We refer to the computer we’re connecting from (e.g. a laptop) as our local machine, and the computer we are connecting to as the remote system.

1. Connecting with SSH
2. Connecting with VS Code
3. Port Forwarding (accessing remote services)
4. Multi-Hop Connections
5. SSH Public Key Authentication

Note

While I do some data analysis on my laptop, I do the majority of my hands-on data science work remotely either on my research group’s computer in the CCI data center or on a high-end gaming PC in my office.