Docs
AI
PyTorch
📈 Training
Running Jupyter Notebook/Lab on a remote server

Running Jupyter Notebook/Lab on a remote server

SSH Tunneling

  • Also known as port forwarding. I.e. setup a network tunnel (a connection for data to flow) from a local point to remote point.

  • Example:

    ssh username@xx.xx.xx.xx -NL 1234:localhost:1234

    1234:localhost:1234 means that any network request you send to port 1234 in your current system will be automatically forwarded to localhost:1234 from the remote system.

    1*uGLPZIeLPkvvaRkVG1-tkw

Run Jupyter Notebook on a remote server

0. Log on the remote server vis SSH

1. Run Jupyter Notebook from remote machine

To launch Jupyter Notebook from remote server, type the following in the remote server console:

jupyter notebook --no-browser --port=<remote_port>
  • --no-browser: this starts the notebook without opening a browser
  • --port=<remote_port>: this sets the port for starting your notebook where the default is 8888. When it’s occupied, it finds the next available port.

Note: Please note the port setting. You will need it in the next step.

2. Forward <remote_port> to <local_port> and listen to it

In your remote, the notebook is now running at the port <remote_port> that you specified. What you’ll do next is forward this to port <local_port> of your local machine via SSH tunneling so that you can listen and run it from your browser.

Type the following in the local machine terminal:

ssh -N -L <local_port>:localhost:<remote_port> <remote_user>@<remote_host>

This opens up a new SSH session in the terminal.

3. Fire-up Jupyter Notebook

To open up the Jupyter notebook from your remote machine, simply start your browser and navigate to localhost:<loacl_port>

Overview

overview

Run Jupyter Lab on a remote server

It’s similar to running Jupyter Notebook. Simply replace notebook with lab.

Reference