Connect VSCode remotely to the GPU Server

Set up SSH (Windows)

• Open PowerShell
• Enter ssh-keygen -t ed25519
  • Press enter once
  • If you are asked Overwrite (y/n), enter n
  • Otherwise, just press enter at every prompt
• Enter $u = "alice", replacing alice with your own GPU Server username
• Enter ssh [email protected] "mkdir ~/.ssh"
• Enter type $env:USERPROFILE\.ssh\id_ed25519.pub | ssh [email protected] "mkdir -p ~/.ssh; cat >> ~/.ssh/authorized_keys"
  • If you are asked whether you are sure you want to continue, enter yes
  • Enter your GPU Sever password
• Optional: Enter ssh [email protected] "hostname" to test whether everything worked
  • You should directly get gpu-dhlab without being asked for a password

Set up SSH (Linux and probably also Mac)

• Open a terminal
• Enter ssh-keygen -t ed25519
  • Press enter once
  • If you are asked Overwrite (y/n), enter n
  • Otherwise, just press enter at every prompt
• Enter u="alice", replacing alice with your own GPU Server username
• Enter ssh-copy-id [email protected]
  • If you are asked whether you are sure you want to continue, enter yes
  • Enter your GPU Sever password
• Optional: Enter ssh [email protected] "hostname" to test whether everything worked
  • You should directly get gpu-dhlab without being asked for a password

Configure VSCode

• Install and open VSCode
• Click the ≶ Icon in the bottom left corner
• Select SSH or Connect to Host...
• Select Add new SSH Host...
• Enter [email protected], again replacing alice with your own GPU Server username
• Select the highlighted option

Connect VSCode

• Open VSCode
• Click the ≶ Icon in the bottom left corner
• Select Connect to Host...
• Select gpu-server.dhlab.intra.uni-freiburg.de
• If you are asked to selected the plaform of the remote host, select Linux

Using Jupyter Notebooks

• Create a new file ending in .ipynb
• Open the file
• In the top right click Select Kernel...
• Select Python Environments...
• Select Create Python Environment
• Choose Venv or Conda, whichever you prefer
• Choose /bin/python3

If you need a specific Python version, use uv to manage your virtual environment.