Run the workshops under Windows with WSL

Data Analysis
1

Linux install on Windows with WSL (only Windows 10 or 11)

Install a Linux distribution on your Windows system. We suggest you install Debian GNU/Linux, which is the closest distribution to what is used currently by LIGO/Virgo/KAGRA.

This subsystem is a virtual machine installed in a partitioned area of your disk, so you need at least 10 Gigabytes of space available on your disk.

  • Install a windows subsystem for linux: see instructions here (Windows 10) or alternatively here (Windows 11).

  • Once the distribution is set up, you still need a few key elements:

  1. A display X-server: a good option is VcXsrv. It is a bit tricky to make it work on WSL, but following all of these steps should work. You will have to launch an X-server session every time you switch on your computer.

  2. A web browser, in order to run the Jupyter notebooks. Google Chrome is a safe option.

Once you have all that installed, follow the instructions for the IGWN software setup.

1 Like
2

I think this is a bit outdated due to

  • WSL 2 now supporting Linux GUI applications (no need for VcXsrv),

  • and there is a more lightweight Conda environment odw-py312 maintained for the course (no need to download the IGWN software which takes a lot of space).

This is what worked for me:

  1. Install WSL.
  2. Make sure it is version 2.
  3. Make sure the GPU drivers are installed and updated.
  4. Follow the steps for Linux here to download odw-py312.
  5. Before opening the notebooks, use these commands each time:
    cd ~/odw
    conda activate odw-py312
    jupyter notebook
  6. Paste this to your browser to open the notebooks:
    http://localhost:8888/tree/

(Sorry, I am not able to link more than 2 links per post as I am a new user. I can provide more helpful links if someone needs them.)

1 Like