Hardware requirement for ATS and Watershed_workflow

[suraj shrestha]

Dear all,

I have been running ATS and watershed workflow on a Windows 11 machine via WSL2 (Ubuntu), but I am getting frequent problems. Among them, my Jupyter Notebook kernel crashes whenever I try to run the workflow for a basin of approximately 1,370 sq.km.

My current workstation specs are:

• CPU: Intel i7-13700 (16 cores)

• RAM: 32 GB (assigned to WSL via .wslconfig)

• Storage: 954 GB SSD

• Graphics: 128 MB

I wanted to verify if this hardware is sufficient for a study area of this size, or if these crashes indicate that I need more RAM or a different system setup. Could anyone help me determine the best workstation configuration to solve these stability issues?

Thank you all

Suraj

[Amanzi-ATS Users]

Hi Suraj,

Sorry you haven't gotten a response on this -- I have no experience using it on Windows, so I just don't know what is reasonable.  Hopefully you've found a workaround at this point.  I can say that that is more than sufficient for a watershed of this size on a native Ubuntu or Mac.  We've run WW on 30,000+ km^2 domains on beefy laptops of about this capability.

You might confirm that you aren't throttling the memory to the virtual machine -- I don't know how this works on WSL2, but on most virtual machines you can limit what portion of your machine's resources you're passing onto the virtual machine.

Ethan