I have a couple very large libraries in my flask app's requirements.txt which I upload to Google App Engine for deployment, but I only load small submodules for my app.
I'm trying to understand what the hard disk space looks like for each google cloud instance type to see which one to specify and avoid hitting memory limits.
In terms of Hard Disk space, you'll simply be charged if you exceed the free quota. It isn't controlled by instance class
Per the documentation
Memory is different from hard disk space. All things being equal, you can assume that the more libraries/modules you use, the more memory you'll use up but it also depends on how your app is architected.
For example, if you have stuff which run as tasks when you barely have any traffic and that task uses a library which is large, you could potentially not hit your memory traffic simply because your App is only working on that task. If on the other hand, it's running in the middle of your busy traffic time, then it's using that memory intensive library in addition to serving traffic to regular users