Question: Fetching static content and file storage on deploy

I am doing some experimental image processing using the imagemagick wasm library. Which itself recently ran into issues because of the `Cache` missing module (I know it's being worked on). I had to fix this by just including the `.wasm` build in my repo. A fix was made to fetch the wasm file without using `Cache`. I can use some help trying to understand how a dependency like this wasm file and the fetch works on the edge. When a instance "cold starts" will it download this file? When does ever data-center or region have to download the file? Does each need to download the file when the instance "cold starts"? I'm facing the same issue with some images that are large currently they need to be included in the repository. I'm understand that deno deploy can't write files, so do files just need to be kept in memory? For instance If I have large image assets and I wanted them stored in the cloud (gcp or s3) I can't really fetch them on start they're too large and there's too many. It's my understanding that the fastest and most reliable way to interact with these large files is to have them in the repo, is this true or are there other mechanisms I'm not thinking of?