For a small site, this is not a scary idea, with one caveat:
You MUST abstract the file creation / deletion processing in one mechanism!
Make sure that all of your code uses this mechanism, which should be able to store, retrieve, and delete, and that it is. Then, in the future, if you want to change the way you store your files, you can do this very easily.
PS When I say all your code, I really mean all your code. For example, I will have a GET that matches
^img.mysite.com/(\w+)/(\w+\.jpg)$
It is processed by one method, which calls the mechanism described above.
source share