This is the first time I am posting on this particular StackExchange forum and I hope that I am using the right one for the present question. Briefly, this is what I need to do

1. I am running a web service where users can, amongst other things, upload and store files on the server.
2. What I want to do is to hive off user file storage to a different location so my server (CentOS 64 bit) can concentrate on what it does best - server side scripting and database management.
3. As things stand all user files go into subdirectories in a folder called stash that lies above DOC_ROOT. What I would like to do is
4. Transparently detect all attempts to read/write to stash/sub_folder and get/set file data on a remote server - ideally the latter would be one which replicates files like a CDN so they can be delivered from the closest/fastest location based on where the user's location.
5. Even nicer would be if for all read accesses I could provide a URL that allows the user's browser to fetch the relevant file directly without having to funnel them via my server.

I am a relative newbie when it comes to this sort of thing so I hope that I have phrased this question adequately well. From the little searching I done I gathered that WebDAV can be used to map drives to a different location on the web so perhaps that is a starting point. But if that will work I need to

1. Establish how to get WebDAV up and running on my CentOS 64 bit server.
2. Ideally, identify a service that allows this kind of file storage and provides an API I can use in my own scripting.

I'd much appreciate any help with this.