I’m using Amazon S3 repository for my files. of course larger is the repository size, more I pay per month.
Users, most of the time, only look for the latest files. sometimes they will search for an old one.
With that in mind, I suggest to have a file archieving service that will move old files to an archive repository.
If the user tries to access an archieved file, the server must be clever enough to recover that file from the archive repository.
- Archive repository could be cloud server or a NAS with FTP service.
- Upload files to archive repository whenever the file repository is full or the files are too old(set by the admin. I think 1 month old is enough).
- Download files from archive repository will be slower. That’s fine this is an economical solution.
- Archive repository is not mandatory, it should be disabled by default unless the admin decides to set it up, keeping MM installation simple.
- Only the actual files are transfered the archive repository. Previews, thumbnails could stay in the file repository.