Louis Davidson continues a discussion of SQL Server’s filetable feature:
In my article “Filetables, Great For Business, Great For Hobbyist”, I implemented the basis of a hobby database I use for my Twitter accounts (@disneypicaday and @dollywoodp), where I post daily pictures of theme parks.
As my inventory of prepped pictures has grown, one thing has become a major pain. Duplicates. I have tens of thousands of pictures, and I scan them occasionally to add to my directories. Perhaps not surprisingly, the same pictures often get chosen a second time. I use a naming standard that integrates the files into my database, as well as for the copies to go out for a given day. So the second (and third and fourth) time I choose the same picture, it looks different, though it has the exact same bits.
There are some nice benefits to maintaining file metadata in a queryable format, though do read to the end to see the performance impact.