Reading the documentation, it's not really clear.
What I want is to be able to store and retrieve simple json documents. With CloudSearch it seems possible to store documents in SDF format, and then search for them but it only returns the document ID and a small part (200 chars I think) of the fields specified.
Is there a way to retrieve the full document by ID just using CloudSearch? or is it intended to work as an additional tool for searching and then using your primary storage service?
If you index the id as a literal and search with that exact id then yes you can but it seems like a waste to use CloudSearch in that way. What about S3?