Hi,
we are importing a lot of small docs into. To avoid duplicates we use find_or_initialize_by with fields. Looking at the logs Mongoid has to make a query for each find_or_initialize_by call. This makes me wondering if there is a way to improve performance with Mongoid out of the box.

I could think of not using find_or_initialize_by and instead preloading corresponding data and searching with ruby methods to check if the documents already exists. However I would be nice if Mongoid had some cache I could load before through which it would a first finding attempt before actually hitting the database server.

Cheers,
Martin