MongoDB Query Performance: return all vs select fields

Question

I have unexpected results comparing the performance of various queries in a collection that I made for testing. The collection somewhat imitates my real needs with 10,000 documents, each with 20 fields (each with 5-30 characters). All documents are the same and only have _id different (maybe this is somehow a problem?).

Unlike the official version of MongoDB, indicating which fields to return do not lead to better performance, but much worse.

A typical find is performed in approximately 5 ms.

 db.collection.find().explain() 

User find is performed in approximately 30 ms.

 db.collection.find({},{Field1:1,Field2:1,Field3:1,Field4:1,Field5:1,Field6:1,Field7:1},{}).explain() 

Is the simple query “find everything” and “return everything” faster or am I missing something?

Answer 1

If you return the whole document, this is less than the overhead for the database, because it does not convert the document to return partial. All documents are stored in a database in BSON format. And he returned in the same way.

In your case, a small overhead is expected.

The field limitation is suitable for large amounts of data, when you have 10,000 resulting documents, and it is much faster to convert a document at the DBMS level, rather than transferring through the socket level.