Resume: shredding versus embedding
Resume: shredding vs embedding
Shredding: tables + JOIN
users: PK user_id
positions: FK user_id
education: FK user_id
read: JOIN on every read
write: one row at a time
Embedding: one JSON doc
{ _id: 251, name: ... }
jobs: [ {...}, {...} ]
education: [ {...} ]
} // a single read
read: one lookup
write: rewrite + copies
one-to-few tree? embed it. many-to-many web? reference it.