Parent-Child Relationship

• Similar in nature to the nested model: both allow you to associate one entity with another

• Parent-Child linked documents will be stored in the same shard, possible trouble with data distribution

• With nested datatype all data lives within the scope of parent document

• With parent-child: parent and each child are indexed separately

• Children can be queried, updated, and deleted separately from the parent and from each other

• With large number of child documents parent-child indexing is more effective

• Child documents can be incorporated into query criteria and into the query results

• Parent-Child maps are stored in so called doc values quick in-memory processing and scalable split on disk

• There are still performance concerns voiced in Elastic documentation