How to query a nested document in mongo?

Which technique is preferable aggregations  or has_child query ?

• Lets say we have Parent Document for book with it's name ISBN and  short summary (and some other stuff) and child documents for chapters. I can also save "parent doc" data for every chapter as doc_values (not_analysed string) and then aggregate results at query  time. Why this way? I want to boost score for those books where multiply chapters match. In Elasticsearch which technique is preferable : 1) Using nested documents and query them with has_child - non cached 2) Just store field from parent doc in child doc and apply aggregation - this option is going to take more disk space, but can be cached, no look-up for parent documents is required. Update:My intuition is that aggregation should cause less random IO - since IO is slowest part - aggregations should be faster.   Average number of child documents will be a few hundreds docs. Thanks

• Answer:

  Both solutions have different use so it's not easy to answer the question without knowing the use case.