Querying vs documents

Hi, I have the following structure : (sorry for the silly example but I hope you get what I mean)

Document ID - person_george_king
“George”: {
“lastName”: “King”,
“salary”: “X”,
“city” “nyc”
“children” :“children_george_king”

document id -> children_george_king

"name": "child1",
"age": 5
"name": "child2",
"age": 4,

Basically, in real life, I have 3 document types that the first has reference to the second and the second to the third.
I want to get George and his childern.
Since I know querying is slower than getting a document by ID - should I create multiple requests for best performance or querying will be better in this case? and how


If you maintain sperate documents you must do do JOIN. If you few like above you can embedded as Array

Also if you know documentid you can use SDK get directly document in parallel