Defining a mapping for Join (ES 6.2.2) - CURL

Elastic Stack Elasticsearch
1

What is the best mapping for a join between two documents?

Actually, i'm creating a data warehouse to stock my postgres data:

table: customer
column			| type
---------------------------
id_customer 	| integer
name            | varchar
email           | varchar
enabled 		| char

table: order
column			| type
---------------------------
id_order 		| integer
id_customer 	| integer (foreign key from customer)
value 			| numeric
finishied 		| char

How i create a mapping to this both documents?
Because i will need search in orders when the customer (father) has the property enabled equal to 'Y'.

2

Or what about a document like:

PUT orders/doc/1
{
  "value": 1,
  "finishied": true,
  "customer": {
    "name": "Joe Smith",
    "email": "joe@foo.bar",
    "enabled": true
  }
}
3

Hi @dadoonet.

After a customer update will be necessary reindex all the orders, and some customer can have a huge quantity of orders.

4

After a customer update will be necessary reindex all the orders

Yes.

and some customer can have a huge quantity of orders.

How many? Like more than 1 million orders?
How long would that take?

I mean that for sure you have to pay a price at some point.

Do you want to pay the price at index time? Or at search time?

5

Thanks @dadoonet, your observation make all sense.

Is best pay the price at index time, whereas the client change occurs less frequently than orders.