What means 'index object' cleary?

MongoDB Atlas Atlas Search
1

Hi, mongodb atlas coummnity, I have question.
The atlas search index performance page contains the following.

‘If you create an Atlas Search index for a collection that has or will soon have more than 2.1 billion index objects, you must use numPartitions or shard your cluster. Index partitions are limited to 2.1 billion objects per partition.’

What exactly is the ‘index objects’ you’re talking about over there?
For example, there is one document a in collection A, and the search index is as follows.

A collection

{
    "items": [
        {
            "count": 10,
            "info": {
                "name": "abcd",
                "value": "ABCD"
            }
        },
        {
            "count": 5,
            "info": {
                "name": "xyz",
                "value": "XYZ"
            }
        }
    ],
    "link": {
        "url": "https://www.google.com",
        "alias": "google"
    },
    "score": 100 
}

Search index

{
  "mappings": {
    "dynamic": false,
    "fields": {
      "items": [
        {
          "fields": {
            "count": {
              "indexDoubles": false,
              "representation": "int64",
              "type": "number"
            },
            "info": {
              "fields": {
                "value": {
                  "type": "token"
                }
              },
              "type": "document"
            }
          },
          "type": "embeddedDocuments"
        }
      ],
      "link": {
        "fields": {
          "url": {
            "type": "token"
          },
          "alias": {
            "type": "token"
          }
        },
        "type": "document"
      },
      "score": [
        {
          "indexDoubles": false,
          "representation": "int64",
          "type": "number"
        },
        {
          "indexDoubles": false,
          "representation": "int64",
          "type": "numberFacet"
        }
      ]
    }
  }
}

How many index objects are there in the above case?

2

In your example, there would be 3 index objects - one for the main document, and one for each of the items. This can explode easily, so be sure that the extra index objects via embeddedDocuments is valuable to the search use case. For more information, see the nested documents section here: Data Modeling and Schema Design for Atlas Search | MongoDB

3

Hi, @Erik_Hatcher
Thank you very much for your response.
Now I understand for sure.