Storage size calculations

This page describes the storage size of documents, fields, and index entries in Firestore with MongoDB compatibility.

You can learn about the costs of this storage at the Pricing page.

String size

String sizes are calculated as the number of UTF-8 encoded bytes + 1.

The following are stored as strings:

• Collection name
• Field names
• String field values (including _id)

For example:

• The collection name tasks uses 5 bytes + 1 byte, for a total of 6 bytes.
• The field name description uses 11 bytes + 1 byte, for a total of 12 bytes.

Field value size

The following table shows the size of field values by type.

For example, a boolean field named done would use 6 bytes:

• 5 bytes for the done field name
• 1 byte for the boolean value

Document size

The size of a document is the sum of:

• The string size of the collection name
• The sum of the string size of each field name (except _id)
• The sum of the size of each field value (including _id)
• 48 additional bytes

This example is for a document in collection tasks:

{
  "_id": "my_task_id",
  "type": "Personal",
  "done": false,
  "priority": 1,
  "description": "Learn Cloud Firestore"
}

The total size of the fields is 78 bytes:

So the document size is 6 + 78 + 48 = 132 bytes:

• 6 for the collection name
• 78 bytes for the fields
• 48 additional bytes

Index entry size

The size of an index entry in an index is the sum of:

• The string size of the collection name
• The size of the _id field value
• The sum of the indexed field values
• 48 additional bytes

Consider a document in the tasks collection:

{
  "_id": "my_task_id",
  "type": "Personal",
  "done": false,
  "priority": 1,
  "description": "Learn Cloud Firestore"
}

For an index on the done and priority fields (both ascending), the total size of the index entry in this index is 70 bytes:

• 6 bytes for the collection name tasks
• 11 bytes for the _id field value
• 1 byte for the boolean field value
• 4 bytes for the 32-bit integer field value
• 48 additional bytes

For sparse indexes, if a document doesn't include any of the fields, then no index entry is created. If a document contains at least one of the indexed fields, an index entry is created with absent indexed fields set to NULL.

Change stream event entry size

The size of a change stream event is the total of:

• The sum of the string size of the collection name (x2).
• For insert and update events for a document:
  • The sum of the string size of each field name in the fullDocument or the updateDescription (except _id).
  • The sum of the size of each field value in the fullDocument or the updateDescription. (including _id).
• If applicable to multi-document transactions, an additional 24 bytes for the lsid and txnNumber.
• 92 additional bytes

Consider an example for an insert event for a document in the tasks collection:

{
  "_id": { <Resume Token> },
  "operationType": "insert",
  "clusterTime": <Timestamp>,
  "wallTime": <ISODate>,
  "ns": {
     "db": "db",
     "coll": "tasks"
  },
  "documentKey": {
     "_id": "my_task_id"
  },
  "fullDocument": {
     "_id": "my_task_id",
     "description": "Learn Cloud Firestore"
  },
}

The total size of the change stream event is 149 bytes:

• 92 bytes for general metadata
• 12 bytes based on the collection name tasks (6 bytes) * 2
• 11 bytes for the _id field value
• 12 bytes for the description field name
• 22 bytes for the description field value

Text search index entry size

The size of a text search index entry in an index is the sum of:

• The string size of the collection name
• The size of the _id value
• The sum of bytes from indexed field values (x2)
• 48 additional bytes for general metadata

Consider an example for an insert event for a document with _id my_task_id in the tasks collection:

{
    "_id": "my_place",
     "type": "Restaurant",
     "visited": false,
     "priority": 1,
     "location": GeoPoint(longitude, latitude)
}

The total size of a text search index entry on description is 105 bytes based on:

• 6 bytes for the collection name tasks
• 11 bytes for the _id value
• 44 bytes, based on 22 bytes for the description field x2
• 48 additional bytes for general metadata

Geospatial index entry size

The size of a geospatial index entry in an index is the sum of:

• The string size of the collection name
• The size of the _id value
• 128 bytes for each indexed geo point
• 48 additional bytes for general metadata

Consider an example for an insert event for a document with _id my_place in the places collection:

{
    "_id": "my_place",
     "type": "Restaurant",
     "visited": false,
     "priority": 1,
     "location": GeoPoint(longitude, latitude)
}

The total size of a geospatial index entry on location is 192 bytes based on:

• 7 bytes for the collection name places
• 9 bytes for the document ID
• 128 bytes for the location field
• 48 additional bytes for general metadata

What's next

Learn about pricing.