> ## Documentation Index > Fetch the complete documentation index at: https://docs.ditto.live/llms.txt > Use this file to discover all available pages before exploring further. # Databases and Collections > Ditto stores data records as documents (specifically JSON-like documents) which are gathered together in collections. An database stores one or more collections of documents. ## Databases In Ditto, a database is a grouping of collections. All databases that embed the Ditto SDK and share the same database ID will automatically form a mesh network and synchronize data with each other, either directly or via Ditto Server. ## Collections Ditto stores documents in collections. Collections are analogous to tables in relational databases. Collections allow you to group similar documents together within the same namespace.

### Structuring Collections While it is good practice for documents within a collection to share a similar structure — for instance, all car-related documents stored in the `cars` collection contain fields `'make'`, `'model'`, and `'year'` — structural uniformity is not mandatory. There is no limit to the number of collections you can have in your data model, so create as many collections as you need. ### Managing Collections Every document must be associated with a collection, even if only one document exists within that collection. There are no explicit steps to manage collections; Ditto implicitly creates collections when you first store data for that collection. ### Querying Collections When querying, database interactions occur with collections rather than individual documents. For instance, the query below searches the entire `cars` collection once executed. ```sql theme={null} SELECT * FROM cars ``` You can read more about accessing data within collections in [Accessing Data](/key-concepts/accessing-data).