Data Consistency
MongoDB gives you the flexibility to normalize or duplicate your data to optimize your application. If you duplicate data in your schema, you must decide how to keep duplicated data consistent across multiple collections. Some applications require duplicated data to be made consistent immediately, whereas other applications can tolerate reading stale data.
Use Cases
There are multiple ways to enforce data consistency in your application:
Method | Description | Performance Impact | Use Case |
|---|---|---|---|
Updates to multiple collections occur in a single atomic
operation. | Potentially high, due to read contention | Your application must always return up-to-date data, and
can tolerate potential negative performance impact during
periods of heavy reads. | |
Modify the application schema to embed related data in a
single collection. | Low to moderate, depending on document size and indexes | Your application always reads and updates the related data
at the same time. This solution simplifies your schema and
prevents the need for $lookup operations. | |
When an update occurs in one collection, triggers automatically
update another collection. | Low to moderate, with potential delays in processing triggered
events | Your application can tolerate reading slightly stale data. Users
can potentially see outdated data if they run a query immediately
after an update, but before the trigger finishes updating the
second collection. |
The best way to enforce data consistency depends on your application. To learn more about the benefits and implementation for each approach, refer to the corresponding documentation pages.
Tasks
To enforce data consistency in your application, see the following pages:
Details
The following factors can affect how you enforce data consistency.
Data Staleness
Consider how important it is that your application returns the most up-to-date data. Some applications can return data that is minutes or hours stale with no impact to the user.
For example, in an e-commerce application, a user needs to know immediately whether or not an item is available. This information is ideally kept as consistent as possible, even if it requires frequent updates.
In contrast, analytic queries are typically expected to read slightly stale data. It is not critical to keep analytic data completely consistent.
Your application's tolerance for stale data affects how to best keep data consistent. Frequently updating data in multiple collections reduces the risk that a user reads stale data. However, frequent updates can negatively impact your application's performance. When you enforce data consistency, balance user needs with performance impact.
Referential Integrity
Referential integrity ensures that when an object is deleted, all references to that object are also deleted.
For example, an application has a products collection and a
warehouse collection that contains references to the products
collection. When a product is deleted from the products collection,
the corresponding reference in the warehouse collection should also
be deleted.
If your schema requires referential integrity, incorporate logic into your application to keep references consistent. At minimum, your application logic should prevent errors when attempting to query a reference that does not exist.