Frequently asked questions about materialized views
Before using materialized views, be aware of the known limitations and test them against your application requirements to determine if materialized views are suitable for your environment.
After materialized views are deployed, regular maintenance repairs are required to ensure that base tables and views are consistent. Provided that limitations are validated against the application and Consistency considerations are observed, materialized views can be deployed in production environments.
This choice depends on the use case and requirements. Ensuring consistency between views and tables in the face of complex failures and concurrent updates requires additional mechanisms (such as row locking, view repair, and paired view replication), which requires extra work. In practice, no guarantees are lost when using built-in materialized views versus manually denormalized tables.
One differentiator of doing manual denormalization versus using materialized views is when consistency is less important, or data is never updated or deleted. In these instances, write to multiple tables from the client rather than using materialized views.
DataStax recommends using the most recent version of DSE 6.x and later to get the most stable version of materialized views. Several bugs fixes and performance improvements were added for materialized views in these versions.