cancel
Showing results for 
Search instead for 
Did you mean: 
Data Engineering
Join discussions on data engineering best practices, architectures, and optimization strategies within the Databricks Community. Exchange insights and solutions with fellow data engineers.
cancel
Showing results for 
Search instead for 
Did you mean: 

Materialized Views Compute

NikosLoutas
New Contributor II

When creating a Materialized View (MV) without a schedule, there seems to be a cost associated with the MV once it is created, even if it is not queried.

The question is, once the MV is created, is there already a "hot" compute ready for use in case ad-hoc refreshes of the MV occur?

1 ACCEPTED SOLUTION

Accepted Solutions

BigRoux
Databricks Employee
Databricks Employee

When a Materialized View (MV) is created in Databricks without a refresh schedule, there is no “hot” compute automatically kept ready for ad-hoc refreshes. However, the MV incurs costs associated with storage (vendor cost) because it physically stores precomputed query results in a table-like structure.  In short, a materialized view is a table that incurrs cost when it is created or modified. However, once it is sitting in place (cloud storage) there are no Databricks charges, unless of course you are modifying it or accessing it via Spark (reads, optimizations, etc).  Hope this helps. Lou.

View solution in original post

3 REPLIES 3

BigRoux
Databricks Employee
Databricks Employee

When a Materialized View (MV) is created in Databricks without a refresh schedule, there is no “hot” compute automatically kept ready for ad-hoc refreshes. However, the MV incurs costs associated with storage (vendor cost) because it physically stores precomputed query results in a table-like structure.  In short, a materialized view is a table that incurrs cost when it is created or modified. However, once it is sitting in place (cloud storage) there are no Databricks charges, unless of course you are modifying it or accessing it via Spark (reads, optimizations, etc).  Hope this helps. Lou.

NikosLoutas
New Contributor II

Thanks for the clarification, much appreciated !

BigRoux
Databricks Employee
Databricks Employee

Please select "Accept as Solution" so that others can benefit from this exchange.  Regards, Louis.

Join Us as a Local Community Builder!

Passionate about hosting events and connecting people? Help us grow a vibrant local community—sign up today to get started!

Sign Up Now