You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have searched the existing issues, and I could not find an existing issue for this feature
I am requesting a straightforward extension of existing dbt-snowflake functionality, rather than a Big Idea better suited to a discussion
Describe the feature
I know that dbt adds an order by at the end of models materialized as table having a clustering key, in order to reduce (potentially to zero) the auto-clustering costs.
Couldn't the same logic be applied also to append and delete+insert materializations?
I often apply them to models with time-aggregated data (e.g. monthly summaries), and they could surely benefit for such a feature (and I don't see any drawback: in the worst case, the auto-clustering would re-cluster the new data with the old one as it would have done anyway).
I did some testing on my own project, and found that overriding get_insert_into_sql macro should do the trick.
Describe alternatives you've considered
No response
Who will this benefit?
No response
Are you interested in contributing this feature?
No response
Anything else?
No response
The text was updated successfully, but these errors were encountered:
amychen1776
changed the title
[Feature] <title>sort inserted data to reduce auto-clustering costs
[Feature] Support ordering on Incremental Materializations
Oct 25, 2024
Is this your first time submitting a feature request?
Describe the feature
I know that dbt adds an order by at the end of models materialized as table having a clustering key, in order to reduce (potentially to zero) the auto-clustering costs.
Couldn't the same logic be applied also to append and delete+insert materializations?
I often apply them to models with time-aggregated data (e.g. monthly summaries), and they could surely benefit for such a feature (and I don't see any drawback: in the worst case, the auto-clustering would re-cluster the new data with the old one as it would have done anyway).
I did some testing on my own project, and found that overriding
get_insert_into_sql
macro should do the trick.Describe alternatives you've considered
No response
Who will this benefit?
No response
Are you interested in contributing this feature?
No response
Anything else?
No response
The text was updated successfully, but these errors were encountered: