Support data clustering for table and incremental materializations

[cheyney-w]

Now that Fabric warehouses support data clustering (currently preview), it would be useful if dbt models configured for table and incremental materialization could use this feature.

This should probably be implemented as an optional cluster_by model configuration that can accept a list of column names.

Note: some adapters use the config name cluster_by and others use clustered_by. Without a consistent convention, I think cluster_by makes sense for the Fabric warehouse adapter since it matches the CLUSTER BY clause in the T-SQL CREATE TABLE syntax.

[nickteff]

This would be a great feature to see supported!

[arasdk]

+1

It's been in preview for more than 2 months, and this would be really helpful to tune performance. The Databricks dbt adapter can configure clustering for SQL warehouses, so it would be really useful if Fabric added this also!

[ThomsenS]

This would be very useful for us too!