When I enable data capture on a table a new table gets created to keep the changes in (cdc.schema_TableName_CT). That's a lot of duplicate data. Is it possible to somehow reduce the number of columns in the _CT table to the absolute minimum (unique key)?
Having this unique row identifier you can select the rows in your original table during your ETL process and use the data in there to update/delete/insert in your DWH.
At this moment, if you remove some columns from tracking, no changes are being captured. That sounds obvious of course.
I just want to save disk space for large fast changing tables.
Microsoft is conducting an online survey to understand your opinion of the Technet Web site. If you choose to participate, the online survey will be presented to you when you leave the Technet Web site.