I have a problem in my SQL 2005 Data Warehouse. I'm using a slowly changing dimension data flow transformation in SSIS which will not delete any records from my DB when they are deleted from the source data. I need to keep the dimension keys, but in this one case/one table, I need to delete any records that were deleted in the source data. What I want to do is take a new data load of my dimension table and compare the two dimension tables and delete any records that are not in both tables. Both tables will be identical in every way other than the records I need to delete. Any help would be appreciated, thaks. A little info about my tables, I have project and phase tables, every project must have at least 1 phase. Users can not delete projects but can delete phases, so i can compare my phase tables by linking project number from each table and delete phase numbers that don't show up when grouped by project, I'm just not sure how my SQL code should work.

little rough trying to put words into table structure, are you able to post the schema of those affected tables

and just to be clear, you only need dimensions deleted that are not in both tables?