Siebel Enterprise Integration Manager Administration Guide > Importing Data > Special Considerations for Imports >

Maintaining Denormalized Columns

When updating columns that are the source of denormalized columns in other tables, you must find the records related to the columns being updated and load them as well, in the same batch.

As an example, you are updating the S_SRC table using EIM_SRC. EIM_SRC maps to S_SRC, S_SRC_BU, and S_SRC_POSTN, among others. S_SRC_BU and S_SRC_POSTN both contain the column SRC_NAME, which is denormalized from S_SRC.NAME. So, S_SRC_BU.SRC_NAME and S_SRC_POSTN.SRC_NAME should match S_SRC.NAME.

You have a record in S_SRC, and you want to update its NAME to something else using EIM_SRC. When you load the data of this record with its new NAME into EIM_SRC and then run EIM to update the NAME, EIM does not automatically update the SRC_NAME in the records within S_SRC_BU and S_SRC_POSTN. In order for the EIM engine to update S_SRC_BU.SRC_NAME and S_SRC_POSTN.SRC_NAME with these related records, you must find these related records in S_SRC_BU and S_SRC_POSTN and load them into EIM_SRC as well. The batch number must be the same. Only the user key column data needs to be loaded for these related records.

 Siebel Enterprise Integration Manager Administration Guide 
 Published: 05 January 2004