Using DAC for implementation of two change capture views on same base table

Question

Hello
We have a need to create 2 views (V_INS_CLAIM & V1_INS_CLAIM) on a single table (S_INS_CLAIM) using DAC & Image tables.
The V_INS_CLAIM will be used to load ODS on a hourly basis and
The V1_INS_CLAIM will be used to load DataMart on a daily basis.
For this purpose, we propose using the following approach:
1) Create 2 separate containers - one for ODS and one for Datamart.
2) Use a separate image table suffix (for the same table) on each container e.g.
- Use image suffix 51 on S_INS_CLAIM for the ODS container (will result in image tables having suffix 51)
- Use image suffix 61 on S_INS_CLAIM for the Datamart container (will result in image tables having suffix 61)
3) In the ODS container, the change capture process would create a view V_INS_CLAIM using
the base table S_INS_CLAIM and the I/R/D tables using image suffix 51
For this folder the 'Drop and Create Change Capture Views Always' flag will be set to True
All views for this container will be created with the prefix "V_"
4) For the Datamart container, we propose to set the flag 'Drop and Create Change Capture Views Always' to False.
This would ensure that DAC never drops/creates the view (assuming that the view was already created by the ODS container)
For creating the change capture views, we propose to create the views using sql procedures.
All views for this container will be created with the prefix "V1_"
The above procedure would ensure that we have distinct views created for each target (ODS & Datamart), which are refreshed at different intervals.
Alternatively, instead of creating 2 views with different names in the SIEBEL schema,
we could create 2 schemas - one for ODS and the other for Datamart and create the change capture
views separately in each schema (using separate I/R/D image tables - as mentioned above).
Please let me know which of the above 2 approaches would work and supported by DAC and which would be the better option ?
Thanks

Answer 1

1.) Wrong forum. Go here: Business Intelligence Applications
2.) Soft delete or hard delete? Sound to me like you want to get rid of teh records. I.e. deleting history.
3.) Normal upsert mapping swon't do, you'll need delete mappings (depending on #2)