Page 97 - DCAP603_DATAWARE_HOUSING_AND_DATAMINING

Unit 5: Data Warehouse Research – Issues and Research

data. Data reconciliation is based on a comparison of the data loaded into business intelligence notes
and the application data in the source system. You can access the data in the source system
directly to perform this comparison.
The term productive DataSource is used for DataSources that are used for data transfer in the
productive operation of business intelligence. The term data reconciliation DataSource is used for
DataSources that are used as a reference for accessing the application data in the source directly
and therefore allow you to draw comparisons to the source data.

5.2.1 Model

Figure below shows the data model for reconciling application data and loaded data in the
data flow with transformation. The data model can also be based on 3.x objects (data flow with
transfer rules).

The productive DataSource uses data transfer to deliver the data that is to be validated to BI.
The transformation connects the DataSource fields with the InfoObject of a DataStore object that
has been created for data reconciliation, by means of a direct assignment. The data reconciliation
DataSource allows a VirtualProvider direct access to the application data. In a MultiProvider,
the data from the DataStore object is combined with the data that has been read directly. In a
query that is defined on the basis of a MultiProvider, the loaded data can be compared with the
application data in the source system.
In order to automate data reconciliation, we recommend that you define exceptions in the
query that proactively signal that differences exist between the productive data in BI and the
reconciliation data in the source. You can use information broadcasting to distribute the results
of data reconciliation by e-mail.

LoveLy professionaL university 91