Using Open ODS Views in Modeling Scenarios

In the next step, DataSources are generated based on the Open ODS Views. Let's see how you can use the option Generate DataFlow.

A DataSource can be generated from an Open ODS View.

When you use the option Generate DataFlow on an Open ODS View in a first step, the following happens:

• The system creates a DataSource in the NODESNOTCONNECTED application component. The fields of this DataSource are copied from the fields of the structure of the Open ODS View.

• The source object of the Open ODS View will be changed to the DataSource by the system.

After this first generation step, you can then use the option Generate DataFlow on the Open ODS View (now based on a DataSource) again. This will offer two scenarios:

• Scenario 2A - Generation of an InfoSource and a Transformation
• Scenario 2B - Generation of a DataStore Object (advanced), a Transformation and a Data Transfer Process

When an Open ODS View is based on a DataSource, the following happens when you use the option Generate DataFlow with Target Type InfoSource based on a DataSource:

• The system creates an InfoSource in the NODESNOTCONNECTED application component. The fields of this InfoSource are copied from the fields of the structure of the Open ODS View. Associations to other Open ODS View fields are ignored when creating the InfoSource.

• The system creates a Transformation between the DataSource and the InfoSource.

• The source object of the Open ODS View will be changed to the Transformation by the system.

When an Open ODS View is based on a DataSource, the following happens when you use the option Generate DataFlow with Target Type DataStore Object (advanced).

• The system creates a DataStore Object (advanced) with an active table in the InfoArea of the Open ODS View. The fields of this DataStore Object (advanced) are copied from the fields of the structure of the Open ODS View. Associations to other Open ODS View fields are ignored.

• The system creates a Transformation and DTP between the DataSource and DataStore Object (advanced).

• The source object of the Open ODS View will be changed to the DataStore Object (advanced) by the system.

The source of the Open ODS View is now the active table of the DataStore Object (advanced). Without loading the data, this data model will not see any data. The outbound interface of the original Open ODS View remains unchanged. All Queries on the Open ODS View will remain unchanged, all connections to CompositeProviders are not impacted at all.

Let's have a look at an overview of the modeling scenarios with Open ODS Views that have been covered in this unit.

• The first and simplest scenario is when an Open ODS View defines a virtual data model to expose an underlying database object such as an SAP HANA database table or database view. This virtual data model allows fast integration without the need to create InfoObjects, transformations or a data flow to load data to SAP BW/4HANA in order to report on this data.

• In the second scenario, you create an Open ODS View on source data, but also implement a Transformation. You might do this for example to transform a source field in order to match the InfoObject values that you want to associate. To support the Transformation, a DataSource and an InfoSource are required, which can be automatically generated using the Generate DataFlow option . Also in this scenario, the data still resides in the source system and data is never loaded to SAP BW/4HANA.

• For the third scenario, also using the Generate DataFlow option, you can generate a DataStore Object (advanced) to persist the data. The main reason to do this is to keep a snapshot of the data. In this scenario a DTP is also generated to enable the scheduling of the data load. InfoObjects are optional, because the DataStore Object (advanced) can consist of fields.