The propagation layer sits on top of the inbound and harmonization layer and is responsible for providing data in a highly harmonized, consistent, and complete form for further distribution and multiple use cases. Let's explore how graphical views and SQL Views can be used in this layer to combine data from different sources.
Combining Data with Graphical Views
Views in the propagation layer are typically used to join or union views or tables from the harmonization layer to bring data together from multiple source systems into a single, unified dataset.
When you combine data sets in a union, you can choose from the following options:
- UNION: Removes duplicate rows from the combined result. After merging, it performs a distinct sort, so every row in the output is unique.
- UNION ALL: Keeps all rows, including duplicates. No de-duplication step is performed.
In SAP Datasphere specifically, UNION ALL is the more common choice in the graphical view union node when combining data from multiple sources (like combining CSV files, SAP S/4HANA, and SAP BW/4HANA sales data), because the source systems are distinct and you want every row preserved for aggregation downstream.
You typically use a view with semantic usage relational dataset when the entity has no direct analytical or reporting requirement.
Combining Data with SQL Views
Operations to combine data are also available when using SQL Views.
In a SQL View you use the SQL editor. You can choose between writing a standard SQL query using SELECT statements with operators such as JOIN, UNION and UNION ALL, or use SQL Script to produce a table function.
Note
