With data blending, you can combine data from two or more data sources. Data blending is useful when you have multiple data sources that contain information about the same type of entity.
What is Data Blending?
Data blending is similar to a SQL join query. SQL joins combine data from different tables in a database. With SQL joins, each table must contain at least one column that refers to the same entity.
Likewise, data blending combines data from different data sources. With data blending, each data source must contain information about the same type of entity.
After blending data, you can view and use the combined set of all the data in Data Cube.
Example of Data Blending
Assume that you have three different data sources that contain customer data: a database to track customer orders, a customer relationship management system, and a file server where some additional customer information is stored.
After you connect to each data source in Data Cube, you can use data blending to combine all the information about your customers from all three systems into a single data source. You can then use the blended data source to search, visualize, and develop applications using the many features of Data Cube.
How Does Data Blending Work?
How You Configure Data Blending
When you create a data source in Data Cube, you configure data blending as follows:
-
Select a data source to combine with the current data source. The additional data source must have been crawled at least one time.
-
Map the columns from the current data source that contain the same values as the additional data source.
-
You can configure multiple column mapping schemes to focus on the data that you want to combine. You can also repeat the process to blend multiple data sources.
How Data Cube Blends the Data
When Data Cube crawls the blended data source, it combines the data sources as follows:
-
Data Cube each row in the data sources for matching data based on your mapping definitions.
-
When Data Cube finds a match, it combines the additional data from each matching data source and places all the information in the same row.
When the crawl is complete, you can view the entire combined set of data in Data Cube.
How Data Cube Blends Multiple Data Sources
When you configure multiple data sources, the combination works like an OR Boolean operator.
If any of the data from two or more data sources contains the same value based on your mapping definitions, then any additional information from the matching data sources is combined in the blended data source.
How Data Cube Blends Multiple Columns Between Two Data Sources
When you map more than one set of columns between two data sources, the combination works like an AND Boolean operator.
The data blending combines the data only if each mapped column between the data sources contains the same value.
Supported Data Sources
Data Cube supports all data sources, except HTTP sources and open data sources.