Sometimes we need to combine measures from two fact tables in the same report and apply a filter to one of them. If the filter is on a dimension shared by both fact tables; that’s great. However, if the dimension is only related to one of the facts, then we have a problem: Oracle BI would show null values for a measure that’s not related to the dimension.

Imagine that we have two fact tables: Revenue and Budget. Revenue is recorded for Product but also contains information about General Ledger (GL) Account. Budget is defined for Product but not for GL Account so there’s no relationship between them. Our business model is shown in the image below. Of course in a normal repository there would be more dimensions and facts but they have been left out to clarify the example.

Continue reading this post >

If you are using an Essbase data source, you may be interested in learning more about the relationship between Essbase cubes and your Oracle BI EE platform.

It is quite common to find a dimension with multiple hierarchies in a BI project. This is managed in Oracle BI by creating a Dimension object with multiple child levels at a point that converges into the same leaf level.

This looks quite easy when the physical data source is a relational database. However, when the physical data source is an Essbase cube, the process is a bit more complex.

Continue reading this post >