Facts and Dimensions
Fact Table
The Fact Table identifies the name of the table in the database from which
facts are retrieved. One fact table is usually associated with multiple
dimension tables - this is a star schema. If each dimension table is
in turn associated with other dimension tables, then it is a snowflake
schema. Both schema types are supported. This table name is also used as the
name of the directory within the Home Directory where indexes are stored.
Fact SQL
By default the SQL used for reading the facts is of the form:
SELECT * FROM FactTable
where FactTable is the name of the Fact Table described above.
If only a subset of columns are required then you can save memory on
the server by only selecting those columns.
Dimensions
The Dimensions list shows the dimensions tables that are available for
indexing. At this point we just need to identify what tables are available
to the cube. We will connect them into a star or snowflake structure on
the Indexes panel.
Dimension Dialog
Name
The Name of the database table that holds the dimension information.
You can either type this name, or select a table name using the
chooser button. The chooser will only work correctly if the connection information
is valid. Using the chooser is often preferred as it will also set the
Primary Key, when one can be identified.
Primary Key
The Primary Key is the name of the column in the dimension table that
uniquely identifies each row. This is the column that will be mapped to the
foreign key held by the Fact Table (or another Dimension, if using a
snowflake schema). The value will be filled automatically when using the
chooser button for Dimension Name.
SQL Override
By default the system will use
SELECT * FROM DimensionName
to read dimension data from the database. If there is extra information
in the table that isn't required, you can save memory by overriding the
default SQL. Leaving this blank will revert to the default SQL.