Granted Patent B2

US 9,208,183 · App. 13/843,648 · Granted Dec 8, 2015

Abstract

Claims (42)

1. A method comprising:

storing selected historical persist dimension attribute data utilizing a row insertion without updating all previous versions of the selected persist dimension attribute, the storing including,

updating a dimension record by changing an end date of the dimension record and indicating the record is not current, and

inserting a new record including indicating the new record as current, setting an attribute to a new value and setting a start date, wherein the selected persist dimension attribute data is Type-6 slowly changing dimensional data; and

generating a view, the view including,

current dimension attribute data and persisted dimension attribute data as dual values utilizing a star join, the star join causing row data to be converted to column data, and

at least one row a dual value pair including only persisted dimension attribute data corresponding to the current dimension attribute data of another row.

2. The method of claim 1 , wherein storing selected historical persist dimension attribute data includes,

generating a configuration structure to store information indicating a dimension attribute to be persisted, and

selecting the historical persist dimension attribute based on the stored information.

3. The method of claim 1 , further comprising:

generating a dimension data table including current attribute data, history attribute data, effective date data for the attribute, and data indicating an attribute is current.

4. The method of claim 1 , wherein generating the view includes,

associating foreign key columns of a fact table with a column identifying dimension data of at least one dimension table,

the association being a star join.

5. The method of claim 1 , wherein the view includes a snapshot validity interval.

6. A data warehouse comprising:

a database,

at least one processor, and

at least one memory storing code segments that when executed by the processor cause the processor to,

generate a configuration structure to store information indicating a dimension attribute to be historically persisted,

generate a plurality of data tables in the database, each of the plurality of data tables including an associated schema, the associated schema including current attribute data, history attribute data, effective date data for the attribute, and data indicating an attribute is current,

store selected historical persist dimension attribute data utilizing a row insertion without updating all previous versions of the selected persist dimension attribute, the storing including,

updating a dimension record by changing an end date of the dimension record and indicating the record is not current, and

inserting a new record including indicating the new record as current, setting an attribute to a new value and setting a start date, wherein the selected persist dimension attribute data is Type-6 slowly changing dimensional data, and

generate a view configured to convert the row data to column data based on a star join, the column data being dual values, the view including,

corresponding current attribute data and history attribute data as the dual values, and

at least one row a dual value pair including only persisted dimension attribute data corresponding to the current dimension attribute data of another row.

7. The data warehouse of claim 6 , wherein storing selected persist dimension attribute data includes,

generating a configuration structure to store information indicating a dimension attribute to be historically persisted, and

selecting the historical persist dimension attribute based on the stored information.

8. A method, comprising:

associating foreign key columns of a fact table with a column identifying dimension data of at least one dimension data table, the association being a star join, the dimension data is Type-6 slowly changing dimensional data, and

generating a view configured to convert row data to column data based on the star join, the column data being dual values, the view including,

current dimension attribute data and persisted dimension attribute data as the dual values, and

at least one row a dual value pair including only persisted dimension attribute data corresponding to the current dimension attribute data of another row.

9. The method of claim 8 , wherein each of the dimension data tables includes an associated schema, the associated schema including current attribute data, history attribute data, effective date data for the attribute, and data indicating an attribute is current.

10. The method of claim 8 , wherein the dual values include the current attribute data and the history attribute data.

11. The method of claim 8 , wherein generating the view includes,

associating foreign key columns of a fact table with a column identifying dimension data of at least one dimension table,

the association being a star join.

12. The method of claim 8 , wherein the view includes a snapshot validity interval.

Assignments (16)