Combining multi-dimensional data sources using database operations

1. A computer-implemented method of linking data sources in a multidimensional database using database joins, the method comprising:

filtering join query attributes from a join query, the join query attributes including join query characteristics and join query restrictions;

identifying one or more qualified aggregated data stores that contain all of the join query characteristics and join query restrictions, each of the one or more qualified aggregated data stores redundantly storing, in a multidimensional format, only a part of the multidimensional database based on aggregate filter values for single database characteristics of the multidimensional database that are specified in an aggregated data store definition for each of the one or more qualified aggregated data stores;

selecting a smallest qualified aggregated data store out of the one or more qualified aggregated data stores for use in responding to the join query, the smallest qualified aggregated data store having a smallest number of data records of the one or more aggregated data stores;

extracting a subset of the join query attributes that are relevant to the multidimensional database;

generating a partial join query from the extracted subset of the attributes, the partial join query including database characteristics and database restrictions that are relevant to the multidimensional database;

creating a one-dimensional database view of the multidimensional database based on the smallest qualified aggregated data store;

executing the join query on the multidimensional database according to the one-dimensional database view; and

answering the partial join query using the smallest qualified aggregated data store.

2. A computer-implemented method in accordance with claim 1 , further comprising executing the partial join query on the data sources.

3. A computer-implemented method in accordance with claim 2 , further comprising generating a number of data records from the partial join query.

4. A computer-implemented method of linking two or more multidimensional data sources using database joins, the method comprising:

filtering join query attributes from a join query, the join query attributes including join query characteristics and join query restrictions;

identifying one or more qualified aggregated data stores that contain all of the join query characteristics and join query restrictions for a corresponding one of the multidimensional data sources, each of the one or more qualified aggregated data stores redundantly storing, in a multidimensional format, only a part of the corresponding multidimensional database based on aggregate filter values for single database characteristics of the corresponding multidimensional database that are specified in an aggregated data store definition for each of the one or more qualified aggregated data stores;

selecting a smallest qualified aggregated data store for each of the two or more multidimensional data sources out of the one or more qualified aggregated data stores corresponding to each of the two or more multidimensional data sources for use in responding to the join query, the smallest qualified aggregated data store corresponding to each of the two or more multidimensional data sources having a smallest number of data records of the one or more aggregated data stores;

extracting a subset of the join query attributes that are relevant to the two or more multidimensional data sources;

generating a partial join query from the extracted subset of the join query attributes, the partial join query including data source characteristics and data source restrictions that are relevant to the two or more multidimensional data sources;

creating a one-dimensional database view of the two or more multidimensional data sources based on the smallest qualified aggregated data store;

executing the join query on the two or more multidimensional data sources according to the one-dimensional database view; and

answering the partial join query using the smallest qualified aggregated data store corresponding to each of the two or more multidimensional data sources.

5. A computer-implemented method in accordance with claim 4 , further comprising executing the partial join query on the data sources.

6. A computer-implemented method in accordance with claim 5 , further comprising generating a number of data records from the partial join query.

7. A computer program product, tangibly embodied on a memory device, operable to cause a data processing apparatus to link data sources in a multidimensional database using database joins, the computer program product being configured to:

filter attributes from a join query, the attributes including join query characteristics and join query restrictions;

identify one or more qualified aggregated data stores that contain all of the join query characteristics and join query restrictions, each of the one or more qualified aggregated data stores redundantly storing, in a multidimensional format, only a part of the multidimensional database based on aggregate filter values for single database characteristics of the multidimensional database that are specified in an aggregated data store definition for each of the one or more qualified aggregated data stores;

select a smallest qualified aggregated data store out of the one or more qualified aggregated data stores for use in responding to the join query, the smallest qualified aggregated data store having a smallest number of data records of the one or more aggregated data stores;

extract a subset of the join query attributes that are relevant to the multidimensional database;

generate a partial join query from the extracted subset of the join query attributes, the partial join query including database characteristics and database restrictions that are relevant to the multidimensional database;

create a one-dimensional database view of the multidimensional database based on the smallest qualified aggregated data store;

execute the join query on the multidimensional database according to the one-dimensional database view; and

answer the partial join query using the smallest qualified aggregated data store.

8. A computer program product in accordance with claim 7 , being further configured to execute the partial join query on the data sources.