Data warehouse queries using SPARQL
View Patent ↗Disclosed is a system allowing to query data warehouses using SPARQL. An aspect of the system may support the representation of multidimensional data as virtual graphs. Another aspect of the system may provide mapping of SPARQL queries directed against multidimensional data vis-à-vis the graphs to native queries directed against the multidimensional data. Responses from the native queries may then be translated to a SPARQL response format.
1. A computer-implemented method for querying a multidimensional database configured according to a dimensional model, the method comprising:
at a computing device having one or more processors and memory storing one or more programs for execution by the one or more processors:
receiving a Simple Protocol And Resource Description Framework Query Language (SPARQL) query having one or more query parameters on the multidimensional database comprising references to one or more fact tables and one or more dimension tables in the multidimensional database, the multidimensional database having one or more native query languages to access the multidimensional database, none of which are SPARQL;
generating a native query from the SPARQL query using a mapping file comprising metadata that describes the multidimensional database, including:
automatically mapping one or more columns in the multidimensional database to the one or more query parameters in the SPARQL query; wherein
the mapping is serialized with a target vocabulary; and
the target vocabulary includes one of: a R2RML vocabulary, a D2RQ vocabulary, or a QB4OLAP vocabulary;
wherein the native query is expressed in one of the native query languages of the multidimensional database;
executing the native query on the multidimensional database;
receiving a native result comprising data stored in the multidimensional database resulting from execution of the native query against the multidimensional database; and
generating a SPARQL result from the native result using the mapping file, the SPARQL result representing a response to the SPARQL query.
2. The method of claim 1 wherein the multidimensional database is a data warehouse, the native query language is Structured Query Language (SQL), and the native query is expressed in SQL.
3. The method of claim 1 wherein the multidimensional database is organized in a star schema.
4. The method of claim 1 wherein the multidimensional database is organized in a snowflake schema.
5. The method of claim 1 wherein the native result further comprises one or more aggregations of data stored in the multidimensional database that was retrieved in response to the native query.
6. The method of claim 1 further comprising generating the mapping file including:
extracting from the multidimensional database the metadata that describes the multidimensional database; and
generating from the metadata correspondences between objects of the multidimensional database and Resource Description Framework (RDF) triples that represent the objects of the multidimensional database.
7. The method of claim 1 wherein the native query is expressed as an Multidimensional Expressions (MDX) query.
8. The method of claim 1 , wherein the mapping includes:
mapping a first RDF class specified in the SPARKQLE query with a first column in the multidimensional database; and
mapping a first property corresponding to the first RDF class specified in the SPARQLE query with first values corresponding to the first column in the multidimensional database.
9. A database query system comprising:
a processor; and
a data storage device having stored thereon computer executable program code, which, when executed by the processor, causes the processor to:
receive a SPARQL query having one or more query parameters on the multidimensional database comprising references to one or more fact tables and one or more dimension tables in the multidimensional database, the multidimensional database having one or more native query languages to access the multidimensional database, none of which are SPARQL;
generate a native query from the SPARQL query using a mapping file comprising metadata that describes the multidimensional database, including:
automatically mapping one or more columns in the multidimensional database to the one or more query parameters in the SPARQL query; wherein
the mapping is serialized with a target vocabulary; and
the target vocabulary includes one of: a R2RML vocabulary, a D2RQ vocabulary, or a QB4OLAP vocabulary;
wherein the native query is expressed in one of the native query languages of the multidimensional database;
execute the native query on the multidimensional database;
receive a native result comprising data stored in the multidimensional database resulting from execution of the native query against the multidimensional database; and
generate a SPARQL result from the native result using the mapping file, the SPARQL result representing a response to the SPARQL query.
10. The system of claim 9 wherein the multidimensional database is a data warehouse, the native query language is Structured Query Language (SQL), and the native query is expressed in SQL.
11. The system of claim 9 wherein the multidimensional database is organized in a star schema or a snowflake schema.
12. The system of claim 9 wherein the native result further comprises one or more aggregations of data stored in the multidimensional database that was retrieved in response to the native query.
13. The system of claim 9 wherein the computer executable program code, which, when executed by the processor, further causes the processor to:
extract from the multidimensional database the metadata that describes the multidimensional database; and
generate from the metadata correspondences between objects of the multidimensional database and Resource Description Framework (RDF) triples that represent the objects of the multidimensional database.
14. The system of claim 9 wherein the native query is expressed as an Multidimensional Expressions (MDX) query.
15. A non-transitory computer-readable storage medium having stored thereon computer executable program code, which, when executed by a computer, causes the computer to:
receive a SPARQL query having one or more query parameters on the multidimensional database comprising references to one or more fact tables and one or more dimension tables in the multidimensional database, the multidimensional database having one or more native query languages to access the multidimensional database, none of which are SPARQL;
generate a native query from the SPARQL query using a mapping file comprising metadata that describes the multidimensional database, including:
automatically mapping one or more columns in the multidimensional database to the one or more query parameters in the SPARQL query; wherein
the mapping is serialized with a target vocabulary; and
the target vocabulary includes one of: a R2RML vocabulary, a D2RQ vocabulary, or a QB4OLAP vocabulary;
wherein the native query is expressed in one of the native query languages of the multidimensional database;
execute the native query on the multidimensional database;
receive a native result comprising data stored in the multidimensional database resulting from execution of the native query against the multidimensional database; and
generate a SPARQL result from the native result using the mapping file, the SPARQL result representing a response to the SPARQL query.
16. The non-transitory computer-readable storage medium of claim 15 wherein the multidimensional database is a data warehouse, the native query language is Structured Query Language (SQL), and the native query is expressed in SQL.
17. The non-transitory computer-readable storage medium of claim 15 wherein the multidimensional database is organized in a star schema or a snowflake schema.
18. The non-transitory computer-readable storage medium of claim 15 wherein the native result further comprises one or more aggregations of data stored in the multidimensional database that was retrieved in response to the native query.
19. The non-transitory computer-readable storage medium of claim 15 wherein the computer executable program code, which, when executed by the computer, further causes the computer to:
extract from the multidimensional database the metadata that describes the multidimensional database; and
generate from the metadata correspondences between objects of the multidimensional database and Resource Description Framework (RDFS triples that represent the objects of the multidimensional database.