Integrating RDF data into a relational database system

• US 20060235823A1
• Filed: 04/18/2005
• Published: 10/19/2006
• Est. Priority Date: 04/18/2005
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Chong, Eugene Inseok, Das, Souripriya, Eadon, George, Srinivasan, Jagannathan
• Granted Patent: US 8,719,250 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2455 Query execution
• CPC Class Codes: G06F 16/25 Integrating or interfacing systems involving database management systems

Abstract

The TABLE function mechanism available in a RDBMS is used to integrate RDF models into SQL queries. The table function invocation takes parameters including an RDF pattern, an RDF model, and an RDF rule base and returns result rows to the SQL query that contain RDF triples resulting from the application of the pattern to the triples of the model and the triples inferred by applying the rule base to the model. The RDBMS includes relational representations of the triples and the rules. Optimizations include indexes and materialized views of the representations of the triples, precomputed inferred triples, and a method associated with the TABLE function that rewrites the part of the SQL query that contains the TABLE function invocation as an equivalent SQL string. The latter technique is generally applicable to TABLE functions.

Claims

1. Apparatus for integrating a non-SQL query written in a non-SQL language into a relational database management system of the type wherein an SQL query may contain an invocation of a table function and the table function, when executed, returns a set of rows to the containing query, the apparatus comprising:

a table function invocation in the SQL query which includes a parameter that specifies the non-SQL query; and

query execution code associated with the table function that, when executed, executes the non-SQL query on data that is accessible to the relational database management system and returns results of the execution of the non-SQL query in the returned set of rows.

2. The apparatus set forth in claim 1 further comprising:

SQL query string generating code associated with the table function that, when executed, creates an SQL query string that is equivalent to the non-SQL query, the relational database management system executing the query string generating code and replacing the table function invocation in the containing query with the generated SQL query string prior to executing the containing query.

3. A storage device that is accessible to a processor, the storage device being characterized in that:

the storage device contains code which, when executed by the processor, implements the apparatus set forth in claim 1.

4. Apparatus for integrating a non-SQL query written in a non-SQL language into a relational database management system of the type wherein an SQL query may contain an invocation of a table function and the table function, when executed, returns a set of rows to the containing query, the apparatus comprising:

a table function invocation in the SQL query which includes a parameter that specifies the non-SQL query; and

SQL query string generating code associated with the table function that, when executed, creates an SQL query string that is equivalent to the non-SQL query, the relational database management system executing the query string generating code and replacing the table function invocation in the containing query with the generated SQL query string prior to execution of the containing query.

5. A storage device that is accessible to a processor, the storage device being characterized in that:

the storage device contains code which, when executed by the processor, implements the apparatus set forth in claim 4.

6. A method of integrating a non-SQL query written in a non-SQL language into a relational database system of the type wherein an SQL query may contain an invocation of a table function, the table function, when executed, returning a set of rows to the containing SQL query and the method comprising the steps performed in the relational database system of:

receiving a containing SQL query with a table function invocation which includes a parameter that specifies the non-SQL query;

during execution of the containing SQL query, executing the specified non-SQL query on data that is accessible to the table function; and

returning results of the execution of the non-SQL query in the set of returned rows.

7. The method set forth in claim 6 further comprising the steps performed prior to execution of the received containing SQL query of:

generating an SQL query string that is equivalent to the specified non-SQL;

query; and

replacing the invocation of the table function in the received containing SQL query with the SQL query string.

8. A data storage device, characterized in that:

the data storage device contains code which when executed by a processor performs a method set forth in claim 6.

9. A method of integrating a non-SQL query written in a non-SQL language into a relational database system of the type wherein an SQL query may contain an invocation of a table function, the table function, when executed, returning a set of rows to the containing SQL query and the method comprising the steps performed in the relational database system of:

receiving a containing SQL query with a table function invocation which includes a parameter that specifies the non-SQL query;

prior to execution of the containing SQL query, making an SQL query string that returns results of the non-SQL query; and

replacing the table function invocation in the containing query with the SQL query string.

10. A data storage device, characterized in that:

the data storage device contains code which when executed by a processor performs the method set forth in claim 9.

11. A method of integrating a query on a set of RDF triples into a relational database system, the query being expressed as an RDF pattern, the relational database system being of the type wherein a query may contain an invocation of a table function, the table function, when executed, returning a set of rows to the containing query, and the method comprising the steps performed in the relational database system of:

receiving a containing query that includes a table function invocation with a parameter that specifies the RDF pattern;

during execution of the containing query, obtaining results that are equivalent to those specified by the RDF pattern from RDF triples accessible to the database system and returning the results in the returned rows.

12. The method set forth in claim 11 wherein:

the table function further has a parameter that specifies a RDF rule base of RDF rules;

the RDF rule base is accessible to the relational database management system; and

in the step of obtaining results, the results are further obtained from RDF triples that are inferred using RDF rules from the specified RDF rule base.

13. The method set forth in claim 12 further comprising the step of:

inferring the inferred RDF triples and making them accessible to the relational database management system prior to executing the containing query.

14. The method set forth in claim 12 wherein:

the table function'"'"'s parameters further include a specification of an RDF model; and

in the step of obtaining results, further obtaining results from inferred triples that are inferred by applying RDF rules from the rule base to the RDF triples belonging to the model.

15. The method set forth in claim 11 wherein:

in the step of obtaining results, the results are obtained by querying an RDF triples table of representations of RDF triples in the relational database management system

16. The method set forth in claim 15 wherein:

the relational database system includes one or more optimization objects for optimizing queries on the RDF triples table; and

in the step of obtaining results, the relational database management system uses the optimization objects to optimize the query on the RDF triples table.

17. The method set forth in claim 16 wherein:

the query optimization objects include an index on the RDF triples table.

18. The method set forth in claim 17 wherein:

the RDF triples table includes a subject column whose values represent RDF subjects, a predicate column whose values represent RDF predicates, and an object column whose values represent RDF objects; and

the index is on the predicate, subject, and object columns.

19. The method set forth in claim 16 wherein:

the queries include self-joins on rows from the RDF triples table; and

the query optimization objects include a materialized view of a self-join on the RDF triples table.

20. The method set forth in claim 16 wherein:

the RDF triples table includes a subject column whose values represent RDF subjects, a predicate column whose values represent RDF predicates, and an object column whose values represent RDF objects; and

the query optimization objects include a subject-property matrix materialized join view which has a subject column and one or more predicate columns, the values in the predicate columns of a row of the join view being values of RDF objects which the RDF triples in the RDF triples table relate to the value in the row'"'"'s subject column.

21. The method set forth in claim 15 wherein:

the relational database management system includes an RDF rules table that contains representations of RDF rules belonging to one or more rule bases; and

in the step of obtaining results, the query on the RDF Triples table is additionally applied to inferred RDF triples that are inferred by applying rules represented in the RDF rules table to the representations in the RDF triples table.

22. The method set forth in claim 21 wherein:

the table function'"'"'s parameters further include a specification of at least one of the rule bases; and

in the step of obtaining results, the applied rules belong to the rule base specified in the rule base specification.

23. The method set forth in claim 21 further comprising the step of:

prior to the step of receiving a containing query, making an inferred triples table containing triples inferred by applying the rules to the representations in the RDF triples table; and

in the step of obtaining results, executing the query on the inferred triples from the inferred triples table.

24. The method set forth in claim 11 further comprising the steps performed prior to executing the containing query of:

generating a query string that specifies rows containing results that are equivalent to those specified by the RDF pattern; and

replacing the invocation of the table function with the generated query string.

25. The method set forth in claim 11 wherein:

the step of obtaining the rows is performed by executing code associated with the table function invocation that obtains the results from the RDF triples that are accessible to the relational database system.

26. A data storage device, characterized in that:

the data storage device contains code which, when executed by a processor, performs the method set forth in claim 11.

27. Apparatus for integrating a query expressed as an RDF pattern on a set of RDF triples into a relational database management system of the type wherein a query may contain an invocation of a table function, the table function, when executed, returning a set of rows to the containing query, the apparatus comprising:

an RDF table function in the relational database system whose parameters include a specification of the RDF pattern; and

code associated with the RDF table function that causes rows containing results that are equivalent to those specified by the RDF pattern to be obtained from RDF triples accessible to the database system and returned to the containing query.

28. The apparatus set forth in claim 27 wherein:

the table function'"'"'s parameters further include a specification of a RDF rule base of RDF rules;

the rule base is accessible to the table function; and

the RDF triples include triples that are inferred using RDF rules from the rule base.

29. The apparatus set forth in claim 28 wherein:

the inferred RDF triples are inferred prior to execution of the RDF table function.

30. The apparatus set forth in claim 28 wherein:

the table function'"'"'s parameters further include a specification of an RDF model; and

the RDF triples include RDF triples belonging to the model and inferred triples that are inferred by applying RDF rule from the rule base to the RDF triples belonging to the model.

31. The apparatus set forth in claim 27 further comprising:

a RDF triples table in the relational database management system that contains representations of the RDF triples, the code causes the results to be obtained from the RDF triples table.

32. The apparatus set forth in claim 31 further comprising;

one or more optimization objects for optimizing the query on the representations of the RDF triples in the RDF triples table.

33. The apparatus set forth in claim 32 wherein:

the query optimization objects include an index on the RDF triples table.

34. The apparatus set forth in claim 33 wherein:

the RDF triples table includes a subject column whose values represent RDF subjects, a predicate column whose values represent RDF predicates, and an object column whose values represent RDF objects; and

the index is on the predicate, subject, and object columns.

35. The apparatus set forth in claim 32 wherein:

the queries include self-joins on rows from the RDF triples table; and

the query optimization objects include a materialized view of a self-join on the RDF triples table.

36. The apparatus set forth in claim 32 wherein:

the RDF triples table includes a subject column whose values represent RDF subjects, a predicate column whose values represent RDF predicates, and an object column whose values represent RDF objects; and

the query optimization objects include a subject-property matrix materialized join view which has a subject column and one or more predicate columns, the values in the predicate columns of a row of the join view being values of RDF objects which the RDF triples in the RDF triples table relate to the value in the row'"'"'s subject column.

37. The apparatus set forth in claim 31 further comprising:

an RDF rules table in the relational database management system that contains representations of RDF rules belonging to one or more rule bases, the code additionally causing the results to be obtained from inferred RDF triples that are inferred by applying rules represented in the RDF rules table to the representations in the RDF triples table.

38. The apparatus set forth in claim 37 wherein:

the table function'"'"'s parameters further include a specification of at least one of the rule bases; and

the applied rules belong to the rule base specified in the rule base specification.

39. The apparatus set forth in claim 37 further comprising:

an inferred triples table containing triples inferred by applying the rules to the representations in the RDF triples table,

the code further causing results to be obtained from the inferred triples table.

40. The apparatus set forth in claim 27 wherein:

the code is first code associated with the RDF table function that, when executed, creates a query string that specifies results equivalent to the results specified by the RDF pattern,

the relational database management system executing the first code prior to execution of the containing query and replacing the RDF table function invocation in the containing query with the generated query string.

41. The apparatus set forth in claim 27 wherein:

the code is second code associated with the RDF table function that, when executed, returns the results specified by the RDF pattern, the relational database management system executing the second code when the containing query is executed.

42. A storage device that is accessible to a processor, the storage device being characterized in that:

the storage device contains code which, when executed by the processor, implements the apparatus set forth in claim 27.

43. A materialized view that is an optimization for queries on a relational database table wherein an entity is represented by a plurality of rows, each of the rows relating a type of attribute and a value of the attribute to the entity, the materialized view comprising:

a table wherein the entity is represented by a row, the row having a field for the entity and a field for each type of attribute, with the field for a given type of attribute containing the value of that attribute for the entity.

44. A storage device that is accessible to a processor, the storage device being characterized in that:

the storage device contains code which, when executed by the processor, implements the materialized view set forth in claim 43.