Matches in ScholarlyData for { ?s ?p SAQ (Semantic Archive and Query) is a system for long-term preservation and querying of relational data in terms of RDF. In SAQ relational data in a back-end DBMS is exposed as an RDF view, called the RD-view. SAQ can process arbitrary SPARQL queries to the RD-view. Long-term preservation as RDF of selected parts of a relational database is specified by SPARQL queries to the RD-view. Such queries usually select sets of RDF properties and thus in the query definition a property p is unknown. We call such queries unbound-property queries. This class of queries is also present in the SPARQL benchmarks. We optimize unbound-property queries by introducing a query transformation algorithm called Group Common Terms, GCT. It pulls out from a DNF normalized query those common terms that can be translated to SQL predicates accessing the relational database. Our experiments using the Berlin SPARQL benchmark show that GCT improves substantially the query execution time to a back-end commercial relational DBMS for both selective and unselective unbound-property queries. We compared the performance of our approach with the performance of other systems processing SPARQL queries over views of relational databases and showed that GCT improves scalability compared to the approaches used by the other systems. . }
Showing items 1 to 1 of
1
with 100 items per page.
- 4 abstract "SAQ (Semantic Archive and Query) is a system for long-term preservation and querying of relational data in terms of RDF. In SAQ relational data in a back-end DBMS is exposed as an RDF view, called the RD-view. SAQ can process arbitrary SPARQL queries to the RD-view. Long-term preservation as RDF of selected parts of a relational database is specified by SPARQL queries to the RD-view. Such queries usually select sets of RDF properties and thus in the query definition a property p is unknown. We call such queries unbound-property queries. This class of queries is also present in the SPARQL benchmarks. We optimize unbound-property queries by introducing a query transformation algorithm called Group Common Terms, GCT. It pulls out from a DNF normalized query those common terms that can be translated to SQL predicates accessing the relational database. Our experiments using the Berlin SPARQL benchmark show that GCT improves substantially the query execution time to a back-end commercial relational DBMS for both selective and unselective unbound-property queries. We compared the performance of our approach with the performance of other systems processing SPARQL queries over views of relational databases and showed that GCT improves scalability compared to the approaches used by the other systems. ".