Query processing for heterogeneous relational data integration

The emergence of the Semantic Web has given rise to researches on semantic integration of heterogeneous relational databases using ontology. Semantic-based heterogeneous relational data integration is recently attracting many diverse attentions, since most of the information is still stored in relational databases. This paper focuses on the query processing problem in semantic-based heterogeneous relational databases integration systems. The semantic of query rewriting is defined and algorithms for query decomposing, rewriting and transforming are presented. SPARQL queries over global schema are rewritten into local SQL queries that can be executed on heterogeneous relational databases. The experiment results show that the approach we proposed has nice performance and scalability.