Towards Query Processing over Heterogeneous Federations of RDF Data Sources.

A federation of RDF data sources offers enormous potential when answers or insights of queries are unavailable via a single data source. As various interfaces for accessing RDF data are proposed, one challenge for querying such a federation is that the federation members are heterogeneous in terms of the type of data access interfaces. There does not exist any research on systematic approaches to tackle this challenge. To provide a formal foundation for future approaches that aim to address this challenge, we have introduced a language, called FedQPL, that can be used for representing query execution plans in this setting. With a poster in the conference we generally want to outline the vision for the next generation of query engines for such federations and, in this context, we want to raise awareness in the Semantic Web community for our language. In this extended abstract, we first discuss challenges in query processing over such heterogeneous federations; thereafter, we briefly introduce our proposed language, which we have extended with a few new features that we did not have in the version published originally.