Authors:
Regis Pires Magalhães
1
;
José Maria Monteiro
1
;
Vânia M. P. Vidal
1
;
José A. F. de Macêdo
1
;
Macedo Maia
1
;
Fábio Porto
2
and
Marco A. Casanova
3
Affiliations:
1
Universidade Federal do Ceará (UFC), Brazil
;
2
Laboratório Nacional de Computação Científica (LNCC), Brazil
;
3
Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), Brazil
Keyword(s):
Linked Data, Federated Queries, Query Processing, Data Integration, Mashup.
Related
Ontology
Subjects/Areas/Topics:
Cloud Computing
;
Coupling and Integrating Heterogeneous Data Sources
;
Databases and Information Systems Integration
;
Distributed Database Systems
;
Enterprise Information Systems
;
Non-Relational Databases
;
Query Languages and Query Processing
;
Semantic Web Technologies
;
Services Science
;
Software Agents and Internet Computing
Abstract:
Linked data applications express integrated views using the SPARQL query language. A SPARQL federated query is submitted to a query engine that processes it over the distributed SPARQL endpoints. However, achieving an efficient execution of such a SPARQL federated query is hard. This is mainly due to the fact that query processors have little or no statistical information about the data stored at the endpoints. Moreover, the endpoints, usually, are autonomous and unstable. This paper presents QEF-LD, a query engine that enables the efficient execution of federated queries over multiple Linked Data sources. Experiments demonstrate the feasibility of QEF-LD when compared to available federated query engines.