DEFENDER: A decomposer for queries against federations of endpoints

Gabriela Montoya*, Maria Esther Vidal, Maribel Acosta

*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

2 Citations (Scopus)


We present DEFENDER and illustrate the benefits of identifying promising query decompositions and efficient plans that combine results from federations of SPARQL endpoints. DEFENDER is a query decomposer that implements a two-fold approach. First, triple patterns in a SPARQL query are decomposed into simple sub-queries that can be completely executed on one endpoint. Second, sub-queries are combined into a feasible bushy tree plan where the number of joins is maximized and the height of tree is minimized. We demonstrate DEFENDER and compare its performance with respect to state-of-the-art RDF engines for queries of diverse complexity, networks with different delays, and dataset differently distributed among a variety of endpoints.

Original languageEnglish
Title of host publicationThe Semantic Web : ESWC 2012 Satellite Events - Revised Selected Papers
EditorsAlexandre Passant, Barry Norton, Emanuele Della Valle, Raphael Troncy, Irini Fundulaki, Elena Simperl, Dunja Mladenic
Number of pages5
Publication date1 Jan 2015
ISBN (Print)9783662466407
Publication statusPublished - 1 Jan 2015
EventExtended Semantic Web Conference, ESWC 2012 - Heraklion, Greece
Duration: 27 May 201231 May 2012


ConferenceExtended Semantic Web Conference, ESWC 2012
SponsorEtal, Fluid Operations, MIMOS,,, Yahoo, Fluid Operations
SeriesLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)


Dive into the research topics of 'DEFENDER: A decomposer for queries against federations of endpoints'. Together they form a unique fingerprint.

Cite this