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)

Abstract

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
PublisherSpringer
Publication date1 Jan 2015
Pages480-484
ISBN (Print)9783662466407
DOIs
Publication statusPublished - 1 Jan 2015
EventExtended Semantic Web Conference, ESWC 2012 - Heraklion, Greece
Duration: 27 May 201231 May 2012

Conference

ConferenceExtended Semantic Web Conference, ESWC 2012
Country/TerritoryGreece
CityHeraklion
Period27/05/201231/05/2012
SponsorEtal, Fluid Operations, MIMOS, videolectures.net, Xlike.org, Yahoo, Fluid Operations
SeriesLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7540
ISSN0302-9743

Cite this