Skip to main navigation Skip to search Skip to main content

DEFENDER: A decomposer for queries against federations of endpoints

  • Simón Bolívar Univ.
  • University of Karlsruhe

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

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
Subtitle of host publicationESWC 2012 Satellite Events - Revised Selected Papers
EditorsElena Simperl, Barry Norton, Dunja Mladenic, Emanuele Della Valle, Irini Fundulaki, Alexandre Passant, Raphael Troncy
PublisherSpringer Verlag
Pages480-484
Number of pages5
ISBN (Print)9783662466407
DOIs
StatePublished - 2015
Externally publishedYes
EventExtended Semantic Web Conference on Satellite Events, ESWC 2012 - Heraklion, Greece
Duration: 27 May 201231 May 2012

Publication series

NameLecture Notes in Computer Science
Volume7540
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceExtended Semantic Web Conference on Satellite Events, ESWC 2012
Country/TerritoryGreece
CityHeraklion
Period27/05/1231/05/12

Fingerprint

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

Cite this