Online querying of concept hierarchies in P2P systems

Doka, K; Asiki, A; Tsoumakos, D; Koziris, N

dc.contributor.author	Doka, K	en
dc.contributor.author	Asiki, A	en
dc.contributor.author	Tsoumakos, D	en
dc.contributor.author	Koziris, N	en
dc.date.accessioned	2014-03-01T02:45:43Z
dc.date.available	2014-03-01T02:45:43Z
dc.date.issued	2008	en
dc.identifier.issn	03029743	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32337
dc.subject	Adaptive Algorithm	en
dc.subject	Concept Hierarchy	en
dc.subject	Distributed Storage	en
dc.subject	Experimental Evaluation	en
dc.subject	Indexation	en
dc.subject	IT Management	en
dc.subject	P2p System	en
dc.subject	Prior Knowledge	en
dc.subject.other	Applications.	en
dc.subject.other	Concept hierarchies	en
dc.subject.other	Distributed storages	en
dc.subject.other	Experimental evaluations	en
dc.subject.other	P2P systems	en
dc.subject.other	Prior knowledges	en
dc.subject.other	Query flooding	en
dc.subject.other	Query workloads	en
dc.subject.other	Querying of datums	en
dc.subject.other	Remarkable performances	en
dc.subject.other	Adaptive algorithms	en
dc.subject.other	Direction of arrival	en
dc.subject.other	Internet	en
dc.subject.other	Radio direction finding systems	en
dc.subject.other	Online systems	en
dc.title	Online querying of concept hierarchies in P2P systems	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1007/978-3-540-88871-0_16	en
heal.identifier.secondary	http://dx.doi.org/10.1007/978-3-540-88871-0_16	en
heal.publicationDate	2008	en
heal.abstract	In this paper we describe HIS, a system that enables efficient storage and querying of data organized into concept hierarchies and dispersed over a network. Our scheme utilizes an adaptive algorithm that automatically adjusts the level of indexing according to the granularity of the incoming queries, without assuming any prior knowledge of the query workload. Efficient roll-up and drill-down operations increase the exact-match query ratio by shifting to the most favorable hierarchy level. Combined with soft-state indices created after query misses, our system achieves maximization of performance by minimizing query flooding. Extensive experimental evaluations show that, on top of the advantages that a distributed storage offers, our method answers the large majority of incoming queries without flooding the network and at the same time it manages to preserve the hierarchical nature of data. It shows remarkable performance especially for skewed workloads, which are frequently documented in the majority of Internet-scale applications. These characteristics are maintained even after sudden shifts in the workload. © 2008 Springer Berlin Heidelberg.	en
heal.journalName	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)	en
dc.identifier.doi	10.1007/978-3-540-88871-0_16	en
dc.identifier.volume	5331 LNCS	en
dc.identifier.issue	PART 1	en
dc.identifier.spage	212	en
dc.identifier.epage	230	en