Ítem
Solo Metadatos

Cutting Latency Tail: Analyzing and Validating Replication without Canceling

Mostrar el registro sencillo de la publicación
dc.creatorQiu Z.spa
dc.creatorPérez, Juan F.spa
dc.creatorBirke R.spa
dc.creatorChen L.spa
dc.creatorHarrison P.G.spa
dc.date.accessioned2020-05-26T00:10:42Z
dc.date.available2020-05-26T00:10:42Z
dc.date.created2017spa
dc.description.abstractResponse time variability in software applications can severely degrade the quality of the user experience. To reduce this variability, request replication emerges as an effective solution by spawning multiple copies of each request and using the result of the first one to complete. Most previous studies have mainly focused on the mean latency for systems implementing replica cancellation, i.e., all replicas of a request are canceled once the first one finishes. Instead, we develop models to obtain the response-time distribution for systems where replica cancellation may be too expensive or infeasible to implement, as in 'fast' systems, such as web services, or in legacy systems. Furthermore, we introduce a novel service model to explicitly consider correlation in the processing times of the request replicas, and design an efficient algorithm to parameterize the model from real data. Extensive evaluations on a MATLAB benchmark and a three-tier web application (MediaWiki) show remarkable accuracy, e.g., 7 (4 percent) average error on the 99th percentile response time for the benchmark (respectively, MediaWiki), the requests of which execute in the order of seconds (respectively, milliseconds). Insights into optimal replication levels are thereby gained from this precise quantitative analysis, under a wide variety of system scenarios. © 2017 IEEE.eng
dc.format.mimetypeapplication/pdf
dc.identifier.doihttps://doi.org/10.1109/TPDS.2017.2706268
dc.identifier.issn10459219
dc.identifier.urihttps://repository.urosario.edu.co/handle/10336/24250
dc.language.isoengspa
dc.publisherIEEE Computer Societyspa
dc.relation.citationEndPage3141
dc.relation.citationIssueNo. 11
dc.relation.citationStartPage3128
dc.relation.citationTitleIEEE Transactions on Parallel and Distributed Systems
dc.relation.citationVolumeVol. 28
dc.relation.ispartofIEEE Transactions on Parallel and Distributed Systems, ISSN:10459219, Vol.28, No.11 (2017); pp. 3128-3141spa
dc.relation.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85032457020&doi=10.1109%2fTPDS.2017.2706268&partnerID=40&md5=bc10af0ad3bf606510446e87dbad6990spa
dc.rights.accesRightsinfo:eu-repo/semantics/openAccess
dc.rights.accesoAbierto (Texto Completo)spa
dc.source.instnameinstname:Universidad del Rosariospa
dc.source.reponamereponame:Repositorio Institucional EdocURspa
dc.subject.keywordApplication programsspa
dc.subject.keywordBenchmarkingspa
dc.subject.keywordComputer softwarespa
dc.subject.keywordComputer software selection and evaluationspa
dc.subject.keywordLegacy systemsspa
dc.subject.keywordMATLABspa
dc.subject.keywordWeb servicesspa
dc.subject.keywordEffective solutionspa
dc.subject.keywordMatrix analytic methodsspa
dc.subject.keywordResponse time distributionspa
dc.subject.keywordResponse time variabilityspa
dc.subject.keywordService timespa
dc.subject.keywordSoftware applicationsspa
dc.subject.keywordSoftware quality engineeringspa
dc.subject.keywordSpeculative computingspa
dc.subject.keywordResponse time (computer systems)spa
dc.subject.keywordCorrelated service timesspa
dc.subject.keywordMatrix analytic methodsspa
dc.subject.keywordSoftware quality engineeringspa
dc.subject.keywordSpeculative computingspa
dc.titleCutting Latency Tail: Analyzing and Validating Replication without Cancelingspa
dc.typearticleeng
dc.type.hasVersioninfo:eu-repo/semantics/publishedVersion
dc.type.spaArtículospa
Archivos
Colecciones
Artículos