Ítem
Solo Metadatos

Managing Response Time Tails by Sharding

Mostrar el registro sencillo de la publicación
dc.creatorHarrison, P. G.spa
dc.creatorPatel, N. M.spa
dc.creatorPérez, Juan F.spa
dc.creatorQiu, Z.spa
dc.date.accessioned2020-05-26T00:00:31Z
dc.date.available2020-05-26T00:00:31Z
dc.date.created2019spa
dc.description.abstractMatrix analytic methods are developed to compute the probability distribution of response times (i.e., data access times) in distributed storage systems protected by erasure coding, which is implemented by sharding a data object into N fragments, only K less than N of which are required to reconstruct the object. This leads to a partial-fork-join model with a choice of canceling policies for the redundant N - K tasks. The accuracy of the analytical model is supported by tests against simulation in a broad range of setups. At increasing workload intensities, numerical results show the extent to which increasing the redundancy level reduces the mean response time of storage reads and significantly flattens the tail of their distribution; this is demonstrated at medium-high quantiles, up to the 99th. The quantitative reduction in response time achieved by two policies for canceling redundant tasks is also shown: For cancel-at-finish and cancel-at-start, which limits the additional load introduced whilst losing the benefit of selectivity amongst fragment service times. © 2019 Copyright held by the owner/author(s).eng
dc.format.mimetypeapplication/pdf
dc.identifier.doihttps://doi.org/10.1145/3300143
dc.identifier.urihttps://repository.urosario.edu.co/handle/10336/23233
dc.language.isoengspa
dc.publisherAssociation for Computing Machineryspa
dc.relation.citationIssueNo. 1
dc.relation.citationTitleACM Transactions on Modeling and Performance Evaluation of Computing Systems
dc.relation.citationVolumeVol. 4
dc.relation.ispartofACM Transactions on Modeling and Performance Evaluation of Computing Systems, Vol.4, No.1 (2019)spa
dc.relation.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85074676254&doi=10.1145%2f3300143&partnerID=40&md5=241dbfc47e2d8a85dd826cf3508f0f13spa
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.keywordMultiprocessing systemsspa
dc.subject.keywordProbability distributionsspa
dc.subject.keywordQuality of servicespa
dc.subject.keywordResponse time (computer systems)spa
dc.subject.keywordDistributed storage systemspa
dc.subject.keywordMatrix analytic methodsspa
dc.subject.keywordMean response timespa
dc.subject.keywordNumerical resultsspa
dc.subject.keywordParallel taskspa
dc.subject.keywordPerformancespa
dc.subject.keywordShardingspa
dc.subject.keywordWorkload intensitiesspa
dc.subject.keywordDigital storagespa
dc.subject.keywordParallel task processingspa
dc.subject.keywordPerformancespa
dc.subject.keywordQuality of servicespa
dc.subject.keywordResponse timespa
dc.subject.keywordShardingspa
dc.titleManaging Response Time Tails by Shardingspa
dc.typearticleeng
dc.type.hasVersioninfo:eu-repo/semantics/publishedVersion
dc.type.spaArtículospa
Archivos
Colecciones
Artículos