预览加载中，请您耐心等待几秒...

基于端到端测量的链路状态概率快速推断方法.docx

预览
1/3
2/3
3/3

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于端到端测量的链路状态概率快速推断方法 Abstract: Intoday'sworldoffast-pacedtechnologyandincreasingrelianceondatacommunication,efficientandaccurateinferenceofnetworktopologyandlink-stateprobabilitiesisofparamountimportance.Traditionalmethodsthatrelyonperiodicnetworkmeasurementshavelimitationsduetotheirlowgranularityandhighoverheads.End-to-endmeasurement-basedmethodshavebeenproposedasanalternative.Inthispaper,wepresentafastandaccuratemethodforinferringlink-stateprobabilitiesbasedonend-to-endmeasurement. Introduction: Inferringnetworktopologyandlink-stateprobabilitiesisavitalcomponentofnetworkingandhasavastrangeofpotentialapplications.Theseapplicationsincludetrafficengineering,routing,anomalydetection,andnetworkoptimization.Traditionalmethodsforinferringnetworktopologyandlink-stateprobabilitiesrelyonperiodicnetworkmeasurements.However,suchmethodshavelimitations.Theselimitationsincludelowgranularity,highoverheads,scalability,andpoorefficiency.Furthermore,somenetworkfailuresmaynotbevisibleusingtraditionalnetworkmeasurements,leadingtoincorrectinferences.Thishasledtoresearchersproposingend-to-endmeasurement-basedmethodstoovercometheselimitations. End-to-endmeasurement-basedmethodshavegainedsignificantattentioninrecentyearsduetotheirabilitytoprovidedetailedandmoreaccurateinformationonthehealthandstateofnetworklinks.Thesemethodsrelyonactiveprobing,whichinvolvesinjectingtestpacketsintothenetworkandmeasuringtheirassociatedperformancemetrics.Thetestpacketsareinjectedandmonitoredfromdifferentlocationsinthenetwork,allowingforamorecomprehensiveunderstandingofnetworktopologyandlink-stateprobabilities. Inthispaper,wepresentafastandaccuratemethodforinferringlink-stateprobabilitiesbasedonend-to-endmeasurements.Theproposedmethodisadeparturefromthetraditionalapproaches,whichrelyonperiodicnetworkmeasurements. Methodology: Theproposedmethodologyleveragesaprobabilisticmodelofthenetworklinksandutilizestheobservedend-to-enddatameasurementstoinferthelink-stateprobabilities.Themethodologyreliesonthefollowingste