Application-LayerAnycasting

Size: px
Start display at page:

Download "Application-LayerAnycasting"

Transcription

1 Application-LayerAnycasting SamratBhattacharjee MostafaH.Ammar GIT{CC{96/25 EllenW.Zegura ZongmingFei VirenShah nicatewiththe\best"server,accordingtosomeperformanceorpolicycriteria,inagroup ofcontent-equivalentservers.weexaminethedenitionandsupportoftheanycasting signedtosupportserverreplicationbyallowingapplicationstoeasilyselectandcommu- ageographicallywide-spreadnetwork.theanycastingcommunicationparadigmisde- Serverreplicationisakeyapproachformaintaininguser-perceivedqualityofservicewithin Abstract Weexaminetheuseofvarioustechniquesformaintainingthisinformationatanycast basedlters(e.g.,serverresponsetime)andpolicy-basedlters;wefurtherallowltering processismetricinformationdescribingtherelativeperformanceofreplicatedservers. bothattheanycastresolverandlocaltotheanycastclient.akeyinputtotheltering fromnetwork-layersupport,ourdenitionincludesthenotionoflters,functionsthatare appliedtogroupsofaddressestoaecttheselectionprocess.weconsiderbothmetric- intooneormoreipaddressesusinganycastresolvers.inadditiontobeingindependent paradigmattheapplicationlayer,providingaservicethatmapsanycastdomainnames resolvers. GeorgiaInstituteofTechnology Atlanta,Georgia30332{0280 CollegeofComputing

2 theprovisionofsuchservicesistheabilityofaservicetomeetthedemandsofalargenumber 1Introduction ofusersthataregeographicallywide-spread.itisalsoimportantthattheuser-perceived Asthisviewbecomesmoreprevalent,itbecomesimportanttoprovide,withintheInternet, explicitsupportfortheecientdeliveryofnetworkedservices.animportantconsiderationin TheInternetisincreasinglybeingviewedasprovidingservices,andnotjustconnectivity. qualityofservice(e.g.,responsetime,throughput,reliability)bemaintainedatanacceptable scalabilityoftheservice. [4]andmulticastingofserverresponsesoverthenetwork[5,6].Inthispaperweconsider andcompetitive(inthecaseofcommercialservices)level.thisisoftenreferredtoasthe theserverreplicationtechnique.inparticular,weinvestigatetheanycastingcommunication paradigmwhichhasbeenproposed[7]tosupportserverreplication. service.theseincludeserverreplication[1],caching[2,3],batchingofrequestsattheserver Asoriginallydened[7],anycastingprovides: Therehavebeenseveralapproachesproposedforimprovingthescalabilityofanetworked Inthisdenition,anIPanycastaddressisusedtodeneagroupofserversthatprovidethe sameservice.asenderdesiringtocommunicatewithonlyoneoftheserverssendsdatagrams withtheipanycastaddressinthedestinationaddresseld.thedatagramisthenrouted tosomecriteria)identiedbytheanycastaddress. usinganycast-awarerouterstoatleastoneoftheservers(presumablythe\best"according \astatelessbesteortdeliveryofananycastdatagramtoatleastonehost,and Inourworkweadoptamoregeneralviewofanycastingasacommunicationparadigm preferablyonlyonehost,whichservestheanycastaddress." providingtheanycastingservicedenitionandexaminingtheprovisionofthisservicewithin theanycastingservice1.theoriginalanycastingproposal[7]can,therefore,beviewedas wedierentiatebetweentheanycastingservicedenitionandtheprotocollayerproviding hasthefollowinglimitations: thatisanalogoustothebroadcastandmulticastcommunicationparadigms.inparticular, theiplayer. plicationlayer.ourmotivationderivesfromthefactthatnetwork-layer-supportedanycasting SomepartoftheIPaddressspacemustbeallocatedtoanycastaddresses.ForIPv4, Inthispaperweexaminethedenitionandsupportoftheanycastingparadigmattheap- classofaddresses.ipv6proposals[9]doincludeaspecicaddressspaceallocatedto natingsomeexistingaddressesasanycast(e.g.,withinclassc)orcreatingaseparate severalwaystoallocateanycastaddresseshavebeensuggested[7].theseincludedesig- 1Forexamplemulticastingasacommunicationparadigmrepresentsadesiretosendthesamemessagetoagroup Theselectionoftheservertowhichananycastaddresseddatagramissentismade Theuseofanycastaddressesrequiresroutersupport.Routersmustrecognizeanycast addressesandforwardthemproperly.routersmustcoordinatewithoneanotherto entirelywithinthenetworkwithnooptionforuserselectionorinput. ensurethatdeliveryisusuallymadetoexactlyonehost. anycasting. ofreceivers.themulticastparadigmcanbesupportedusingmulticastroutingatthenetworklayeroritcanbe supportedaboveaunicast-onlynetworkortransportlayerbyusingmultipleunicasts.indeedthislatterapproach washowmanymulticastproblemswereaddressedinearlysystemdesigns(e.g.,theoriginalisisdistributedsystem [8]).Thereisclearlyadistinctionbetweenmulticastasacommunicationparadigmandhowthisparadigmis supported. 1

3 FIGURE1AnycastNameResolutionQuery/ResponseCycle Application Client Anycast Domain Name Specification Client Filter IP Address Anycast Server Filter Anycast Group Response Metric Info Query Anycast Anycast Resolver Filter ConsistentwiththestatelessnatureofIP,thedestinationisdeterminedonaperdatagrambasis.Thus,twosuccessivedatagramssenttoananycastaddressmaybe deliveredtotwodierenthoststhatservetheaddress.aprotocolthatrequiresalldatagramstobedeliveredtoasinglehostcanuseanycastingfordiscoveryofasatisfactory host,butsubsequenttransmissionsshouldusestandardiptotheselectedhost. Becauseofthenetworklayer'sabilitytoecientlydetermineshortestpaths,itiswellsuitedforananycastingservicebasedontheselectionoftheclosest(innumberofhops orotherdistanceordelaymetrics)server.anapplicationlayerapproachisbettersuited athandlingavarietyofothermetricssuchasserverthroughputorturnaroundtime. Whereasnetwork-layersupporthingesaroundtheuseofanycastIPaddresses,ourapplicationlayersupportmakesuseofanycastdomainnames(ADNs).ThefunctionofanapplicationlayeranycastingserviceisthustomapanADNintooneormore(multicastorunicast)IP addresses.animportantfeatureofapplication-layeranycastingisthatitdoesnotrequire modicationstonetworklayeroperations. Thispaperfocusesonthedesignofaninfrastructuretoprovideanapplication-layeranycastingservice.OurdesigncentersaroundtheuseofanycastresolverstoperformtheADN toipaddressmapping.clientsinteractwiththeanycastresolversaccordingtoabasic query/responsecycleillustratedinfigure1:aclientgeneratesananycastquery,theresolver processesthequeryandreplieswithananycastresponse.akeyfeatureofthesystemisthe presenceofmetricdatabases,associatedwitheachanycastresolver,containingperformance dataaboutservers.theperformancedatacanbeusedintheselectionofaserverfroma group,basedonuser-speciedperformancecriteria.weconsidermetricdeterminationtechniquesthatcanbeusedtomaintaintheanycastresolverdatabases.inaddition,weinvestigate howanycastclientapplicationsmayinterfacewiththeanycastresolver. Therestofthepaperisstructuredasfollows.Section2discussesrelatedwork.Sec- 2

4 denitionsofreplicatedservice,anycastgroupsanddomainnames,andanycastresolverfunctionality.insection4weelaborateontheanycastquery/responsecycle,discussingthedesign tion3describesinmoredetailthecomponentsofapplication-layeranycasting,providingour 2RelatedWork thenetworkortheresolvers.weconcludethepaperinsection6withanassessmentofthe meritsofapplication-layeranycastingandadiscussionoffuturework. andoperationoftheinterfacetoananycastresolver.section5considersasetofmetricsand metricdeterminationmechanismstoallowaccuratemeasurementswhilenotundulyloading adecade.initially,withlowtomoderateserverloads,theproblemwashowtondthe Theserverorresourcendingproblemhasbeenthesubjectofmuchinvestigationforover desiredresourceoverthenetworkknowingonlyitsnameorproperty.manytechniqueswere proposedandinvestigated.theseinclude:1)theuseofmulticastorbroadcastcommunication ServiceLocationProtocolwhichallowsausertospecifyasetofserviceattributeswhichcan recently,theservicelocationworkinggroupoftheietfisconsideringthedesignofthe atsiteswheretheresourceisfrequentlyaccessed[15].thisearlyworkdealtwithsituations wherethereistypicallyasingleinstanceoftheresource.thecaseofamobileresourcewas addressedthroughinterestingtechniquessuchastheuseofforwardingaddresses[16].more theresource(e.g.,[12,13,14])and3)theuseofcachingofaresource'slocation(notcontent) to\touch"allthelocationswheretheresourcemayresideinanattempttondit(e.g., beboundtoaserver'snetworkaddressinadynamicfashion[17]. [10,11]),2)theuseofvariousnameserverarchitecturesinordertolookupthelocationof MendezandMilliken[7]proposingtheideaofanycastinganddiscussingitsnetwork-layer supportand2)arecentstudybyguytonandschwartz[18]whichaddressestheproblem equivalentservers.twonotablestudiesinthisareaare:1)theoriginalworkbypartridge, ndingproblem:itisnowimportanttondthe\best"serverfromamongmanycontent- withtheworld-wideweb,theinternethasexperiencedadramaticgrowthintheuseand resultinginthedesiretoreplicate(ormirror)servers.thisaddsanewdimensiontotheserver provisionofinformationservices.thishasresultedinheavydemandsbeingplacedonservers Beginningwithinitialserviceslikeftp,archie,andgopherandculminatingmorerecently oflocatingthenearestserver.thelatterworkalsopresentsaclassicationof\best"-server locationschemes.theworkisrelatedtoearlierworkontheharvestsystem[19]which usedintheharvestsysteminordertoimprovethescalabilityoftheservice.anotherrelated theguytonandschwartzclassicationwhichconsiderstheoriginaldenitionofanycastingas andsearchingthroughthisinformation.toolsforcachingandreplicationofindicesarealso providesasetoftoolsforgatheringinformationfromvariousserversandecientlyindexing projectisthesonarnetworkproximityservice[20]inwhichtheauthorsdeneaservice whichcanreturntheclosest(inhops)serverfromamongaprovidedlist. anetwork-layer-supportedservice.ourworkinvestigatesthecompletedesignofapplicationlayeranycastingandnotjustthemetricprobingaspectasdiscussedin[18].wealsoconsider Bychoosingtodeneanycastingasacommunicationparadigm,wedeviatesomewhatfrom Theconceptofprobingthenetworkandtheserverstodetermineperformancemeasuresis aclosermatchtotheapplicationrequirement. usingavarietyofmetrics(notjusthop-distanceasprovidedbysonar)inordertoprovide 3

5 relatedtovarioustools,andsystemsthatareusedfornetworkmanagementpurposes.examplesoftheseincludethetraceroute[21],andmwatch[22]toolsandmanagementsystemsthat servicedubbedontimedelivery2whichprobesawebsiteatregularintervalsandprovides integratesuchtoolssuchasthefremontsystem[23].remotemeasurementandmonitoring areportontheperformanceofthewebsiteasperceivedbytheirclient. ofsystemperformancehasalsobeenexploredaspartoftheextensiveworkondistributed systemmonitoring[24,25].recently,thecompanytimedancersystemshasbeenoeringa observations.thisisrelatedtothepushcachingidea[26]whereserversareinchargeof theideaofhavingtheserver\push"ontothenetworkitsownlocallygeneratedperformance pushingthedesiredinformationontoremotecachesandtheserverpushmechanism[27] implementedinnetscapebrowsers. server.modicationstothednsresolutionmechanismsaremadetoallocatetheloadamong suchaschememultipleserversareclusteredandappeartotheoutsideworldasinglelogical Toavoidthescalabilityproblemsinherentinprobingforperformancewealsoexplore theservers. 3GeneralizedApplication-LayerAnycasting Finally,anycastingisrelatedtothetechniqueusedtobuildscalableHTTPservers[28].In ofthisparadigm. 3.1WhatIsaReplicatedService communicationparadigm.latersectionsareconcernedwithissuesrelatingtotherealization Inthissectionwedescribeinmoredetailthedenitionofapplication-layeranycastingasa Ananycastdomainname(ADN)isusedtoidentifyaparticularnetworkservicethatcanbe canbeconstruedasbeingreplicatedservers.dependingontheapplication,thewebservers thoughinourdenitionreplicatedserversdonotnecessarilyhavetobeidentical. iftheycontainequivalentcontentand/orfunctionalityfromanapplicationperspective.itis truethatserverswithidenticalcontentorfunctionalityareconsideredreplicasofeachother, providedbymultiplelocationsonthenetwork.wesaythattwoserversreplicateeachother forcnninteractive,timemagazineandusatodaycouldconstituteasetofreplicated primaryexampleisquorumconsensusapplicationsthatariseindistributedsystems.quorum serversaswell. websearchsites(e.g.,lycos,infoseek,yahoo)carrydierentinformation,they,nevertheless, cated,whereallserversprovideidenticalfunctionality.ontheotherhand,whereasthemany Insomecasesmultipleserversarerequiredtoprovideasingleinstanceofaservice.A TheNetworkTimeProtocol(NTP)isanexampleofasimpleservicethatiswidelyrepliplications,suchasthereadingofstrongly-consistentreplicateddata,mayrequirepermission coordinatornodes.ifmutuallyexclusiveexecutionofoperationsisdesired,e.g.,aswould operationneedstocollectpermissionfromamajorityofthecoordinatornodes.otherap- berequiredwhenupdatingstrongly-consistentreplicateddata,thenthenodeexecutingthe Anoperationmayproceedtocompletiononlyifitisgrantedpermissionfromanumberof consensusisageneralclassofsynchronizationprotocolsfordistributedsystems[29,30,31,32]. 2Seehttp://

6 fromacertainnumberofnodes,notnecessarilyamajority.byincludingallthecoordinator nodesinamulticastgroupwithasinglegroupaddress,multicastcommunicationcanbeused torunaparticularcomputation.thesetofavailablehostscanthenbemadetoforman content,e.g.,computeservers.inthiscasetheuserisinterestedinndingaserveronwhich reasons:faulttoleranceandloadbalancing. byaquorumcollectortocommunicatewiththesetofcoordinatornodes. theappropriatequorumtobeabletodelivertheservice.thisreplicationisdonefortheusual nodeslargerthantherequiredquorum.thisallowsanysubsetofthecoordinatornodeswith anycastgroup.theperformancemeasureofinteresttoselectthebestserverwouldnormally Wealsoextendournotionofservicetoincludemoregeneralfunctionalitythaninformation Thistypeofserviceistypically\replicated"bymakingthethenumberofcoordinator becpuload.aninterestingvariationiswheretheuserhasaparallelprogramtorunthat requirestheuseofmultiplehosts.thisisanotherinstanceofmultiplehostsprovidingasingle instanceofaservice;makingtheanycastgroupacollectionofpre-formedmulticastgroups eachcontainingenoughhoststoexecutetheparallelprogram. thegrouptothe\best"addressinthecollectionaccordingtosomecriteria. groups4.inbothcases,theanycastingserviceprovidesamappingfromtheadnrepresenting grouprepresentsagroupofserversandinthelatteritrepresentsacollectionofmulticast 3.2AnycastGroups Ananycastdomainname(ADN)uniquelyidentiesa(potentiallydynamic)collectionofIP unicastipaddressesorentirelyofmulticastipaddresses3.intheformercasetheanycast addresses,whichconstitutesananycastgroup.ananycastgroupcanbemadeupentirelyof metric. theclassofquorumconsensusapplicationsthatwedescribedabove.fortheseapplications,it addressesinthegrouprepresentstheaddressofoneofthereplicatedserversprovidingthe concerned),andtheanycastingservicemakesadecisionbasedonperformanceand/orpolicy service.reachinganyoneofthemisacceptable(asfarasservicecontentandfunctionalityis forwardandderivesfromtheoriginalmotivationoftheanycastingparadigm.eachoftheip Weallowacollectionofmulticastaddressestoformananycastgroupinordertosupport ThemotivationforcollectingasetofunicastIPaddressesintoananycastgroupisstraight- makessensetodenemulticastgroupsmadeupofsubsetsofthesetofallcoordinators,assign eachsubsetamulticastaddressandcollectthemulticastaddressesintoananycastgroup.the applicationthenwouldusetheadnassociatedwiththisgrouptorefertothecoordination service.theanycastingimplementationwouldthenmaptheadnintoamulticastgroup withthedesiredquality(e.g.,leastaveragecoordinatorloadordistancefromclient). thispapersincewecannotconceiveofanapplicationthatwoulduseit. aliases)5.inthiscasetheanycastingserviceprovidesamappingfromanadnintoahost facilitatetheprocessofselectingfromamongthegroupmembersaswillbediscussedlater. lookup.ontheotherhand,knowingthedomainnamesthatmakeupananycastgroupmay domainnameoralias.obtainingtheipaddressinthiscaserequiresanadditionaldnsserver 3Althoughthegeneralcaseofamixtureofunicastandmulticastaddressesispossible,wedonotconsideritin 4AfurthergeneralizationofourdenitionallowsthecollectionofanycastIPaddressesintoananycastgroup Itisalsopossibletodeneananycastgroupasacollectionofserverdomainnames(or representedbyanadn.werelegateconsiderationofthis\nestedanycasting"tofuturework. anycastgroupisacollectionofhostsandnotacollectionofgroupsofhosts. 5Becausethereisnoexistingconventionfornamingmulticastgroupsthiscurrentlyonlymakessenseifthe 5

7 FIGURE2AnycastResolverArchitecture Local Anycast Resolver 3: Request for ADN X Members and Metrics Authoritative Anycast Resolver for ADN X 5: Caches ADN X 2: Determines Authoritative X = IP addr. 1 Membership and Metrics; IP addr. 2 Initiates Metric Collection... IP addr. 3 systemingeneral,andtheanycastresolverarchitectureinparticular.weproposeanapproach 3.3AnycastDomainNamesandResolverArchitecture 4: List ADN X Members and Metrics 1: Request ADN X 6: Response DNS[12],allowingustoconsiderissuesrelatedtoanycastresolverdesignseparatelyfrom Internetinfrastructure. inthispaperthatisderivedfromtheinternetnaminganddirectoryservicearchitecture.such anapproachmakesitstraightforwardtointegrateouranycastingarchitectureintotheexisting Theissueofthestructureofanycastdomainnamesinuencestheoperationoftheanycasting withtheoperationofdns. othernameserviceissues.inreality,thefunctionsofananycastresolvercouldbeintegrated Weprefertoviewtheanycastresolveraslogicallydistinctfromothernameserverslike Ananycastdomainname(ADN)isoftheform<Service>%<DomainName>.Sucha Client guredwiththeaddressofitslocalanycastresolver(inthesamewaylocaldnsserversare oftheadnidentiestheservicewithintheauthoritativeresolver. namewilltypicallybeusedasanargumenttoalibrarycallthatinvokestheanycastingservice indicatesthelocationoftheauthoritativeanycastresolverforthisadn.theservicepart andresultsinthemappingofthisadntoanipaddress.thedomainnamepartofthesystem localresolver.determiningtheaddressoftheauthoritativeanycastresolverforaparticular partoftheadnandobtainstheanycastgroup'sinformationwhichisthencachedinthe ADN,itcanprocessthequeryimmediatelyandreturntheappropriaterepsonse.Otherwise, thelocalresolverdeterminestheaddressoftheauthoritativeresolverforthedomainname congured).ananycastclientmakesitsinitialanycastquerytoitslocalresolver.ifthe resolverisauthoritativefortheadninthequeryorifithascachedinformationaboutthe TheanycastresolverarchitectureisshowninFigure2.EachnetworklocationisprecondomaincanbedoneusingthesametechniqueusedforDNStodetermineanauthoritative nameserver[12]. Thishierarchicalnamingallowsuserstodenetheirownanycastgroupsandmaintain 6

8 suchgroupsinlocalanycastresolvers.bypropagatingtheadnofalocally-denedanycast group(includingthenameofthedomaininwhichitsauthoritativeresolverresides)allows otherstomakeuseofthisanycastgroup. ADNtoIPaddress.Thisinformationincludes: 1.ThelistofIPaddressesthatformparticularanycastgroups.Authoritativeresolvers 2.Themetricinformationassociatedwitheachmemberoftheanycastgroup.ThisinformationismaintainedindependentlyateachanycastresolverthathastheADNgroumation,metricsmaintainedattheauthoritativeresolverare,ingeneral,oflittlevalue membershipinformationcached.becauseofthelocalsignicanceofthemetricinfor- Ananycastresolvermaintainstheinformationnecessarytoperformthemappingfrom maintainthedenitivelist,whereaslocalresolverscachethisinformation. 1.Theanycastgroupmembershipinformationisupdatedaccordingtosomemembership Theinformationmaintainedintheresolverisupdatedusingtwoseparatemechanisms: astheybegintogathertheirownmetricinformation. memberlistforagivenadn.localresolverscanusethisinformationas\hints"initially informationwheneveritreceivesarequestfromanotherresolverfortheanycastgroup tootherresolvers.theauthoritativeresolvermayprovideitslocallymaintainedmetric shouldbe)dependentontheparticularsoftheservice.forexample,thelistofftp protocol.wedonotdiscussthedetailsofsuchaprotocolherebecauseitcan(and locationsforapopularwebbrowsercanbedisseminatedfromahomelocationtovarious 4InteractingwithAnycastResolvers 2.Theserviceperformancemetricinformationwouldtypicallyneedtobemaintaineddynamically.WeproposeanddiscusssomemechanismstoachievethisinSection5. beadministrativelyconguredinanycastservers. anycastservers.forthewebsearchservice,ontheotherhand,thelistofserverscan First,theanycastclientgeneratesaquerythatispassedontotheanycastresolver.After processingthequery,theresolvergeneratesaresponsewhichissentbacktotheclient.in general,anadditionalprocessingstepisperformedattheclienttoyieldthenalipaddress Ananycastresolverisconsultedthroughtheuseofanycastqueriesandtheresolverresponds withanycastresponses.thebasicanycastingquery/responsecycleisillustratedinfigure1. areappliedtotheinformationmaintainedabouttheanycastgrouptoobtainanipaddress. result. Wedistinguishbetweenthreebasictypesoflters:content-independentlters,metric-based Inourproposedapproachtheanycastingserviceisaccomplishedthroughasetofltersthat 4.1FilteringandDecisionMaking resolver.aseparate(butstraightforward)protocolneedstobedenedforsuchanexchange. Anothermessageexchangemayberequiredbetweenalocalresolverandanauthoritative 7

9 ltersandpolicy-basedlters.wefurtherdistinguishbetweentwolocationsforltering: withintheresolverandattheuser.(seefigure1.) oftheanycastgroupofsomegivensize. basedsolelyonmembershipinformationandnotbasedonanyothercriteriamaintainedor knownbytheserver.examplesofsuchltersinclude:1)theselectionofanymemberat random,2)theselectionofallmembersoftheanycastgroup,or3)theselectionofasubset ciatedwiththemembersoftheanycastgroup.thefollowingvariationsareofinterest: Content-independentlterscanbeusedtospecifyaselectionofanycastgroupmembers Selectthebestanycastgroupmember(s)accordingtoafunctionofoneormoremetrics. Selectthebestanycastgroupmember(s)accordingtoasinglecriterion. Metric-basedltersspecifyselectionaccordingtothevaluesofoneormoremetricsasso- Selectthebestanycastgroupmember(s)resultingfromthesequentialapplicationof Forexample,aweightedsumofmetricscanbeused,allowingaclienttocontrolthe importanceofeachmetricintheoverallevaluationofananycastgroupmember. thebroadrangeofothercriteriathatmightaecttheselectionofanaddress.policylters Policy-basedltersarenotbasedonmeasurementsofperformance,butratherencompass metricisusedtofurtherrenetheselectionset.forexample,onemightrstselecta weblocationthathasthefastestresponsetime,andthen(amongstthefastest)choose thelocationthatistheleastnumberofhopsaway. process.inthecaseoftwolters,therstmetricisgiventoppriorityandthesecond lters.bycomposingltersinseries,onecanspecifyastrictpriorityintheselection arelikelytobeboolean,inthesensethatanaddresseithermeetsorfailsthepolicycriteria. Policylterstendtoruleoutgroupsofaddresses;forexample,onemightbeinterestedin thenapplypolicy,however,someothercompositionofthetwotypesofltersmayalsobe useful.applyingpolicyltersmaybeeasierifthemembershipoftheanycastgroupisknown accessingallserverscontainingincometaxinformationthatarenotrunbytheirs.perhaps bydomainnamesratherthanjustipaddress.iftheanycastgroupmembersaremaintained themostnaturalinteractionofmetricandpolicy-basedltersistorstapplymetricsand cantaketheresolverlistandfurtherapplylterstoselectasinglemember. byipaddress,mappingtheseaddressesintodomainnames(e.g.,throughtheuseofdns beginswiththesetofanycastgroupmembersandapplieslterstoproducealist.theclient aretwolocationsforlters:intheanycastresolverandattheanycastclient.theresolver providetheanycastclientwithalistofaddressesthatmeetthespeciedcriteria.thusthere PointerQueries)mayberequiredbeforetheapplicationofapolicylter. wheretheyareapplied.inourgeneralizedversionofanycasting,theanycastservicemay privacyregardingtheaddressselectionprocess.followingthetaxexampleabove,onemay Theprovisionofltersattheclientallowsformorelocalcontrol,includingincreased Relatedtotheissueoftheorderinwhichltersareappliedistheissueofthelocation notwanttomakeknowntothesystemthefactthatirs-runserversaretobeavoided.thus thepolicylterthatexemptsirs-runserverscouldbeappliedlocallyratherthanspecied tothesystem.anotherexamplecouldbetheuseofalocally-preferredsubsetofserversin thenusedasanalltertotheresultsobtainedfromtheanycastingservice. theformofalocally-denedanycastgroup.themembershipofthislocally-denedgroupis 8

10 theltersexercisedinthesystemwillbepredominantlymetric-based. membersissuccessfulifthenumberofaddressesproducedisacceptabletotheclientprocedure.inallcasesthereturnofexactlyoneipaddressforanapplicationtouseshouldbe Obviously,thenecessaryinformationmustexistatthelocationwherethelterisapplied. SuccessfulFilteringWesaythatasequenceofltersappliedtothesetofanycastgroup Itseemslikelythattheltersexercisedattheuserwillbepredominantlypolicy-based,while anacceptableoutcome.ideally,theapplicationprocedurethatinvokestheanycastingservice Thisisbecause,ingeneral,itisnotalwayspossibletoknowaprioritheoutcomeofthe shouldalsobecapableofdealingwiththecasewheremultipleornoaddressesarereturned. applicationofasetoflterstoananycastgroup.inthecasewheremultipleaddressesare Incasetheanycastserviceinvocationreturnsnoaddresses,afallbackpositionneedstobe returnedfromtheanycastserviceinvocation,theapplicationcanarbitrarilypickoneofthe programmedintotheapplicationprocedure.forexample,itcouldretrywithanothersetof returnedaddresses.thisisequivalenttoapplyingyetanother(content-independent)lter. 4.2FilterSpecication Wenowturnourattentiontotheissueofhowaclientmayspecifyalterfortheanycast ltersoracontent-independentlteraskingforanygroupmemberatrandom. resolvertouse.localltersarerelativelyeasytodealwithsincetheyaretypicallyconceived, speciedandappliedinthesamelocation.resolverlters,ontheotherhand,areconceived availablethiswaythroughwell-knownidentiers.inthesecondtypeoflterspecication, theformofafunctionofmetricsoraproceduredescribingsomeelaboratesequenceoflter alterthatisalreadybuiltintotheresolver.inthiscasealltheclienthastodoisrefer theclientprovidesaproceduraldescriptionoftheoperationofthelter.thiscanbein tothislterusingsomeagreeduponidentier.weexpectthatmanypopularlterswillbe applications.aninterestingpossibilityistheuseofjava[33]astheprogramminglanguageto ataclientbutarerunbytheresolver,whichaddscomplexitytotheirspecication. specifylters,withthecommunicationofthejavaprogramtotheanycastresolverincluded Weenvisiontwotypesoflterspecications.Intherstcase,theclientdesirestouse Argumentstothiscallcanbeusedtospecifythelteringdesired,eitherbyidentifyinga thatcanbeusedbyanapplicationtoinvoketheprocessofadntoipaddressmapping. alternatives.intherst,anewapplicationlayerinterface(api)functioncallisprovided, aspartofanycastclient/resolverinteraction. mannersimilartohowthegethostbyname()functioncalltypicallyinvokesthednsservice. server.anexampleofthisinthecontextofthesocketsapiisthedenitionofa(new) getanyhostbyname()functioncallthatcanbeusedtoinvoketheanycastingserviceina built-inlterorbypointingtosomelocalfunctionorproceduretobecommunicatedtothe Anotherimportantissueiswhereandhowlterspecicationsaremade.Weexploretwo Names(MQ-ADN)toconveythedesiredlteraspartoftheADNbeingsenttotheanycast server.ametric-qualiedadnisoftheform<filter-specification>.<adn>where Filter-Specificationprovidesinformationabouttheltertobeusedbytheserver.This isrelativelystraightforwardifabuilt-inlterisbeingspecied.forexample,thename AnalternativemethodforspecifyingltersistouseMetric-QualiedAnycastDomain 9

11 insuchcasestohavethemq-adnsgeneratedautomaticallyusinglibraryfunctionsthatare madeavailableattheapi. complexurlsgeneratedbycgiprograms[34]toperformdatabasesearches.itisnecessary ADN.Usingmetric-qualiedADNsismorecomplexwhenaproceduraldescriptionofthelter isdesired.onepossibilityistoallowcomplexfilter-specicationinthemq-adn,akintothe ServerLoadbuilt-inlterontheanycastgrouprepresentedbythewwwsearch%cc.gatech.edu ServerLoad.wwwsearch%cc.gatech.educouldconveytotheresolverthedesiretousethe needtopickanftpserverandthenissuethecommand: >ftpftp.freebsd.org applicationsneednotberewritten.manyoftheseapplicationstakedomainnamesasinput; oftheanycastingservice.inthiscase,themq-adnisaugmentedbythesux\.any". onewouldsimplyneedtosubstituteamq-adninstead,toallowtheapplicationtomakeuse UsingtheaugmentedMQ-ADN,itwouldbepossiblefortheusertoissuethecommand: Forexample,auserwantingthemostrecentversionofFreeBSDusingftpwouldordinarily MQ-ADNscanbeenhancedtoallowexistingAPIcallstobeleftintact,thusexisting software,withthemembershipoftheanycastgroupmaintainedatthecc.gatech.edudomain's anycastresolver. whereftpfreebsdisananycastgroupconsistingofserversthatcontainthedesiredfreebsd >ftpthroughput.ftpfreebsd%cc.gatech.edu.any case,thenananycastresolverqueryisformulated,afterparsingthenametoobtainthe callandcheckiftheargumentisanadn(i.e.,endswiththe.anysux).ifthatisthe 4.3AnImplementationusingAugmentedMetric-QualiedADNs softwareinthisinstance. Inourimplementation(showninFigure3)weinterceptcallstothegethostbyname()socket WenextexplainhowourprototypeimplementationallowstheuseofunmodiedFTP theusualprocedureiscalleddirectly.thisallowsustousetraditionalapplicationswithout resolverlterspecication.thisqueryissenttotheanycastresolverwhichreturnsasetof 5MetricsandMetricDetermination modicationandgivestheoptionofusingthemwithorwithouttheanycastingfeature6. returnthedesiredipaddress.ifthegethostbyname()argumentisotherthananadnthen hostdomainnames.thesetispassedthroughaclientltertoreduceittoasinglehost domainname.thisnameisthenusedasinputtotheoriginalgethostbyname()procedureto metricdatawillnottypicallybesevere;ratherthanselectingthe\best"server,theservice values.further,theperformancepenaltyassociatedwithout-of-dateorslightlyinaccurate mayidentifya\nearly-best"server.asapracticalmatter,evenmetricsthataresuciently formancemetricswith\reasonable"accuracy,withoutundulyloadingthenetworkorthe servers.notethatthemeasurementsneednotbeperfectlyaccurate;theanycastmechanismreliesuponrelativeorderingtodeterminethebestserver,ratherthanabsolutemetric Theuseofmetric-basedltersintheanycastingservicerequirestheabilitytomeasureper- library.wesuccessfullyusedthistechniquetoaugmentmanyapplicationswithanycastingcapabilities,including weareunabletosubstituteourownlibrarysubroutines. themosaicwebbrowser.however,someapplications,suchasthenetscapebrowser,arestaticallylinked,andthus 6Useofthistechniqueonexistingapplicationsrequiresdynamiclinkingoftheapplicationcodetoourown 10

12 FIGURE3TheStructureofanImplementationusingMetric-QualiedADNs Gethostbyname other.any accuratetoallowonetoavoidtheworstoftheservers,e.g.,thosethataredownorcurrently Anycast Server accessingreplicatedservers.wegiveexamplesofmeasurementstakenonvarioussystems; thesemeasurementsindicatethatsome,butnotall,metricsandserversexhibitsucient Client Filter performancevariationtowarrantanycasting.wenextturntotheissueofcollectingmetric unreachable,willmakeforausefulservice. Webeginthissectionbyexaminingasetofmetricsthatmaybeofinteresttoclientsin Domain Name theloadthattheyplaceonthenetworkandservers,andintheaccuracyofthemeasurements. DNS Weexaminethreeofthemethods proxyprobing,userexperienceandserverpush in information.wedescribefourpossiblewaystomaintaininformation;thesemethodsdierin moredetailwithinanexampleenvironment.weconcludebysummarizingtheperformance andsystemoverheadcharacteristicsofeachofthemethods. IP Address serverresponsetime,2)server-to-userthroughput,3)serverload,and4)processorload7.we focusonthesemetricsbecausetheyarelikelytobeofinteresttoclientsinselectingfromaset batch,real-time)andthepreferencesoftheusers. measured.metricsofinterestwilldependonthecharacteristicsoftheservice(e.g.,interactive, 5.1Metrics ofreplicatedservers.further,theyrepresentdiversityinthecomponents(server,path,client) involvedindeterminingthemetricvalueanddiversityinthetimescaleonwhichthemetric Thearchitecturedescribedthusfarissucientlygeneraltosupportanymetricthatcanbe valuechanges.weenvisionthatadditionalmetrics(e.g.,packetloss,delayjitter)wouldalso 7Theoriginalanycastingproposal[7]wasprimarilyconcernedwithaserverdistance(hopcount)metric;measuringserverdistancehasalsobeenstudiedextensivelybyGuytonandSchwartz[18]. Tobetterunderstandhowanycastingwillperform,weexaminefourspecicmetrics:1) 11

13 FIGURE4HTTPResponseTimeatNews-OrientedServers 4.5e+07 5e+07 ESPNet CNN FN USA Today CNN 4e e+07 beofinteresttocertaintypesofapplications. 3e+07 aresignicantdierencesacrossserversatvarioustimes,otherwisetheselectionofaserver 2.5e+07 couldjustaswellberandom.further,thevariationinametricmustoccuronatimescale 2e+07 thatispracticaltotrackusingmeasurementtools.wenowpresentexamplemeasurements foreachofthefourmetrics,withtheaimofbetterunderstandingwhichmetricsareamenable 1.5e+07 toanycasting. 1e ServerResponseTime Theuseofmetricstoidentifythe\best"serverispredicatedontheassumptionthatthere 5e+06 browsing.wemeasureserverresponsetimebymeasuringtheround-triptimeforaquery8. Serverresponsetimeisofinterestforinteractiveclient-serverapplicationssuchasweb 7:15 A.M. 2:00 P.M. 20:50 P.M. WehavemeasuredHTTPserverresponsetimeatfournews-orientedservers Time EST9.Thex-axisintheplotistheprobeindex;they-axisisthemeasuredresponsetimein workstationsonourownresearchgroup'ssubnet.atypicalresultisdepictedinfigure4, espnet.sportszone.com, forvariousperiodsoverseveralweeks.themeasurementsweretakenfromrelativelyunloaded representingservermeasurementsevery5minutesduringa13hourperiodfrom8:00to21:00 constructed;mostoftenweused\anycasttest(testversion)"followedbyoneormorecarriagereturns.aprotocol specicconnectionwasinitiatedtotheserver'sspecicserviceport(tcpport80incaseofhttpservers).the microseconds.each25ticksonthex-axisisapproximatelytwohours. orlunchtimeoverthethreeustimezones.thecnnnewsserverisalsoactiveinthelater afternoon.theseresultsdemonstrateseveralordersofmagnitudedierencebetweenthebest timetocreatetheconnection,sendthequery,andthetimefortheservertoprocessandrespondtothequerywere andworstserversatagiventime,andalsodemonstratethatnooneserverisconsistently 8Specically,aquerystringconsistingofasetofcharactersnotlikelytobeintheserver'svocabularywas Theseserversshowtime-of-daytrends,withsignicantactivityfrom11:30to15:00EST, measuredusingtheunixgettimeofday()systemcall.theserverresponsewasparsedtomakesuretheserver tripnetworklatencyandthetimefortheservertoforkaprocessandrespondtothequery.inthecasewhere actuallyrespondedtothequery,andtimeouts(ifany)werelogged.thus,thequeryservestomeasuretheround theserverloadislight,thepathlatencywilldominatethismetric.notethatasimilarmethodcouldbeusedto 12 Response Time (Microseconds)

14 FIGURE5FTPThroughputunderVaryingLoad 35 ATM Ethernet timemetric. best.theseserversappeartobefairlygoodcandidatesforanycastingbasedonaresponse fromourmeasurementmachinetotheserverssharemanylinks,thuschangesintheloadof tionsinresponsetimebyanorderofmagnitude,rangingfromabout5secondsto25seconds. Interestingly,theresponsetimesatdierentserverswerehighlycorrelated,alltendingtohave thatthepredominantfactordeterminingresponsetimeforthesesearchenginesislatency withinthenetwork,notloadorprocessingattheservers.wehaveveriedthatthepaths roughlythesameperformanceatagiventime.weinferfromthis(andotherdatacollected) 5 theselinkswillhaveasimilareectonallresponsetimes. Weperformedsimilarmeasurementsonasetofsearchengines.Weagainobservedvaria- 0 Serversthatareinacompetitivebusiness(e.g.,searching)haveaneconomicincentiveto Theseexperimentsillustratethatsomeserversandmetricsarenotamenabletoanycasting Number of Requests provisionadequatelytoavoidserverover-loading.auserwillprobablydojustaswellto selectoneoftheseserversarbitrarily. connectionbetweentheclientandtheserver(ethernetoratm).eachtransferconsistedof functionofserverload(measuredasthenumberofclientsbeingserved)andtypeofnetwork transferssuchasftp.wehavemeasuredserver-to-userthroughputusingftpclientsthat wehaveinstrumented10.figure5demonstrateshowserver-to-clientthroughputvariesasa 5.1.2Server-to-UserThroughput signicantamountsofdata(e.g.,webpageswithgraphics)andformoretraditionalbatch-style 10Theclientwasmodiedsothatitdidnotwritereceiveddatatothedisk;wedidthistoavoidhavingthedisk Server-to-userthroughputisausefulmetricforinteractiveapplicationsthatmusttransfer writebethebottleneckinthethroughputmeasurement. 13 Average Throughput at Client

15 FIGURE6FTPThroughputwithVaryingDistance 10 Internal [2 hops] Internal [3 hops] Internet [15 hops] Internet [16 hops] 8 10Mbytes. 6 4 canvaryconsiderably,bothacrossserversand(toalesserextent)onasingleserverover time.theserversthatareclosertotheclientgenerallyhavehigherthroughputandgreater doneforeachserver;thex-axisindicatestheindexofthetransfer,andthey-axisindicates Figure6demonstratesvariationinserver-to-clientthroughputbasedonthe(hop-count) 2 thethroughputtendstobe. distancebetweenserverandclient.tentransfersofapproximately100kbyteseachare themeasuredthroughput.theseresultsindicatethatthroughputbetweenserverandclient ServerLoad reasonable(inverse)proxyforthroughput;thefurthertheserverisfromtheclient,thelower variationinthroughputovertime.notethattheseresultssuggestthathopcountmaymakea Client Access Index applications.weobtainapproximatedataabouttheserverstatusbypost-processingthelogs becauseitismorepracticaltotrack,andmayprovidesucientinformationtobenetsome fromanhttpserver.specically,thelogsthatweaccessedcontainedarrivaltimeinformation forhttprequests.wecountedtheaggregatenumberofarrivalsintimeintervalsoflength Serverloadisamorecoarsemetricthanresponsetimeorthroughput.Weincludethismetric Tseconds,anddenedtheservertobe\loaded"ifthenumberofrequestsinaninterval exceededr.clearlytheparticularresultswillbesensitivetothechoiceoftandr,which Wehaveoverlaidresultsfromthreedierentdays Saturday,May11-Tuesday,May14, shouldbechosenbasedonservercapabilitiesandtimescaleofuctuationinload TheseresultsusetimeintervalT=60secondsandthresholdR=85requests.Wenote Figure7showstheloadstatusofanhttpserverwithinourdepartment( 14 Server Throughput (Mbps)

16 FIGURE7LoadStatusofHTTPServer Sat May Sun May Mon May Tue May ProcessorLoad Measurementofprocessorloadisausefulmetricforanapplicationthatisquitedierent fromtheserveraccessdescribedthusfar.forexample,aprocessmanagerresponsiblefor fourtracesexhibittime-of-dayvariations. distributingtasksontomachinescanviewasetofprocessorsasareplicatedservice,with thattheactivityisdistinctlydierentontheweekenddaysversustheweekdays.further,all servers(forgeandlennon)overatypical24hourperiod,withmeasurementstakenevery selectionofthebestmachinecorrespondingtotheleastloadedprocessor. network.figure8showstypicalresultdepictingtheload(asreportedbyrup)ontwocompute Wehaveexperimentedwithmeasuringprocessorloadforvariousmachinesonourcampus 12:00 A.M. 12:00 P.M. 12:00 P.M. Time 5minutes.Thisportionofthedatarunsfromabout2amto2am.Thesevaluesalsoshow Theprevioussubsectionindicatesthatthereareserversandmetricsthatcanbenetfrom 5.2MetricCollectionTechniques ofprocessors.inthisexamplebothmachineshavefourprocessorseach. thenight(roughly2am).toproperlycomparetheloadsonthevariousmachines,thedata time-of-daytrends,withoneoftheservers(forge)seeingsignicantactivityinthemiddleof anycasting.wenextconsiderhowtoecientlyandaccuratelymaintaindatabasesofthe shouldbenormalizedusingsomemeasureofprocessingcapability,suchasnumberandtype performanceinformationinanycastservers'database: metricvalues.wehaveidentiedfourpossibleapproachestomaintainingreplicatedserver or more accesses this minute?

17 FIGURE8ProcessorLoadonComputeServers Forge Lennon RemoteServerPerformanceProbing:Inthistechniqueprobingagentsareinchargeof ServerPush:Itmaybeadvantageousinsomecircumstancestohavethereplicated partsofthenetworkandtheserverthattheclientrequestwill\exercise".wehave asetofclients.theseprobesneedtobedesignedtomimic(asmuchaspossible)the experimentedwiththistechniqueinourimplementation.insection5.3wegivean exampleofourresultsandexaminetheissueofprobelocationandaccuracy. periodicallyqueryingthereplicatedserverstodeterminetheperformancethatwillbe experiencedifaclientweretorequestservice.eachprobingagentactsasaproxyfor 2 0 servertoupdateitsperformanceinformationonlywheninterestingchangesareobserved. theanycastservers.theprimaryadvantageofthistechniqueisscalability:itallowsthe serverssend(orpush)therelevantperformanceinformation(asmeasuredlocally)onto 2:00 A.M. 2:00 A.M. Time thatmaintaininformationabouttheserver.theanycastresolverscanjoinwell-known multicastgroupsforeachserverthattheyareinterestedin,allowingtheserversto Further,theupdateinformationcanbe(networklayer)multicasttoallanycastresolvers disseminateperformanceinformationwithoutknowingtheidentitiesoftheresolvers. Themulticastgroupsmaybeorganizedinotherways,forexamplewithonemulticast alternativesdierinthenumberofmulticastgroupsrequired,theeaseofmaintaining groupperanycastdomainname,oronemulticastgroupperanycastgroup.these themulticastgroupsandtheamountofnetworktracrequiredfordisseminatingserver information. Wehaveconsideredseveralvariationsonserverpushbasedonthetypeand/orfrequency ofinformationthatispushed.insection5.4wedescribeanexperimentthatconsiders pushinggoodnews,pushingbadnewsandperiodicallypushingcurrentstate. 16 Load

18 3.ProbingforLocally-MaintainedServerPerformance:Anotherpossibilityistohaveeach 4.UserExperience:Thelasttechniqueismotivatedbytheobservationthatuserscurrently niques.sinceprobesmerelyreadfromalocally-maintainedle,theymayrepresentless readablele.remoteprobinglocationscanthenreadtheinformationinthele(as sense,thisisahybridbetweentheprobing-for-performanceandtheserver-pushtech- opposedtoattemptingtoexercisetheserver)toobtainthedesiredinformation.ina replicatedservermaintainitsownlocallymonitoredperformancemetricsinaglobally serverperformance.theprimaryadvantageofthismethodisthattheinformationis time.collectinginformationaboutpastexperienceoersacoarsemethodofmaintaining collectedforfree;noadditionalburdenisplacedontheserverorthenetwork.the particularservertobeunreachable,thatserverislikelytobeavoidedforaperiodof makeserveraccessdecisionsbased,inpart,onpastexperience.thatis,ifonendsa ofaburdenontheserverthantheprobing-for-performanceapproach. costsassociatedwiththreeofthesetechniques. 5.3Example:ProxyProbingandUserExperience Wenextdescribeseveralexperimentsthatexamineperformanceandsystemoverhead amongclients.forexample,agatewayintoacampusmightmaintainserverperformance quantityandaccuracyoftheinformationcanbeincreasedbysharingofexperience Ideally,theprobeswhichcollectmetricinformationwouldberuninasfewlocationsaspossible informationbasedontheexperienceofallclientsonthecampus. andtheserver,thusfewerprobeswillcauselessoverheadformetricmonitoring.again,it yetstillprovideaccuratedatatobeusedinltering.eachprobeputsloadontothenetwork comparisonreecttheviewattheuser. isnotnecessarythattheabsolutevaluesofthemetricsbeprecise,onlythattherelative runateachuser,andthuswouldfollowapathfromtheusertotheserver.forscalability,it ispreferabletorunpath-dependentprobeson\proxies"thatprovidemetricdataforagroup fromalocationthatincludes(somepartof)thepath.attheextreme,theseprobescouldbe ofusers.theseproxiesmightbechosentocoverallhostsonasubnetwork,withinarouting domainorwithinageographicregion.asaproxycoversalargersetofhosts,thenumberof proxies(andthusnetworkloadduetoprobing)decreases,howeversodoestheaccuracyof theproxydatarelativetoanarbitraryuserinthegroup. Metricswhicharedependentonthepathfromtheusertotheserverwillneedtobeprobed maintainmetricdatawithnoadditionalloadonthenetworkorservers.weexpectmetric thesameasoccursintheproxyprobingmethod.thustheexperimentdescribednextalso providessomeinformationabouttheaccuracyofuserexperience. datathatismaintainedusingexperiencetobelessaccuratethandatathatismaintainedbya moreregularmeasurementprocess.therearetwosourcesfortheinaccuracy:(1)theservers performanceatanotherclient.wenotethatthissecondformofinaccuracyisessentially maynotbeaccessedregularly,and(2)theexperienceatoneclientmaynotaccuratelyreect Theexperienceofaclient(oragroupofclientsthatshareinformation)canbeusedto thefollowingexperimentalsetup.weranprobesonvemachines,threeonourlocalcampus Wehaveexploredthequestionofproxylocationandaccuracyoftheproxydatausing 17

19 local local local remote remote BestSecondThirdFourthFifthSixthSeventh theothertwowerechosentogetgoodgeographicaldistribution12.measurementsweretaken everythreeminutesforabout13hoursfrom23:00estto12:00estonthefollowingday. timetoasetofsevenserverslocatedaroundtheworld.fiveofthemarefreebsdservers; networkandtwoonthenetworkofanothercampus11.wemeasuredhttpserverresponse Inordertocomparetheaccuracyofdierentproxies,thetimesatwhichthereadings Table1:AccuracyofProxies betweenthetwocampuseswheretheproxieswerelocated). processesasquicklyaspossiblefromamachinerunningshellsoneveryproxy.moreelaborate theprobesareoccurringwithinsecondsofoneanother(modulodierencesinsystemclocks synchronizationiscertainlypossible,howeveraninspectionofthetimelogsindicatesthat inaccuracycausedbytakingthemeasurementsatdierentlocationsfromtheinaccuracy causedbytakingthemeasurementsatdierenttimes.)thisisaccomplishedbystartingthe aretakenmustberelativelysynchronized.(thatis,wewouldliketoisolatetheissueof thedatalescontainedtime-consistentprobemeasurementsacrossallproxies. probes.wepost-processedthedatalestondthemissesateachproxyandremovethe correspondingtimevaluesfromthedatalesfortheotherproxies.the\clean"versionsof thebaseandbyeachproxy,basedonthemeasurementdata.wethenrecordhowoftenthe othersasproxies.foreachmeasurementpoint,wedeterminethebestserverasselectedby proxiesagreeordisagreewiththebaseontheselectionofthebestserver.moreprecisely, Inthissetofdata,wedidndthattheproxiesoccasionallymissedregularlyscheduled wecounthowofteneachproxyagreeswiththebase(i.e.,selectsthesamebestserver),how oftentheproxyselectsthesecondbestbaseserver,andsoon.table1showsthepercentage oftimethateachproxypickseachserver. Toassesstheaccuracyoftheproxies,wedesignateonemachineasthe\base"andall withinourownlocalcampusnetwork.local1isthebasemachine.remote1andremote2 asmightbeobtainedwithoutanycastingsupport wouldpickthebestserver1/7thor localcampusagreemoreoftenandmoreaccuratelywiththebasethantheproxiesonthe remotecampus.eventheproxiesontheremotecampusgivefairlygoodresults,selectingthe bestorsecondbestserverabout50%ofthetime.(notethatarandomselectionofservers aretwomachinesondierentsubnetsoftheremotecampus'network.theproxiesonthe 14%ofthetime.) Themachineslabeledlocal1,local2andlocal3arethreehostsondierentsubnets minnie.cs.adfa.oz.au, amorediversesetofmachinelocations,howeverwewerelimitedinouraccesstowidelydispersedsystems. chinesorfaraway.wehavedividedtheanalysisaboveintotwoparts:onecontainingthe 12Theserversweprobedwere: 11Thetwocampusesareseparatedbyseveralstatesand3-4networkhops.Theseexperimentsshouldberunwith Theresultsareaectedbywhethertheserversarerelativelynearthemeasurementma- 18

20 local local local remote remote BestSecondThird Table2:AccuracyofProxies-NorthAmericanServers local local local remote remote BestSecondThirdFourth 3showtheresultsforthesetwogroupsofservers.Theadvantageofnearbyproxiesismore threenorthamericanserversandtheothercontainingthefourforeignservers.tables2and aslongastotheforeignservers,andhavemorecommonalityforthenearbyproxies. signicantforthenorthamericanservers;thepathsfromtheproxiestotheseserversarenot Alloftheproxiesdoquitewellfortheforeignservers,witheachproxyagreeingwiththe Table3:AccuracyofProxies-ForeignServers baseover50%ofthetime.theroutestotheforeignmachineshavemanyhopsincommon; forexample,allmachinessharethesamelast15hopstotheserverinindia. 5.4Example:ServerPush Todiscusstheperformanceoftheserverpushtechnique,wemustrstdenethealgorithm thattheserverwillusetodeterminewhatinformationtopushandwhen.ingeneral,we overheadoftheupdatingmechanism. \interesting"withsomeconstraintonthemaximumfrequencyofupdatessoastoboundthe wanttheservertopushstateinformationwheneverthestatehaschangedsucientlytobe TheservermeasuresitsstateovereachintervalI.Ifthestatechangesfromtheprevious thesamecriteria.wehaveadoptedthelinkstateupdatealgorithmusedinthearpanet[35] changebyatleastc,cisreducedbyr.(notethatwhencbecomes0,thestatewillbe measurementbyatleastc,thestateispushedandcisresettot.ifthestatedoesnot andexperimentedwiththeperformanceandoverheadwithavarietyofparameters.the andareductionfactorr.thealgorithmmaintainsacurrentthresholdc,initializedtot. updatealgorithmisparameterizedbyameasurementintervali,amaximumthresholdt Notethatthetaskofupdatinglinkstateinadistributedroutingenvironmenthasprecisely pushedandcwillberesettot.)thealgorithmwillsendupdatesatleasteveryt=rtime 19

21 FIGURE9PerformanceofServerPushAlgorithms Threshold = 10 Threshold = 20 Threshold = 30 Threshold = unitsandatmosteveryitimeunits. 40 surementquantityisthenumberofconnectionsinitiatedinthemeasurementinterval.inour experimentsweusei=1minute.weprocesshttpserverlogstodeterminethepushed 30 valuesforvaryingtandr. 20 R.Ateachoneminuteinterval,wecomparethevalueoftheserverloadbasedonthepushed 10 thedierencebetweentheestimateandthetruevalue,andplotafrequencydistribution.in informationtothetruevalueofloadextractedfromthelogs.werecordthemagnitudeof WehaveimplementedthisserverpushalgorithmforanHTTPserver,wherethemea- Figure9weusedR=5minutesandvaryingT;wealsocollecteddataforR=10minutes. InFigure9weexaminetheaccuracyoftheupdatingmechanismasafunctionofTand forvariousvaluesoftandr.thenumberofupdatemessagesgeneratedisameasure Table4givestheaveragevalueoftheinaccuracyandthenumberofupdatemessages, Error Magnitude summarizestheperformancedatagivenbythefrequencydistribution.thetradeoisclear:a oftheloadplacedonthenetworkbythemechanism;theaveragevalueoftheinaccuracy moreaccuratealgorithmalsoincursalargeroverhead.interestingly,increasingthemaximum intervalbetweenupdatesfrom5to10hasrelativelylittleeectontheaccuracyforaconstant networkand/ortheserver. themetricvalue,thedesiredaccuracyoftheinformation,andthecostofburdeningthe scaleonwhichthemetricvaries,thecomponents(server,networkpath,client)thatdetermine Whichtechniqueismostappropriatewilldependonanumberoffactors,includingthetime- updatethreshold. 5.5ComparisonofMetricCollectionTechniques 20 Error Frequency

22 ThresholdBetweenUpdatesErrorUpdates UpdateMax.IntervalAverageNumberof Table4:LoadandAccuracyofServerPush LoadModLoadNetPath NetServerServerExercisesAccuracy 379 Probing ServerPush ReadingServerLogPTpYes UserExperienceNoneNo 2PTpNoModerateYes TsYes Low No Moderate Table5:ComparisonofMetricCollectionTechniques (Seenoteintext) None YesLow/Varies High rstthreecolumnsaremeasuresofsystemoverhead.thenetloadcolumnrepresentsthe Pisthenumberofproxies,Tpistheperiodofproxyprobing,Tsistheperiodofserver numberofmessagesgeneratedperunittimetoobtainthemetricdatafromoneserver,where push.notethattheserverpushmessagesaremulticastratherthanunicast.theservermod ServerLoadcolumnexpresses(relatively)howmuchadditionalloadisplacedontheserverby columnindicateswhethertheservermustbemodiedtoallowthemetrictobecollected.the thecollectionofthemetricdata.thelasttwocolumnsareperformancemeasures,indicating Table5.5summarizesthefourtechniquesbasedonperformanceandcostdimensions.The whetherthemethodexercisesnetworkpath,and(relatively)howaccuratelythemethodis methodscanbemadetoexercisethenetworkpathbymeasuringhopcountorlatencyforthe abletomaintainthemetricsthatitcanevaluate.notethattheserverpushandserverlog metricupdatemessage. 21

PowerLink Bandwidth Aggregation Redundant WAN Link and VPN Fail-Over Solutions

PowerLink Bandwidth Aggregation Redundant WAN Link and VPN Fail-Over Solutions Bandwidth Aggregation Redundant WAN Link and VPN Fail-Over Solutions Find your network example: 1. Basic network with and 2 WAN lines - click here 2. Add a web server to the LAN - click here 3. Add a web,

More information

Alteon Global Server Load Balancing

Alteon Global Server Load Balancing Alteon Global Server Load Balancing Whitepaper GSLB Operation Overview Major Components Distributed Site Monitoring Distributed Site State Protocol Internet Topology Awareness DNS Authoritative Name Server

More information

Building Nameserver Clusters with Free Software

Building Nameserver Clusters with Free Software Building Nameserver Clusters with Free Software Joe Abley, ISC NANOG 34 Seattle, WA, USA Starting Point Discrete, single-host authoritative nameservers several (two or more) several (two or more) geographically

More information

Deploying IP Anycast. Core DNS Services for University of Minnesota Introduction and General discussion

Deploying IP Anycast. Core DNS Services for University of Minnesota Introduction and General discussion Deploying IP Anycast Core DNS Services for University of Minnesota Introduction and General discussion Agenda Deploying IPv4 anycast DNS What is ANYCAST Why is ANYCAST important? Monitoring and using ANYCAST

More information

ECE 4321 Computer Networks. Network Programming

ECE 4321 Computer Networks. Network Programming ECE 4321 Computer Networks Network Programming Name Space System.Net Domain Name System (DNS) To resolve computer naming Host database is split up and distributed among multiple systems on the Internet

More information

Ethereal Lab: DNS PART 1. 1.Run nslookup to obtain the IP address of a Web server in Asia. I performed nslookup for www.rediff.com

Ethereal Lab: DNS PART 1. 1.Run nslookup to obtain the IP address of a Web server in Asia. I performed nslookup for www.rediff.com Ethereal Lab: DNS PART 1 1.Run nslookup to obtain the IP address of a Web server in Asia. I performed nslookup for www.rediff.com Screenshot taken after question 1 2. Run nslookup to determine the authoritative

More information

NET0183 Networks and Communications

NET0183 Networks and Communications NET0183 Networks and Communications Lecture 25 DNS Domain Name System 8/25/2009 1 NET0183 Networks and Communications by Dr Andy Brooks DNS is a distributed database implemented in a hierarchy of many

More information

Lecture 2 CS 3311. An example of a middleware service: DNS Domain Name System

Lecture 2 CS 3311. An example of a middleware service: DNS Domain Name System Lecture 2 CS 3311 An example of a middleware service: DNS Domain Name System The problem Networked computers have names and IP addresses. Applications use names; IP uses for routing purposes IP addresses.

More information

BEST PRACTICES FOR IMPROVING EXTERNAL DNS RESILIENCY AND PERFORMANCE

BEST PRACTICES FOR IMPROVING EXTERNAL DNS RESILIENCY AND PERFORMANCE BEST PRACTICES FOR IMPROVING EXTERNAL DNS RESILIENCY AND PERFORMANCE BEST PRACTICES FOR IMPROVING EXTERNAL DNS RESILIENCY AND PERFORMANCE Your external DNS is a mission critical business resource. Without

More information

CS514: Intermediate Course in Computer Systems

CS514: Intermediate Course in Computer Systems : Intermediate Course in Computer Systems Lecture 7: Sept. 19, 2003 Load Balancing Options Sources Lots of graphics and product description courtesy F5 website (www.f5.com) I believe F5 is market leader

More information

EECS 489 Winter 2010 Midterm Exam

EECS 489 Winter 2010 Midterm Exam EECS 489 Winter 2010 Midterm Exam Name: This is an open-book, open-resources exam. Explain or show your work for each question. Your grade will be severely deducted if you don t show your work, even if

More information

Blocking DNS Messages is Dangerous

Blocking DNS Messages is Dangerous Blocking DNS Messages is Dangerous Florian Maury, Mathieu Feuillet October 5-6, 2013 F Maury, M Feuillet Blocking DNS Messages is Dangerous October 5-6, 2013 1/25 ANSSI Created in 2009, the ANSSI is the

More information

DNS (Domain Name System) is the system & protocol that translates domain names to IP addresses.

DNS (Domain Name System) is the system & protocol that translates domain names to IP addresses. Lab Exercise DNS Objective DNS (Domain Name System) is the system & protocol that translates domain names to IP addresses. Step 1: Analyse the supplied DNS Trace Here we examine the supplied trace of a

More information

Configuration Examples for the D-Link NetDefend Firewall Series

Configuration Examples for the D-Link NetDefend Firewall Series Configuration Examples for the D-Link NetDefend Firewall Series Scenario: How to configure Anti-Spam on NetDefend Firewall Platform Compatibility: All NetDefend Firewall Series Last update: 2008-03-13

More information

State of the Cloud DNS Report

State of the Cloud DNS Report transparency for the cloud State of the Cloud DNS Report Basic Edition April 2015 2015 Table of Contents Overview Introduction 3 Anycast vs. Unicast DNS 3 Provider Overview & Current News 4 Provider Marketshare

More information

Understand Names Resolution

Understand Names Resolution Understand Names Resolution Lesson Overview In this lesson, you will learn about: Domain name resolution Name resolution process steps DNS WINS Anticipatory Set 1. List the host name of 4 of your favorite

More information

The secret life of a DNS query. Igor Sviridov <sia@nest.org> 20120522

The secret life of a DNS query. Igor Sviridov <sia@nest.org> 20120522 The secret life of a DNS query Igor Sviridov 20120522 Preface Nowadays, when we type URL (or is it a search string? ;-) into a browser (or mobile device) many things happen. While most of

More information

LAN TCP/IP and DHCP Setup

LAN TCP/IP and DHCP Setup CHAPTER 2 LAN TCP/IP and DHCP Setup 2.1 Introduction In this chapter, we will explain in more detail the LAN TCP/IP and DHCP Setup. 2.2 LAN IP Network Configuration In the Vigor 2900 router, there are

More information

A Task-Based Adaptive-TTL approach for Web Server Load Balancing *

A Task-Based Adaptive-TTL approach for Web Server Load Balancing * A Task-Based Adaptive-TTL approach for Web Server Load Balancing * Devarshi Chatterjee Zahir Tari RMIT University School of Computer Science and IT Melbourne, Australia zahirt@cs cs.rmit.edu.au * Supported

More information

State of the Cloud DNS Report

State of the Cloud DNS Report transparency for the cloud State of the Cloud DNS Report Basic Edition August 2015 2015 Table of Contents Overview Introduction 3 Anycast vs. Unicast DNS 3 Provider Overview & Current News 4 Provider Marketshare

More information

Wireshark DNS. Introduction. nslookup

Wireshark DNS. Introduction. nslookup Wireshark DNS Introduction The Domain Name System (DNS) translates hostnames to IP addresses, fulfilling a critical role in the Internet infrastructure. In this lab, we ll take a closer look at the client

More information

Non-authoritative answer: home.web.cern.ch canonical name = drupalprod.cern.ch. Name: drupalprod.cern.ch Address: 137.138.76.28

Non-authoritative answer: home.web.cern.ch canonical name = drupalprod.cern.ch. Name: drupalprod.cern.ch Address: 137.138.76.28 1. Run nslookup to obtain the IP address of a Web server in Europe. frigate:desktop drb$ nslookup home.web.cern.ch Server: 130.215.32.18 Address: 130.215.32.18#53 Non-authoritative answer: home.web.cern.ch

More information

BELNET: Service Level Description Version (29/7/2009)

BELNET: Service Level Description Version (29/7/2009) BELNET: Service Level Description Version (29/7/2009) Table of Contents I Introduction...3 A Goal of this document... 3 B Publishing... 3 C Definitions... 3 II Network Services...5 A Connectivity (BELNET

More information

CS 348: Computer Networks. - DNS; 22 nd Oct 2012. Instructor: Sridhar Iyer IIT Bombay

CS 348: Computer Networks. - DNS; 22 nd Oct 2012. Instructor: Sridhar Iyer IIT Bombay CS 348: Computer Networks - DNS; 22 nd Oct 2012 Instructor: Sridhar Iyer IIT Bombay Domain Name System Map between host names and IP addresses People: many identifiers: name, Passport #, Internet hosts:

More information

PS S S: Sensor. PS: Persistent State F: Forecaster Workstation 1 S PS. F Client

PS S S: Sensor. PS: Persistent State F: Forecaster Workstation 1 S PS. F Client ResourcePerformanceForecastingServicefor TheNetworkWeatherService:ADistributed Metacomputing auniversityofcalifornia,sandiegoanduniversityoftennessee,knoxville RichWolskia;1NeilT.Springb;2JimHayesb;3 buniversityofcalifornia,sandiego

More information

Domain Name System. CS 571 Fall 2006. 2006, Kenneth L. Calvert University of Kentucky, USA All rights reserved

Domain Name System. CS 571 Fall 2006. 2006, Kenneth L. Calvert University of Kentucky, USA All rights reserved Domain Name System CS 571 Fall 2006 2006, Kenneth L. Calvert University of Kentucky, USA All rights reserved DNS Specifications Domain Names Concepts and Facilities RFC 1034, November 1987 Introduction

More information

HTG XROADS NETWORKS. Network Appliance How To Guide: EdgeDNS. How To Guide

HTG XROADS NETWORKS. Network Appliance How To Guide: EdgeDNS. How To Guide HTG X XROADS NETWORKS Network Appliance How To Guide: EdgeDNS How To Guide V 3. 2 E D G E N E T W O R K A P P L I A N C E How To Guide EdgeDNS XRoads Networks 17165 Von Karman Suite 112 888-9-XROADS V

More information

IPv6 for AT&T Broadband

IPv6 for AT&T Broadband IPv6 for AT&T Broadband Chris Chase, AT&T Labs Sept 15, 2011 AT&T Broadband ~15 million subscribers Legacy DSL, PPP subscribers, ATM aggregation Not many CPE IPv6 capable Customer owned, unmanaged CPE

More information

Candidates should attempt FOUR questions. All questions carry 25 marks.

Candidates should attempt FOUR questions. All questions carry 25 marks. UNIVERSITY OF ABERDEEN Exam 2010 Degree Examination in ES 3567 Communications Engineering 1B Xday X Notes: 9.00 a.m. 12 Noon (i) CANDIDATES ARE PERMITTED TO USE APPROVED CALCULATORS (II) CANDIDATES ARE

More information

Simulation of DNS(Domain Name System) Using SimLib

Simulation of DNS(Domain Name System) Using SimLib Simulation of DNS(Domain Name System) Using SimLib Submitted by Prem Tamang Submitted to Dr. Lawrence J. Osborne Table of Contents 1. Introduction 3 2. Motivation and Challenges. 5 3. Assumptions 5 4.

More information

Implementing, Managing and Maintaining a Microsoft Windows Server 2003 Network Infrastructure: Network Services Course No.

Implementing, Managing and Maintaining a Microsoft Windows Server 2003 Network Infrastructure: Network Services Course No. COURSE OVERVIEW This five-day instructor-led course provides students with the knowledge and skills to implement and manage a Microsoft Windows Server 2003 network The course is intended for systems administrator

More information

USHA. Email Notification Setting. User Manual

USHA. Email Notification Setting. User Manual USHA Email Notification Setting User Manual 1 Email Notification configuration... 3 1.1 Mail Server Table... 3 1.1.1 Mail Server... 3 1.1.2 User Account and User Password... 4 1.1.3 Sender s Email Address...

More information

Harness Your Internet Activity!

Harness Your Internet Activity! Harness Your Internet Activity Random Subdomain Attacks Plaguing the Internet Agenda Brief Intro Covered at last OARC Attack overview Latest data Progress on open dns proxies in home gateways Impact of

More information

- Domain Name System -

- Domain Name System - 1 Name Resolution - Domain Name System - Name resolution systems provide the translation between alphanumeric names and numerical addresses, alleviating the need for users and administrators to memorize

More information

BGP and Traffic Engineering with Akamai. Caglar Dabanoglu Akamai Technologies AfPIF 2015, Maputo, August 25th

BGP and Traffic Engineering with Akamai. Caglar Dabanoglu Akamai Technologies AfPIF 2015, Maputo, August 25th BGP and Traffic Engineering with Akamai Caglar Dabanoglu Akamai Technologies AfPIF 2015, Maputo, August 25th AGENDA Akamai Intelligent Platform Peering with Akamai Traffic Engineering Summary Q&A The Akamai

More information

Why contribute passive DNS data to ISC?

Why contribute passive DNS data to ISC? Join The Global Passive DNS (pdns) Network Today & Gain Effective Tools To Fight Against Cyber Crime Why contribute passive DNS data to ISC? ISC - the Public Benefit Company that works to sustain the spirit

More information

E2BN Direct - Network Services for Schools and Academies

E2BN Direct - Network Services for Schools and Academies E2BN Direct - Network Services for Schools and Academies E2BN Direct provides internet, web filtering and other services directly to Schools and Academies. We use proven products backed by dedicated support,

More information

ABSTRACT Acknowledgments List of Abbreviations Contents ABSTRACT 3 Acknowledgments 5 List of Abbreviations 7 List of Figures 15 List of Tables 23 1 Introduction 25 2 Motivation and background 29 3 Overview

More information

DNSSEC and DNS Proxying

DNSSEC and DNS Proxying DNSSEC and DNS Proxying DNS is hard at scale when you are a huge target 2 CloudFlare DNS is big 3 CloudFlare DNS is fast 4 CloudFlare DNS is always under attack 5 CloudFlare A secure reverse proxy for

More information

CA Nimsoft Monitor. Probe Guide for DNS Response Monitoring. dns_response v1.6 series

CA Nimsoft Monitor. Probe Guide for DNS Response Monitoring. dns_response v1.6 series CA Nimsoft Monitor Probe Guide for DNS Response Monitoring dns_response v1.6 series Legal Notices This online help system (the "System") is for your informational purposes only and is subject to change

More information

How to set up the Integrated DNS Server for Inbound Load Balancing

How to set up the Integrated DNS Server for Inbound Load Balancing How to set up the Integrated DNS Server for Introduction Getting Started Peplink Balance has a built-in DNS server for inbound link load balancing. You can delegate a domain s NS/SOA records, e.g. www.mycompany.com,

More information

School Connectivity Services Updata

School Connectivity Services Updata Schools Connectivity Services Updata is a leading provider of IT services, support and solutions for schools, working in partnership with over 3000 schools, academies and local authorities to support their

More information

Introduction to the Domain Name System

Introduction to the Domain Name System CHAPTER 14 The Domain Name System (DNS) handles the growing number of Internet users. DNS translates names, such as www.cisco.com, into IP addresses, such as 192.168.40.0 (or the more extended IPv6 addresses),

More information

Fig 1A portion of the Internet domain name space.

Fig 1A portion of the Internet domain name space. 1 Asst. Lecturar ahmed S. Kareem DNS The Domain Name System Although programs theoretically could refer to hosts, mailboxes, and other resources by their network (e.g., IP) addresses, these addresses are

More information

Distributed Denial of Service Attacks

Distributed Denial of Service Attacks Distributed Denial of Service Attacks Steve Crocker Chair, SSAC June 25, 2007 San Juan, Puerto Rico 1 Agenda Types of Attacks DDoS attacks Amplified DDoS attacks - 2006 Estonia - May 2007 What do Do 2

More information

Guardian Digital Secure Mail Suite Quick Start Guide

Guardian Digital Secure Mail Suite Quick Start Guide Guardian Digital Secure Mail Suite Quick Start Guide Copyright c 2004 Guardian Digital, Inc. Contents 1 Introduction 1 2 Contacting Guardian Digital 2 3 Purpose of This Document 3 3.1 Terminology...............................

More information

The Domain Name System (DNS)

The Domain Name System (DNS) The Domain Name System (DNS) Columbus, OH 43210 Jain@CIS.Ohio-State.Edu http://www.cis.ohio-state.edu/~jain/ 24-1 Overview Naming hierarchy hierarchy Name resolution Other information in name servers 24-2

More information

Current Counter-measures and Responses by the Domain Name System Community

Current Counter-measures and Responses by the Domain Name System Community Current Counter-measures and Responses by the Domain Name System Community Paul Twomey President and CEO 22 April 2007 APEC-OECD Malware Workshop Manila, The Philippines 1 What I want to do today in 15

More information

Citrix NetScaler Global Server Load Balancing Primer:

Citrix NetScaler Global Server Load Balancing Primer: Citrix NetScaler Global Server Load Balancing Primer: Theory and Implementation www.citrix.com Background...3 DNS Overview...3 How DNS level GSLB works...4 Basic NetScaler GSLB Configuration...8 Accepting

More information

DNS. The Root Name Servers. DNS Hierarchy. Computer System Security and Management SMD139. Root name server. .se name server. .

DNS. The Root Name Servers. DNS Hierarchy. Computer System Security and Management SMD139. Root name server. .se name server. . Computer System Security and Management SMD139 Lecture 5: Domain Name System Peter A. Jonsson DNS Translation of Hostnames to IP addresses Hierarchical distributed database DNS Hierarchy The Root Name

More information

Chapter 25 Domain Name System. 25.1 Copyright The McGraw-Hill Companies, Inc. Permission required for reproduction or display.

Chapter 25 Domain Name System. 25.1 Copyright The McGraw-Hill Companies, Inc. Permission required for reproduction or display. Chapter 25 Domain Name System 25.1 Copyright The McGraw-Hill Companies, Inc. Permission required for reproduction or display. 25.2 Figure 25.1 Example of using the DNS service 25-1 NAME SPACE To be unambiguous,

More information

BIG IP Global Traffic Manager (GTM) v.11

BIG IP Global Traffic Manager (GTM) v.11 BIG IP Global Traffic Manager (GTM) v.11 This two day course gives networking professionals a functional understanding of the BIG IP GTM system as it is commonly used. The course covers installation, configuration,

More information

Akamai CDN, IPv6 and DNS security. Christian Kaufmann Akamai Technologies DENOG 5 14 th November 2013

Akamai CDN, IPv6 and DNS security. Christian Kaufmann Akamai Technologies DENOG 5 14 th November 2013 Akamai CDN, IPv6 and DNS security Christian Kaufmann Akamai Technologies DENOG 5 14 th November 2013 Agenda Akamai Introduction Who s Akamai? Intelligent Platform & Traffic Snapshot Basic Technology Akamai

More information

Active Directory. By: Kishor Datar 10/25/2007

Active Directory. By: Kishor Datar 10/25/2007 Active Directory By: Kishor Datar 10/25/2007 What is a directory service? Directory Collection of related objects Files, Printers, Fax servers etc. Directory Service Information needed to use and manage

More information

Content Delivery Networks

Content Delivery Networks Content Delivery Networks Terena 2000 ftp://ftpeng.cisco.com/sgai/t2000cdn.pdf Silvano Gai Cisco Systems, USA Politecnico di Torino, IT sgai@cisco.com Terena 2000 1 Agenda What are Content Delivery Networks?

More information

Application-layer protocols

Application-layer protocols Application layer Goals: Conceptual aspects of network application protocols Client server paradigm Service models Learn about protocols by examining popular application-level protocols HTTP DNS Application-layer

More information

GLOBAL SERVER LOAD BALANCING WITH SERVERIRON

GLOBAL SERVER LOAD BALANCING WITH SERVERIRON APPLICATION NOTE GLOBAL SERVER LOAD BALANCING WITH SERVERIRON Growing Global Simply by connecting to the Internet, local businesses transform themselves into global ebusiness enterprises that span the

More information

THE MASTER LIST OF DNS TERMINOLOGY. v 2.0

THE MASTER LIST OF DNS TERMINOLOGY. v 2.0 THE MASTER LIST OF DNS TERMINOLOGY v 2.0 DNS can be hard to understand and if you re unfamiliar with the terminology, learning more about DNS can seem as daunting as learning a new language. To help people

More information

Best Practices in DNS Anycast Service-Provision Architecture. Version 1.1 March 2006 Bill Woodcock Gaurab Raj Upadhaya Packet Clearing House

Best Practices in DNS Anycast Service-Provision Architecture. Version 1.1 March 2006 Bill Woodcock Gaurab Raj Upadhaya Packet Clearing House Best Practices in DNS Service-Provision Architecture Version 1.1 March 2006 Bill Woodcock Gaurab Raj Upadhaya Packet Clearing House It s all Large ISPs have been running production anycast DNS for more

More information

Week 3 / Paper 2. Bernhard Ager, Wolfgang Mühlbauer, Georgios Smaragdakis, Steve Uhlig ACM IMC 2010.

Week 3 / Paper 2. Bernhard Ager, Wolfgang Mühlbauer, Georgios Smaragdakis, Steve Uhlig ACM IMC 2010. Week 3 / Paper 2 Comparing DNS Resolvers in the Wild Bernhard Ager, Wolfgang Mühlbauer, Georgios Smaragdakis, Steve Uhlig ACM IMC 2010. Main point How does ISP DNS compare with Google DNS and OpenDNS?

More information

Domain Name System (DNS) Services

Domain Name System (DNS) Services 12 Domain Name System (DNS) Services Contents Overview..................................................... 12-3 Host and Domain Names.................................... 12-3 Host Tables...............................................

More information

DNS and BIND. David White

DNS and BIND. David White DNS and BIND David White DNS: Backbone of the Internet Translates Domains into unique IP Addresses i.e. developcents.com = 66.228.59.103 Distributed Database of Host Information Works seamlessly behind

More information

Send document comments to nexus7k-docfeedback@cisco.com.

Send document comments to nexus7k-docfeedback@cisco.com. CHAPTER 4 This chapter describes how to configure the Domain Name Server (DNS) client. This chapter includes the following sections: Information About DNS Clients, page 4-1 Licensing Requirements for DNS

More information

How to Add Domains and DNS Records

How to Add Domains and DNS Records How to Add Domains and DNS Records Configure the Barracuda NextGen X-Series Firewall to be the authoritative DNS server for your domains or subdomains to take advantage of Split DNS or dead link detection.

More information

John S. Otto Fabián E. Bustamante

John S. Otto Fabián E. Bustamante John S. Otto Fabián E. Bustamante Northwestern, EECS AIMS-4 CAIDA, SDSC, San Diego, CA Feb 10, 2012 http://aqualab.cs.northwestern.edu ! CDNs direct web clients to nearby content replicas! Several motivations

More information

Introduction to Network. Topics

Introduction to Network. Topics Introduction to Security Chapter 3 The Internet 1 Topics The Internet Addressing Client Routing 2 The Internet User s View Internet 3 National, International, and large regionalisps ISP ISP ISP The Internet

More information

THE UNIVERSITY OF AUCKLAND

THE UNIVERSITY OF AUCKLAND COMPSCI 742 THE UNIVERSITY OF AUCKLAND SECOND SEMESTER, 2008 Campus: City COMPUTER SCIENCE Data Communications and Networks (Time allowed: TWO hours) NOTE: Attempt all questions. Calculators are NOT permitted.

More information

Monitoring the DNS. Gustavo Lozano Event Name XX XXXX 2015

Monitoring the DNS. Gustavo Lozano Event Name XX XXXX 2015 Monitoring the DNS Gustavo Lozano Event Name XX XXXX 2015 Agenda 1 2 3 Components of the DNS Monitoring gtlds Monitoring other components of the DNS 4 5 Monitoring system Conclusion 2 Components of the

More information

DNS Architecture Case Study: Resiliency and Disaster Recovery

DNS Architecture Case Study: Resiliency and Disaster Recovery DNS Architecture Case Study: Resiliency and Disaster Recovery Cricket Liu VP, Architecture Infoblox Company Background Large U.S.-based company, Company Co. (company.com) Three categories of sites Headquarters

More information

Configuring your network settings to use Google Public DNS

Configuring your network settings to use Google Public DNS Configuring your network settings to use Google Public DNS When you use Google Public DNS, you are changing your DNS "switchboard" operator from your ISP to Google Public DNS. In most cases, the IP addresses

More information

THE MASTER LIST OF DNS TERMINOLOGY. First Edition

THE MASTER LIST OF DNS TERMINOLOGY. First Edition THE MASTER LIST OF DNS TERMINOLOGY First Edition DNS can be hard to understand and if you re unfamiliar with the terminology, learning more about DNS can seem as daunting as learning a new language. To

More information

You can specify IPv4 and IPv6 addresses while performing various tasks in this feature. The resource

You can specify IPv4 and IPv6 addresses while performing various tasks in this feature. The resource The feature enables the configuration of a Virtual Private Network (VPN) routing and forwarding instance (VRF) table so that the domain name system (DNS) can forward queries to name servers using the VRF

More information

DNS Amplification Attacks as a DDoS Tool and Mitigation Techniques

DNS Amplification Attacks as a DDoS Tool and Mitigation Techniques DNS Amplification Attacks as a DDoS Tool and Mitigation Techniques Klaus Steding-Jessen jessen@cert.br! Computer Emergency Response Team Brazil - CERT.br Network Information Center Brazil - NIC.br Brazilian

More information

How To Guide Edge Network Appliance How To Guide:

How To Guide Edge Network Appliance How To Guide: How To Guide Edge Network Appliance How To Guide: ActiveDNS v 4.01 Edge Network Appliance How To Guide: ActiveDNS 2007 XRoads Networks 17165 Von Karman, Suite 112 888-9-XROADS v 4.01 updated 09/11/07 Table

More information

Interconnecting IPv6 Domains Using Tunnels

Interconnecting IPv6 Domains Using Tunnels Interconnecting Domains Using Tunnels Version History Version Number Date Notes 1 30 July 2002 This document was created. 2 19 May 2003 Updated the related documents section. This document describes how

More information

Wireshark Lab: DNS v6.01

Wireshark Lab: DNS v6.01 Wireshark Lab: DNS v6.01 Supplement to Computer Networking: A Top-Down Approach, 6 th ed., J.F. Kurose and K.W. Ross Tell me and I forget. Show me and I remember. Involve me and I understand. Chinese proverb

More information

HTG XROADS NETWORKS. Network Appliance How To Guide: DNS Delegation. How To Guide

HTG XROADS NETWORKS. Network Appliance How To Guide: DNS Delegation. How To Guide HTG X XROADS NETWORKS Network Appliance How To Guide: DNS Delegation How To Guide DNS Delegation (The Simple Redundancy Solution) The key requirement when performing DNS based network redundancy and load

More information

SSAC Advisory SAC008 DNS Distributed Denial of Service (DDoS) Attacks

SSAC Advisory SAC008 DNS Distributed Denial of Service (DDoS) Attacks SSAC Advisory SAC008 DNS Distributed Denial of Service (DDoS) Attacks A Report from the ICANN Security and Stability Advisory Committee (SSAC) March 2006 Page 1 of 16 Executive Summary In early February

More information

CA Unified Infrastructure Management

CA Unified Infrastructure Management CA Unified Infrastructure Management Probe Guide for IIS Server Monitoring iis v1.7 series Copyright Notice This online help system (the "System") is for your informational purposes only and is subject

More information

USG40HE Content Filter Customization

USG40HE Content Filter Customization USG40HE Content Filter Customization This guide is designed to help with the setup of the ZyWALL s content filtering feature. Supported Devices USG40HE Firmware version 4.10(AALA.0) or later Overview Content

More information

Configuring Basic Settings

Configuring Basic Settings CHAPTER 9 This chapter describes how to configure basic settings on your ASASM that are typically required for a functioning configuration. This chapter includes the following sections: Configuring the

More information

Application Note. SIP Domain Management

Application Note. SIP Domain Management Application Note SIP Domain Management 28 March 2008 Table of Contents 1 WHAT IS A SIP DOMAIN?... 1 2 LOCAL SIP DOMAIN... 2 3 OTHER SIP DOMAIN... 3 4 DNS CONSIDERATIONS... 4 5 USING A PUBLIC DNS... 5 6

More information

Classifying DNS Heavy User Traffic by using Hierarchical Aggregate Entropy. 2012/3/5 Keisuke Ishibashi, Kazumichi Sato NTT Service Integration Labs

Classifying DNS Heavy User Traffic by using Hierarchical Aggregate Entropy. 2012/3/5 Keisuke Ishibashi, Kazumichi Sato NTT Service Integration Labs Classifying DNS Heavy User Traffic by using Hierarchical Aggregate Entropy 2012/3/5 Keisuke Ishibashi, Kazumichi Sato NTT Service Integration Labs Motivation Network resources are consumed by a small number

More information

Date 07/05/2014 10:20:22. CENTREL Solutions. Author. Version 1.02. Product XIA Configuration Server [6.0.0.25996]

Date 07/05/2014 10:20:22. CENTREL Solutions. Author. Version 1.02. Product XIA Configuration Server [6.0.0.25996] Date 07/05/2014 10:20:22 Author CENTREL Solutions Version 1.02 Product XIA Configuration Server [6.0.0.25996] Page 3 of 28 Description Item Name Microsoft Windows Server 2012 R2 Datacenter DEMO-DHCP-12R2

More information

Applications and Services. DNS (Domain Name System)

Applications and Services. DNS (Domain Name System) Applications and Services DNS (Domain Name Service) File Transfer Protocol (FTP) Simple Mail Transfer Protocol (SMTP) Malathi Veeraraghavan Distributed database used to: DNS (Domain Name System) map between

More information

Moonv6 Test Suite DRAFT

Moonv6 Test Suite DRAFT Moonv6 Test Suite DHCP Interoperability Test Suite DRAFT Technical Document Revision 0.1 IPv6 Consortium 121 Technology Drive, Suite 2 InterOperability Laboratory Durham, NH 03824-3525 Research Computing

More information

DNS Cache Poisoning Vulnerability Explanation and Remedies Viareggio, Italy October 2008

DNS Cache Poisoning Vulnerability Explanation and Remedies Viareggio, Italy October 2008 DNS Cache Poisoning Vulnerability Explanation and Remedies Viareggio, Italy October 2008 Kim Davies Internet Assigned Numbers Authority Internet Corporation for Assigned Names & Numbers Agenda How do you

More information

Hostnames. HOSTS.TXT was a bottleneck. Once there was HOSTS.TXT. CSCE515 Computer Network Programming. Hierarchical Organization of DNS

Hostnames. HOSTS.TXT was a bottleneck. Once there was HOSTS.TXT. CSCE515 Computer Network Programming. Hierarchical Organization of DNS Hostnames CSCE 515: Computer Network Programming ------ Address Conversion Function and DNS RFC 1034, RFC 1035 Wenyuan Xu http://www.cse..edu/~wyxu/ce515f07.html Department of Computer Science and Engineering

More information

CA Nimsoft Monitor. Probe Guide for IIS Server Monitoring. iis v1.5 series

CA Nimsoft Monitor. Probe Guide for IIS Server Monitoring. iis v1.5 series CA Nimsoft Monitor Probe Guide for IIS Server Monitoring iis v1.5 series Legal Notices Copyright 2013, CA. All rights reserved. Warranty The material contained in this document is provided "as is," and

More information

DNS for Internet Firewalls

DNS for Internet Firewalls DNS for s 1 DNS Security Do not rely on DNS names to make security-related decisions DNS can be spoofed Use IP addresses whenever possible Note that IP addresses can also be spoofed It s just a little

More information

Lotus Domino 8 Monitoring and Maintenance

Lotus Domino 8 Monitoring and Maintenance Lotus Domino 8 Monitoring and Maintenance Course Title Course Code Lotus Domino 8 Monitoring and Maintenance DSMM8 Duration 02 days Course Fee Call to Request Instructor Certified Lotus Instructor or Certified

More information

No. Time Source Destination Protocol Info 1 0.000000 192.168.1.28 192.168.1.2 DNS Standard query A weather.noaa.gov

No. Time Source Destination Protocol Info 1 0.000000 192.168.1.28 192.168.1.2 DNS Standard query A weather.noaa.gov /tmp/dump/dump02_arp_dns-weather_syn_fin complete-session - Ethereal Page 1 1 0.000000 192.168.1.28 192.168.1.2 DNS Standard query A weather.noaa.gov Frame 1 (76 bytes on wire, 76 bytes captured) Arrival

More information

Tools for penetration tests 1. Carlo U. Nicola, HT FHNW With extracts from documents of : Google; Wireshark; nmap; Nessus.

Tools for penetration tests 1. Carlo U. Nicola, HT FHNW With extracts from documents of : Google; Wireshark; nmap; Nessus. Tools for penetration tests 1 Carlo U. Nicola, HT FHNW With extracts from documents of : Google; Wireshark; nmap; Nessus. What is a penetration test? Goals: 1. Analysis of an IT-environment and search

More information

Owner of the content within this article is www.msexchange.org Written by Marc Grote www.it-training-grote.de

Owner of the content within this article is www.msexchange.org Written by Marc Grote www.it-training-grote.de Owner of the content within this article is www.msexchange.org Written by Marc Grote www.it-training-grote.de Exchange 2003 SMTP Namespace Sharing Written by Marc Grote - mailto:grotem@it-training-grote.de

More information

TCP/IP Network Connectivity and ION Meters

TCP/IP Network Connectivity and ION Meters 70072-0170-04 TECHNICAL NOTE 12/2007 TCP/IP Network Connectivity and ION Meters This technical note describes how PowerLogic ION8800, ION8600, ION7550 and ION7650 meters are able to interact with several

More information

Using the Domain Name System for System Break-ins

Using the Domain Name System for System Break-ins Using the Domain Name System for System Break-ins Steven M. Bellovin Presented by: Thomas Repantis trep@cs.ucr.edu CS255-Computer Security, Winter 2004 p.1/37 Overview Using DNS to spoof a host s name

More information

Web Application Hosting Cloud Architecture

Web Application Hosting Cloud Architecture Web Application Hosting Cloud Architecture Executive Overview This paper describes vendor neutral best practices for hosting web applications using cloud computing. The architectural elements described

More information

Network Layers. CSC358 - Introduction to Computer Networks

Network Layers. CSC358 - Introduction to Computer Networks Network Layers Goal Understand how application processes set up a connection and exchange messages. Understand how addresses are determined Data Exchange Between Application Processes TCP Connection-Setup

More information

MOC 6419: Configuring, Managing, and Maintaining Windows Server 2008

MOC 6419: Configuring, Managing, and Maintaining Windows Server 2008 1 of 6 1/6/2010 3:23 PM MOC 6419: Configuring, Managing, and Maintaining Windows Server 2008 This five-day instructor-led course combines five days worth of instructor-led training content from the Network

More information

Create a New Account Contents

Create a New Account Contents Create a New Account Contents Goals...2 Create a New Domain Account...2 Domain Information...3 Package...4 Settings...5 Reseller Settings...6 DNS Settings...6 Mail Routing Settings...7 Conclusion...7 1

More information