Data Cached in Client 1 Data in Broadcast Program Database in Server. Hot. Cold. e broadcast program a a Queries in Client 1: i, a, c, e, d, g, i, j

Size: px
Start display at page:

Download "Data Cached in Client 1 Data in Broadcast Program Database in Server. Hot. Cold. e broadcast program a a Queries in Client 1: i, a, c, e, d, g, i, j"

Transcription

1 DynamicDataDeliveryinWireless QinglongHu1,DikLunLee1,andWang-ChienLee2 CommunicationEnvironments 2GTELaboratoriesIncorporated,40SylvanRoad,Waltham,MA02254, 1UniversityofScienceandTechnology,ClearWaterBay,HongKong, Abstract.Informationbroadcasting,cachingoffrequentlyaccesseddata, andpull-baseddatadeliveryarecommonlyusedtechniquestoreduce dataaccesstimeofwirelessinformationservices.mostofthestudiesin theliteraturefocusedeitheronindividualtechniqueoracombination aredisseminatedthroughvariousstoragemediumsaccordingtothedynamicallycollecteddataaccesspatterns.performanceevaluationofthe modelisconductedbysimulationstudies. ofthemwithsomerestrictiveassumptions.inthispaper,weproposea inanintegratedmanner.aparticularfeatureofourmodelisthatdata dynamicdatadeliverymodelwherethesethreetechniquesworktogether 1Introduction isaccesstimewhichistheaveragetimefromamobilecomputerrequestsa dataitemuntilthedataisreceived.toreduceaccesstimeinamobileenvironment,threemajorclassesoftechniques,namely,caching,push-basedinformation Acriterionoftenusedtoevaluatethedataaccesseciencyofamobilesystem quentlyaccesseddatashouldbecachedintheclient,thelessfrequentlyaccessed cientinformationretrievalinawirelesscommunicationenvironment,mostfre- [IVB94,AAFZ95,AFZ97,SRB97]. broadcastandpulled-baseddatadeliveryhavebeeninvestigatedintheliterature datasubsetistemporarilystoredontheair(throughbroadcastchannels),and therestofthedatacanbepulledfromtheserverviaexplicitclientrequests. Toreducethedisadvantagesoftheabovethreetechniquesandtoachieveef- Datacachingandpush-baseddatadisseminationcanalleviatepull-basedrequestconsiderably,asmostfrequentlyaccessdatacanberetrievedeitherfrom broadcastchannelarebothlimitedresourcesinamobileenvironmentandtheir eciencydependsheavilyontheprecisenessofthedataset,datadistribution shouldbemanagedcarefullytoachievehighutilizationoftheresources.the keypointistodynamicallydecidewhichsubsetofdatashouldbecached,which subsetshouldbebroadcastontheair. broadcast,andpull-basedtechniquestoachieveecientinformationdisseminationbycontinuouslyadjustingtheclientcacheandthebroadcastscheduleto matchtheaccesspatterns.inmobilesystems,theclientscaneithermonitorthe theclientcacheorthebroadcastchannel.sincetheclientcacheandwireless Thispaperinvestigatesthemethodstointegratedatacaching,information

2 broadcastchannelforthearrivalofthedesireddataorissueapullrequesttothe serverforthedataandrespectively,theservercaneitherbroadcastdataaccordingtobroadcastscheduleorbroadcastpulleddatainstead.theclientsmanage thecacheviatraditionalpagereplacementpolicy(i.e.,lru).thehotspotof simulationstudyshowsthattheaccesspatternscanbeaccuratelyestimatedand theuncacheddatacanbeobtainedbyclientmonitoringthecachemisses.our purepull-baseddataservice. inwhichtheserverknowstheexactclientaccesspatternsandisbetterthan basedonittheservercandynamicallyconstructbroadcastprogramtooptimize systemperformance.theresultingperformanceisconsistentwiththeidealcase accesspatterns,whichisdierentfromtheworkin[aafz95].in[aafz95], Acharyaetalassumedthatthedataaccesspatternsareknownandremain unchange.therefore,axedbroadcastschedule,whichisorganizedaccording totheuseraccesspatterns,isrepeatedineverybroadcastcycle.thecached Ourworkisbasedontheassumptionthattheserverdoesnotknowthedata datashouldbetailoredaccordingtothebroadcastscheduleinadditiontothe LRUpolicy:datasetwhoselocalprobabilityofaccessissignicantlygreater thantheirbroadcastfrequency.bytakingthosetwofactorsintoconsideration, withotherschemes.[srb97]investigatedthedynamicadjustmentofthehotspotforthebroadcastprogram.however,datacachingandmultipledisksof dierentsizesandspeedsforthebroadcastprogramwerenotconsidered. showedthathixissuperiorforcachinginthebroadcastsystemscompared Achayaetal[AAFZ95]developedapagereplacementpolicycalledHIXand theevaluationofthedatadeliverysystem.section4presentsexperimentsand disseminationmethods.section3describesthesimulationmodeldevelopedfor theresultsderivedfromthismodel.finally,section5concludesthepaper. Therestofthepaperisorganizedasfollows.Section2proposesdynamicdata 2DynamicDataBroadcastScheduling downlinkchannelforalltheclientsandanshareduplinkchannelforrequests) Oursystemmodelisbasedonthesatellitebroadcastscenario(i.e.,oneshared thepullrequeststotheserver.onceaclientisallocatedtheuplinkchannel,itwill andasymmetriccommunication(i.e.,thedownlinkchannelfromservertoclients hasfarmorecapacitythanthereversedirection). occupythechanneluntiltheserviceends.uponreceiptofthepullrequest,the serverbroadcasttheanswertotheclient.pull-baseddatadeliverycanguarantee allrequesteddatatobenallyobtained.however,bandwidthmaybewasted Forpull-baseddatadelivery,clientscompetefortheuplinkchanneltosend clientpopulation,thedownlinkchannelcaneasilygetcongestedwithseparate withclientrequests. pulledpages.inaddition,theservermustbeinterruptedcontinuouslytodeal becausethepulledpagesareusuallyofinteresttoonlyoneclient1.withalarge 1Although,thisproblemismitigatedtosomeextent,bythefactorthatanypage whichispulledbyoneclientcanbeaccessedonthebroadcastchannelbyanyother client.

3 time.therefore,clientaccesspatternsmustbecollecteddynamically.basedon encesarethattheclientaccesspatternsareunknowninitiallyorcanchangewith theaccesspatterns,theserverconstructsthebroadcastprogram. Themodelextendsthebroadcastdiskmodelproposedin[AFZ97].Thedier- handlingclientpullrequestswhichisimitatedbyanitequeueintheserver. ThequeueisservicedinaFIFOfashionandtheservicerateisdeterminedby thepullbandwidthofthesystem.wheneverthequeueisfull,allclientsare blockedandpullrequestswillbeacceptedbytheserveronlywhenthequeue Similarto[AFZ97],weassumethattheserverhaslimitedthroughputin request. hasspaceavailableagain.therequestforapagethatisalreadyinthequeueis droppedbecausetheprocessingoftheearliermessagewillalsosatisfythenew thecycleversion),isbroadcastatthebeginningofeachcycle.eachactiveclient datadelivery,thecompletebroadcastschedule(i.e.,alistofdataidentiersand keepsonmonitoringtheschedulesandretrievesthebroadcastscheduleintoits cache2.basedontheschedules,theclientcanchoosetheoptimaldatadelivery Inordertorealizedynamicalselectionbetweenpush-basedandpull-based method.thatis,ifthedesireddataitemswillappearwithinacertainbroadcast slots,thentheclientkeepsonmonitoringthebroadcastchannel;otherwise,the clientissuesapullrequesttotheserverandmonitorsthechannelforthearrival ofthedataitem.theserverdeterminesthenextdataitemforbroadcastfrom eitherthebroadcastprogramorthepullrequestswaitinginthequeue.weadopt isbroadcast,acoinweightedbythepercentageofpullbandwidthistossedand dependingontheoutcome,eithertherequestedpageattheheadofthequeueis broadcastortheregularbroadcastprogramcontinues.ifthequeueisempty(i.e., arandomslotallocationschemeproposedin[afz97],inwhichbeforeeverypage frombroadcastprogramdespitethatitistheturnforpulledpages.therefore, nopullrequestsfromtheclients),thenextcandidateforbroadcastingisapage thepercentageofbandwidthallocatedforpullrequestsisanupperbound. [EPW93]inthemodel.InLRU-Kscheme,thecachemanagementkeepstrack thecacheddata.toimprovethehitratioofthecache,weadoptlru-kscheme ofthetimesoflastkreferencestopopulardataitemsandselectionisbasedon accordingtodatacachedandpulledrequests,itisthesecondhotspotbesides Toecientlyutilizethebroadcastchannel,thebroadcastscheduleistailored LRU-Kpolicewhendynamicbroadcastprogramsareusedinthesystem. thepastkhitshistory.in[hll98]weshowthathixpolicydoesworsethan result,issuessuchascacheinvalidationwhichiscloselyrelatedtodataupdates, sumingthattherearenoupdateseitherbytheserverorattheclients.asa arenotdiscussedinthepaper.readersinterestedinthissubjectmayreferto [BI94,HL98,WYC96]. Tobeconcentratedontheissuesweliketoaddressinthispaper,weas- 2Noticethatmonitoringtheschedulescanbequitepowerconsumptionandisnot exible,inalaterstudy[hll98],weintroduceindexingmechanismstoavoidclient keepingonmonitoringthechannelfortheschedules.

4 clientswithoutanyperformancedegradation,itiswellsuitedfordisseminating datatomanyclients.thebroadcastchannelisalinearmedium:aclientwaitsfor Sincebroadcastdatacanbeaccessedconcurrentlybyanarbitrarynumberof 2.1BroadcastScheduling ofdatabroadcast.tomaintainshortaccesslatency,theamountofbroadcast notcachedshouldbeprovidedforbroadcasting). dataneedstobesmall(i.e.,onlythemostfrequentlyaccesseddatawhichare thedesireddataonthechannel.hence,theaccesslatencydependsonthevolume gram)needstobeconstructed.suchthattheserverperiodicallybroadcastsdata accordingtothisbroadcastschedule.asimplemethodcalledatbroadcastisto broadcasteachdataonlyonceineachcycle.notethattheclientsaccesspatterns areusuallyskewed(i.e.,somedataareaccessedmorefrequentlywhiletheothersarelessfrequentlyaccessed),anotherschedulingmethodfordatabroadcast calledbroadcastdiskswasproposedin[aafz95].forbroadcastdisks,theserver dividestheleintogroupsofdataitemswithdierentbroadcastfrequenciesto theclients,theexpectedwaittimeforadataitemonthatdiskistheshortest. imitatemultipledisksspinningatdierentspeeds.thefastestdiskisclosestto forclientstoaccessthem.thebroadcastfrequenciesofdisksareinproportion Thedataitemsontheslowestdiskarefarthestsinceitwilltakealongertime Oncethedatasetforbroadcastisdetermined,abroadcastschedule(pro- forasymmetriccommunicationenvironmentswheretheclientaccesspatterns andretrievedesireddataitemsfromthechannel.broadcastdisksareecient programoftheserver.accordingtotheprogram,clientsmonitorthebroadcast tem.toretrievedatafrombroadcastdisks,theclientsmustknowthebroadcast totheprobabilityofaccessinordertogaintheoptimalaccesstimeofthesys- areskewed.dataontheaircanfurtherbeorganizedasbroadcastdiskswith fastdiskscontainingfrequentlyaccesseddataandslowdisksforlessfrequently accesseddata. namicbroadcastprogramchangesonlythecontentoftheprogram,whileall adoptsbdiskorflatapproachisdenotedasbp.weassumethatthedy- disks(bdisk)andatbroadcast(flat),whiledatadeliverysystemwhich ofdisksisone.hereafter,thetermbroadcastprogramreferstobothbroadcast Flatbroadcastcanberegardedasaspecialbroadcastdiskswherethenumber otherparametersremainxed. allbroadcastbandwidthisdedicatedtopulledpages.onacachemiss,theclient patterninformationisrequiredbytheserverandtheserverworksinapassive thearrivalofthedesireddataitem.itiscalledpure-pull.obviouslynoaccess immediatelysendsapullrequestforthepagetotheserverandthenwaitsfor Aspecialscenarioappearswhenthebroadcastprogramisempty.Therefore, way.thisapproachissimplebuttheserverandtheuplinkchannelmaybeeasily 2.2CollectionofClientAccessPatterns congestedbyalotofseparatepullrequests. metricenvironments[jbea97,wyc96].torecordclientaccessinformation, Bitvectorisawidelyusedmethodindeliveringcompactinformationinasym- eachclientmaintainsabitvectorinwhicheachbitrepresentsadataiteminthe

5 Data Cached in Client 1 Data in Broadcast Program Database in Server Hot Cold i c e a b c d e f g a b c d e f g h i j b d a 7-pages, 3-disks c g e broadcast program a a currentbroadcastprogram(figure1).wheneveranewbroadcastprogramis c a f b Queries in Client 1: i, a, c, e, d, g, i, j cycleversionnumberandresetstheirbitvectorsforthecurrentcycle.every retrieved(i.e.,atthebeginningofabroadcastcycle),theclientsupdatethenew Fig.1.ExampleofaDataStorage. Cache Hit: i, c, e Air Hit: a, d, g Missing: j dataitemonthechannelismarkedtoindicatewhetheritisfromthebroadcast a b c d e f g Bit Vector: Pull Request: j programoritisfromthepullrequests.ifthereisaqueryansweredbybroadcast data(referredtoairhit),thenthecorrespondingbitisset.retrievingdatafrom The MFA value items a,d, g, j increases, while that all other remains unchanged. theclientcache(referredascachehit)andpulleddataonthebroadcastchannel foundfromboththecacheandthebroadcastprogram(calledairmiss),anexplicitpullrequestwillbeissuedbytheclientviatheuplinkchannel.thus,the havenoinuenceonthebitvector.however,whenthedesireddatacannotbe ofaclientallcanbeansweredeitherbycacheddataorbybroadcastdata,the clientaccessstatisticsispiggybackedatthesametime.notethatiftherequests itemsinthebroadcastprogramwhichisquitesmall,theoverheadfortheuplink thebitvector.sincethelengthofthebitvectorequalstothenumberofdata clientaccesspatternwillnotbedeliveredtotheserver. bandwidthislow.afterthisbitvectorispiggybackedtotheserver,itscontent isresetanditisreadyforthecollectionoffutureclientaccess. ThepullrequestconsistsofthedesireddataitemID,thecycleversion,and correspondingtowhichbroadcastprogram.inaddition,eachdataitemisassociatedwithamostfrequentlyaccessed(mfa)counterwhichisinitiallyreset. Thecounterofadataitemisincrementedbyonewheneverthedataitemis pulledbyaclientorthebitcorrespondingtothatdataiteminthereceived Usingthereceivedcycleversionnumber,theserverknowsthebitvector bitvectorisset.asaresult,theservercollectsstatisticalinformationabout theclientaccesspatterns.thisstatisticshelpstheserverdynamicallybuildthe broadcastprogramaccordingtothesystemworkload.inthispaper,theserver selectspageswithhighmfavaluesforbroadcast.notethatacachehithas noinuenceonthemfavalue.therefore,onceadataitemiscachedinthe client,anyfurtherrequestsforitfromtheclientwillhavenocontributiontoits MFAvalueintheserver.Hence,cacheddatanormallyhavelowMFAvalues, thoughtheyaremostfrequentlyretrievedbytheclient.astheaccessstatistics isaccumulated,cacheddatawilleventuallynotbeselectedintothebroadcast programoronlybeplacedonaslowerspinningdisk.intuitively,onlythosedata queryforthemwillhavehighmfavalues.inthisway,theservercantailorthe itemswhichareaccessedbyalargenumberoftheclientsthatoccasionallymake broadcastprogramtotheneedofaparticularclient.

6 3SimulationModeling Server Broadcast Messager Broadcast Channel Database Broadcast program b d a c Push g e Pull MUX a a c a f b Message Manager Page Receiver Access Server Pattern Queue No Client 1 Client 2 Client n modeledbyaprocess3.theprocessconsistsoftwosub-processes,namelythe Yes Bit Double? Dropped Query Generator Query Generator Query Generator Vector QueryGeneratorandthePullManager.TheQueryGeneratorrunsacontinuous Inthispaper,asimulationmodelshowninFigure2isused.Eachclientis Fig.2.SimulationModel. No Yes Q Full? Blocked... loopthatrandomlyrequestsapageaccordingtoaspecicdistribution.itrst Thresh. Thresh. Thresh. Cache Filter Cache Filter Cache Filter checkstheclient'scachetoseewhetheravalidcopyofthedesireddataitems Request Request Request existsthere.iftheanswerisyes,animmediatereplyisdeliveredtotheuser. Uplink Channel processingaswellastherelativespeedsofthecpuandthebroadcastmedium. hasaxed-sizecacheandlru-kpagereplacementpolicyisadopted.after eachrequest,thequerygeneratorwaitsforaperiodofthinktimelongand thenmakesthenextrequest.thinktimeisaparameterusedtomodelworkload Otherwiseasub-processPullManagerisactivated.Weassumethattheclient Afteractivated,thePullManagerissuesapullrequesttotheserverforthe neededdataitemifthedataitemcannotbeobtainedfromthebroadcastdisks orthecostofaccessdataitemfromthedisksexceedsacut-ovaluespeciedby thesystem.ineithercase,theclientmonitorsthebroadcastchanneluntilthe desiredpagearrives.asingleprocess,calledpagereceiver,isusedtomodelall therequestedone,itisbroughtintotheclient'scache. clientsinthesystem.thepagereceivercontinuouslymonitorsthebroadcast seewhethertheincomingpageisthedesiredpageofthatclient.ifthepageis channelandbecomesactivewhenapagearrives.itcheckseachclientinturnto (withparameter)isfrequentlyusedtomodelskewedaccesspatternswhere isaparameternamedaccessskewcoecientandcanvaryfromzerotoone. usedin[knu81])andgaussian(usedin[srb97]aswell).thezipfdistribution Forthedataaccesspatterns,weusetwodierentdistributions:Zipf(also 3In[AFZ97],theentireclientpopulationwasmodeledbytwoclientprocesses,areal clientandavirtualclient.sincetheclient-servermodelisnotmemorylessbecauseof contentionontheuplinkchannelamongtheclients. datacaching,theinterleavingofpushandpullslots,andtheboundedserverqueue. Usingasingleclientprocess(i.e.,virtualclient)maynotexactlyreectthetrue

7 Thedistributionbecomesincreasingly"skewed"asincreases.TheGaussian distribution,normal(;),isusedtomodelthedynamicchangesintheaccess patternswiththecenterofhot-spotandthewidthofhot-spot.duringthe experiments,thevalueofcanvarytocreatetheeectofdynamicworkload theparametersusedtomodeltheresourceandthedataaccesspatternofeach client. andthevalueofreectstheskewofclientaccesspatterns.table1summarizes CacheSize PullReqSizeSizeofapullrequestinbytes ThinkTimeMeanthinktime(inseconds)betweenqueriesforeachclient ThreshFactorValueforclienttoselectbetweenbroadcastandon-demand Clientcachesize(indataitems) Table1.ClientParameterSetting. Zipfdistributionparameter Widthofhot-spot Manager.Aboundedqueueisbuiltfortheuplinkrequestmessage.Whenthe Theserverismodeledbytwoprocesses:BroadcastingManagerandMessage Centerofhot-spot arrivalrateofuplinkrequestsexceedtheservicerateoftheserver,(e.g.thequeue inthequeuearedropped.thequeueisservedinafifofashionandtheservice rateisdeterminedbytheparameter,pullbw.pullbwdeterminesthepercentage isfull),allrequestsareblocked.requestsforpagesthatarealreadyrequested cycle.onlythecontentofthebroadcastprogramistailoredtomeettheaccess broadcastprogramisdynamicallydeterminedbythebroadcastingmanagerper ofthebroadcastslotsallocatedforpagesexplicitlypulledbytheclients.the fromthebroadcastprogram.themessagemanagerreceivespullrequestsfrom thequeue.however,whenthequeueisempty,allbroadcastpagesareselected theclients.fromthesepullmessages,clientaccesspatternscanbeobtained broadcastingpageisselectedalternatelyfromeitherthebroadcastprogramor patternscollectedfromtheclientsviatheuplinkchannel.inthismodel,thenext fromtheserver. DataItemSizeSizeofadataiteminbytes NumClient DatabaseSizeSizeofdatabaseindataitems DownlinkBWDownlinkfromtheservertotheclientchannelbandwidth Numberofclientsinacell Table2.SystemParameterSetting. UplinkBW PullPercentagePercentageofbroadcastslotsallocatedforpullrequests QueneSize ProgramSizeNumberofdistinctdataitemsinbroadcastprogram DiskNum Uplinkfromtheclienttotheserverchannelbandwidth DiskSizei DiskFreqi Sizeofdownlinkchannelqueue(indataietms) NumberofdisksforBDISK Sizeofdiski(indataitems) castprogramarelistedintable2.additionally,theparametersthatdescribe Parameterswhichdescribetheserverresourceandthestructureofthebroad- Relativebroadcastfrequencyofdiski thecongurationandthephysicalresourceinacellarealsoincludedinthe

8 table.weassumethatthereisauplinkchannelfromtheclientstotheserver andashareddownlinkchannel(alsocalledbroadcastchannel)forthereverse communication.thedatabaseismodeledasacollectionofdatabasesizedata itemsofdataitemsizebyteseach.therearenumchannelchannelsinacell. Fortheexclusivechannelallocationapproach,allNumChannelchannelswork ineitheron-demandmodeorbroadcastmode.forbpapproach,thechannel allocationmadebetweenbroadcastandon-demandmodesisxedsuchthatthe ratiobetweenthenumberofbroadcastchannelsandthenumberofon-demand channelsispullvspush.thecommunicationisasymmetric,i.e.,thebandwidth ofdownlinkfromtheservertotheclientdownlinkbwisgreaterthanthatof uplinkfromtheclienttotheserveruplinkbw. 4ExperimentsandResults Parameters DataItemSize DatabaseSize3000dataitemsQueueSize Table3.SystemParameterSetting. PullReqSize DownlinkBW 1000bytesNumClient 10bytesPullPecentage ValuesParameters 10;;2000 Values ThinkTime ProgramSize500dataitemsDiskSizei bpsUplinkBW 10secondsThreshFactorDiskSize1=100,DiskSize2=400 1%ofdownlinkbandwidth 100% 1=3 Inthissection,weinvestigatetheperformanceofsimulatedBPtogetherwith DiskNum CacheSize 300dataitems 50dataitems 2DiskFreqi randomlyselectedinthedatabase DiskFreq1=2,DiskFreq2= Pure-Pullapproach.Theprimaryperformancemetricistheaverageclientaccess time.thesimulationisimplementedusingcsim[sch92].table3denesthe systemparametersettingfortheexperiments.theservercontainsacollection ofself-identifyingdataitemsofequalsize.theclientaccesspatternsareskewed, (i.e.,eitherzipforgaussiandistribution)andtherequesteddatasetsamong Theserverdatabaseconsistsof3000dataitems(orpages),eachdataitemis theclientsarelargelyoverlapped.theclientpopulationvariesfrom10to bytesinsize.Thebroadcastrateis100Kbps,whiletheuplinkrateis1Kbps (1%ofthebroadcastrate).Theclientcachesizeis50pages.Theback-channel diskis400pages,withrelativespinspeedsbeing2and1,respectively. forbroadcast-diskwherethesizeofthefasterdiskis100pages,whiletheslower consecutivequeriesis10seconds.thenumberofdistinctpagesinthebroadcast program(forbothbdiskandflat)is500pages.two-diskbroadcastisused queuecanholdupto100distinctrequests.themeanthinktimebetweentwo sisonthedynamicworkloadenvironments.therearetwocircumstanceswhen dynamicworkloadcanoccur.therstcaseiswhentheclientsjointhesystem thesecondcase,anexistingclientchangesitsaccesshot-spotwithtime.obviously,dynamicworkloadonlyaectstheclientcacheforthepurepullapproach. withanemptycacheandtheserverdoesnotknowtheclientaccesspatterns.in WeexaminetheperformanceofthePure-PullapproachandBPwithempha-

9 cachebutthecontentofthebroadcastprogramaswell. theentirebroadcastslotsarededicatedtothepulledpagesinpure-pullapproach.forbothflatandbdiskapproaches,thethresholdissetto100%. ForBP,onethirdofthebroadcastslotsareallocatedforpulledpages,while ForBP(FLATorBDISK),dynamicworkloadhaseectsonnotonlytheclient Thatis,theclientrstchecksthewholebroadcastprogramtoseewhetherthe desiredpageswillappearwithinthereminingpartofthecurrentbroadcastprogram.iftheanswerisno,thentheclientissuesanexplicitpullrequesttothe server.otherwise,theclientkeepsonmonitoringthebroadcastchannelforthe desireddata. populations.thezipfdistributionparameterissetto0.95.wedeneaclientto thezipfdistributionandevaluatetheadaptivenessofbpfordierentclient Intherstsetofexperimentsweassumethattheclientaccesspatternsfollow 4.1PerformanceforChangingUserPopulationSize beatstablestagewhichisachievedbywaiting2000accessesmoreaftertheclient cachelledandasystemisatstablestagewhenallclientsinthesystemareat stablestage.forasystematstablestage,thewarm-upeectsintheclientcache andthebroadcastprogramareeliminated,suchthattheclientcachescontain thehottestaccessspotandtheserverhasinformationontheaccesspatterns. atinitialstage,boththeclientsandtheserverhavenoknowledgeoftheclient Inthecontrast,theinitialstageofthesystemisfromthetimewhenthesystem beginstoworktothetimewhenthesystemisatstablestage.henceasystem accesspatterns approachesatstablestagewiththatatinitialstageinfigure3.obviously, Fig.3.AccessTimeVSClientPopulation. B_DISK (Initial) B_DISK (Steady) 150 (Initial) (Steady) clientsandtheserverhavetheknowledgeofaccessprobabilityinformation, stage.thisindicatesthatafteracertainperiodof"learning"stageboththe ToseetheadaptabilityofBP,wecomparethesystemperformanceforBP which,inturn,guaranteestheclientstocacheimportantpagesandtheserverto tailorthebroadcastprogramaccordingtoclientdemand.bdiskalwaysgives bothflatandbdiskatsteadystatealwaysperformbetterthanatinitial Number of Clients abetterperformancethanflatapproachatbothstableandinitialstage.this isconsistentwiththeresultsobtainedforstaticbroadcastprogram[afz97]. Access Time

10 system(boththeclientsandtheserver)knowstheexactclientaccesspatterns. idealbroadcastdiskswithrespecttoflatandbdisk.fortheidealones,the thatthemostfrequentlyaccesseddatasubsetiscachedintheclient,theless BPdisseminatesinformationaccordingtotheoptimalaccessdistributionsuch Toprovidecomparisonbaselines,weintroducetheidealBP:idealatand frequentlyaccesseddatasubsetisprovidedbybroadcast,andtherestispulled fromtheserverbyexplicitrequests.thebroadcastprogramforbroadcastdisks isalsoconstructedaccordingtotheaccessprobabilityofthedatasubset.the idealbpistheultimateperformancegoalofoursystem B_DISK (Steady) FLAT,andBDISK,areshowninFigure4.Notethattheresultsshownare Theexperimentalresultsforthethreedatadeliverymethods:Pure-Pull, Fig.4.AccessTimeVSClientPopulation. 150 Ideal B_DISK (Steady) 100 Ideal numberofclientsisbeyond600,pure-pullperformsworsethanbdisk,and timeforthismethodincreasesrapidlyastheclientpopulationgrows.whenthe whenthesystembecomesevenmoreheavilyloaded(i.e.,thenumberofclients obtainedwhenthesystemisatstablestage.asexpected,thepure-pullapproach isgoodonlywhenthesystemworkloadislight(leftsideofthegure).theaccess Number Clients isgreaterthan700),pure-pullgivesworseperformancethanflatandbdisk. However,asclientpopulationsbecomelarge,thecurveforPure-Pullbecomes broadcastchannelformsahot-spot.asaresult,onepulleddataitemcansatisfy at.thereasonisasthenumberofclientsincreasesthepulleddatasetinthe alargenumberofclientrequests.sincethebasicpagereplacementalgorithm forclientcachingisthesameforbpandpure-pull,theimprovementofthe performanceforbpoverpure-pullisaresultofthetailoredbroadcastprogram. andbdiskmayneverachievetheidealperformance,thedynamicbroadcast onesverywell.thereisacertaingapbetweentherealapproacheswiththeideal approaches.thedierenceisduetotheimpreciseaccesspatternscollectedby theserverandtheinaccuratelrukpagereplacementstrategy.though,flat ItisobviousthatbothFLATandBDISKcanmanagetofollowtheideal 4.2PerformanceforChangingHot-Spot programisstillaneectiveapproachbecauseofitsadaptivenesstothesystem. ToevaluatetheperformanceofBPindynamicworkloads,weassumethatthe clientaccesspatternsfollowthegaussiandistribution.suchthatvariesto createtheeectofdynamicworkloadandwhichreectstheskewofclient Access Time

11 accesspatternsis100or300.inthisway,wecanmodeltheeliminationof accesshot-spotsandthegenerationofanewoneinanotherpartofthedatabase (randomlyselected).tointerprettheimpactofsuchhot-spotmigrationonthe systemperformance,weassumedthatallclientshavethesamehot-spotsand theychangeaccessdemandatthesametime.clientsstaywitheachhot-spot fordurationperiodsofsimulationtime Fig.5.MigrationofHot-spot. 250 B_DISK (Skewed=100) FLAT (Skewed=100) Pure-Pull (Skewed=100) 200 B_DISK (Skewed=300) FLAT (Skewed=300) (Skewed=300) andflat.sincewithsmallerduration,thebroadcastprogramofbpmaynot 300,showtwodierentdataaccesspatterns.Formoredynamicworkload(left clientpopulationis1000.twosetofcurves,labeledasskewed100andskewed InFigure5,weshowtheresultsobtainedasafunctionofDuration,wherethe theclientschangeaccesshot-spottoanothernewone.asaresult,thebenet reachsteadystateorthesteadybroadcastprogramisobtainedshortlybefore sideofthegureswheredurationisshort),pure-pulloutperformsbothbdisk (x 10,000) performancethanthepure-pullmethod,whichcanbeobservedintherightpart ofthegure.however,asthedurationincreasesto4106,anyfurtherincrease oftheaircachedoesn'thaveenoughtimetobecomeeective.onlywhenthe ofthedurationwon'timprovetheperformanceofbothbpandpure-pull.it systemreachessteadystateandstaytherelongenough,wouldbpgivebetter accesspattern=100thantheaccesspattern=300,becausefor=100 ComparedwithPure-Pull,BPimprovesthesystemperformancemoreforthe isbecausethatthesystemreachesthesteadystateafterthatdurationperiod. therearemoreairhitsthan=300. 5Conclusions Inthispaper,weproposedadynamicdatadeliverysystemforasymmetriccommunicationenvironments,whereinformationcanberetrievedbydierentmeans (i.e.,clientcache,broadcastchannels,ortheserverdatabasebyexplicitpull sothatonlythemostdesireddatasubsetisscheduledtobroadcast.pullrequests termineddynamicallybytheaccesspatternsoftheclients,notbyprecompiled userproles.thebroadcastprogramistailoredaccordingtoaccessfrequencies requests),toachievethemaximumadvantages.thewaydataarestoredisde- areneededonlywhenthedesireddatacannotbefoundintheclientcacheor thebroadcastprogram. Access Time

12 datadeliverysystem.wefoundthatthesystem(clientsandserver)canadaptto thedynamicallychangingworkloadverywell.generally,broadcastdisksandat broadcasthavebetterperformancethanpurepullapproachwhendataaccess patternsareskewed.forskeweddataaccesspatternsandlessdynamicworkloads,broadcastdisksoutperformatbroadcast.however,purepulltechnique ofdierentbroadcastprogramstructure(i.e.,thesizeofbroadcastprogrametc.) ontheperformanceofdatadelivery.moreover,wewillincludethedissemination ofupdatesinthehierarchicaldatacachingsystem. tweenpullandpushbandwidthaccordingtotheserverworkloadandtheimpact Asfuturework,wewillinvestigatewaystodynamicallyadjusttheratiobe- Asimulationmodelhasbeendevelopedtoevaluatetheperformanceofthe isabetterchoiceforrandomdataaccesspatternsorhighlydynamicworkloads. References [AAFZ95]S.Acharya,R.Alonso,M.Franklin,andS.Zdonik.Broadcastdisks:Data [AFZ97]S.Acharya,M.Franklin,andS.Zdonik.Balancingpushandpullfordata managementforasymmetriccommunicationsenvironments.inproceedings SanJose,California,May1995. oftheacmsigmodconferenceonmanagementofdata,pages199{210, [BI94] D.BarbaraandT.Imielinksi.Sleepersandworkaholics:Cachingstrategies ofdata,pages183{194,tuscon,arizona,may1997. broadcast.inproceedingsoftheacmsigmodconferenceonmanagement [EPW93]E.J.O'Neil,P.E.O'Neil,andG.Weikum.Thelru-kpagereplacementalgorithmfordatabasediskbuering.InProceedingsoftheACMSIGMOD formobileenvironments.inproceedingsoftheacmsigmodconference onmanagementofdata,pages1{12,minneapolis,minnesota,may1994. [HL98] [HLL98]Q.L.Hu,D.L.Lee,andW.-C.Lee.Hierarchicaldatacachinginasymmetric ConferenceonManagementofData,pages297{306,Washington,D.C,May Q.L.HuandD.L.Lee.Cachealgorithmsbasedonadaptiveinvalidation [IVB94]T.Imielinski,S.Viswanathan,andB.R.Badrinath.Energyeciencyindexingonair.InProceedingsoftheInternationalConferenceonSIGMOD, reportsformobileenvironments.clustercomputing,1(1):39{48,feb [JBEA97]J.Jing,O.Bukhres,A.K.Elmargarmid,andR.Alonso.Bit-sequences: pages25{36,1994. communicationenvironments.inunderpreparation,1998. Anewcacheinvalidationmethodinmobileenvironments.ACM/Baltzer [Sch92]H.Schwetman.Csimuser'sguide(version17).MCCCorporation,1992. [SRB97]K.Stathatos,N.Roussopoulos,andJ.S.Baras.Adaptivedatabroadcast [Knu81]D.Knuth.TheArtofComputerProgramming,volume3.Addison-Wesley, CA,1981. MobileNetworksandApplications,2(II),1997. [WYC96]K.-L.Wu,P.S.Yu,andM.-S.Chen.Energy-ecientcachingforwireless 326{335,Athens,Greece,August1997. inhybridnetworks.inproceedingsofthe23rdvldbconference,pages mobilecomputing.in12thinternationalconferenceondataengineering, pages336{345,feb.26-march11996.

Broadcast Disks: Data Management for Asymmetric Communication Environments

Broadcast Disks: Data Management for Asymmetric Communication Environments Appears in Proceedings of the ACM SIGMOD Conference, San Jose, CA, May 1995 Broadcast Disks: Data Management for Asymmetric Communication Environments Swarup Acharya Brown University sa@cs.brown.edu Rafael

More information

A novel push-and-pull hybrid data broadcast scheme for wireless information networks

A novel push-and-pull hybrid data broadcast scheme for wireless information networks Title A novel push-and-pull hybrid data broadcast scheme for wireless information networks Author(s) Hu, JH; Yeung, LK; Feng, G; Leung, KF Citation The 2000 IEEE International Conference on Communications

More information

FAST 11. Yongseok Oh <ysoh@uos.ac.kr> University of Seoul. Mobile Embedded System Laboratory

FAST 11. Yongseok Oh <ysoh@uos.ac.kr> University of Seoul. Mobile Embedded System Laboratory CAFTL: A Content-Aware Flash Translation Layer Enhancing the Lifespan of flash Memory based Solid State Drives FAST 11 Yongseok Oh University of Seoul Mobile Embedded System Laboratory

More information

Implementing and Maintaining Microsoft SQL Server 2005 Reporting Services COURSE OVERVIEW AUDIENCE OUTLINE OBJECTIVES PREREQUISITES

Implementing and Maintaining Microsoft SQL Server 2005 Reporting Services COURSE OVERVIEW AUDIENCE OUTLINE OBJECTIVES PREREQUISITES COURSE OVERVIEW This three-day instructor-led course teaches students how to implement a ing Services solution in their organizations. The course discusses how to use the ing Services development tools

More information

Implementing and Maintaining Microsoft SQL Server 2008 Reporting Services

Implementing and Maintaining Microsoft SQL Server 2008 Reporting Services Course 6236A: Implementing and Maintaining Microsoft SQL Server 2008 Reporting Services Length: 3 Days Published: December 05, 2008 Language(s): English Audience(s): IT Professionals Level: 200 Technology:

More information

Bandwidth consumption: Adaptive Defense and Adaptive Defense 360

Bandwidth consumption: Adaptive Defense and Adaptive Defense 360 Contents 1. 2. 3. 4. How Adaptive Defense communicates with the Internet... 3 Bandwidth consumption summary table... 4 Estimating bandwidth usage... 5 URLs required by Adaptive Defense... 6 1. How Adaptive

More information

About Me: Brent Ozar. Perfmon and Profiler 101

About Me: Brent Ozar. Perfmon and Profiler 101 Perfmon and Profiler 101 2008 Quest Software, Inc. ALL RIGHTS RESERVED. About Me: Brent Ozar SQL Server Expert for Quest Software Former SQL DBA Managed >80tb SAN, VMware Dot-com-crash experience Specializes

More information

Windows Server 2008 R2 Hyper V. Public FAQ

Windows Server 2008 R2 Hyper V. Public FAQ Windows Server 2008 R2 Hyper V Public FAQ Contents New Functionality in Windows Server 2008 R2 Hyper V...3 Windows Server 2008 R2 Hyper V Questions...4 Clustering and Live Migration...5 Supported Guests...6

More information

Instant Revocation. Jon A. Solworth. 16 June 2008. Dept. of Computer Science and Center for RITES University of Illinois at Chicago

Instant Revocation. Jon A. Solworth. 16 June 2008. Dept. of Computer Science and Center for RITES University of Illinois at Chicago Instant Revocation Jon A. Solworth Dept. of Computer Science and Center for RITES University of Illinois at Chicago 16 June 2008 Certificates and Revocation Part I Certificates and Revocation Certificates

More information

Redpaper. Performance Metrics in TotalStorage Productivity Center Performance Reports. Introduction. Mary Lovelace

Redpaper. Performance Metrics in TotalStorage Productivity Center Performance Reports. Introduction. Mary Lovelace Redpaper Mary Lovelace Performance Metrics in TotalStorage Productivity Center Performance Reports Introduction This Redpaper contains the TotalStorage Productivity Center performance metrics that are

More information

Digital Certificate IP Address Test Procedure

Digital Certificate IP Address Test Procedure Digital Certificate IP Address Test Procedure Testing Firewall Access to New CA IP Addresses May 11 th to 15 th Verizon is making some test servers with the new Certificate Authority (CA) system IP addresses

More information

Microsoft Access 2007

Microsoft Access 2007 How to Use: Microsoft Access 2007 Microsoft Office Access is a powerful tool used to create and format databases. Databases allow information to be organized in rows and tables, where queries can be formed

More information

TECHNICAL NOTE. The following information is provided as a service to our users, customers, and distributors.

TECHNICAL NOTE. The following information is provided as a service to our users, customers, and distributors. page 1 of 11 The following information is provided as a service to our users, customers, and distributors. ** If you are just beginning the process of installing PIPSPro 4.3.1 then please note these instructions

More information

dotmailer for Dynamics Frequently Asked Questions v 6,0

dotmailer for Dynamics Frequently Asked Questions v 6,0 for Dynamics Frequently Asked Questions v 6,0 Page 1 Contents Introduction... 2 Why should I use the Microsoft Dynamics CRM Connector for dotmailer?... 3 What software needs to be installed?... 3 Can I

More information

Analyzing Big Data with Splunk A Cost Effective Storage Architecture and Solution

Analyzing Big Data with Splunk A Cost Effective Storage Architecture and Solution Analyzing Big Data with Splunk A Cost Effective Storage Architecture and Solution Jonathan Halstuch, COO, RackTop Systems JHalstuch@racktopsystems.com Big Data Invasion We hear so much on Big Data and

More information

Database!Fatal!Flash!Flaws!No!One! Talks!About!!!

Database!Fatal!Flash!Flaws!No!One! Talks!About!!! MarcStaimer,President&CDSDragonSlayerConsulting W h i t e P A P E R DatabaseFatalFlashFlawsNoOne TalksAbout AndHowtoAvoidThem WHITEPAPER DatabaseFatalFlashFlawsNoOneTalksAbout AndHowtoAvoidThem DatabaseFatalFlashFlawsNoOneTalksAbout

More information

Support Desk Help Manual. v 1, May 2014

Support Desk Help Manual. v 1, May 2014 Support Desk Help Manual v 1, May 2014 Table of Contents When do I create a ticket in DataRPM?... 3 How do I decide the Priority of the bug I am logging in?... 3 How do I Create a Ticket?... 3 How do I

More information

Spam Testing Methodology Opus One, Inc. March, 2007

Spam Testing Methodology Opus One, Inc. March, 2007 Spam Testing Methodology Opus One, Inc. March, 2007 This document describes Opus One s testing methodology for anti-spam products. This methodology has been used, largely unchanged, for four tests published

More information

Westek Technology Snapshot and HA iscsi Replication Suite

Westek Technology Snapshot and HA iscsi Replication Suite Westek Technology Snapshot and HA iscsi Replication Suite Westek s Power iscsi models have feature options to provide both time stamped snapshots of your data; and real time block level data replication

More information

Τhe SAS BI delivers business-critical answers ahead of the competition Yannis Salamaras Senior Business Intelligence Consultant SAS Greece & Cyprus

Τhe SAS BI delivers business-critical answers ahead of the competition Yannis Salamaras Senior Business Intelligence Consultant SAS Greece & Cyprus Τhe SAS BI delivers business-critical answers ahead of the competition Yannis Salamaras Senior Business Intelligence Consultant SAS Greece & Cyprus The Value of the Information What s wrong with this picture?

More information

TRACE PERFORMANCE TESTING APPROACH. Overview. Approach. Flow. Attributes

TRACE PERFORMANCE TESTING APPROACH. Overview. Approach. Flow. Attributes TRACE PERFORMANCE TESTING APPROACH Overview Approach Flow Attributes INTRODUCTION Software Testing Testing is not just finding out the defects. Testing is not just seeing the requirements are satisfied.

More information

Comparison of Hybrid Flash Storage System Performance

Comparison of Hybrid Flash Storage System Performance Test Validation Comparison of Hybrid Flash Storage System Performance Author: Russ Fellows March 23, 2015 Enabling you to make the best technology decisions 2015 Evaluator Group, Inc. All rights reserved.

More information

Keep an eye on your PostgreSQL clusters

Keep an eye on your PostgreSQL clusters Keep an eye on your PostgreSQL clusters Open PostgreSQL Monitoring & PostgreSQL Workload Analyzer Julien Rouhaud Dalibo - www.dalibo.org pgconf.ru 2015 - February [ 1 / 39 ] Monitoring? Service availability

More information

Pervasive data access in wireless and mobile computing environments

Pervasive data access in wireless and mobile computing environments WIRELESS COMMUNICATIONS AND MOBILE COMPUTING Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/wcm.424 Pervasive data access in wireless and mobile computing environments

More information

Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at

Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at distributing load b. QUESTION: What is the context? i. How

More information

This course will also teach how to create various kinds of dashboards using Reporting Services.

This course will also teach how to create various kinds of dashboards using Reporting Services. Implementing and Maintaining Microsoft SQL Server 2008 Reporting Services Length : 3 Days (24 Hours) Language(s) : English Audience(s) : IT Professionals Level : 200 Technology : Microsoft SQL Server 2008

More information

The FlexiSchools Online Order Management System Installation Guide

The FlexiSchools Online Order Management System Installation Guide The FlexiSchools Online Order Management System Installation Guide FlexiSchools May 2012 Page 1 of 18 Installation Pack Welcome to the FlexiSchools system. You will have received a disc containing: Sato

More information

Printers connected to a computer running GARO printer driver and Windows OS (2000, XP, 2003 Server, Vista 32/64-bit)

Printers connected to a computer running GARO printer driver and Windows OS (2000, XP, 2003 Server, Vista 32/64-bit) Title: Model: ipf500, ipf600, ipf700, ipf5000, ipf5100, ipf6100, ipf8000, and ipf9000 Note: This information is intended for end users. This document describes how to retrieve status information from the

More information

JaM - PHP Error Monitoring Extension

JaM - PHP Error Monitoring Extension JaM - PHP Error Monitoring Extension jess.portnoy@kaltura.com April 20, 2016 The Need Big complex PHP based systems have a lot of moving parts. It is very common for something to malfunction without being

More information

SQL diagnostic manager Management Pack for Microsoft System Center. Overview

SQL diagnostic manager Management Pack for Microsoft System Center. Overview Overview What is so cool about the SQL diagnostic manager Management Pack? The SQL diagnostic manager (SQLdm) Management Pack integrates key monitors and alerts used by SQL Server DBAs with Microsoft's

More information

Boost SQL Server Performance Buffer Pool Extensions & Delayed Durability

Boost SQL Server Performance Buffer Pool Extensions & Delayed Durability Boost SQL Server Performance Buffer Pool Extensions & Delayed Durability Manohar Punna President - SQLServerGeeks #509 Brisbane 2016 Agenda SQL Server Memory Buffer Pool Extensions Delayed Durability Analysis

More information

Gladstone Health & Leisure Technical Services

Gladstone Health & Leisure Technical Services Gladstone Health & Leisure Technical Services Plus2 Environment Server Recommendations Commercial in Confidence Database Server Specifications Database server specifications are based on sizes in use on

More information

Deployment Planning Guide

Deployment Planning Guide Deployment Planning Guide August 2011 Copyright: 2011, CCH, a Wolters Kluwer business. All rights reserved. Material in this publication may not be reproduced or transmitted in any form or by any means,

More information

CIO Update: Microsoft's Business Intelligence Strategy Is a Work in Progress

CIO Update: Microsoft's Business Intelligence Strategy Is a Work in Progress IGG-05282003-04 B. Hostmann, K. Strange Article 28 May 2003 CIO Update: Microsoft's Business Intelligence Strategy Is a Work in Progress Microsoft s SQL Server and related business intelligence (BI) products

More information

Who am I? Copyright 2014, Oracle and/or its affiliates. All rights reserved. 3

Who am I? Copyright 2014, Oracle and/or its affiliates. All rights reserved. 3 Oracle Database In-Memory Power the Real-Time Enterprise Saurabh K. Gupta Principal Technologist, Database Product Management Who am I? Principal Technologist, Database Product Management at Oracle Author

More information

Oracle Database Scalability in VMware ESX VMware ESX 3.5

Oracle Database Scalability in VMware ESX VMware ESX 3.5 Performance Study Oracle Database Scalability in VMware ESX VMware ESX 3.5 Database applications running on individual physical servers represent a large consolidation opportunity. However enterprises

More information

Accelerating Application Performance on Virtual Machines

Accelerating Application Performance on Virtual Machines Accelerating Application Performance on Virtual Machines...with flash-based caching in the server Published: August 2011 FlashSoft Corporation 155-A W. Moffett Park Dr Sunnyvale, CA 94089 info@flashsoft.com

More information

How to Scale out SharePoint Server 2007 from a single server farm to a 3 server farm with Microsoft Network Load Balancing on the Web servers.

How to Scale out SharePoint Server 2007 from a single server farm to a 3 server farm with Microsoft Network Load Balancing on the Web servers. 1 How to Scale out SharePoint Server 2007 from a single server farm to a 3 server farm with Microsoft Network Load Balancing on the Web servers. Back to Basics Series By Steve Smith, MVP SharePoint Server,

More information

Enterprise Edition. Hardware Requirements

Enterprise Edition. Hardware Requirements Enterprise Edition Hardware Requirements For Blackbaud FundWare 6 to 100+ user environments Revised August 6 th, 2008 Table of Contents Supported Platforms... 3 6 to 25 User Environment... 4 26 to 35 User

More information

Hierarchical Bloom Filters: Accelerating Flow Queries and Analysis

Hierarchical Bloom Filters: Accelerating Flow Queries and Analysis Hierarchical Bloom Filters: Accelerating Flow Queries and Analysis January 8, 2008 FloCon 2008 Chris Roblee, P. O. Box 808, Livermore, CA 94551 This work performed under the auspices of the U.S. Department

More information

Pulsar Realtime Analytics At Scale. Tony Ng April 14, 2015

Pulsar Realtime Analytics At Scale. Tony Ng April 14, 2015 Pulsar Realtime Analytics At Scale Tony Ng April 14, 2015 Big Data Trends Bigger data volumes More data sources DBs, logs, behavioral & business event streams, sensors Faster analysis Next day to hours

More information

DATABASE VIRTUALIZATION AND INSTANT CLONING WHITE PAPER

DATABASE VIRTUALIZATION AND INSTANT CLONING WHITE PAPER DATABASE VIRTUALIZATION AND INSTANT CLONING TABLE OF CONTENTS Brief...3 Introduction...3 Solutions...4 Technologies....5 Database Virtualization...7 Database Virtualization Examples...9 Summary....9 Appendix...

More information

REQUEST FOR PROPOSAL NO. AHS16-1. ADDENDUM No. 1 ANSWERS TO WRITTEN QUESTIONS RECEIVED FROM INTERESTED VENDORS FOR

REQUEST FOR PROPOSAL NO. AHS16-1. ADDENDUM No. 1 ANSWERS TO WRITTEN QUESTIONS RECEIVED FROM INTERESTED VENDORS FOR REQUEST FOR PROPOSAL NO. AHS16-1 ADDENDUM No. 1 ANSWERS TO WRITTEN QUESTIONS RECEIVED FROM INTERESTED VENDORS FOR PROVIDER INFORMATION MANAGEMENT DATABASE OPEN - March 11 th, 2016 RESPONSE TO QUESTIONS

More information

New Advanced RAID Level for Today's Larger Storage Capacities: Advanced Data Guarding

New Advanced RAID Level for Today's Larger Storage Capacities: Advanced Data Guarding White Paper October 2000 Prepared by Storage Products Group Compaq Computer Corporation Contents Introduction...2 What customers can expect from Compaq RAID ADG solution...3 RAID ADG Features and Benefits...3

More information

etoken Enterprise For: SSL SSL with etoken

etoken Enterprise For: SSL SSL with etoken etoken Enterprise For: SSL SSL with etoken System Requirements Windows 2000 Internet Explorer 5.0 and above Netscape 4.6 and above etoken R2 or Pro key Install etoken RTE Certificates from: (click on the

More information

Idera SQL Diagnostic Manager Management Pack Guide for System Center Operations Manager. Install Guide. Idera Inc., Published: April 2013

Idera SQL Diagnostic Manager Management Pack Guide for System Center Operations Manager. Install Guide. Idera Inc., Published: April 2013 Idera SQL Diagnostic Manager Management Pack Guide for System Center Operations Manager Install Guide Idera Inc., Published: April 2013 Contents Introduction to the Idera SQL Diagnostic Manager Management

More information

Creating A Highly Available Database Solution

Creating A Highly Available Database Solution WHITE PAPER Creating A Highly Available Database Solution Advantage Database Server and High Availability TABLE OF CONTENTS 1 Introduction 1 High Availability 2 High Availability Hardware Requirements

More information

User Installation Guide

User Installation Guide The will provide a step-by-step walkthough of how to download and install the application, activate each feature of the product, install any of the feature's prerequisites, extend the license, and deactivate

More information

Kafka & Redis for Big Data Solutions

Kafka & Redis for Big Data Solutions Kafka & Redis for Big Data Solutions Christopher Curtin Head of Technical Research @ChrisCurtin About Me 25+ years in technology Head of Technical Research at Silverpop, an IBM Company (14 + years at Silverpop)

More information

Symantec Encryption Solutions for Email, Powered by PGP Technology

Symantec Encryption Solutions for Email, Powered by PGP Technology Symantec Encryption Solutions for Email, Powered by PGP Technology Data Sheet: Encryption The Problem with Email Are you worried that users are emailing sensitive information openly? According to Osterman

More information

Step-by-step installation guide for monitoring untrusted servers using Operations Manager ( Part 3 of 3)

Step-by-step installation guide for monitoring untrusted servers using Operations Manager ( Part 3 of 3) Step-by-step installation guide for monitoring untrusted servers using Operations Manager ( Part 3 of 3) Manual installation of agents and importing the SCOM certificate to the servers to be monitored:

More information

Chapter 6: Broadcast Systems. Mobile Communications. Unidirectional distribution systems DVB DAB. High-speed Internet. architecture Container

Chapter 6: Broadcast Systems. Mobile Communications. Unidirectional distribution systems DVB DAB. High-speed Internet. architecture Container Mobile Communications Chapter 6: Broadcast Systems Unidirectional distribution systems DAB DVB architecture Container High-speed Internet Prof. Dr.-Ing. Jochen Schiller, http://www.jochenschiller.de/ MC

More information

XenData Product Brief: SX-550 Series Servers for LTO Archives

XenData Product Brief: SX-550 Series Servers for LTO Archives XenData Product Brief: SX-550 Series Servers for LTO Archives The SX-550 Series of Archive Servers creates highly scalable LTO Digital Video Archives that are optimized for broadcasters, video production

More information

Top 10 Performance Tips for OBI-EE

Top 10 Performance Tips for OBI-EE Top 10 Performance Tips for OBI-EE Narasimha Rao Madhuvarsu L V Bharath Terala October 2011 Apps Associates LLC Boston New York Atlanta Germany India Premier IT Professional Service and Solution Provider

More information

Chapter-15 -------------------------------------------- Replication in SQL Server

Chapter-15 -------------------------------------------- Replication in SQL Server Important Terminologies: What is Replication? Replication is the process where data is copied between databases on the same server or different servers connected by LANs, WANs, or the Internet. Microsoft

More information

T H E O P E N S O U R C E E L E A R N I N G B L O G

T H E O P E N S O U R C E E L E A R N I N G B L O G Page 1 of 8 Share 1 More Next Blog» Create Blog Sign In T H E O P E N S O U R C E E L E A R N I N G B L O G A D I S C U S S I O N O F O P E N S O U R C E E L E A R N I N G T O O L S C O V E R I N G M O

More information

Sun 8Gb/s Fibre Channel HBA Performance Advantages for Oracle Database

Sun 8Gb/s Fibre Channel HBA Performance Advantages for Oracle Database Performance Advantages for Oracle Database At a Glance This Technical Brief illustrates that even for smaller online transaction processing (OLTP) databases, the Sun 8Gb/s Fibre Channel Host Bus Adapter

More information

Question 3.1.1. Question 3.2.1. Question 3.3.1. EdTech 552: Lab 3 Answer Sheet

Question 3.1.1. Question 3.2.1. Question 3.3.1. EdTech 552: Lab 3 Answer Sheet Question 3.1.1 Question Answers a. 123 01111011 b. 202 11001010 c. 67 01000011 d. 7 00000111 e. 252 11111100 f. 91 01011011 g. 116.127.71.3 01110100.01111111.01010001.00000011 h. 255.255.255.0 11111111.11111111.11111111.00000000

More information

Oracle Database In-Memory The Next Big Thing

Oracle Database In-Memory The Next Big Thing Oracle Database In-Memory The Next Big Thing Maria Colgan Master Product Manager #DBIM12c Why is Oracle do this Oracle Database In-Memory Goals Real Time Analytics Accelerate Mixed Workload OLTP No Changes

More information

MailEnable Installation Guide

MailEnable Installation Guide MailEnable Installation Guide MailEnable Messaging Services for Microsoft Windows 2000/2003/2008 Installation Guide for: MailEnable Standard Edition MailEnable Professional Edition MailEnable Enterprise

More information

Using ODVA Common Industrial Protocol to Enhance Performance White Paper

Using ODVA Common Industrial Protocol to Enhance Performance White Paper Monitor & Control Multiple Groups Using ODVA Common Industrial Protocol to Enhance Performance White Paper Monitor & Control Multiple Groups Using ODVA Common Industrial Protocol to Enhance Performance

More information

Recovering from a Hard Drive Failure

Recovering from a Hard Drive Failure APPENDIXB This appendix describes how to recover from a hard drive failure on a Multiservices Platform Series device. The process for doing so depends on the physical security product that you are using

More information

In-Memory Analytics: A comparison between Oracle TimesTen and Oracle Essbase

In-Memory Analytics: A comparison between Oracle TimesTen and Oracle Essbase In-Memory Analytics: A comparison between Oracle TimesTen and Oracle Essbase Agenda Introduction Why In-Memory? Options for In-Memory in Oracle Products - Times Ten - Essbase Comparison - Essbase Vs Times

More information

EMC Backup and Recovery for Microsoft SQL Server 2008 Enabled by EMC Celerra Unified Storage

EMC Backup and Recovery for Microsoft SQL Server 2008 Enabled by EMC Celerra Unified Storage EMC Backup and Recovery for Microsoft SQL Server 2008 Enabled by EMC Celerra Unified Storage Applied Technology Abstract This white paper describes various backup and recovery solutions available for SQL

More information

Evaluating NoSQL for Enterprise Applications. Dirk Bartels VP Strategy & Marketing

Evaluating NoSQL for Enterprise Applications. Dirk Bartels VP Strategy & Marketing Evaluating NoSQL for Enterprise Applications Dirk Bartels VP Strategy & Marketing Agenda The Real Time Enterprise The Data Gold Rush Managing The Data Tsunami Analytics and Data Case Studies Where to go

More information

MAX Auto Dialer Copyright 2011 Main Telecom, Inc.

MAX Auto Dialer Copyright 2011 Main Telecom, Inc. Table of Contents 1. Introduction to MAX Auto Dialer... 2 1.1 Overview: 2 1.2 Feature: 2 1.3 Benefits 2 2. System Requirements and Installation:... 2 2.1 System Requirements: 2 2.1.1. Hardware Requirements:...

More information

SATA 6G PCIe Card User Manual Model: UGT-ST622

SATA 6G PCIe Card User Manual Model: UGT-ST622 SATA 6G PCIe Card User Manual Model: UGT-ST622 All brand names and trademarks are properties of their respective owners www.vantecusa.com Contents: Chapter 1: Introduction... 3 1.1 Product Introduction...

More information

IFS CLOUD UPLINK INSTALLATION GUIDE

IFS CLOUD UPLINK INSTALLATION GUIDE IFS CLOUD UPLINK INSTALLATION GUIDE ABSTRACT This guide describes how to install IFS Cloud Uplink. UPLINK VERSION 4.13 PREPARE THE WEB SERVER THAT SERVES IFS EXTENDED SERVER Since the user credentials

More information

NE-2273B Managing and Maintaining a Microsoft Windows Server 2003 Environment

NE-2273B Managing and Maintaining a Microsoft Windows Server 2003 Environment NE-2273B Managing and Maintaining a Microsoft Windows Server 2003 Environment Summary Duration Vendor Audience 5 Days Microsoft IT Professionals Published Level Technology 05 October 2005 200 Microsoft

More information

Guidelines for Creating Reports

Guidelines for Creating Reports Guidelines for Creating Reports Contents Exercise 1: Custom Reporting - Ad hoc Reports... 1 Exercise 2: Custom Reporting - Ad Hoc Queries... 5 Exercise 3: Section Status Report.... 8 Exercise 1: Custom

More information

Private Cloud Storage for Media Applications. Bang Chang Vice President, Broadcast Servers and Storage bang.chang@xor-media.com

Private Cloud Storage for Media Applications. Bang Chang Vice President, Broadcast Servers and Storage bang.chang@xor-media.com Private Cloud Storage for Media Bang Chang Vice President, Broadcast Servers and Storage bang.chang@xor-media.com Table of Contents Introduction Cloud Storage Requirements Application transparency Universal

More information

The Evolution of Microsoft SQL Server: The right time for Violin flash Memory Arrays

The Evolution of Microsoft SQL Server: The right time for Violin flash Memory Arrays The Evolution of Microsoft SQL Server: The right time for Violin flash Memory Arrays Executive Summary Microsoft SQL has evolved beyond serving simple workgroups to a platform delivering sophisticated

More information

LARGEST UK NETWORK SOLVES BIG PROBLEM WITH SMALL SOLUTION

LARGEST UK NETWORK SOLVES BIG PROBLEM WITH SMALL SOLUTION LARGEST UK NETWORK SOLVES BIG PROBLEM WITH SMALL SOLUTION ITV s production editing team edits hundreds of hours of online video every month. IT management needed a solution that would improve overall performance

More information

Scaling Sitecore for Load

Scaling Sitecore for Load Scaling Sitecore for Load introduction In 2012, SolutionSet rebuilt the California Lottery website from the ground up and learned a great deal about building a Sitecore infrastructure to withstand targeted

More information

Lab 2. CS-335a. Fall 2012 Computer Science Department. Manolis Surligas surligas@csd.uoc.gr

Lab 2. CS-335a. Fall 2012 Computer Science Department. Manolis Surligas surligas@csd.uoc.gr Lab 2 CS-335a Fall 2012 Computer Science Department Manolis Surligas surligas@csd.uoc.gr 1 Summary At this lab we will cover: Basics of Transport Layer (TCP, UDP) Broadcast ARP DNS More Wireshark filters

More information

McAfee Content Security Reporter 2.0.0

McAfee Content Security Reporter 2.0.0 Product Guide Revision A McAfee Content Security Reporter 2.0.0 For use with epolicy Orchestrator 4.6.5 Software COPYRIGHT Copyright 2013 McAfee, Inc. Do not copy without permission. TRADEMARK ATTRIBUTIONS

More information

Virtuoso and Database Scalability

Virtuoso and Database Scalability Virtuoso and Database Scalability By Orri Erling Table of Contents Abstract Metrics Results Transaction Throughput Initializing 40 warehouses Serial Read Test Conditions Analysis Working Set Effect of

More information

NetIQ. How to guides: AppManager v7.04 Initial Setup for a trial. Haf Saba Attachmate NetIQ. Prepared by. Haf Saba. Senior Technical Consultant

NetIQ. How to guides: AppManager v7.04 Initial Setup for a trial. Haf Saba Attachmate NetIQ. Prepared by. Haf Saba. Senior Technical Consultant How to guides: AppManager v7.04 Initial Setup for a trial By NetIQ Prepared by Haf Saba Senior Technical Consultant Asia Pacific 1 Executive Summary This document will walk you through an initial setup

More information

Data Warehouse and Business Intelligence Testing: Challenges, Best Practices & the Solution

Data Warehouse and Business Intelligence Testing: Challenges, Best Practices & the Solution Warehouse and Business Intelligence : Challenges, Best Practices & the Solution Prepared by datagaps http://www.datagaps.com http://www.youtube.com/datagaps http://www.twitter.com/datagaps Contact contact@datagaps.com

More information

Oracle Big Data, In-memory, and Exadata - One Database Engine to Rule Them All Dr.-Ing. Holger Friedrich

Oracle Big Data, In-memory, and Exadata - One Database Engine to Rule Them All Dr.-Ing. Holger Friedrich Oracle Big Data, In-memory, and Exadata - One Database Engine to Rule Them All Dr.-Ing. Holger Friedrich Agenda Introduction Old Times Exadata Big Data Oracle In-Memory Headquarters Conclusions 2 sumit

More information

Online Courses. Version 9 Comprehensive Series. What's New Series

Online Courses. Version 9 Comprehensive Series. What's New Series Version 9 Comprehensive Series MicroStrategy Distribution Services Online Key Features Distribution Services for End Users Administering Subscriptions in Web Configuring Distribution Services Monitoring

More information

SWISSBOX REVISITING THE DATA PROCESSING SOFTWARE STACK

SWISSBOX REVISITING THE DATA PROCESSING SOFTWARE STACK 3/2/2011 SWISSBOX REVISITING THE DATA PROCESSING SOFTWARE STACK Systems Group Dept. of Computer Science ETH Zürich, Switzerland SwissBox Humboldt University Dec. 2010 Systems Group = www.systems.ethz.ch

More information

Understanding Neo4j Scalability

Understanding Neo4j Scalability Understanding Neo4j Scalability David Montag January 2013 Understanding Neo4j Scalability Scalability means different things to different people. Common traits associated include: 1. Redundancy in the

More information

Perfmon counters for Enterprise MOSS

Perfmon counters for Enterprise MOSS Perfmon counters for Enterprise MOSS # Counter What does it measure or can tell us Threshold [Action taken if] Notes PROCESSOR RELATED COUNTERS 1 Processor(_Total)\% Measures average processor utilization

More information

Using Microsoft Performance Monitor. Guide

Using Microsoft Performance Monitor. Guide Using Microsoft Performance Monitor Guide December 2005 The information contained in this document represents the current view of Compulink Management Center, Inc on the issues discussed as of the date

More information

MS 10978A Introduction to Azure for Developers

MS 10978A Introduction to Azure for Developers MS 10978A Introduction to Azure for Developers Description: Days: 5 Prerequisites: This course offers students the opportunity to learn about Microsoft Azure development by taking an existing ASP.NET MVC

More information

Using Ad-Hoc Reporting

Using Ad-Hoc Reporting Using Ad-Hoc Reporting The purpose of this guide is to explain how the Ad-hoc reporting function can be used to produce Management Information from client and product data held in the Key. The guide will

More information

Data warehousing with PostgreSQL

Data warehousing with PostgreSQL Data warehousing with PostgreSQL Gabriele Bartolini http://www.2ndquadrant.it/ European PostgreSQL Day 2009 6 November, ParisTech Telecom, Paris, France Audience

More information

Geeks with...sql Monitor

Geeks with...sql Monitor Case Study Geeks with...sql Monitor How Geekswithblogs.net uses SQL Monitor to look after their servers and keep users around the world happy. Introducing Geekswithblogs.net Problems faced How SQL Monitor

More information

Clusters in the Cloud

Clusters in the Cloud Clusters in the Cloud Dr. Paul Coddington, Deputy Director Dr. Shunde Zhang, Compu:ng Specialist eresearch SA October 2014 Use Cases Make the cloud easier to use for compute jobs Par:cularly for users

More information

Release Notes. Medtech32. Medtech Fax Solution (for Windows 7 and later)

Release Notes. Medtech32. Medtech Fax Solution (for Windows 7 and later) Release Notes Medtech32 Medtech Fax Solution (for Windows 7 and later) (June 2014) IMPORTANT NOTE Medtech recommends that all Medtech upgrades and database back-up and restore processes are performed by

More information

Preview of Oracle Database 12c In-Memory Option. Copyright 2013, Oracle and/or its affiliates. All rights reserved.

Preview of Oracle Database 12c In-Memory Option. Copyright 2013, Oracle and/or its affiliates. All rights reserved. Preview of Oracle Database 12c In-Memory Option 1 The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any

More information

An Oracle White Paper November 2010. Oracle Business Intelligence Standard Edition One 11g

An Oracle White Paper November 2010. Oracle Business Intelligence Standard Edition One 11g An Oracle White Paper November 2010 Oracle Business Intelligence Standard Edition One 11g Introduction Oracle Business Intelligence Standard Edition One is a complete, integrated BI system designed for

More information

4 Channel 6-Port SATA 6Gb/s PCIe RAID Host Card

4 Channel 6-Port SATA 6Gb/s PCIe RAID Host Card 4 Channel 6-Port SATA 6Gb/s PCIe RAID Host Card User Manual Model: UGT-ST644R All brand names and trademarks are properties of their respective owners www.vantecusa.com Contents: Chapter 1: Introduction...

More information

How To Build A Cloud Based Data Hub For A Networked Network (Networking) System (Network)

How To Build A Cloud Based Data Hub For A Networked Network (Networking) System (Network) The Versatile Content Distribution System Highly Efficient Content Distribution The SkyScraper system from Triveni Digital is a highly convenient and efficient platform for content distribution via any

More information

Memory Database Application in the Processing of Huge Amounts of Data Daqiang Xiao 1, Qi Qian 2, Jianhua Yang 3, Guang Chen 4

Memory Database Application in the Processing of Huge Amounts of Data Daqiang Xiao 1, Qi Qian 2, Jianhua Yang 3, Guang Chen 4 5th International Conference on Advanced Materials and Computer Science (ICAMCS 2016) Memory Database Application in the Processing of Huge Amounts of Data Daqiang Xiao 1, Qi Qian 2, Jianhua Yang 3, Guang

More information

How to Setup SQL Server Replication

How to Setup SQL Server Replication Introduction This document describes a scenario how to setup the Transactional SQL Server Replication. Before we proceed for Replication setup you can read brief note about Understanding of Replication

More information

Configuring Security Features of Session Recording

Configuring Security Features of Session Recording Configuring Security Features of Session Recording Summary This article provides information about the security features of Citrix Session Recording and outlines the process of configuring Session Recording

More information

W H I T E P A P E R : T E C H N I C A L. Understanding and Configuring Symantec Endpoint Protection Group Update Providers

W H I T E P A P E R : T E C H N I C A L. Understanding and Configuring Symantec Endpoint Protection Group Update Providers W H I T E P A P E R : T E C H N I C A L Understanding and Configuring Symantec Endpoint Protection Group Update Providers Martial Richard, Technical Field Enablement Manager Table of Contents Content Introduction...

More information

Getting Started with Clearlogin A Guide for Administrators V1.01

Getting Started with Clearlogin A Guide for Administrators V1.01 Getting Started with Clearlogin A Guide for Administrators V1.01 Clearlogin makes secure access to the cloud easy for users, administrators, and developers. The following guide explains the functionality

More information