专业英语论文云计算和数字图书馆.docx

专业英语论文云计算和数字图书馆.docx

《专业英语论文云计算和数字图书馆.docx》由会员分享，可在线阅读，更多相关《专业英语论文云计算和数字图书馆.docx（14页珍藏版）》请在冰豆网上搜索。

专业英语论文云计算和数字图书馆

CloudComputingTechnologyandDigitalLibrary

Abstract：

Asarapidlydevelopingnewinformationtechnologies，Cloudcomputinghasbeenappliedinmanyfieldsathomeandabroad，itwillalsohelptheconstructionanddevelopmentoftheDigitalLibrarybringingtheimpactwhichcannotbeignored.Onthebasisofintroducingtheprincipleofcloudcomputing,discusstheinfluenceofCloudcomputingontheUniversityDigitalLibraryandsomeproblemswhichshouldbepayattentionto.Thearticlefirstlyanalyzestheconceptofcloudcomputing,theconceptanddifficultiesofdigitallibrary,andthecloudontheDigitalLibraryfromtheinformationresources,informationusers,informationworkersandinformationfacilities.Intheend,itprospectsthedevelopmentforthefuture.Cloudcomputinghasbroughtusanewperspectivetolookatthecurrentresource-sharingproblem,cloudcomputingcanbeappliedtodigitallibraryresourcestoimproveinformationsharingcapabilitiesandimproveresourceutilization.Ourresultsshowthatacloudimplementationmaybeafeasiblealternativeforitscontinuedoperationandgrowth.

KeyWords:

CloudComputing,DigitalLibrary,CloudComputingApplications

Introductionsection

WiththepopularityofInternetandupgradingnetworktechnology,Ithasquietlyenteredourlivesbecomingapartofpeople'slives,changingourwayoflife.Weusedlearnandacquireknowledgeintheclassroom,booksandlibraries.Nowwecanquerytheinformationwewantatcomfortablehome.Computertechnologybroughtfreshbloodtotheoldlibraryscience,alsobroughtinevitableproblems.Today,theshortcomingsmaliciousattackorasingleservermodewillleadirreparabledamagetothelostoflibraries’importantinformation.Nowwewillleveragecloudcomputingtechnologytohelpavoidlossofdigitallibraries,andprovidegoodserviceforreadersbetterandmorehumaneway.

DigitalLibraryisathingproducedcertainlywiththedevelopmentoftheInternet,theadvantageisobviouscomparedwiththetraditionallibrary.ItmakestheInformationstoragespacegreatlyreducedwithoutdamage,andInformationretrievalmoreconvenient.achievingthetargetofremotetransmissionofinformationandinformationSharing,soallcollegesanduniversitiesareactivelybuildingthedigitallibrary.

However,CurrentdigitalLibraryisfacedwithaseriesofproblemssuchassecurityrisks,difficultiesinsystemmaintenanceandlargecapitalinvestment,leadingtothedevelopmentofdigitalLibraryexperienceddifficulties.Technologyisadvancingrapidly,nowadays,theemergenceofcloudcomputingtechnologyhasbroughtnewhopeandvitalitytotheDigitalLibrary,itwillhelpdigitallibraryoutofdilemma,andprovidethenecessaryservicesforreadersbetter.

Part1CloudComputingOverview

Cloudcomputingisadistributedprocessing,parallelprocessingandgridcomputingdevelopment,itaimedtointegratenumbersoflower-costcomputingentitiesintoonesystemwithasuper-computingpowerthroughthenetwork.Intheremotedatacenter,tensofthousandsofcomputersandserversconnectedintoacomputergoestowork,makingtherelatedcalculationdistributedinthelargenumbersofdistributedcomputerratherthanthelocalcomputer.Therefore,ThepurposeofthecloudcomputingisdependonB/SArchitecture,transferringthecalculationofthepressurefromtheclienttotheserver.Usersonlyneedtopaylittleservicefeecanbeconfiguredcomputers,laptops,cellphones,etcaccessingtothe"cloud"center,operationingaccordingtotheirownneeds.Theimplementationofcloudcomputingwillbringusmorecomputingpower,lowercostsandpeople-orientedservice.

Cloudcomputingasarelativelynewconcept,onceapprancedcanbepursuitofindustryandacademia.Inparticular,since2007,includingMicrosoft,IBM,GOOGLE,etc.,themajorcompanyhassuccessfullylaunchedacloudcomputingservice,whichcanprovideuserswithonlinedatastorage,informationtransfer,largeamountsofdataparallelprocessing,dataindexingandqueryservices,andbroughtgreatbenefits.InChina,thedevelopmentofcloudcomputingisveryrapidly,anumberofCloudComputingCenterhasestablished,developedavarietyofcloudcomputingproducts,serveringinvariousfieldsandlevelsofusers.Nationallibrarycommunityarecloselywatchingthecloud,LibrarySocietyofChinaandtheShanghaiSocietyforLibraryScienceAcademicCommitteeheldsymposiumseveraltimesaboutthisrelatedtopicsof"cloudcomputingandlibrary".

1.1Thecharacteristicsofcloudcomputing

1,Largescale."Cloud"hasaconsiderablesize.Googlecloudcomputingalreadyhasmorethan100millionservers.Amazon,IBM,Microsoft,Yahooandothers’"cloud"hashundredsofthousandsofservers.Privatecompaniestypicallyhavehundredsofthousandsofcloudserver."Cloud"cangiveusersanunprecedentedcomputingpower.

2,Virtualization.Cloudcomputingallowsusersinanylocation,usingavarietyofterminalaccessingtoapplicationservices.Alltheresourcesrequestedarefromthe"cloud",insteadoffixedphysicalentities.Applicationisrunninginthesomewhereof"cloud".butinfactusersdonotunderstandandworryaboutthespecificlocationofrunningapplications.Onlyneedalaptoporamobilephone,webservicescanachieveallweneed,eventhetaskofSupercomputing.

3,Highreliability."Cloud"usesthemeasuressuchasmultiplecopiesofdatafaulttolerance,computingnodesareinterchangeablewiththestructuretoprotectthereliabilityofservices.makingcloudcomputingmorereliablethanusingthelocalcomputer.

4,Universal.Cloudcomputingisnotthespecificapplication,ever-changingapplicationcanbeconstructedunderthesupportoftheof"cloud",witha"cloud"cansupportdifferentapplicationsrunning.

5,Highscalability."Cloud"’ssizeisscalabledynamically,meetingtheneedsofthescaleofgrowthofApplicationsandusers.

6,On-demandservice."Cloud"isahugeresourcepool,Youshouldbuyondemand.Cloudcanbebilledlikerunningwater,electricityandgas.

7,Extremelycheap.Verylow-costnodescanbeusedtoformthecloudsasthespecialfaulttolerancemeasuresof"cloud"."Cloud"’scentralizedandautomatedmanagementmakelargenumbersofbusinesswithouttheburdenofthehighcostofdatacentermanagement."Cloud"oftheuniversalresourceutilizationimprovesignificantlycomparedtotraditionalsystems.Souserscanfullyenjoythelow-costadvantageof"cloud",oftenonlytakesafewhundreddollarsorafewdaystocompletethetaskusedtotaketensofthousandsofdollars,severalmonthstocomplete.

Part2Backgroundandrelatedwork

Cloudcomputingandinfrastructurehavebeendiscussedinthecontextofinformationretrievalsystemswhicharerelatedtogridcomputinganddistributedcomputing.Inthesediscussions,thefocushasbeeneitheroncloudcomputingfordatastorageinfrastructureorthecomputeinfrastructure.Adiscussionofcloudcomputingforscalabilityinpreprocessing,harvesting,transformationandstorageisprovidedbyinthecontextofcrawlingtheWebwithSindice.Inthefollowingsubsections,webrieflyoverviewthearchitectureoftheSeerSuiteapplicationframeworkanditsdeploymentasCiteSeerx.Thissectionismeanttoprovideanunderstandingoftheapplicationrequirementsthatarecrucialforunderstandingthechallengesandtheneedforinfrastructurevirtualizationandabstraction.

Figure1providesanoverviewofSeerSuitearchitecture.SeerSuiteincludescomponentsofbothwebsearchsystemsanddigitallibraries.SeerSuiteconsistsofseveralcomponentsandserviceslooselycoupledtogetherwithRESTandSOAinterfaces.WegroupcomponentsoftheframeworkintoWebapplication,datastorage,extraction,ingestion,andmaintenancesystems.Thesecomponentscanfunctionasstandaloneapplications.

Fig.1.SeerSuiteArchitecture

1）WebApplication:

TheprimaryinterfacebetweenusersandSeerSuiteistheWebapplication.Userscansearch,browse,andtraversethecollectionthroughthisinterface.QueriesinSeerSuitearesupportedthroughaninterfaceoverRESTtoanindex,aninvertedfiledatastructurewhichallowsforfastandaccurateretrieval.Resultsobtainedfromtheindexareprocessedbytheapplicationandwiththedatabasedisplayssearchresults.Userscanviewdocumentmetadatafromsummarypages.Thesesummarypagesallowausertodownloadacachedcopyorbedirectedtoanothersource,viewcitationsfromthedocument,viewtagscreatedbyusers,and

trackchangestothedocument.

Theautonomouscitationindexingfeatureenablesuserstorankdocumentsbasedoncitationstothedocumentinsidethecollectionaswellasbrowsethecollectionthroughcitationlinks.Thesefeaturesrequirethecitationgraphstructureofthecollectiontobestoredinthedatabase.Inadditiontoproviding

useraccessthroughwebpages,SeerSuitesupportsanOpenArchivesandanApplicationProgrammingInterface,whichprovideprogrammaticaccesstothedatabase.CiteSeerxsupportsmyCiteSeer,apersonalizationportalwhichwithsupportofthedatabase,allowsuserstostorequeries,documentportfolios,tagdocumentsandmakecorrectionstodocumentmetadata.

2）CrawlingandMetadataExtraction:

ThedocumentacquisitionprocessofSeerSuiteincludestwomajortasks.Thefirstistoobtaindocumentsrelevanttothecollection.Afocusedcrawlertraversesthewebtofinddocumentsrelevanttoaparticulartopic.InSeerSuite,afocusedcrawlerscanswebpagesfromacrawllistandfetchesdocuments,mostlyPDF,embeddedinthesepages.Beforemetadataextractioncantakeplace,theincomingdocumentsarefilteredandconvertedinto