Getting to Know Relational Databases【外文翻译】.doc

1、1本科毕业论文外文翻译外文文献原文GETTINGTOKNOWRELATIONALDATABASESMATERIALSOURCEBEGINNINGC2008DATABASES，FROMNOVICETOPROFESSIONALNOTESINAPRESSAUTHORVIDYAVRATAGARWALANDJAMESHUDDLESTONNOWTHATYOUHAVEGOTTENTOKNOWTHETOOLSYOULLUSEINTHISBOOK,WELLSTEPBACKABITTOGIVEYOUABRIEFINTRODUCTIONTOTHEIMPORTANTCONCEPTSOFTHEPCDATABASEWOR

2、LDBEFOREDIVINGINTOTHEEXAMPLESWHATISADATABASEINVERYSIMPLETERMS,ADATABASEISACOLLECTIONOFSTRUCTUREDINFORMATIONDATABASESAREDESIGNEDSPECIFICALLYTOMANAGELARGEBODIESOFINFORMATION,ANDTHEYSTOREDATAINANORGANIZEDANDSTRUCTUREDMANNERTHATMAKESITEASYFORUSERSTOMANAGEANDRETRIEVETHATDATAWHENREQUIREDADATABASEMANAGEMEN

3、TSYSTEMDBMSISASOFTWAREPROGRAMTHATENABLESUSERSTOCREATEANDMAINTAINDATABASESADBMSALSOALLOWSUSERSTOWRITEQUERIESFORANINDIVIDUALDATABASETOPERFORMREQUIREDACTIONSLIKERETRIEVINGDATA,MODIFYINGDATA,DELETINGDATA,ANDSOFORTHDBMSSSUPPORTTABLESAKARELATIONSORENTITIESTOSTOREDATAINROWSAKARECORDSORTUPLESANDCOLUMNSAKAFI

4、ELDSORATTRIBUTES,SIMILARTOHOWDATAAPPEARSINASPREADSHEETAPPLICATIONARELATIONALDATABASEMANAGEMENTSYSTEM,ORRDBMS,ISATYPEOFDBMSTHATSTORESINFORMATIONINTHEFORMOFRELATEDTABLESRDBMSISBASEDONTHERELATIONALMODELCHOOSINGBETWEENASPREADSHEETANDADATABASEIFDATABASESAREMUCHLIKESPREADSHEETS,WHYDOPEOPLESTILLUSEDATABASE

5、APPLICATIONSADATABASEISDESIGNEDTOPERFORMTHEFOLLOWINGACTIONSINANEASIERANDMOREPRODUCTIVEMANNERTHANASPREADSHEETAPPLICATIONWOULDREQUIRERETRIEVEALLRECORDSTHATMATCHPARTICULARCRITERIAUPDATEORMODIFYACOMPLETESETOFRECORDSATONETIME2EXTRACTVALUESFROMRECORDSDISTRIBUTEDAMONGMULTIPLETABLESWHYUSEADATABASEFOLLOWINGA

6、RESOMEOFTHEREASONSWEUSEDATABASESCOMPACTNESSDATABASESHELPINMAINTAININGLARGEAMOUNTSOFDATA,ANDTHUSCOMPLETELYREPLACEVOLUMINOUSPAPERFILESSPEEDSEARCHESFORAPARTICULARPIECEOFDATAORINFORMATIONINADATABASEAREMUCHFASTERTHANSORTINGTHROUGHPILESOFPAPERLESSDRUDGERYMAINTAININGFILESBYHANDISDULLWORKUSINGADATABASECOMPL

7、ETELYELIMINATESSUCHMAINTENANCECURRENCYDATABASESYSTEMSCANEASILYBEUPDATEDANDSOPROVIDEACCURATEINFORMATIONALLTHETIMEANDONDEMANDBENEFITSOFUSINGARELATIONALDATABASEMANAGEMENTSYSTEMRDBMSSOFFERVARIOUSBENEFITSBYCONTROLLINGTHEFOLLOWINGREDUNDANCYRDBMSSPREVENTHAVINGMULTIPLEDUPLICATECOPIESOFTHESAMEDATA,WHICHTAKES

8、UPDISKSPACEUNNECESSARILYINCONSISTENCYEACHREDUNDANTSETOFDATAMAYNOLONGERAGREEWITHOTHERSETSOFTHESAMEDATAWHENANRDBMSREMOVESREDUNDANCY,INCONSISTENCYCANNOTOCCURDATAINTEGRITYDATAVALUESSTOREDINTHEDATABASEMUSTSATISFYCERTAINTYPESOFCONSISTENCYCONSTRAINTSDATAATOMICITYINEVENTOFAFAILURE,DATAISRESTOREDTOTHECONSIST

9、ENTSTATEITEXISTEDINPRIORTOTHEFAILUREFOREXAMPLE,FUNDTRANSFERACTIVITYMUSTBEATOMICACCESSANOMALIESRDBMSSPREVENTMORETHANONEUSERFROMUPDATINGTHESAMEDATASIMULTANEOUSLYSUCHCONCURRENTUPDATESMAYRESULTININCONSISTENTDATADATASECURITYNOTEVERYUSEROFTHEDATABASESYSTEMSHOULDBEABLETOACCESSALLTHEDATASECURITYREFERSTOTHEP

10、ROTECTIONOFDATAAGAINSTANYUNAUTHORIZEDACCESSTRANSACTIONPROCESSINGATRANSACTIONISASEQUENCEOFDATABASEOPERATIONSTHATREPRESENTSALOGICALUNITOFWORKINRDBMSS,ATRANSACTIONEITHERCOMMITSALLTHECHANGESORROLLSBACKALLTHEACTIONSPERFORMEDUNTILTHEPOINTATWHICHFAILUREOCCURREDRECOVERYRECOVERYFEATURESENSURETHATDATAISREORGA

11、NIZEDINTOACONSISTENT3STATEAFTERATRANSACTIONFAILSSTORAGEMANAGEMENTRDBMSSPROVIDEAMECHANISMFORDATASTORAGEMANAGEMENTTHEINTERNALSCHEMADEFINESHOWDATASHOULDBESTOREDCOMPARINGDESKTOPANDSERVERRDBMSSYSTEMSINTHEINDUSTRYTODAY,WEMAINLYWORKWITHTWOTYPESOFDATABASESDESKTOPDATABASESANDSERVERDATABASESHERE,WELLGIVEYOUAB

12、RIEFLOOKATEACHOFTHEMDESKTOPDATABASESDESKTOPDATABASESAREDESIGNEDTOSERVEALIMITEDNUMBEROFUSERSANDRUNONDESKTOPPCS,ANDTHEYOFFERALESSEXPENSIVESOLUTIONWHEREVERADATABASEISREQUIREDCHANCESAREYOUHAVEWORKEDWITHADESKTOPDATABASEPROGRAMMICROSOFTSQLSERVEREXPRESS,MICROSOFTACCESS,MICROSOFTFOXPRO,FILEMAKERPRO,PARADOX,

13、ANDLOTUSREPRESENTAWIDERANGEOFDESKTOPDATABASESOLUTIONSDESKTOPDATABASESDIFFERFROMSERVERDATABASESINTHEFOLLOWINGWAYSLESSEXPENSIVEMOSTDESKTOPSOLUTIONSAREAVAILABLEFORJUSTAFEWHUNDREDDOLLARSINFACT,IFYOUOWNALICENSEDVERSIONOFMICROSOFTOFFICEPROFESSIONAL,YOUREALREADYALICENSEDOWNEROFMICROSOFTACCESS,WHICHISONEOFT

14、HEMOSTCOMMONLYANDWIDELYUSEDDESKTOPDATABASEPROGRAMSAROUNDUSERFRIENDLYDESKTOPDATABASESAREQUITEUSERFRIENDLYANDEASYTOWORKWITH,ASTHEYDONOTREQUIRECOMPLEXSQLQUERIESTOPERFORMDATABASEOPERATIONSALTHOUGHSOMEDESKTOPDATABASESALSOSUPPORTSQLSYNTAXIFYOUWOULDLIKETOCODEDESKTOPDATABASESGENERALLYOFFERANEASYTOUSEGRAPHIC

15、ALUSERINTERFACESERVERDATABASESSERVERDATABASESARESPECIFICALLYDESIGNEDTOSERVEMULTIPLEUSERSATATIMEANDOFFERFEATURESTHATALLOWYOUTOMANAGELARGEAMOUNTSOFDATAVERYEFFICIENTLYBYSERVINGMULTIPLEUSERREQUESTSSIMULTANEOUSLYWELLKNOWNEXAMPLESOFSERVERDATABASESINCLUDEMICROSOFTSQLSERVER,ORACLE,SYBASE,ANDDB2HEREARESOMEOT

16、HERCHARACTERISTICSTHATDIFFERENTIATESERVERDATABASESFROMTHEIRDESKTOPCOUNTERPARTSFLEXIBILITYSERVERDATABASESAREDESIGNEDTOBEVERYFLEXIBLETOSUPPORTMULTIPLEPLATFORMS,RESPONDTOREQUESTSCOMINGFROMMULTIPLEDATABASEUSERS,AND4PERFORMANYDATABASEMANAGEMENTTASKWITHOPTIMUMSPEEDAVAILABILITYSERVERDATABASESAREINTENDEDFOR

17、ENTERPRISES,ANDSOTHEYNEEDTOBEAVAILABLE24/7TOBEAVAILABLEALLTHETIME,SERVERDATABASESCOMEWITHSOMEHIGHAVAILABILITYFEATURES,SUCHASMIRRORINGANDLOGSHIPPINGPERFORMANCESERVERDATABASESUSUALLYHAVEHUGEHARDWARESUPPORT,ANDSOSERVERSRUNNINGTHESEDATABASESHAVELARGEAMOUNTSOFRAMANDMULTIPLECPUS,ANDTHISISWHYSERVERDATABASE

18、SSUPPORTRICHINFRASTRUCTUREANDGIVEOPTIMUMPERFORMANCESCALABILITYTHISPROPERTYALLOWSASERVERDATABASETOEXPANDITSABILITYTOPROCESSANDSTORERECORDSEVENIFITHASGROWNTREMENDOUSLYTHEDATABASELIFECYCLETHEDATABASELIFECYCLEDEFINESTHECOMPLETEPROCESSFROMCONCEPTIONTOIMPLEMENTATIONTHEENTIREDEVELOPMENTANDIMPLEMENTATIONPRO

19、CESSOFTHISCYCLECANBEDIVIDEDINTOSMALLPHASESONLYAFTERTHECOMPLETIONOFEACHPHASECANYOUMOVEONTOTHENEXTPHASE,ANDTHISISTHEWAYYOUBUILDYOURDATABASEBLOCKBYBLOCKBEFOREGETTINGINTOTHEDEVELOPMENTOFANYSYSTEM,YOUNEEDTOHAVESTRONGALIFECYCLEMODELTOFOLLOWTHEMODELMUSTHAVEALLTHEPHASESDEFINEDINPROPERSEQUENCE,WHICHWILLHELPT

20、HEDEVELOPMENTTEAMTOBUILDTHESYSTEMWITHFEWERPROBLEMSANDFULLFUNCTIONALITYASEXPECTEDTHEDATABASELIFECYCLECONSISTSOFTHEFOLLOWINGSTAGES,FROMTHEBASICSTEPSINVOLVEDINDESIGNINGAGLOBALSCHEMAOFTHEDATABASETODATABASEIMPLEMENTATIONANDMAINTENANCEREQUIREMENTANALYSISREQUIREMENTSNEEDTOBEDETERMINEDBEFOREYOUCANBEGINDESIG

21、NANDIMPLEMENTATIONTHEREQUIREMENTSCANBEGATHEREDBYINTERVIEWINGBOTHTHEPRODUCERANDTHEUSEROFTHEDATATHISPROCESSHELPSINCREATINGAFORMALREQUIREMENTSPECIFICATIONLOGICALDESIGNAFTERREQUIREMENTGATHERING,DATAANDRELATIONSHIPSNEEDTOBEDEFINEDUSINGACONCEPTUALDATAMODELINGTECHNIQUESUCHASANENTITYRELATIONSHIPERDIAGRAMPHY

22、SICALDESIGNONCETHELOGICALDESIGNISINPLACE,THENEXTSTEPISTOPRODUCE5THEPHYSICALSTRUCTUREFORTHEDATABASETHEPHYSICALDESIGNPHASEINVOLVESTABLECREATIONANDSELECTIONOFINDEXESDATABASEIMPLEMENTATIONONCETHEDESIGNISCOMPLETED,THEDATABASECANBECREATEDTHROUGHIMPLEMENTATIONOFFORMALSCHEMAUSINGTHEDATADEFINITIONLANGUAGEDDL

23、OFTHERDBMSDATAMODIFICATIONDATAMODIFICATIONLANGUAGEDMLCANBEUSEDTOQUERYANDUPDATETHEDATABASEASWELLASSETUPINDEXESANDESTABLISHCONSTRAINTSSUCHASREFERENTIALINTEGRITYDATABASEMONITORINGASTHEDATABASEBEGINSOPERATION,MONITORINGINDICATESWHETHERPERFORMANCEREQUIREMENTSAREBEINGMETIFTHEYARENOT,MODIFICATIONSSHOULDBEM

24、ADETOIMPROVEDATABASEPERFORMANCETHUSTHEDATABASELIFECYCLECONTINUESWITHMONITORING,REDESIGN,ANDMODIFICATIONMAPPINGCARDINALITIESTABLESARETHEFUNDAMENTALCOMPONENTSOFARELATIONALDATABASEINFACT,BOTHDATAANDRELATIONSHIPSARESTOREDSIMPLYASDATAINTABLESTABLESARECOMPOSEDOFROWSANDCOLUMNSEACHCOLUMNREPRESENTSAPIECEOFIN

25、FORMATIONMAPPINGCARDINALITIES,ORCARDINALITYRATIOS,EXPRESSTHENUMBEROFENTITIESTOWHICHANOTHERENTITYCANBEASSOCIATEDVIAARELATIONSHIPSETCARDINALITYREFERSTOTHEUNIQUENESSOFDATAVALUESCONTAINEDINAPARTICULARCOLUMNOFADATABASETABLETHETERMRELATIONALDATABASEREFERSTOTHEFACTTHATDIFFERENTTABLESQUITEOFTENCONTAINRELATE

26、DDATAFOREXAMPLE,ONESALESREPINACOMPANYMAYTAKEMANYORDERS,WHICHWEREPLACEDBYMANYCUSTOMERSTHEPRODUCTSORDEREDMAYCOMEFROMDIFFERENTSUPPLIERS,ANDCHANCESARETHATEACHSUPPLIERCANSUPPLYMORETHANONEPRODUCTALLOFTHESERELATIONSHIPSEXISTINALMOSTEVERYDATABASEANDCANBECLASSIFIEDASFOLLOWSONETOONE11FOREACHROWINTABLEA,THEREI

27、SATMOSTONLYONERELATEDROWINTABLEB,ANDVICEVERSATHISRELATIONSHIPISTYPICALLYUSEDTOSEPARATEDATABYFREQUENCYOFUSETOOPTIMALLYORGANIZEDATAPHYSICALLYFOREXAMPLE,ONEDEPARTMENTCANHAVEONLYONEDEPARTMENTHEADONETOMANY1MFOREACHROWINTABLEA,THERECANBEZEROORMORERELATED6ROWSINTABLEB,BUTFOREACHROWINTABLEB,THEREISATMOSTONE

28、ROWINTABLEATHISISTHEMOSTCOMMONRELATIONSHIPANEXAMPLEOFAONETOMANYRELATIONSHIPOFTABLESINNORTHWINDISSHOWNINFIGURE31NOTETHECUSTOMERSTABLEHASACUSTOMERIDFIELDASTHEPRIMARYKEYINDICATEDBYTHEKEYSYMBOLONTHELEFT,WHICHHASARELATIONWITHTHECUSTOMERIDFIELDOFTHEORDERSTABLECUSTOMERIDISCONSIDEREDAFOREIGNKEYINTHEORDERSTA

29、BLETHELINKSHOWNBETWEENTHECUSTOMERSANDORDERSTABLESINDICATESAONETOMANYRELATIONSHIP,ASMANYORDERSCANBELONGTOONECUSTOMERHERE,CUSTOMERSISREFERREDTOASTHEPARENTTABLE,ANDORDERSISTHECHILDTABLEINTHERELATIONSHIPMANYTOMANYMMFOREACHROWINTABLEA,THEREAREZEROORMORERELATEDROWSINTABLEB,ANDVICEVERSAMANYTOMANYRELATIONSH

30、IPSARENOTSOEASYTOACHIEVE,7ANDTHEYREQUIREASPECIALTECHNIQUETOIMPLEMENTTHEMTHISRELATIONSHIPISACTUALLYIMPLEMENTEDINAONEMANYONEFORMAT,SOITREQUIRESATHIRDTABLEOFTENREFERREDTOASAJUNCTIONTABLETOBEINTRODUCEDINBETWEENTHATSERVESASTHEPATHBETWEENTHERELATEDTABLESTHISISAVERYCOMMONRELATIONSHIPANEXAMPLEFROMNORTHWINDI

31、SSHOWNINFIGURE32ANORDERCANHAVEMANYPRODUCTSANDAPRODUCTCANBELONGTOMANYORDERSTHEORDERDETAILSTABLENOTONLYREPRESENTSTHEMMRELATIONSHIP,BUTALSOCONTAINSDATAABOUTEACHPARTICULARORDERPRODUCTCOMBINATIONNOTETHOUGHRELATIONSHIPSAMONGTABLESAREEXTREMELYIMPORTANT,THETERMRELATIONALDATABASEHASNOTHINGTODOWITHTHEMRELATIO

32、NALDATABASESARETOVARYINGEXTENTSBASEDONTHERELATIONALMODELOFDATAINVENTEDBYDREDGARFCODDATIBMINTHE81970SCODDBASEDHISMODELONTHEMATHEMATICALSETTHEORETICCONCEPTOFARELATIONRELATIONSARESETSOFTUPLESTHATCANBEMANIPULATEDWITHAWELLDEFINEDANDWELLBEHAVEDSETOFMATHEMATICALOPERATIONSINFACT,TWOSETSRELATIONALALGEBRAANDR

33、ELATIONALCALCULUSYOUDONTHAVETOKNOWORUNDERSTANDTHEMATHEMATICSTOWORKWITHRELATIONALDATABASES,BUTIFYOUHEARITSAIDTHATADATABASEISRELATIONALBECAUSEIT“RELATESDATA,”YOULLKNOWTHATWHOEVERSAIDITDOESNTUNDERSTANDRELATIONALDATABASESUNDERSTANDINGKEYSTHEKEY,THEWHOLEKEY,ANDNOTHINGBUTTHEKEY,SOHELPMECODDRELATIONSHIPSAR

34、EREPRESENTEDBYDATAINTABLESTOESTABLISHARELATIONSHIPBETWEENTWOTABLES,YOUNEEDTOHAVEDATAINONETABLETHATENABLESYOUTOFINDRELATEDROWSINANOTHERTABLETHATSWHEREKEYSCOMEIN,ANDRDBMSSMAINLYWORKWITHTWOTYPESOFKEYS,ASMENTIONEDEARLIERPRIMARYKEYSANDFOREIGNKEYSAKEYISONEORMORECOLUMNSOFARELATIONTHATISUSEDTOIDENTIFYAROWPR

35、IMARYKEYSAPRIMARYKEYISANATTRIBUTECOLUMNORCOMBINATIONOFATTRIBUTESCOLUMNSWHOSEVALUESUNIQUELYIDENTIFYRECORDSINANENTITYBEFOREYOUCHOOSEAPRIMARYKEYFORANENTITY,ANATTRIBUTEMUSTHAVETHEFOLLOWINGPROPERTIESEACHRECORDOFTHEENTITYMUSTHAVEANOTNULLVALUETHEVALUEMUSTBEUNIQUEFOREACHRECORDENTEREDINTOTHEENTITYTHEVALUESMU

36、STNOTCHANGEORBECOMENULLDURINGTHELIFEOFEACHENTITYINSTANCETHERECANBEONLYONEPRIMARYKEYDEFINEDFORANENTITYBESIDESHELPINGINUNIQUELYIDENTIFYINGARECORD,THEPRIMARYKEYALSOHELPSINSEARCHINGRECORDSASANINDEXAUTOMATICALLYGETSGENERATEDASYOUASSIGNAPRIMARYKEYTOANATTRIBUTEANENTITYWILLHAVEMORETHANONEATTRIBUTETHATCANSER

37、VEASAPRIMARYKEYANYKEYORMINIMUMSETOFKEYSTHATCOULDBEAPRIMARYKEYISCALLEDACANDIDATEKEYONCECANDIDATEKEYSAREIDENTIFIED,CHOOSEONE,ANDONLYONE,PRIMARYKEYFOREACHENTITYSOMETIMESITREQUIRESMORETHANONEATTRIBUTETOUNIQUELYIDENTIFYANENTITYAPRIMARYKEYTHATCONSISTSOFMORETHANONEATTRIBUTEISKNOWNASACOMPOSITEKEYTHERE9CANBE

38、ONLYONEPRIMARYKEYINANENTITY,BUTACOMPOSITEKEYCANHAVEMULTIPLEATTRIBUTESIE,APRIMARYKEYWILLBEDEFINEDONLYONCE,BUTITCANHAVEUPTO16ATTRIBUTESTHEPRIMARYKEYREPRESENTSTHEPARENTENTITYPRIMARYKEYSAREUSUALLYDEFINEDWITHTHEIDENTITYPROPERTY,WHICHALLOWSINSERTIONOFANAUTOINCREMENTEDINTEGERVALUEINTOTHETABLEWHENYOUINSERTA

39、ROWINTOTHETABLEFOREIGNKEYSAFOREIGNKEYISANATTRIBUTETHATCOMPLETESARELATIONSHIPBYIDENTIFYINGTHEPARENTENTITYFOREIGNKEYSPROVIDEAMETHODFORMAINTAININGINTEGRITYINTHEDATACALLEDREFERENTIALINTEGRITYANDFORNAVIGATINGBETWEENDIFFERENTINSTANCESOFANENTITYEVERYRELATIONSHIPINTHEMODELMUSTBESUPPORTEDBYAFOREIGNKEYFOREXAM

40、PLE,INFIGURE31EARLIER,THECUSTOMERSANDORDERSTABLESHAVEAPRIMARYKEYANDFOREIGNKEYRELATIONSHIP,WHERETHEORDERSTABLESCUSTOMERIDFIELDISTHEFOREIGNKEYHAVINGAREFERENCETOTHECUSTOMERIDFIELD,WHICHISTHEPRIMARYKEYOFTHECUSTOMERSTABLEUNDERSTANDINGDATAINTEGRITYDATAINTEGRITYMEANSTHATDATAVALUESINADATABASEARECORRECTANDCO

41、NSISTENTTHEREARETWOASPECTSTODATAINTEGRITYENTITYINTEGRITYANDREFERENTIALINTEGRITYENTITYINTEGRITYWEMENTIONEDPREVIOUSLYIN“PRIMARYKEYS”THATNOPARTOFAPRIMARYKEYCANBENULLTHISISTOGUARANTEETHATPRIMARYKEYVALUESEXISTFORALLROWSTHEREQUIREMENTTHATPRIMARYKEYVALUESEXISTANDTHATTHEYAREUNIQUEISKNOWNASENTITYINTEGRITYEIT

42、HEDBMSENFORCESENTITYINTEGRITYBYNOTALLOWINGOPERATIONSINSERT,UPDATETOPRODUCEANINVALIDPRIMARYKEYANYOPERATIONTHATCREATESADUPLICATEPRIMARYKEYORONECONTAININGNULLSISREJECTEDTHATIS,TOESTABLISHENTITYINTEGRITY,YOUNEEDTODEFINEPRIMARYKEYSSOTHEDBMSCANENFORCETHEIRUNIQUENESSREFERENTIALINTEGRITYONCEARELATIONSHIPISD

43、EFINEDBETWEENTABLESWITHFOREIGNKEYS,THEKEYDATAMUSTBEMANAGEDTOMAINTAINTHECORRECTRELATIONSHIPS,THATIS,TOENFORCEREFERENTIALINTEGRITYRIRIREQUIRESTHATALLFOREIGNKEYVALUESINACHILDTABLEEITHERMATCHPRIMARYKEYVALUESINAPARENTTABLEORIFPERMITTEDBENULLTHISISALSOKNOWNASSATISFYINGAFOREIGNKEYCONSTRAINT10NORMALIZATIONC

44、ONCEPTSNORMALIZATIONISATECHNIQUEFORAVOIDINGPOTENTIALUPDATEANOMALIES,BASICALLYBYMINIMIZINGREDUNDANTDATAINALOGICALDATABASEDESIGNNORMALIZEDDESIGNSAREINASENSE“BETTER”DESIGNSBECAUSETHEYIDEALLYKEEPEACHDATAITEMINONLYONEPLACENORMALIZEDDATABASEDESIGNSUSUALLYREDUCEUPDATEPROCESSINGCOSTSBUTCANMAKEQUERYPROCESSIN

45、GMORECOMPLICATEDTHESETRADEOFFSMUSTBECAREFULLYEVALUATEDINTERMSOFTHEREQUIREDPERFORMANCEPROFILEOFADATABASEOFTEN,ADATABASEDESIGNNEEDSTOBEDENORMALIZEDTOADEQUATELYMEETOPERATIONALNEEDSNORMALIZINGALOGICALDATABASEDESIGNINVOLVESASETOFFORMALPROCESSESTOSEPARATETHEDATAINTOMULTIPLE,RELATEDTABLESTHERESULTOFEACHPRO

46、CESSISREFERREDTOASANORMALFORMFIVENORMALFORMSHAVEBEENIDENTIFIEDINTHEORY,BUTMOSTOFTHETIMETHIRDNORMALFORM3NFISASFARASYOUNEEDTOGOINPRACTICETOBEIN3NF,ARELATIONTHEFORMALTERMFORWHATSQLCALLSATABLEANDTHEPRECISECONCEPTONWHICHTHEMATHEMATICALTHEORYOFNORMALIZATIONRESTSMUSTALREADYBEINSECONDNORMALFORM2NF,AND2NFREQ

47、UIRESARELATIONTOBEINFIRSTNORMALFORM1NFLETSLOOKBRIEFLYATWHATTHESENORMALFORMSMEANFIRSTNORMALFORM1NFINFIRSTNORMALFORM,ALLCOLUMNVALUESARESCALARINOTHERWORDS,THEYHAVEASINGLEVALUETHATCANTBEFURTHERDECOMPOSEDINTERMSOFTHEDATAMODELFOREXAMPLE,ALTHOUGHINDIVIDUALCHARACTERSOFASTRINGCANBEACCESSEDTHROUGHAPROCEDURETH

48、ATDECOMPOSESTHESTRING,ONLYTHEFULLSTRINGISACCESSIBLEBYNAMEINSQL,SO,ASFARASTHEDATAMODELISCONCERNED,THEYARENTPARTOFTHEMODELLIKEWISE,FORAMANAGERSTABLEWITHAMANAGERCOLUMNANDACOLUMNCONTAININGALISTOFEMPLOYEESINEMPLOYEESTABLEWHOWORKFORAGIVENMANAGER,THEMANAGERANDTHELISTWOULDBEACCESSIBLEBYNAME,BUTTHEINDIVIDUAL

49、EMPLOYEESINTHELISTWOULDNTBEALLRELATIONSANDSQLTABLESAREBYDEFINITIONIN1NFSINCETHELOWESTLEVELOFACCESSIBILITYKNOWNASTHETABLESGRANULARITYISTHECOLUMNLEVEL,ANDCOLUMNVALUESARESCALARSINSQLSECONDNORMALFORM2NFSECONDNORMALFORMREQUIRESTHATATTRIBUTESTHEFORMALTERMFORSQLCOLUMNSTHATARENTPARTSOFKEYSBEFUNCTIONALLYDEPENDENTONAKEYTHATUNIQUELYIDENTIFIESTHEMFUNCTIONALDEPENDENCEBASICALLYMEANSTHATFORAGIVENKEY11VALUE,ONLYONEVALUEEXISTSINATABLEFORACOLUMNORSETOFCOLUMNSFOREXAMPLE,IFATABLECONTAINEDEMPLOYEESANDTHEIRTITLES,ANDMORETHANONEEMPLOYEECOULDHAVETHESAMETITLEVERYLIKELY,AKEYTHATUNIQUELYIDENTIFIEDEMPLOYE