《電子與通信工程專業(yè)英語》課件第4章

Unit4ApproachestoOpticalInternetPacketSwitchingNEWWORDSANDPHRASES

NOTES

EXERCISES

參考譯文

EXTENSIVETEXT

1．Introduction

Theconvergenceoftelecommunicationsanddatacommunicationshascausedaparadigmshiftinthenetworkingenvironment.ThemassiveexplosionintrafficgeneratedbytheInternethasdriventhecurrenttrend,withtheInternetProtocol(IP)becomingthedominantprotocolfordatacommunicationsaswellasrepresenting,inthelongerterm,averystrongcandidatefortheconvergenceofdatacommunicationswithtelecommunications.Intandem,thedevelopmentofwavelength-divisionmultiplexed(WDM)techniquesonpoint-to-pointlinkshavebeguntoutilizethemassiveopticalbandwidthofinstalledsingle-modeopticalfibersmoreefficiently.WDMwasinitiallyarapidsolutiontothesevererouteexhaustionproblemsbroughtaboutbyexponentiallyincreasingtraffic.Thefuturedeploymentofopticalcross-connectedWDMtransportnetworks,initiallyforprotectionandbypassbutultimatelymanagingopticallightpathsdynamicallyinmultipleringormesharchitectures,willpotentiallymodifytheroleofthenetworkfunctionalitiesprovidedbythesynchronousopticalnetwork/synchronousdigitalhierarchy(SONET/SDH)layer[1].Thus,IPoverWDMhasbecomeaveryimportantareaofstudy,encompassingawiderangeofsolutionstosupportingpredominantlyIPtrafficoverWDMopticalpaths.MuchresearcheffortfocusesondevelopinganelegantsolutiontothemismatchbetweenthetransmissioncapacitiesofferedbytheWDMopticallayerandtheprocessingpowerofrouters.IProutersperformfourmaintasks:

(1)

Routing:providingnetworkconnectivityinformationthroughroutingtables.

(2)

Forwarding:definingtheoutputofeachincomingpacket(basedontheroutingtables).

(3)

Switching:directingeachpackettotheproperoutput(definedbytheforwardingprocess).

(4)

Buffering:resolvingcontentionbystoringpacketswhenmorethanonewishestogotothesameoutputatonceduetotheunschedulednatureoftheirarrival.

Currently,theforwardingprocessimpliesmajorthroughputlimitations,withthesizeoftheroutingtablesandfrequencyoftheirupdatesbeingmajorissues.Suchproblemsarecurrentlyaddressedandmanaged,butthetimeneededfortablelookupsetsafundamentallimitonrouterthroughput.Muchworkhasconcentratedonthedevelopmentofdatastructuresandalgorithmsforminimizingthelookuptimegivenroutingtableandmemoryspaceconstraints.

Thereisnodoubtthatmassivestrideshavebeenachievedinhigh-throughputrouterdesigns(>1

Tb/s).Nevertheless,despitetheseimpressiveadvances,thereisstillafearthatelectronicswitchingsystemsexhibitlimitedupgradeflexibility.GiventhatWDMallowscheapandeasyincrementalincreasesofthetransmissionbandwidth,frequentupgradesofthetransport-layertransmissioncapacitycanbeenvisagedtomatchincreasingdemand,inturnplacingheavydemandsontheswitchingprocess[2].

2．Opticalpacketswitching

Anumberofotherapproachestoobviateoramelioratetheforwardingbottleneckarebeingresearched,utilizingopticalpacketorbursttechniques,perhapswithelectronicbuffering,toimplementIPoverWDM.However,thestrategydetailedhereimplementscontentionresolution(buffering)directlyintheopticaldomaintoyieldWDMopticalpacketswitching.Theobjectiveistoshiftthebulkoftheswitchingburdenintotheopticaldomain,permittingcompatiblescalingoftheswitchingcapabilitywithWDMtransmissioncapacity.Thusfar,thisstrategyhasassumedahybridsolution,achievingdecouplingbetweenthethroughputandtherouting/forwardingprocesses.Transmissionandswitchingareexecutedintheopticaldomain,whileroutingandforwardingarecarriedoutelectronically,wheretherelativelycomplexpacketheaderprocessingoccursindependentoftheopticalpayload[3].ThisdecouplingeffectivelypermitstheopticalpacketlayertosupportarangeofnetworkprotocolswhileharnessingthepowerofWDMtransmission.However,itmustbenotedthatthisisalsochanging,withtherecentdemonstrationofrudimentaryheaderprocessingfunctionsdirectlyintheopticaldomain.Theserelievesomeoftheburdenplacedonelectronicprocessing,therebyreducingcontrolsignalsetuptimeandmanaginglatencymoreeffectively.

Withanextensiveopticalpacketlayer,theinterfacetoIPandotherprotocolsiscrucial.Encapsulation,theadditionofdeliveryinformationtothedatabytheopticalpacketlayer,willoccuratinterworkingunits(IWUs)ateachinterfacetotheelectronicclientlayer.EncapsulationpermitsarangeofprotocolssuchasIPandasynchronoustransfermode(ATM)tobemappedintotheopticalpayloads,whichmaybeofeitherfixedorvariableduration.IPhidesthecomplexityofthephysicallayer(includingopticalpacketswitching),providingaunifiedinterfacetohigherlayers,regardlessoftheunderlyingnetworktype.Inadditiontoencapsulation,theIWUscreateheadersforproperroutingwithintheopticalpackerlayer,andmultiplextrafficfromdifferentinputlinksforonwardtransmissioninopticalpacketsforthesamedestination,ensuringanentirelyopticalend-to-endconnectionpath.Opticalpacketsprovideafurthermultiplexingtier,allowingtheaggregationoftrafficflowspriortotransmissionovertheopticallayer,andalsopotentiallyobviatingtheneedforSDHasanadaptationlayerforIPtrafficonWDMlinks.Opticalpacketnetworkthereforeoffersapotentialsolutiontoprovidingbothconnectionlessandconnection-orientednetworkingcapacities,flexibleintermsofbandwidthmanagementandfuture-proofwithregardtobandwidthgrowth.NostandardsexistyetformappingprotocolssuchasIPandATMintotheopticalpacketlayer.

Ashintedabove,therearetwoprincipalapproachestoopticalpacketswitching,bothwithapplicationstotheInternet:

(1)

Employingfixed-lengthopticalpackets,withmanycorrespondingtooneIPdatagram,requiringIWUstofragmentandreassemblethepacketseitherattheedgesofthelayerorontheinputsandoutputsoftheswitch.

(2)

Employingavariable-lengthopticalpacketforeachIPdatagram.

Mostreportedresearchtodatausesfixed-durationopticalpackets.Hence,forthepurposesofthisarticle(inordertoillustratethefunctionalityofanopticalpacket-switchedlayer),thebulkofthesubsequentmaterialwillbeconfinedtofixed-lengthpacketswhereboththeheaderandpayloadareencodedonthesamewavelength.Itisassumedthatthedestinationswitchoutputforeachpacketisderivedfromtheheaderafteropto-electronicconversion;theheadermaythusbeatalowerbitratetoallowitselectronicmanipulation.Duetothenatureofopticalbuffering,thepayloaddurationisfixed,whateveritscontent;thenetworkthroughputisproportionaltopayloadbitratewhichmayvaryfrom10Gb/sandup,witheasyupgradecapability.Aswillalsobecomeapparent,thewavelengthdimensioniscrucialnotonlyfortransmissioncapacitybutalsoinexecutingpracticalcontentionresolution.Thearticlewillendwithsomeforward-lookingconceptsaddressingtherequirementstoswitchandbuffervariable-lengthopticalpackets.

3．Thedesignofopticalpacketswitches

Agenericopticalpacket-switchednodestructureconsistsofthreesubblocks(Fig4.1):

(1)

Aninputinterfaceconsistingofan(optical)synchronizerwhichalignsincomingpacketsinrealtimeagainstaclock.

(2)

Aswitchingcorewhichroutesthepacketstotheirproperoutputsandexecutescontentionresolution.

(3)

Anoutputinterfacewhichinsertsanewheaderandmayhavetoregeneratethedata.

Fig4.1Agenericopticalpacket-switchednodestructure

Packetformatisafundamentalconsiderationinanypackettransmissionsystem,definedbytherequirementsofthelayerwithrespecttonetworkfunctionalities,andisalsocrucialforopticaldomainimplementations[4].Forexample,considerthepacketformatdefinedbytheKEOPSproject,uponconsiderationofthedelay-throughputperformanceoftheopticalpacket-switchedlayerunderdifferenttrafficflowsaswellasconsiderationofthenodeandnetworkroutingrequirements(Fig4.2).Throughoutitisassumedthattimeisdividedintoequaltimeslots,eachcontainingoneopticalpacket,andthepayloadmaycontaindatafrom622Mb/sto10Gb/s.

Fig4.2Anexampleofanopticalpacketformat(asdefinedbyKEOPS)

Insummary,theheaderfieldis14bytes:8forroutinginformation;3foridentificationofpayloadtype;flowcontrolinformation,packetnumberingforsequenceintegritypreservation,andheadererrorchecking.Theformatalsoindirectlyindicatestheneedforopticalsynchronizersastheinputinterfacetothenode.Itisassumedthatwhenpacketsentertheswitch,theirboundariesarealignedsothateachpacketisalignedwithitstimeslot.Suchsynchronizationisgenerallyarequirementforcorrectswitchoperation,achievinglowpacketloss;approachestoachievingpacketsynchronismattheinputstotheswitchesconstituteaseparatetopicandwillnotbediscussedfurther.Thus,2bytesarerequiredtoaidinthesynchronizationprocessas“tag”uniquelyidentifyingthestartofthepayload.Guardbandsaccountforswitchingtimesoftheconstituentopto-electronicdevicesaswellaspayloadpositionjitter.Thepayloaddurationresultsfromatrade-offbetweentransmissionefficiency(thelongerthepayloadduration,thehighertheefficiency)andthepracticallimitationsonoveralllengthoftheopticalfiberdelaylinebuffers.Atthistime,delaylinesaretheonlyviableapproachtoimplementingcontentionresolutiondirectlyintheopticaldomain.

Anopticalpacketswitchhasthreeprincipalfunctions:switching,buffering,and,optionally,headertranslation.InIP-orientedsystems,thelatterfunctionmaybereplacedbyroutingandforwarding,whichwerediscussedearlier.Switchingensuresthateachpacketemergesatthecorrectoutput,dependingontheinformationcontainedinthepacketheader.Althoughfixed-lengthpacketsarrivingontheinputsmustbesynchronized,thereisnocoordinationbetweenpacketstreamsarrivingondifferentinputs.Henceoneormorepacketsmayarriveduringthesametimeslotondifferentinputswishingtogotothesameoutput.Forthisreason,bufferingisrequired,whereoneormorepacketsarestoredwhileothersaretransmittedtothedesiredoutput.

HeadertranslationrepresentsacentralprocessintheATMtransmissionsystemsstrategy,while,asdiscussedpreciously,routingandforwardingareequallyimportantforIPsystems.Althoughtheseschemesoffergreatfunctionalityandflexibility,theyarenotusedineveryopticalpacket-switchingsystemsincedirectopticalheadertranslationisintheearlystagesofdevelopment.Thus,theheaderisusuallyatalowerbitratethanthepayloadtofacilitateelectronicdecodingandinterpretationofheaderinformation.

NEWWORDSANDPHRASES

convergence n. 會合，會聚

exhaustion n. 耗竭；耗盡；耗損；衰竭

exponentially adv. 指數(shù)地

mesh n. 網(wǎng)狀，網(wǎng)眼

elegant adj. 優(yōu)良的，雅致的；簡潔的

stride n. 進(jìn)展，進(jìn)步；步幅

envisaged vt. 正視

obviatevt. 消除，排除(危險(xiǎn)、障礙等)，回避，預(yù)防

ameliorate vt. 改善，改進(jìn)

bottleneck n. 瓶頸

contention n. 爭奪，爭論

hybrid adj. 雜種的，混合的

decouple vt. 解耦，去耦合

harness vt. 治理，利用

rudimentary adj. 根本的，低級的

latency n. 等待時間；執(zhí)行時間

unified adj. 同一的，統(tǒng)一的

future-proof 經(jīng)得起未來考驗(yàn)的

manipulation n. 操作，處理

withrespectto 關(guān)于，至于

NOTES

[1]Thefuturedeploymentofopticalcross-connectedWDMtransportnetworks,initiallyforprotectionandbypassbutultimatelymanagingopticallightpathsdynamicallyinmultipleringormesharchitectures,willpotentiallymodifytheroleofthenetworkfunctionalitiesprovidedbythesynchronousopticalnetwork/synchronousdigitalhierarchy(SONET/SDH)layer.“initiallyforprotectionandbypassbutultimatelymanagingopticallightpathsdynamicallyinmultipleringormesharchitectures,”是原因狀語，用來解釋前面網(wǎng)絡(luò)使用的目的。“providedbythesynchronousopticalnetwork/synchronousdigitalhierarchy(SONET/SDH)layer.”修飾“networkfunctionalities”，它是省略了whichare的后置定語從句。

本句可譯為：光交叉互聯(lián)WDM傳輸網(wǎng)絡(luò)最初是為了保護(hù)和旁路多環(huán)網(wǎng)或網(wǎng)狀網(wǎng)，但最后(演變?yōu)?動態(tài)管理光路，它可能改變網(wǎng)絡(luò)功能的角色，這些網(wǎng)絡(luò)功能是由同步網(wǎng)絡(luò)/同步數(shù)字體系層提供的。

[2]GiventhatWDMallowscheapandeasyincrementalincreasesofthetransmissionbandwidth,frequentupgradesofthetransport-layertransmissioncapacitycanbeenvisagedtomatchincreasingdemand,inturnplacingheavydemandsontheswitchingprocess.

“Giventhat…”意為“假設(shè)……則……”，表示某種假設(shè)情況，后面是前面假設(shè)條件成立后的結(jié)果?！癕atch”意為“匹配”?！癲emandson”意為“對……的要求”。

本句可譯為：假設(shè)WDM允許傳輸帶寬輕易、頻繁地不斷增長，傳輸層傳輸容量的頻繁升級則可看做是匹配日益增長的容量需求，進(jìn)而就會對交換過程施加繁重的需求壓力。

[3]Transmissionandswitchingareexecutedintheopticaldomain,whileroutingandforwardingarecarriedoutelectronically,wheretherelativelycomplexpacketheaderprocessingoccursindependentoftheopticalpayload.“while”引導(dǎo)并列從句，和前面的“Transmissionandswitchingareexecutedintheopticaldomain,”是并列從句；“where”引導(dǎo)定語從句，修飾“electronically”。

本句可譯為：傳輸和交換在光域?qū)崿F(xiàn)，而路由和轉(zhuǎn)發(fā)在電域?qū)崿F(xiàn)，相對復(fù)雜的分組頭獨(dú)立于光載荷在電域進(jìn)行處理。

[4]Packetformatisafundamentalconsiderationinanypackettransmissionsystem,definedbytherequirementsofthelayerwithrespecttonetworkfunctionalities,andisalsocrucialforopticaldomainimplementations.“providingtheleastsignalattenuationandthehighesttransmissionspeedsofanyfibercabletype.”為分詞短語作結(jié)果狀語，可以理解為whichcanprovidetheleastsignalattenuationandthehighesttransmissionspeedsofanyfibercabletype的縮寫。

本句可譯為：分組的格式在任何分組傳輸系統(tǒng)中都是一個基本的、需要考慮的事項(xiàng)，對于光域的應(yīng)用也是至關(guān)重要的。分組的格式是由某層根據(jù)網(wǎng)絡(luò)功能的需求決定的。

EXERCISES

I．Translatethefollowingwordsorphases.

IP WDM IWU

SONET/SDH hybridsolution opticalpayload

光分組交換 光電轉(zhuǎn)換 光纖延遲線

傳輸容量

II．TranslatethefollowingparagraphsintoChinese.

(1)

Withanextensiveopticalpacketlayer,theinterfacetoIPandotherprotocolsiscrucial.Encapsulation,theadditionofdeliveryinformationtothedatabytheopticalpacketlayer,willoccuratinterworkingunits(IWUs)ateachinterfacetotheelectronicclientlayer.

(2)

Intandem,thedevelopmentofwavelength-divisionmultiplexed(WDM)techniquesonpoint-to-pointlinkshavebeguntoutilizethemassiveopticalbandwidthofinstalledsingle-modeopticalfibersmoreefficiently.WDMwasinitiallyarapidsolutiontothesevererouteexhaustionproblemsbroughtaboutbyexponentiallyincreasingtraffic.

(3)

Thus,2bytesarerequiredtoaidinthesynchronizationprocessas“tag”uniquelyidentifyingthestartofthepayload.Guardbandsaccountforswitchingtimesoftheconstituentopto-electronicdevicesaswellaspayloadpositionjitter.Thepayloaddurationresultsfromatrade-offbetweentransmissionefficiency(thelongerthepayloadduration,thehighertheefficiency)andthepracticallimitationsonoveralllengthoftheopticalfiberdelaylinebuffers.Atthistime,delaylinesaretheonlyviableapproachtoimplementingcontentionresolutiondirectlyintheopticaldomain.

(4)

Switchingensuresthateachpacketemergesatthecorrectoutput,dependingontheinformationcontainedinthepacketheader.Althoughfixed-lengthpacketsarrivingontheinputsmustbesynchronized,thereisnocoordinationbetweenpacketstreamsarrivingondifferentinputs.Henceoneormorepacketsmayarriveduringthesametimeslotondifferentinputswishingtogotothesameoutput.Forthisreason,bufferingisrequired,whereoneormorepacketsarestoredwhileothersaretransmittedtothedesiredoutput.

參考譯文

第四單元光互聯(lián)網(wǎng)分組交換方法

1．簡介

無線電通信和數(shù)據(jù)通信的融合導(dǎo)致了網(wǎng)絡(luò)環(huán)境的模式變遷，Internet產(chǎn)生的大量業(yè)務(wù)推動了這種趨勢。隨著IP協(xié)議正在成為數(shù)據(jù)通信的主要協(xié)議，在較長的時期內(nèi)，IP協(xié)議也是代表無線通信和數(shù)據(jù)通信融合的強(qiáng)大的支持方。一前一后地，點(diǎn)對點(diǎn)鏈路的WDM(波分復(fù)用技術(shù))的發(fā)展已經(jīng)開始利用已有的單模光纖的大帶寬使其利用率更高。最初，WDM為呈指數(shù)增長的業(yè)務(wù)的嚴(yán)重路由耗盡問題提供了快速解決的方法。光交叉互聯(lián)WDM傳輸網(wǎng)絡(luò)最初是為了保護(hù)和旁路多環(huán)網(wǎng)或網(wǎng)狀網(wǎng)，但最后(演變?yōu)?動態(tài)管理光路，它可能改變網(wǎng)絡(luò)功能的角色，這些網(wǎng)絡(luò)功能是由同步網(wǎng)絡(luò)/同步數(shù)字體系層提供的。因此，IPoverWDM已經(jīng)成為一個重要的研究領(lǐng)域，它包括了支持主要的IP業(yè)務(wù)在WDM光鏈路上傳輸?shù)脑S多方法。許多研究工作集中在開發(fā)一種優(yōu)良的解決辦法，這種辦法用來解決WDM光層的傳輸能力和路由器的處理能力之間的不匹配問題。IP路由器完成四個主要的功能：

(1)路由：通過路由表提供網(wǎng)絡(luò)連接信息。

(2)轉(zhuǎn)發(fā)：定義每個輸入分組的輸出端口(基于路由表)。

(3)交換：把每個分組輸出到正確的輸出端口(由轉(zhuǎn)發(fā)過程定義)。

(4)緩存：由于分組到來的隨機(jī)性，當(dāng)多個分組想要同時通過同一出口時，可通過存儲分組解決沖突。

目前，隨著路由表尺寸的變大和更新頻率問題，轉(zhuǎn)發(fā)過程成為了限制吞吐量的主要問題。這些問題目前已得到解決，但是路由表的吞吐量(依然)基本受限于路由表查詢時間。大量的工作集中在開發(fā)數(shù)據(jù)結(jié)構(gòu)和算法，這些數(shù)據(jù)結(jié)構(gòu)和算法(的作用)是為了最小化路由表的查詢時間和存儲空間限制。

毫無疑問，在大吞吐量路由器設(shè)計(jì)上已經(jīng)取得了巨大的進(jìn)展(大于1

Tb/s)。不過，盡管取得了這些驚人的進(jìn)步，但人們?nèi)匀粚﹄娮咏粨Q系統(tǒng)有限的升級靈活性存有疑慮。假設(shè)WDM允許傳輸帶寬輕易、頻繁地不斷增長，傳輸層傳輸容量的頻繁升級則可看做是匹配日益增長的容量需求，進(jìn)而就會對交換過程施加繁重的需求壓力。

2．光分組交換

許多其他的解決或消除轉(zhuǎn)發(fā)瓶頸的方法正在進(jìn)行研究。利用光分組或突發(fā)技術(shù)，或者與電緩存一起來實(shí)現(xiàn)IPoverWDM。但是，這里講述的策略是直接在光域解決沖突(緩存)來讓步于WDM光分組交換。其目的是把交換負(fù)擔(dān)轉(zhuǎn)移到光域，允許交換能力和WDM傳輸容量之間進(jìn)行有比例的協(xié)調(diào)。因此這種策略可認(rèn)為是一種混合解決方法，它解除了吞吐量和路由/轉(zhuǎn)發(fā)過程之間的耦合關(guān)系。傳輸和交換在光域?qū)崿F(xiàn)，而路由和轉(zhuǎn)發(fā)在電域?qū)崿F(xiàn)，相對復(fù)雜的分組頭獨(dú)立于光載荷在電域進(jìn)行處理。這種去耦合方式使得光分組層能有效地支持很多網(wǎng)絡(luò)協(xié)議，同時也利用了WDM的傳輸能力。但是需要注意的是：(這種情形)隨著最近的一些演示情況在發(fā)生變化，這些演示中，有一些分組頭頭部處理功能可初步直接在光域進(jìn)行，這減輕了電處理的負(fù)擔(dān)，因而也有效地減少了控制信號的建鏈時間和管理時間。

有了一個擴(kuò)展的光分組層，則它與IP協(xié)議和其他協(xié)議的接口就是至關(guān)重要的。在每個互連單元(IWU)與電客戶層的接口處光分組層進(jìn)行傳輸信息的疊加，即封裝。封裝允許大量的協(xié)議數(shù)據(jù)比如IP和ATM數(shù)據(jù)映射到光載荷中，這些光載荷的持續(xù)時間可能是固定或可變的。IP隱藏了物理層的復(fù)雜性(包括光分組交換)，對高層提供統(tǒng)一的接口，而不管底層網(wǎng)絡(luò)是什么類型的。除了封裝功能，互連單元(IWU)也為(光分組)在光分組層的正確路由生成頭部，同時為了轉(zhuǎn)發(fā)而復(fù)用來自不同輸入鏈路的、去往同一目的地址的分組業(yè)務(wù)，從而保證了一個全光的端到端的鏈路(的正確運(yùn)行)。光分組提供了一個更高的復(fù)用等級，它允許業(yè)務(wù)流在光層傳輸之前進(jìn)行聚合，潛在地消除了對SDH的需求(SDH是IP業(yè)務(wù)在WDM鏈路上傳輸時需要的適配層)。因此，光分組網(wǎng)絡(luò)提供了一種潛在的解決辦法：能傳輸無連接和面向連接業(yè)務(wù)，并且具有靈活的帶寬管理能力和經(jīng)得起未來考驗(yàn)的帶寬增長能力。目前還沒有制定IP和ATM數(shù)據(jù)映射到光分組層的標(biāo)準(zhǔn)。

正如上面所提到的，有兩種主要的近似于光分組交換的方法，都可以應(yīng)用到Internet中：

(1)使用固定長度的光分組，很多個這樣的分組對應(yīng)一個IP數(shù)據(jù)報(bào)，需要IWU在層間或交換機(jī)的輸入和輸出端進(jìn)行分段和重組。

(2)每個IP數(shù)據(jù)報(bào)使用一個不定長度的光分組。

大部分已經(jīng)報(bào)道的研究都使用固定長度的光分組。因此，為了說明光分組層的功能，下面的資料都限于固定長度分組，分組的頭部和載荷都使用同樣的波長來編碼。假設(shè)每個分組的目的交換端口是光/電轉(zhuǎn)換后從頭部獲取的，那么，為了使得電操作(成為)可能，其頭部的速率應(yīng)當(dāng)較低。由于光緩存的天然特性，載荷的持續(xù)時間是固定的，不管它的內(nèi)容是什么，網(wǎng)絡(luò)的吞吐量和負(fù)載比特率是成比例的。負(fù)載比特率的范圍可從10Gb/s到更高，具有很大的上升空間。顯而易見，波長的個數(shù)不但對傳輸容量是至關(guān)重要的，而且對在實(shí)際運(yùn)用時的沖突解決也是至關(guān)重要的。為了交換和緩存變長光分組，本文將提出一些有前瞻性的概念來分析解決這些需求，并以這些概念結(jié)尾。

3．光分組交換機(jī)的設(shè)計(jì)

一個通用的光分組交換機(jī)的節(jié)點(diǎn)結(jié)構(gòu)包括三個子部分(見圖4.1)：

(1)輸入接口，包括一個用來對準(zhǔn)實(shí)時時鐘的輸入分組(光的)同步裝置。

(2)交換核心，用來把分組路由到它們正確的輸出端口，并執(zhí)行沖突解決。

(3)輸出接口，它插入新的分組頭，并且可再生數(shù)據(jù)。

圖4.1通用光分組交換機(jī)節(jié)點(diǎn)結(jié)構(gòu)分組的格式在任何分組傳輸系統(tǒng)中都是一個基本的、需要考慮的事項(xiàng)，對于光域的應(yīng)用也是至關(guān)重要的。分組的格式是由某層根據(jù)網(wǎng)絡(luò)功能的需求決定的。例如，KEOPS工程定義的分組格式既考慮了不同業(yè)務(wù)流情況下的光分組交換層的延遲-吞吐量性能，也考慮了節(jié)點(diǎn)和網(wǎng)絡(luò)路由的需要(見圖4.2)。自始至終它都假設(shè)時間被分成相等的時隙，每個時隙容納一個光分組，并且有效載荷可能包含從622Mb/s到10Gb/s速率的數(shù)據(jù)。

圖4.2光分組交換格式例子(KEOPS的定義)概要地說，頭域有14個字節(jié)：8個字節(jié)是路由信息，3個字節(jié)是有效載荷的類型標(biāo)識、流量控制信息、分組序號完整性保護(hù)編碼和頭部錯誤校驗(yàn)。分組格式間接地表明了對光同步裝置作為節(jié)點(diǎn)輸入接口的需要。假設(shè)當(dāng)分組進(jìn)入交換機(jī)時，它們的邊界是對準(zhǔn)的，那么每個分組和它的時隙也是對準(zhǔn)的。一般來說，這樣的同步過程對于正確的交換操作和低的分組丟失率是必需的。在輸入端到交換機(jī)之間同步的方法可以構(gòu)成一個單獨(dú)的命題，這里不作深入討論。因此，在同步過程中，使用2個字節(jié)作為“標(biāo)簽”特別標(biāo)識有效載荷的開始來幫助達(dá)到同步。保護(hù)頻帶負(fù)責(zé)組成光電設(shè)備的交換時間以及有效載荷位置的抖動。有效載荷持續(xù)時間源于傳輸效率(持續(xù)時間越長，效率越高)和光纖延遲線緩存總長的實(shí)際限制之間的折中平衡。此時，光纖延遲線是唯一能直接在光域解決沖突的可行辦法。

光分組交換有三個主要功能：交換、緩存和頭部翻譯(可選)。在基于IP的系統(tǒng)中，后面的功能(頭部翻譯)可能被路由和轉(zhuǎn)發(fā)所替換，這個問題前面討論過了。交換依靠包含在分組頭中的信息來保證每個分組能能被傳送到正確的輸出端。盡管到達(dá)輸入端的定長分組一定是同步過的，但是來自不同輸入端的分組流也達(dá)不到(完全)一致。因此，一個或多個分組可能在同一時隙到達(dá)不同輸入端，并去往相同的輸出端。由于這個原因，緩存是必需的，在這里一個或多個分組被存儲起來，而其他的分組被傳送到希望的輸出端口。

頭部翻譯功能在ATM傳輸策略中是一個重要的過程，而路由和轉(zhuǎn)發(fā)功能在IP系統(tǒng)中也是同樣重要的，這點(diǎn)前面已討論過了。盡管這些機(jī)制提供了重要的功能和靈活性，但是并沒有體現(xiàn)在每個光分組交換系統(tǒng)中，因?yàn)橹苯拥墓夥纸M頭部翻譯功能還處在發(fā)展的初級階段，還有很長的路要走。因此，為了方便電解碼和頭部信息的翻譯，頭部的比特率通常比有效載荷的比特率低。

EXTENSIVETEXT

OpticalBurstSwitchingfortheNextGenerationInternet

Currently,ourdemandsfornetworkbandwidthincreasedaily.WeaccesstheInternettocheckemail,readthenews,checkstockquotes,takeclassesonline,finddrivingdirections,playgamesonline,buyflowersandthislistgoesonandon.Inordertomeetourever-increasingnetworkbandwidthneeds,onesolutionistodesignandbuildtheNextGenerationInternetwithanopticalcoreinfrastructure,withfastconnectionprovisioningtimeandunprecedentedhighdataratesof100terabitspersecondandhigher.Anopticalnetworkisbuiltbyinterconnectingvariousopticalswitcheswithwavelength-divisionmultiplexing(WDM)fibers,i.e.,fibersthatcansimultaneouslytransmitdataoverdifferentwavelengths.Manyoftoday’scommercialopticalnetworks,however,donotutilizetheWDMtechnologyefficiently.Oneexampleisthepopularsynchronousopticalnetwork(SONET)rings,whichconsistofSONETadd/dropmultiplexers(ADM)linkedwithWDMfiberssoastoformaring.Inthiscase,WDMissimplyusedasatransportmedium.Thatis,thedataistransmittedoptically,buttheopticalsignalisterminatedateveryADM.ThesignalisthentranslatedtoelectronicsforprocessingandthentranslatedbacktoanopticalsignalforforwardingtothenextADM.TheproblemisthemismatchbetweentheelectronicprocessingspeedsandtheopticaltransmissionratesthatleadtoanundesirablebottleneckateachSONETnode.Toresolvethebottleneckproblem,anall-opticalWDMnetworkwherethedatatravelsfromthesourcetothedestinationentirelyintheopticaldomainhasbeenproposed.Withrespecttothecurrentstateofthetechnology,anOpticalBurstSwitched(OBS)Networkisoneofthemostpromisingall-opticalarchitecturesfortheNextGenerationInternet.Itefficientlysupportsthetransmissionofburstytrafficoveranall-opticalinfrastructure.OBSisstillbeingdevelopedandithasnotbeenstandardizedyet.ThisarticledescribesthemainfeaturesofanOBSnetwork,itsbenefitsaswellasitschallenges.

1．TheOBSbasics

AnOBSnetworkconsistsofcorenodesandend-devicesinterconnectedbyWDMfibersasshowninFig4.3.AnOBScorenodeconsistsofanopticalcrossconnect(OXC),anelectronicswitchcontrolunit,androutingandsignalingprocessors.AnOXCisanon-blockingswitchthatcanswitchanopticalsignalfromaninputporttoanoutputportwithoutconvertingthesignaltoelectronics.TheOBSenddevicesareelectronicInternetProtocol(IP)routers,asynchronoustransfermode(ATM)switches,orframerelayswitches,equippedwithanOBSinterface.EachOBSend-deviceisconnectedtoaningressOBScorenode.Theend-devicecollectstrafficfromvariouselectronicnetworks(suchasATM,IPandframerelay).ItsortsthetrafficperdestinationOBSend-deviceaddressandassemblesitintolargervariable-sizeunits,calledbursts.Foreachburst,theend-devicealsoconstructsacontrolpacket,whichcontainsinformationabouttheburst,suchastheburstlengthandtheburstdestinationaddress.Thiscontrolpacketisimmediatelysentalongtherouteoftheburstanditiselectronicallyprocessedateachnode.

Fig4.3TheOBSnetworkarchitecture

Thefunctionofthecontrolpacketistoinformthenodesoftheimpendingdataburstandtosetupanend-to-endopticalpathbetweenthesourceandthedestination.Afteradelaytime,knownastheoffset,theend-devicealsotransmitstheburstitself.Thebursttravelsasanopticalsignalovertheend-to-endopticalpathsetupbyitscontrolpacket.Thisopticalpathistorndownafterthebursttransmissioniscompleted.ThisseparationofthecontrolinformationandtheburstdataisoneofthemainadvantagesofOBS.Itfacilitatesefficientelectroniccontrolwhileitallowsforagreatflexibilityintheformatandtransmissionrateoftheuserdata.Thisisbecausetheburstsaretransmittedentirelyasanopticalsignal,whichremainstransparentthroughoutthenetwork.Ingeneral,thetimeittakesthecontrolpackettoreachthedestinationend-deviceisequaltotheend-to-endpropagationdelayplusthesumofalltheprocessingdelaysatalltheintermediatecorenodes.Ontheotherhand,thetimeittakesforabursttoreachthedestinationend-deviceisonlyequaltotheend-to-endpropagationdelay.ThereasonisthattheburstistransmittedasanopticalsignalthattraversestheOBSswitcheswithoutanyprocessingorbufferingdelays.Inviewofthis,thetransmissionofaburstisdelayedbyanoffset.ThiswayitalwaysarrivesatanOBSnode,afteritsswitchcontrolunithashadthechancetoprocessthecontrolpacketassociatedwiththeburstandconfiguretheopticalswitchfabric.Theoffset,therefore,isafunctionofthenumberofnodesthatthecontrolpackethastotraverseendtoend.

2．Burstaggregationalgorithm

Theburstaggregationalgorithmattheend-devicescangreatlyimpacttheoverallOBSnetworkoperationbecauseitsetstheburstcharacteristicsand,therefore,itshapestheburstarrivaltraffic.Thealgorithmhastoconsiderthefollowingparameters:apre-settimer,amaximumburstlengthandaminimumburstlength.Thetimerdetermineswhentheend-deviceistoassembleitscollectedtrafficintoanewburst.Themaximumandtheminimumburstlengthparametersshapethesizeofthebursts.Weneedtosetamaximumburstlengthsinceverylongburstsholdontotheresourcesofthenetworkforalongtime.Thus,theycausetheunfairlossofotherbursts.Ontheotherhand,theminimumburstlengthisalsonecessarybecauseveryshortburstsmaygiverisetotoomanycontrolpackets.ThissituationcanoverloadthecontrolunitoftheOBSnode.Theburstaggregationalgorithmmayusebit-paddingifthereisnotenoughdatatoassembleaminimumsizeburst.OnewaytoprovideclassesoftrafficinOBSistoimplementpriorityqueuesattheedgeofthenetworkduringtheburstaggregation.Basedontheclassofservice,theend-devicessorttheupperlayertrafficintodifferentqueues.Asaresult,eachend-devicewillhaveC*Npriorityqueues,whereCisthenumberofserviceclassesandNisthenumberofpossibledestinations.Anappropriateschedulingalgorithmguaranteesthatthesequeuesareservedaccordingtotheirpriority.

3．Signaling,routingandwavelengthallocation

Signalingisanimportantaspectinanynetwork.Itspecifieshowtheconnectionsareestablishedanditdetermineswhetherornottheresourcesareutilizedefficiently.InmostOBSvariants,thesignalingofconnectionsisaccomplishedusingaone-waysignalingscheme:theburstistransmittedafteranoffsetwithoutanyknowledgeofwhethertheopticalpathhasbeensuccessfullyestablishedend-to-end,asillustratedinFig4.4.

Fig4.4One-waysignaling

End-devicesAandBareconnectedviatwocoreOBSnodes.TheverticallinesinFig4.4representatimelinesoastoshowtheactionstakenbyeachnode.End-deviceAtransmitsacontrolpackettoitsingressOBSnode.Thecontrolpacketisprocessedattheingressnode.Iftheconnectioncanbeaccepted,itisforwardedtothenextnode.Thecontrolpacketisreceivedby