THE SIMULTANEOUS GLOBALIZATION AND LOCALIZATION OF NETWORKS has changed the nature and relevance ofcommunication networks dramatically for a broad audience.
This change is evidence that communications systems have become the most significant element of mission-critical infrastructure for consumer, business, and government market segments.
While it iscommonly recognized that carrier-class service has become a minimum network requirement for delivering services, its limitations become clear when disrupted by human error, act of war or natural disasters. Minimizing the risk of such disruptions requires a new approach to building a survivable optical network.
White PaperSurvivable Optical NetworksCoreDirector Multiservice Optical Switch Control Plane enables a protected worldwide networkTHE SIMULTANEOUS GLOBALIZATION AND LOCALIZATIONOF NETWORKS has changed the nature and relevance ofcommunication networks dramatically for a broad audience. Thischange is evidence that communications systems have become themost significant element of mission-critical infrastructure forconsumer, business, and government market segments. While it iscommonly recognized that carrier-class service has become aminimum network requirement for delivering services, its limitationsbecome clear when disrupted by human error, act of war or naturaldisasters. Minimizing the risk of such disruptions requires a newapproach to building a survivable optical network.Network globalization has been an ongoing strategic goal forresearch and education institutions, government, and enterprises.Government-funded research and education networks rely on theinfrastructure for global data interchange during bandwidth-intensiveexperiments, many of which share supercomputers and cannottolerate service interruptions. Likewise, large financial organizationstransacting in world markets depend on global telecommunicationsinfrastructures and require low latency and close to 100 percentuptime, as even a few seconds of network failure can cost millions ofdollars. The recent increase in high profile natural disasters highlightsthe critical role of a survivable network to the stability, longevity, andin some cases survival of global companies. Due to the critical natureof international bandwidth, many operators rely on pre-existingsubmarine cable network restoration plans, with their associated highcosts, limited protection and slow restoration. Localization of networks is being driven heavily by consumer marketdemand for broadband services, specifically video services, and theproliferation of business services within Metropolitan Area Networks(MANs). Local network reliability has become more important asoperators grow high-value, revenue-generating business servicessuch as IP-based voice, private line, and Virtual Private Network(VPN) services. Cable network operators, for example, arepositioning network reliability and dependability as key features oftheir rapidly expanding business services portfolios in North America.Sensitivity to cost, however, demands a strong business rationale andproven savings. Network and usage trends indicate a growing reliance oncommunication networks, which can be detrimental to the end-userin the event of a catastrophe. Operators seeking to implement globalsurvivable optical networks must evaluate the necessity, complexityand cost. Other considerations would include how to managemultiple protection mechanisms for each service and integrate intoan existing network management infrastructure. Many operators areturning to automation schemes to reduce costs, deliver new servicesand enhance network survivability. This paper discusses the effect of industry standards-basedintelligent control plane technology within Ciena s CoreDirectorMultiservice Optical Switch, which offers network automation andcost reductions, while improving network survivability. Optical Control PlanesAN OPTICAL CONTROL PLANE, in very simple terms, is softwarethat controls all configurable features of a network element and/or anentire network. This capability includes the simple configuration,activation, and deactivation of circuits from individual circuits to afull mesh that make up an entire network. Fully meshed networksenable interconnection of a single node to every other node.Managing such a complex network requires a control plane that,given some parameters, can make decisions on its own. Known as anintelligent control plane, this software-based automation helpsimprove network survivability, scalability and cost. The intelligentcontrol plane also enables the creation of a fully automated meshedoptical network, which is the single most important ingredient in aUntitled Document2Survivable Optical Networkssurvivable optical network. For example, when a Ciena CoreDirectornetwork is expanded, the inventory of available circuits and ports isautomatically discovered and placed into context of the entirenetwork in which it functions. Once installed, the new capacity ismade available for new services, and any new circuits are addedautomatically to the pool of circuits that may be used for restoration.CoreDirector manages this process real-time and automates circuitprovisioning and eliminates manual inventory management. Key tothis automation is the software, which requires years of fieldexperience to achieve the high levels of reliability, resiliency andefficiency.Control Plane StandardsTHE INTERNATIONAL TELECOMMUNICATIONS UNION (ITU)established the Automatically Switched Optical Network (ASON)standard to guide the development of common optical controlplanes for intelligent optical networks. G.ASON requirements andrecommendations describe how a suite of control plane protocolsshould react to service requests, automatically provisioning end-to-end network resources across a multi-technology, multi-vendoroptical network. This provisioning allows the network s scope andcapabilities to be increased without a corresponding increase inmanpower. CoreDirector intelligent control plane software iscompliant with the ITU G.ASON standard and is field-proven for feature-rich functionality, reliability, andscalability based on nearly seven years of live networkdeployment and operational experience.CoreDirector BenefitsTHE MOST WIDELY DEPLOYED INTELLIGENTOPTICAL SWITCH, CoreDirector was the first toimplement and deploy control plane functions fordiscovery, routing and signaling. CoreDirectorstandards-based signaling and routing protocols havebeen deployed by over 25 carriers around the world,with as many as 150 nodes per network. Carrier analysesof operational expenses show up to 85 percent savingsafter deploying an intelligent optical network based on CoreDirectorsystems. The control plane s level of automation also improvesnetwork resource utilization, manages equipment, automatesprovisioning and manages inventory real-time, lowering capital costsup to 70 percent and reducing provisioning and/or change intervalsfrom months to minutes, allowing carriers to optimize time-to-revenue.Additional features implemented on Ciena s CoreDirector extendvalue-added functionality beyond the CoreDirector network. Supportfor standards defined by the Optical Internetworking Forum (OIF)include the Optical User-to-Network Interface (O-UNI) and theExternal Network-to-Network Interface (E-NNI), which allow theCoreDirector network to extend its intelligence to provision servicesacross a multi-vendor network. ONE OF THE KEY BENEFITS of a CoreDirector and its G.ASON-based intelligent control plane is its ability to offer multiple levels ofrestoration at the network level and the circuit level. Such additionalfeatures on CoreDirector are essential to enabling survivable opticalnetworks. Whether the network is made up of circuits in a MAN,Wide Area Network (WAN), or undersea cable network,Figure 1. Mesh-based restoration in actionUntitled Document3CoreDirector s intelligent control plane allows the creation ofmultiple restoration schemes down to the individual circuit leveleven while overlaid with network level restoration schemes. Theintegration of network-level and circuit-level restoration enables50ms restoration even in a fully mesh-based network. During acatastrophic failure, a network would automatically begin asequence of events to restore services while preserving therestoration schemes of the individual circuits. In the event acustomer demands a specific restoration path for individualcircuits, the intelligent control plane would factor thoserequirements into its restoration plans and restore services.Based upon field experience, few catastrophic events are noticedby the network, except for those services located at the failurepoint. CoreDirector s robust software represents the industry smost resilient, and fastest-restoring survivable optical network. FastMesh RestorationFASTMESH IS COREDIRECTOR S MESH TOPOLOGY ANDRESTORATION IMPLEMENTATION, which can be deployed aloneor concurrently with traditional SONET/SDH 50ms line protectionmechanisms. Linear and ring protection can be implementedconcurrently with FastMesh restoration to provide the highest levelsof service availability. FastMesh continually monitors the network andmakes restoration connections on an end-to-end basis. In case of acatastrophic event, this function continues to restore networkconnections, ensuring the highest priority traffic is carried firstdespite multiple network failures. This function operatesindependently of underlying linear and ring protection schemes,offering a multi-tiered protection hierarchy for each end-to-endconnection. The flexibility of FastMesh enables macro level networkprotection, granular protection by circuit as well as pre-determinedrestoration routes for fastest restoration. Varying levels of restorationservices can be sold to customers as a premium service along withtheir Service Level Agreement (SLA) requirements. Networkoperators benefit from high levels of network protection and highmargins from various levels of protection. Figure 1 illustrates howFastMesh routes around a single network failure.Figure 2 shows the conceptual advantages of intelligent meshrestored networks and how automation reduces the time for servicerestoration compared to traditional networks. Traditional ringnetworks are designed to restore service quickly in the event of asingle failure but usually cannot manage a second failure, requiringmanual intervention. Mesh restored networks can handle multiplefailures maximizing active circuit time, customer satisfaction andcontinuous generation of revenue. Link AggregationLINK AGGREGATION ALLOWS MULTIPLE LINKS BETWEENTWO NODES TO BE MANAGED AS A SINGLE LINK, andsupports intelligent network scaling, faster restoration times andnetwork simplification. For customers requiring submarine cablediversity without the complexity of managing multiple circuits, linkaggregation is ideal. Submarine cable restoration architecturalschemes, typically controlled only by the cable operator, may beexpensive and limit restoration to only one other cable system.CoreDirector FastMesh can aggregate capacity from multiple cablesystems into a bandwidth pool and optimize usage and restorationdecisions, lowering resource costs, providing greater levels ofSurvivable Optical NetworksFast RestoreManual RestoreFast ProtectSLAPenaltyRepairFailure 1Failure 2Failure 1Request ServiceDelayedTurn-upControl Plane MeshTraditional SONET/SDH RingFast RestoreFast Turn-upService Session tFigure 2. Revenue impact of traditional versusmesh-based restorationUntitled Document4flexibility, and creating the highest level of network survivability. Anexample of multiple circuits sharing undersea cable capacity is shownin Figure 3.Multi-vendor InteroperabilityThe OIF has created O-UNI specifications to allow subtendednetwork elements, which also support O-UNI, to request the setup ortear-down of light paths across the CoreDirector network. Uponreceiving a request from a subtended network element, the OpticalSwitching and Routing Protocol (OSRP) will perform the operationautomatically.Automation of the request, setup, and tear-down processes reducesoperational costs and allows rapid response to service requests andgeneration of additional revenue. Major features of the initialCoreDirector O-UNI implementation include: In-band and out-of-band Resource Reservation-Traffic Engineering(RSVP-TE) for signaling Neighbor discovery Support across all CoreDirector SONET/SDH optical interfaces Support for IPv4 addressing, with potential extension to IPv6 andNetwork Service Access Point (NSAP) Handling for numbered and un-numbered links Neighbor discovery Support on mesh, ring or linear CoreDirector topologiesOIF External Network-to-Network Interface (E-NNI)OPTICAL CONTROL PLANE INTEROPERABILITY, based on OIF E-NNI/ITU-T G.7713.2, expands the end-to-end automatedprovisioning capabilities of the intelligent optical network acrossdisparate optical control domains. This interoperability allows rapidservice deployment worldwide across multiple carrier networks andvaried vendor equipment. E-NNI is an ASON standard in progress forcontrol plane signaling and routing of Label-Switched Paths (LSPs)between the optical control domains of a carrier s network, orbetween the network and another carrier s network. Current CoreDirector implementation of E-NNI includes: Extensions to RSVP-TE for signaling Extensions to Open Shortest Path First-Traffic Engineering (OSPF-TE) for routing Interoperability with O-UNI 1.0 Support across all CoreDirector SONET/SDH optical interfaces IPv4 addressing, with potential extension to IPv6 and NSAP Handling of both numbered and un-numbered linksSurvivable Optical NetworksSubmarine Mesh NetworkCable SystemCable SystemCable SystemOC-192DS3GbE1GbESTM-1STM-64#1#2#3Figure 3. Link aggregation combined with mesh architectures eliminates multiple points of failure Untitled Document5IETF GMPLSCONTROL PLANE INTEROPERABILITY using the InternetEngineering Task Force s (IETF) Generalized Multi-Protocol LabelSwitching (GMPLS) Internal Network to-Network Interface (I-NNI)signaling and routing combines O-UNI and CoreDirector sintelligent control plane to enable automated end-to-end serviceprovisioning within an optical control domain. GMPLS is a standardin progress for common control plane signaling and routing of LSPswithin the optical control domain of a carrier s network. The initialCoreDirector implementation of GMPLS is based on the peer-to-peermodel, in which all elements on the optical network, such as opticalswitches, routers and Add/Drop Multiplexers (ADMs), share thedatabase for the entire optical network, shown in Figure 4.Financial Benefits of CoreDirector Intelligent Optical NetworksThe financial benefits are most apparent in the following areas: Reduced cost for networking equipment, space and power Improved resource utilization and service availability via mesh networking Reduced operations costs via automationCiena and customer studies show that significant CapitalExpenditure (CAPEX) and Operations Expense (OPEX) savings canbe achieved through deployment of a CoreDirector network.Figure 5 represents a Ciena study that compared legacytechnology to an intelligent optical network built withCoreDirector, resulting in CAPEX and OPEX savings andgeneration of additional revenues for a six-year Net PresentValue (NPV) of 218M. The major assumptions of the Cienastudy are: A 15-node network 10 Gb/s links between nodes 30 to 40G aggregate traffic per node CoreDirector network scenario mixing mesh and ring restoration Five-day service provisioning for legacy network Labor force sized for a typical carrierFigure 6 shows the change in service provisioning time for an existing CoreDirector customer after deploying a CoreDirectorintelligent network. Average service provisioning time wassignificantly reduced as the footprint of the CoreDirector intelligentnetwork increased. Reducing the time required to provision servicesresulted in incremental revenue generation. Survivable Optical NetworksCDCDCDCDCDABEFGHIFigure 4. CoreDirector extends routing intelligence toGMPLS enabled elements 62 90 66 218Total ImpactAdditionalRevenuesOPEXSavingsCAPEXSavingsFigure 5. Sample CAPEX and OPEX reduction withintelligent control plane optical switchUntitled Document6Reduced Operations Costs via AutomationThe automation of operating functions such as service provisioning,equipment inventory and topology updates allows carriers toreduce the number of network operation personnel and respondmore quickly to requests for new services, moves, adds and changes.Automated service activation can reduce time to revenue and reduceservice provider operational expenses associated with manualprovisioning. This automated process also enables a very simplemanagement structure.SummaryTHE NEED FOR SURVIVABLE NETWORKS GOES WELLBEYOND THE TYPICAL CRITICAL NETWORK INFRASTRUCTUREsuch as air traffic control and government, military, and medicalcommunications. The definition for those that require near-zerodowntime has expanded to include educational entities, largeenterprise, multi-national corporations and large financialinstitutions. Operators that wish to sustain and grow their high-margin revenue streams, such as those associated with global voiceand data services, rely on their global infrastructure even moreduring catastrophic failures. An intelligent control plane expandscarrier-class operator networks to help operators thrive, scale newservice, lower costs, and most importantly survive single,multiple, or catastrophic failures. As communication networkscontinue to grow, businesses that implement survivable opticalnetworks will garner higher margins for services, reduce CAPEX andOPEX, and retain the best customers. Ciena s CoreDirector Multiservice Optical Switch possesses thetechnologies necessary for implementing survivable networks. As aresult of the company s innovation in intelligent switching, Ciena isthe market leader in optical switch deployment. CoreDirectorsystems are deployed in some of the world s largest carriers,undersea cable systems and cable MSOs, and enable survivableoptical networks by offering the industry s most robust control planetechnology. Customers are reaping the financial benefits ofCoreDirector s intelligent networking functionality and provenreliability for carrier-class networks.Survivable Optical Networks100% Legacy Network100% Intelligent Optical NetworkNetwork EvolutionMinutesDaysWeeksMonthsHours*Assumes access is availableCiena may from time to time make changes to the products or specifications contained herein without notice. 2007 Ciena Corporation. All rights reserved. WP023 5.2007Figure 6. Provisioning time comparisonFor more information about Ciena, please call 1-800-207-3714 (US), 1-410-865-8671 (outside US), or +353.1.2436711 (international) or visit www.ciena.com