Making Sense of Metro DWDM, Part I
By: Lawrence Gasman, Communications Industry Researchers Inc. (WWW.CIR-INC.COM)
Contents
The metro DWDM opportunity proposition
Assessing the proposition
The high cost of metro DWDM
Special considerations
LAN internetworking
Path restoration and protection switching
Time slot assignment
Since mid-1998, the dense wavelength division multiplexing (DWDM) industry has begun to focus on the opportunities in the "metro" segment of the network. This is the segment between the access network and major backbone facilities. The metro market segment is getting so much attention because the long-haul DWDM market is maturing, the metro space is a potentially huge market, and a level playing field for new entrants vs. incumbent suppliers. In a recent report, Communications Industry Researchers, Inc. projected that DWDM system expenditures by CLECs and ILECs would be about $255 million this year, but would grow to $1.9 billion by 2003.
The metro DWDM opportunity proposition
The long-distance DWDM market shows signs of early maturity. Deployment of DWDM by the major long-distance carriers is not accelerating, and the largest long-distance networks are settling on a few DWDM suppliers, creating barriers to entry for others, especially start-ups. For this reason, DWDM vendors have started to look around for new opportunities.
Metro networks appear to offer vendors new DWDM opportunities. The metro DWDM market is potentially huge because metro network operators already consume billions of dollars of time division multiplexing (TDM, including synchronous optical network [SONET] and T-carrier) equipment every year.
DWDM vendors argue that if they could just grab a little of the metro network equipment market, they could do very well. Metro network operators appear to need DWDM-class bandwidth, because metro networks are squeezed between the access network, which is expanding in capacity with digital subscriber line (DSL) equipment, cable modems, and ATM access devices, and long-haul backbones that are being pumped up with DWDM.
Because the technical requirements for a metro DWDM box are significantly different from those of a long-haul DWDM system, established long-haul DWDM suppliers may not have significant competitive advantages over new entrants in the metro space.
Assessing the proposition
Metro networks will need more bandwidth. But when? Will DWDM be the most effective way to provide that bandwidth? For a few service providers—perhaps as few as 20—the answer to these questions appears to be "immediately," and "yes," respectively. These carriers are already deploying DWDM in the metro area, and most of them are competitive local exchange carriers (CLECs) deploying the equipment in new, or "green field" networks. But the incumbent local exchange carriers (ILECs), in whose networks the greatest potential for DWDM sales clearly lies, mostly just dabble with DWDM.
Many service providers recognize that DWDM is the leading edge optical transmission technology, but the technology's cost and immaturity for metro applications still drives them towards fiber/SONET deployments in lieu of DWDM. The speed and degree to which vendors can overcome these interrelated problems will determine the size of the metro DWDM market over the next few years.
The high cost of metro DWDM
Serious cost-related factors will continue to delay the mass introduction of DWDM into metro networks. These factors include inappropriate engineering of DWDM systems, the low cost of deploying fiber, and the lack of a value proposition presented by minimizing costly regenerators as in long-haul networks.
Until very recently, local exchange carriers have found that vendors have designed DWDM systems primarily to meet the needs of long-distance providers. The systems were usually too expensive for metro applications because they supported higher data rates over longer distances than required by metro networks. This problem is going away quickly as more vendors bring metro-optimized DWDM systems to market. These newer systems are less costly because they do not need the amplifiers used for long-haul applications and because they can use lower cost transceivers for lower data rates than long-haul systems. On the other hand, metro DWDM systems require special features that long-haul systems do not, such as interfaces to Fast Ethernet, Gigabit Ethernet, video systems, etc., which adds to the systems' cost.
Proponents tout DWDM deployment as a cost-effective alternative to pulling in new fiber. Although fiber deployment is expensive in the largest metropolitan areas, pulling new fiber in other areas is often relatively cheap. DWDM at current prices often does not make economic sense for metro network operators to deploy, when they factor in the additional cost of SONET add/drop multiplexers (ADM) to carry traffic on the DWDM channels. With the cost of labor relatively static, the metro DWDM systems may have to come down in price by 40% to 50% to make them attractive for mass deployment. In addition, it is important to remember that while l particular spans in LEC networks may be in need of bandwidth boosts, on the whole networks have quite a lot of spare capacity, as shown in the exhibit below.
In long-haul networks, the elimination of regenerators provides a compelling economic for DWDM deployment. Although eliminating costly regenerators creates a "slam-dunk" value proposition for deploying DWDM in long haul networks, the scenario is absent in the regenerator-free metro area.
Special considerations
Even an overnight drop in the price of DWDM equipment would not lead to a scramble by metro network operators to deploy it. Vendors must also address special technical issues with metro DWDM systems. Issues that metro DWDM vendors currently contend with include support for local area network protocols, signal path restoration, protection switching, and time slot assignment.
LAN internetworking
Metro DWDM systems must support standard local area network (LAN) internetworking services and interfaces. In the carrier network backbone, different services usually are encapsulated into SONET frames. Because of this, carriers have primarily sought equipment with "open" interfaces to established SONET products.
Unlike backbone networks, metro networks often support services in their native form, instead of having them encapsulated into SONET. Network operators will require metro DWDM platforms to support native LAN protocols, because this facilitates the service provider's ability to generate revenues from high-end LAN internetworking services.
In addition to "mass market" LAN interfaces, carriers may also require metro DWDM systems to support high-end channel interfaces—Escon, Ficon, and Fibre Channel—associated with the large data centers in financial institutions, utilities, and government agencies. These data centers will likely be the primary customers of carriers provisioning services over metro DWDM platforms.
Path restoration and protection switching
SONET rings are either unidirectional path-switched rings (UPSRs) or bidirectional line-switched rings (BLSRs), as defined by Bellcore (Telcordia) standards. UPSRs offer 1:1 redundancy throughout the ring; BLSRs provide a degree of redundancy through extra capacity inherent in the SONET standards. Few metro DWDM vendors provide equivalent functionality, yet the ability to do so will put vendors at a major competitive advantage.
Protection switching is the feature that many industry experts herald as SONET's critical advantage over DWDM, but the issue may be somewhat exaggerated. DWDM equipment manufacturers will likely resolve the disparity in protection switching performance between SONET and DWDM within a year.
The protection issue also appears to have less to do with DWDM itself than the fact that DWDM will be used in conjunction with SONET in the near term. In an entirely DWDM environment, most next-generation systems would shift signals over to the backup fiber in about 25 milliseconds or less, whereas SONET ADMs take 50 milliseconds. Where SONET and DWDM are used in conjunction, operators turn off the DWDM protection switching to avoid conflict with the SONET protection scheme. This is inefficient and results in a slower-than-SONET response.
Thus the argument that DWDM will continue to have deployment problems because of protection switching issues could also be used to argue that there is a need to move quickly to DWDM-only metro systems. Given the huge embedded base of SONET and the commitment that the Bell companies have made to SONET, pushing DWDM-only solutions in the metro segment is now misguided, although it may not be within a few years. Nevertheless, vendors that capitalize on DWDM's intrinsic protection switching capabilities will likely enjoy significant competitive advantage.
Time slot assignment
Yet another advantage claimed by SONET proponents is time slot assignment. SONET ADMs are programmed to assign a time slot from one customer to another. Nortel Networks' (Brampton, ON) metro DWDM solution, for instance, contains an electrical crossconnect that reassigns wavelengths between customers. The only alternative appears to be to physically moving cards in the DWDM system chassis, which harkens back to patch panels and is hardly an acceptable solution in the next-generation network space.
Part II of this article lays out how metro DWDM will succeed.