Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index


THought this might be of interest to some of you----


The Long and Short of WDM 
Posted April 02, 1999 10:00 AM PST 
The bandwidth explosion. The data wave. The capacity crunch. Whatever name is
used, the phenomenal growth in network traffic levels, and the seemingly
never-ending demand for more bandwidth, are key issues defining the 21st
century telecommunications landscape. This "bandwidth challenge" has
wavelength-division multiplexing (WDM) into the spotlight in recent years, to
the extent that some observers have called it the fastest-growing
carrier-equipment market in telecommunications history. The ability to provide
a dramatic jump in capacity, while simultaneously reducing the cost per
transmitted bit, rapidly has made WDM an established part of the long distance
transport network.
Impressive progress, certainly. But now as the embryonic stage of WDM's
evolution draws to a close, it is pertinent to assess the new directions
technology will take, in particular the much-vaunted extension of WDM to the
short-haul local exchange carrier (LEC) environment. Unfortunately, it is here
that a great deal of confusion--not to mention hype--seems to exist in the
The case for WDM in the long haul now is well understood. The increased
capacity and cost benefits on long distance routes mean that it has become a
"no-brainer" to deploy WDM in such applications. When long-haul carriers such
as Sprint Corp. and AT&T Corp. first started purchasing WDM systems in large
numbers to relieve exhausted fiber in 1996 and 1997, the next step seemed
obvious--extension of WDM to the metropolitan and access regions of the
network. Industry watchers predicted the market for so-called "metro WDM" to
grow in the same explosive fashion in 1998 that occurred in the long distance
environment in 1996 and 1997.
However, as of today, that explosion has not occurred. Many LECs are in trials
with the technology, and the first commercial applications in metropolitan
environments have begun, but widespread commercial deployment has not
as rapidly as predicted.
So what happened? One factor creating confusion is the "short-haul" WDM
environment being referred to as if it were a homogeneous market. This in the
past led to the mistaken view that once WDM became cost-effective for one
network, it would necessarily become cost-effective for all 
metropolitan/interoffice networks, feeding predictions of a market explosion.
However, the local environment is not homogenous. Metropolitan, interoffice
access networks consist of a diverse range of architectures, bit rates,
patterns and protocols, in contrast to the long distance environment that is
evolving from an established, standardized synchronous optical network (SONET)
base. The economics of WDM compared to alternative solutions are, therefore,
much more complex.
Cost is clearly a major issue. Although the cost of systems continues to fall,
WDM has not broken significant price barriers in the short-haul market yet.
Unlike the long distance environment, there has been no obvious "killer
application"--an economically compelling reason to deploy WDM in the majority
of cases, over all other solutions. 
In the short term, therefore, the decision to deploy WDM in the local
environment appears to be driven by simple network congestion and cost
considerations--where WDM is cheaper and/or more readily available than the
alternatives for increasing capacity, then it will be deployed. This is
occurring in those networks that are "fiber poor"--i.e. where existing fiber
capacity is running out and there is a lot of traffic on interoffice
facilities. However in "fiber rich" networks--where there are spare conduits
for laying new fiber and/or network congestion is not so acute--WDM may not be
deployed widely in the near term. It is fair to say that there will be room
a variety of solutions and the choice of technology will be decided on a
case-by-case basis.
However, despite the relative lack of success thus far for metro WDM, simply
looking at the technology in terms of point-to-point capacity increase is
misleading. It is the move toward optical networking, or "photonics," that
be the key to the success of WDM in the local environment. 
Optical networking, deploying WDM ring architectures as opposed to simple
point-to-point WDM, potentially will provide a common method for transport
regardless of signal format in the metro environment, giving LECs the ability
to deliver new services more quickly while reducing the costs associated with
There will be an increasing need for this approach, due to changes in the
industry itself. The emergence of so many competitive LECs (CLECs) since the
Telecommunications Act of 1996 is engendering a new business approach in the
local environment, opening up many windows of opportunity for new, lean,
aggressive organizations, as well as exerting pressure on the incumbents to
improve their own networks. The ability to provide flexible allocation of
amounts of capacity to multiple points in the network, making more intelligent
use of scarce bandwidth resources, will be a key success factor. In this
environment, flexible access rings deploying WDM will start to become a source
of competitive advantage. 
With the introduction of optical networking techniques comes the potential for
wavelength services, opening up many possibilities for LECs. Service providers
will be able to assign the end customer a set of wavelengths and route those
wavelengths through the network independent of whatever service is being
carried. The marketing of transport on a wavelength basis will enable
to offer end users a way to enter the high-speed backbone in native formats,
avoiding the costs of aggregating multiple service types--for example,
deploying datacomm solutions such as gigabit Ethernet directly on an optical
How close is the WDM industry to this vision? There continue to be
in the price and functionality of WDM systems from vendors, but there still
challenges to overcome. In particular, concerns surrounding network
performance monitoring and vendor interoperability must be addressed before
LECs will have full confidence in optical networking solutions,
particularly as
they deploy large numbers of wavelength channels in the network. 
What is clear, however, is that the foundations for the future all-optical
network must be laid today. The way in which WDM develops in the network in
1999 will be critical to the way in which bandwidth is supplied, priced and
managed in the next decade. 
Copyright  1999 Virgo Publishing, Inc. By Barry Flanigan, senior analyst at
London-based Ovum Ltd. and lead author of Ovum's recent report, "WDM: Global
Strategies for Next Generation Networks." For more information visit This article appears in the April 1, 1999 edition of X-CHANGE
magazine. For more information about X-CHANGE, please click on the logo at the
top of this article.

Colin K. Mick
The Mick Group
2130 Hanover St,
Palo Alto, CA  94306
voice: (650) 856-3666
FAX: (650) 494-3737
email: ckm@xxxxxxxxxxxxx