Some articles on the market for optical components and Metro
DWDM.
Worldwide Market for Optical Components To Reach $23B by 2003
Posted on March 14, 2000
The worldwide market for optical components used in long distance telecom and CATV applications will grow from $6.6 billion in 1999 to more than $23 billion by 2003, so says San Francisco-based telecom market research firm RHK.The terrestrial DWDM segment of this market will expand fastest, growing more than 50 percent a year. This growth in optical components is driven by demand for transmission at speeds of at least 2.5 Gbps, significant advances in price-to-performance ratios in optical components and strong traffic growth.RHK's research looks at components used in terrestrial and submarine systems including: sonet/SDH, cable television systems, WDM/DWDM and optical networks. While sonet/SDH component demand drives the market in 2000, the demand for WDM/optical networking components will exceed all other segments by 2002.
"All the market segments that RHK analyzed are benefiting from strong growth in traffic, which has surpassed voice traffic in volume," says Jay Liebowitz, director of RHK's optical components service. "We will continue to see service providers deploy additional capacity as rapidly as possible through 2003 in an effort to keep pace with demand and competition."
Other Fiber Optics News Headlines:
telecomdirect.pwcglobal.com
Will Metro DWDM Pull a Vanishing Act?
Posted on January 4, 2000
After a Couple of Years as the Next Big Thing, Metro DWDM now has Detractors
A few years ago, dense wavelength-division multiplexing (DWDM) suddenly emerged in the optical long-haul market as if from a puff of smoke.
The traffic demands for Internet data were beginning to explode, and the likes of Sprint Corp. and MCI WorldCom Inc. faced an onslaught of bandwidth-hungry customers. Promised OC-192 technology to boost the bandwidth of synchronous optical network (SONET) was delayed. In an industry where technology changes very slowly, DWDM, which multiplies the capacity of optical networks by transmitting on multiple colors, or wavelengths, of light, found
overnight success.
It seemed DWDM was destined to migrate with equal explosiveness to metropolitan area networks (MANs). There, fiber exhaust is an issue for many carriers, as is the need to carry a variety of services, each on their own wavelengths. But two years after the introduction of the first metro DWDM gear, the technology is still looking for
broad acceptance. Predictions of metro DWDM sales are being
revised downward daily--some foresee as little as $1 billion in sales by 2005--and an atmosphere of gloom has settled over the market.
Cost Conscious
Where DWDM lost momentum was over the issue of cost. Whereas
the technology provided an immediate cost benefit over stringing thousands of miles of new fiber in the long-haul, fiber is cheap and abundant in metro environments. The high cost of the equipment initially offered as "metro DWDM" left no margins for service providers, in part because it was only slightly modified from the long-haul onfigurations.
"In the local exchange, service providers will sacrifice waves for cost," says Mat Steinberg, optical network analyst with RHK Inc., a South San Francisco, Calif., consulting firm. "In the backbone there is a big channel war going on, but in the local part of the network it is not about channels. They are important, but it is about keeping the cost down."
Randolph Nicklas, director of data architecture with NEXTLINK Communications Inc., Bellevue, Wash., says DWDM is considered only when fiber is used up. "If we are looking at fiber exhaust, and we haven't done any build, it is of interest. But, for us, why add cost when we can just pick up another pair of fiber strands in our bundle? I'm not sure that makes lot of sense."
Metro-specific designs emerged in 1999 from the established
providers--Alcatel Network Systems, Richardson, Texas; Ciena Corp., Linthicum, Md.; Fujitsu Network Communications Inc., Richardson, Texas; Lucent Technologies Inc., Murray Hill, N.J.; Nortel Networks Inc., Richardson, Texas; Optical Networks Inc., San Jose, Calif.; and Tellabs Inc., Lisle, Ill.--amid predictions the market would finally take off. But as the year draws to a close, service providers are still waiting or trying other solutions.
Ted Huf, vice president of engineering, Adelphia Business Solutions, Coudersport, Pa., says the company opted for OC-192 on a recent project instead of DWDM. "When we made the first evaluation, we didn't feel that the product had really come into being, and prices were still high. So we made the decision to put in OC-192 SONET, and later
as we need more bandwidth, we can upgrade to DWDM."
The question is how cheap can metro DWDM become. Mark
Lutkowitz, president, Trans-Formations Inc., a Birmingham, Ala., research and consulting firm specializing in the transmission equipment market, says margins are already squeezed on long-haul DWDM. "Ciena is the only one where we can look at [the financials] publicly, and their margins are nothing to write home about," he says. "On the metro side, cost is even more of an issue. I have to wonder who will make profits on metro DWDM when all is said and done."
Despite the slow short-term development, Lutkowitz is still among those who see the long-term likelihood that DWDM will dominate the market in metro as it does in long haul, and that the metro market will ultimately be much larger than long haul. Lutkowitz adds that established vendors, such as Nortel, seem to realize that returns from metro sales will be limited, at least for the near term, and are trying to sell as much SONET as possible in the metro space.
Managing Blindfolded
Besides price, the shortcoming of metro DWDM systems that
troubles operators most is the perceived primitive state of monitoring and management tools. Metro DWDM users today often complain that trying to manage a DWDM network is like doing magic tricks while wearing a blindfold. The technology gives its users precious little "view" into the network. Service providers know if the network is on or off, and with a ring topology, there is some protection switching
capability in addition to the protection the SONET muxes on each wavelength afford.
"If you can't understand on a wave-by-wave basis how the system is performing," says Nick Tanzi, senior vice president and chief operating officer for Metromedia Fiber Network Inc., White Plains, N.Y., "how do you in turn determine when problems are emerging?" Tanzi adds the problem is exacerbated in the metro arena because "in
the metro area you have very mission-critical, high-bandwidth applications."
Some service providers feel the next big opportunity for vendors in this arena is to build network management platforms that allow service providers to manage each wave on a very granular basis. The ability to subdivide a fiber into more waves adds efficiency and eventually will improve the cost/performance ratio, but service providers will also
need tools to manage that new environment.
Read more
telecomdirect.pwcglobal.com
Posted: 12/15/1999
Will Metro DWDM Pull a Vanishing Act?
After a Couple of Years as the Next Big Thing, Metro DWDM now has Detractors
By Charlotte Wolter
Cost Conscious
Where DWDM lost momentum was over the issue of cost. Whereas the technology provided an immediate cost benefit over stringing thousands of miles of new fiber in
the long-haul, fiber is cheap and abundant in metro environments. The high cost of the equipment initially offered as "metro DWDM" left no margins for service providers, in part because it was only slightly modified from the long-haul configurations.
"In the local exchange, service providers will sacrifice waves for cost," says Mat Steinberg, optical network analyst with RHK Inc., a South San Francisco, Calif., consulting firm. "In the backbone there is a big channel war going on, but in the local part of the network it is not about channels. They are important, but it is about keeping
the cost down."
Randolph Nicklas, director of data architecture with NEXTLINK Communications Inc.,
Bellevue, Wash., says DWDM is considered only when fiber is used up. "If we are looking at fiber exhaust, and we haven't done any build, it is of interest. But, for us, why add cost when we can just pick up another pair of fiber strands in our bundle? I'm not sure that makes lot of sense."
Metro-specific designs emerged in 1999 from the established providers--Alcatel Network Systems, Richardson, Texas; Ciena Corp., Linthicum, Md.; Fujitsu Network Communications Inc., Richardson, Texas; Lucent Technologies Inc., Murray Hill, N.J.; Nortel Networks Inc., Richardson, Texas; Optical Networks Inc., San Jose, Calif.; and Tellabs Inc., Lisle, Ill.--amid predictions the market would finally take off. But as the year draws to a close, service providers are still waiting or trying other solutions.
Ted Huf, vice president of engineering, Adelphia Business Solutions, Coudersport, Pa., says the company opted for OC-192 on a recent project instead of DWDM. "When we
made the first evaluation, we didn't feel that the product had really come into being, and prices were still high. So we made the decision to put in OC-192 SONET, and later
as we need more bandwidth, we can upgrade to DWDM."
The question is how cheap can metro DWDM become. Mark Lutkowitz, president, Trans-Formations Inc., a Birmingham, Ala., research and consulting firm specializing in
the transmission equipment market, says margins are already squeezed on long-haul DWDM. "Ciena is the only one where we can look at [the financials] publicly, and their margins are nothing to write home about," he says. "On the metro side, cost is even more of an issue. I have to wonder who will make profits on metro DWDM when all is said and done."
Despite the slow short-term development, Lutkowitz is still among those who see the long-term likelihood that DWDM will dominate the market in metro as it does in long haul, and that the metro market will ultimately be much larger than long haul. Lutkowitz adds that established vendors, such as Nortel, seem to realize that returns from metro sales will be limited, at least for the near term, and are trying to sell as much SONET as possible in the metro space.
Managing Blindfolded
Besides price, the shortcoming of metro DWDM systems that troubles operators most is the perceived primitive state of monitoring and management tools. Metro DWDM users today often complain that trying to manage a DWDM network is like doing magic tricks while wearing a blindfold. The technology gives its users precious little "view" into the network. Service providers know if the network is on or off, and with a ring topology, there is some protection switching capability in addition to the protection the SONET muxes on each wavelength afford.
"If you can't understand on a wave-by-wave basis how the system is performing," says Nick Tanzi, senior vice president and chief operating officer for Metromedia Fiber
Network Inc., White Plains, N.Y., "how do you in turn determine when problems are emerging?" Tanzi adds the problem is exacerbated in the metro arena because "in the
metro area you have very mission-critical, high-bandwidth applications."
Some service providers feel the next big opportunity for vendors in this arena is to build network management platforms that allow service providers to manage each wave
on a very granular basis. The ability to subdivide a fiber into more waves adds efficiency and eventually will improve the cost/performance ratio, but service providers will also need tools to manage that new environment.
Metro DWDM Applications
While metro DWDM as a general network strategy is being questioned, the technology has become established in specialized applications. It still plays its traditional role of relieving situations where a fiber's capacity has been subscribed and it would be too expensive to string new fiber. The problem in the local environment has been that its intended customers, namely competitive local exchange carriers (CLECs), as new operators, have constructed their networks with abundant fiber. The Bell
operating companies (BOCs), by contrast, who are sometimes short on legacy fiber, have been reluctant to try the new technology except where fiber exhaust makes it mandatory.
A typical CLEC user would be an operator such as e.spire Communications Inc., Annapolis Junction, Md., which has strung its own 144-fiber bundles along the long
stretches between its Washington and New York networks.
However, within New York, the company has only 16 leased fibers. There,the density of the locale and the need to get the most out of a few fiber strands provided the motivation to go to metro DWDM no matter what the cost, says Albert Ng, vice president of engineering.
Data services, specifically gigabit Ethernet, can be transmitted directly on a DWDM wave without intervening SONET equipment. Gigabit Ethernet is an optical technology
so it can be transmitted directly at a significant cost savings. Also, at 1 gigabit per second (gbps), gigabit Ethernet does not fit neatly into any of the SONET multiplexes, being too big for OC-12 (622 megabits per seconds [mbps]) and too small for OC-48 (2.5gbps).
"Gigabit Ethernet manufacturers are producing what are, in effect, bit-rate-independent transponders for DWDM systems," says Mike Guess, vice president of engineering, IXC Communications Inc., Austin, Texas. "So you have a transponder that is not fixed to a SONET rate and is not looking for an OC-48 input, for example. It is looking just for an optical input and will clock its laser off whatever is the optical input."
However, Guess points out, the optical character that is so advantageous for gigabit Ethernet isn't there for the two other most common local area network (LAN) traffic data types, 100Base-T and 10Base-T, both of which are electrical. Also, DWDM is much more effective at high data rates, "and gigabit Ethernet is at the low end of what we consider high-speed nowadays," says Guess. Fujitsu has equipment that
multiplexes 10Base-T streams for SONET. "You really need a combination of those two solutions [DWDM and new approaches to SONET] to make a data-friendly metro [DWDM] optical network."
Metromedia, which built its business leasing dark fiber for private networks, is now beginning to sell DWDM wavelengths, especially in corporate data applications. One of the most in-demand services is transparent LAN, in which a wide area network (WAN) is used to link offices of a company such that there appears to be a single LAN.
With metro DWDM, it is just becoming feasible in terms of cost to offer transparent LAN at high bandwidths. Companies now have 100Base-T or even gigabit Ethernet backbones within the enterprise, but can afford only 45mbps DS-3 connections
between locations. "With DWDM, they keep the same router and switch [on their premises]," says Tanzi. "Instead of a DS-3 trunk card, they put in an OC-12 and, voila, they have a 622mbps connection between facilities."
ith a dark fiber pricing model, the cost of the wave can remain the same even if the customer later chooses to upgrade to an OC-48.
ADVA Optical Networking Inc., the U.S. subsidiary of ADVA Optical Networking AG, Munich, Germany, offers DWDM solutions designed for point-to-point data applications. It transports such protocols as gigabit Ethernet, fibre channel and Enterprise Systems Connection architecture (ESCON), IBM's fiber optic architecture. Priced for environments where carriers lease dark fiber for $2,000 a month, ADVA
claims the system will be cost-effective with four channels lit per fiber. The units are priced at approximately $25,000 per wavelength. If a carrier charges $2,500 a month for a
managed DWDM wavelength, the system can pay for itself within 18 months with four channels sold.
Metromedia is supporting the same services on Nortel's Optera product. "We use Optera to allow customers to create end-to-end optical infrastructure. It allows them to
take disparate network protocols and collapse them over one transport system, and it allows them to scale infrastructure from one channel to 32 channels, or from 10 gigabits
to 320 gigabits, so there is scalability embedded in the platform.
x-changemag.com
|
