Metro Access: A Mixed Bag
According to "Optical Access in the Public Network," a report from Communications Industry Researchers Inc., gigabit Ethernet, 10-Gbit/s Ethernet, and techniques for running IP directly over lightwaves will comprise at least 40 percent of ports shipped for metro access in the U.S. by 2004.
The result will be a mixed bag in the metro space, where bandwidth bottlenecks are forming as a result of carriers' inability to support faster, more sophisticated transmission methods in the optical backbone or the customer premises.
"The metro/access space has not kept pace with developments," the report states. "With multi-gigabit premises networks and multi-terabit long-haul networks already a reality, the bandwidth crunch in the metro/access network can only get worse."
As a result, CIR says, vendors are going to build on a trend that's already begun, by continuing to offer products that support a range of access options for metro customers, some based on Sonet, some on other techniques, such as putting IP directly over fiber, without requiring framing by Sonet or any other protocol.
In fact, the big loser in metro access, CIR says, is ATM, which has proven too expensive and complicated to be a good access solution.
Some recent developments seem to validate CIR's claim. Maple Optical Systems, for instance, is working on a product that uses MPLS (multiprotocol label switching) to direct IP over fiber without ATM or Sonet (see Maple Gets $50M for MPLS Box).
Progress will not be made at the cost of Sonet, however -- at least in the short term. CIR says Sonet will continue to grow at double-digit rates for the next five years, thanks to carriers' existing investments and their need for Sonet's reliability.
But Sonet will change. It will get faster, with OC192 taking off as products emerge in 2003. Sonet also will start to become part of the overall mixed-bag solution to the access problem, instead of the solution itself. Products that support Sonet also will support a variety of other service options, such as gigabit Ethernet and native IP over light.
This kind of multipronged approach to access will allow equipment vendors to sell one platform to a range of carriers, including incumbents who have a lot invested in Sonet, as well as CLECs who wish to peddle gigabit- and 10-Gbit/s Ethernet services as a Sonet alternative.
Bottom line? There will be a jumble of access methods in the metro space for the foreseeable future -- and an environment where Sonet remains robust.
-- Mary Jander, senior editor, Light Reading http://www.lightreading.com