LightWire's odyssey takes it to Dallas, Miami
LightWire's promise of an economical, extended use of fiber to "mini-nodes" in an architecture designed to reduce powering costs and eliminate in-line amplifiers is finally being kept after a lengthy trial and series of upgrades and tweaks.
AT&T Broadband's Salt Lake City trial of LightWire over 600 miles of plant produced the expected results and prompted the company to take the latest version of LightWire to its Miami and Dallas markets, the company says.
LightWire's odyssey began in Salt Lake in 1999. After nearly two years on trial, it is now ready to be built into other systems, albeit in select markets. LightWire extends fiber to mini nodes at interface points on existing coaxial cable plant and can eliminate all in-line amplifiers between the node and end users. This design reduces bandwidth competition over coaxial serving areas and cuts maintenance and performance issues associated with traditional coaxial amplifiers.
AT&T's initial purpose for LightWire was to cut traditional costs of powering and the number of amplifiers required when fiber is driven deeper into networks, resulting in what the company says are significant cost savings in network powering and maintenance.
"We saw the savings in the design around powering and the fewer number of coaxial amps needed. That's what we got out of the trial. We'll continue to build LightWire in Salt Lake City, and we were pleased with the outcome there, so we're building it in Miami and Dallas," says Greg Braden, executive vice president of engineering and telephony operations for AT&T Broadband.
LightWire has undergone a series of subtle upgrades and changes during its trial period. It has also prompted questions from a number of industry observers as to its cost-efficiency in underground plant versus aerial, which comprises about 70 percent of AT&T's plant.
"If you put it in aerial plant, you can get close to cost parity with a new housing development, for instance. But with underground plant that's in good condition, it can be incredibly expensive. So, it has to be used on a market-by-market basis," says Tony Werner, president and CEO of Aurora Networks and an early champion of LightWire while at AT&T.
Despite the difference in underground versus aerial costs, Werner remains bullish on LightWire's future. "Without a doubt, it will be a mainstream architecture for companies beyond AT&T. A lot of operators are looking at plant extension and going down to fiber-deep, passive architectures. LightWire is just too compelling cost-wise. It's very attractive," he says.
LightWire's cost-savings can be significant, Braden maintains. Its ROI (return on investment) is set at three to four years based on data from its Salt Lake City trial. The cost to deploy LightWire runs $40 more per home versus an HFC network, and on the back end, the savings per household is about $11 a year on the lifecycle management side.
"What we're doing differently is paying closer attention to lifecyle costs versus deployment costs," Braden notes.
Yet he admits LightWire isn't for everybody, or every system. "There's lots of work being done in Miami and Dallas relative to full upgrades, so those were natural markets to extend. Given the peculiar nature of each market, we'll apply either HFC, LightWire or OXiom as the engineering variables dictate," he says.
OXiom is the latest and greatest version of LightWire and builds on LightWire's advances. Says Braden: "It gets into the daisy chain design of mini-fiber nodes with less fiber management and easier upgrades. There's lots of development work being done on both LightWire and OXiom." AT&T will roll out OXiom in Pittsburgh this year and use it in the system's fiber extension build there, he notes.
Meanwhile, LightWire is destined for select markets which will be determined on a system-by-system basis, and there are no plans to send it to the scrap heap, the company insists. "LightWire isn't antiquated. We'll continue to see an evolution of architectural systems as economics dictate and LightWire is still fully serviceable," Braden says.
AT&T's strategy is to deploy one of three architectures going forward: traditional HFC infrastructure; LightWire; and OXiom, which is considered the most advanced architecture in the LightWire class. It provides bandwidth to the end of the network and allows easier network extensions to expand plant into new housing developments, for instance, at less cost with less fiber management.
"OXiom is based on learning from LightWire and the technology continues to evolve. Moving forward, we'll use more OXiom than LightWire," says Xiaolin Lu, VP of AT&T Broadband's engineering group.
Early discussions about the effectiveness of LightWire during its pre-trial days convinced Lu and AT&T that the architecture could indeed do what it promised.
"We did estimates before we went to Salt Lake City for the trial and we learned that field engineers liked the balancing and sweeping of the network with LightWire. It was very easy. And troubleshooting was easy, too. We expected and got a lot of data. That's why we're rolling it into other markets," Lu says.
Armed with data from the Salt Lake trial, Braden adds that AT&T is more confident about how and where LightWire can be most effective. "Now that the case has been proven (in Salt Lake), we know we can replicate it, and we'll take any information on the learning curve and use it. But to the extent where we see other technologies come along, we'll use those, too."
Both LightWire and OXiom architectures are available on the open market today, and AT&T lays no proprietary claim to either.
Customers, he notes, won't see a difference with either architecture. "From a customer perspective, their experience is the same, and we're anxious to take advantage of what LightWire and OXiom bring. Our philosophy is built on improvements we can make with our architectures to reduce costs. LightWire is step one, and OXiom is step two," Braden concludes.
Step one, Werner reflects, was the understanding that AT&T "was working with new opticals and unique architectures" which required a different way of thinking and spurred the development of LightWire.
"The genesis of LightWire was the realization that we were working with new architectures, and we could save money over traditional maintenance and operational costs when we build fiber several times deeper with fewer amps," Werner says.
The Salt Lake trial, he notes, was LightWire's rite of passage. "There were no really big surprises, only refinement going forward, and the economies came in like the company expected. It went very smoothly."
Just how smoothly LightWire's move into the Dallas and Miami markets goes is likely to depend on what Lu and Braden term "engineering variables."
"Whether we use LightWire, OXiom or traditional HFC plant depends on the network situation and the depth of the fiber architecture. We'll go into markets based on their network architecture," says Lu.
LightWire's opportunity to reduce the cost of fiber deployment and maintenance, while eliminating strings of amps, may very well allow it to fulfill its pre-trial days promises.
Contact Craig Kuhl via e-mail at firstname.lastname@example.org