Triple-play over LAN still off in the distance
It will be some time before multiple applications run over a single medium in a structured environment.
It will be some time before multiple applications run over a single medium in a structured environment.
For professionals in the information-transport systems (ITS) industry today, the word “nanometer” is relevant as a term used to describe the wavelengths at which optical signals are transported-850 nm, 1300 nm, and others.
On a much larger scale, the nanometer is being hailed as the basis for all of technology’s future. Nanotechnology, a concept that has existed for more than 20 years, is defined as the theories and techniques that permit the production and manipulation of minute objects measuring the size of atoms.
And how large is an atom? Just about one nanometer.
The promise of nanotechnology is limitless, its advocates say, citing all kinds of medical and other biological enhancements that this earth-changing technology will eventually deliver. The two scientific fields most frequently mentioned in relation to nanotech are biotechnology and information technology (IT). In the realm of IT, nanotech promises to make products lighter, more powerful, and less expensive. As it stands today, vast amounts of time, effort, and money have been invested in the promise, and much more time and effort will go in before results come out.
Last year, IT author and consultant George Spafford, in a column on EarthWeb.com entitled “Surviving the great technology convergence,” (http://itmanagement.earthweb.com/erp/article.php/3400321) compared the impending convergence of nanotechnology, biotechnology, and information technology to the convergence of video, audio, and data. Video and audio, he reminds us, can be digitized and treated as data.
That topic of discussion is much closer to home for ITS professionals than the great promise of nanotechnology. While we perhaps as much as anyone will benefit from the eventual flourishing of nanotech, for now, we will continue to be most concerned with nanometers in relation to optical transmission wavelengths and reasonably priced 850-nm vertical-cavity surface-emitting lasers.
In his column, Spafford takes an “of course” attitude toward voice, data, and video convergence. Of course it can be done; the technology to do so has existed for some time. But while it can happen, how often it actually is happening is another matter.
A broader view
Before looking at the convergence of voice, data, and video in LAN environments, we can take a more straightforward look at the concept by examining a hot-button topic in the outside-plant public network. The current initiative by Regional Bell Operating Companies (RBOCs) to bring fiber-to-the-home/fiber-to-the-premises has some form of convergence at its heart.
Once the RBOCs received assurance from the Federal Communications Commission (FCC) that they would not have to share their optical-fiber lines with competitors, they kicked their fiber-to-the-premises initiatives into high gear. The primary business proposition for these RBOCs, with relation to their residential customers anyway, is to offer telephone service, Internet access, and television/video-the so-called “triple play” of services. In many cases, residential customers subscribe to all three, representing the potential for the RBOC to pick up at least the television/video business and, in many cases, the Internet access business as well.
According to figures from KMI Research (www.kmiresearch.com), on average, consumers pay $30.54 per month for Internet access, $54.03 per month for video service, and $54.78 per month for their landline telephone service. If you add an average of $69.94 per month for wireless service-not contingent upon fiber-to-the-premises access but still offered by RBOCs-these RBOCs have the potential to generate an average of $209 per month in revenue from a single residential subscriber. The benefits to the company are obvious and, based on the recent expenditures these companies have made to install fiber deep into the public network, the revenue potential is worth the significant capital investment.
Business and technology
With those business ramifications as a backdrop, singlemode fiber is the medium that RBOCs are installing to the premises in order to deliver these triple-play services. Simply put, it’s the medium capable of carrying all three elements of the triple play over the distances required to reach customer premises. When you consider that phone companies used singlemode fiber for years as the long-haul medium to carry voice-only traffic, it is no shock that singlemode is the one that meets the requirements for their new initiatives.
Richard Mack, vice president and general manager of KMI Research, has perspective on the financial and technical aspects of long-haul singlemode systems. “Singlemode optical fiber actually is less expensive than multimode,” he explains. Traditionally, he adds, the cost of connectivity for singlemode optical-fiber cable has been higher than for multimode cable.
“The difference between singlemode and multimode connectivity has dropped as the prices for both have dropped,” he explains. “Singlemode terminations are not that much more than multimode terminations. Today, multimode terminations are less than a dollar apiece while singlemode terminations are less than a dollar-and-a-half apiece.”
The major price chasm, of course, is in transmit-and-receive devices where, on average, the difference between short-wave (multimode) and long-wave (singlemode) is tens of dollars per connection.
But does the fiber-to-the-premises scenario analogize to what goes on in a single-building or campus LAN? In many cases it does not, for one reason: distance.
Campus networks that are not exorbitantly lengthy might be able to use multimode rather than singlemode fiber, even if they are trying to emulate RBOCs by sending triple-play services over a single medium. The term “multimode” comes from that type of fiber’s ability to carry more than one (multi) waves (modes) of light down its core. It has significant bandwidth potential, and carries laser-based signals that are generated on a lower end of the nanometer scale (typically, 850 nm) than the signals that singlemode fiber carries (typically 1310 or 1550 nm).
The flipside for multimode is that it can support such transmission only at much shorter distances than can singlemode. In a campus LAN environment, however, multimode just might have enough distance capability to support a user’s needs.
“For premises distances, up to a 2-kilometer campus, the bandwidth-by-distance equation for multimode might do the user OK,” adds Mack.
He adds that in a triple-play transmission package, video delivery is the wild card. “Video transmission is variable, and can be anywhere from 6 to 40 Mbits. Most office applications do not need high-resolution video, thereby decreasing the amount of bandwidth needed.” In any event, he says, “singlemode to the horizontal is not needed.” Horizontal cabling distances, which have maximum limits per the specifications of the Telecommunications Industry Association (www.tiaonline.org), fall within multimode fiber’s distance capabilities.
Toes in the water
The prospect of triple-play-over-LAN is not pure fantasy, at least according to those who run the University of New Hampshire’s Interoperability Lab (www.iol.unh.edu). In March, three of the lab’s consortia collaborated to run week-long tests on the multi-vendor transmission of voice, data, and video services over a variety of media, including shielded and unshielded twisted-pair, multimode fiber, and wireless.
Gerard Goubert manages UNH-IOL’s Wireless and VoX (Voice over Anything) Consortia, two of the three involved in last month’s trials. The lab’s Bridging Consortium also participated. Goubert explains that the testing was not designed to determine which medium best handles triple-play services, but was more like a test of testing. “All the infrastructure, including the cable and fiber plant, had been specified and tested to ensure they were good,” he says.
Beyond that, no measurement was taken specifically to gauge each medium’s transmission performance. The ambitions of this testing were aimed at higher network layers than the physical. “We used a mix of different protocols,” Goubert says. “The interesting thing is, people have tested voice, video, and data. But they have done it all separately. The point of this testing was to plug it all in and test it.”
Testing, then, became a primary focus of this experiment. Indeed, the list of vendors whose equipment was used reads like a sampling of network testing manufacturers and includes Azimuth (www.azimuth.net), ClearSight Networks (www.clearsightnet.com), Fluke Networks (www.flukenetworks.com), Spirent Communications (spirentcom.com), and Veriwave (www.veriwave.com). As its name indicates, UNH-IOL spends significant time researching the interoperability among different pieces of equipment, including, in this case, test equipment.
Not ready for prime time
The fact that some of the most pertinent results of these week-long lab experiments relate to interoperability among test equipment is an indication that a triple-play package of services in a LAN environment is very much in its infancy. By any measure, the convergence of all three technologies is not a standardized, structured offering.
Shortly after the testing ended, Goubert was quoted as saying, “Enterprises are already deploying VoIP, wireless, and even triple play by adding to their networks piece by piece. But until now there hasn’t been a neutral proving ground where all the companies could put aside their differences and make sure they work with each other. Obviously, these are very heterogeneous, complex networks. What’s needed to instill confidence on the IT side is a forum ground where the vendors can come together in a realistic mixed environment and fix anything that’s less than optimal, ideally before their products get into the field.”
Perhaps somewhere within that proving ground will be indications of the media types’ capabilities within a triple-play environment.
Editor’s note: This article cites research figures from and quotes an executive of KMI Research. KMI is owned by PennWell, which publishes Cabling Installation & Maintenance.
PATRICK McLAUGHLIN is chief editor of Cabling Installation & Maintenance.