The demand for increased bandwidth is an ever-growing need for network engineers and a nightmare for end users. The complex graphics, interactive courseware and instructional gaming technologies that are now available tend to bottleneck an educational network and leave teachers and students feeling frustrated with their systems. It is estimated that "bandwidth requirements will grow by a factor of 100 to 200 times in the next two years," according to Anil Khatod, president of the Optical Networking division at Nortel Networks (McGarvey, Oct. 18, 1999). Nortel recently introduced technology that will enable fiber optic lines to deliver 6.4 terabits of data per second. Since most of us would be happy just to have 100Mbps to the desktop the thought of 6.4 terabits flying around almost defies comprehension yet it will most likely get faster as technology improves. How is it that Nortel can promise a terabit rate? What exactly does that mean to an educator in today’s classroom? This paper seeks to answer those questions by identifying what is in use today and then delineate what the newest advances will enable a person to do.

The network backbones, the basic structures that connect us on the Internet are composed of fiber optic cable. A fiber optic cable consists of a bundle of glass threads, each of which is capable of transmitting messages modulated onto light waves. Currently the Synchronous Optical Network (SONET) has been in wide use and Asynchronous Transfer Mode (ATM) also continues in use. The following is a very brief description of both technologies.

Asynchronous Transfer Mode (ATM) is a network technology based on transferring data in cells or packets of a fixed size. The small, (53 bytes - 48 bytes) constant cell size allows ATM equipment to transmit video, audio, and computer data over the same network, and assure that no single type of data takes over the line.

Some people used to think that ATM held the answer to the Internet bandwidth problem. However, Ethernet can now handle Gigabit speeds and the promise that ATM held did not turn out to be the panacea that people were expecting. ATM works by creating a fixed route between two points whenever data transfer begins. This differs from Transmission Control Protocol/Internet Protocol (TCP/IP) in which messages are divided into packets and each packet can take a different route from source to destination. This difference makes it easier to track data usage across an ATM network, but it makes it less adaptable to sudden surges in network traffic.

The ZD Webopedia gives the following definition of SONET. "SONET defines interface standards at the physical layer of the Open System Interconnection (OSI) seven-layer model. The standard defines a hierarchy of interface rates that allow data streams at different rates to be multiplexed. Multiplexing allows several different data streams to be carried on one line. Dense Wave Division Multiplexing (DWDM) used in optical networks allows transmission signals to be carried on various wavelengths. SONET establishes Optical Carrier levels (OC) from 51.8 Mbps to 2.48 Gbps. Prior rate standards used by different countries specified rates that were not compatible for multiplexing. With the implementation of SONET, communication carriers throughout the world can interconnect their existing digital carrier and fiber optic systems." http://www.zdwebopedia.com.

This has certainly boosted bandwidth and carrier efficiency in the past several years. However, the wavelength signals still had to be converted to electrical signals so they could be dropped off or added to another network depending on how that network was configured. This conversion requires another piece of equipment called an Add-Drop multiplexer (ADM). Prior to the use of an ADM the entire signal had to be converted and also brought down to the lowest transmission speed thus creating bottlenecks in the network.

The use of IP over SONET and ATM is discussed in a White Paper by the Communications Industry Researchers, Inc. (CIR). The paper can be found at the following link:

http://www.cir-inc.com/registered/whitepapers/op_net_wp.asp . The CIR notes that in IP over SONET data packets are no longer condensed into an ATM cell for transfer on the SONET. Rather, data packets are directly mapped to the SONET frame eliminating the ATM portion. By doing this the extra data that is identified with the ATM cell is eliminated. This extra data can account for 10 percent of the available bandwidth according to the CIR. So using IP over SONET can gain some bandwidth.

The following diagram from the White Paper depicts the bandwidth efficiencies of SONET and ATM.

The Bandwidth Efficiencies of IP-over-SONET on an OC-3 Link
 

	IP-over-ATM	IP-over-SONET
Nominal speed link	155.52 Mbps	155.52 Mpbs
Speed after SONET management overhead	149.76 Mbps	149.76 Mbps
Speed after ATM cell tax	135.63 Mbps	N/A
Speed after IP overhead	123 Mbps	135 Mbps
Advantage to IP-over-SONET = 12 Mbps

The authors point out that a gain of 12 Mbps does not seem to be much, "but if you think of it as eight T1’s gone to waste, the loss appears more significant" (CIR, p.2).

So what is on the horizon to break up the continuing problem of bottlenecks as more data is passed across the networks? Many companies today are espousing the benefits of an all -optical network.

This picture depicts a possible set-up for the optical network configuration. Data comes in from another network, possibly a LAN, via the IP router or ATM switch. It runs through the optical switch to the dense wave division multiplexing terminal where it is changed to photonic form for high-speed transfer across the optical network. According to McGarvey, (October 18,1999), Nortel Networks is using optical equipment to create 80 separate wavelengths each capable of carrying data at the rate of 80 gigabits per second on one fiber strand. Mathematically that calculates out to an astounding rate of 6.4 terabits per second. The key to this conduit is the movement of the light waves along the data path. Dense Wave Division Multiplexing (DWDM) equipment is what is used to put so many wavelengths onto the fiber strands. Tunable lasers have made it possible to create very precise wavelengths thereby increasing the number that can be placed on a fiber cable. The wavelengths still have to cross long distances, such as Chicago to Texas, so optical amplifiers are used to boost the wavelength along its path. The amplifiers are fiber that has been fortified with Erbium. Erbium is an element that strengthens the optical wavelength as it goes along its path. Previously, signals had to be regenerated as they moved over long distances. By using an optical amplifier the signal regeneration equipment is removed so operation and maintenance costs are reduced (Wilson, 8 June, p2.). The bottlenecks still occur where data has to be converted. By keeping the data in photonic form until the last possible switch data will move faster across the network. Two companies, Monterey Networks (www.montereynetworks.com) and Tellium (www.tellium.com) are already producing the optical switches that will be used to move data to the network edge. However, there will continue to be a bottleneck once the data reaches the final switch and has to be converted back to electronic form. Perhaps sometime in the not too distant future, 50 years or so, networks will be completely optical. Companies involved in optical networking concede that the vision is still a bit ahead of the technology. It has been said that optical technology is currently where electronic technology was back in the 1950’s (Inter@active Week, 25 Oct 99, p.11). Given the giant leaps that have been made in the last decade it may be assumed that optical technology will advance at a much quicker rate. For more information on optical networks the following sites offer basic tutorials for the optical networking neophyte. http://www.zdnet.com/intweek/supplements/optical/intro.html

http://fiddle.ee.vt.edu/courses/ee4984/proj_95/choudary.html

While all the technological advances continue what do they mean for educational institutions? Industry and universities all over the world are collaborating on projects that will mesh industry’s need for trained personnel and the universities need for research. In Canada, the Ontario Regional Area Network, Onet Networking, was established in 1988 as " a not-for-profit Ontario corporation that exists to obtain sophisticated and cost-effective network facilities for use by its members" http://www.onet.on.ca/onetabout.html . They have established the beginnings of an optical network that serves the entire province from industry to education to healthcare. Some of the applications for education include videoconferencing, distance education, and packetized voice transmission. On the health care side the access to medical imaging and remote diagnostics will now be possible. The slide presentation can be viewed at: http://enfm.utcc.utoronto.ca/c2/onetii
 

Closer to home but nonetheless global in scope is the Internet2 project. The mission of Internet2 is to "facilitate and coordinate the development, deployment, operation and technology transfer of advanced, network-based applications and network services to further U.S. in research and higher education and accelerate the availability of new services and applications on the Internet" (http://www.internet2.edu/html/mission.html#). One of the many objectives is to demonstrate the enhanced delivery of education that can occur when the infrastructure can support such mediums. A recent meeting of the technical working group in June 1999 noted a discussion of optical networks and how wave division multiplexing will enable the engineers to go beyond the limitations of fiber as it now used. A conference attendee noted that a few years ago the tele-presentation that was given would not have been possible but now is possible due to the gigabit Ethernet set-up that was in use.  The collaboration that is now possible through the current research networks that support the Internet2 program is astounding. Scientists can send huge volumes of data across networks to colleagues thousands of miles away, equipment sharing is being done "over the wire", and peers collaborate in real-time without the jerkiness of <30fps video delivery. In medical institutions the patient can be in a hospital across country while the surgeon operates on him from another state. Our minds have yet to realize the full potential of the growing network especially as we add data that transfers at light speed.  While some educators are still trying to figure out how to get a PowerPoint presentation to work the technological innovations march on. It behooves every teacher to at least stay abreast of what is happening in the research and development arena. The very latest technical advances will take time to trickle down to universities and school systems but the flow will not be as slow as it once was. As the optical backbones are put in place the advances to the desktop will not be far behind. Reaching out and touching your neighbor across the ocean without leaving your classroom will be akin to the first transatlantic phone call. It is destined for the new millennia.     References Choudary, A. (1995). EE 4984 Telecommunications project 1: All optical networks. [On-Line]. Available: http://fiddle.ee.vt.edu/courses/ee4984/proj_95/choudary.html Communications Industry Researchers, Inc. (1999). Three approaches to optical networking in an era of convergence. [On-Line]. Available: http://www.cir-inc.com/registered/whitepapers/op_net_wp.asp Internet2 Project. [On-Line]. Available: www.internet2.edu McGarvey, J. (1999, October 11). Lighting up the net’s core. Interactive Week, 6, 52. McGarvey, J. (1999, October 18). Nortel breaks bandwidth boundary. Interactive Week, 6, 18. McGarvey, J. (1999, October 25). Start-up claims optical network breakthrough. Interactive Week, 6, 11. Mokbel, S. (1999). Ontario’s optical network: Building the SMART province. [On-Line]. Available: http://enfm.utcc.utoronto.ca/c2/onetii Wilson, C. (1997, June 2). Optical networking: Technology on the fast track. [On-Line]. Available: http://www.zdnet.com/intweek/print/970602/inweek00001.html Wilson, C. (1998, June 8). Optical networking 101. [On-Line]. Available: http://www.zdnet.com/intweek/supplements/optical/intro.html Wilson, C. (1999, October 25). Quantum cracks an optical window. Interactive Week, 6, 11.