Having looked at the differences between the various switching methods, we can now take a better look at the technologies used to create WANs. Several technologies, including the following, can be used to implement WANs:
Dial-up modem connections
ISDN
T-carrier lines
Fiber Distributed Data Interface (FDDI)
ATM
X.25
Frame Relay
Synchronous Optical Network (SONET)/OC-x
These technologies vary in terms of cost, complexity, and switching methods. We'll start our discussion by looking at perhaps the simplest of WAN technologies—the modem.
Today, people are more likely to associate a modem with a dial-up Internet service provider (ISP) account than with a WAN technology. But the reality is that for many years, and still today, modems have been used to provide WAN capabilities.
EXAM TIP
Modem WAN Links Modem WAN links are inexpensive and easy to implement compared to other WAN connectivity methods.
The biggest drawback of a modem connection is the speed, which is limited to 56Kbps. There are, however, a few advantages to modem connections. The cost of a modem link depends on the distance covered. In parts of the world such as North America, where local calls are free, modem links can provide an inexpensive WAN solution where there would otherwise be no WAN connectivity at all. All that is needed to create the modem WAN link is a phone line and a modem at each end of the link. You also need software to enable, support, and configure the link, but all modern network operating systems include this functionality, so this is not a problem.
ISDN is a dial-up technology capable of transmitting voice and data simultaneously over the same physical connection. Using ISDN, users are able to access digital communication channels via both packet- and circuit-switching connections. ISDN is much faster than a regular modem connection. To access ISDN, a special phone line is required, and this line is usually paid for through a monthly subscription. You can expect these monthly costs to be significantly higher than those for a dial-up modem account.
To establish an ISDN connection, you dial the number for the end of the connection, much as you would with a conventional phone call or modem dial-up connection. A conversation between the sending and receiving devices is then established. The connection is dropped when one end disconnects or hangs up. The line pickup of ISDN is very fast, allowing a connection to be established, or brought up, very quickly—much more quickly than a conventional phone line.
NOTE
B-ISDN Broadband ISDN (B-ISDN) is an enhanced version of ISDN and is capable of faster transmission rates than ISDN. It is implemented with fiber-optic media.
ISDN has two defined interface standards—Basic Rate Interface (BRI) and Primary Rate Interface (PRI)—which are discussed in the following sections.
BRI defines a communication line that utilizes three separate channels. There are two B (that is, bearer) channels of 64Kbps each and one D (that is, delta) channel of 16Kbps. The two B channels are used to carry digital information, which can be either voice or data. The B channels can be used independently to provide 64Kbps access or combined together to utilize the entire 128Kbps. The D channel is used for out-of-band signaling.
EXAM TIP
2B+D BRI ISDN is sometimes referred to as 2B+D. This abbreviation simply refers to the available channels.
To use BRI ISDN, the connection point must be within 5,486 meters (18,000 feet) of the ISDN provider's BRI service center. In addition, to use BRI ISDN, special equipment is needed, such as ISDN routers and ISDN terminal adapters. Figure 7.4 shows a standard ISDN router.
PRI is a form of ISDN that is generally carried over a T1 line (called E1 in Europe) and can handle transmission rates of up to 1.544Mbps. PRI is composed of 23 B channels (30 in Europe), each providing 64Kbps for data/voice, and one 64Kbps D channel.
NOTE
Leased Lines ISDN is considered a leased line because access to ISDN is leased from a service provider.
Table 7.2 compares BRI and PRI ISDN.
Characteristic | PRI | BRI |
---|---|---|
Speed | 1.544Mbps | 128Kbps |
Channels | 23B+D | 2B+D |
Transmission carrier | T1 | PSTN |
T-carrier lines are high-speed lines that can be leased from telephone companies. T-carrier lines can support both voice and data transmissions and are often used to create point-to-point private networks. Four distinct types of T-carrier lines are available:
T1— T1 lines offer transmission speeds of 1.544Mbps, and they can be used to create point-to-point dedicated digital communication paths. T1 lines have commonly been used for connecting LANs.
T2— T2 leased lines offer transmission speeds of 6.312Mbps. It accomplishes this by using 96 64Kbps B channels.
T3— T3 lines offer transmission speeds of up to 44.736Mbps, using 672 64Kbps B channels.
T4— T4 lines offer impressive transmission speeds of up to 274.176Mbps by using 4,032 64Kbps B channels
Of these T-carrier lines, the ones commonly associated with networks are T1 and T3 lines, which are discussed further in the following sections.
T1 (also known as a leased line) is actually a dedicated digital circuit that is leased from the telephone company. This creates an always-open, always-available line between you and whomever you choose to connect to when you establish the service. T1 lines also eliminate the “one call per wire” limitation by using a method called multiplexing, or muxing. Using a device called a multiplexer, the signal is broken into smaller pieces and assigned identifiers. Multiple transmissions are divided by the multiplexer and transmitted across the wire simultaneously. When the signals reach their destination, they are put back in the proper order and converted back into the proper form.
Having T1 service used to be the way to show someone you were serious about your particular communication needs. T1 lines were expensive; however, their prices have fallen in the past few years, as other technologies have begun to rival their transmission rates. T1 offers speeds up to 1.544Mbps. The obvious advantages of a T1 line are its constant connection—no dial-up or other connection is required because it is always on—and it can easily be budgeted because it has a fixed monthly cost. In addition, the transfer rate is guaranteed because it, like a telephone call, is a private circuit. Many companies use T1 lines as their pipelines to the Internet.
NOTE
E1 Lines In Europe, the service provided by telephone companies that is similar to the T1 service is called E1. E1 supports speeds of 2.048Mbps
For a time, the speeds offered by T1 lines were sufficient for all but a few organizations. As networks and the data they support expanded, T1 lines did not provide enough speed for many organizations. T3 service answered the call by providing transmission speeds of 44.736Mbps.
NOTE
E3 Lines In Europe, the service provided by telephone companies that is similar to the T3 service is called E3. E3 supports speeds of 34.368Mbps.
T3 lines are dedicated circuits that provide very high capacity and are generally used by large companies, ISPs, or long-distance companies. T3 service offers all the strengths of a T1 service (just a whole lot more), but the costs associated with T3 limits its use to the few organizations that have the money to pay for it.
NOTE
Fractional T Due to the cost of a T-carrier solution, it is now possible to lease portions of a T-carrier service. Known as fractional T, you can subscribe and pay for service based on 64Kbps channels.
FDDI was introduced in the mid-1980s. FDDI is an American National Standards Institute (ANSI) topology standard that uses fiber-optic cable and token-passing media access. Recall from Chapter 1, “Introduction to Networking” that the token-passing method requires systems that are sending data on the network to have access to a token. The data is attached to the token and transported throughout the network.
FDDI can be used over both multimode and single-mode fiber cable and can reach transmissions speeds of up to 100Mbps. FDDI combines the strengths of Token Ring, the speed of Fast Ethernet, and the security of fiber-optic cable. Although not widely deployed, FDDI is used for creating network backbones and connecting private LANs to create WANs.
FDDI and the IEEE 802.5 standard share some common features. For instance, both standards use a token-passing access method, and both can use fiber-optic media. However, despite their surface similarities, if you dig a little deeper, there are some significant differences.
NOTE
CDDI The Copper Distributed Data Interface (CDDI) standard defines FDDI over copper cable rather than fiber-optic cable. CDDI has an even lower level of popularity than FDDI.
As mentioned previously, the FDDI standard uses a token-passing access method similar to that of the IEEE 802.5 standard, with one very notable difference. The original 802.5 standard specifies that only a single data frame can be attached to a token. However, a computer in an FDDI network can transmit as much data on the token as possible within a specified period. When the specified time has expired, the computer releases the token to the ring, and then it must wait until the token returns before it can send more data.
Another key difference between standards 802.5 and FDDI is that FDDI uses a dual-ring configuration. The first, or primary, ring is used to transfer the data around the network, and the secondary ring is used for redundancy and fault tolerance; the secondary ring waits to take over if the primary ring fails. If the primary ring fails, the secondary ring kicks in automatically, with no disruption to network users. Figure 7.5 shows an FDDI dual-ring configuration.
Even though the second ring sits dormant, you can connect network devices to both rings. Network devices that attach to both rings are referred to as Class A stations, or dual attached stations (DASs). Network devices that connect to a single ring are called Class B stations, or single attached stations (SASs). Class A stations are the more reliable of the two because they continue to function in the event that one of the rings fails—a technique known as wrapping. SASs, on the other hand, are not fault tolerant; if the ring attached to the device fails, the device becomes isolated from the network. Figure 7.6 shows an example of DASs and SASs on an FDDI network.
NOTE
Wrapping In the FDDI topology, wrapping refers to the capability of a network device to continue to operate if one of the rings fails.
If you are implementing or troubleshooting an FDDI network, you need to keep in mind a few factors. The practical limitations of an FDDI network are 500 workstations and a maximum of 100 kilometers of cable. The FDDI specification calls for multimode fiber-optic cable with a 62.5-micrometer core. Two kilometers (6,561 feet) is the maximum cable segment length; to cover a longer distance with FDDI, a repeating device is needed every two kilometers (6,561 feet).
The FDDI standard uses a technique called beaconing to detect faults on a network. When a computer on an FDDI network detects an error, it sends a continual signal called a beacon to its immediately upstream neighbor until it hears a beacon response from the upstream neighbor. This process continues until the only computer system still beaconing is the one immediately downstream from the one with the fault. Of course, the upstream system cannot respond to the beacon because there is a fault between it and the sending computer, disabling the connection between the two devices. To identify the location of the cable break, the network administrator looks for the computer sending the beacons and then looks upstream for the problem.
FDDI has a few significant advantages, some of which stem directly from the fact that it uses fiber-optic cable as its transmission media.
The following list contains some of the advantages associated with FDDI:
Immune to electromagnetic interference (EMI)— Fiber is not susceptible to the influences of EMI.
Secure— Fiber is more secure than copper-based media. Eavesdropping and tapping into the line are far more difficult with fiber-optic cable than with copper-based cable.
Long cable distances— Fiber-optic cable has a transmission range of more than two kilometers (6,561 feet).
In addition to the advantages provided by the fiber-optic cable, FDDI itself has a few strong points:
Fault-tolerant design— By using a dual-ring configuration, FDDI is able to provide some fault tolerance. If one cable fails, the other can be used to transmit the data throughout the network.
Speed due to the use of multiple tokens— Unlike the IEEE 802.5 standard, FDDI uses multiple tokens, which increases the overall network speed.
Beaconing— FDDI uses beaconing as a built-in error-detection method, making finding faults such as cable breaks a lot easier.
FDDI also has a few key drawbacks, including the following:
High cost— The costs associated with FDDI and the devices and cable needed to implement an FDDI solution are very costly—too costly for many small organizations.
Implementation difficulty— FDDI setup and management can be very complex, requiring trained professionals with significant experience to manage and maintain the cable and infrastructure.
Because of the cost, FDDI is implemented in only a limited number of environments. As a result, your chances of encountering FDDI in the real world are relatively low.
Most of us got our first look at ATM in the early 1990s, when it was introduced. ATM was heralded as a breakthrough technology for networking because it was an end-to-end solution, ranging in use from a desktop to a remote system. Though promoted as both a LAN and WAN solution, ATM did not live up to its hype due to associated implementation costs and a lack of standards. The introduction of Gigabit Ethernet, which offered great transmissions speeds and compatibility with existing network infrastructure, further dampened the momentum of the ATM bandwagon. ATM has, however, found a niche with some ISPs and is also commonly used as a network backbone.
ATM is a packet-switching technology that provides transfer speeds ranging from 1.544Mbps to 622Mbps. It is well suited for a variety of data types, such as voice, data, and video. Using fixed-length packets, or cells, that are 53 bytes long, ATM can operate much more efficiently than variable-length-packet packet-switching technologies such as Frame Relay. Having a fixed-length packet allows ATM to be concerned only with the header information of each packet. It does not need to read every bit of a packet to determine the beginning and end of the packet. ATM's fixed cell length also makes it easily adaptable to other technologies as they develop. Each cell has 48 bytes available for data, with 5 bytes reserved for the ATM header.
ATM is a circuit-based network technology because it uses a virtual circuit to connect two networked devices. Two types of circuits are used in an ATM network:
Switched virtual circuits (SVCs)— SVCs are set up only for the duration of a conversation or data transmission. An SVC is a temporary connection that is dropped when the transmission is complete.
Permanent virtual circuits (PVCs)— PVCs are permanently established virtual circuits between two devices.
In the following sections we look at some of the characteristics of ATM.
ATM is compatible with the most widely used and implemented networking media types available today, including single-mode and multimode fiber, coaxial cable, unshielded twisted-pair, and shielded twisted-pair. Although it can be used over various media, the limitations of some of the media types make them impractical choices. ATM can also operate over other media, including FDDI, T1, T3, SONET, OC-3, and Fiber Channel.
X.25 was one of the original packet-switching technologies, but today it has been replaced in many applications by Frame Relay. Various telephone companies, along with network providers, developed X.25 in the mid-1970s to transmit digital data over analog signals on copper lines. Because so many different entities had their hands in the development and implementation of X.25, it works well on many different kinds of networks with different types of traffic. X.25 is one of the oldest standards, and therein lie both its greatest advantage and its greatest disadvantage. On the upside, X.25 is a global standard that can be found all over the world. On the downside, its maximum transfer speed is 56Kbps—which is quite reasonable when compared to other technologies in the mid-1970s but quite slow and cumbersome today.
Because X.25 is a packet-switching technology, it uses different routes to get the best possible connection between the sending and receiving device at a given time. As conditions on the network change, such as increased network traffic, so do the routes that the packets take. Consequently, each packet is likely to take a different route to reach its destination during a single communication session. The devices that make it possible to use X.25 service are called packet assemblers/disassemblers (PADs). A PAD is required at each end of the X.25 connection.
Frame Relay was designed to provide standards for transmitting data packets in high-speed bursts over digital networks, using a public data network service. Frame Relay is a packet-switching technology that uses variable-length packets. Essentially, Frame Relay is a streamlined version of X.25. It uses smaller packet sizes and fewer error-checking mechanisms than X.25, and consequently it has less overhead than X.25.
A Frame Relay connection is built by using PVCs that establish end-to-end circuits. This means that Frame Relay is not dependent on the best-route method of X.25. Frame Relay can be implemented on 56Kbps, T1, T3, and ISDN lines.
In 1984 the U.S. Department of Justice and AT&T reached an agreement stating that AT&T was a monopoly that needed be divided into smaller, directly competitive companies. This created a challenge for local telephone companies, which were then faced with the task of connecting to an ever-growing number of independent long-distance carriers, each of which had a different interfacing mechanism. Bell Communications Research answered the challenge by developing SONET, a fiber-optic WAN technology that delivers voice, data, and video at speeds in multiples of 51.84Mbps. Bell's main goals in creating SONET were to create a standardized access method for all carriers within the newly competitive U.S. market and to unify different standards around the world. SONET is capable of transmission speeds between 51.84Mbps and 2.488Gbps
One of Bell's biggest accomplishments with SONET was that it created a new system that defined data rates in terms of Optical Carrier (OC) levels. Table 7.3 contains the OC levels you should be familiar with.
OC Level | Transmission Rate |
---|---|
OC-1 | 51.84Mbps |
OC-3 | 155.52Mbps |
OC-12 | 622.08Mbps |
OC-24 | 1.244Gbps |
OC-48 | 2.488Gbps |
Table 7.4 summarizes the main characteristics of the various WAN technologies discussed in this chapter. You can use this table as an aid in reviewing before you take the Network+ exam.