CompTIA Network+ N10-006 Quick Reference: WAN Technologies
Date: Aug 13, 2015
Even if your company is very small, it will not take long before you have employees or branch offices that cannot be connected with fast and relatively inexpensive local area network (LAN) equipment and cabling. This is unfortunate, because “fast and inexpensive” are great terms to hear in computer networking. Thankfully, there have been many advancements in wide area networking (WAN) that make it more tolerable and affordable. Whenever you hear WAN, think about distances beyond the floor of your office building. You might even want to think of the global Internet itself, and don’t forget about those satellites beaming signals from space!
Fiber
WAN connections that require a high bandwidth capacity or the capability to span a large distance might use fiber-optic cabling. In addition to the massively long distances that are supported, fiber-optic cabling provides great immunity from electromagnetic interference (EMI).
Synchronous Optical Network (SONET) is a Layer 1 technology that uses fiber-optic cabling as its media. Because SONET is a Layer 1 technology, it can be used to transport various Layer 2 encapsulation types, such as Asynchronous Transfer Mode (ATM). And because SONET uses fiber-optic cabling, it offers high data rates, typically in the 155 Mbps to 10 Gbps range, and long-distance limitations, typically in the 20 km to 250 km range. Optical carrier transmission rates, such as OC3 (close to 155 Mbps) and OC12 (close to 622 Mbps) are examples of specifications for digital signal transmission bandwidth.
The term SONET is often used synonymously with the term Synchronous Digital Hierarchy (SDH), which is another fiber-optic multiplexing standard. Although these standards are similar, SONET is typically utilized in North America, whereas SDH has greater worldwide popularity.
A SONET network can vary in its physical topology. For example, devices can connect as many as 16 devices in a linear fashion (similar to a bus topology) or in a ring topology. A metropolitan-area network (MAN) often uses a ring topology. The ring might circumnavigate a large metropolitan area. Sites within that metropolitan area could then connect to the nearest point on the SONET ring.
A SONET network uses a single wavelength of light, along with time-division multiplexing (TDM) to support multiple data flows on a single fiber. This approach differs from dense wavelength division multiplexing (DWDM), which is another high-speed optical network commonly used in MANs. DWDM uses as many as 32 light wavelengths on a single fiber, and each wavelength can support as many as 160 simultaneous transmissions using more than eight active wavelengths per fiber. Coarse wavelength division multiplexing (CWDM) uses fewer than eight active wavelengths per fiber. Current standards make even more transmissions possible.
Frame Relay
Frame Relay is beginning to decline but is still worthy of inclusion here, especially when you consider regions of the globe outside the United States.
Frame Relay sites are interconnected using virtual circuits (VC). So a single router interface can have multiple VCs. Frame Relay is a Layer 2 technology, and a router uses locally significant identifiers for each VC. These identifiers are called data-link connection identifiers (DLCI). Because DLCIs are locally significant, DLCIs at the different ends of a VC do not need to match (although they could).
If a VC is always connected, it is considered to be a permanent virtual circuit (PVC). However, some VCs can be brought up on an as-needed basis, and they are referred to as switched virtual circuits (SVC).
Unlike a dedicated leased line, Frame Relay shares a service provider’s bandwidth with other customers of its service provider. Therefore, subscribers might purchase an SLA (previously described) to guarantee a minimum level of service. In SLA terms, a minimum bandwidth guarantee is called a committed information rate (CIR).
During times of congestion, a service provider might need a sender to reduce his transmission rate to the CIR. A service provider can ask a sender to reduce his rate by setting the backward explicit congestion notification (BECN) bit in the Frame Relay header of a frame destined for the sender that needs to slow down. If the sender is configured to respond to BECN bits, it can reduce its transmission rate by as much as 25 percent per timing interval (which is 125 ms by default). CIR and BECN configurations are both considered elements of Frame Relay Traffic Shaping (FRTS). A device that does packet shaping is referred to as a packet shaper.
Another bit to be aware of in a Frame Relay header is the discard eligible (DE) bit. Recall that a CIR is a minimum bandwidth guarantee for a service provider’s customer. However, if the service is not congested, a customer might be able to temporarily transmit at a higher rate. However, frames sent in excess of the CIR have the DE bit in their header set. Then, if the Frame Relay service provider experiences congestion, it might first drop those frames marked with a DE bit.
Satellite
Some locations do not have WAN connectivity options, such as DSL connections or cable modems, commonly available in urban areas. However, these locations might be able to connect to the Internet or to a remote office, using satellite communications. This occurs when a transmission is bounced off of a satellite, received by a satellite ground station, and then sent to its destination using either another satellite hop or a wired WAN connection.
Broadband Cable
Cable television companies have a well-established and wide-reaching infrastructure for television programming. This infrastructure might contain both coaxial and fiber-optic cabling. Such an infrastructure is called a hybrid fiber-coax (HFC) distribution network. These networks can designate specific frequency ranges for upstream and downstream data transmission. The device located in a residence (or a business) that can receive and transmit in those data frequency ranges is known as a cable modem.
The frequency ranges typically allocated for upstream and downstream data are 5 MHz to 42 MHz upstream and 50 MHz to 860 MHz downstream.
Although the theoretical maximum upstream/downstream bandwidth limits are greater (and dependent on the HFC distribution network in use), most upstream speeds are limited to 2 Mbps, with downstream speeds limited to 10 Mbps. As HFC distribution networks continue to evolve, greater bandwidth capacities become available.
The frequencies dedicated to data transmission are specified by a Data-Over-Cable Service Interface Specification (DOCSIS) version. Although DOCSIS is an international standard, European countries use their own set of frequency ranges, their own standard known as Euro-DOCSIS.
DSL/ADSL
Commonplace in many residential and small-business locations (also known as small office/home office or SOHO locations), digital subscriber line (DSL) is a group of technologies that provide high-speed data transmission over existing telephone wiring. DSL has several variants, which differ in data rates and distance limitations.
Three popular DSL variants are
- Asymmetric DSL (ADSL)
- Symmetric DSL (SDSL)
- Very-high-bit-rate DSL (VDSL)
Asymmetric DSL (ADSL) is popular Internet-access solution for residential locations. Note that ADSL enables an existing analog telephone to share the same line used for data for simultaneous transmission of voice and data. The maximum distance from a DSL modem to a DSL access multiplexer (DSLAM) is 18,000 feet. This limitation stems from a procedure that telephone companies have used for decades to change the impedance of telephone lines. A DSLAM acts as an aggregation point for multiple connections, and it connects via an ATM network back to a service provider’s router. The service provider authenticates user credentials, obtained via PPPoE, using an authentication server. Also, the service provider has a DHCP server to distribute IP address information to end-user devices (for example, a PC or a wireless router connected to a DSL modem). The term asymmetric in asymmetric DSL implies that the upstream and downstream speeds can be different. Typically, downstream speeds are greater than upstream speeds in an ADSL connection. The theoretical maximum downstream speed for an ADSL connection is 8 Mbps, and the maximum upstream speed is 1.544 Mbps (the speed of a T1 circuit).
Whereas ADSL has asymmetric (unequal) upstream and downstream speeds, by definition, SDSL has symmetric (equal) upstream and downstream speeds. Another distinction between ADSL and SDSL is that SDSL does not allow simultaneous voice and data on the same phone line. Therefore, SDSL is less popular in residential installations because an additional phone line is required for data. Also, SDSL connections are usually limited to a maximum distance of 12,000 feet between a DSL modem and its DSLAM.
VDSL boasts a much higher bandwidth capacity than ADSL or SDSL, with a common downstream limit of 52 Mbps and a limit of 12 Mbps for upstream traffic.
VDSL’s distance limitation is 4,000 feet of telephone cable between a cable modem and a DSLAM. This constraint might seem too stringent for many potential VDSL subscribers, based on their proximity to their closest telephone central office (CO). However, service providers and telephone companies offering VDSL service often extend their fiber-optic network into their surrounding communities. This enables VDSL gateways to be located in multiple communities. The 4,000-foot limitation then becomes a distance limitation between a DSL modem and the nearest VDSL gateway, thus increasing the number of potential VDSL subscribers.
ISDN
Integrated Services Digital Network (ISDN) is a digital telephony technology that supports multiple 64-kbps channels (known as bearer channels [B channels]) on a single connection. ISDN was popular back in the 1980s and was used to connect private branch exchanges (PBX), which are telephone switches owned by and operated by a company, to a central office. ISDN has the capability to carry voice, video, or data over its B channels. ISDN also offers a robust set of signaling protocols: Q.921 for Layer 2 signaling and Q.931 for Layer 3 signaling. These signaling protocols run on a separate channel in an ISDN circuit (known as the delta channel, data channel, or D channel).
A PRI circuit is an ISDN circuit built on a T1 or E1 circuit. Recall that a T1 circuit has 24 channels. Therefore, if a PRI circuit is built on a T1 circuit, the ISDN PRI circuit has 23 B channels and one 64-kbps D channel. The 24th channel in the T1 circuit is used as the ISDN D channel (the channel used to carry the Q.921 and Q.931 signaling protocols, which are used to set up, maintain, and tear down connections).
Also, recall that an E1 circuit has 32 channels, with the first channel being reserved for framing and synchronization and the seventeenth channel used for signaling. Therefore, an ISDN PRI circuit built on an E1 circuit has 30 B channels and one D channel, which is the seventeenth channel.
Some ISDN circuits are four-wire circuits and some are two-wire. Also, some devices in an ISDN network might not natively be ISDN devices, or they might need to connect to a four-wire ISDN circuit or a two-wire ISDN circuit.
ATM
Asynchronous Transfer Mode (ATM) is a Layer 2 WAN technology that operates using the concept of PVCs and SVCs. However, ATM uses fixed-length cells as its protocol data unit (PDU), as opposed to the variable frames used by Frame Relay. An ATM cell contains a 48-byte payload and a 5-byte header.
An ATM cell’s 48-byte payload size resulted from a compromise between the different countries as an international standard for ATM was being developed. Some countries, such as France and Japan, wanted a 32-byte payload size because smaller payload sizes worked well for voice transmission. However, other countries, including the United States, requested a 64-byte payload size because they felt such a size would better support the transmission of both voice and data. In the end, the compromise was to use the average of 32 bytes and 64 bytes (that is, 48 bytes).
Although ATM uses VCs to send voice, data, and video, those VCs are not identified with DLCIs. Instead, ATM uses a pair of numbers to identify a VC. One of the numbers represents the identifier of an ATM virtual path. A single virtual path can contain multiple virtual circuits.
A virtual path is labeled with a virtual path identifier (VPI), and a virtual circuit is labeled with a virtual circuit identifier (VCI). Therefore, an ATM VC can be identified with a VPI/VCI pair of numbers. For example, 100/110 can be used to represent a VC with a VPI of 100 and a VCI of 110.
Interconnections between ATM switches and ATM endpoints are called user-network interfaces (UNI), and interconnections between ATM switches are called network-node interfaces (NNI).
PPP/Multilink PPP
A common Layer 2 protocol used on dedicated leased lines is Point-to-Point Protocol (PPP). PPP has the capability to simultaneously transmit multiple Layer 3 protocols (for example, IP and IPX) through the use of control protocols (CP). IP, as an example, uses the IP control protocol (IPCP).
Each Layer 3 CP runs an instance of PPP’s Link Control Protocol (LCP). Four primary features offered by LCP include the following:
- Multilink interface—PPP’s multilink interface feature enables multiple physical connections to be bonded together into a logical interface. This logical interface allows load balancing across multiple physical interfaces. This is referred to as Multilink PPP.
- Looped link detection—A Layer 2 loop (of PPP links) can be detected and prevented.
- Error detection—Frames containing errors can be detected and discarded by PPP.
Authentication—A device at one end of a PPP link can authenticate the device at the other end of the link. Three approaches to perform PPP authentication are as follows:
- Password Authentication Protocol (PAP)—PAP performs one-way authentication (a client authenticates with a server). A significant drawback to PPP, other than its unidirectional authentication, is the security vulnerability of its clear text transmission of credentials, which could permit an eavesdropper to learn the authentication credentials being used.
- Challenge-Handshake Authentication Protocol (CHAP)—Like PAP, CHAP performs a one-way authentication. However, authentication is performed through a three-way handshake (challenge, response, and acceptance messages) between a server and a client. The three-way handshake enables a client to be authenticated without sending credential information across a network.
- Microsoft Challenge-Handshake Authentication Protocol (MS-CHAP)—MS-CHAP is a Microsoft-enhanced version of CHAP that offers a collection of additional features, including two-way authentication.
MPLS
Multiprotocol Label Switching (MPLS) is growing in popularity as a WAN technology used by service providers. This is due in part to MPLS’s capability to support multiple protocols on the same network. MPLS also has the capability to perform traffic engineering (which allows traffic to be dynamically routed within an MPLS cloud based on current load conditions of specific links and availability of alternative paths).
MPLS inserts a 32-bit header between Layer 2 and Layer 3 headers. Because this header is shimmed between the Layer 2 and Layer 3 headers, it is sometimes referred to as a shim header. Also, because the MPLS header resides between the Layer 2 and Layer 3 headers, MPLS is considered to be a Layer 2 1/2 technology.
The 32-bit header contains a 20-bit label. This label is used to make forwarding decisions within an MPLS cloud. Therefore, the process of routing MPLS frames through an MPLS cloud is commonly referred to as label switching.
An MPLS frame does not maintain the same label throughout the MPLS cloud. Instead, an LSR receives a frame, examines the label on the frame, makes a forwarding decision based on the label, places a new label on the frame, and forwards the frame to the next LSR. This process of label switching is more efficient than routing based on Layer 3 IP addresses. The customer using a provider’s network and the MPLS transport across that network is not normally aware of the details of the exact MPLS forwarding that is done by the service provider.
GSM/CDMA
Some cellular-phone technologies (for example, Long-Term Evolution [LTE], which supports a 100-Mbps data rate to mobile devices and a 1-Gbps data rate for stationary devices) can be used to connect a mobile device (such as a smartphone) to the Internet. Other technologies for cellular phones include the older 2G EDGE, which provides slow data rates. EDGE stands for Enhanced Data Rates for GSM Evolution. 2G EDGE was improved upon with 3G, as well as the newer 4G, LTE, and Evolved High-Speed Packet Access (HSPA+). The term tethering is commonly used with today’s smartphones. Tethering enables a smartphone’s data connection to be used by another device, such as a laptop. Also, mobile hotspots are growing in popularity, because these devices connect to a cellphone company’s data network. It makes that data network available to nearby devices (typically, a maximum of five devices) via wireless networking technologies. This, for example, enables multiple passengers in a car to share a mobile hotspot and have Internet connectivity from their laptops when riding down the road. Code Division Multiple Access (CDMA) and Global System for Mobiles (GSM) are the two major radio systems used in cellphones.
Dialup
Dialup Internet access uses the facilities of the public switched telephone network (PSTN) to establish a connection to an Internet service provider (ISP) by dialing a telephone number on a conventional telephone line. The user’s computer or router uses an attached modem to encode and decode link layer frames and control information into and from audio frequency signals, respectively. Software of the computer encapsulates or extracts Internet protocol packets from the data stream. Despite the proliferation of high-speed Internet access (broadband), dialup Internet access can be used where other forms are not available, such as in rural or remote areas.
WiMAX
Worldwide Interoperability for Microwave Access (WiMAX) provides wireless broadband access to fixed locations (as an alternative to technologies such as DSL) and mobile devices. Depending on the WiMAX service provider, WiMAX coverage areas could encompass entire cities or small countries.
Metro-Ethernet
Ethernet ports (using an RJ-45 connector) are very common and less expensive than specialized serial ports and associated cables. Service providers can offer an Ethernet interface to their customers for their WAN connectivity. The service provider configures the logical connections (in the provider network) required to connect the customer to sites. The technology used in the provider’s network is hidden from the customer, allowing what appears to be Ethernet connectivity to each of the customer sites. Actual throughput between sites is controlled by the provider based on the level of service purchased by the customer. Metro-Ethernet is certainly exciting when you consider the speeds possible and the use of such a familiar technology!
Leased Lines
A dedicated leased line is typically a point-to-point link interconnecting two sites. All the bandwidth on that dedicated leased line is available to those sites. This means that, unlike a packet-switched connection, the bandwidth of a dedicated leased line connection does not need to be shared among multiple service provider customers.
WAN technologies commonly used with dedicated leased lines include digital circuits, such as T1, E1, T3, and E3 circuits. These circuits use multiplexing technology to simultaneously carry multiple conversations in different 64-kbps channels. A single 64-kbps channel is called a Digital Signal 0 (DS0).
When one of these circuits comes into your location, it terminates on a device called a channel service unit/data service unit (CSU/DSU). Also, be aware that a customary Layer 2 protocol used on dedicated leased lines is PPP. A common connection type used to join to a CSU/DSU is an RJ-48C, which looks similar to an RJ-45(Ethernet) connector. Figure 3-1 shows a dedicated leased line.
Figure 3-1 A Dedicated Leased Line
T1—T1 circuits were originally used in telephony networks, with the intent of one voice conversation being carried in a single channel (that is, a single DS0). A T1 circuit is composed of 24 DS0s, which is called a Digital Signal 1 (DS1). The bandwidth of a T1 circuit is 1.544 Mbps.
T1 circuits are popular in North America and Japan.
E1—An E1 circuit contains 32 channels, in contrast to the 24 channels on a T1 circuit. Only 30 of those 32 channels, however, can transmit data (or voice or video). Specifically, the first of those 32 channels is reserved for framing and synchronization, and the seventeenth channel is used for signaling (that is, setting up, maintaining, and tearing down a call).
Because an E1 circuit has more DS0s than a T1, it has a higher bandwidth capacity. Specifically, an E1 has a bandwidth capacity of 2.048 Mbps.
Unlike a T1 circuit, an E1 circuit does not group frames together in an SF or ESF. Instead, an E1 circuit groups 16 frames together in a multiframe.
E1 circuits are popular outside North America and Japan.
T3—In the same T-carrier family of standards as a T1, a T3 circuit offers an increased bandwidth capacity. Although a T1 circuit combines 24 DS0s into a single physical connection to offer 1.544 Mbps of bandwidth, a T3 circuit combines 672 DS0s into a single physical connection, which is called a Digital Signal 3 (DS3). A T3 circuit has a bandwidth capacity of 44.7 Mbps.
E3—Just as a T3 circuit provides more bandwidth than a T1 circuit, an E3 circuit’s available bandwidth of 34.4 Mbps is significantly more than the 2.048 Mbps of bandwidth offered by an E1 circuit. A common misconception is that the bandwidth of an E3 is greater than the bandwidth of a T3 because an E1’s bandwidth is greater than a T1’s bandwidth. However, that is not the case—a T3 has a greater bandwidth (that is, 44.7 Mbps) than an E3 (that is, 34.4 Mbps).
CSU/DSU—Although far less popular than they once were, analog modems allowed a phone line to come into a home or business and terminate on analog modems, which provided data connections for devices such as PCs. These analog modems supported a single data conversation per modem.
However, digital circuits (for example, T1, E1, T3, or E3 circuits) usually have multiple data conversations multiplexed together on a single physical connection. Therefore CSU/DSU, a digital modem, is needed, as opposed to an analog modem. This digital modem must be able to distinguish between data arriving on various DS0s.
A CSU/DSU circuit can terminate an incoming digital circuit from a service provider and send properly formatted bits to a router. A CSU/DSU uses clocking (often provided by the service provider) to determine when one bit stops and another starts. Therefore, the circuit coming from a service provider and terminating on a CSU/DSU is a synchronous circuit (in which the synchronization is made possible by clocking).
Circuit Switched Versus Packet Switched
A circuit-switched connection is brought up on an as-needed basis. In fact, a circuit-switched connection is analogous to a phone call, for which you pick up your phone, dial a number, and a connection is established based on the number you dial. As discussed earlier in this chapter, Integrated Services Digital Network (ISDN) can operate as a circuit-switched connection, bringing up a virtual circuit on demand. This approach to on-demand bandwidth can be a cost savings for some customers who need only periodic connectivity to a remote site.
A packet-switched connection is similar to a dedicated leased line because most packet-switched networks are always on. However, unlike a dedicated leased line, packet-switched connections enable multiple customers to share a service provider’s bandwidth.