Introduction to Wide Area Networks

What is it that makes something a wide area network (WAN) instead of a local area network (LAN)? Distance is the first thing that comes to mind, but these days, wireless LANs can cover some serious turf! So, is it bandwidth? Here again, in many places really big pipes can be had for a price, so that's not it either. Well, what then? Perhaps one of the best ways to tell a WAN from a LAN is that you generally own a LAN infrastructure, but you lease a WAN infrastructure from a service provider.

I've already talked about a data link that you usually own (Ethernet), but now I'm going to talk about the data links you most often don't own but instead lease from a service provider.

The key to understanding WAN technologies is to be familiar with the different WAN terms and connection types often used by service providers to connect your networks.

Defining WAN Terms

Before ordering a WAN service type, it is a good idea to understand the following terms, which are commonly used by service providers and shown in Figure 10.1.

FIGURE 10.1 WAN terms

Customer Premises Equipment (CPE) Customer premises equipment (CPE) is equipment that's owned by the subscriber and located on the subscriber's premises.

Demarcation Point The demarcation location is the spot where the service provider's responsibility ends and the CPE begins. It's generally a RJ45 plug in a telecommunications closet owned and installed by the telecommunications company (telco). The customer is responsible for the cable (extended demarc) from this box to the CPE, which is usually a connection to a CSU/DSU. The CSU/DSU in the diagram (Figure 10.1) plugs into the demarcation location (demarc) and is the service provider's last point of responsibility for the circuit.

Local Loop The local loop connects the demarc to the closest switching office, called a central office.

Central Office (CO) This point connects the customers to the provider's switching network. A central office (CO) is sometimes referred to as a point of presence (POP).

Toll Network The toll network is a trunk line inside a WAN provider's network. This network is a collection of switches and facilities owned by the ISP.

It is important to familiarize yourself with these terms, because they are crucial to understanding WAN technologies.

WAN Connection Types

A WAN can use a number of different connection types, and this section will provide you with an introduction to the various types of WAN connections you'll find on the market today. Figure 10.2 shows the different WAN connection types that can be used to connect your LANs (DTE) over a DCE network.

FIGURE 10.2 WAN connection types

The following list explains the WAN connection types:

Leased Lines Typically, these are referred to as a point-to-point connection or dedicated connection. A leased line is a preestablished WAN communications path from the CPE through the DCE switch to the CPE of the remote site, allowing DTE networks to communicate at any time with no setup procedures before transmitting data. When cost is no object, it's really the best choice. It uses synchronous serial lines up to 45Mbps. HDLC and PPP encapsulations are frequently used on leased lines, and I'll go over them with you in detail in a bit.

Circuit Switching When you hear the term circuit switching, think phone call. The big advantage is cost—you pay only for the time you actually use it. No data can transfer before an end-to-end connection is established. Circuit switching uses dial-up modems or ISDN and is used for low-bandwidth data transfers.

Packet Switching This is a WAN switching method that allows you to share bandwidth with other companies to save money. Packet switching can be thought of as a network that's designed to look like a leased line yet the cost is more like the cost of circuit switching. There is a downside: if you need to transfer data constantly, forget about this option. Just get yourself a leased line. Packet switching will work well only if your data transfers are bursty in nature. Frame Relay is a packet-switching technology. Speeds can range from 56Kbps to T3 (45Mbps).

WAN Support

Cisco supports pretty much every WAN service available, and you can see this with the encapsulation ? command from any serial interface (your output may vary depending on the IOS version you are running).

Router#config t
Enter configuration commands, one per line. End with CNTL/Z.
Router(config)#int s0/0
Router(config-if)#encapsulation ?
  atm-dxi         ATM-DXI encapsulation
  bstun           Block Serial tunneling (BSTUN)
  frame-relay     Frame Relay networks
  hdlc            Serial HDLC synchronous
  lapb            LAPB (X.25 Level 2)
  ppp             Point-to-Point protocol
  sdlc            SDLC
  sdlc-primary    SDLC (primary)
  sdlc-secondary  SDLC (secondary)
  smds            Switched Megabit Data Service (SMDS)
  stun            Serial tunneling (STUN)
  x25             X.25

You cannot configure Ethernet or any type of LAN technologies as an encapsulation on a serial interface.

In this section, I will define the most prominent WAN protocols used today: Frame Relay, DSL, HDLC, PPP, and ATM. Usually, though, the only layer 2 WAN protocols configured on a serial interface these days (or for the exam) are HDLC, PPP, and Frame Relay.

Frame Relay A packet-switched technology that emerged in the early 1990s, Frame Relay is a Data Link and Physical layer specification that provides high performance. Frame Relay is a successor to X.25, except that much of the technology in X.25 used to compensate for physical errors (noisy lines) has been eliminated. Frame Relay can be more cost-effective than point-to-point links and can typically run at speeds of 64Kbps up to 45Mbps (T3). Frame Relay provides features for dynamic bandwidth allocation and congestion control, replacing leased lines with permanent virtual circuits (PVCs) through a packet-switching network.

DSL Digital subscriber line (DSL) is a technology used by traditional telephone companies to deliver advanced services (high-speed data and backward compatibility to analog video) over twisted-pair copper telephone wires. It typically has lower data-carrying capacity than Hybrid Fiber and Coax (HFC) networks, and data speeds can be limited in range by line lengths and quality (fiber backbone from the provider and cable to the home or business). Digital subscriber line is not a complete end-to-end solution but rather a Physical layer transmission technology such as dial-up, cable, or wireless. DSL connections are deployed in the last mile of a local telephone network—the local loop. The connection is set up between a pair of DSL modems on either end of a copper wire that is between the customer premises equipment (CPE) and the digital subscriber line access multiplexer (DSLAM). A DSLAM is the device located at the provider's central office (CO) and concentrates connections from multiple DSL subscribers.

HDLC High-Level Data-Link Control (HDLC) was derived from Synchronous Data Link Control (SDLC), which was created by IBM as a Data Link connection protocol. HDLC is a protocol at the Data Link layer, and it has very little overhead, so it is the recommended WAN encapsulation (and the default) if you have two Cisco routers on point-to-point links. HDLC wasn't intended to encapsulate multiple Network layer protocols across the same link. The HDLC header carries no identification of the type of protocol being carried inside the HDLC encapsulation. Because of this, each vendor that uses HDLC has its own way of identifying the Network layer protocol, which means that each vendor's HDLC is proprietary for its equipment.

PPP Point-to-Point Protocol (PPP) is an industry-standard protocol. Because all multiprotocol versions of HDLC are proprietary, PPP can be used to create point-to-point links between different vendors' equipment. It uses a Network Control Protocol (NCP) field in the Data Link header to identify the Network layer protocol. It allows authentication and multilink connections and can be run over asynchronous and synchronous links.

ATM Asynchronous Transfer Mode (ATM) was created for time-sensitive traffic, providing simultaneous transmission of voice, video, and data. ATM uses cells instead of packets that are a fixed 53 bytes long. It also can use isochronous clocking (external clocking) to help the data move faster.