WAN Connectivity

Anyone familiar with the concept of a WAN has no doubt seen the “cloud”. In just about any network diagram that includes WAN links or a connection to the Internet, the shape of a cloud will usually appear. In the case of a WAN, the cloud represents the switching equipment and links of the service provider network, over which data will travel between locations. The cloud is just an abstraction, meant to symbolize the fact that you don’t necessary know (or care about) the path that data takes through this equipment in reaching its destination. In other words, data leaves one site, enters the cloud, and then somehow reaches the other. The path over which our data travels within their network cloud is the responsibility of the service provider.

While functions within the service provider network aren’t usually of concern to the customer, there are still some terms and concepts relating to them that you should be familiar with. Consider the network outlined in the figure below. In it, a company with two locations has WAN link connecting to their two offices. At this point, the actual WAN technology in use isn’t important.

Figure: Network diagram showing the physical connections to and within a service provider’s network.

The physical connection between the locations shown in Figure 11-2 consists of a number of elements. At both company offices, the service provider will install a connection point (usually in the form of an RJ-45 jack) that physically connects a circuit to their nearest switching office. This jack is known as the demarcation point, and represents the point at which the service provider’s responsibility is said to end – in order words, the provider will ensure that the link functions correctly up to that point. The other end of this link ultimately connects to the service provider’s nearest switching facility, also known as the Central Office (CO). In general, a CO will be located within approximately 5 kilometers of the customer premises. These links (between a CO and customer premises) are part of what is known as the “local loop” or “last mile”. The local loop may consist of a variety of technologies, including fiber optics, traditional twisted pair wiring, and more.

Central Offices act as distribution points within a service provider network. In any given city, there will be many COs, interconnected to one another using high-capacity trunk links (usually fiber optics). While these links will typically connect COs to one another, some COs also serve as interconnection points to other service providers, such as national or international carriers. WAN connections that need to use the facilities of multiple service providers generally incur higher monthly fees than those that stay within a single provider’s network.

On the customer’s side of the demarcation point is where you will find what is referred to as Customer Premises Equipment (CPE). The term CPE is often used quite loosely, but traditionally refers to equipment that is owned and operated by the customer for the purpose of connecting to the service provider’s network. However, the term CPE can also be used to describe just about any piece of equipment that resides at the customer’s location. Furthermore, many companies choose to lease equipment (such as routers or DCE devices) from their service providers – this equipment is still considered to be CPE.
Before physically connecting to a service provider network, a company needs to determine the type of WAN service or connectivity that they require. For example, a customer may want a dial-up link between locations, or an “always-on” dedicated connection. Three main WAN connectivity techniques exist, each with relative advantages and disadvantages in terms of speed, performance, and cost. These include point-to-point, circuit switched, and packet switched links.

WAN Technologies and the OSI Model

WAN technologies are considered to exist and function at the three lower layers of the OSI model – Physical, Data Link, and Network. While not all WAN technologies have elements that function at the Network Layer, some (like X.25 and ISDN) do. The figure below provides an overview of how the WAN technologies that you’ll look at in this chapter map to the OSI model.

Figure: A high-level overview of how various WAN technologies map to the Physical, Data Link, and Network layers of the OSI model.

Introduction to WAN Technologies

A Wide Area Network (WAN) is best described as a data network that covers a relatively broad geographic distance. Unlike a LAN, which is usually localized within a relatively small area such as a single office, building, or small campus, a WAN will typically span distances of anywhere from a few, to thousands of kilometers. Besides distance, LANs and WANs are generally differentiated by network ownership. With a LAN, companies typically own and operate all of the equipment that interconnects devices – switches, routers, wiring, and so forth. This is usually not the case with a WAN, where a service provider (such as a local telecommunications carrier) generally owns the network links and switching equipment.

In order to interconnect geographically dispersed locations, companies will usually provision services from a telecommunications carrier, generally renting or leasing links on a monthly basis. The speed and cost of these links can vary greatly, depending upon bandwidth requirements, distances to be spanned, and available technologies. As a general rule, WANs are implemented at speeds much slower than their LAN counterparts, with higher bandwidth requirements significantly increasing monthly costs. This is a function of simple economics – providers have made huge investments in building their networks and are doing their best to make money. For companies, choosing the most appropriate WAN technology involves finding a cost-effective solution that meets their performance, reliability, and scalability requirements.

Wide Area Network (WAN) Technologies

By and large, the LAN marketplace is dominated by one technology – Ethernet. When it comes to interconnecting multiple locations, however, a variety of different Wide Area Network technologies are available. The decision to use one of these technologies over another is usually based on a number of factors including cost, performance, reliability, and availability. In some cases, a company’s WAN will not be based on a single technology, but many. For example, an organization may choose to connect some offices in a permanent fashion, while having other offices connect using various dial-up techniques. In this chapter, you’ll take a look at many of the most common WAN technologies that you’ll need to be familiar with for the CCNA and CCDA exams, including those based on point-to-point, circuit switched, and packet-switched connections.

The material to be covered in this chapter includes:

  • Introduction to WAN technologies
  • High Level Data Link Control Protocol (HDLC)
  • Point-to-Point Protocol (PPP)
  • Integrated Services Digital Network (ISDN)
  • Frame Relay