Two primary types of connections are used to create WANs: point-to-point and switched services.

Point-to-point services include many Physical Layer services and protocols such as dial-up networking, ISDN-PRI, FT1, and T1. These point-to-point links connect networks in a secure fashion; however, they offer limited options for connectivity. If a WAN needs any-to-any communication among nodes, the number of point-to-point links increases sharply with the number of endpoints.

Switched services provide more flexibility than point-to-point services, and are used to connect many networks in a large WAN. Two of the most common WAN switching protocols are frame relay and ATM. These public switching networks can set up permanent or temporary virtual circuits between any combination of endpoints. Many companies are also using the Internet as a WAN backbone. While the Internet is inherently insecure, corporations can use encryption to create VPNs across the public network.

The type of hardware a WAN requires depends on the type of service used to provide connectivity. Analog connections require a modem at each end of the line. A T1 or FT1 line requires a CSU/DSU to convert signals from the format used by the telephone system to that used by a LAN. In addition, a router is usually placed between a LAN and its WAN link, to keep unnecessary traffic off the expensive wide area connection.

WAN services are significantly more expensive than LANs, and the cost of a WAN link is directly related to its bandwidth and quality. A WAN designer always tries to get the highest bandwidth at the lowest cost; however, one must also consider other factors that may be more important than price. Thus, many organizations use more costly services that provide greater security, or better support for time-sensitive traffic.

[ Previous Section ] [ Unit Contents ] [ Next Unit ]