In this lesson we'll learn how to connect sites together and how to connect to the Internet.
If you have a network with multiple sites then you probably have at least one subnet per site. In the last lesson we learned how data is routed between networks. In the examples we assumed ethernet was used to connect the networks together. In most cases when connecting sites together ethernet is not used. This is because we don't have the right of way to run ethernet cables on the telephone poles on the side of the street.
In lesson 1 we learned that one of the characteristics of a WAN is the medium is not locally owned. This is because we have no way to run a cable between the sites ourselves. We need to use a third party and lease a line. We can lease lines from phone companies, cable companies, or satellite companies. A leased line acts like a point to point connection between the sites. In reality the connection connects to a device at the central office like a phone switch, telco switch, or WAN switch. The central office is your leased line providers local office. The connection between your office and the central office is referred to as the local loop.
Each site will need a connection to their local provider's central office. Once the two sites are connected the provider will facilitate the connection between the sites using their network.
Once you have the physical connection between the devices you need to prepare the data to travel over the WAN link. We've learned a frame is used at the data link layer to encapsulate the data and transmit it over the connection. On LANs we saw how ethernet frames are used to transfer the data. On the WAN link the type of frame created depends on the protocol used. Two common protocols used in connections between two sites are PPP (Point to Point Protocol) and HDLC (High-level Data Link Control).
When an ethernet frame is received on your router the framing header and trailer are discarded revealing the IP packet. When the router decides to send the data out the WAN link it has to wrap the packet in a new frame. If HDLC is used as the data link layer protocol then an HDLC frame is created. In the header there is a 1 byte field for addressing, but it isn't used. There is no need for a destination address when sending data over a point to point connection. There is only one place to go.
HDLC is an older protocol than PPP, PPP supports some advanced features. Cisco uses a proprietary version of HDLC that adds some enhancements. If you use Cisco's HDLC it won't work with a device using the original HDLC.
If you want to connect more than two sites together you can use a frame relay network. Frame relay networks create virtual circuits between sites.
A frame relay network turns your provider into a big switch that uses DLCI (Data Link Connection Identifier) addresses to move data between sites. The virtual circuits are identified by the DLCI and represent links between sites. In the image below we see each site has a connection to the frame relay networks, but there are only VC's (Virtual Circuits) between Queensbury and each site. If someone wants to send data from Wilton to Albany it will have to travel over the VC to Queensbury then the VC to Albany.
The provider of the frame relay network uses the DLCI address in the frame header to send the data to the correct site. Your sending router looks at the IP packet and decides where to send it. It then creates a frame relay frame with the correct address in it to get the the destination.
As you can see the advantages of the layered approach to networking becomes apparent when you realize multiple data link layer protocols can be used as your data travels from the source to the destination. It's this flexibility that's given us interoperability between different networks.
When connecting to the Internet you have a few choices for connectivity. In the past you would use a modem to establish a dial up connection, but these connections were slow and prevented phone use while connected. A replacement for the dial up connection is DSL (Digital Subscriber Line). DSL uses the existing phone connection to connect to the Internet, but in a different way than dial up. It will transmit data at frequencies above 4000 hz. It turns out 0-4000 hz is all we need to communicate with voice so anything above that is fair game for data transmission. At the phone company they use a device called a DSLAM (DSL Access Multiplexer) that splits the signal into the voice network and data.
DSL can operate at different speeds depending on how much you pay and how far you are from the CO. The farther away you are the slower the service.
Another method for connecting to the Internet is using a cable connection. Cable TV is a broadband technology that sends multiple data signals into your house at the same time. Each data signal represents a channel on the TV. Cable modems will use a channel to transmit the data. The latest DOCSIS (Data Over Cable Services Interface Specifications) standard allows multiple channels to be bound together creating greater throughput.
Each of these connections are referred to as asymmetric connections, meaning they're capable of receiving more than they send.