Chapter 13

Wide Area Networks


Home | Chapters | Homework | Grades | Final Project | Calendar | Forums | Syllabus

Forum Policy | Assignment Policy | Games | Tests | E-Mail Mr Hull | FTP


Main | Chapter 1 | Chapter 2 | Chapter 3 | Chapter 4 | Chapter 5 | Chapter 6 | Chapter 7

Chapter 8 | Chapter 9 | Chapter 10 | Chapter 11 | Chapter 12 | Chapter 13 | Chapter 14 | Chapter 15


Summary

So far we have mostly talked about LAN's now we are going to discuss what happens when you have LAN's at multiple locations that you want to connect together.  This is where WAN technology comes in to play.  This chapter is in introduction.  We will discuss some of the terms and ideas behind WAN's.


Terminology

Below are a few of the terms that are used in a WAN.

  • Link: The direct connection of to nodes on a network.

  • End-to-End Routing: When a node sends a message to another node and the two nodes are not directly connected, the message must pass through multiple nodes to get to it's destination.  End-to-end routing determines how this is done.

  • Path: The combination on Links used to do End-to-End Routing.

  • Hop: The number of nodes a message has to go through to get to it's destination

  • Store-and-Forward: As a message is sent from node to node the preceding node will keep a copy of the data sent until it receives and acknowledgment from the receiving node.

  • Session: A communications between two users of a network. (Session layer of OSI model)

  • Packet Switching: The technology of transmitting a message in one or more fixed-length data packets.


WAN Network Topologies

We have already talked about topologies in Chapter 4.  A WAN can be set up in the Ring, Bus or Star configuration just like a LAN but there are a few more that are supported.

Hierarchical Network is a topology that matches the geographical layout of your organization.  In Chapter 7 we learned about a Network Directory.  This topology might mimic your Network Directory.

Interconnected (Plex) Network contains multiple connections to different nodes.  This will help increase speed and an alternate route may be able to be used if there is a failure.

Combination Network uses multiple topologies.  You might have an Interconnected with a ring topology network branching off.  This topology is never one set topology but a network that uses two or more of the topologies that we have talked about in class.  This might be the result of two company's joining together with different topologies.  They could join the to networks together and form a Combination Network.


WAN Data Link Control Protocols and Bit Synchronous Protocols

In Chapter 4 we talked about the Data Link layer of the OSI model.  We learned about the difference between asynchronous and synchronous protocols.  Click here to review that section.  WAN's typically use synchronous protocols to communicate.  The first bit synchronous data link protocol was introduced by IBM in 1972.  Since then many have different protocols have been introduced.

  • SDLC - Synchronous data link protocol (from IBM)

  • ADCCP - Advanced data communications control procedure (pronounced addcap)

  • HDLC - High-level data link control (ISO)

  • LAPB - Link Access procedure balanced.  (X.25)

All of these bit synchronous protocols operate similarly.  We will look closely at an HDLC frame.

High-level Data Link Control uses a flag field at both the beginning and end of the frame to maintain synchronization.  This would be the equivalent of the STX field that we learned about in Chapter 4.  The HDLC Frame can be broken up into 5 unique sections. Flag , Address, Control, Information (Data), CRC (Frame Check Sequence)

  • The Flag field is an 8 bit number with the pattern 01111110.  This field is found at the beginning and end of a frame.  If a lot of frames are in sequence the flag field doesn't have to repeat.  The end flag for one frame can be the start flag for the next frame.

An HDLC Frame

  • The Address block contains the address, this can be 1 or more octets.  Since this is a data link layer protocol the address you will find in the section are MAC addresses.  If this is a point to point connection (balanced) there will be no address.

  • The Control field determines the frame type.  There are three possibilities.

    • Unnumbered - An unnumbered frame type is used to initialize or disconnect a link.  This frame type can have data but it doesn't have too.  If it does contain data it will be found in the information field.  The first two bits in the control field in this type of frame are 1's.

    • Informational - There is data in this frame type which is found in the information field.  A frame of this type has additional information in it's control field which is designed to improve speeds.  In an 8 bit control field the first bit is 0.  The next three bits are the Ns field.  The fifth bit is the P/F bit, and the last three bits are the Nr field bits.  When the frame is sent the Ns field increases by one, when the frame is received the Nr field is increased by one.  This allows the receiving node to only have to reply once every 7 frames saying data was received properly.  The P/F bit is the Poll/Final bit.  The Poll bit is used when a primary station wants to poll a secondary station.  The Final bit is used by the secondary station to indicate a response or the end of a transmission. 

    • Supervisory - In this frame type there is no data being transmitted.  This frame type is used to acknowledge frames, request retransmission, or to ask for suspension of transmission.  The first two bits in the control field in this type of frame are 1 and 0.

  • Information or data block contains the data being transmitted.  This field is optional.  Since the flag field is in the pattern 01111110 you can't have any data using that pattern.  To prevent this from happening the sending node will insert a 0 after 5 consecutive 1's.  This way there will never be a 01111110 pattern found in the information field.  The receiving node will examines the next bit after 5 consecutive 1's, if it is a 0 it will discard the 0.  If it is a 1 it knows it is the flag field. 

  • The CRC or Frame Check Sequence is for error detection.


WAN Network Layer Functions and Message Routing on a WAN

The Network Layer is responsible for end to end routing.  The Data Link Layer protocols we have looked at are responsible for transmitting data down one link.  We will now look at routing and how the network layer determines the best path for end to end routing on a WAN.

Two LANs connected forming a WAN usually don't have a direct cable connecting them together.  You would lease a line from a local carrier to supply the connection.  That connection will go through multiple hops before it gets to the destination LAN.  The way data is routed between those hops is what we will focus on now.  We will be talking about hardware devices called routers which are layer 3 devices.  To understand the different types of routing protocols you should understand that there routers have tables inside that tell where to route packets depending on the destination.  For example if you have a router with 3 ports and a packet comes in on port 1 destined for LAN A.  The router will look at the destination address and compare it to its table which says Port 2 goes to LAN A.  It will then send the packets out Port 2.  Now that we have a general understanding of routers lets look at different routing protocols.

  • Centralized Routing - With this routing protocol there is one central router that all other routers communicate with updating their status.  If one router fails or has a busy line the central router can update other routers tables with new routes to take.  If the central router fails you can have a backup takes it's place. If there is no backup, routing will still occur but no tables will be updated if one fails.

  • Distributed Routing - Each router is responsible for determining the best path for the packets.  This can be done through any of the remaining routing protocols.

  • Static Routing - Each router has a table in it that cannot be changed unless a person where to manually update them.

  • Weighted Routing - In a router that has multiple paths to one destination might set a weight to each line that leads to that destination.  If one line was a fast 1.54mbs T1 and the other a slow 64kbs frame relay line

  • Adaptive Routing - AKA Dynamic Routing, will send data down any one line that is free.  The only rule is that it will not send data down the link it came in on.  The router does not make any effort to determine if the link it is sending down is the most efficient route.

  • Broadcast Routing - This will send packets out all ports.


Packet Distributed Networks

A packet distributed network is the process of segmenting data into specific sized packets, routing those packets to their destination, and reassembling them to recreate the message.

  • Advantages:

    • The user can be charged for the amount of data sent as apposed to the connection time. 

    • A PDN can give you access to many different locations.

    • Connections are usually via a local telephone call.

    • Network maintenance and error recovery are the responsibility of the PDN.

  • Disadvantages:

    • Shared connection, you might have to compete for circuits, this can cause a slower connection.

    • If you are transmitting a lot of data the cost might be higher then a leased line.

    • Since you don't have control over the network you are not able to optimize the networks.

X.25 is a type of PDN that was created when data lines were not as reliable as they are today.  This protocol has a lot of error detection built in to allow the data to get through the poor lines.  X.25 is a network layer protocol.  Frame Relay takes advantage of today's improved lines and reduces the amount of error detection.  Consequently you will see higher speeds in Frame Relay then X.25.


Comparing WANs with LANs

In Chapter 1 we talked about the differences between WAN's and LAN's.  We will take another look at the differences now that we have looked at WAN's.  The most common reason to use WAN's is they can cover a larger area then LAN's. 

  • Topology, Protocols, and Routing: There are different ways nodes on a network are physically arranged on WAN's vs. WAN's.  In LAN's you you would find star, bus and ring topologies, in a WAN you would find hierarchical, and interconnected.  In both you will find combination networks.  You will also fin stars, buses and rings in a WAN but they not as often.  The protocols to send data on the wire on a LAN are CSMA/CD and token passing.  WAN's use bit synchronous data link protocols such as HDLC.  Routing in a WAN is more complex then on a LAN, routing on a WAN can be any done using any of the routing protocols discussed in this chapter.

  • Media: The media on a WAN is rented and may switch between multiple media types.  A LAN usually has one type of media.  You might find fiber for the long hauls and copper to the desktops.

  • Ownership: The media in a LAN is usually locally owned.  The media in a WAN is owned by a central company or multiple companies and you lease it.

  • Transmission Speed: LAN's are connected at higher speeds. 


More Information

 

Click Here to download the slides for this chapter

(NOTE: You must have PowerPoint or PowerPoint Viewer if you don't have  either Click Here to download PowerPoint Viewer.)


Home | Chapters | Homework | Grades | Final Project | Calendar | Forums | Syllabus

Forum Policy | Assignment Policy | Games | Tests | E-Mail Mr Hull | FTP