[Cisco Networking Academy] Del Mar College
ITNW 1392-Beginning Router Configuration
 InstructorMichael P. Harris
Sem2Les5

Semester 2, Lesson 5 Notes:

The Network Layer (layer 3) and Routing

The network layer (layer 3) interfaces to networks and provides best effort end-to-end packet delivery services to its user, the transport layer (layer 4). The network layer sends packets from the source network to the destination based on IP routing tables.

After the router determines which path to use, it can proceed with "switching the packet": taking the packet it accepted on one interface and forwarding it to another interface or port that reflects the best path to the packet's destination.

The switching function allows the router to accept a packet on one interface and forward it on a second interface. The path determination function enables the router to select the most appropriate interface for forwarding a packet. The node portion of the address refers to a specific port on the router that leads to an adjacent router in that direction.

The success of dynamic routing depends on two basic router functions:

  1. Maintenance of a routing table.
  2. Timely distribution of knowledge in the form of routing updates to other routers. Dynamic routing relies on a routing protocol to share knowledge.

1. Distinguish between Routed and Routing Protocols.
"Routed protocols" are used between routers to direct user traffic. Any network protocol that provides enough information in its network layer address to be forwarded from host-to-host based on the addressing scheme is a routed protocol. The most relevant example for us is IP. "Routing protocols" are used only between routers to maintain routing tables. Examples are RIP and IGRP. Routing protocols support routed protocols by providing a mechanism for sharing routing information messages amongst routers.
2. How does "multiprotocol routing" enable the Internet to exist?
There are many different host systems all over the world, with many different 'routed protocols'. Multiprotocol routing -- by which a router can pass traffic from Novell, Apple, DEC, IP and many other networks -- allows these diverse hosts to communicate. Routers are capable of supporting multiple independent routing protocols and maintaining routing tables for several routed protocols concurrently. This capability allows a router to deliver packets from several routed protocols over the same data links.

"Algorithms" are rules for deciding the best path through a network. "Metrics" are numbers which measure the desirability of a particular path through a network.

3. List and briefly describe at least seven examples of routing "metrics".

The metrics most commonly used by routers include:

  1. Bandwidth -- the data capacity of a link
  2. Delay -- length of time required to move a packet from a source to destination
  3. Load -- amount of activity on a network resource such as a router or a link
  4. Reliability -- usually refers to the error rate of each network link
  5. Hop count -- Number of passages of a packet through the output port of one router
  6. Ticks -- Delay on a data link using IBM PC clock ticks
  7. Cost -- arbitrary value, usually based on bandwidth or other measurements, that is assigned by the network administrator
4. Briefly describe 'distance vector' routing.
"Distance vector" (also known as Bellman-Ford algorithms) - based routing algorithms pass periodic copies of a routing table (emphasizing hop counts) fron router to routers; each router receives a routing table from it's direct neighbors. Network discovery for distance vector involves exchange of routing tables; problems can include 'Slow time to convergence', 'routing loops', and 'counting to infinity'. Distance vector algorithms do not allow a router to know the exact topology of an internetwork.

"Convergance" - The speed and ability of a group of internetworking devices running a specific routing protocol to agree on the topology of an internetwork after a change in that topology.

  • "hold-down timers" - is a solution to counting to infinity and routing loops which has routers ignore network update for some period. State into which a router is placed so that routers will niether advertise the route nor accept advertisements about the route for a specific length of time (the hold-down period). Hold-down is used to flush bad information about a route from all routers in the network. A route is typcally placed in hold-down when a link in that route fails.
  • "split horizons" - Routing technique in which information about routes is prevented from exiting the router interface through which that information was received. Split-horizon updates are useful in preventing routing loops.
  • "defining a maximum" - a solution to routing loops in which once the metric reaches some number - for example hop count reaches 15 - the network is considered unreachable.
5. Briefly decribe 'link-state' routing.
"Link-state" (also called shortest path first) routing uses 'link-state advertisements' (LSA), put inot topological database, the SPF (shortest path first) algorithm, the resulting SPF tree, and finally a routing table of paths and ports to each network. Instead of simply passing routing tables, emphasizing hop counts to nearest neighbors (as in distance vector routing), link-state routing shares a complex database of topology information amongst all routers in the network. For link-state, routers calculate the shortest path to others routers; problems can include 'Processing and memory requirements', 'Too much bandwidth consumed', 'unsynchronized/inconsistant updates', and 'inconsistant path decisions'. Engineers have implimented this "link-state concept" in "Open Shortest Path First" (OSPF) routing.

"LSA" - Link-state advertisement. Broadcast packet used by link-state protocols that contain information about neighbors and path costs. LSAs are used by the receiving routers to maintain their routing tables. Some times called a link-state packet (LSP). Routers with different sets a LSAs will calculate routes based on different topological data. Then routes become unreachable.

"Balanced hybrid" protocols combine aspects of the link-state and distance vector algorithms. The balanced hybrid routing protocol uses distance vectors with more accurate metrics to determine the best paths to destination networks. However, it differs from most distance vector protocols by using topology changes to trigger routing database updates. The balanced hybrid routing type converges more rapidly, like the link-state protocols. However, it differs from these protocols by using fewer resources such as bandwidth, memory, and processor overhead. Examples of balanced hybrid protocols are OSI's Intermediate System (IS-IS) routing and Cisco's Enhanced Interior Gateway Routing Protocol (Enhanced IGRP).

"Static route" - Route that is explicitly configured and entered into the routing table. Static routes take precedence over routes chosen by dymanic routing protocols.

"SPF" - Shortes path first algorithm. Routing algorithm that iterates on length of path to determine a shortest-path spanning tree. Commonly used in link-state routing algorithms. Sometimes called Dijkstra's algorithm. See also link-state algorithm.

"Static routing" has several useful applications.

  1. Dynamic routing tends to reveal everything known about an internetwork. For security reasons, it might be appropriate to hide parts of an internetwork. Static routing allows anm internetwork administrator to specify what is advertised about restricted partitions.
  2. When a network is accessable by only one path, a static route to the network can be sufficient. This type of partition is called a "Stub network". Configuring static routing to a stub network avoids the overhead of dynamic routing.

"Default route" - Routing table entry that is used to direct frames for which a next hop is not explicitly listed in the routing table.

Compare "distance-vector routing" to "link-state routing":

  • Distance-vector routing gets all topological data from the routing table information of its neighbors. Link-state routing obtains a wide view of the entire internetwork topology by accumulating all necessary LSAs.
  • Distance vector routing determines the best path by adding to the metric value it receives as tables move from router to router. For link-state routing, each router works sperately to calculate its own shortest path to destinations.
  • With most distance vector routing protocols, updates for topology changes come in periodic table updates. These tables pass from router to router, usually resulting in slower convergence.
  • With link-state routing protocols, updates are usually tirggered by topology changes. Relatively small LSAs passed to all other routers usually result in faster time to convergence on any internetwork topology change.
[Previous]Lesson 4         [Index]Index         [Next]Next
[Top of Page]     This page is maintained by:   Michael P. Harris
[Viking Home Page] viking.delmar.edu
Last updated April 2		 [E-mail] mpharris@surf.delmar.edu
Copyright © 1999