+ All Categories
Home > Documents > Dynamic Routing Protocols

Dynamic Routing Protocols

Date post: 18-Jan-2016
Category:
Upload: liza
View: 24 times
Download: 1 times
Share this document with a friend
Description:
Dynamic Routing Protocols. Static Routing. Easy to configure Easy to troubleshoot Doesn’t scale well Limited redundancy options. Routing Protocols. - PowerPoint PPT Presentation
Popular Tags:
37
RCC College Dynamic Routing Protocols
Transcript
Page 1: Dynamic Routing Protocols

RCC College

Dynamic Routing Protocols

Page 2: Dynamic Routing Protocols

RCC College

Static Routing

Easy to configureEasy to configure Easy to troubleshootEasy to troubleshoot Doesn’t scale wellDoesn’t scale well Limited redundancy optionsLimited redundancy options

Page 3: Dynamic Routing Protocols

RCC College

Routing Protocols

A routing protocol is the language a router A routing protocol is the language a router speaks with other routers in order to share speaks with other routers in order to share information about the reachability and information about the reachability and status of networks.status of networks.

Routing protocols also determine the next Routing protocols also determine the next best path if the current best path to a best path if the current best path to a destination becomes unusable.destination becomes unusable.

Page 4: Dynamic Routing Protocols

RCC College

Routing Algorithm specifies:

A procedure for passing reachability information A procedure for passing reachability information about networks to other routersabout networks to other routers

A procedure for receiving reachability information A procedure for receiving reachability information from other routersfrom other routers

A procedure for determining optimal routes based A procedure for determining optimal routes based on the reachability information it has and for on the reachability information it has and for recording this information in a routing tablerecording this information in a routing table

A procedure for reacting to, compensation for, and A procedure for reacting to, compensation for, and advertising topology changes in an internetworkadvertising topology changes in an internetwork

Page 5: Dynamic Routing Protocols

RCC College

Example: 3-router internetwork

Page 6: Dynamic Routing Protocols

RCC College

Metrics

When there are multiple routes to the same When there are multiple routes to the same destination, a router must have a destination, a router must have a mechanism for calculating the best pathmechanism for calculating the best path

A metric is a variable assigned to routes to A metric is a variable assigned to routes to rank them from best to worstrank them from best to worst

Different protocols use different, and Different protocols use different, and sometimes multiple, metricssometimes multiple, metrics

Page 7: Dynamic Routing Protocols

RCC College

Hop Count

A hop count metrics simply counts router A hop count metrics simply counts router hops.hops.

Is the A-B link really the best path? Is the A-B link really the best path? What if the A-B link is a DS-0 and the A-C What if the A-B link is a DS-0 and the A-C and C-B links are T-1s?and C-B links are T-1s?

Page 8: Dynamic Routing Protocols

RCC College

Bandwidth

A bandwidth metric would choose a higher-A bandwidth metric would choose a higher-bandwidth path over a lower-bandwidth bandwidth path over a lower-bandwidth link.link.

What if the higher-bandwidth link has a What if the higher-bandwidth link has a higher delay and we transfer voice traffic?higher delay and we transfer voice traffic?

What if the T1 links are highly loaded and a What if the T1 links are highly loaded and a 56Kb link is lightly loaded? 56Kb link is lightly loaded?

Page 9: Dynamic Routing Protocols

RCC College

Load

This metric reflects the amount of traffic This metric reflects the amount of traffic utilizing the links along the path. The best utilizing the links along the path. The best path is the one with the lowest load.path is the one with the lowest load.

Unlike hop count and bandwidth, the load Unlike hop count and bandwidth, the load on a router changes, and therefore the on a router changes, and therefore the metric will change. If the metric changes metric will change. If the metric changes too frequently, too frequently, route flapping - route flapping - the frequent the frequent change of preferred routes - may occur.change of preferred routes - may occur.

Page 10: Dynamic Routing Protocols

RCC College

Delay

Delay Delay is a measure of the time it takes a is a measure of the time it takes a packet to travel a route.packet to travel a route.

A routing protocol using delay as a metric A routing protocol using delay as a metric would choose the path with the least delay would choose the path with the least delay as the best path.as the best path.

Delay changes with traffic load.Delay changes with traffic load.

Page 11: Dynamic Routing Protocols

RCC College

Reliability

ReliabilityReliability measures the likelihood that the measures the likelihood that the link will fail in some way and can be either link will fail in some way and can be either variable or fixed.variable or fixed.

Examples: the number of times a link has Examples: the number of times a link has failed or the number of errors it has failed or the number of errors it has received within a certain period of time.received within a certain period of time.

Page 12: Dynamic Routing Protocols

RCC College

Cost

CostCost is configured by the network is configured by the network administrator to reflect more- or less- administrator to reflect more- or less- preferred routes.preferred routes.

The term The term costcost is often used as a generic is often used as a generic term when speaking of route choices. term when speaking of route choices.

Page 13: Dynamic Routing Protocols

RCC College

Convergence

The process of bringing all route tables to The process of bringing all route tables to the state of consistency is called the state of consistency is called convergence.convergence.

The time it takes to share information The time it takes to share information across an internetwork and for all routers to across an internetwork and for all routers to calculate best paths is called the calculate best paths is called the convergence time.convergence time.

Page 14: Dynamic Routing Protocols

RCC College

Distance Vector Routing Protocols

Most routing protocols fall into one of two Most routing protocols fall into one of two classes: classes: distance vectordistance vector or or link state.link state.

Distance vector algorithms are based on the Distance vector algorithms are based on the work done of Bellman, Ford, Fulkerson and work done of Bellman, Ford, Fulkerson and occasionally referred to as occasionally referred to as Bellamn-FordBellamn-Ford or or Ford-Fulkerson.Ford-Fulkerson.

Page 15: Dynamic Routing Protocols

RCC College

Distance Vector Routing Protocols (cont’d) The name The name distance vectordistance vector is delivered from is delivered from

the fact that routes are advertised as vectors the fact that routes are advertised as vectors of distance, where distance is defined in of distance, where distance is defined in terms of metric, and direction is defined in terms of metric, and direction is defined in terms of the next-hop router.terms of the next-hop router.

Example: Destination A is at a distance of 5 Example: Destination A is at a distance of 5 hops away, in the direction of next-hop hops away, in the direction of next-hop router X. router X.

Page 16: Dynamic Routing Protocols

RCC College

Distance vector routing protocols:

Routing Information Protocol (RIP) for IPRouting Information Protocol (RIP) for IP Cisco’s Internet Gateway Routing Protocol Cisco’s Internet Gateway Routing Protocol

(IGRP)(IGRP) Novell’s IPX RIPNovell’s IPX RIP AppleTalk’s Routing Table Maintenance AppleTalk’s Routing Table Maintenance

Protocol (RTMP) Protocol (RTMP)

Page 17: Dynamic Routing Protocols

RCC College

Common Characteristics of DVPs

Periodic updatesPeriodic updates NeighborsNeighbors Broadcast UpdatesBroadcast Updates Full Routing Table UpdatesFull Routing Table Updates

Page 18: Dynamic Routing Protocols

RCC College

Periodic Updates Periodic Updates Periodic Updates mean that at the end of a mean that at the end of a

certain time period, updates will be certain time period, updates will be retransmitted.retransmitted.

These periods typically range from 10 to 90 These periods typically range from 10 to 90 seconds.seconds.

If updates are sent too infrequently, If updates are sent too infrequently, convergence time may be unacceptably high; if convergence time may be unacceptably high; if updates are sent too frequently, network traffic updates are sent too frequently, network traffic may be unacceptably high.may be unacceptably high.

Page 19: Dynamic Routing Protocols

RCC College

Neighbors

In the context of routers, In the context of routers, neighborsneighbors always always mean routers sharing a common data link.mean routers sharing a common data link.

A distance vector routing protocol sends its A distance vector routing protocol sends its updates to neighboring routers and depends updates to neighboring routers and depends on them to pass the update information on them to pass the update information along to their neighbors. along to their neighbors.

For this reason, distance vector routing is For this reason, distance vector routing is said to use hop-by-hop updates. said to use hop-by-hop updates.

Page 20: Dynamic Routing Protocols

RCC College

Broadcast Updates

When a router first becomes active on a When a router first becomes active on a network, how does it find other routers and network, how does it find other routers and how does it announce its own presence?how does it announce its own presence?

The simplest method - to send the updates The simplest method - to send the updates to the broadcast address 255.255.255.255.to the broadcast address 255.255.255.255.

Hosts and other devices uninterested in the Hosts and other devices uninterested in the routing updates will simply drop the routing updates will simply drop the packets.packets.

Page 21: Dynamic Routing Protocols

RCC College

Full Routing Table Updates

Most distance vector protocols take the very Most distance vector protocols take the very simple approach of telling their neighbors simple approach of telling their neighbors everything they know by broadcasting the everything they know by broadcasting the entire routing tableentire routing table

Neighbors receiving these updates glean the Neighbors receiving these updates glean the information they need and discard information they need and discard everything else.everything else.

Page 22: Dynamic Routing Protocols

RCC College

Routing by Rumor

Page 23: Dynamic Routing Protocols

RCC College

Route Invalidation Timers

What if a router goes down? Information What if a router goes down? Information about a path is no longer valid, but there is about a path is no longer valid, but there is no router to inform about this fact.no router to inform about this fact.

The problem is handled by setting a route The problem is handled by setting a route invalidation timer for each entry in the invalidation timer for each entry in the routing table.routing table.

Typical periods for route timeouts range Typical periods for route timeouts range from 3 to 6 update periods.from 3 to 6 update periods.

Page 24: Dynamic Routing Protocols

RCC College

Split Horizon

Each router broadcasts its entire route table Each router broadcasts its entire route table to every neighbor every update period.to every neighbor every update period.

Is this really necessary?Is this really necessary? A router pointing back to the router from A router pointing back to the router from

which packets were received is called a which packets were received is called a reverse routereverse route..

Split horizon Split horizon is a technique for preventing is a technique for preventing reverse routes between two routers.reverse routes between two routers.

Page 25: Dynamic Routing Protocols

RCC College

Split Horizon (cont’d)

Split Horizon prevents routing loops Split Horizon prevents routing loops problems.problems.

2 categories of split horizon: simple split 2 categories of split horizon: simple split horizon and split horizon with poisoned horizon and split horizon with poisoned reverse.reverse.

Page 26: Dynamic Routing Protocols

RCC College

Poisoned Reverse

When sending updates out of a particular When sending updates out of a particular interface, designate any network that was interface, designate any network that was learned from updates received on that learned from updates received on that interface as unreachable.interface as unreachable.

Split horizon with poisoned reverse is Split horizon with poisoned reverse is considered safer and stronger than simple considered safer and stronger than simple split horizon - it helps to remove corrupted split horizon - it helps to remove corrupted routing information. routing information.

Page 27: Dynamic Routing Protocols

RCC College

Counting to Infinity

Sometimes split horizon doesn’t resolve the loop problems.Sometimes split horizon doesn’t resolve the loop problems. The way to alleviate the effect of counting to infinity is to The way to alleviate the effect of counting to infinity is to

define infinity.define infinity. For example, RIP defines infinity as 16 hops. As soon as For example, RIP defines infinity as 16 hops. As soon as

hop count increases to 16, the network is considered hop count increases to 16, the network is considered unreachable.unreachable.

Setting a maximum hop count to 15 helps to solve the Setting a maximum hop count to 15 helps to solve the counting-to-infinity problem, but convergence will be very counting-to-infinity problem, but convergence will be very slow. Given an update period of 30 seconds, a network slow. Given an update period of 30 seconds, a network could take up to 7.5 minutes to reconverge.could take up to 7.5 minutes to reconverge.

Page 28: Dynamic Routing Protocols

RCC College

Triggered updates Triggered updatesTriggered updates, also known as , also known as flash flash

updatesupdates, are very simple:, are very simple: If a metric changes for better or for worse, a If a metric changes for better or for worse, a

router will immediately send out an update router will immediately send out an update without waiting for its update timer to expire.without waiting for its update timer to expire.

Reconvergence will occur quicker.Reconvergence will occur quicker. Regular updates may still occur along with Regular updates may still occur along with

triggered updates.triggered updates.

Page 29: Dynamic Routing Protocols

RCC College

Holddown Timers Holddown timers Holddown timers introduce a certain amount introduce a certain amount

of skepticism to reduce the acceptance of bad of skepticism to reduce the acceptance of bad routing information.routing information.

If the distance to a destination increases (for If the distance to a destination increases (for example, the hop count increases from 2 to example, the hop count increases from 2 to 4), the router sets a holddown timer for that 4), the router sets a holddown timer for that route.route.

Until the timer expires, the router will not Until the timer expires, the router will not accept any new updates for the route.accept any new updates for the route.

Page 30: Dynamic Routing Protocols

RCC College

Asynchronous Updates As a few routers become synchronized, As a few routers become synchronized,

collisions will begin to occur, further collisions will begin to occur, further contributing to system delays, and eventually contributing to system delays, and eventually all routers sharing the broadcast network all routers sharing the broadcast network may become synchronized.may become synchronized.

Asynchronous updates may be maintained by Asynchronous updates may be maintained by a small random time, or a small random time, or timing jittertiming jitter, which , which is added to each update period as an offset.is added to each update period as an offset.

Page 31: Dynamic Routing Protocols

RCC College

Link State Routing Protocols Unlike the routing-by-rumor approach of Unlike the routing-by-rumor approach of

distance-vector, link state routers have firsthand distance-vector, link state routers have firsthand information from all their peer routers.information from all their peer routers.

Each router originates information about itself, Each router originates information about itself, its directly connected links, and the state of its directly connected links, and the state of those links.those links.

This information is passed around from router to This information is passed around from router to router, each router makes a copy of it, but never router, each router makes a copy of it, but never changes it.changes it.

Page 32: Dynamic Routing Protocols

RCC College

Link State Routing Protocols

Every router has identical information about Every router has identical information about the internetwork, and each router will the internetwork, and each router will independently calculate its own best path.independently calculate its own best path.

Link state protocols, sometimes called Link state protocols, sometimes called shortest path first shortest path first or or distributed database distributed database protocols, are built around a well-known protocols, are built around a well-known algorithm from graph theory, algorithm from graph theory, E.W. E.W. Dijkstra’s Dijkstra’s shortest path algorithm.shortest path algorithm.

Page 33: Dynamic Routing Protocols

RCC College

Link-state routing protocols:

Open Shortest Path First (OSPF) for IPOpen Shortest Path First (OSPF) for IP The ISO’s Intermediate System to The ISO’s Intermediate System to

Intermediate System (IS-IS) for IPIntermediate System (IS-IS) for IP Novell’s NetWare Link Service Protocol Novell’s NetWare Link Service Protocol

(NLSP)(NLSP)

Page 34: Dynamic Routing Protocols

RCC College

Basic functionality of LSRP Each router establishes a relationship - an Each router establishes a relationship - an adjacencyadjacency - with - with

each of its neighbors.each of its neighbors. Each router sends Each router sends link state advertisements (LSAs)link state advertisements (LSAs), ,

sometimes called sometimes called link state packets (LSPs), link state packets (LSPs), to each to each neighbor.neighbor.

Each router stores a copy of all the LSAs it has seen in a Each router stores a copy of all the LSAs it has seen in a database.database.

The completed The completed topology databasetopology database, also called , also called link state link state database, database, describes a graph of the internetwork.describes a graph of the internetwork.

Each router calculates the shortest path to each network Each router calculates the shortest path to each network and enters this information into the route table.and enters this information into the route table.

Page 35: Dynamic Routing Protocols

RCC College

Autonomous System (AS)

On the Internet, an autonomous system (AS) is the unit of router policy, either a single network or a group of networks that is controlled by a common network administrator.

An autonomous system is assigned a globally unique number, called an Autonomous System Number (ASN).

Page 36: Dynamic Routing Protocols

RCC College

Interior and Exterior Gateway Protocol

Networks within an autonomous system communicate routing information to each other using an Interior Gateway Protocol.

An autonomous system shares routing information with other autonomous systems using an Exterior Gateway Protocol.

Page 37: Dynamic Routing Protocols

RCC College

Want to know more?

Jeff Doyle ‘Routing TCP/IP, volume 1’Jeff Doyle ‘Routing TCP/IP, volume 1’ Perlman, R. ‘Interconnections: Bridges and Perlman, R. ‘Interconnections: Bridges and

Routers.’Routers.’


Recommended