Chapter 11 Routing

Objectives Routing BasicsRouting Basics Why Routing Protocols are NecessaryWhy Routing Protocols are Necessary Distance-Vector RoutingDistance-Vector Routing Link-State RoutingLink-State Routing The Context of Different Routing ProtocolsThe Context of Different Routing Protocols

Path determination Path determination is a Layer 3 router function.Path determination is a Layer 3 router function. Routers evaluate various paths and determine the best path for data to be sent.Routers evaluate various paths and determine the best path for data to be sent. Routing information can be configured by the administrator (static routes) or collected dynamically by routing protocols.Routing information can be configured by the administrator (static routes) or collected dynamically by routing protocols.

End-to-End communications Network addresses represent the various paths between routers. By learning consistent paths between end stations, routers can reduce broadcasts and increase efficiency.

Network and Host Addressing Network addresses allow the data to be forwarded from one network to another. Host addresses are assigned locally and allow the delivery of data to individual hosts.

Router Functions Routers perform path determination and packet switching.Routers perform path determination and packet switching. –Path determination examines the network portion of the address to determine which interface to send the data out of. –Packet switching allows the packet to be accepted into one interface and be forwarded out of another interface.

Routed vs. Routing Protocols

Routed Protocols A routed protocol is any network protocol that provides enough information in its network layer address to allow a packet to be forwarded from one host to another host based on the addressing scheme. Examples of routed protocols – –IP – –IPX – –Apple Talk

Routing Protocols Routing protocols support a routed protocol by providing mechanisms for sharing routing information. Routing protocol messages move between the routers. A routing protocol allows the routers to communicate with other routers to update and maintain tables. TCP/IP examples of routing protocols are:   RIP (Routing Information Protocol)   IGRP (Interior Gateway Routing Protocol)   EIGRP (Enhanced Interior Gateway Routing Protocol)   OSPF (Open Shortest Path First)

Multi-protocol Routing

Multi-protocol routing Routers are capable of supporting multiple independent routing protocols and maintaining routing tables for several routed protocols. This capability allows a router to deliver packets from several routed protocols over the same data links.

Static vs. Dynamic Routing Dynamic route knowledge works differently. A network administrator enters configuration commands to start dynamic routing. The routing process automatically updates the router knowledge whenever new information is received from the internetwork. Changes in dynamic knowledge are exchanged between routers as part of the update process.

Static vs. Dynamic Routing Static route knowledge is administered manually by a network administrator who enters it into a router's configuration. The administrator must manually update this static route entry whenever an internetwork topology change requires an update.

Static Routing Static routing allows you to reveal only what you want about a secure network. Dynamic routing reveals everything about a network. When a network is accessible by only one route, static routing eliminates unneeded overhead from flooding the network.

Default Route Default routes are used by a router for any unknown destinations. These paths are determined statically.Default routes are used by a router for any unknown destinations. These paths are determined statically. An organizations Internet connection is normally setup as a default route.An organizations Internet connection is normally setup as a default route.

Advantage of Dynamic Routing If the route between A and D was static, no data could flow. Dynamic routing allows A to discover the path fault and reconfigure to B. When A-D path is good, A will reconfigure path to original configuration.

Dynamic Routing The success of dynamic routing depends on two basic router functions: –Maintenance of a routing table –Timely distribution of knowledge, in the form of routing updates, to other routers

Dynamic Routing Dynamic routing relies on a routing protocol to share knowledge among routers. A routing protocol defines the set of rules used by a router when it communicates with neighboring routers.

Dynamic Routing For example, a routing protocol describes:  How to send updates  What knowledge is contained in these updates  When to send this knowledge  How to locate recipients of the updates

How dynamic routes are chosen Routing protocols use algorithms to the best path for data.Routing protocols use algorithms to the best path for data. Routing tables are updated to show the best path.Routing tables are updated to show the best path. The algorithm generates a number called metric to determine the best path.The algorithm generates a number called metric to determine the best path. The lower the metric number, the better the path.The lower the metric number, the better the path.

Routing Metrics Metrics may be determined by a single characteristic or by several complex characteristics.

Routing Metrics Examples of common routing metrics are: –Bandwidth –Delay –Load –Reliability –Hop Count –Ticks –Cost

Classes of routing protocols All routing protocols are classified as one of three groups:All routing protocols are classified as one of three groups: –Distance Vector –Link-State –Hybrid

Classes of routing protocols- Distance Vector The distance vector routing approach determines the direction (vector) and distance to any link in the internetwork.

Classes of routing protocols- Link-State The link state approach recreates the exact topology of the entire internetwork (or at least the portion in which the router is situated). This approach is also called shortest path first.

Classes of routing protocols- Hybrid The balanced hybrid approach combines aspects of the link state and distance vector algorithms.

Convergence Whenever the topology of a network changes because of growth, reconfiguration, or failure, the network knowledge base must also change. The knowledge needs to reflect an accurate, consistent view of the new topology. This view is called convergence.

Convergence When all routers in an internetwork are operating with the same knowledge, the internetwork is said to have converged. Fast convergence is a desirable network feature. It reduces the period of time in which routers would continue to make incorrect/wasteful routing decisions.

Distance Vector Routing Distance-vector based routing algorithms pass periodic copies of a routing table from router to router. These regular updates between routers communicate topology changes.Distance-vector based routing algorithms pass periodic copies of a routing table from router to router. These regular updates between routers communicate topology changes.

Distance Vector Routing Each router receives a routing table from it’s directly connected neighbor. Each router then computes the distance vector (number of hops, etc.) to each location. This continues until all routers have same information. (convergence)

Distance Vector Routing Directly connected routers have a vector of 0. A-B=0, A-D=0 Router B reports to router A that it has a path to Router C. A-C=1 Router C reports to Router B it has a path to Router D. B-D=1 Router B reports to Router A it has a path to Router D. A-D=2

Distance Vector Routing

Distance Vector Routing- Topology Changes Each router will periodically sent its routing table to it’s adjacent routers. As topology changes occur, the routers affected update their own routing table and then forward changes to their directly connected neighbors.

Routing Loops Routing loops can occur if a network's slow convergence of a new configuration causes inconsistent routing entriesRouting loops can occur if a network's slow convergence of a new configuration causes inconsistent routing entries

Routing Loops Just before the failure of Network 1, all routers have consistent knowledge and correct routing tables. The network is said to have converged. Assume for the remainder of this example that Router C's preferred path to Network 1 is by way of Router B, and the distance from Router C to Network 1 is 3.

Routing Loops When Network 1 fails, Router E sends an update to Router A. Router A stops routing packets to Network 1. However, Routers B, C, and D continue to do so because they have not yet been informed of the failure. When Router A sends out its update, Routers B and D stop routing to Network 1. Router C, however, has not received an update. To Router C, Network 1 is still reachable via Router B.

Routing Loops Now Router C sends a periodic update to Router D, indicating a path to Network 1 by way of Router B. Router D changes its routing table to reflect this good, but incorrect, information, and propagates the information to Router A. Router A propagates the information to Routers B and E, and so on. Any packet destined for Network 1 will now loop from Router C to B to A to D and back to again to C.

Routing Loops Without countermeasures, this will continue until some other process stops the action. This is called count to infinity.

Solutions for routing loops- Defining a maximum One solution to routing loops is to set a maximum value for a metric. (i.e. Number of hops<16.) If the distance vector metric exceeds the maximum, it is discarded.

Solutions for routing loops- Split Horizon Split horizon specifies that it is never useful to send information about a route from which an update originally came. When router A triggers an update because of network 1 being down, no other router can tell A that it has a path to network 1.

Split Horizons

Solutions for routing loops- Hold-Down Timers When an update is triggered because a route goes down, the router puts a hold- down timer on that path. Updated routes with a better metrics are allowed, routes with higher value metrics are ignored during the hold-down time.

Hold-Down Timers

Distance Vector Summary Based on algorithmsBased on algorithms Distance vector numbers assigned to each pathDistance vector numbers assigned to each path Periodically updates entire routing tablePeriodically updates entire routing table Routers only update directly connected neighborsRouters only update directly connected neighbors Routing loops caused by slow convergence Count to Infinity example of loop Loops solved by setting maximum metric values, split horizons and hold- down timers RIP & IGRP are distance vector routing protocols

Link-State Routing A second popular routing protocol is Link-State routing.A second popular routing protocol is Link-State routing. Also known as Shortest Path First (SPF).Also known as Shortest Path First (SPF). Maintains complex database of complete topology.Maintains complex database of complete topology. Detailed information about routers, interfaces and links connecting them.Detailed information about routers, interfaces and links connecting them.

Link-State Routing Link-state routing uses: Link-state advertisements (LSAs) A topological database The SPF algorithm, and the resulting SPF tree A routing table of paths and ports to each network OSPF is an example of a Link-state routing protocol.

Link-State Routing Routers exchange LSA’s with each, beginning with neighbors. Each router receives LSA’s from all routers and forms a database and topology tree, with itself as the root. The tree consists of all paths to each router. Best paths are placed in the routing table, with each router acting independently.

Link-State Routing- Requirements for convergence Keep track of neighborsKeep track of neighbors Construct LSA packetsConstruct LSA packets Send out LSA’s to all routersSend out LSA’s to all routers Record received LSA’s into databaseRecord received LSA’s into database Complete map of internetworkComplete map of internetwork

Link-state routing- concerns Processor and memory requirementsProcessor and memory requirements –Router’s memory requirements higher than for Distance Vector –Database and tree information must be stored –Dijkstra's algorithm for SPF requires processing proportional to the number of links x the number of routers. Bandwidth requirements –Bandwidth consumed during initial link-state floods the network –After initial convergence, bandwidth requirements are minimal.

Link-state routing- problems Biggest concern is not all routers having same LSA’s. Routers connected to slower links will not receive updates as quickly as others Confusion about which LSA to use can cause network disruptions.

Comparing Link-state and Distance Vector Routing

Balanced-Hybrid Routing Takes the best of Distance Vector and link-state routing methods Chooses routes based on distance vectors Converges rapidly using change-based updates Cisco’s EIGRP and OSI’s IS-IS are hybrids

LAN-to-LAN routing

LAN-to-WAN routing

Path selection and switching of multiple protocols and media