A routing protocol can be analyzed as a process. A protocol is formalized through the
Request for Comments (RFC) process. This process involves open written commentary on
a proposed technology in an effort to bring about standardization of that technology. An
example of a routing protocol standard would be OSPF as defined in RFC 2328.
The basic reason you are here is to understand how, through the use of OSPF, you can get packets from one host to another over a network. This process can be summarized, from a router's perspective, as consisting of two steps, forwarding and routing:
In summary, a dynamic routing protocol is a set of standardized rules that allow routers to determine routes. The routing protocol builds routing tables that tell the router the optimal path to a destination. Routing protocols compare numeric values known as metrics to determine the optimal route (sometimes referred to as path). Metrics are numeric values that represent route path characteristics. Metrics can be thought of as costs; therefore, the numeric value is explained as the cost of transiting a link. This information is then stored in the routing table to be used by the router when determining the best route to a destination network.
Dynamic routing is the most commonly used method of routing. Associated with this type of routing are several terms and characteristics that define how it operates, such as convergence and accuracy.
In dynamic routing, the routing table must consistently reflect accurate and up-to-date information concerning the network topology. The amount of time that it takes for changes to be reflected in every network router's routing table is known as the convergence time. Convergence, in this context, is the act of approaching the state where all routing tables have congruent information and are in steady state.
Having a routing protocol with a fast convergence time is desirable because disruption of routing can occur during the time that a router spends calculating the new optimal path.
The basic reason you are here is to understand how, through the use of OSPF, you can get packets from one host to another over a network. This process can be summarized, from a router's perspective, as consisting of two steps, forwarding and routing:
- Forwarding-Refers to the process of a router receiving a packet on an interface and then knowing which interface to retransmit that packet out of so that the packet can continue to its destination. However, forwarding is entirely dependent on the router knowing where to send the packet; if it does not know this, it just discards the packet.
- Routing-For the router to know in which direction to forward packets, it requires a route or road map that illustrates the path from source to destination. This route can be generated by either a static or dynamic configuration; at this point, you need to be concerned with only the dynamic aspect of routing. OSPF is a dynamic routing protocol, and when a router is running OSPF, it dynamically develops routes to all destinations within a network. You can view these routes when you view a router's routing table.
- Routed protocols-Protocols that forward data via routers. A router must be able to interpret the logical network as specified by the routed protocol for the router to operate properly. The most commonly known routed protocol is Internet protocol (IP); other examples include AppleTalk and DECnet. Routed protocols rely on the routing protocols for transport over a LAN or WAN.
- Dynamic routing protocols-Protocols that accomplish dynamic routing with a routing algorithm. A dynamic routing protocol supports a routed protocol and maintains routing tables. A dynamic routing protocol dynamically exchanges information about paths or topology of the network by distributing routing information throughout a network. Examples of dynamic routing protocols include OSPF, interior gateway routing protocol (IGRP), and routing information protocol (RIP).
In summary, a dynamic routing protocol is a set of standardized rules that allow routers to determine routes. The routing protocol builds routing tables that tell the router the optimal path to a destination. Routing protocols compare numeric values known as metrics to determine the optimal route (sometimes referred to as path). Metrics are numeric values that represent route path characteristics. Metrics can be thought of as costs; therefore, the numeric value is explained as the cost of transiting a link. This information is then stored in the routing table to be used by the router when determining the best route to a destination network.
Basic Routing Protocol Operation
Consider an example of a router that is initially configured with two networks to which it directly connects. The router has only these two networks in its routing tables. However, other networks beyond the initial two are not entered into the routing table because they do not directly connect to the router. So how does the router recognize these other networks? This can be accomplished in the following ways:- Static routing-A manually defined and installed type of route within the router as
the only path to a given destination. This type of routing forces the destination within
routing tables. This type of routing also takes precedence over routes chosen by
dynamic routing protocols. Figure below shows an example of where to place a static
route and how a static route is used. Static routing is not an effective stand-alone
solution in a medium- to large-sized network because the work needed to make the
network run something is very large. The most common use of static routes is in stub
networks, as shown in Figure below. In conjunction with a dynamic routing protocol
such as OSPF, this synergy works well.
- Default routing-A type of route within the router that is manually defined as the
path to take when no route to the destination is known. The router to which this
information is sent is also known as the router or gateway of last resort. Figure below
shows the use of a default route, which can make routing easy.
In Figure below, all traffic that Router B receives is forwarded to Router A's S0 interface if it is not for the E0 network, 192.168.254.0/24. - Dynamic routing-Uses routing algorithms that analyze incoming routing update messages from one or more routing protocols to determine the optimal path to a destination. This type of routing has the greatest advantage in that routing automatically adapts to a change in the network's topology.
Dynamic routing is the most commonly used method of routing. Associated with this type of routing are several terms and characteristics that define how it operates, such as convergence and accuracy.
In dynamic routing, the routing table must consistently reflect accurate and up-to-date information concerning the network topology. The amount of time that it takes for changes to be reflected in every network router's routing table is known as the convergence time. Convergence, in this context, is the act of approaching the state where all routing tables have congruent information and are in steady state.
Having a routing protocol with a fast convergence time is desirable because disruption of routing can occur during the time that a router spends calculating the new optimal path.
