This tutorial explains static and dynamic routing, Administrative distance, Autonomous System, Routing update, Routing Metric and Routing protocols in detail with examples.
Routing protocol shares routing update that contains network information. Routers keep this information in routing tables. Routing tables are used to determine the best path for packet. Routing protocols uses several concepts to determine the best path.
IP routing is the process of moving data packets between different networks. By default two different IP networks cannot communicate with each other. They need a mediator device that can switch packet between them. Router takes this responsibility. Routers interfaces are associated with different networks. This association is kept in routing table. Routers use it to take switching decision.
Following figure illustrates a simple routing example.
- 10.0.0.0/8 Network is connected on router’s F0/1 interface.
- 126.96.36.199/8 Network is connected with router’s F0/2 interface.
- Laptop sends a packet to PC.
- Router receives this packet in F0/1 interface.
- Router checks destination address field in packet.
- Packet has 188.8.131.52/8 address in destination address field.
- IP address 184.108.40.206/8 belongs to 220.127.116.11/8 network.
- Router checks routing table for matching network.
- Routing table has an entry for 18.104.22.168/8 network.
- 22.214.171.124/8 Network is associated with F0/2 interface of router.
- Based on this information router moves this packet from F0/1 to F0/2.
- F0/2 interface sends this packet to its destination.
It is the easiest example of IP routing. In above example we used only one router to explain the process. In real life there could be several routers in the way of destination address. These routers may create certain path to the destination. Routing protocols insure that packet takes the best path from these paths.
To accomplish this task, a router will need to perform following:
- Learn about the directly connected routers on its interface
- Find the networks that they have
- Build routing table from this information
- Choose best network from routing table to send the packet
No additional configuration is required for directly connected networks. Router already knows how to get it. Such as in above example both networks 10.0.0.0/8 and 126.96.36.199/8 were directly connected to the router’s interfaces.
Routers learn about remote networks from an administrator or from neighbor routers.
When routers learn from an administrator, it is called static routing. In static routing we have to add all network locations manually. If any change occurs in network, administrator is responsible to update it by hand in all routers.
Advantage of static routing
- It is easy to implement.
- It is most secure way of routing, since no information is shared with other routers.
- It puts no overhead on resources such as CPU or memory.
Disadvantage of static routing
- It is suitable only for small network.
- If a link fails static route cannot reroute the traffic.
When routers learn from neighboring router through the routing protocols, it is called dynamic routing. In dynamic routing routers add network locations automatically form the routing information. If any change occurs in network, affected routers update others via routing information.
Advantage of dynamic routing
- It is suitable for all type of networks.
- Automatically build routing tables.
- Reroute the traffic from possible network, in link failure condition.
Disadvantage of dynamic routing
- It is hard to implement.
- It is less secure, since it shares routing updates with other routers.
- It puts additional overhead on resources such as CPU, memory and link bandwidth.
Depending on network requirement we can use either static routing or dynamic routing. Even more we can use a combination of both dynamic and static routing.
Routing update is a mechanism of sharing information with neighboring routers. In a particular time duration router advertise its routing information through broadcast or multicast. Different protocols have different time intervals. Some protocols use broadcast for routing updates while some uses multicast. Routing updates contain all necessary information for routing protocol such as learned network, timers, AS, AD, matrix values, interface details etc.
Autonomous System (AS) is a collection of routers that share same routing table information. AS is a boundary line for routing protocol. It could be your company, or group of companies. It is defined by a numeric value. To distinguish between different AS, Internet Assigned Numbers Authority (IANA) provides a range from 1 to 65535. There are two types of AS, private and public. Private AS numbers are used for internal network. Public AS numbers are used for internet backbone.
Interior Gateway Protocols
Interior gateway protocols (IGPs) are used to share routing updates between routers in same AS. Some examples of IGPs are RIPv1, RIPv2, EIGRP, IGRP, and OSPF.
Exterior Gateway Protocols
Exterior gateway protocols (EGPs) are used to share routing updates between different AS. Example of EGPs is Border Gateway Protocol (BGP).
Administrative distance (AD) is the trustworthiness of routing update received from a neighbor router. If a router receives two routing updates for same path from two different routing protocols then router will check the AD value to choose the best path. AD is a numeric value from 0 to 255. If one update has lower AD value than other, then the route with the lowest AD will be placed in the routing table
|Route source||Default AD value|
|Direct connected interface||0|
Lower AD value is more believable by router. 0 is considered as the most trustworthiness network while 255 is considered as invalid route and it will be never used.
If two routing updates for same network have same AD value then metric will use to choose the best path. Metric is a measurement to calculate best path. Route with the lowest metric will be chosen. Different routing protocols use different metrics. It may use single metric or multiple metrics. For example EIGRP uses bandwidth, delay, load, MTU and reliability while RIP only uses hop count as metric.
|EIGRP||Bandwidth||Capacity of link in Kbps|
|EIGRP||Delay||Time to reach in destination|
|EIGRP||Load||Path that is least utilize|
|EIGRP||MTU||Path that support largest frame size|
|EIGRP||Reliability||Path that have least down time|
|OSPF||Cost||Inverse of bandwidth links|
|RIP||Hop count||Hops ( Routers) in the way of destination|
There are three types of routing protocols:-
- Distance Vector
- Link State
Distance vector routing protocol uses distance (metric value) and direction (vector) to find the best path to destination network. Router receives routing update from neighboring router and these neighboring routers receive updates from their neighboring routers until the destination network. Every router in the way of destination network called hop. Each time a packet goes through a router, it add one in hop count value. Route with the least hop count value will be chosen as best path and will be placed in routing table. RIP is the example of distance vector routing protocol. These protocol shares entire routing table to the directly connected neighbors.
Link state routing protocols use more composite metric to locate the best path for destination network. It maintains three separate tables. First table keeps track of directly connected neighbors. Second table determines the entire network topology. Third is the routing table that keeps actual path. OSPF is the example of link state protocol. Link state protocols share their own links to all other routers in network.
Hybrid routing protocols are the mix of distance vector and link state protocol. To locate more accurate path, it uses aspect from both distance vector and link state. EIGRP is the example of hybrid routing protocols.
All three types of protocol have their own advantage and disadvantage. They take different approach in sharing routing updates and in choosing the best path. In next articles of this section we will explain these protocols in details with examples.