In Link-State Routing, aka Shortest Path First (SPF), each #Router needs to know the entire topology of the network with the aid of a database. The database stores link states, that is information about the state of a router’s links, which include IP Address, type of network (Ethernet or serial point-to-point link), cost, and neighbours. This database will be identical to all routers using the same protocol albeit processed independently. The #exchange of topology information (called flooding) only happened upon the initial set-up of the network and when there is a topological change in the network (new router added or router is removed). Since only information specific to the change needs to be propagated, combined with the hierarchical model used by the protocol, it limits the scope of the route changes that occur and provides better scaling than Distance Vector Routing Protocol. Though there still a need for keep alive message between router and its neighbours to detect potential topological changes.
Note: Since flooding requires a lot of computing resources especially in a large network, it is advice to logically divide the routers into areas so that flooding will only occur within that particular area.
Most Link-State Routing Protocols run Dijkstra’s Algorithm# to calculate the possible shortest path to determine the network routing.
It is best used in a Hierarchical Network and fast convergence# of the network is crucial.
Examples of such protocol:
- Open Shortest Path First (OSPF)#
- Intermediate System to Intermediate System (IS-IS)#
- OSPFv3 (designed for IPv6 Address)