Routing - Distance Vector vs Link State
There are 2 types of routing protocols. Distance Vector vs Link State.
Distance vector protocols use a vector of distance and direction to find the best path, where distance is defined in terms of a metric and direction is defined as the next-hop router. Distance vector protocols broadcast the entire routing table at periodic intervals and are slow to converge because of hold-down timers. Examples include RIP and IGRP. Distance Vector routing protocols advertise all the known routes out off each interface.
- Route Poisoning – Gives a “down” route a infinite metric “16” to show it is down.
- Split Horizon – Updates cannot be sent out the same interface that they came in on.
- Split Horizon with Poison Reverse – Failed Routes are given a infinite metric but failed routes are advertised out the interface which the update came from.
- Hold-Down timer – When the router learns that a route has failed ignore any other update for that route for the certain period.
- Triggered Updates – When a route fails it sends a flash update regarding the routes status.
Link-state protocols use a topological database that is created on each router. This database keeps track of directly attached neighbors, the entire network, and the routing table. Link-state updates are typically multicast to all neighbors. (OSPF is a link-state protocol.)