How does a link-state protocol differ from a distance-vector protocol?

Link-state protocols are distinct from distance-vector protocols primarily because they maintain a complete topology of the network. This means that each router using a link-state protocol has a full representation of the entire network layout, including each link and the state of each link. When a link-state router discovers changes in the network, it shares information about the state of its connected links with all other routers in the network, allowing them to build a consistent and up-to-date view of the network topology.

In contrast, distance-vector protocols only share information about the distance to other routers, as opposed to the overall state of the links within the network. This results in each router relying on its neighbors to inform it of the network layout, which can lead to issues such as routing loops and inconsistencies in the network view.

While the bandwidth usage of link-state protocols can often be lower in environments with numerous changes, this is not always the case, as the initial exchange of link-state information can be substantial. Furthermore, the claim about convergence speed varies; in many cases, link-state protocols converge faster than distance-vector protocols due to their sharing of complete topology information. Therefore, the key differentiator is indeed that link-state protocols maintain a comprehensive view of the network topology.