Link - State Algorithm
Link - State Algorithm
Each router identifies and forms adjacencies with directly connected routers. It uses Hello
packets to establish and maintain these neighbor relationships.
3. Flooding of LSAs
The LSA is flooded throughout the network so that every router receives the same set of
link-state information. Each router stores these LSAs in its Link-State Database (LSDB).
The LSDB is essentially a topology map of the entire network, showing how routers and
networks are interconnected.
Using the LSDB, each router independently runs Dijkstra's algorithm to compute the
shortest path tree, which determines the best routes to all destinations.
Based on the computed shortest paths, the router updates its routing table, which is used to
forward data packets.
Examples of Link-State Routing Protocols
Protocol Description
OSPF (Open Shortest Path A widely used interior gateway protocol (IGP) that uses
First) link-state principles. Supports hierarchical design with
areas.
IS-IS (Intermediate System to Originally developed for OSI networks, now also
Intermediate System) supports IP routing. Used in large service provider
networks.
Each router has a complete and synchronized map of the network. This leads to better
route decisions compared to distance-vector protocols.
2. Faster Convergence
Link-state protocols respond quickly to network changes. When a link goes down or
changes, only the affected part of the network is recalculated, minimizing downtime.
3. Highly Scalable
Supports hierarchical network designs. For example, OSPF uses areas (like Area 0, Area
1, etc.) to divide the network into manageable sections.
4. Loop-Free by Design
By using Dijkstra’s algorithm and complete network knowledge, link-state routing avoids
routing loops which are common in poorly configured distance-vector protocols.
Updates are triggered by changes, not sent periodically. Only the specific change (LSA) is
shared, reducing unnecessary traffic.
• When a router starts, it floods LSAs to all routers, which may cause a temporary
spike in network traffic.
• The router must store complete topology data instead of just next-hop information.
• This can be a burden in memory-constrained devices.