What is hierarchical routing?

In hierarchical routing, the routers are divided into regions. Each router has complete details about how to route packets to destinations within its own region, but it does not have any idea about the internal structure of other regions.

As networks grow larger, the number of routers increases significantly. In traditional routing algorithms like Link State (LS) and Distance Vector (DV), every router needs to maintain information about all other routers. This causes routing tables to become extremely large, reducing efficiency. Hierarchical routing solves this scalability problem by organizing routers into a structured hierarchy.

How Hierarchical Routing Works

In hierarchical routing, routers are classified into groups called regions. Each router maintains detailed routing information only for routers within its own region and stores a single entry for each external region. This dramatically reduces the size of routing tables while maintaining connectivity across the entire network.

Multi-Level Hierarchies

For huge networks, a two-level hierarchy may be insufficient. Networks can be organized into multiple levels: regions grouped into clusters, clusters into zones, zones into groups, and so on. Each additional level further reduces routing table sizes.

Example: Two-Level Hierarchy

Consider router 1A in a five-region network. Without hierarchy, its routing table would have 17 entries for all individual routers. With hierarchical routing, the table is reduced to just 7 entries:

Trade-offs

While hierarchical routing significantly reduces table sizes, it comes with increased path length. The routing may not always choose the optimal path because it routes based on regions rather than individual routers. For example, traffic from region 1 to region 5 might take a longer path through region 3 instead of the shorter path through region 2.

Scalability Benefits

Consider a network with 720 routers. Without hierarchy, each router needs 720 routing entries. With hierarchical organization:

• Two-level hierarchy − 24 regions of 30 routers each results in 53 entries per router (30 local + 23 remote)
• Three-level hierarchy − 8 clusters of 9 regions of 10 routers each results in only 25 entries per router (10 local + 8 regions + 7 clusters)

Conclusion

Hierarchical routing provides an elegant solution to the scalability problem in large networks by organizing routers into regional hierarchies. While it trades some path optimality for significant reductions in routing table sizes, it enables efficient routing in networks that would otherwise be unmanageable.