Why Use OSPF Routing

Open Shortest Path First (OSPF) is a link-state interior gateway protocol (IGP) widely deployed in enterprise and service provider networks. Unlike distance-vector protocols such as RIP, OSPF builds a complete view of the topology using link-state advertisements (LSAs) and runs Dijkstra’s SPF algorithm to compute loop-free shortest paths. This design gives OSPF fast convergence, rich control over route propagation, and excellent scalability for large, segmented networks.

1. Link-State vs Distance-Vector

Traditional distance-vector protocols advertise routes as simple hop counts to neighbors, which then propagate those routes further. This can lead to slow convergence and issues such as counting to infinity. OSPF, as a link-state protocol, takes a more sophisticated approach:

• Each router floods LSAs describing its local links and costs to all routers in the OSPF area.
• All routers maintain an identical Link-State Database (LSDB), representing the full topology.
• Dijkstra’s Shortest Path First (SPF) algorithm is run locally to compute the routing table.

Because every router has a synchronized view of the network, OSPF can converge much faster after a topology change than distance-vector protocols.

2. Cost Metric and Path Selection

OSPF does not rely on simple hop counts. Instead, it uses a cost metric derived from interface bandwidth. By default (in many vendors), cost is calculated using a reference bandwidth divided by the interface bandwidth. On Cisco IOS, the default reference bandwidth is 100 Mbps:

OSPF cost = reference-bandwidth / interface-bandwidth

For example, a 100 Mbps interface gets cost 1, while a 10 Mbps link gets cost 10. You can tune this behavior:

router ospf 1
 auto-cost reference-bandwidth 10000   ! Treat 10 Gbps as cost 1
!

This allows network engineers to prefer higher-capacity links and create deterministic routing behavior across the topology.

3. Areas and Hierarchical Design

One of OSPF's biggest advantages is support for a hierarchical routing design using areas. Every OSPF domain must contain a backbone area (Area 0), and additional non-backbone areas can be attached to it via Area Border Routers (ABRs).

• Backbone Area (0): Core of the OSPF domain; all other areas must connect logically to Area 0.
• Regular Areas: Receive a summarized view of the rest of the network, reducing LSDB size.
• Stub, Totally Stubby, and NSSA Areas: Special area types that control which external or inter-area routes are injected, optimizing memory and CPU usage on edge routers.

This hierarchy makes OSPF extremely scalable compared to flat routing protocols. Instead of every router knowing every external prefix, OSPF can summarize at area boundaries and reduce routing-table size.

4. Fast Convergence and Reliable Neighbor Relationships

OSPF forms explicit neighbor adjacencies and tracks them using Hello packets and timeouts. This makes link failure detection and recovery deterministic and tunable.

Key parameters:

• Hello interval: How often OSPF Hello packets are sent.
• Dead interval: Time after which a neighbor is declared down if no Hellos are received.

interface GigabitEthernet0/0
 ip ospf hello-interval 5
 ip ospf dead-interval 20

By adjusting these timers (with care), you can make OSPF react faster to failures, improving convergence time for latency-sensitive applications.

5. Designated Router (DR) and BDR on Multi-Access Networks

On multi-access segments such as Ethernet LANs, having every router fully adjacent to every other router would create excessive LSA flooding. OSPF solves this by electing a Designated Router (DR) and Backup Designated Router (BDR) on each broadcast network.

• All other routers (DROTHERs) form full adjacencies only with the DR and BDR.
• LSAs are sent to a well-known multicast address and processed primarily by the DR/BDR.
• This significantly reduces the number of adjacencies and control-plane overhead.

Priority values can be used to influence the DR/BDR election on a per-interface basis:

interface GigabitEthernet0/1
 ip ospf priority 200   ! Higher value increases chance of becoming DR

6. OSPF Configuration Example (Cisco-Style)

Below is a simple configuration to illustrate how OSPF is typically deployed on an enterprise router. The example assumes two interfaces connecting to different subnets and participation in Area 0.

router ospf 1
 router-id 1.1.1.1
 log-adjacency-changes

 ! Advertise directly connected networks into OSPF Area 0
 network 10.0.0.0 0.0.0.255 area 0
 network 10.0.1.0 0.0.0.255 area 0

 ! Passive interfaces: don't form adjacencies on user-facing ports
 passive-interface GigabitEthernet0/2
!
interface GigabitEthernet0/0
 ip address 10.0.0.1 255.255.255.0
 ip ospf 1 area 0
!
interface GigabitEthernet0/1
 ip address 10.0.1.1 255.255.255.0
 ip ospf 1 area 0

Using passive-interface on non-routing interfaces reduces unnecessary OSPF Hellos and improves security by preventing rogue routers from joining the process.

7. External Routes and Redistribution

OSPF can import external routes (for example, from BGP, static routes, or another IGP) and advertise them into the OSPF domain as Type 5 (External) LSAs or Type 7 (NSSA) LSAs. These external routes are tagged as either E1 or E2:

• E2 (default): Uses only the external metric; internal path cost to the ASBR is not considered.
• E1: Adds internal cost to reach the ASBR plus the external metric, giving more accurate path selection in complex topologies.

router ospf 1
 redistribute static subnets metric-type 1   ! Inject static routes as OSPF E1

8. Operational Visibility and Troubleshooting

OSPF provides rich operational commands for troubleshooting and monitoring.

View the OSPF neighbor table:

show ip ospf neighbor

This displays neighbor states (FULL, 2-WAY, EXSTART, etc.), design role (DR/BDR/DROTHER), and adjacency status.

Inspect the OSPF database:

show ip ospf database

The LSDB can be examined to confirm that all routers share the same topology view and that LSAs are aging and refreshing correctly.

Review calculated routes:

show ip route ospf

Use this to verify which prefixes are installed as OSPF routes, what their metrics are, and through which next-hops they are reached.

Final Thoughts

OSPF remains a cornerstone routing protocol in modern networks because it balances fast convergence, scalability, and design flexibility. Its link-state architecture, area hierarchy, and tunable metrics make it suitable for everything from campus networks to large enterprise backbones. If you need deterministic, resilient, and well-organized routing behavior inside your autonomous system, OSPF is almost always the protocol of choice.