OSPF Cost Calculation

OSPF (Open Shortest Path First) uses a link-state routing algorithm (Dijkstra's SPF) to compute the shortest path tree to every destination. Unlike RIP's hop count or EIGRP's composite metric, OSPF uses a dimensionless metric called cost. The cost is inversely proportional to the bandwidth of the interface: higher bandwidth = lower cost. This design allows OSPF to prefer faster paths over slower ones, which is more realistic for modern networks where bandwidth is often the bottleneck, not hop count.

The standard formula on Cisco IOS is:

Cost = Reference Bandwidth / Interface Bandwidth

Where:

The result is an integer; fractional values are rounded down. For example: - 10 Mbps Ethernet: cost = 100 / 10 = 10 - 100 Mbps FastEthernet: cost = 100 / 100 = 1 - 1 Gbps GigabitEthernet: cost = 100 / 1000 = 0.1 → truncated to 1 (since cost must be at least 1) - T1 (1.544 Mbps): cost = 100 / 1.544 ≈ 64.8 → truncated to 64

Wait—1 Gbps gets cost 1? That's the same as 100 Mbps! This is a problem: OSPF cannot distinguish between 100 Mbps and 1 Gbps with default reference. That's why engineers often change the reference bandwidth to a higher value, typically 10000 (10 Gbps) or 100000 (100 Gbps), using the command auto-cost reference-bandwidth under the OSPF process. For example, with reference 10000: - 100 Mbps: cost = 10000 / 100 = 100 - 1 Gbps: cost = 10000 / 1000 = 10 - 10 Gbps: cost = 10000 / 10000 = 1

#### Show the OSPF cost on an interface:

R1# show ip ospf interface gigabitEthernet 0/0/0
GigabitEthernet0/0/0 is up, line protocol is up
  Internet Address 10.0.0.1/24, Area 0
  Process ID 1, Router ID 1.1.1.1, Network Type BROADCAST, Cost: 1
  Transmit Delay is 1 sec, State DR, Priority 1
  Designated Router (ID) 1.1.1.1, Interface address 10.0.0.1
  Backup Designated router (ID) 2.2.2.2, Interface address 10.0.0.2
  Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5
    oob-resync timeout 40
    Hello due in 00:00:05
  Supports Link-local Signaling (LLS)
  Cisco NSF helper support enabled
  IETF NSF helper support enabled
  Index 1/1/1, flood queue length 0
Next 0x0(0)/0x0(0)/0x0(0)
  Last flood scan length is 1, maximum is 25
  Last flood scan time is 0 msec, maximum is 4 msec
  Neighbor Count is 1, Adjacent neighbor count is 1
    Adjacent with neighbor 2.2.2.2  (Backup Designated Router)
  Suppress hello for 0 neighbor(s)

The line Cost: 1 shows the OSPF cost for this interface.

#### Show the OSPF route with cost:

R1# show ip route ospf
     10.0.0.0/8 is variably subnetted, 3 subnets, 2 masks
O       10.0.1.0/24 [110/2] via 10.0.0.2, 00:12:34, GigabitEthernet0/0/0
O       10.0.2.0/24 [110/3] via 10.0.0.2, 00:12:34, GigabitEthernet0/0/0

The [110/2] shows the administrative distance (110) and the OSPF metric (cost) of 2 for that route.

#### Show the LSDB to see cost in LSAs:

R1# show ip ospf database router 1.1.1.1

     OSPF Router with ID (1.1.1.1) (Process ID 1)

                Router Link States (Area 0)

  LS age: 120
  Options: (No TOS-capability, DC)
  LS Type: Router Links
  Link State ID: 1.1.1.1
  Advertising Router: 1.1.1.1
  LS Seq Number: 80000005
  Checksum: 0x1234
  Length: 48
  Number of Links: 2

    Link connected to: a Transit Network
     (Link ID) Designated Router address: 10.0.0.1
     (Link Data) Router Interface address: 10.0.0.1
      Number of TOS metrics: 0
       TOS 0 Metrics: 1

    Link connected to: a Stub Network
     (Link ID) Network/subnet number: 10.0.1.0
     (Link Data) Network Mask: 255.255.255.0
      Number of TOS metrics: 0
       TOS 0 Metrics: 1

The TOS 0 Metrics: 1 shows the cost of that link.

You have a data center with servers connected via 10 Gbps links and older 1 Gbps links. With default OSPF reference bandwidth (100 Mbps), both link types get cost 1, so OSPF sees them as equal cost and load balances. But you want to prefer the 10 Gbps links for primary traffic and use 1 Gbps as backup. Solution: Change reference bandwidth to 10000 (10 Gbps) on all routers. Then 10 Gbps gets cost 1, 1 Gbps gets cost 10. Now OSPF will prefer the 10 Gbps path. If the 10 Gbps link fails, traffic shifts to 1 Gbps. This is a common enterprise scenario. Misconfiguration: if you forget to change reference on one router, it may calculate different costs and cause routing asymmetry or suboptimal paths.

You have two WAN links between sites: a primary MPLS circuit (100 Mbps) and a backup DSL (10 Mbps). You want the primary to be preferred. With default reference, 100 Mbps gets cost 1, 10 Mbps gets cost 10. That's fine. But suppose the primary link is a satellite link with high latency but same bandwidth as DSL. Bandwidth alone doesn't reflect latency. You might manually set a higher cost on the satellite link to avoid it. Use ip ospf cost 20 on the satellite interface. This overrides the automatic calculation. In production, manual cost is often used for policy-based routing without changing bandwidth.

Company A uses reference 100 (default), Company B uses reference 1000. When they merge their OSPF networks, routers will compute different costs for the same links, causing inconsistent path selection. The fix is to standardize the reference bandwidth across the entire OSPF domain. This is a real-world challenge during network mergers. The engineer must plan a maintenance window to update all routers. Failure to do so can cause routing loops or suboptimal routing. Always document the reference bandwidth used in your network.

OSPF SPF calculation runs when a link state change occurs. The cost change triggers an LSA update and SPF recalculation. In large networks, frequent cost changes can cause CPU spikes. Use route summarization and stub areas to limit LSDB size. Also, avoid changing costs on many interfaces simultaneously.