A large enterprise network has a partial mesh network with multiples redundant links. OSPF is used as IGP and it is implemented in a single-are

In a single OSPF area with many routers and redundant links, every router must store and process the entire LSDB and run Dijkstra's SPF algorithm upon each topology change, resulting in high CPU load and slow convergence. The partial mesh with multiple redundant links increases the number of LSAs flooded throughout the area, amplifying both the LSDB size and the frequency of SPF triggers. This correctly identifies the operational symptom that results from the described single-area design at enterprise scale.

Breaking the domain into separate OSPF areas is a valid remedy for the scalability problem, not a description of the symptom the network is experiencing.

Converting to hub-and-spoke is a topology change that reduces redundancy, not a description of the problem exhibited by the current partial-mesh design.

Replacing OSPF with BGP is a design decision proposed as a solution, not a symptom that results from the described single-area OSPF configuration.

Reducing the number of links is a remediation option to lower LSA flooding, not a description of the observed problem in the current network.

Upgrading hardware resources is a workaround to tolerate high resource usage, not an accurate characterization of the symptom caused by the design.