| The best way to understand the intricacies of the OSPF protocol is to compare it with a more familiar interior routing protocol: the Routing Information Protocol version 1 (RIPv1). OSPF implements a Dijkstra Shortest Path First algorithm to populate a routing table, whereas RIPv1 uses the Bellman-Ford distance vector algorithm. Here is a list of the fundamental differences between and features of the RIP and OSPF interior routing protocols: -
OSPF converges a lot faster than RIPv1 because it propagates changes immediately. This results in more timely and accurate routes as well as fewer packets being lost while the network topology is converging. -
RIPv2 and OSPF both support Variable Length Subnet Mask (VLSM), whereas RIPv1 does not. -
RIP networks cannot grow larger than 15 hops; OSPF networks are technically unlimited in size. -
RIP uses much more bandwidth because of its distance-vector behavior. RIP broadcasts the entire routing table every 30 seconds by default. OSPF sends a multicast update only when a change to the topology occurs. -
RIP does not understand factors such as delay and bandwidth cost. RIP bases the routing decision completely on hop count as a metric, which can result in pinhole congestion. -
OSPF can utilize VLSM masks for route summarization, which can reduce the number of routing table entries, as well as the amount of update traffic and router overhead. -
OSPF enables you to tag external routes coming from other autonomous systems to help identify and manage routing between OSPF and other non-OSPF systems for administrative purposes. -
Both OSPF and RIP can load balance packets across equal cost paths (four by default). -
The Cisco IOS offers a richer set of tools for configuring, monitoring, and troubleshooting with OSPF. On the other hand, RIP is less difficult to configure, monitor, and troubleshoot than the more complex and configurable OSPF routing protocol. -
Although OSPF is run on a wide variety of routing devices, it can consume a considerable amount of CPU and memory resources, especially on high-end routers where many shortest path first (SPF) calculations are performed for multiple OSPF processes. -
OSPF and RIPv2 ”but not RIPv1 ”offer authentication in the packet.  | It is highly recommended that you download the entire RFC 2328 and read this 100+-page document at least one time through. Although it is not leisure reading, the benefits are substantial for additional exam preparation. For instance, it is an excellent source for a complete terminology and definition of the OSPF protocol version 2. |
Now that you have a general idea of the many advantages of OSPF in scalable networks, it is imperative to master the many terms and concepts used with OSPF. Refer to the following list throughout this chapter and the rest of the book: -
Link ” The interface between an OSPF router and a connected network. -
Link-state ” Condition of the link between two routers sharing link-state advertisements. Information is used by OSFP routers to build SPF topology. Depending on the implementation, the link can be in the down, init, two-, exstart, exchange, loading, or full state. -
Link-state advertisement (LSA) ” OSPF broadcast packets that contain updates used to maintain routing tables that contain data about neighbors and path costs. -
Area ” Grouping of routers and networks with the same area ID number and link-state information. All OSPF routers must be in an area. -
Autonomous System (AS) ” Collection of routers that share OSPF routing information and are under the same network administration. For example, an entire company could be an AS administered by a single IT group . On the other hand, a large corporation or organization might divide administration into ASs across geographical, political, departmental, or logical boundaries for the sake of managing costs, migration, and other corporate operations. -
Cost ” The metric OSPF uses. Represented by a numerical value assigned to a link based on the speed of the connection output. A lower cost is preferable to a higher cost. -
Router ID ” Unique number on a Cisco router that is typically represented by the highest configured IP address, the highest configured IP address loopback address, or manual assignment. -
Neighbor ” Two routers that have links on a common network. -
Hello ” A type of packet used by OSPF to establish and manage neighbor relationships. -
Topology table ” A table that holds every link in the entire network. Also known as the link-state database. -
Adjacency ” Condition that occurs when two OSPF routers have traded information to the point where they have identical topology tables. -
Routing table ” Table created when the SPF algorithm is run on the link-state database. Also called the forwarding database . -
Designated router (DR) ” OSPF router with many duties in a multi-access network, including generating LSAs. Having a DR lowers the number of adjacencies necessary in a multi-access topology and reduces traffic and topology database size. -
Backup designated router (BDR) ” Receives the same data as the DR but functions as a standby DR in the case of a DR failure. -
Backbone ” The primary path for traffic on a network; the part of the network that is most often sourced from as well as transmitted to other networks. -
Area border router (ABR) ” A router that sits on the border of more than one OSPF area and connects those routers to the backbone. -
Autonomous system boundary router (ASBR) ” An area Border Router (ABR) that sits between an OSPF AS and a non-OSPF network (such as IP) and runs multiple routing protocols. An ASBR can also link two OSPF networks from different autonomous systems. -
Router priority ” An 8-bit number (1 “255) that designates the priority of a router in the DR/BDR election process and can be manually reconfigured as necessary. |
