Foundation Summary

     

The "Foundation Summary" section of each chapter lists the most important facts from the chapter. Although this section does not list every fact from the chapter that will be on your exam, a well-prepared candidate should, at a minimum, know all the details in each "Foundation Summary" before going to take the exam.

Table 5-2 is a summary of IP routing protocols and the update timers.

Table 5-2. A Summary of IP Routing Protocols and the Update Timers

Protocol	Update Timer	Technology
RIPv1	Every 30 seconds for entire routing table.	Distance vector.
RIPv2	Every 30 seconds for entire routing table.	Distance vector.
OSPF	Incremental with only the network change. However, 30 minutes after the last update was received, a compressed version of the table is propagated.	Link state.
EIGRP	Incremental updates with network change only.	Advanced distance vector, sometimes called enhanced distance vector or a hybrid routing protocol.
IGRP	Updates every 90 seconds with incremental updates as needed.	Distance vector.
BGP-4	Incremental with only the network change.	Path vector, sometimes referred to as a type of distance vector routing protocol.
IS-IS	Incremental with only the network change. However, the router that originated the LSP must periodically refresh its LSPs to prevent the remaining lifetime on the receiving router from reaching 0. The refresh interval is 15 minutes. This means that approximately 15 minutes after the last update was received, a compressed list of all the links the router has knowledge of is sent to all routers.	Link state.

Table 5-3 summarizes the major differences between distance vector routing protocols and link-state routing protocols.

Table 5-3. Distance Vector Routing Protocols Versus Link-State Routing Protocols

Distance Vector	Link-State
Sends its entire routing table at periodic intervals out of all interfaces (typically, this is based in seconds). Sends triggered updates to reflect changes in the network.	Sends incremental updates when a change is detected . OSPF will send summary information every 30 minutes, regardless of whether incremental updates have been sent in that time.
Typically involves updates sent using a broadcast address to everyone on the link.	Typically involves updates sent to those routers participating in the routing protocol domain, via a multicast address.
Uses a metric based on how distant the remote network is to the router. (IGRP does not conform to this as a proprietary solution.)	Is capable of using a complex metric, referred to by OSPF and IS-IS as cost.
Has knowledge of the network based on information learned from its neighbors.	Has knowledge of the network based on information learned from every router in the area.
Includes a routing table that is a database viewed from the perspective of each router.	Has a topological database that is the same for every router in the area. The routing table that is built from this database is unique to each router.
Uses the Bellman Ford algorithm for calculating the best path.	Uses the Dijkstra algorithm.
Does not consume many router resources, but is heavy in the use of network resources.	Uses many router resources, but is relatively low in its demand for network resources.
Maintains one domain in which all the routes are known.	Has a hierarchical design of areas that allow for summarization and growth.
Is not restricted by addressing scheme.	For effective use, the addressing scheme should reflect the hierarchical design of the network.
Involves slower convergence because information of changes must come from the entire network (but indirectly). Each routing table on every intervening router must be updated before the changes reach the remote end of the network.	Involves quicker convergence because the update is flooded immediately throughout the network.

Table 5-4 summarizes the differences between RIPv1 and OSPF. RIPv1, as the first distance vector routing protocol, and OSPF, as the first link-state routing protocol, are very familiar to most in the networking industry and thus easily used as examples for comparison.

Table 5-4. RIPv1 Versus OSPF

RIPv1	OSPF
Is a simple protocol to design, configure, and maintain.	Is a complex protocol to design and, in some instances, to configure and maintain.
Does not require a hierarchical addressing scheme.	If full benefits of the protocol are to be harnessed, should use a hierarchical IP addressing scheme.
Does not pass the subnet mask in the routing update and therefore is not capable of classless routing or VLSM.	Carries the mask in the update and therefore can implement VLSM, summarization, and classless routing.
Is limited to a 15-hop diameter network.	Is unlimited in the diameter of the network, although it is suggested that an area not exceed more than 50 networks.
Does not acknowledge routing updates; just repeats them periodically (every 30 seconds).	Acknowledges updates.
Has a routing table that is sent out of every interface every 30 seconds (by default).	Involves updates sent as required (when changes are seen) and every 30 minutes after no change has been seen.
Can transmit information about the network in two messages: the routing update and the triggered update.	Has protocols for discovering neighbors and forming adjacencies, in addition to protocols for sending updates through the network. These protocols alone add up to nine message types.
Uses hop count as a metric, the number of routers to process the data.	Uses cost as a metric. Cost is not stated in the RFCs, but it has the capacity to be a complex calculation, as seen in Cisco's implementation.

Table 5-5 summarizes the major differences between all available IP routing protocols.

Table 5-5. Comparison Chart for IP Routing Protocols

	RIPv1	RIPv2	IGRP	EIGRP	OSPF	IS - IS	BGP
Distance Vector							Path vector
Link State
Classless
Classful
VLSM
Manual Summarization
Automatic Summarization at IANA boundary
Metric	Hop	Hop	Composite	Composite	Cost	Cost	Path attributes
Max Hop Count	15	15	255	255
Update Timers	30 sec	30 sec + triggered	90 sec + triggered	Triggered	30 min + triggered	Synchronized every 15 min on broadcast media+ triggered	Triggered
Hello	None	None	None	60 sec <T1, everything else 5 sec	30 sec WAN, everything else 10 sec	10 sec	60 sec
Dead Time	180 sec, flush after 240 sec	180 sec, flush after 240 sec	3 * hello	3 * hello	4 * hello	Hold for 30 sec	180 sec