Technical Considerations

Table 2-2 provides a list of technical issues to consider when selecting a routing protocol.

Table 2-2 Important Technical Considerations: IS-IS Versus OSPF

OSPF

IS-IS

Fast Convergence

Yes

Yes

Routing Updates

Fast, change only

Fast, change only

VLSM Support and CIDR

Yes

Yes

Load Sharing

Yes, equal cost

Yes, equal cost

Metric Range

0-65,535

0-1023

Static Metric Pieces

Sum of bandwidth

Sum of bandwidth

Dynamic Metric Pieces

None

None

Scalability

Very Strong

Strong

Physical Media Support

All types

Most, but some issues

Extensibility

Yes w/Opaque LSAs

All routing protocols have three important characteristics when dealing with the issue of convergence:

1 Detecting that a change has occurred

2 Adapting to that change

3 Updating the network topology to reflect the change

IS-IS and OSPF detect certain types of network changes instantly. In general, any change that can be detected by a physical change (such as loss of carrier) is detected immediately by any routing protocol.

In addition, both routing protocols use hello packets as keepalives and to detect other failures (such as the loss of an adjacent router or the degradation of an interface to the point where it should no longer be used). Both protocols cause adjacent routers to exchange information periodically.

After the routing protocol has detected the topology change, it needs to adjust the routing tables to accommodate the new topology. OSPF and integrated IS-IS both have mechanisms for updating routing tables. If the topology change were within the area, all the existing routes affected by the change would be discarded and a new routing table would be generated. In general, OSPF and integrated IS-IS converge in less than 2 seconds. The amount of CPU time required to do the recompilation is strongly affected by the number of routes and the amount of redundancy in the network.

Routing Updates

All routing protocols exchange routing information dynamically. The three most important questions concerning the operation of routing updates are as follows:

• When are they sent? — All three routing protocols exchange periodic hellos and full topology information when a router starts up and periodically thereafter, depending how they are configured. RIP floods the full topology table every 30 seconds. OSPF floods the full topology table every 30 minutes. Integrated IS-IS floods the full topology table every 15 minutes to ensure synchronization.

• What is in them? — Within an area, OSPF and integrated IS-IS exchange changed link-state information. Between areas, OSPF and integrated IS-IS exchange changed routes.

• Where are they sent?—Changed information in a RIP network is broadcast to all of its neighbors after the network has finished updating its topology. Changed information in OSPF and integrated IS-IS propagates throughout the area in which the change occurred. If route summarization is not done, change information might also propagate to the backbone and into other areas.

VLSM and CIDR Support

OSPF and integrated IS-IS include support for variable-length subnet masks (VLSMs) and classless interdomain routing (CIDR). VLSM is required to support route summarization. In addition, VLSM and CIDR also enable network administrators to use their address space more effectively.

Load Sharing

Today's networks are commonly designed with redundant paths. This has two positive benefits: rerouting in case of failure and load sharing. All routing protocols supported by Cisco provide load sharing across as many as six equal-cost paths. The default for OSPF is to use four equal-cost paths, but if you have more, you must configure OSPF to use them.

Metrics

The quality of route selection is essentially controlled by the value of the metrics placed upon the various routes. Two components are important in how a routing protocol uses metrics: the range of the values the metric can express and how the metric is computed.

OSPF uses a flat metric with a range of 16 bits. This results in OSPF having a metric range that is from 0 to 65,535. By default, OSPF metrics are assigned as an inverse of the bandwidth available on an interface—normalized to give the Fiber Distributed Data Interface (FDDI) a metric of 1. OSPF computes the cost of a path by summing the metrics for each hop on that path.

62 Chapter 2: Introduction to OSPF

Integrated IS-IS uses a flat metric. The metric range is 0 to 1023. By default, all integrated IS-IS metrics are 10. Network administrators need to configure nondefault values. Integrated IS-IS computes the cost of a path by summing the metrics for each hop on that path.

Scalability

In IS-IS, ISO 10589 states that 100 routers per area and 400 L2 routers should be possible. The biggest scaling issue now seems to be the overhead of the flooding in large meshed networks, for example, flat ATM clouds with many attached routers that form a full mesh.

OSPF, on the other hand, scales well regardless of the size of the network. However, to make the network operate optimally, you should implement physical and logical areas as needed.

Physical Media Support

Both OSPF and IS-IS support point-to-point links and LANs in similar ways. However, IS-IS has no nonbroadcast multiaccess (NBMA) support and expects the router operating system to present the NBMA network as either a LAN or a set of point-to-point links; this can be problematic. OSPF can overcome this issue, but with Cisco routers, considerable configuration is needed.

Extensibility

Protocols must be able to grow and expand to meet the ever-changing and evolving network environment. OSPF has been able to do this through the recent inclusion of opaque LSAs; however, all routers must understand these LSAs to effectively adapt to the dynamic network environment. IS-IS floods unknown LSAs and ignores them.

Both protocols support traffic engineering, so networks can benefit from Multiprotocol Label Switching (MPLS) regardless of which protocol is chosen.

Was this article helpful?

0 0

Post a comment