background image
Determining IGRP/EIGRP Routing, Characteristics, and Convergence 233
the techniques of summarization, the OSPF processing load on a router can be reduced.
With summarization configured properly, there are fewer LSA packets to be sent and
acknowledged. The following sections describe how bits can be split to efficiently summarize
a major network number.
Bit-Splitting the Address Space
To efficiently summarize, networks should be assigned in bit boundaries. To divide a major
network number across more than one area, use bit splitting. To divide the network into four
areas, you need 2 bits. With 2 bits, the result is 00, 01, 10, and 11.
To differentiate 16 bits, split 4 bits. Table 7-3 shows the number of bits used. Bit splitting
borrows subnet bits to designate areas. To differentiate two areas, split 1 bit.
Discontiguous Subnets
Subnets become discontiguous when they are separated by one or more subnets. When two
major networks are separated by another major network, the network flow lacks continuity and
becomes discontiguous. OSPF supports discontiguous subnets. The wise CCDP will avoid this
design whenever possible.
Determining IGRP/EIGRP Routing, Characteristics, and
Convergence
Cisco developed IGRP to be a proprietary protocol that was a simple, robust, and more scalable
routing protocol than RIP. Although RIP worked fine in small, homogenous networks, it was
found to be harshly restricted in larger internetworks because of its limit of 16 hop-counts.
Indeed, the use of the hop-count as its only metric ensures that RIP will never have the
flexibility it needs to perform successfully in a large, complex network. IGRP was intended to
overcome the shortcomings and limitations of RIP. Because IGRP is a distance-vector protocol,
it shares several features that are common to RIP. Like RIP, IGRP broadcasts its entire routing
Table 7-3
Bit Splitting
Area
Bits
4
2
8
3
16
4
32
5
64
6
128
7
87200333.book Page 233 Wednesday, August 22, 2001 2:37 PM