IP Address Assignment

IP Address Assignment
IP address assignment is a crucial aspect of network design. If you assign IP addresses to the network
arbitrarily, there will be no way that
route summarization
can take place. Your network
will not scale because the routing tables on all the routers in your network will become very
large and will have performance problems. As the number of routes grows, the amount of memory
needed to store and manipulate those routes increases. There is also increased CPU and network
utilization. When you design a network with summarization in mind, you will reduce the
size of the routing tables in all routers in your network, which allows them to do their job more
efficiently. Plus, readdressing a network to resolve these problems can be very difficult, costly,
and time-consuming.
If you are using RFC 1918 address space within your network, then it’s a little easier because
you have such a huge block of address space to allocate from. If you don’t have the luxury of
a large address space and you need to allocate from a smaller block, then you need to plan very
well. You also need to leave plenty of address space for growth.
We will be using the 10.0.0.0/8 IP address space from RFC 1918 in our examples of a large
national network. Figure 11.4 illustrates the sample network that we will be allocating
addresses for.
You can see that there are four major regions at the core of the network and that each region
has two core routers. Each core router is connected to a core router in each other region, which
allows redundant connections among all regions. Each region is going to be allocated
1
⁄
4
of the
total address space available within that region. Instead of seeing thousands of routes for each network
within a given region, the other core routers will see only a single aggregate route from that
region. Table 11.1 shows the breakdown of the address space and which regions it is allocated to.