Networking Technologies

Chapter 6: Creating Supernets

 

Objectives:

Chapter 6 teaches you about supernets. The objectives important to this chapter are on page 6-1:

1. Describe the purpose of supernets.
2. Identify the IP address criteria used for supernets.
3. Describe routing considerations when creating supernets.

Concepts:

Supernets exist because of the growth of networks. The concept was invented to allow the use of larger numbers of hosts on a network than are allowed by standard IPv4 addressing rules. Page 6-2 states that supernets were invented to "extend the 32-bit IP address lifetime". This is also true.

As noted in other chapters, all available class A and class B licenses network addresses have been assigned. If you get a class C address, your network can only have about 254 hosts on it. (Theoretically, 256, but you can't use some addresses.) This is not enough for some users. (Other reasons for the creation of supernets appear in the bullets on page 6-2.)

Supernets combine multiple, sequential class C network addresses into single logical networks. The networks used will fall into the range 192 through 223. This is the opposite of subnetting, so it should be logical that you begin the opposite way: you borrow bits from the network portion of the IP address in order to use them as host bits. Every bit you borrow for use as a host bit doubles the theoretical number of hosts for your network. The phrase used to describe the combining of addresses into a supernet is address aggregation.

Supernets have rules that are different from subnets.

• Supernets are created from class C network addresses.
• Network addresses used must be consecutive.
• In each pair of network addresses, the third octet of the first address must be an even number. Another way of saying this is that the binary version of the third octet of this address must end with a 0. For example, if you own the network addresses 201.41.18.0 and 201.41.19.0, you can make a supernet. If you own the addresses 201.41.19.0 and 201.41.20.0, you cannot make a supernet. (I will leave the binary conversion to you as an exercise.)
• Supernet masks are subnet masks that use fewer 1 bits than classful masks. A standard class C subnet mask is 255.255.255.0. If you borrow one bit for supernetting, your mask is 255.255.254.0, and your network can have 512 (theoretical) hosts instead of 256. This mask uses 23 bits for the network identifier, instead of the standard 24 bits for class C. Routers will be informed of this shorter mask, and they must support it. (The notation hinted at in the text is a network address, followed by a slash, followed by the number of network bits in the mask. More on this on page 6-6, in the discussion of CIDR.)

Page 6-4 shows a relatively simple example of combining two class C networks into one, using the network addresses 201.41.18.0 and 201.41.19.0. As noted above, they qualify, because they are consecutively numbered, and the one with the lower number has an even number for its third octet. One bit is borrowed from the network bits, giving the new mask 23 network bits.

Page 6-5 shows a chart concerning the aggregation of more than two network addresses. Allow me to rephrase it here:

Combining Multiple Class C Network Addresses

Bits borrowed from network bits:	7	6	5	4	3	2	1	0
Subnet Mask:	128	192	224	240	248	252	254	255
Number of Consecutive Network Addresses Needed:	128	64	32	16	8	4	2	0
Number of Hosts Possible:	16384- 32768	8192- 16383	4096- 8190	2048- 4095	1024- 2047	512- 1023	256- 511	<256

You can solve a story problem with this chart. Suppose you know the number of hosts you would like to have on a network. Find the number in one of the ranges in the bottom row. The cells above it will tell you how many network addresses you need, what the subnet mask will be and how many bits you will borrow.

Another concept from page 6-4 is the broadcast address. The broadcast address for your supernet is obtained by setting all the host bits to 1s.

The last concept in this chapter is Classless Inter-Domain Routing (CIDR). CIDR is a router standard that allows the use of classless subnet masks. In the CIDR standard, IP addresses are followed by a slash and the decimal number of bits used in the network portion of the subnet mask, as in the example 220.12.78.0/23. If your routers do not support CIDR, you cannot have supernets.