How is it possible, with millions of host on the internet the world over, to ensure that
each computer has a unique IP address? The secret that each network that wants on the
internet must register with the Internetwork Information Center
(InterNIC). In turn the InterNIC assigns that network a block of IP address that the
administrator can then dole out to each computer ( or in case of Internet Service Provider
to each customer). These blocks come in three classes: A, B, and C.
In a class A network, the InterNIC assigns the first
(that is, the leftmost) 8 bits of the addresses: The first bit is 0, and remaining 7 bits
are an assigned number. Two to the power of 7 is 128, so 128 class A networks are
possible. The dotted decimal versions of these IP addresses begin with the number 0 (that
is, 00000000) through 127 (that is 01111111). However, 0 isn't used and 127 is used for
other purpose, so there are really only 126 possibilities.
|NOTE: NETWORKS IDs AND HOST IDs
| The numbers assigned by the InterNIC are called
network IDs, and the numbers assigned by the network administrator are called host IDs.
For example, consider the following address from a class A network:184.108.40.206. The
network ID is 115 (or it's sometimes written as 220.127.116.11) and the host ID is 123.234.1.
The number 126 might seem small,but consider that the remaining
24 addresses bits are available for the network to assign locally. In each quad, you have
254 possible numbers (0 and 255 aren't used) so you have 254X 254X254 possible
addresses to assign, which comes out to a little more than 16 million. In other words, you
need to have large system to rate class A network.( If you have such a system, don't
bother partitioning the InterNIC for a block of IP addresses, because all the
classes A networks were snapped up long ago by behemoths such as IBM)
Here is the layout of the IP addresses used by class a networks.
In class B networks, the InterNIC assigns the first 16 bits of
the address: The first two bits are 10, and the remaining is 14 bits are assigned number.
This allows for total 16,384( 2 to the power of 14) class B networks, all of which have a
first quad dotted decimal value between 128 ( that is, 10000000) and 91 ( that is,
10111111). Note that, as with class A networks, all the possible class B numbers have been
Again, the network administrator can dole out the
remaining 16 bits to the network hosts. Given 254 possible values in each of the two
quads, that produces 64,516 possible IP addresses. Here is the layout of class B network
In a class C network, the InterNIC
assigns the first 24 bits of the address: The first three bits are 110, and the remaining
21 bits are an assigned number. So the total number of class C networks is 2,097,152 ( 2
to the power of 21) and 223 ( that is, 11011111).
This leaves only the remaining 8 bits
in the fourth quad for network administrator to assign addresses to local computers.
Again, 0 and 255 aren't used, so class C has a total of 254 possible IP addresses. Here is
the layout of class C network IP address:
|NOTE: WHAT HAPPENED TO THE RESET OF THE
| Because the first quad of an IP address is 8 bits, the range of the
possible values should be between 0 and 255, but class A, B, and C networks usurp only 0
through 223. What happened to 224 through 255 ? Well, the values between 224 and 239 are
used for special multicast protocols ( these are D class addresses), and the values
between 240 and 255 are used for experimental purposes ( these are the E class addresses).
|NOTE: CLASS C
ADDRESSES ARE GOING FAST!
| We mentioned that the addresses blocks for class
A and class B networks are long gone, but with more than 2 million class C blocks
available, there's plenty to go around, right? Wrong! These blocks are being gobbled up
quickly, and it's predicted that the InterNIC will run out before too long .
So, We have a serious problem here.What will
be the solution for the running out IP addresses? Try to answer this question!
If you know the answer, please contact us to tell
us about it.