Subnetting
Subnetting is used to divide a single network into multiple smaller ones. This allows you to use routers to control traffic between networks. When you configure a network, you receive a range of IP addresses from your ISP or you use internal addressing that are not routable. If you have multiple physical locations or a large single location, you need to subnet the address range you are using for easier network management.
IP addresses are viewed using dotted decimal notation (i.e. 10.10.2.238 reads ten dot ten dot two dot two-three-eight), but in order to truly understand subnetting you need to be about to use binary notation. Why? Because when data is routed from one network to another it is routed using binary numbers, not decimal numbers.
Why Subnet?
A single network is fine for a small company with a limited number of computers in a limited location. However, if a company has many larger networks spread among a number of geographical locations, subnetting is going to be necessary. Using subnetting splits one large network into a number smaller, easier to manage networks. Subnetting can:
Reduce collisions
o As you (hopefully) recall, Ethernet networks use CSMA/CD as an access method. If two computers attempt to transmit data at the same time, a collision occurs. The more computers on a network subnet, the more collisions. Collisions slow down communication within in a network. When a collision occurs each computer waits a random period of time before attempting to send data again.
o On a very large network the actual throughput may be only 30% to 40% of capacity because of loss due to collisions. In other words, a 10 Mbps network might only run at 3-4 Mbps due to collision activity.
o Subnetting a network reduces network traffic by isolating traffic to specific network segments, reducing collisions.
Limit broadcast messages which take up bandwidth
o Broadcast messages are generated by a number of network services. A broadcast message is received by all computers on a network. For example, on a DHCP network when a computer logs in it sends broadcast messages out looking for a DHCP server. When there are few computers on a network, this is no big deal. As the network grows, though, broadcast messages may slow down the network substantially.
o Broadcast messages are addressed to the IP address 255.255.255.255. This address is processed by every single computer on the network.
o Broadcast messages are not sent across subnets because routers do not forward packets sent to the 255.255.255.255 address. Therefore, subnetting a network will isolate broadcast messages within a subnet, reducing broadcast traffic.
Control traffic and bandwidth use
o When a network is subnetted routers are placed between subnets for passing traffic from one network to another. Subnetting a network allows you to control what traffic goes where.
Binary and TCP/IP
IP addresses are expressed in dotted decimal notation. It’s easier for people to use and the software can convert it to binary for the network. However, networks must read all IP addresses in binary. Therefore, you need to be able to convert an IP address from dotted decimal notation to binary notation for custom subnetting. If you need to review how to convert decimal numbers to binary and binary to decimal, please access the document at http://teechur.msvl.k12.wa.us/w2svr/unit15/binary.htm.
Binary and Subnet
Masks
When we think of subnet masks, we think of them in dotted decimal notation because that’s how we enter them into the TCP/IP properties. However if you convert a ddn subnet mask into binary you’ll see how the masking actually works.
For example the subnet mask 255.255.254.0 can be re-written in binary to be:
11111111.11111111.11111110.00000000
Look at the following subnet masks
|
255.255.255.0 |
11111111.11111111.11111111.00000000 |
|
255.255.252.0 |
11111111.11111111.11111100.00000000 |
|
255.255.240.0 |
11111111.11111111.11110000.00000000 |
Notice how the 1s and the 0s are all grouped together? This is where the masking takes place. The 1s mask off, or identify, the network part of the IP address. The 0s identify the host (individual computer or node) part of the IP address.
So to learn what part of an IP address is the host ID and what part is a network ID you do the following:
- Convert the IP address into binary.
- Convert the subnet mask into binary.
- Line them up. The numbers in the IP address corresponding with the 1s are the network Ids.
- The numbers in the IP address corresponding with the 0s are the host Ids.
See the example below.
|
|
Decimal |
Binary |
|
IP Address |
192.168.5.20 |
11000000.10101000.00000101.00010100 |
|
Subnet Mask |
255.255.255.0 |
11111111.11111111.11111111.00000000 |
|
Network ID |
192.168.5.0 |
11000000.10101000.00000101.00000000 |
|
Host ID |
0.0.0.20 |
00000000.00000000.00000000.00010100 |
The process your computers uses to find the network ID is called ANDing. This is a mathematical operation that compares two binary digits and gives a result of 1 or 0. If both binary digits being compared have a value of 1, the result is 1. If one digit is 1 and the other 0, or both are 0 then the result is 0.
Recap
1 and 1 = 1
1 and 0 = 0
0 and 0 = 0
Practice ANDing
Find the network ID of several IP addresses based on a given subnet mask
|
|
Decimal |
Binary |
|
IP Address |
130.179.16.67 |
|
|
Subnet Mask |
255.255.255.0 |
|
|
Network ID |
|
|
|
Host ID |
|
|
|
|
Decimal |
Binary |
|
IP Address |
10.10.238.2 |
|
|
Subnet Mask |
255.224.0.0 |
|
|
Network ID |
|
|
|
Host ID |
|
|
|
|
Decimal |
Binary |
|
IP Address |
130.179.16.67 |
|
|
Subnet Mask |
255.255.0.0 |
|
|
Network ID |
|
|
|
Host ID |
|
|
Address Classes Review
There are 5
different address classes. You can determine which class any IP address is in
by examining the first 4 bits of the IP address.
Class A addresses begin with 0xxx, or 1 to 126 decimal.
Class B addresses begin with 10xx, or 128 to 191 decimal.
Class C addresses begin with 110x, or 192 to 223 decimal.
Class D addresses begin with 1110, or 224 to 239 decimal.
Class E addresses begin with 1111, or 240 to 254 decimal.
Addresses beginning
with 01111111, or 127 decimal, are reserved for loopback and for internal
testing on a local machine. [You can test this: you should always be able to
ping 127.0.0.1, which points to yourself] Class D addresses are reserved for
multicasting. Class E addresses are reserved for future use. They should not be
used for host addresses.
Now we can see how
the Class determines, by default, which part of the IP address belongs to the
network (N) and which part belongs to the node (n).
Class A -- NNNNNNNN.nnnnnnnn.nnnnnnn.nnnnnnn
Class B -- NNNNNNNN.NNNNNNNN.nnnnnnnn.nnnnnnnn
Class C -- NNNNNNNN.NNNNNNNN.NNNNNNNN.nnnnnnnn
In the example,
140.179.220.200 is a Class B address so by default the Network part of the
address (also known as the Network Address) is defined by the first two
octets (140.179.x.x) and the node part is defined by the last 2 octets
(x.x.220.200).
Subnetting a Network
To subnet a network,
you take some bits from the host ID and give them to the network ID. As the
manager of a network, you have the power and freedom to do this. (Feel the
power. FEEL THE POWER!)
A Class B address is
very large and generally needs to be subnetted to handle routing between
different physical locations. To keep subnetting simple, bits are often taken
from the host ID in a group of eight. This keeps the entire octet in intact.
The table below shows an example of subnetting a Class B address by taking
eight bits from the host ID and giving them to the network ID. Using an entire
octet for subnetting gives you 256 possible subnets. Traditionally the subnets
using all 0s or all 1s are discarded, leaving 254 usable subnets.
|
|
Decimal |
Binary |
|
Original network |
135.25.0.0 |
10000111.00011001.00000000.00000000 |
|
Original subnet mask |
255.255.0.0 |
11111111.11111111.00000000.00000000 |
|
New subnet mask |
255.255.255.0 |
11111111.11111111.11111111.00000000 |
|
Subnet 1 |
135.25.0.0 |
10000111.00011001.00000000.00000000 |
|
Subnet 2 |
135.25.1.0 |
10000111.00011001.00000001.00000000 |
|
Subnet 3 |
135.25.2.0 |
10000111.00011001.00000010.00000000 |
|
Subnet 4 |
135.25.3.0 |
10000111.00011001.00000011.00000000 |
|
Subnet 5 |
135.25.4.0 |
10000111.00011001.00000100.00000000 |
|
Subnet 6 |
135.25.5.0 |
10000111.00011001.00000101.00000000 |
|
Subnet 7 |
135.25.6.0 |
10000111.00011001.00000110.00000000 |
|
Subnet 256 |
135.25.255.0 |
10000111.00011001.11111111.00000000 |
Doing this instead
of having one great big network, you can subnet into a bunch of smaller
networks. The problem is, we rarely need 256 extra networks, and each of those
networks has a very limited number of nodes available. Not a lot of use for 256
(or 254 assuming you don’t use all 1s or all 0s) teensy little networks, is
there?
So instead you’ll
generally “borrow” only the number of bits you need. This is called custom
subnetting. In this case you
use only SOME of the octet to get the number of networks you need.
Example: Your
company leased the class B address 173.25.0.0. This is way too large for one
network. You need to split it into ten networks.
To calculate the numbers
of subnets that can be created by a certain number of bits use the formula 2n-2.
In this formula n represents the number of bits and -2 is used because we
generally do not use all 1s or all 0s in our subnets, so we subtract 2
un-useable networks. So let’s try the formula!
2n-2>=10
(In other words, 2n-2 has to be greater
than or equal to 10)
2n>=10+2
2n>=12
26=64
6 is too big so
let’s try 5
2n>=10+2
2n>=12
25-=32
since 2x2x2=8 23 would be too small, but 24 is just
right.
To keep from having
to use a calculator, just use this table. Looking at the table we can easily
see that 5=32 which, even when we subtract 2, is larger than or equal to 10.
|
x |
2x |
2x in
Decimal |
|
0 |
20 |
1 |
|
1 |
21 |
2 |
|
2 |
22 |
4 |
|
3 |
23 |
8 |
|
4 |
24 |
16 |
|
5 |
25 |
32 |
|
6 |
26 |
64 |
|
7 |
27 |
128 |
|
8 |
28 |
256 |
|
|
Decimal |
Binary |
|
Original network |
173.25.0.0 |
10101101.00011001.00000000.00000000 |
|
Original subnet mask |
255.255.0.0 |
11111111.11111111.00000000.00000000 |
|
In order to create ten smaller
networks we’re going to borrow 4 bits from the host ID. |
||
|
New subnet mask |
255.255.240.0 |
11111111.11111111.11111000.00000000 (convert the middle octet to binary) |
|
Subnet 1 |
173.25.0.0 |
10101101.00011001.00000000.00000000 |
|
Subnet 2 |
173.25.8.0 |
10101101.00011001.00001000.00000000 |
|
Why do we put 8 in the third
octet? Because starting with the end of that octet we find where the first 1
from the right sits in place value. It sits in the 8s place, so we use that
number for our network address for our first subnet. We then increase that octet
by 8 for each subsequent network. |
||
|
Subnet 3 |
173.25.16.0 |
10101101.00011001.00010000.00000000 |
|
Subnet 4 |
173.25.24.0 |
10101101.00011001.00011000.00000000 |
|
Subnet 5 |
173.25.32.0 |
10101101.00011001.00100000.00000000 |
|
Subnet 6 |
173.25.40.0 |
10101101.00011001.00101000.00000000 |
|
Subnet 7 |
173.25.48.0 |
10101101.00011001.00110000.00000000 |
|
Subnet 8 |
173.25.56.0 |
10101101.00011001.00111000.00000000 |
|
Subnet 9 |
173.25.64.0 |
10101101.00011001.01000000.00000000 |
|
Subnet 10 |
173.25.72.0 |
10101101.00011001.01001000.00000000 |
|
Subnet 11 |
173.25.80.0 |
10101101.00011001.01010000.00000000 |
|
Subnet 12 |
173.25.88.0 |
10101101.00011001.01011000.00000000 |
|
Subnet 13 |
173.25.96.0 |
10101101.00011001.01100000.00000000 |
Custom Subnetting Practice
Given the class B
address of 172.20.0.0 determine the new subnet mask at the new subnet addresses
for 10 subnets. This will subnet a single large network into 10 smaller
networks.
|
|
Decimal |
Binary |
|
Original network |
173.20.0.0 |
10101100.00010100.00000000.00000000 |
|
Original subnet mask |
255.255.0.0 |
11111111.11111111.00000000.00000000 |
|
New subnet mask |
255.255.240.0 |
|
|
Subnet 1 |
|
|
|
Subnet 2 |
|
|
|
Subnet 3 |
|
|
|
Subnet 4 |
|
|
|
Subnet 5 |
|
|
|
Subnet 6 |
|
|
|
Subnet 7 |
|
|
|
Subnet 8 |
|
|
|
Subnet 9 |
|
|
|
Subnet 10 |
|
|
Subnet Hosts
Hosts are computers
or other objects (such as printers) that need IP addresses on a network. Once
you’ve determined your network addresses you have to know what IP addresses you
can assign to your hosts. In our example above we split our 173.25.0.0 network
into 10 separate networks. We know the network Ids, so now we need the IP
address range we can use on each network.
|
|
Decimal |
IP |
Binary |
|
Original network |
173.25.0.0 |
|
10101101.00011001.00000000.00000000 |
|
Original subnet mask |
255.255.0.0 |
|
11111111.11111111.00000000.00000000 |
|
New subnet mask |
255.255.240.0 |
|
11111111.11111111.11111000.00000000 (convert the middle octet to binary) |
|
Subnet 1 |
173.25.0.0 |
173.25.0.1-173.25.7.254 |
10101101.00011001.00000000.00000000 |
|
See how we did that? We started
with adding 1 to the end of the network ID and then took 1 away from the
third octet of the starting address for the next subnet. If we convert the 2nd
IP address into binary we see it comes out to be 10101101.00011001.00000111.11111110.
The IP 173.25.15.255 would result in all 1s on the host ID. That address is
the broadcast address for this subnet, so it can’t be assigned to any
computers. See how that works? Pretty
nifty, huh? |
|||
|
Subnet 2 |
173.25.8.0 |
173.25.8.1-173.25.15.254 |
10101101.00011001.00001000.00000000 |
|
Subnet 3 |
173.25.16.0 |
173.25.16.1-173.25.23.254 |
10101101.00011001.00010000.00000000 |
|
You do the rest! |
|||
|
Subnet 4 |
173.25.24.0 |
|
10101101.00011001.00011000.00000000 |
|
Subnet 5 |
173.25.32.0 |
|
10101101.00011001.00100000.00000000 |
|
Subnet 6 |
173.25.40.0 |
|
10101101.00011001.00101000.00000000 |
|
Subnet 7 |
173.25.48.0 |
|
10101101.00011001.00110000.00000000 |
|
Subnet 8 |
173.25.56.0 |
|
10101101.00011001.00111000.00000000 |
|
Subnet 9 |
173.25.64.0 |
|
10101101.00011001.01000000.00000000 |
|
Subnet 10 |
173.25.72.0 |
|
10101101.00011001.01001000.00000000 |
|
Subnet 11 |
173.25.80.0 |
|
10101101.00011001.01010000.00000000 |
|
Subnet 12 |
173.25.88.0 |
|
10101101.00011001.01011000.00000000 |
|
Subnet 13 |
173.25.96.0 |
|
10101101.00011001.01100000.00000000 |
Finding Valid Hosts
Once you have
calculated your network Ids for your subnets you want to find out the valid
hosts. For the subnets you found in the custom subnetting practice, find
starting IP, ending IP, and the broadcast address for each subnet.
|
|
Decimal |
Binary |
|
|
Original network |
173.20.0.0 |
10101100.00010100.00000000.00000000 |
|
|
Original subnet mask |
255.255.0.0 |
11111111.11111111.00000000.00000000 |
|
|
New subnet mask |
|
|
|
|
Subnet |
Starting IP |
Ending IP |
Broadcast IP |
|
Subnet 1 |
|
|
|
|
Subnet 2 |
|
|
|
|
Subnet 3 |
|
|
|
|
Subnet 4 |
|
|
|
|
Subnet 5 |
|
|
|
|
Subnet 6 |
|
|
|
|
Subnet 7 |
|
|
|
|
Subnet 8 |
|
|
|
|
Subnet 9 |
|
|
|
|
Subnet 10 |
|
|
|
How do I Know How Many Hosts on a Subnet
You can use your old
friend, the 2n-2= formula except in a slightly way.
- Count the number of host bits (the
number of 0s) in the subnet mask. That will be n.
- Now plug that into the formula: 2n-2=number
of hosts per network.
|
Host Bits |
Formula |
Usable Hosts |
|
6 |
26-2 |
64-2=62 |
|
8 |
28-2 |
256-2=254 |
|
10 |
|
|
|
12 |
|
|
You fill in the
rest!
Supernetting
Supernetting is the
opposite of subnetting. In subnetting you take one large network and create
smaller ones. In supernetting you take a number of smaller networks and create
one (or fewer) large networks. Instead of borrowing bits from the host ID, you
borrow bits from the network ID.
Why would you do
this? Let’s say your company has a large network and purchases or leases two
class C IP address ranges. They don’t want two networks, they want only one.
They can use supernetting to create one large network out of the two small
ones.
Another benefit of
supernetting is reducing routing complexity. You have to have a router between
every subnet on a network. On an old 10 Mbps you might have subnetted into
multiple networks to reduce collisions. If you update your hardware to 100 Mbps
switches, NICs, and wiring, you can buy fewer routers and combine class C
network Ids to create fewer larger networks, thus reducing complexity and
saving the company money! Sweet!
Example:
|
|
Decimal |
Binary |
|
Original network 1 |
192.168.10.0 |
11000000.10101000.00001010.00000000 |
|
Original network 2 |
192.168.11.0 |
11000000.10101000.00001011.00000000 |
|
Original subnet mask |
255.255.255.0 |
11111111.11111111.11111111.00000000 |
|
Supernetted network |
192.168.10.0 |
11000000.10101000.00001010.00000000 |
|
New subnet mask |
255.255.254.0 |
11111111.11111111.11111110.00000000 |
|
First host |
192.168.10.1 |
11000000.10101000.00001010.00000001 |
|
Last host |
192.168.11.254 |
11000000.10101000.00001011.11111110 |
|
Broadcast |
192.168.11.255 |
11000000.10101000.00001011.11111111 |
Supernetting four
Class C networks
|
|
Decimal |
Binary |
|
Original network 1 |
192.168.76.0 |
11000000.10101000.01001100.00000000 |
|
Original network 2 |
192.168.77.0 |
11000000.10101000.01001101.00000000 |
|
Original network 3 |
192.168.78.0 |
11000000.10101000.01001110.00000000 |
|
Original network 4 |
192.168.79.0 |
11000000.10101000.01001110.00000000 |
|
Original subnet mask |
255.255.255.0 |
11111111.11111111.11111111.00000000 |
|
Supernetted network |
192.168.76.0 |
11000000.10101000.01001100.00000000 |
|
New subnet mask |
255.255.252.0 |
11111111.11111111.11111100.00000000 |
|
First host |
192.168.76.1 |
11000000.10101000.01001100.00000001 |
|
Last host |
192.168.79.254 |
11000000.10101000.01001111.11111110 |
|
Broadcast |
192.168.79.255 |
11000000.10101000.01001111.11111110 |
Fill in the following table showing supernetting
of three networks. Borrow 2 bits from the network ID.
|
|
Decimal |
Binary |
|
Original network 1 |
192.168.10.0 |
11000000.10101000.00001010.00000000 |
|
Original network 2 |
192.168.11.0 |
11000000.10101000.00001011.00000000 |
|
Original network 3 |
192.168.12.0 |
11000000.10101000.01001110.00000000 |
|
Original subnet mask |
255.255.255.0 |
|
|
Supernetted network |
192.168.10.0 |
11000000.10101000.00001010.00000000 |
|
New subnet mask |
255.255.252.0 |
|
|
First host |
192.168.10.1 |
|
|
Last host |
192.168.12.254 |
|
|
Broadcast |
192.168.12.255 |
|