It basically just means that he doesn't know what he's talking about.
- How does subnetting work?
first you must understand what an IP address is, an IP address (eg. 10.0.0.1 is a decimal representation of four 8 bit binary numbers. in the example of 10.0.0.1 the binary is
00001010 . 00000000 . 00000000 . 00000001
a subnet is noted in the same way, (four decimal numbers represent four binary octets).
for example 255.0.0.0 (class A networks subnet)
is
11111111 . 00000000 . 00000000 . 00000000
these two numbers work together to show the address space on a network.
the way this works is that the numbers undergo a AND opperation to determine the complete address space.
(and operation means that result is only true eg 1 & 0=0 1 & 1=1)
00001010 . 00000000 . 00000000 . 00000001
11111111 . 00000000 . 00000000 . 00000000 (bitwise AND opperation)
---------------------------------------------
00001010 . 00000000 . 00000000 . 00000000
10.0.0.0 - 10.255.255.255 is the address space as any address inside of that range gives the same resultant with the bitwise AND opperation
another example
172.16.0.5
with the subnet 255.255.0.0
IP = 10101100 . 00010000 . 00000000 . 00000101
SUB 11111111 . 11111111 . 00000000 . 00000000
result of bitwise and
10101100 . 00010000 . 00000000 . 00000000
another address in the same range
172.16.254.1 (with same subnet)
10101100 . 00010000 . 11111110 . 00000001 (ip)
11111111 . 11111111 . 00000000 . 00000000 (subnet)
10101100 . 00010000 . 00000000 . 00000000 (bitwise and result)
the bitwise and result is in the same, so you know it's in the same subnet.
however, 192.17.0.1 / 255.255.0.0 looks like this
10101100 . 00010001 . 00000000 . 00000001 (ip)
11111111 . 11111111 . 00000000 . 00000000 (subnet)
10101100 . 00010001 . 00000000 . 00000000 (resultant)
here the resultant is different so you know it's not in the same subnet
of course subnets needn't just stay as 255.0.0.0 (class A) 255.255.0.0 (class B) or 255.255.255.0 (Class C).
you can use other subnets to divide ranges
for example 16.8.2.4 with subnet 255.255.255.254
00010000 . 00001000 . 00000010 . 00000100 (IP)
11111111 . 11111111 . 11111111 . 11111110 (subnet)
00010000 . 00001000 . 00000010 . 00000100 (resultant)
it's on the same address as 16.8.2.5
00010000 . 00001000 . 00000010 . 00000101 (IP)
11111111 . 11111111 . 11111111 . 11111110 (subnet)
00010000 . 00001000 . 00000010 . 00000100 (resultant)
but not as 16.8.2.6
00010000 . 00001000 . 00000010 . 00000110 (IP)
11111111 . 11111111 . 11111111 . 11111110 (subnet)
00010000 . 00001000 . 00000010 . 00000110 (resultant)
or 16.8.2.3
00010000 . 00001000 . 00000010 . 00000011 (IP)
11111111 . 11111111 . 11111111 . 11111110 (subnet)
00010000 . 00001000 . 00000010 . 00000010 (resultant)
so you see that the IP subnet 255.255.255.254 allows for just 2 addresses in the network.
- What is the subnet mask and how does it provide information regarding the associated IP address?
it's kind of described above, the subnet mask is just a collection of octets that are used in a logical function to show what addresses are in a range.
Supernetting is when instead of dividing the host by using host bits to create another network you borrow network bits.
consider a class B network
172.16.0.1 subnet 255.255.0.0
as discussed above you can divide this network easily into multiple ranges
172.16.0.1 subnet 255.255.255.0 divides the 65,000 possibly address ranges into smaller possible chunks.
so with this mask you can have
172.16.0.1
on a seperate network to 172.16.1.1
(with mask 255.255.255.0)
with supernetting you use network bits in the mask rather than host bit to extend the network. using a different mask
172.16.0.1 (255.255.0.0) gives range 172.16.0.1 -> 172.16.255.254 (65,000 addresses)
172.16.0.1 (255.254.0.0) gives range 172.16.0.1 -> 172.17.255.254 (130,000 addresses).
supernetting allows you to conjoin networks that are separate address spaces to what you'd normally subnet with.