MCSA: IPV6 Subnetting Part 2


Uploaded by itfreetraining on 22.07.2012

Transcript:
Welcome to the second video on subnetting and routing for IPv6 asked for by the viewers.
In the last video I looked at how routing and subnetting worked in IPv6. In this video
I will look at how you can subnet global IPv6 addresses.
The process of subnetting in IPv6 I think is a lot easier than in IPv4 but at first
the formulas involved may look a bit difficult and confusing. Before I get started with the
mathematics involved, I will first look at an example of an online tool that is available
that makes the process very easy. Let’s say that you are given the following
IPv6 address. Remember that the subnet ID is always 16 bits. 5 bits in the subnet ID
have already been used since the network prefix is 53. This leaves us 11 bits. In this case
I will divide the subnet ID into two more levels.
To do this, I will use the online IPv6 subnet tools provided on www.subnetonline.com. There
are a lot of tools on this web site that take the hard work out of the subnetting and other
IP related administration. In this case, I will select subnet calculators and select
the option IPv6 subnet calculator. This tool will work out the ranges for each
level when given the number of subnets you require. It is important to factor in some
future expansion when putting in these figures. To use the tool, enter in the address and
select the network prefix. For the first level I will enter in 5 subnets and press add level.
Notice that the number of bits used is shown. When allocating levels like this, consider
if you want to use more or less bits of the subnet ID. Once you start deploying your network
it is a lot harder to change later on if you did not allocate enough bits at each level.
For the next level I will enter in 20 subnets which will require 5 more bits. To calculate
the subnets, press the button, calculate subnets. Once the subnets are calculated, when I expand
down to the first level I can see the 5 subnets each with a network prefix of 56. The point
to notice here is the first subnet is 3000, the second 3100 and 3200 and so on. Each subnet
is being incremented by 100 in hexadecimal. I chose 5 subnets because I know the increment
value is very simple; however, in most cases the difference between each subnet may not
be as simple. If I expand down to the second level, notice the sequence of subnet starts
from 0, then 8, 10, 18, 20 and so forth. The challenge thus when dividing up IPv6 into
subnets is to work out the increment value. Once you have that value you can easily work
out the subnet address range of each subnet. Now Let’s now consider the same network prefix
again, but this time doing the mathematics by hand. The network prefix that we have been
given is 53. This means the first 5 bits are already in use and so cannot be subnetted
by us. In some cases, your company may have already
used some of the bits in the subnet for routing or in some cases this will be the network
prefix given by the ISP. When IPv6 was designed, the idea was the ISP would allocate addresses
with network prefixes of 48 for their customers. However, in the real world this may not happen
as the ISP is free to allocate the address space as they see fit.
This leaves us with 11 bits for our subnet ID. Let’s say for this example we need 30
subnets or at least we think that this is the most we are ever going to expand to. The
next step is to work out how many bits are required to support 30 subnets. To do is,
work out 2 to the power of the number of bits that will give us a value over 30. In this
case, 2 to the power of 5 results in 32. This means 5 bits will support 32 networks, so
I will use 5 bits of the subnet ID for the new network prefix. This makes the network
prefix for the new subnets 58. The next step is to work out the Subnetted
Address Prefix for the new subnets. The first one is simple: it will be the address that
we were given with a different network prefix of 58.
In order to work out the rest of the address ranges, we will need to know what the increment
value will be for each address range. There are a number of different ways to achieve
this. We have already looked at an online tool that can help us, but in the exam if
you get asked a question on this you won’t have access to this site. In your exam you
may have access to the Windows calculator. Some exams allow it and some do not. If you
do have access to the Windows calculator there is a very fast way of working out this value.
If I open Windows calculator from the start menu, the first thing that I need to do is
select programmer mode from the view menu. If you have an older version of the Windows
calculator this option may not be available, but you can get the same result selecting
scientific mode. When we change from the first network range
to the second network range the first binary digit will change from 0 to 1. Working out
what this value is will give us the amount that we need to increment each time.
To do this, first select binary mode. Enter in 1 to represent the second subnet. Now the
value needs to be padded with 0’s depending on which network prefix we are using. In this
case, the first 6 digits of our subnet ID are not part of the prefix so I will add in
6 zeros. Once this is done, select the option hexadecimal. There it is, 40, the value that
needs to be added on to work out each new address range.
Notice that in programmer mode you have a break down in binary of the number. This is
useful when working with binary, particularly when working with IPv4 subnet masks. Now that
I have the increment value, it is a simple matter to add that to each address to work
out the ranges for all the subnets. As you can see here, the sequence of networks goes
3000, 3040, 3080 and seeing it is hexadecimal the next one will be 30C0.
For the sake of completeness, let’s add another level of subnets using 3 bits. In
this case I will use some of the standard formulas for working out network address ranges
used in some literature. You may come across some variations of these formulas depending
which literature you look at. The first formula is f equals m – 48. The
value m is the prefix length of the address that is being subnetted; in this case, it
will be 58. This gives us a value for f of 10. F is the number of bits in the subnet
that are fixed and we can’t use to subnet. The next formula works out the increment value
between different address ranges. This function is i equals 2 to the power of 16 minus f plus
s. The s in the formula is the number of bits you are adding to the prefix, in this case
3. When I plug in all the values this will give
me 2 to the power of 3 which equals 8. Thus the increment value is 8 for the new subnets.
If I expand the first subnet this gives me the following address ranges 3000, 3008, 3010,
3018 and so on. The same technique could be used for all the other address range to work
out the address ranges. This video and the last video cover subnetting
and routing for IPv6. Calculating the increment value of the address is the hardest part,
but if you use an online IPv6 calculator or the Windows calculator I think that you will
find the process is very easy. I personally think that the process is easier than working
with subnet masks in IPv4 since the global IPv6 dedicates 16 bits of the subnet mask.
Remember that the network prefix is only used for routing and thus on the local network,
regardless whether the IP address is subnetted or not, you will always use a network prefix
of 64. This makes subnetting a lot easier as you never need to work with subnet masks
like in IPv4. Well I hope you have enjoyed these 2 videos
requested by the viewers and found them helpful. For more of our completely free Microsoft
certification courses, please see our YouTube channel or web site. Thanks for watching.