You follow the same steps in performing VLSM as you did when performing classical subnetting. Consider Figure 2-1 as you work through an example. Figure 2-1. Sample Network Needing a VLSM Address Plan
A Class C network192.168.100.0/24is assigned. You need to create an IP plan for this network using VLSM. Once again, you cannot use the N bits192.168.100. You can use only the H bits. Therefore, ignore the N bits, because they cannot change! The steps to create an IP plan using VLSM for the network illustrated in Figure 2-1 are as follows:
The remainder of the chapter details what is involved with each step of the process. Step 1 Determine How Many H Bits Will Be Needed to Satisfy the Largest NetworkA is the largest network with 50 hosts. Therefore, you need to know how many H bits will be needed:
You need 6 H bits to satisfy the requirements of Network A. If you need 6 H bits and you started with 8 N bits, you are left with 8 6 = 2 N bits to create subnets:
All subnetting will now have to start at this reference point, to satisfy the requirements of Network A. Step 2 Pick a Subnet for the Largest Network to UseYou have 2 N bits to work with, leaving you with 2N or 22 or 4 subnets to work with:
If you add all zeros to the H bits, you are left with the network numbers for the four subnets:
All of these subnets will have the same subnet mask, just like in classful subnetting. Two borrowed H bits means a subnet mask of:
or
or
The /x notation represents how to show different subnet masks when using VLSM. /8 means that the first 8 bits of the address are network, the remaining 24 bits are H bits. /24 means that the first 24 bits are network, the last 8 are hostthis is either a traditional default Class C address, or a traditional Class A network that has borrowed 16 bits, or even a traditional Class B network that has borrowed 8 bits! Pick one of these subnets to use for Network A. The rest of the networks will have to use the other three subnets. For purposes of this example, pick the .64 network.
Step 3 Pick the Next Largest Network to Work WithNetwork B = 27 hosts Determine the number of H bits needed for this network:
You need 5 H bits to satisfy the requirements of Network B. You started with a pattern of 2 N bits and 6 H bits for Network A. You have to maintain that pattern. Pick one of the remaining /26 networks to work with Network B. For purposes of this example, select the .128/26 network:
But you need only 5 H bits, not 6. Therefore, you are left with:
where:
Because you have this extra bit, you can create two smaller subnets from the original subnet:
Converted to decimal, these subnets are as follows:
You have now subnetted a subnet! This is the basis of VLSM. Each of these sub-subnets will have a new subnet mask. The original subnet mask of /24 was changed into /26 for Network A. You then take one of these /26 networks and break it into two /27 networks:
The mask now equals:
or
or
Pick one of these new sub-subnets for Network B:
Use the remaining sub-subnet for future growth, or you can break it down further if needed. You want to make sure the addresses are not overlapping with each other. So go back to the original table.
You can now break the .128/26 network into two smaller /27 networks and assign Network B.
The remaining networks are still available to be assigned to networks, or subnetted further for better efficiency. Step 4 Pick the Third Largest Network to Work WithNetworks C and Network D = 12 hosts each Determine the number of H bits needed for these networks:
You need 4 H bits to satisfy the requirements of Network C and Network D. You started with a pattern of 2 N bits and 6 H bits for Network A. You have to maintain that pattern. You now have a choice as to where to put these networks. You could go to a different /26 network, or you could go to a /27 network and try to fit them into there. For the purposes of this example, select the other /27 network.160/27:
But you only need 4 H bits, not 5. Therefore you are left with:
where:
Because you have this extra bit, you can create two smaller subnets from the original subnet:
Converted to decimal, these subnets are as follows:
These new sub-subnets will now have new subnet masks. Each sub-subnet now has 4 N bits and 4 H bits, so their new masks will be:
or
or
Pick one of these new sub-subnets for Network C and one for Network D.
You have now used two of the original four subnets to satisfy the requirements of four networks. Now all you need to do is determine the network numbers for the serial links between the routers. Step 5 Determine Network Numbers for Serial LinksSerial links between routers all have the same property in that they only need two addresses in a networkone for each router interface. Determine the number of H bits needed for these networks:
You need 2 H bits to satisfy the requirements of Networks E, F, G, and H. You have two of the original subnets left to work with. For purposes of this example, select the .0/26 network:
But you need only 2 H bits, not 6. Therefore, you are left with:
where:
Because you have 4 N bits, you can create 16 sub-subnets from the original subnet:
You need only four of them. You can hold the rest for future expansion, or recombine them for a new, larger subnet:
These can all be recombined into the following:
Going back to the original table, you now have the following:
Looking at the plan, you can see that no number is used twice. You have now created an IP plan for the network, and have made the plan as efficient as possible, wasting no addresses in the serial links and leaving room for future growth. This is the power of VLSM! |