Suppose that Company ABCD owns Class B address 130.15 (i.e., this is the network address) and needs:
- 4 LANs with 300-500 systems (Large)
- 20 LANs with 50-100 systems (Medium)
- 1000 LANs with 10-20 systems (Small)
1 -Describe how you would determine the subnet masks for each of the above subnetworks. (Suggestion: use the number of hosts, instead of number of LANs, required by each LAN as the basis to determine the subnet mask)
2-What are the address ranges of the hosts in each subnet? (For example, hosts in the first Large subnet can range from 18.104.22.168 to 22.214.171.124)
3-When “designing” your subnets and the address range do not forget to leave a portion of unused addresses for each subnet to enable future additions or changes.
4-Do not use all ‘1s’ for subnet or host number. Do not use all ‘0s’ for host numbers. You may use all `0s’ for subnet number.
5- In the case of CIDR, answer the following questions:
5.1 - Would you still need 16 bits (i.e., a class for the network part?
5.2 - How would the address be distributed? Use the CIDR notation.
The answer to sunny's question was a series of emails, around 6 or so and got very messy due to the fact you wouldnt ever split up such a large network using one class B address because your likely to have remote lans over wans, and possibly a man (metropolitan area network).
The question is purely theory based and I consider invalid for implementation.
If you like, you can email Sunny and he can give you the information, alternative, leave a msg here and I'll see what I can gather from the emails and send you a few lines of information to help you see how it was tackeled.
I would have posted it here, but its really way to long.