|Subnet Mask||CIDR||Net Bits||Host Bits||# of Nets||Hosts/Net||Total Hosts|
[*] According to classic IP routing rules you are not able to use the
subnets with all zeros or ones in the network portion. However most
modern machines have no trouble actually using the upper and lower
subnets. Steven C. Jensen
[**] Note that you should not use a host address with all zeros as that means this host in many IP implementations (and is considered the broadcast address in some antiquated systems [SunOS-2?]), and neither can you use a host address with all ones, as that's the broadcast address for the subnet. The host address of all zeros, in combination with the network address, are also used to specify the complete subnet address (RFC 1105).
This means that a 31-bit netmask is essentially useless since it leaves only two addresses per net: one for the network number, and the other for the broadcast address. In theory though you could specify a host address as the same as the network address if the IP implementation of the host(s) in question does not map an all-zeros host number to be equivalent to the localhost. I wouldn't advise this though unless you're really strapped for subnets. It's far safer and more general to use /30 nets instead.
[***] An all-one's netmask (i.e. all 32 bits) specifies a host address.
Valid subnets on a network 192.168.100.0/26 (i.e. with a subnet mask of
Subnet 1:- 192.168.100.64 with addresses 192.168.100.65 -> 192.168.100.127
Subnet 2:- 192.168.100.128 with addresses 192.168.100.129 -> 192.168.100.191
You can also see the effect of subnetting your network.
Visit the Class B Subnetting summary table.
This page has been accessed times since 22 Jan. 1997.
What'sa matter? Doncha got a back button?
(How about going back here or here instead!)