subnetting

The answer is too long to post here. Here's a pretty good tutorial:

http://compnetworking.about.com/library/weekly/aa043000b.htm

or do a search for subnet tutorial on the web.

Cheers.

RW

Answer 1 is correct in that it can be a complex process. With
that said here are the basics:

In TCP/IP an address is made up of 2 parts. The network and the
host addresses. As an example:

IP Address: 131.107.16.200
Subnet Mask: 255.255.0.0
______________________________

Network ID: 131.107.y.z
Host ID: w.x.16.200

The "MASK" screens out the network ID. Then it can be
determined if that is the local network. If it is it send data to the
host. If not it send data to the configured "gateway" to routed to
the correct network.

The issue is that any given Network ID only allows for a fixed
number of hosts. What if your company is given a Class B
network with a 255.255.0.0 mask? You would have 65,534
hosts. That may be to much braodcast traffic so you could
custom subnet. In other words take to address block given and
break it into smaller subnets by customizing the mask. The each
subnet would be seperated by a router. Broadcast traffic does
not pass through a router and so traffic could be better
managed.

So, in the same example you could change the mask:

IP Address: 131.107.16.200
Subnet Mask: 255.255.240.0
______________________________

Network ID: 131.107.y.z
Subnet ID: w.x.16.z
Host ID: w.x.y.200

Now you would have multiple networks (or subnets) with fewer
allowed host each. Think of it as departmentalizing ip traffic.

Now, Before someone starts correcting things, I must say:

THIS IS A VERY SIMPLE DESCRIPTION.

Search Google for "subnet tutorial" and you will find a lot of info.

razz