How does subnetting solve the problem of IPv4 address shortage?

Question

Reading articles on the web, they all say that "subnetting solves the problem of IPv4 address shortage by dividing a large physical network into multiple smaller networks". I do not understand how this works.

If you are dividing a large physical network into multiple sub-networks, doesn't total the number of hosts/addresses remain the same? (product of number of networks and number of hosts would remain the same)

Also, if you are borrowing 1 or more bits from the host portion of the IP address to create a subnet mask, aren't you further reducing the number of hosts that can be connected to a network?

Answer 1

Consider a network that needs 300 addresses. A class C block is too small, so they would get a class B, even though they're not using over 65,000 addresses. If you have 10 networks, that's a lot of wasted addresses. You can see that one would quickly run out of address space.

By subnetting, you can allocate addresses to networks in smaller amount. So in this case, each network would get a /23 allocation, only wasting about 200 addresses per network.

Answer 2

Likely that should say subnetting of classful networks solves the problem of IPv4 address shortage. Classful networking is long dead. It died in 1993 with the introduction of CIDR. Please leave it dead and buried.

Obsolete classful networking wasted vast quantities of IP addresses. Subnetting reduced the waste, so it became the norm.

Of course, CIDR and private addressing only postponed running out of v4 addresses, they never solved the problem.