CIDR, ARP, DHCP, ICMP
CIDR, therefore, tries to balance the desire to minimize the number of routes that a router needs to know against the need to hand out addresses efficiently. To do this, CIDR helps us to aggregate routes. That is, it lets us use a single entry in a forwarding table to tell us how to reach a lot of different networks.
As noted above it does this by breaking the rigid boundaries between address classes. To understand how this works, consider our hypothetical organization with 16 class C network numbers.
Instead of handing out 16 addresses at random, we can hand out a block of contiguous class C addresses. Suppose we assign the class C network numbers from 192.4.16 through 192.4.31. Observe that the top 20 bits of all the addresses in this range are the same (11000000 00000100 0001). Thus, what we have effectively created is a 20-bit network number—something that is between a class B network number and a class C number in terms of the number of hosts that it can support.
In other words, we get both the high address efficiency of handing out addresses in chunks smaller than a class B network, and a single network prefix that can be used in forwarding tables. Observe that, for this scheme to work, we need to hand out blocks of class C addresses that share a common prefix, which means that each block must contain a number of class C networks that is a power of two.
CIDR requires a new type of notation to represent network numbers, or prefixes as they are known, because the prefixes can be of any length. The convention is to place a /X after the prefix, where X is the prefix length in bits. So, for the example above, the 20-bit prefix for all the networks 192.4.16 through 192.4.31 is represented as 192.4.16/20.
By contrast, if we wanted to represent a single class C network number, which is 24 bits long, we would write it 192.4.16/24. Today, with CIDR being the norm, it is more common to hear people talk about “slash 24” prefixes than class C networks. Note that representing a network address in this way is similar to the h mask, value i approach used in sub netting, as long as masks consist of contiguous bits starting from the most significant bit (which in practice is almost always the case).
Address Translation (ARP)
In the previous section we talked about how to get IP datagrams to the right physical network but glossed over the issue of how to get a datagram to a particular host or router on that network. The main issue is that IP datagrams contain IP addresses, but the physical interface hardware on the host or router to which you want to send the datagram only understands the addressing scheme of that particular network.
Thus, we need to translate the IP address to a link-level address that makes sense on this network (e.g., a 48-bit Ethernet address). We can then encapsulate the IP datagram inside a frame that contains that link-level address and send it either to the ultimate destination or to a router that promises to forward the datagram toward the ultimate destination. One simple way to map an IP address into a physical network address is to encode a host’s physical address in the host part of its IP address.
For example, a host with physical address 00100001 01001001 (which has the decimal value 33 in the upper byte and 81 in the lower byte) might be given the IP address 220.127.116.11. While this solution has been used on some networks, it is limited in that the network’s physical addresses can be no more than 16 bits long in this example; they can be only 8 bits long on a class C network. This clearly will not work for 48-bit Ethernet addresses.
Host Configuration (DHCP)
Ethernet addresses are configured into the network adaptor by the manufacturer, and this process is managed in such a way to ensure that these addresses are globally unique. This is clearly a sufficient condition to ensure that any collection of hosts connected to a single Ethernet (including an extended LAN) will have unique addresses.
Furthermore, uniqueness is all we ask of Ethernet addresses. IP addresses, by contrast, not only must be unique on a given internetwork but also must reflect the structure of the internetwork. As noted above, they contain a network part and a host part, and the network part must be the same for all hosts on the same network.
Thus, it is not possible for the IP address to be configured once into a host when it is manufactured, since that would imply that the manufacturer knew which hosts were going to end up on which networks, and it would mean that a host, once connected to one network, could never move to another.
For this reason, IP addresses need to be reconfigurable. In addition to an IP address, there are some other pieces of information a host needs to have before it can start sending packets. The most notable of these is the address of a default router—the place to which it can send packets whose destination address is not on the same network as the sending host.
Error Reporting (ICMP)
The next issue is how the Internet treats errors. While IP is perfectly willing to drop datagrams when the going gets tough—for example, when a router does not know how to forward the datagram or when one fragment of a datagram fails to arrive at the destination—it does not necessarily fail silently.
IP is always configured with a companion protocol, known as the Internet Control Message Protocol (ICMP), that defines a collection of error messages that are sent back to the source host whenever a router or host is unable to process an IP datagram successfully.
For example, ICMP defines error messages indicating that the destination host is unreachable (perhaps due to a link failure), that the reassembly process failed, that the TTL had reached 0 and that the IP header checksum failed, and so on. ICMP also defines a handful of control messages that a router can send back to a source host. One of the most useful control messages, called an ICMP-Redirect, tells the source host that there is a better route to the destination.
ICMP-Redirects are used in the following situation. Suppose a host is connected to a network that has two routers attached to it, called R1 and R2, where the host uses R1 as its default router. Should R1 ever receive a datagram from the host, where based on its forwarding table it knows that R2 would have been a better choice for a particular destination address, it sends an ICMP Redirect back to the host, instructing it to use R2 for all future datagrams addressed to that destination. The host then adds this new route to its forwarding table.