Now that you've seen how IP-addresses are made up, you may be wondering how they are used on an Ethernet to address different hosts. After all, the Ethernet protocol identifies hosts by a six-octet number that has absolutely nothing in common with an IP-address, doesn't it?
Right. That's why a mechanism is needed to map IP-addresses onto Ethernet addresses. This is the so-called Address Resolution Protocol, or ARP. In fact, ARP is not confined to Ethernets at all, but is used on other types networks such as ham radio as well. The idea underlying ARP is exactly what most people do when they have to find Mr. X. Ample in a throng of 150-people: they go round, calling out his name, confident that he will respond if he's there.
When ARP wants to find out the Ethernet address corresponding to a given IP-address, it uses a feature of Ethernet known as ``broadcasting,'' where a datagram is addressed to all stations on the network simultaneously. The broadcast datagram sent by ARP contains a query for the IP-address. Each receiving host compares this to its own IP-address, and if it matches, returns an ARP reply to the inquiring host. The inquiring host can now extract the sender's Ethernet address from the reply.
Of course you might wonder how a host may know on which of the zillions of Ethernets throughout the world it is to find the desired host, and why this should even be an Ethernet. These questions all involve what is called routing, namely finding out the physical location of a host in a network. This will be the topic of the following section.
For a moment, let's talk about ARP a little longer. Once a host has discovered an Ethernet address, it stores it in its ARP cache, so that it doesn't have to query for it the next time it wants to send a datagram to the host in question. However, it is unwise to keep this information forever; for instance, the remote host's Ethernet card may be replaced because of technical problems, so the ARP entry becomes invalid. To force another query for the IP-address, entries in the ARP cache are therefore discarded after some time.
Sometimes, it is also necessary to find out the IP-address associated with a given Ethernet address. This happens when a diskless machine wants to boot from a server on the network, which is quite a common situation on local area networks. A diskless client, however, has virtually no information about itself-- except for its Ethernet address! So what it basically does is broadcast a message containing a plea for boot servers to tell it its IP-address. There's another protocol for this, named Reverse Address Resolution Protocol, or RARP. Along with the BOOTP protocol, it serves to define a procedure for bootstrapping diskless clients over the network.
Thu Mar 7 23:22:06 EST 1996