Welcome to part 1 in a 3 part series of articles about routers and routing. In part 1 we will cover how routers works, part 2 talks about the routing table and part 3 covers IP routing.
So, how does a router work? Well, before we get into that you should know the basics of what the IP address is, subnet mask and default gateway before continuing. If you don’t have a read of networking basics first.
How Routers Work
Routing is the process of forwarding IP packets from one network to another. A router is a device that joins networks together and routes traffic between them. A router will have at least two network cards (NICs), one physically connected to one network and the other physically connected to another network. A router can connect any number of networks together providing it has a dedicated NIC for each network.
Having two networks with one router is extremely simple and easy to configure. Once we move onto bigger networks with multiple routers however, things get a little more complicated. For example, if we have just 3 networks we can connect them in two different ways; one way is to daisy chain them together using only 2 routers. The other way would be to connect them directly to each other using 3 routers.
In configuration 1 if router A or B went down no machines from subnet A would be able to communicate with subnet C as there is only one path to it. But, if we add an additional router between subnets A and C we now have two routes to Subnet C which makes our network more efficient.
Routers don’t just route traffic to other networks, they learn which are the fastest routes and use them first. Using configuration 2 from above subnet A has two routes to subnet C; one directly through Router C (1 hop) and one through Router A then B (2 hops). When sending traffic from subnet A to C we obviously want it to try going directly through Router C first. This is the quickest and most efficient route but how does the router know this? It knows by using something called a metric value. Each route the router knows of has a metric value assigned to it. A metric value is basically a preference number. If there are two routes to the same destination then the one with the lowest metric is assumed to be the most efficient. Routers will always use this route first until it fails, in which case it will then try the route with the next lowest metric and so on. Routers store all this information in a routing table which is covered in Part 2 – The Routing Table