Section Two $$ Routing Tables

You may be wondering how a host or router determines where to transmit an IPv$4$ packet, especially because any particular network can contain dozens, hundreds, or thousands of hosts, and the answer lies in the routing table. A routing table is a set of rules used by a host to formulaically determine the next$$hop destination of an IPv$4$ packet. In particular, each host uses its routing table to determine which of its network adapters it will use for the transmission, and which host $($technically$,$ which network adapter$)$ will be the next$$hop recipient. The term $$table$$ in $$routing table$$ is used primarily because the set of rules defined in a routing table is most often displayed in a tabular format, which consists of specific columns information, with one rule per row. Each rule defines the next hop for a particular network as defined by a network address and a subnet mask. Computers and devices are concerned with the rules themselves, and we as human beings view the rules in a tabular format, for understandability. A routing table contains all the information necessary for a host to determine where transmit an IPv$4$ packet.

It is important to understand that routing a packet from a sender to a receiver is distinct from a routing protocol. A routing protocol specifies exactly how routers share route information with each other, so that routers are able to communicate and configure their routing tables. Routing protocols treat each network as a single entity, and therefore exclusively consider networks, but not hosts on the network, when computing routes. Routing, on the other hand, uses the route information defined by a routing protocol, or route information defined through some other means such as manual or DHCP route assignment, in order to move an IPv$4$ packet from a sender to a receiver. A router may settle upon a set of routes by using a routing protocol, then route a copious number of packets using that set of routes. There are many different routing protocols in use today, but they all perform the same function $$ to determine viable and efficient routes for IPv$4$ packets, and to update the routing table on each router to reflect this.

Let us take a look at the structure of a routing table. The significant columns of information in a routing table are the destination network$$s network address, the destination network$$s subnet mask, the network adapter to be used to reach the destination network, the IPv$4$ address of the next hop to reach for the destination network, and a metric that is used to determine which rule is preferable in the event two or more rules can be used. Each of these is explained in turn below.

The Destination Network$$s Network Address

What it Means: The network address is specified in this column so that the sending host knows which network is specified by the rule. Recall that a network can contain many hosts, and so any one rule in a routing table can identify the next$$hop recipient for all hosts in a network.

Example: $35\mathrm{.}26\mathrm{.}1\mathrm{.}0$ is an example of a network address.

Common Title: When routing tables are printed, you will often see this column titled $$Network

Destination$$ or $$Destination$,$ though different operating systems and tools may use a different title.

The Destination Network$$s Subnet Mask

What it Means: The subnet mask determines which of the bits in an IPv$4$ address are used in the network identifier, and which are used in the host identifier. Recall that the network identifier

Page$7$of$12$

uniquely identifies a particular network, and the host identifier uniquely identifies a host on that network.

Example: $255\mathrm{.}255\mathrm{.}252\mathrm{.}0$ is an example of a subnet mask.

How it Works: The sending host collectively uses the destination network$$s address and subnet mask to determine if a particular IPv$4$ address matches a rule. For example, given a destination IPv$4$ address of $192\mathrm{.}168\mathrm{.}1\mathrm{.}0,$ a rule with network address $192\mathrm{.}168\mathrm{.}1\mathrm{.}0$ and subnet mask $255\mathrm{.}255\mathrm{.}255\mathrm{.}0$ matches the destination address, but a rule with network address $192\mathrm{.}168\mathrm{.}5\mathrm{.}0$ and subnet mask $255\mathrm{.}255\mathrm{.}255\mathrm{.}0$ does not match the destination address.

It is important to note that if an IPv$4$ address matches two or more rules, the most specific rule $$ the one with the most number of $1$ bits in the subnet mask $$ is the rule that is IPv$4$ uses to determine the next hop. For example, if one matching rule specifies $10\mathrm{.}0\mathrm{.}0\mathrm{.}0/8$ for its network address and subnet mask, and a second matching rule specifies $10\mathrm{.}2\mathrm{.}8\mathrm{.}0/24,$ the second matching rule wins. You may now legitimately wonder what happens when two or more rules may match the same IPv$4$ address and also have the same number of $1$ bits in the subnet mask. In such a case, another piece of information in the routing table $$ the metric $$ is used to settle the dispute. The metric is discussed in more detail in its own section, so you may refer to that section for more details. Because more than one rule may match any particular IPv$4$ address, it is important that one of the rules still wins so