A subnet is a logical partition of an IP network into multiple, smaller network segments. It is typically used to subdivide large networks into smaller, more efficient subnetworks.
The internet is composed of many networks that are run by many organizations. In turn, each organization's network can be composed of many smaller networks, or subnets. Each subnet allows its connected devices to communicate with each other, and routers are used to communicate between subnets. The size of a subnet depends on the connectivity requirements and the network technology employed. A point-to-point subnet allows two devices to connect, while a data center subnet might be designed to connect many more devices.
Each organization is responsible for determining the number and size of the subnets it creates, within the limits of the address space available for its use. The details of subnet segmentation within an organization remain local to that organization. Network devices in one organization on the internet do not need to know the details of the subnet segmentation in use by other organizations.
Subnetting is the logical segmentation of a network address space into segments that are appropriate for the size of an organization's internal networks, improving address allocation efficiency. It is described in Request for Comments 950 (RFC 950) and is tightly linked to IP addresses, subnet masks, and CIDR notation.
An IP address is divided into two fields: a Network Prefix (also called the Network ID) and a Host ID. The default point that separates the Network Prefix and the Host ID depends on whether the address is a Class A, Class B or Class C address. Figure 1 shows an IPv4 Class B address, 172.16.37.5. Its Network Prefix is 172.16.0.0, and the Host ID is 37.5.
The subnet mechanism uses a portion of the Host ID field to identify individual subnets. Figure 2 shows the third octet of the 172.16.0.0 network being used as a Subnet ID. A subnet mask is used to identify the part of the address that should be used as the Subnet ID. The subnet mask is applied to the full network address using a binary AND operation, resulting in the Subnet ID. In the binary AND operation, only when two bits are both 1, the result is 1 (1 AND 1 = 1).
In Figure 2, the AND of the IP address and the mask produces the Subnet ID. Any remaining address bits identify the Host ID. The subnet in Figure 2 is identified as 172.16.2.0, and the Host ID is 5. In practice, network staff will typically refer to a subnet by just the Subnet ID. It would be common to hear someone say, "Subnet 2 is having a problem today," or, "There is a problem with the dot-two subnet."
The Subnet ID is used by routers to determine the best route between subnetworks. Figure 3 shows the 172.16.0.0 network, with the third octet as the Subnet ID. Four of the 256 possible subnets are shown connected to one router. Each subnet is identified by its Subnet ID, or the subnet address with the Host ID set to .0 -- e.g., 172.16.2.0. The router interfaces are assigned the Host ID of .1 -- e.g, 172.16.2.1. When the router receives a packet addressed to a host on a different subnet than the sender -- host A to host C, for example -- it knows the subnet mask and uses it to determine the Subnet ID of host C. It examines its routing table to find the interface connected to host C's subnet and forwards the packet on that interface.
A subnet itself also may be segmented into smaller subnets, giving organizations the flexibility to create smaller subnets for things like point-to-point links or for subnetworks that support a few devices. The example below uses an 8-bit Subnet ID. The number of bits in the subnet mask depends on the organization's requirements for subnet size and the number of subnets. Other subnet mask lengths are common. While this adds some complexity to network addressing, it significantly improves the efficiency of network address utilization.
A subnet can be delegated to a suborganization, which itself may apply the subnetting process to create additional subnets, as long as sufficient address space is available. Subnetting performed by a delegated organization is hidden from other organizations. As a result, the subnet ID field length and where subnets are assigned can be hidden from the parent (delegating) organization, a key characteristic that allows networks to be scaled up to large sizes.
In modern routing architectures, routing protocols distribute the subnet mask with routes and provide mechanisms to summarize groups of subnets as a single routing table entry. Older routing architectures relied on the default Class A, B, and C IP address classification to determine the mask to use. CIDR notation is used to identify Network Prefix and Mask, where the subnet mask is a number that indicates the number of ones in the Mask (e.g., 172.16.2.0/24). This is also known as Variable-Length Subnet Masking (VLSM) and Classless Inter-Domain Routing (CIDR). Subnets and subnetting are used in both IPv4 and IPv6 networks, based on the same principles.
Continue Reading About subnet (subnetwork)
- Whether you need a subnetting cheat sheet or you're preparing for your CCNA exam, this tip and quiz on subnetting shortcuts can help.
- IP addressing and subnetting are an important part of networks. Here's what network administrators need to know.
- Understand how subnetting in IPv6 differs from that of IPv4 and what you need to do in order to calculate the subnet of an IP address in IP version 6.