Summary
Highlights
Every network has three main address categories: the Network ID, usable host addresses, and the Broadcast address. The Network ID is where the host portion of the address is all zeros. The Broadcast address is where the host portion is all ones. Usable host addresses fall within the range between the Network ID (one bit above) and the Broadcast address (one bit below).
IPv4 addresses are 32 bits long, divided into four 8-bit groups called octets. These addresses are split into network and host portions. The network portion is defined by the subnet mask, which consists of a string of ones followed by a string of zeros. The host portion, represented by zeros in the subnet mask, is where devices get their unique addresses. IP addresses are typically written in dotted decimal form, with each octet ranging from 0 to 255. Prefix length, or slash notation, indicates the number of ones in the subnet mask, defining the network portion.
Three main characteristics define IPv4 address types: unicast (one-to-one communication), broadcast (one-to-all communication within a network segment), and multicast (one-to-a-specific-group communication). Special IPv4 addresses include private IP addresses (non-routable on the internet), loopback addresses (for self-testing a device's network stack), and link-local addresses (automatically assigned when no DHCP server is available).
Historically, IPv4 used classful addressing (Class A, B, C) with fixed subnet lengths. Today, classless addressing (CIDR) is used, allowing for flexible prefix lengths. Global IP address assignments are managed by organizations called Regional Internet Registries (RIRs), overseen by IANA (Internet Assigned Numbers Authority). RIRs like ARIN, LACNIC, AFRINIC, APNIC, and RIPE NCC distribute IP addresses for different geographical regions.
Network segmentation, primarily through subnets, is crucial for managing broadcast domains. Broadcasts don't cross Layer 3 boundaries (routers), meaning subnets isolate broadcast traffic. Segmenting networks reduces broadcast traffic, improves performance, enhances security, and allows for better traffic behavior analysis. While physical segmentation is possible, virtual segmentation using VLANs (discussed in later courses) is more common and efficient.
Subnetting allows for breaking down larger networks into smaller, more manageable subnets. The number of subnets and hosts depends on how many bits are borrowed from the host portion. Cisco's method involves 'ANDing' the IP address with the subnet mask to determine the network address. Counting host bits by doubling from one (1, 2, 4, 8...) helps determine the number of available hosts per subnet, remembering to subtract two (for network and broadcast addresses).
Efficient subnetting involves matching subnet sizes to the actual number of hosts required, and Variable Length Subnet Masks (VLSM) enables this. VLSM allows using different subnet mask lengths for different parts of a network, preventing IP address waste. For example, interconnections between routers often only need two usable hosts, making a /30 subnet ideal, while larger segments (e.g., LANs in buildings) will require larger subnets based on their host counts.
Effective IPv4 network design requires considering current and future growth, determining the need for private vs. public addresses, and planning for address assignment. A thorough network audit, analyzing requirements, and organizing the network are essential. This includes deciding how IP addresses will be assigned (e.g., DHCP for end-users, static or DHCP with reservation for servers and management devices).