ScholarGate
Assistant

Internet Protocol and Addressing

The Internet Protocol provides connectionless, best-effort delivery of packets across interconnected networks, identifying every interface with an IP address whose structure enables the hierarchical routing that makes the global Internet scalable.

Definition

The Internet Protocol is the network-layer protocol that delivers datagrams from a source to a destination host across an internetwork using a best-effort, connectionless service; an IP address is the hierarchical numeric identifier assigned to a network interface for this purpose.

Scope

This topic covers the network-layer datagram service: the IPv4 and IPv6 packet formats, the meaning and structure of IP addresses, subnetting and classless inter-domain routing (CIDR) for hierarchical address allocation, fragmentation, and supporting mechanisms such as DHCP for address assignment and NAT for address sharing. It treats addressing as the foundation that routing builds on, and excludes the routing algorithms and protocols themselves, which are covered separately.

Core questions

  • What service model does IP provide, and what does best-effort delivery mean?
  • How are IPv4 and IPv6 addresses structured, and how do their packet formats differ?
  • How do subnetting and CIDR enable hierarchical, aggregatable addressing?
  • How are addresses assigned to hosts, for example via DHCP?
  • What problems led to IPv6 and to mechanisms such as NAT?

Key concepts

  • best-effort delivery
  • IPv4 and IPv6 packet formats
  • IP addresses
  • network and host portions
  • subnetting
  • classless inter-domain routing (CIDR)
  • fragmentation and reassembly
  • DHCP
  • network address translation (NAT)

Key theories

Best-effort connectionless datagram service
IP forwards each datagram independently with no guarantee of delivery, ordering, or timing; this simplicity keeps the network core stateless and robust, pushing reliability and ordering to the end hosts' transport layer.
Hierarchical addressing and CIDR
IP addresses are partitioned into network and host parts so that routers can aggregate many addresses under a single prefix; classless inter-domain routing generalizes this with variable-length prefixes, curbing routing-table growth.
Address-space expansion with IPv6
IPv6 replaces IPv4's 32-bit addresses with 128-bit addresses and a simplified header, resolving address exhaustion and adding features such as stateless autoconfiguration while coexisting with IPv4 during a long transition.

Clinical relevance

IP addressing is the foundation of reachability on the Internet: every connected device needs an address, and the way addresses are allocated and aggregated determines the size and stability of the global routing system. Address exhaustion drove the deployment of NAT and the ongoing migration to IPv6, decisions that shape network design, security, and connectivity for billions of devices.

History

IPv4 was specified in RFC 791 (1981) with a 32-bit address space and an initial class-based allocation scheme. Rapid Internet growth led to CIDR (mid-1990s) to slow routing-table and address exhaustion, and to NAT to share scarce addresses. IPv6, with 128-bit addresses, was designed in the 1990s and finalized as RFC 8200 (2017), and its deployment has grown steadily since.

Debates

NAT as a stopgap versus end-to-end addressing
Network address translation extended IPv4's life by letting many hosts share an address, but it breaks the end-to-end addressing principle and complicates peer-to-peer connectivity; proponents of IPv6 argue restoring globally unique addresses is the proper fix.

Key figures

  • Jon Postel
  • Vinton Cerf
  • Steve Deering
  • Robert Hinden

Related topics

Seminal works

  • rfc791
  • rfc8200
  • kurose2021

Frequently asked questions

What does best-effort delivery mean?
Best-effort means IP tries to deliver each datagram but makes no promises: packets may be lost, duplicated, delayed, or reordered. This keeps the network simple and resilient, and leaves it to higher layers, such as TCP, to add reliability and ordering when an application needs them.
Why do we still use IPv4 if IPv6 exists?
IPv6 adoption requires updating hosts, routers, and applications across the entire Internet, so the transition has been gradual. Techniques like NAT let IPv4 keep working despite address scarcity, and IPv4 and IPv6 run side by side; the migration continues rather than happening all at once.

Methods for this concept

Related concepts