ScholarGate
Assistant

Network Edge and Core

The network edge comprises the hosts and access networks where applications run and traffic originates, while the network core is the mesh of routers and links that carries packets between edges; together they describe the large-scale structure of the Internet.

Definition

The network edge is the collection of end systems and the access networks that attach them to the Internet; the network core is the mesh of packet switches (routers) and links that interconnects these access networks and forwards traffic between end systems.

Scope

This topic covers the structural decomposition of a network into edge and core. At the edge it treats end systems (clients and servers) and the access networks — DSL, cable, fiber-to-the-home, wireless, and enterprise Ethernet — that connect them. In the core it treats the interconnected routers and the tiered structure of Internet service providers and Internet exchange points. It excludes the routing algorithms and forwarding mechanisms that operate within the core, which belong to the network-layer area.

Core questions

  • What end systems and access technologies make up the network edge?
  • How do residential, mobile, and enterprise access networks connect hosts to the Internet?
  • What devices and links constitute the network core?
  • How are Internet service providers organized into tiers and interconnected?
  • What role do Internet exchange points and peering play in the core?

Key concepts

  • end systems (hosts)
  • access networks (DSL, cable, fiber, wireless)
  • network core
  • packet switches and routers
  • Internet service providers (ISPs)
  • tier-1, regional, and access ISPs
  • Internet exchange points
  • peering and transit

Key theories

Edge-core decomposition
The Internet is usefully divided into an edge of end systems and access networks, where data originates and terminates, and a core of routers that forward traffic between edges; this framing clarifies where applications, access technology, and forwarding live.
Tiered ISP structure and peering
The core is organized as a loose hierarchy of access, regional, and global transit providers interconnected through transit and peering relationships at Internet exchange points, an economic and topological structure that emerged rather than being centrally designed.

Clinical relevance

The edge-core view underlies practical questions about Internet access and connectivity: choosing and provisioning broadband technologies, understanding where bottlenecks and last-mile limits arise, and reasoning about how content reaches users through tiers of providers. It frames discussions of net neutrality, peering disputes, and the placement of content and compute closer to the edge.

History

The early ARPANET had a small core of interface message processors connecting a handful of hosts. As the Internet commercialized in the 1990s, a market of interconnected ISPs replaced the single backbone, producing today's tiered structure with global transit providers, regional networks, and a proliferation of access technologies at the edge, knit together at Internet exchange points.

Key figures

  • James F. Kurose
  • Keith W. Ross
  • Larry L. Peterson

Related topics

Seminal works

  • kurose2021
  • peterson2021

Frequently asked questions

What is the difference between the network edge and the network core?
The edge is where end systems live and where data is produced and consumed, together with the access links that connect them. The core is the interior mesh of routers and high-capacity links that carries traffic between different parts of the edge. Applications run at the edge; forwarding happens in the core.
What is an access network?
An access network is the link or set of links that connects an end system to the first router of its provider — for example a DSL or cable line, a fiber-to-the-home connection, a cellular link, or an enterprise Ethernet LAN. It is the 'last mile' between a host and the wider Internet.

Methods for this concept

Related concepts