The Internet is designed with built-in redundancy and resilience to ensure continuous operation even when parts of the network are disrupted. Its distributed architecture allows data to be rerouted dynamically through alternative paths using routing protocols such as BGP (Border Gateway Protocol). Routers detect when a specific route is no longer reachable and adjust their routing tables to forward packets through different autonomous systems. Additionally, major backbone providers often have multiple physical connections, including undersea cables and satellite links, to ensure alternate paths are available. Large-scale data centres and ISPs typically maintain failover infrastructure, such as backup power supplies and geographically distributed mirrors of services. This means if one data centre or route is impacted by a disaster or attack, requests can be redirected to a functioning location. Overall, the Internet’s decentralised and self-healing design, along with proactive monitoring and automation, enables it to withstand localised failures and maintain service globally.

Internet Exchange Points (IXPs) are physical locations where different networks—particularly autonomous systems (ASes)—interconnect and exchange data traffic directly. Their primary purpose is to facilitate peering relationships, where multiple ISPs, content providers, and enterprise networks share data without relying on intermediary transit networks. This reduces latency, lowers costs, and improves routing efficiency, since traffic takes a more direct path between networks. IXPs are critical for regional Internet performance because they allow local traffic to stay local rather than being routed through distant locations. For example, when two users in the same country access each other’s services, the traffic may be routed through a nearby IXP instead of travelling internationally. IXPs also reduce dependency on Tier 1 providers by allowing smaller networks to connect and share bandwidth affordably. Their presence enhances the scalability, speed, and stability of the Internet by optimising how networks communicate and handle large volumes of data.

Fibre-optic cable is preferred for the Internet backbone because it supports significantly higher bandwidth, faster data transmission speeds, and longer distances compared to traditional copper cable. It transmits data using light pulses through strands of glass or plastic, allowing it to carry terabits of data per second with minimal signal degradation. Fibre-optic cables are also immune to electromagnetic interference, making them ideal for use in environments with high electrical noise or long-distance transmission requirements. In contrast, copper cables transmit data using electrical signals, which experience greater attenuation and are more susceptible to signal loss over distance. This requires copper-based systems to rely on more frequent repeaters to boost signals, increasing cost and complexity. Additionally, fibre-optic infrastructure offers better security since it is harder to tap undetected. These advantages make fibre optics the standard choice for high-capacity networks, including the Internet backbone that interconnects global data centres and ISPs.

Tier 1 and Tier 3 ISPs differ mainly in their network size, coverage, and dependency on other networks. A Tier 1 ISP is a large-scale network provider that has global reach and peering agreements with all other Tier 1 networks, meaning it does not pay for Internet transit. These ISPs form the backbone of the Internet and manage high-capacity infrastructure such as undersea fibre-optic cables and international routing hubs. Examples include companies like Level 3 and NTT Communications. A Tier 3 ISP, on the other hand, is a local or regional provider that purchases access to the Internet through Tier 2 or Tier 1 providers. Tier 3 ISPs serve end users, such as homes and small businesses, and do not maintain their own large-scale routing infrastructure. They rely heavily on upstream providers to route traffic beyond their service area. Tier 3 ISPs focus on customer service and last-mile delivery, while Tier 1s manage core global transit.

Devices determine the route for sending data using a process called routing, which relies on routing tables and routing protocols. When a packet is sent, the source device passes it to its local router. This router consults its routing table, which lists destination addresses and the best-known paths to reach them. If the destination is not on the local network, the packet is forwarded to the next router along the route. Routers communicate with each other using protocols like OSPF (Open Shortest Path First) for internal networks and BGP (Border Gateway Protocol) for inter-AS routing. These protocols allow routers to share information about network topology, link availability, and path costs. BGP, in particular, evaluates policies, AS path lengths, and administrative rules to determine optimal paths. At each hop, routers make new decisions based on updated routing tables, continuing the process until the packet reaches its destination. This hop-by-hop forwarding allows data to dynamically traverse the most efficient available route.