Why the Internet Is Reliable
Route redundancy — so failure is not fatal
Imagine your school bus driver suddenly hits a roadblock on the usual route.
Does the bus give up and go home? No. The driver simply takes an alternative road to reach the same destination. That's how the internet works too.
What Is Route Redundancy?
The internet does not rely on a single path to send data. For any two points, there are multiple different routes a packet can take.
Route A fails
A fiber optic cable is cut or a router goes offline.
Route B takes over
Routers automatically detect the failure and redirect data around it in milliseconds.
Key point
Because there are always backup routes, the internet can survive failures in individual cables, routers, or even entire data centers. That's why it can keep working even when things break.
Interactive Diagram
Launch the interactive diagram to see this in action.
Open Interactive DiagramThe interactive diagram for this chapter illustrates route redundancy with a mesh of connected routers. You will see multiple paths between a source and destination. Clicking on any path will highlight it and show the sequence of routers it traverses. When you trigger a simulated failure (by clicking a router or cable), the diagram animates the rerouting process: affected paths turn red, routers recalculate, and alternative paths glow green as traffic shifts. The diagram also displays latency measurements for each path so you can compare the primary route with its backups.
Introduction
The internet was designed with a fundamental principle: no single point of failure should be able to take it down. This resilience is built into every layer, from the physical cables on the ocean floor to the routing protocols that decide where your data goes next. In this chapter, we explore why the internet keeps working even when things go wrong.
How It Works
Think about how many ways you can get from your house to school. There's the main road, a side street, a path through the park, and maybe a shortcut behind the shops. If one road is closed for construction, you can take a different one and still arrive on time. The internet works the same way. When you visit a website, your data can travel along many different paths. If one path breaks — maybe a cable gets cut — your data simply takes another route. That's why you can still watch videos and play games even when something breaks somewhere in the world.
Route redundancy is the internet's built-in backup system. Large networks called Autonomous Systems (ASes) connect to each other at Internet Exchange Points (IXPs). Each AS uses the Border Gateway Protocol (BGP) to share information about which networks it can reach and by what paths. When a router or cable fails, neighboring routers detect the outage within seconds and recalculate routes using BGP. The new path might be longer or slower, but data still gets through. This automatic rerouting happens without any human intervention, often in under a second. It's like a GPS navigation system that instantly finds a new route when you miss a turn.
Internet resilience is achieved through multiple layers of redundancy. At the physical layer, submarine cable systems like Marea (connecting the US to Europe) include multiple fiber pairs and diverse landing stations. At the network layer, BGP maintains a routing table of available paths, using attributes like AS path length, MED, and local preference to select optimal routes. When a BGP session drops due to a link failure, the NLRI (Network Layer Reachability Information) is withdrawn, triggering route recomputation across affected ASes. Convergence time varies from milliseconds (with BFD) to tens of seconds. Additional resilience comes from anycast addressing (used by DNS root servers and CDNs), multi-homed ISPs, and MPLS fast-reroute. The internet's decentralized, packet-switched architecture means no single cable, router, or data center is essential to its overall function.
Deeper Dive
Now that you understand the basics of why the internet is reliable, let's connect the pieces. Redundancy is one of the most important ideas in this chapter. It works together with Route to make the whole system run smoothly.
Think of Packet like a team where every member has a specific job. When one part sends information, another part receives it, checks it, and passes it along. This step-by-step teamwork is what makes technology reliable, even when many devices are involved.
Key Insight
Understanding how Redundancy and Route connect helps you explain why the internet is reliable to a friend using your own words — not just memorizing definitions.
Advanced
At a deeper level, why the internet is reliable involves rules and patterns that engineers use worldwide. Redundancy follows standards so different brands and devices can still work together. That is why your phone, school laptop, and game console can all connect to the same network or use the same apps.
Route does not happen in a straight line. Systems often use backup paths, error checking, and retries so information arrives correctly. When something fails, smart Packet design helps the system recover instead of shutting down completely.
Scientists and engineers keep improving these systems every year — making them faster, safer, and more energy-efficient. The ideas you learn in this chapter are the same building blocks used in real data centers, robots, apps, and websites around the world.
Vocabulary Table
| Term | Definition |
|---|---|
| Redundancy | Having backup systems or paths available in case the primary one fails |
| Route | The path data takes from source to destination across a network |
| Packet | A unit of data transmitted over a packet-switched network |
| BGP | Border Gateway Protocol — the protocol that exchanges routing info between ASes |
| Failover | Automatic switching to a backup system when the primary system fails |
| Latency | The time it takes for data to travel from source to destination |
| ISP | Internet Service Provider — the company that gives you access to the internet |
| Backbone | The core high-capacity network infrastructure that connects major networks |
| Autonomous System | A large network or group of networks under a single administrative control |
| Resilience | The ability of a network to maintain service during failures or disruptions |
Fun Facts
- 1
There are over 400 submarine cable systems spanning more than 1.3 million kilometers across the ocean floor — and most have at least one backup cable nearby.
- 2
When a major cable was cut near Egypt in 2008, internet traffic was automatically rerouted through other cables within minutes, affecting only a small portion of users.
- 3
BGP routing tables now contain over 900,000 routes, giving the internet an enormous number of possible paths between any two points.
- 4
Google, Amazon, and Microsoft each have their own private backbone networks connecting their data centers, so traffic never touches the public internet.
- 5
The internet was originally designed by the US Department of Defense to survive a nuclear attack — redundancy was a core requirement from the very beginning.
Misconceptions
Misconception: "The internet is one big network."
The internet is a network of thousands of independently operated networks (ASes) that voluntarily interconnect. No single company or government runs the entire internet.
Misconception: "If a cable is cut, the internet goes down for everyone."
Because of route redundancy, traffic is automatically rerouted through alternative paths. A single cable cut rarely causes widespread outages — only localized slowdowns are common.
Misconception: "More routes always mean faster internet."
Alternative routes may be longer or have higher latency. Redundancy is about reliability, not speed. In fact, a backup route might be slower but still functional.
Misconception: "Cloud services are immune to internet failures."
Even cloud providers experience outages. Redundancy reduces risk but doesn't eliminate it. AWS, Google Cloud, and Azure have all had regional failures despite massive redundancy.
Knowledge Check
Multiple Choice
- What does BGP stand for? A) Basic Gateway Protocol B) Border Gateway Protocol C) Backbone Gateway Protocol D) Broad Global Protocol
- What happens when a router detects that a path has failed? A) The internet stops working B) The router waits for a technician C) It automatically recalculates a new route D) It sends a signal to the user
- What is an Autonomous System? A) A single computer B) A network under one administrative control C) A type of router D) A fiber optic cable
True or False
- The internet has a single master cable that connects all countries. Answer: False
- Redundant routes help the internet keep working when parts of it break. Answer: True
Matching
Match each term to its description:
1. BGP — Routing protocol for the internet
2. ISP — Company that provides internet access
3. Failover — Automatic switch to backup
4. Latency — Delay in data transmission
Fill in the Blanks
- ________ is the ability of a network to maintain service during failures. Answer: Resilience
- ________ are large networks under a single administrative control that exchange routing information. Answer: Autonomous Systems
Critical Thinking
- If route redundancy makes the internet reliable, why do internet outages still happen?
Consider what kinds of failures redundancy cannot protect against, and whether there's a limit to how much redundancy is practical. - Would you design a city's road system the same way the internet's routing is designed? Why or why not?
Think about traffic congestion, alternative routes, and how redundancy affects both reliability and complexity. - Some countries have laws requiring internet companies to store data locally. How might this affect route redundancy and internet resilience?
Consider how data localization affects the number of available paths and the internet's global interconnectedness.
Mini Projects
Project 1: Redundancy Map
Research the submarine cable map (visit submarinecablemap.com) and choose a region (e.g., the Atlantic). Identify three different cables that connect the same two continents. Draw a simple map showing each cable's landing points and write a paragraph explaining how these cables provide redundancy.
Project 2: Failover Simulation
Draw a network diagram with at least 6 routers arranged in a mesh. Label a source and a destination. Use a marker to simulate a failure on the primary path — draw an X through a router or connection. Then trace at least two alternative paths the data could take. Label each path with the number of hops and estimate which would be fastest.
Teacher Notes
Learning Objectives
Students should be able to define route redundancy, explain why the internet can survive failures, name the protocol used for inter-network routing (BGP), and describe at least two real-world examples of redundancy at different network layers.
Discussion Prompts
Ask students: "Have you ever experienced an internet outage? What do you think caused it?" Discuss how redundancy affects our daily dependence on the internet. Compare to other redundant systems: backup generators, spare tires, duplicate keys.
Extended Activity
Organize the class into groups representing different ASes. Give each group a few paper packets. Have them physically pass packets between groups along designated routes. Introduce a failure (remove a group or connection) and observe how groups adapt and find new paths.