AP CSP 4.2 Fault Tolerance | Redundancy and Reliable Networks
Fault Tolerance
After this lesson, you will be able to:
- Define fault tolerance as a system's ability to keep working when one or more components fail
- Explain redundancy as having multiple paths or copies so a failure does not stop the system
- Identify a single point of failure in a described network and explain why it is dangerous
- Explain why the Internet is fault tolerant because of redundant routing
- Describe the tradeoff between more redundancy and higher cost and resource use
A small company runs its entire website through a single router in the closet. It is fast and cheap and works fine, until the day that one router dies and the whole site goes dark for everyone at once. A rival company spends a little more and wires every server to two separate routers. When one of theirs fails, customers never even notice. Both networks had a failure. Why did only one of them go down, and what did the second company buy with that extra wiring?
What Fault Tolerance Actually Means
Fault tolerance is the ability of a system to keep working, possibly at reduced performance, even when one or more of its components fail. A fault-tolerant system does not promise that nothing ever breaks. It promises that when something breaks, the whole system does not stop. The web page still loads, the packet still arrives, the service stays up, just maybe a little slower while the broken part is worked around.
The mechanism that makes this possible is redundancy: having more than one path or more than one copy of a component, so that if one fails, another can take over. A network with only one cable between two points has no redundancy. A network with two independent cables between those same points is redundant, because losing one cable still leaves a working path. Redundancy is what you build in ahead of time so that fault tolerance is what you get when a failure actually happens.
| Term | What it means |
|---|---|
| Redundancy | Having multiple paths or copies of a component so the system can still function if one part fails. |
| Fault tolerance | The ability of a system to keep working, possibly at reduced performance, when one or more components fail. |
| Single point of failure | A component whose failure alone stops the entire system, because there is no backup path or copy for it. |
When a question asks why a system is fault tolerant, the credited answer is almost always redundancy: there is more than one path or copy, so a failure can be worked around. Fault tolerance is the outcome; redundancy is the reason.
Single Points of Failure
A single point of failure is a component whose failure, all by itself, brings down the whole system. If every request in a network has to pass through one router, one cable, or one server, then that one thing is a single point of failure: the moment it dies, everything that depends on it stops. Redundancy is precisely the tool for removing single points of failure. When you add a second path around a component, that component is no longer the only way through, so its failure no longer takes down the system.
To spot a single point of failure in a described network, look for a component that every path must go through. If cutting one node or one link disconnects part of the network from the rest, that node or link is a single point of failure. If you can cut any one component and traffic still has another route, then there is no single point of failure at that spot.
Redundancy is not the same as making one path faster. Upgrading your single router to a faster model, or buying more bandwidth on your one cable, improves performance but adds zero fault tolerance: that faster router is still a single point of failure, and when it dies the system still goes fully down. Fault tolerance comes from having another path, not a better single path.
Why the Internet Keeps Working: Redundant Routing
The Internet is a huge, fault-tolerant network, and the reason is redundant routing. Devices on the Internet are connected by many paths, not one. When you send data, it is broken into packets, and each packet is routed hop by hop from one router to the next toward its destination. If a router or a connection along the way fails, the routers around it can send packets along a different available path. The data still gets through; it just takes another route.
This is why no single broken cable or dead router takes down the whole Internet. Because there are multiple paths between most points, the failure of any one path can be worked around. In general, connecting nodes with more than one path increases fault tolerance, and a network where every pair of important nodes has at least two independent paths has no single point of failure between them.
- One path between two nodes: if that path fails, they are cut off. Not fault tolerant.
- Two or more independent paths between two nodes: if one fails, traffic reroutes over another. Fault tolerant.
The Tradeoff: Redundancy Is Not Free
If redundancy is so good, why not build infinite backups everywhere? Because redundancy costs money and resources. Every extra path means extra cables, extra routers, extra servers, extra electricity, and extra maintenance. A second data center that mostly sits idle waiting for the first one to fail is expensive to run. So designers face a tradeoff: more redundancy generally means more fault tolerance, but also higher cost and more resources used.
The right amount of redundancy depends on how important reliability is. A hospital network or a bank cannot tolerate downtime, so they pay for heavy redundancy. A hobby blog might accept the risk of a single server to save money. The exam wants you to recognize both directions of this relationship: adding redundant components increases fault tolerance, and it also increases cost.
How Topic 4.2 Shows Up on the MCQ Exam
Fault tolerance questions on the multiple-choice section usually come in a few predictable shapes, and knowing them turns these into fast points. The most common is the network diagram or description where you must identify a single point of failure: look for the one component that every path passes through, the one whose removal disconnects part of the network. A close cousin asks which change would make a network more fault tolerant, and the answer is almost always the one that adds an additional path or a redundant component between nodes, not the one that upgrades a single link.
The trap the exam loves is confusing redundancy with speed. Answers that make one path faster, add more bandwidth to a single cable, or use a more powerful single server improve performance but do not add fault tolerance, because a single point of failure remains. Remember the exact framework claim: the Internet is fault tolerant because of redundant routing, and connecting nodes with more than one path increases fault tolerance. Watch for stem words like MOST, LEAST, and EXCEPT that flip which option you should pick.
Get a free AP CSP question every day
Join 3,000+ students. Daily practice, study tips, and exam strategies.
Which statement best describes what it means for a network to be fault tolerant?
A network currently connects a company's main office to its data center over a single leased line. Which change would MOST increase the network's fault tolerance?
In a network, computers X, Y, and Z each connect only to switch S, and switch S is the only device that connects to the rest of the company network. Which statement is correct?
Consider these claims about redundancy and fault tolerance:
- I. Adding redundant paths between nodes can increase a network's fault tolerance.
- II. The Internet is fault tolerant largely because packets can be rerouted along alternate paths when a router or connection fails.
- III. Redundancy is free, so a well-designed network should always add as much of it as possible.
A streaming company decides to run its service from three geographically separate data centers instead of one, so the service stays up if a whole data center goes offline. Which statement best describes the consequence of this decision?
A message is sent across the Internet from one city to another. Partway through the transmission, a router on the original route fails, yet the complete message still arrives at its destination. Which of the following best explains this outcome?
Redundant Routing
Cut the Connection - why the Internet keeps working when links fail
How to play: tap a link to cut it (tap again to restore). The map instantly reroutes and shows if A can still reach B.
Frequently Asked Questions
🔗 Continue studying
Fault tolerance is best taught at the whiteboard with node-and-edge diagrams: draw a network, ask students to find the single point of failure, then have them add one link that removes it. Push the tradeoff explicitly by giving each redundant link a dollar cost and a fixed budget, so students feel that more fault tolerance is not free. The Superpack includes diagram worksheets, a cut-the-link routing activity, and a set of single-point-of-failure exam drills. View what's included →
Get in Touch
Whether you're a student, parent, or teacher — I'd love to hear from you.
Just want free AP CS resources?
Enter your email below and check the subscribe box — no message needed. Students get daily practice questions and study tips. Teachers get curriculum resources and teaching strategies.
Message Sent!
Thanks for reaching out. I'll get back to you within 24 hours.
Prefer email? Reach me directly at [email protected]