Fault tolerance refers to how a computer continues to work correctly despite a system failure. You should know that no matter how many performance tests a computer goes through before getting sold, it can still experience system failures. But computers are designed so that despite system errors, they can still work correctly and give the correct results as much as possible.

Computer errors can stem from three major components—hardware, software, and power sources. Most often than not, when failures occur, an application or your entire computer will shut down and restart. After that, most recently saved copies of the files you worked on when the fault occurred will be available.

Read More about Fault Tolerance

Fault-tolerant systems, as mentioned in the fault tolerance definition above, have backup components that instantly take the place of the parts that failed to ensure no loss of service.

How Can You Ensure Fault Tolerance?

Here are three ways to ensure fault tolerance in various system components.

  • Hardware: These components typically use backups that are identical or equivalent systems. You can, for instance, make a server fault-tolerant by using an identical backup server that runs parallel to it and mirrors all its operations.
  • Software: These components can use backups like other software instances. An example would be continuously replicating a customer information database on another computer. That way, if the primary database goes down, your operations won’t suffer because users will automatically be redirected to the backup database.
  • Power sources: You can ensure fault tolerance by using alternative power sources. Most companies, for instance, have power generators that take over in case electric power goes out.

Any component can be made fault-tolerant through redundancy or by creating backups.

fault tolerance

Is Fault Tolerance the Same as High Availability?

High availability is a system’s ability to avoid loss of service by minimizing downtime. It has to do with system uptime as a percentage of the total running time. A system is considered highly available if it has 99.999% uptime.

Although a fault-tolerant system is a highly available one, not all highly available systems are fault-tolerant. Fault tolerance covers more than resilience to downtime. It also ensures no data is corrupted or lost.

If the difference between the two remains unclear, here’s an example. Consider a car with a readily available spare tire. A flat tire will cause the car to stop, but you’ll experience little downtime since you can easily replace it. That car is thus highly available, but it isn’t fault-tolerant. It will only become fault-tolerant if its tires self-heal.

What Should You Consider When Ensuring Fault Tolerance?

Three primary elements should be considered when ensuring fault tolerance, namely:

  • Downtime: To ensure uninterrupted business operations, all systems must be fault-tolerant.
  • Scope: To ensure fault tolerance, a company must have backups for all its hardware, software, and power sources. And all these must be configured to automatically take over when their primary counterparts fail.
  • Cost: Given the assurance of uninterrupted operations, creating a fault-tolerant network can be costly. Instead of just purchasing one server, for example, you’d need two—one as the primary server and the other as a backup.

Ensuring fault tolerance requires companies to answer these questions about every system they use:

  • How critical is the component? Will the business continue to operate without it? If not, the system needs to be made fault-tolerant.
  • How likely is the component to fail? Does the system tend to fail? If it does, it may need a backup to become fault-tolerant.
  • How expensive is it to make the component fault-tolerant? Is the cost of making the system fault-tolerant less than what you’ll incur if it fails? If you’re likely to lose revenue and even customers if it fails, then the cost you’ll pay to make it fault-tolerant is worth it.

Trivia

Did you know that the first fault-tolerant computer was SAPO, built in 1951 in Czechoslovakia by Antonín Svoboda? His creation was used to develop systems for military, space, and industrial use.

—

In today’s technology-driven world, no business can go offline for much longer than minutes, which could translate to productivity and, consequently, revenue loss. That makes fault tolerance a must despite the work, time, effort, and cost it entails.

Key Takeaways

  • Fault tolerance is how a computer continues to work correctly despite a system failure.
  • While fault tolerance is often mistaken for high availability, the concepts differ.
  • The three elements of fault-tolerant systems are hardware, software, and power sources.
  • Redundancy is the most effective way to ensure fault tolerance.