0% completed
What is Redundancy?
On This Page
The Simplest Version
What Redundancy Buys You
- Improved reliability
- Enhanced fault tolerance
- Increased availability
- Simplified maintenance
- Disaster recovery
Where the Word Shows Up
What It Costs
Every system has parts that everything else depends on. One database. One load balancer. One machine running the service that authenticates users.
A part like that is called a single point of failure. When it stops, the whole system stops, no matter how healthy everything else is.
Redundancy is the fix. It means duplicating critical components or functions so that the system keeps working when one copy fails. That duplication buys three things at once: reliability, availability and fault tolerance.
The Simplest Version
Suppose your service runs as one instance in production. The machine loses power. Every request now fails until somebody notices and rebuilds it.
Now run two instances instead. One loses power, and traffic moves to the other. Users see nothing.
The move from the failed component to the working one is called a failover. Nothing about the failure changed. What changed is that the system had somewhere else to go.
What Redundancy Buys You
1. Improved reliability
A redundant system keeps functioning when an individual component fails. Fewer component failures turn into outages, so the services people depend on stay up.
2. Enhanced fault tolerance
Redundant components let the system tolerate a fault and recover from it. This is what keeps downtime small in systems where downtime is expensive.
3. Increased availability
Because a spare is always ready, the system can serve requests continuously, through both failures and planned maintenance. For anything expected to work around the clock, that is the whole point.
4. Simplified maintenance
This one is easy to overlook and it is worth as much as the others. With duplicates in place, you can take one component out of service, patch it, upgrade it, restart it, and put it back, while the redundant components keep serving. Without redundancy, every upgrade is an outage.
5. Disaster recovery
Keep the copies in more than one place and you are protected from more than a dead machine. If a whole location is lost, a copy somewhere else can be brought up. We come back to this in the last lesson of the chapter.
Where the Word Shows Up
You have already met redundancy in this course without it being named.
- A load balancer sends traffic to several backend servers, and removes any that fails its health check. Those servers are redundant copies of each other.
- A CDN keeps the same asset on many edge servers, so one being unreachable does not make the asset unreachable.
- A DNS name resolves to several addresses, and a resolver that cannot reach one tries the next.
The pattern is always the same. Find the part that everything depends on, and make sure there is more than one of it.
What It Costs
Redundancy is not free, and an interview answer that skips the cost is incomplete.
You pay for hardware you hope never to use. A standby that sits idle still consumes a machine, a licence and an operations budget. You also pay in complexity: something has to detect the failure, decide the component is really dead rather than briefly slow, and move traffic across without sending it to both.
That detection problem is exactly what heartbeats solve, and getting the timeout wrong in either direction has real consequences.
💡 Interviewers listen for whether you can name the single point of failure in your own design. Draw the boxes, then ask out loud: "if this one dies, what happens?" A design where the answer is "everything stops" is not finished, and noticing that yourself is worth more than being told.
Key takeaway: Redundancy means duplicating critical components so that the failure of one does not stop the system. It removes single points of failure and gives you reliability, fault tolerance, availability, maintenance without downtime, and a path to disaster recovery. It costs money and adds the problem of detecting failure correctly.
Redundancy keeps spare parts ready. The next lesson, What is Replication?, covers the related idea that keeps spare data ready, and the two are easy to confuse.
On This Page
The Simplest Version
What Redundancy Buys You
- Improved reliability
- Enhanced fault tolerance
- Increased availability
- Simplified maintenance
- Disaster recovery
Where the Word Shows Up
What It Costs