Post 36 | Redundancy vs Backup: When Failsafes Fail You
Post 36 | Redundancy vs Backup: When Failsafes Fail You
In the age of automation and cloud computing, we often place blind faith in systems we barely understand — confident that someone, somewhere, has designed things to “just work.”
And for the most part, they do.
But what happens when they don’t?
Let me tell you a story — not from mythology this time, but from a typical Monday morning.
Imagine an office. One of those “paperless” ones where everything — files, projects, schedules — lives on a central server. One day, a hard disk fails. Panic flickers… then fades. The system keeps running. There was a failover disk. Redundancy worked.
Confidence restored.
Until a week later, someone realizes a folder was accidentally deleted before the failure. And with it, three weeks of work. The deletion, replicated faithfully across all redundant drives, was permanent. Redundancy had done its job — it had preserved the system’s state.
But not its history.
This is the silent gap many overlook: the difference between redundancy and backup. Two words often used interchangeably. Two concepts that are anything but.
Understanding Redundancy: The Illusion of Safety
Redundancy is designed to ensure continuity. It is the equivalent of having multiple engines on a plane — if one fails mid-air, the others keep you flying. In IT systems, this often takes the form of RAID (Redundant Array of Independent Disks) — a setup where your data is spread across multiple drives. If one drive fails, the data survives on another.
This sounds like safety. And to an extent, it is.
But redundancy is not designed to remember. It does not question what it is storing. If a file is corrupted, deleted, or encrypted by ransomware, redundancy ensures those changes are preserved — across all drives — with perfect fidelity.
It’s like having multiple scribes copying the same manuscript. If the original page is torn or miswritten, every copy reflects that mistake. The system never asks: “Was this intentional?”
The Role of Backups: Memory, Not Just Mirrors
Backups serve a different purpose altogether. They are not designed for continuity, but for recovery. A proper backup is a snapshot in time — a historical record of what your system looked like before something went wrong.
It is your digital journal. And like any good journal, it stores multiple entries: yesterday, last week, last month.
Unlike redundancy, a good backup system is:
- Independent: It is stored separately from the system it protects — often offline or in the cloud.
- Versioned: It retains older versions of files, allowing rollback to a clean state.
- Immutable: When designed properly, it cannot be altered by malware or human error.
If redundancy is the aircraft’s backup engine, backup is the parachute — something you never hope to use, but deeply regret not having when the entire system goes down.
Why Confusing the Two Can Be Costly
A system that relies solely on redundancy assumes that failure is always mechanical. It protects against drive malfunctions but not against human mistakes, malicious code, or corrupted data.
And in reality, most data loss isn’t due to hardware failures.
It’s due to a document overwritten.
A folder deleted.
An email clicked.
A file locked and encrypted with no way out.
These are moments when redundancy offers no answers. Only backup can restore what was.
Navigating the Difference: What You Should Do
To build a truly resilient system, both layers are essential. Here’s how to think about it:
- Use redundancy for uptime — to keep systems running during failure.
- Use backups for restoration — to bring systems back to a known good state after an incident.
Adopt the 3-2-1 rule of backups:
- 3 total copies of your data
- 2 stored on different types of media
- 1 stored offsite
And most importantly — test your backups. A backup that cannot be restored is no backup at all. It’s just a comforting illusion.
The Final Thought: Failsafe ≠ Safe From Failure
In cybersecurity, as in life, the danger isn’t always what you can see. Sometimes, it’s what you assumed was covered.
Redundancy will keep your system alive.
Backup will keep your history intact.
When the unexpected happens — and it will — you don’t just need to keep flying.
You need a way to land.