RAID — Redundant Array of Independent Disks — combines two or more physical drives so that a computer treats them as a single storage unit. Different RAID levels trade off between performance, storage capacity, and fault tolerance, allowing organisations to choose the balance that suits their needs.
What problem does RAID solve?
Hard drives fail. Even solid-state drives fail. In a server storing thousands of customer records, a single drive failure can cause catastrophic data loss and downtime. RAID addresses this in two ways:
- Redundancy — storing copies of data on multiple drives so that if one fails, the data survives.
- Performance — spreading data across multiple drives so that read and write operations happen in parallel, faster than a single drive could manage.
The trade-off: some RAID levels use extra drives purely for protection, reducing the usable storage capacity compared to the total raw capacity installed.
What is RAID 0 (striping)?
RAID 0 splits (stripes) data evenly across two or more drives. When saving a file, chunks of data are written simultaneously to both drives; when reading, chunks are retrieved in parallel.
Example with 2 × 1 TB drives:
| Drive A | Drive B |
|---|---|
| Block 1 | Block 2 |
| Block 3 | Block 4 |
| Block 5 | Block 6 |
Usable capacity: 2 TB (all capacity used for data — nothing reserved for redundancy) Performance: roughly double the read/write speed of a single drive Fault tolerance: none — if either drive fails, all data is lost
RAID 0 is used where speed matters and data can be easily recreated — for example, video editing scratch disks or gaming asset caches. It is not suitable for storing irreplaceable data.
What is RAID 1 (mirroring)?
RAID 1 mirrors data — every write is duplicated identically to two or more drives. At any moment, both drives hold exactly the same data.
Example with 2 × 1 TB drives:
| Drive A | Drive B |
|---|---|
| Block 1 | Block 1 (copy) |
| Block 2 | Block 2 (copy) |
| Block 3 | Block 3 (copy) |
Usable capacity: 1 TB (only half the raw capacity, because everything is duplicated) Read performance: can be slightly improved (read from whichever drive responds first) Write performance: similar to a single drive (both drives must confirm each write) Fault tolerance: high — one drive can fail completely and the system continues without data loss
RAID 1 is used where data safety is paramount — database servers, financial systems, and operating system drives. When a failed drive is replaced, the surviving drive is copied to the new one automatically (called rebuilding or resilvering).
What is RAID 5 (striping with parity)?
RAID 5 combines striping and redundancy without dedicating an entire drive to mirroring. Data is striped across three or more drives, and parity information is calculated and distributed across all drives.
What is parity? Parity is a computed value that allows a missing piece of data to be reconstructed. If you know two out of three values and their combined parity, you can recover the third.
Example with 3 × 1 TB drives:
| Drive A | Drive B | Drive C |
|---|---|---|
| Data 1 | Data 2 | Parity(1,2) |
| Data 3 | Parity(3,4) | Data 4 |
| Parity(5,6) | Data 5 | Data 6 |
Usable capacity: 2 TB (the equivalent of one drive's worth is used for distributed parity) Performance: good read performance; write performance slightly reduced (parity must be recalculated on each write) Fault tolerance: one drive can fail — parity blocks on remaining drives allow reconstruction of all lost data
RAID 5 is the most widely used configuration in business servers because it provides a good balance of speed, capacity efficiency, and protection.
Comparing the three key RAID levels
| Feature | RAID 0 | RAID 1 | RAID 5 |
|---|---|---|---|
| Minimum drives | 2 | 2 | 3 |
| Usable capacity | 100% of total | 50% of total | (n-1)/n of total |
| Read speed | Excellent | Good | Excellent |
| Write speed | Excellent | Good | Good |
| Survives 1 drive failure? | No | Yes | Yes |
| Main use case | Performance | Safety | Balanced |
How does parity allow data recovery?
Suppose drives A and B hold Data 1 = 10101100 and Data 2 = 01100011. The parity value is calculated using XOR (exclusive OR) — a bitwise operation where the output is 1 wherever the inputs differ:
Data 1: 10101100
Data 2: 01100011
Parity: 11001111 (XOR of Data 1 and Data 2)
If Drive A fails, RAID 5 recovers Data 1 by XOR-ing Data 2 with the Parity:
Data 2: 01100011
Parity: 11001111
Result: 10101100 ← original Data 1, fully recovered
This is how parity provides fault tolerance without a full copy of the data.
Frequently asked questions
Is RAID a replacement for backups?
No — and this is a critical point. RAID protects against hardware failure (a drive dying), but it does not protect against accidental deletion, file corruption, ransomware (which encrypts data on all mirrors simultaneously), or fire/flood destroying all drives. RAID and backups serve different purposes and are both necessary in any serious data management strategy.
What happens when a drive in RAID 5 fails?
The system continues running in a degraded state, using parity calculations to reconstruct the missing data on the fly. Performance drops because every read requires extra calculation. The failed drive must be replaced promptly — if a second drive fails before the array has finished rebuilding, all data is lost. This is why some organisations use RAID 6, which tolerates two simultaneous drive failures.
Why does RAID 5 need at least three drives?
RAID 5 distributes parity across all drives. With only two drives, parity would have to go on one of the two — effectively becoming RAID 1 mirroring. Three drives allow parity to be spread (rotating which drive holds parity for each stripe), giving the capacity and performance benefits of striping while still providing one-drive fault tolerance.
Is RAID used with SSDs as well as hard disk drives?
Yes — RAID configurations work with SSDs, HDDs, or a mix. SSDs are much faster than HDDs, so RAID levels that improve write performance (RAID 0, RAID 5) offer proportionally smaller gains with SSDs than with slower HDDs. RAID 1 with SSDs is popular for system drives because it combines the speed of flash storage with data protection against drive failure.
Need help with systems architecture topics for your GCSE? Professor Turing at aitutors.me will explain storage, memory, and every hardware concept clearly.