RAID Laboratory
Build it. Break it. Recover it.
Your mission
Slide real drives into a NAS, choose RAID 0, 1, 5 or 6, and watch your files split into blocks, mirrors and parity. Then pull drives out while it's running and see which arrays survive. Finish by building storage to order, and find out the one thing no RAID can protect you from.
Explore
Save some files to the NAS, then switch between RAID 0, 1, 5 and 6 and watch where every block of your files ends up.
This is a NAS with four drives. RAID decides how your files are spread across them: for speed, for safety, or both. Press Save files to the NAS and watch the blocks fly in. Then try every RAID level and tap anything that glows.
0 of 7 discoveries made
Experiment
Pull drives out while the NAS is running. Which arrays survive, which ones lose everything, and what can bring a deleted file back?
Break-it experiments
Try each one. They tick themselves off.
0 of 5 experiments done
Challenge
Six rounds: build storage to a customer's order, work out usable capacity, and decide what RAID can and can't protect.
Storage engineer's notes
- N drives, each of size S (all the same size)
- RAID 0: N × S, survives 0 failures
- RAID 1: two mirrored drives, S usable (50%), survives 1
- RAID 5: (N − 1) × S, survives 1
- RAID 6: (N − 2) × S, survives 2
- RAID is not a backup: use 3-2-1
Finish all 6 rounds to finish this stage.
Understand
Here is what the NAS was showing you, in the words you need for the exam.
What is RAID?
RAID (Redundant Array of Independent Disks) combines multiple physical drives into one logical storage system to improve performance, provide fault tolerance, or both, depending on the RAID level. Your files are cut into blocks, and the RAID level decides where the blocks go: spread across the drives (striping), copied (mirroring), or stored together with parity blocks that let lost data be rebuilt.
The four RAID levels
| Level | How it stores data | Usable capacity | Survives | Four 4 TB drives |
|---|---|---|---|---|
| RAID 0 | Striping: blocks spread across all drives, no redundancy. Fastest reads and writes. | N × S | 0 drive failures | 16 TB |
| RAID 1 | Mirroring: identical copies of the data on both drives. | S (50% of two drives) | 1 drive failure | 4 TB (two drives) |
| RAID 5 | Data plus one distributed parity block per stripe; the parity moves from drive to drive. | (N − 1) × S | 1 drive failure | 12 TB |
| RAID 6 | Data plus two different parity blocks per stripe (P and Q), also distributed. | (N − 2) × S | 2 drive failures | 8 TB |
N = number of drives, S = size of one drive. The capacities are nominal: they ignore formatting overhead.
What parity does
A parity block is extra information calculated from the data blocks in its stripe. If one drive fails, each missing block can be reconstructed from the remaining data blocks and the parity. RAID 6 stores two different parity calculations (P and Q), not two copies, so it can rebuild even when two drives fail.
Degraded and rebuilding
When a drive fails but the array survives, it keeps working in a degraded state: the data is still available, but there is less protection left. After a new drive is inserted, the array rebuilds the missing blocks onto it. If more drives fail than the level can tolerate, the stored data is generally lost.
RAID is not a backup
| RAID (redundancy) | Backup (independent copy) | |
|---|---|---|
| What it is | Several drives working together in one system | A separate copy of your data, on a different device and/or in a different place |
| Protects against | Certain drive failures (up to two in RAID 6): keeps data available | Accidental deletion, ransomware, theft and fire: lets you recover data |
| Deleted a file? | Deleted on every drive at once | Restore it from the copy |
The 3-2-1 backup principle
- 3 copies of your data
- on 2 different storage media (or types)
- with 1 copy off-site (another home, the office, or cloud storage)
Use RAID for availability, and backup (e.g. 3-2-1) for real data protection.
Speed vs safety
- RAID 0: uses all the space and is fast, but if any one drive fails the whole array is generally lost.
- RAID 1: very safe and simple, but half the raw capacity is used for the copy.
- RAID 5 and 6: give up one or two drives' worth of space for parity, in exchange for surviving one or two failures.
Words to know
- Striping
- Splitting data into blocks spread across several drives (RAID 0).
- Mirroring
- Keeping identical copies of the data on two drives (RAID 1).
- Parity
- Extra information that lets lost data be reconstructed (RAID 5: P, RAID 6: P and Q).
- Fault tolerance
- How many drive failures an array can survive without losing data.
- Raw capacity
- All the drives added together: N × S.
- Usable capacity
- What is left for your files after copies or parity.
- Degraded
- Still working after a failure, but with less protection.
- Backup
- A separate, independent copy of your data that you can restore from.
Exam tip
For capacity questions, write the formula first, then the numbers: five 6 TB drives in RAID 5 give (5 − 1) × 6 TB = 24 TB, in RAID 6 (5 − 2) × 6 TB = 18 TB, in RAID 0 5 × 6 TB = 30 TB. Always state the fault tolerance too (0, 1 or 2 drives). And if a question says "RAID, so no backup needed", the answer is no: RAID does not protect against accidental deletion, ransomware, theft or fire.