Resource Cheatsheet
A+ Core 1 3.4: RAID
Professor Messer A+ 220-1201 objective 3.4 — RAID 0 striping, RAID 1 mirroring, RAID 5 and RAID 6 parity, and nested RAID 1+0, plus why RAID is not backup.
- Updated
- Reading time
- 5 min read
- Detail
- A+ Core 1
- Detail
- RAID
On this page
In one line: RAID 0 stripes for speed with zero redundancy, RAID 1 mirrors for full redundancy at double the storage cost, RAID 5 and 6 stripe with one or two parity blocks to survive one or two lost drives, and RAID 1+0 stripes mirrored pairs — but none of them is a substitute for backup.
Companion to: Professor Messer, CompTIA A+ 220-1201 Core 1, objective 3.4 (RAID). The full version is the RAID class notes; the section overview is the Section 3 sheet.
RAID is not backup
Never conflate the two. RAID protects against a drive hardware failure. It does nothing against accidental deletion, corruption, or ransomware — those replicate across the array exactly as faithfully as legitimate data. A separate backup process is always required, regardless of RAID level.
RAID levels at a glance
| Level | Also called | Min drives | Survives losing | Usable capacity | Speed | CPU overhead |
|---|---|---|---|---|---|---|
| RAID 0 | Striping | 2 | 0 drives | 100% (all drives) | Fastest | None |
| RAID 1 | Mirroring | 2 | 1 (either) | 50% | Normal | None |
| RAID 5 | Striping + 1 parity | 3 | 1 | (n−1) drives | Good | Yes |
| RAID 6 | Striping + 2 parity | 4 | 2 (simultaneous) | (n−2) drives | Good | More |
| RAID 1+0 / RAID 10 | Stripe of mirrors | 4 | 1 per mirrored pair | 50% | Fast | None |
How each works
RAID 0 (striping): data split across all drives — write speed benefits from parallel writes. Lose any one drive → the whole array’s data is gone.
RAID 1 (mirroring): every byte duplicated on a second drive. Lose either drive → the other has a complete, unaffected copy. Costs double the storage.
RAID 5 (striping + parity): stripes like RAID 0, but one drive’s worth of each stripe holds parity instead of data — and which physical drive holds it rotates stripe by stripe. Lose one drive → reconstruct it from the rest + parity, in real time, running degraded. Parity math costs CPU, especially during recovery.
RAID 6 (striping + dual parity): RAID 5 plus a second independent parity block per stripe. Survives two simultaneous lost drives. The extra drive adds fault tolerance, not extra usable capacity.
RAID 1+0 / RAID 10 (nested): RAID 0 striping across mirrored pairs (“a stripe of mirrors”). Minimum 4 drives (2 pairs). Survives losing one drive from each pair simultaneously — but not both drives of the same pair at once.
Choosing a level
| Priority | Level |
|---|---|
| Max speed/capacity, data is disposable | RAID 0 |
| Simple full redundancy, budget for 2x storage | RAID 1 |
| Balanced fault tolerance + storage efficiency | RAID 5 |
| Survive two simultaneous failures | RAID 6 |
| Speed + strong redundancy, budget allows 2x storage | RAID 1+0 |
🔐 Security notes
- “We have RAID” ≠ “we have backups.” Ransomware and accidental deletion propagate through a RAID array just as reliably as real data.
- A degraded array is a live risk window, not just slower — a further failure during that window (RAID 5 on N−1, RAID 6 on N−2) can mean real data loss. Replace failed drives and monitor for degraded state.
- Encrypt deliberately, at drive level or above the controller — the choice changes what a stolen single drive exposes and how key management works across a rebuild.
- Decommission every drive in the array, not just one — even a parity-only drive holds recoverable information derived from real data.
- Rebuild load increases second-failure risk on same-age, same-wear drives — a known real-world failure pattern, and part of why RAID 6/10 exist for higher-stakes systems.
Practice drills
1. Why does RAID 0 have zero redundancy?
Data is split across drives, so each drive holds only part of every file — losing any one drive leaves the data incomplete.
2. What's the storage cost of RAID 1?
Double the storage — every byte is written twice.
3. How does RAID 5 recover from a lost drive?
Reconstructs the missing data from the remaining data plus parity, running in a degraded state with a CPU-driven performance cost.
4. RAID 6 vs RAID 5 — what's the actual gain from the extra drive?
Fault tolerance for a second simultaneous lost drive — not extra usable capacity; the added drive’s space goes entirely to parity.
5. What does RAID 1+0 combine, and what's its minimum?
RAID 0 striping across RAID 1 mirrored pairs — minimum 4 drives.
6. RAID 1+0 with 3 mirrored pairs — which specific failure combination loses data?
Both drives of the same mirrored pair failing together. Losing one drive from each of several different pairs is survivable.
7. Does RAID protect against ransomware or accidental deletion?
No. RAID protects against drive hardware failure only — corruption and deletion replicate across the array just like legitimate data.
8. RAID 5 array missing one of four drives — accessible? What next?
Yes, accessible, running degraded. Replace the failed drive promptly so it can rebuild.
Key takeaways
- RAID trades among speed, capacity efficiency and fault tolerance — no level gives you all three.
- RAID 0 = speed, zero redundancy. RAID 1 = full redundancy, half capacity.
- RAID 5 survives one drive loss via rotating parity; RAID 6 survives two, at the cost of one more drive with zero extra capacity.
- RAID 1+0 survives one loss per mirrored pair, not both drives of the same pair.
- RAID is never a substitute for backup — say it every time.