Guide
RAID 0, 1 or 5: which one makes sense for home use.
RAID combines several drives to gain speed, redundancy, or both — but each level makes a different trade-off, and choosing wrong can leave you with less protection than you think you have.
What RAID is, in one sentence
RAID (Redundant Array of Independent Disks) is a method for combining two or more physical disks so the operating system sees them as a single unit, gaining speed, redundancy, or both — depending on how data is spread across the disks. Not every RAID level offers protection against failures; some prioritize pure performance, sacrificing safety.
RAID 0: speed, with no protection at all
RAID 0 splits data across the available disks (striping) without duplicating anything, which allows parallel reads and writes and delivers the best performance of the three levels. Usable capacity is the sum of all disks. The catch: if a single disk in the array fails, you lose the data on every disk, not just the one that failed. RAID 0 offers no redundancy at all — in practice, it's the opposite of a safe backup.
RAID 1: simple redundancy through mirroring
RAID 1 writes an identical copy of the data to every disk in the array (mirroring). If one disk fails, the other keeps working with no data loss, and you can replace the damaged disk without interrupting access to the information. Usable capacity equals that of the smallest disk in the pair (with two 4 TB disks, you get 4 TB usable, not 8). It's the simplest and most reliable option for protecting data you can't afford to lose, like family photos or important documents.
RAID 5: the balance between capacity and protection
RAID 5 spreads data across multiple disks along with parity information, which lets you rebuild a failed disk's data using the information on the remaining disks. It needs at least 3 disks, and usable capacity is (number of disks − 1) × the smallest disk's capacity — with four 4 TB disks, you get 12 TB usable, not 16. It's the most space-efficient of the three options for anyone who needs both protection and good use of capacity, though write performance is somewhat lower than RAID 0 and rebuilding after a failure can take hours.
Quick comparison
| Level | Minimum disks | Usable capacity | Fault tolerance | Best for |
|---|---|---|---|---|
| RAID 0 | 2 | 100% (sum of all) | None | Pure speed, non-critical data |
| RAID 1 | 2 | 50% (one disk's capacity) | 1 disk | Critical data, simplicity |
| RAID 5 | 3 | (n-1)/n of total | 1 disk | Capacity/protection balance |
Why RAID isn't a backup
This is the most common and most dangerous misunderstanding about RAID: it protects against a disk's physical failure, but not against any other kind of data loss. If you accidentally delete a file, that deletion instantly replicates across every disk in the RAID. If ransomware encrypts your files, it encrypts the copy on every disk too. If there's a fire or theft, you lose the entire array at once, no matter how many disks it had. RAID solves availability against hardware failure; a real backup (ideally following the 3-2-1 rule) solves everything else — see the full picture in the backup guide.
Verifiable sources
Technical definitions of each RAID level, their usable capacity and fault tolerance are documented by TechTarget, a standard reference source for enterprise storage infrastructure documentation.