RAID (storage)
Also known as: RAID, redundant array of independent disks, RAID array
RAID (redundant array of independent disks) is a way of combining multiple drives into one storage system for better reliability, speed, or both.
RAID 1 mirrors data on two drives, so if one fails, the other keeps working. RAID 5 and RAID 6 spread data and recovery information across several drives, and RAID 10 combines mirroring and striping. It's common in servers and network storage devices. While RAID keeps systems running through drive failures, it doesn't protect against deletion, ransomware, or theft, so it's never a substitute for backups.
RAID (storage), explained at three levels
At home
RAID 1 is like writing your notes in two notebooks at the same time. If you lose one, the other has everything. Other kinds of RAID spread notes across several notebooks with clues that rebuild a missing one. But if you write something wrong, every notebook gets the mistake, which is why RAID helps when a drive breaks, not when files are deleted.
At work
RAID combines several drives in a server or network storage device so they act as one. Depending on the level, it copies data or spreads it with extra recovery information, so the system keeps running if a drive fails. Businesses care because it avoids downtime from a single drive failure, but it does not protect against deletion, ransomware or theft, so backups are still needed.
At home
RAID combines several drives so the computer sees them as one. How they combine depends on the level. RAID 1 mirrors data, writing everything to two drives, so if one dies the other has a complete copy. RAID 0 splits data across drives for speed but has no protection at all. Other levels store calculated parity information, extra data that lets the array rebuild a missing drive's contents from the remaining drives.
At work
RAID 5 spreads data and parity across three or more drives and survives one failure, while RAID 6 uses double parity to survive two. RAID 10 stripes data across mirrored pairs for speed and resilience. During a rebuild after a failure, the array reads every remaining drive in full, which can take a day or more on large drives and stresses them. Hardware controllers, software RAID and file systems such as ZFS implement RAID differently.
At home
RAID protects you from one kind of problem: a drive failing. It does nothing if you delete a file, ransomware encrypts your data, or the NAS itself is stolen or damaged, because those changes are faithfully applied to every drive. Drives bought together also tend to wear out around the same time. In my view, replace a failed RAID drive quickly, before a second one goes, and always keep a separate backup.
At work
With today's large drives, RAID 5 rebuilds expose arrays to a second failure or unreadable sector, so RAID 6, RAID 10 or erasure coding are safer for capacity arrays. Monitor drive health and alert on failures, keep spare drives on hand, and use CMR drives rated for arrays. Schedule regular scrubs to detect silent corruption, and back up RAID volumes separately, since RAID is about availability, not recovery from deletion, corruption or ransomware.
Real-world examples
At home
Thomas's home storage box used RAID 1. When one drive failed, his photos stayed available while he replaced it, though he still kept a cloud backup.
At work
Fatima's Halifax MSP found a client's server running RAID 5 with a failed drive nobody had noticed for months. She replaced it and added alerts.
How it compares
| feature | RAID (storage) | 3-2-1 backup rule |
|---|---|---|
| Protects against | Drive failure | Deletion, ransomware, disasters |
| Keeps history | No | Yes, older versions |
| Location | Same device | Multiple locations |
| Main benefit | Uptime | Recovery |
RAID keeps you running. Backups get your data back.
For analysts
When reviewing storage, I'd confirm RAID status monitoring and separate backups exist. Many small businesses assume RAID alone protects them.
History & origin
David Patterson, Garth Gibson, and Randy Katz at the University of California, Berkeley, described RAID in a 1988 paper titled 'A Case for Redundant Arrays of Inexpensive Disks,' proposing that many cheap drives could outperform expensive large ones. The industry later changed 'inexpensive' to 'independent.' RAID became standard in servers through the 1990s and in home and small business network storage devices in the 2000s. As drives grew to many terabytes, rebuilding a failed RAID 5 array took longer and carried more risk, pushing many to RAID 6 or other approaches. Newer file systems such as ZFS built similar protections into the file system itself. In my view, 'RAID is not a backup' is one of the most important sentences in IT.