RAID (Redundant Array of Independent Disks)
Definition
RAID is a data storage virtualization technology that combines multiple physical disk drives into a single logical unit for performance, redundancy, or both. The term was originally “Redundant Array of Inexpensive Disks,” later changed to “Independent.”
RAID levels define how data is distributed across disks, with different trade-offs between performance, capacity, and fault tolerance.
Common RAID Levels
| Level | Name | Min Disks | Redundancy | Performance | Use Case |
|---|---|---|---|---|---|
| 0 | Striping | 2+ | None | Read/Write fastest | Performance, no fault tolerance |
| 1 | Mirroring | 2 | 1 disk | Read fast, Write same | Small volumes, boot drives |
| 5 | Striping + Parity | 3+ | 1 disk | Read fast, Write slower | General-purpose storage |
| 6 | Striping + Dual Parity | 4+ | 2 disks | Read fast, Write slower | Critical data, 2-disk failure tolerance |
| 10 | Mirror + Stripe | 4+ | Half disks | Read/Write fast | High-performance, redundant storage |
| 50 | Stripe + Stripe | 6+ | 1 per stripe | High performance | Large-scale storage arrays |
Key Concepts
- Hot spare: Unused disk that automatically rebuilds if another fails
- Rebuild: Reconstructing data on replacement disk after failure
- ZFS/Btrfs: Software RAID alternatives with additional features (snapshots, compression)
- Hardware RAID: Dedicated RAID controller card
- Software RAID: OS-level implementation (Linux mdadm, ZFS)
Related Terms
- Nas — different from RAID; RAID protects against disk failure, not data loss
References
- Wikipedia: https://en.wikipedia.org/wiki/RAID
- Linux RAID Wiki: https://raid.wiki.kernel.org/