What Is RAID?
RAID (Redundant Array of Independent Disks) combines multiple physical hard drives into a single logical volume to improve performance, reliability, or both. Different RAID levels make different trade-offs between usable capacity, redundancy, and speed. The right RAID level depends on what you're storing, how much downtime you can tolerate, and your budget. RAID is widely used in network-attached storage (NAS) devices like Synology and QNAP, in homelabs running TrueNAS or UnRAID, and in enterprise servers.
How to Choose the Right RAID Level
- 2 drives, mirrored backup: Use RAID 1 for a simple one-to-one mirror. 50% efficiency, survives one failure.
- 3–5 drive NAS: RAID 5 gives the best balance of capacity (~67–80% efficiency) and one-drive fault tolerance. See our RAID levels guide for detailed comparisons.
- 6+ drive arrays: RAID 6 is strongly recommended. Rebuilds with modern 16TB+ drives can take days, and RAID 6 protects against a second failure during that window. This is especially critical for Plex media servers running 24/7.
- Database / VM workloads: RAID 10 delivers excellent random I/O at the cost of 50% capacity efficiency.
- Scratch / cache: RAID 0 maximizes speed and capacity but offers no redundancy — use only with separate backups.
Why Rebuild Time Matters
When a drive in a RAID 5 or RAID 6 array fails, the array enters a degraded state and must reconstruct the missing data from parity once a replacement drive is installed. With drives in the 16TB to 24TB range, this rebuild can take 24 to 96 hours of continuous reading from every other drive in the array. During this period, those drives are under sustained heavy load, which is precisely when a second failure becomes most likely. This is the core reason RAID 6 (which can tolerate two failures) is the recommended choice for any modern array of six or more large drives. Use this calculator to estimate rebuild times for your specific configuration — knowing the rebuild window helps you decide between RAID 5, 6, or 10.
RAID Is Not a Backup
RAID protects against hardware failure, not against the most common causes of data loss: accidental deletion, ransomware, software bugs, controller failure, fire, or theft. The 3-2-1 backup rule remains the gold standard: three copies of your data, on two different media types, with at least one copy stored offsite or in the cloud. Use RAID for uptime and to reduce the pain of a single drive failure — not as your only line of defense.
CMR vs SMR in RAID Arrays
Always use CMR (Conventional Magnetic Recording) drives in any RAID configuration. SMR (Shingled Magnetic Recording) drives use overlapping tracks that require background rewrite operations, which can cause sustained-write performance to collapse to a few MB/s. In a RAID rebuild, this often results in stalls that last days or fail entirely, potentially destroying the array. Stick to enterprise-class NAS drives (WD Red Plus/Pro, Seagate IronWolf/IronWolf Pro, Toshiba N300, WD Ultrastar, Seagate Exos) which are all CMR.