RAID
What Is RAID?
RAID—Redundant Array of Independent Disks—is a method of combining multiple physical drives into a logical storage configuration. Depending on the RAID level, drives can be striped for performance, mirrored for redundancy, protected with parity, or combined using nested RAID architectures.
Different RAID levels make different tradeoffs between usable capacity, performance, fault tolerance, drive count, rebuild behavior and hardware cost. There is no single RAID level that is best for every workload.
Compare RAID Configurations
Calculate usable capacity, storage efficiency, fault tolerance, hot-spare impact, write penalty, estimated performance and hardware cost.
Open RAID Calculator →Understand the Building Blocks of RAID
Most RAID architectures are built from three core concepts: striping, mirroring and parity.
Striping
Data is distributed across multiple drives so storage I/O can be spread across the array. RAID 0 is the simplest striped RAID level.
Mirroring
Data is duplicated across drives. RAID 1 uses mirroring, while RAID 10 combines mirrored pairs with striping.
Parity
Parity provides redundant information that can reconstruct missing data after supported drive failures. RAID 5 and RAID 6 are common parity RAID levels.
Explore Individual RAID Levels
Learn how each RAID architecture handles capacity, redundancy, performance and drive failures.
RAID 0
Striping without redundancy for maximum usable capacity and parallel I/O.
Explore RAID 0 →RAID 1
Mirrored drives that maintain redundant copies of the same data.
Explore RAID 1 →RAID 5
Striping with distributed single parity and one-drive fault tolerance.
Explore RAID 5 →RAID 6
Dual distributed parity that can tolerate two member-drive failures.
Explore RAID 6 →RAID 10
Stripes across mirrored pairs for strong write performance and mirror-based recovery.
Explore RAID 10 →RAID 50
RAID 0 striping across multiple RAID 5 parity groups.
Explore RAID 50 →RAID 60
RAID 0 striping across multiple dual-parity RAID 6 groups.
Explore RAID 60 →RAID Level Comparison
| RAID | Minimum Drives | Usable Capacity | Fault Tolerance | Architecture |
|---|---|---|---|---|
| RAID 0 | 2 | N × smallest drive | None | Striping |
| RAID 1 | 2 | One mirror member's usable capacity | At least 1 member in a two-drive mirror | Mirroring |
| RAID 5 | 3 | (N − 1) × smallest drive | 1 drive | Striping + single parity |
| RAID 6 | 4 | (N − 2) × smallest drive | 2 drives | Striping + dual parity |
| RAID 10 | 4 | Approximately 50% of raw capacity | Depends on mirror-pair failure placement | Striped mirrors |
| RAID 50 | 6 | One parity-equivalent per RAID 5 group | 1 drive per RAID 5 group | Stripe across RAID 5 groups |
| RAID 60 | 8 | Two parity-equivalents per RAID 6 group | 2 drives per RAID 6 group | Stripe across RAID 6 groups |
Compare RAID Architectures
Side-by-side comparisons make it easier to evaluate capacity, fault tolerance, write overhead and rebuild behavior.
RAID 5 vs RAID 6
Compare usable capacity, single vs dual parity, write overhead and protection during rebuilds.
Compare RAID 5 & RAID 6 →RAID 6 vs RAID 10
Compare capacity efficiency, write performance, fault tolerance and rebuild characteristics.
Compare RAID 6 & RAID 10 →RAID 50 vs RAID 60
Compare striped RAID 5 and RAID 6 groups for capacity, performance and group-level protection.
Compare RAID 50 & RAID 60 →Go Beyond RAID Capacity
RAID planning should also account for write overhead, degraded-array behavior, rebuild time and recovery resources.
RAID Write Penalty
Learn why RAID 5, RAID 6 and RAID 10 require different numbers of back-end operations for small writes.
Understand Write Penalty →RAID Rebuilds
Understand degraded arrays, reconstruction, rebuild time, performance impact and RAID-level differences.
Learn About RAID Rebuilds →RAID Hot Spares
Learn how global and dedicated hot spares can reduce the delay before a RAID rebuild begins.
Learn About Hot Spares →Architecture & Data Protection
Hardware RAID vs Software RAID
Compare dedicated RAID controllers with operating-system RAID for performance, cache, CPU usage, portability, NVMe support and cost.
Compare Hardware & Software RAID →RAID Is Not a Backup
Understand the difference between drive redundancy, backups, snapshots, ransomware protection and disaster recovery.
Learn RAID vs Backup →What Should You Consider Before Choosing RAID?
The correct RAID configuration depends on more than drive count. Consider usable capacity requirements, workload type, read/write ratio, latency sensitivity, acceptable drive-failure tolerance, rebuild behavior, hot-spare strategy, controller capabilities, drive cost and independent backup requirements.
RAID Calculator
Enter your drive count and drive capacity to compare RAID configurations. Evaluate usable capacity, parity overhead, storage efficiency, fault tolerance, hot-spare impact, estimated performance and hardware cost before configuring your server or storage system.
RAID FAQ
What does RAID stand for?
RAID commonly stands for Redundant Array of Independent Disks. It combines multiple drives into logical storage configurations with different performance, capacity and redundancy characteristics.
Which RAID level is best?
There is no universally best RAID level. The correct configuration depends on performance requirements, capacity, drive count, fault tolerance, rebuild strategy and workload.
Which RAID levels provide redundancy?
RAID 1, RAID 5, RAID 6, RAID 10, RAID 50 and RAID 60 provide redundancy. RAID 0 does not provide drive-failure protection.
Does RAID improve performance?
RAID can improve read or write performance by distributing I/O across multiple drives, but results depend on RAID level, drives, controller, cache and workload.
What happens when a RAID drive fails?
A redundant RAID level may continue operating in a degraded state. A replacement drive or hot spare can then be used to rebuild the missing data.
Is RAID a backup?
No. RAID provides storage redundancy and availability, while backups provide independent recovery copies. Important data should use both when appropriate.
RAID Protects Availability—Not Every Form of Data Loss
RAID redundancy can protect against supported drive failures, but it does not replace independent backups. Accidental deletion, ransomware, application corruption, filesystem damage, catastrophic system loss and failures beyond the RAID level's tolerance can still result in data loss.
Learn Why RAID Is Not a Backup →