Tech calculator
RAID Capacity Calculator
Compare raw and usable storage for RAID 0, 1, 5, 6, and 10 configurations.
Usable capacity
6 TB
How RAID capacity depends on drive count and level
RAID (Redundant Array of Independent Disks) distributes data and redundancy across multiple drives. The calculator accepts a number of drives (a whole number with no decimals), the smallest drive size in TB, and a RAID level of 0, 1, 5, 6, or 10. The drive count must meet minimum requirements per level: RAID 0 and 1 require at least 2 drives, RAID 5 requires at least 3, RAID 6 requires at least 4, and RAID 10 requires at least 4 and must be even (pairs of drives). The calculator validates all inputs before computing usable and raw capacities.
The main output shows usable capacity: the space available for user data after accounting for redundancy. Raw capacity is the total of all drives. The difference between raw and usable is capacity reserved for redundancy and parity. RAID 0 sacrifices redundancy for maximum space; RAID 1 and 10 mirror data using half the drives; RAID 5 and 6 use parity blocks to recover from drive failures.
Worked example: 4 drives of 2 TB each in RAID 5
Four 2 TB drives provide 4 × 2 = 8 TB of raw storage. In RAID 5, the formula subtracts one drive for parity: (4 − 1) × 2 = 3 × 2 = 6 TB usable. The remaining 8 − 6 = 2 TB is distributed as parity blocks across all drives, allowing recovery if any single drive fails.
RAID 5 stores one redundant parity block per stripe. With 4 drives, if drive 2 fails, the data on drives 1, 3, and 4 plus parity reconstruct the missing drive. The cost is 25 percent of raw capacity (2 TB out of 8 TB) given away to redundancy. If space was the only concern, RAID 0 would offer the full 8 TB, but any drive failure loses all data.
RAID levels compared: trade-offs between redundancy and capacity
RAID 0 (striping, no redundancy) uses all drive capacity. Four 2 TB drives yield 8 TB usable. One drive failure loses the entire array. Used for cache, temporary data, or high-performance non-critical workloads where speed and capacity are worth the risk.
RAID 1 (mirroring) and RAID 10 (mirrored pairs) use half the capacity. Four 2 TB drives yield 4 TB usable in RAID 1 (two active mirrors) or RAID 10 (two mirrored pairs). Any single drive can fail without data loss. Used for critical databases and active systems where redundancy is essential, and the performance overhead of checking two drives is acceptable.
RAID 5 uses (N−1)/N of capacity for N drives. Four drives give 75 percent usable (6 TB), five drives give 80 percent (8 TB). Rebuilding a failed drive requires reading from all other drives and recalculating parity, taking many hours for large drives. RAID 6 adds a second parity block, allowing two simultaneous failures. Four drives yield 50 percent usable (4 TB); six drives yield 66 percent (8 TB). The extra parity costs capacity but adds safety during the long rebuild window.
Real-world RAID configurations with capacity examples
- 2 × 4 TB drives in RAID 1: 4 TB usable, 8 TB raw, 4 TB redundancy. Typical for critical single-server backups.
- 3 × 6 TB drives in RAID 5: 12 TB usable, 18 TB raw, 6 TB parity. Common for small NAS systems.
- 4 × 8 TB drives in RAID 10: 16 TB usable, 32 TB raw, 16 TB mirroring. Suitable for active database clusters.
- 6 × 10 TB drives in RAID 6: 40 TB usable, 60 TB raw, 20 TB parity. Supports two simultaneous drive failures during rebuild.
- 8 × 2 TB drives in RAID 5: 14 TB usable, 16 TB raw, 2 TB parity. Large array with many drive slots but smaller individual capacity.
Frequently asked questions
Why does RAID 6 use twice as much capacity for redundancy as RAID 5?
RAID 5 stores one parity block per stripe. RAID 6 stores two independent parity blocks (P and Q), doubling the overhead. Four 6 TB drives yield 18 TB usable in RAID 5 but only 12 TB in RAID 6 because two drives' worth of space is reserved. RAID 6 protects against two simultaneous failures.
If a drive fails, can I replace it and get my data back?
Yes, in RAID 1, 5, 6, and 10. The remaining drives contain enough parity or mirrored data to rebuild the failed drive. RAID 0 has no redundancy, so one failure loses everything. Rebuild time depends on drive size: a 10 TB drive might take 10 to 30 hours to rebuild, during which the array is vulnerable to a second failure.
Should I use mismatched drive sizes in a RAID array?
No. RAID capacity is limited by the smallest drive to prevent data imbalance. A 4-drive array with one 2 TB and three 4 TB drives provides only 6 TB usable in RAID 5 (3 × 2 TB), wasting the extra space on the larger drives. Use identical drives when possible.
Does formatted capacity equal the usable capacity the calculator shows?
No. Usable capacity shown here is the raw RAID space available. Formatting introduces overhead: a typical filesystem reserves 5 to 10 percent for metadata and inode tables. A 6 TB RAID array might format to only 5.4 TB. Check the filesystem type for exact overhead.
Is RAID a backup solution?
No. RAID protects against hardware failure, not data loss from accidental deletion, corruption, or malware. A ransomware infection affects all drives simultaneously. Separate backups to external storage or cloud services are essential for data protection.
Can I grow a RAID array by adding more drives?
Some RAID levels support expansion. RAID 5 and 6 can often be expanded to a larger array through online resize (adding drives one at a time and rebalancing). RAID 0, 1, and 10 offer more limited expansion options depending on controller and firmware. Check your hardware specifications before assuming growth is possible.