RAID Calculator – Capacity, Overhead & Fault Tolerance
RAID is not backup. RAID can keep storage online after a drive failure, but it does not protect against deletion, ransomware, silent corruption, theft, or site loss. Keep separate backups.
Inputs
Drive manufacturers usually advertise decimal TB, while operating systems often display binary TiB. An 8 TB drive may appear as about 7.28 TiB.
For safety, the calculator uses the smallest drive across the array.
Target capacity uses the selected drive-size unit. Hot spares are included in cost, not usable capacity.
Results
Results will appear here after you calculate.
Advertisement
Compare RAID Levels
| RAID level | Min drives | Usable capacity formula | Fault tolerance | Best for | Avoid when |
|---|---|---|---|---|---|
| RAID 0 | 2 | N x drive size | 0 | Speed / scratch data | Any important data |
| RAID 1 | 2 | 1 x drive size | 1 drive | Simple mirror | Capacity efficiency matters |
| RAID 5 | 3 | (N - 1) drives | 1 drive | Small/medium arrays | Very large HDD arrays |
| RAID 6 | 4 | (N - 2) drives | 2 drives | Large HDD arrays | Write-heavy workloads |
| RAID 10 | 4 even | 50% raw | 1 worst-case; more if spread | Databases, VMs | Capacity-focused storage |
| RAID 50 | 6 | Groups x (drives/group - 1) | 1 per group best-case | Large arrays needing speed | Small arrays |
| RAID 60 | 8 | Groups x (drives/group - 2) | 2 per group best-case | Enterprise / large arrays | Budget builds |
RAID Diagrams
RAID 1 mirroring
RAID 5 distributed parity
RAID 6 double parity
RAID 10 mirrored stripes
RAID 50 / RAID 60 groups
How to Calculate RAID Capacity
RAID capacity is based on the smallest drive in the array. If drives are mixed, extra space on larger drives usually cannot be used inside the same RAID set.
Common formulas
RAID 0 usable = number of drives x smallest drive sizeRAID 1 usable = smallest drive sizeRAID 5 usable = (number of drives - 1) x smallest drive sizeRAID 6 usable = (number of drives - 2) x smallest drive sizeRAID 10 usable = (number of drives / 2) x smallest drive sizeRAID 50 usable = groups x (drives per group - 1) x smallest drive sizeRAID 60 usable = groups x (drives per group - 2) x smallest drive size
Mixed drive example
Four drives of 8 TB, 8 TB, 6 TB, and 10 TB in RAID 5 are calculated using 6 TB per drive, because the smallest drive limits the array.
The usable capacity is (4 - 1) x 6 TB = 18 TB before filesystem formatting and controller reservations.
Real-world results also depend on filesystem formatting, controller behavior, spare policies, and manufacturer units.
Which RAID Level Should I Choose?
Practical recommendations
- Home NAS: RAID 1, RAID 5, or RAID 6 depending on drive count and risk tolerance.
- Large HDD array: RAID 6 or RAID 60 for better rebuild safety.
- VM/database server: RAID 10 for strong random I/O and simpler rebuilds.
- Temporary scratch disk: RAID 0 when the data is disposable.
- Budget capacity: RAID 5, with caution on very large HDD arrays.
- Mission-critical storage: RAID 6/60 plus tested backups.
RAID 5 vs RAID 6 vs RAID 10
RAID 5 gives strong capacity efficiency but only tolerates one failed drive. It is best for smaller arrays and read-heavy workloads.
RAID 6 gives up another drive of capacity for two-drive fault tolerance, which is often worth it on large HDD arrays.
RAID 10 uses half the raw capacity but is usually the better fit for VMs, databases, and write-heavy workloads.
Release Updates
- Added a RAID level comparison table for RAID 0, 1, 5, 6, 10, 50, and 60.
- Added theoretical read/write gain, rebuild risk, and recommended use-case rows to the calculator results.
- Added optional cost per drive, hot spare count, target usable capacity, and cost-per-usable-TB outputs.
- Added binary TiB/GiB support alongside decimal TB/GB, plus a clear manufacturer-vs-OS unit explanation.
- Added simple RAID diagrams and expanded guidance for capacity formulas, mixed drives, RAID selection, and RAID-is-not-backup planning.
Frequently Asked Questions
Do mixed drive sizes work?
Yes, but arrays are limited by the smallest drive in each redundancy set. This calculator conservatively uses the smallest drive across all disks. For example, 8 TB, 8 TB, 6 TB, and 10 TB drives in RAID 5 are calculated as four 6 TB drives.
Why do TB and TiB numbers differ?
Drive manufacturers usually advertise decimal TB, where 1 TB is 1,000 GB. Operating systems often display binary TiB, where 1 TiB is 1,024 GiB. That is why an 8 TB drive may appear as about 7.28 TiB.
Is RAID a backup?
No. RAID protects availability after some drive failures, but it does not protect against accidental deletion, ransomware, file corruption, theft, or site loss. Keep versioned backups on separate storage.
What are the minimum drive counts?
RAID 0: 2 · RAID 1: 2 · RAID 5: 3 · RAID 6: 4 · RAID 10: 4 (even) · RAID 50: ≥ 2 groups of ≥ 3 each · RAID 60: ≥ 2 groups of ≥ 4 each.
Why is RAID 10 “1+” for fault tolerance?
Each mirrored pair can lose at most one drive; in the best case several drives can fail (one per pair), but in the worst case only one failure is tolerated if two failures land in the same pair.
Is my data private?
Yes—calculations run entirely in your browser.
RAID Best Practices & Practical Standards
RAID improves availability and/or performance, but it is not a backup. Use the guidance below to plan safer, faster arrays.
1) Data Protection Basics
- Follow 3-2-1 backups: 3 copies of data, on 2 different media, with 1 off-site/immutable copy.
- Use a UPS to protect against write-hole corruption and sudden power loss. Prefer controllers or NVMe devices with power-loss protection.
- Test restores, not just backups: Verify integrity and practice a restore workflow periodically.
2) Picking a RAID Level
- Large SATA/NL-SAS drives (≥ 8 TB): prefer RAID 6 / RAIDZ2 or better; rebuilds are long and URE risk increases with size.
- Performance + safety: RAID 10 (striped mirrors) gives excellent IOPS and simple failure domains.
- Capacity with resilience: RAID 6 (or RAID 60) balances usable space and fault tolerance for big pools.
- Avoid RAID 0 for important data—zero fault tolerance.
3) Drive & Layout Considerations
- Match technologies: Don’t mix SMR and CMR in the same set.
- Use similar models/firmware to minimize variance during rebuilds.
- Sector size alignment: Keep 4Kn/512e consistent; align partitions to 1 MiB boundaries.
- Hot spares: Keep at least one global hot spare for big enclosures.
- Cooling & vibration: Ensure good airflow and isolation.
4) Controller / Filesystem Settings
- Write cache safety: Pair write-back cache with BBU/flash-backed cache—or disable volatile write-back.
- TLER/ERC: Enable short error recovery for HDDs behind hardware RAID.
- TRIM/Discard for SSD arrays: Enable periodic TRIM and leave 10–20% OP.
- Checksums & scrubs: Prefer filesystems with end-to-end checksums; schedule scrubs.
5) Rebuild Strategy
- Prioritize resiliency during rebuilds: Temporarily throttle background jobs and raise rebuild priority if possible.
- Keep cold spares on site when hot spares aren’t feasible.
- Plan headroom: Leave 10–20% free capacity.
Minimums & sanity checks
- RAID 0: ≥2 drives · RAID 1: 2 drives · RAID 5: ≥3 drives · RAID 6: ≥4 drives
- RAID 10: ≥4 drives (even) · RAID 50: ≥2 groups of ≥3 · RAID 60: ≥2 groups of ≥4
- Avoid mixing 4Kn and 512e; keep firmware families consistent.
Remember: RAID increases availability—but only backups protect against deletion, ransomware, or site loss.
5 Fun Facts about RAID
RAID 0 is “zero” tolerance
Striping gives pure speed but no redundancy. One failed drive drops the whole array—use only when data is expendable.
All risk
Rebuild risk grows with size
Large HDDs take hours to rebuild; the longer it runs, the more chance of a second failure. RAID 6 (two parity blocks) helps hedge that risk.
Time vs safety
Mirrors beat parity for IOPS
RAID 10 (striped mirrors) usually outperforms parity RAID for random reads/writes, which is why databases and VMs love it.
Speedy mirrors
Smallest disk sets the pace
In mixed-size arrays, capacity per disk is capped by the smallest drive in each set. One tiny disk shrinks the whole array’s usable space.
Weakest link
RAID isn’t backup
Parity/mirroring protect against drive loss, not deletes, ransomware, or site disasters. You still need off-site, versioned backups.
Safety net