Plan RAID like a pro. This free RAID Calculator helps you size arrays and compare trade-offs across RAID 0/1/1E/5/6/10/50/60. Enter disk count, per-disk capacity and performance, choose your disk type (HDD/SSD/NVMe), and instantly see usable capacity, overhead, fault tolerance, write penalty, mixed IOPS, read/write throughput, rebuild time, and an estimated URE (Unrecoverable Read Error) risk during rebuild. Results are computed locally in your browser—no data is sent to our servers.

Highlights:

• Supports group sizing for RAID 50/60 to reflect how parity groups affect performance and resilience.
• Uses conservative presets for common drive classes, which you can override at any time.
• Shows decimal TB (10¹²) capacity to match how vendors label drives (you can mentally convert to TiB if needed).
• URE risk is an estimate; real controllers perform re-reads/ECC/scrubs, so field results can be better than the theoretical model.

Tip: As arrays grow, RAID 6/60 often provides a safer balance than RAID 5/50 due to dual parity, while RAID 10 favors latency and predictable performance at the cost of capacity.

FAQ

What is URE risk during rebuild?
It’s the chance of hitting an unrecoverable read error while the controller is reading surviving disks to reconstruct a failed one. Higher capacity disks, larger arrays, and single-parity levels (RAID 5/50) increase this risk.

Why do RAID 50/60 ask for “disks per group”?
Because performance and resilience are determined per parity group. The calculator models per-group parity (one for RAID 5, two for RAID 6) and scales the totals by the number of groups formed.

What is write penalty?
Parity and mirroring require extra I/O per write. Typical penalties: RAID 1/10 = ×2, RAID 5 = ×4, RAID 6 = ×6. This affects sustained random write IOPS.

RAID 1 vs. RAID 1E vs. RAID 10 — which one?
RAID 1 mirrors in pairs; RAID 1E mirrors across an odd number of disks; RAID 10 stripes across mirrored pairs. RAID 10 generally offers the best predictable latency and rebuild characteristics, at the cost of capacity efficiency.

Decimal TB vs. TiB — why do numbers look bigger?
Vendors use decimal terabytes (1 TB = 10¹² bytes). Operating systems often show tebibytes (1 TiB = 2⁴⁰ bytes ≈ 1.0995 TB). The calculator uses decimal TB for consistency with drive labels.

HDD vs. SATA/SAS SSD vs. NVMe — which preset should I use?
Use the disk class closest to your hardware, then adjust IOPS/MB/s to your device specs or benchmarks. NVMe offers higher throughput and IOPS; SAS SSDs often provide better sustained mixed workloads than SATA SSDs.

Is this calculator accurate for every controller?
It’s an engineering estimate. Real-world results vary with controllers, caches, queue depths, workloads, firmware, and background tasks (scrubs, patrol reads).