RAID-Z2 Calculator
Calculate usable storage capacity for ZFS RAID-Z2 arrays. Accounts for double parity, metadata overhead, slop space, and configurable free-space reserves. Supports multiple vdevs and TB/TiB display.
Usable Capacity
102.87 TB
71.4% efficiency • 6 data drives • 2 parity drives
RAID-Z2 efficiency by drive count (single vdev)
| Drives | Usable (raw) | Efficiency |
|---|---|---|
| 4 | 36.0 TB | 50.0% |
| 5 | 54.0 TB | 60.0% |
| 6 | 72.0 TB | 66.7% |
| 8 | 108.0 TB | 75.0% |
| 10 | 144.0 TB | 80.0% |
| 12 | 180.0 TB | 83.3% |
| 16 | 252.0 TB | 87.5% |
| 20 | 324.0 TB | 90.0% |
How RAID-Z2 capacity is calculated
RAID-Z2 uses two drives per vdev for parity, similar to traditional RAID 6. The remaining drives store your data.
Raw Usable = (Drives per vdev − 2) × Drive Size × Vdevs
After Metadata = Raw Usable × (1 − metadata%)
After Slop = After Metadata × (1 − slop%)
Final = After Slop × (1 − free reserve%)
Example: 8 × 18 TB drives in a single RAID-Z2 vdev: Raw usable = 6 × 18 = 108 TB. After 1.6% metadata and 3.2% slop ≈ 102.8 TB (93.5 TiB).
Frequently asked questions
What is RAID-Z2?
RAID-Z2 is a ZFS data protection scheme that uses double parity (similar to RAID 6). It can survive two simultaneous drive failures without data loss. It requires a minimum of 4 drives per vdev, with 2 drives worth of space used for parity.
How many drives can fail in RAID-Z2?
RAID-Z2 can tolerate 2 simultaneous drive failures per vdev. If you have multiple vdevs, each can independently tolerate 2 failures. However, losing drives across different vdevs simultaneously could still result in data loss.
What is a vdev?
A vdev (virtual device) is the basic building block of a ZFS storage pool. A RAID-Z2 pool consists of one or more vdevs, each providing its own redundancy. More vdevs improve write performance but require more total drives.
Why is my usable capacity less than expected?
Beyond the 2-drive parity overhead, ZFS reserves space for metadata (typically 1.6%), slop space (~3.2% internal reservation), and you may also want a free-space reserve (commonly 10–20%) to maintain write performance. This calculator accounts for all these overheads.
What is the difference between TB and TiB?
TB (terabyte) uses base-10: 1 TB = 1,000,000,000,000 bytes. TiB (tebibyte) uses base-2: 1 TiB = 1,099,511,627,776 bytes. Drive manufacturers use TB, but operating systems often report in TiB. A "1 TB" drive shows as ~0.91 TiB in your OS.
What is ZFS slop space?
ZFS reserves approximately 3.2% (1/32) of pool capacity as "slop space" for internal metadata and operations. This space is unavailable for user data and ensures the pool continues functioning properly even when nearly full.
How many drives per vdev is optimal for RAID-Z2?
The sweet spot is typically 6–10 drives per vdev. Fewer than 6 gives poor space efficiency (50% with 4 drives). More than 12 increases rebuild times and reduces random write performance. 8 drives per vdev is a popular choice.
RAID-Z2 vs RAID-Z1 vs RAID-Z3?
RAID-Z1 tolerates 1 failure (like RAID 5), Z2 tolerates 2 (like RAID 6), and Z3 tolerates 3. For pools with 5+ drives, Z2 is recommended as rebuild times can be long enough for a second failure. Z3 is for very large pools (10+ drives).