RAID-Z Calculator: usable ZFS capacity for TrueNAS
By Devin Chua · Data checked by Owen Nakamura · Updated With 6 × 20 TB in RAID-Z2 you get about 72.8 TiB after parity (80 TB in drive-maker terms), and roughly 70.5 TiB after the ZFS reserve — the figure TrueNAS actually reports. Two drives go to parity, so two can fail without data loss. ZFS sizes the vdev to the smallest drive and has no SHR, so mixed drive sizes cost real capacity. Set your drives below; the calculator shows Z1, Z2 and Z3 with the cost to fill in Canadian dollars.
As an Amazon Associate we earn from qualifying purchases on Amazon.ca.
RAID-Z2 (double parity)
Multiple vdevs split the drives into equal groups. Each group carries its own parity, so fault tolerance rises and usable capacity falls.
That is 80 TB in drive-maker terms. After the ZFS reserve of about 2.5 TB, in practice 70.49 TiB remains. That smaller number is what you see later in TrueNAS.
Drives: from CA$5,970 for 6 × 20 TB (CA$49.75/TB overall)
Why a dedicated RAID-Z calculator
The vendor calculators stop at classic RAID and SHR; none of them does ZFS. Yet RAID-Z is what most self-built NAS boxes and every TrueNAS install actually run, and its capacity maths differs from RAID 5/6 in one important way: ZFS holds back a reserve. That reserve is why the number you compute on paper is never quite the number the pool reports, and almost no tool shows it. This one does.
How RAID-Z capacity is computed
- RAID-Z1: (drives per vdev minus 1) × smallest drive. One drive may fail.
- RAID-Z2: (drives per vdev minus 2) × smallest drive. Two may fail.
- RAID-Z3: (drives per vdev minus 3) × smallest drive. Three may fail.
- The ZFS reserve: about 1/32 of the pool is held back so it stays writable when full. The reported figure is lower again by that much.
Pick the parity level for the rebuild, not just the failure count
The real argument for RAID-Z2 over Z1 is the resilver. When a drive fails, ZFS reads every remaining drive in full to rebuild — and large modern drives make that a long, heavy operation during which a second failure would end a Z1 pool. With drives of 16 TB and up, or vdevs past about six drives, RAID-Z2 is the calm default and RAID-Z3 the choice for very wide vdevs of archival data.
Mixed drive sizes and ZFS
ZFS has no flexible RAID. A RAID-Z vdev sizes to its smallest member, so a 20 TB drive added to a vdev of 16 TB drives contributes only 16 TB until the rest are upgraded. If you want to use mismatched drives fully, Unraid is the system that does that; the RAID calculator covers SHR for Synology.
RAID-Z is still not a backup
RAID-Z survives one, two or three drives failing and nothing else. ZFS adds real protection against silent corruption through checksums and scrubs, which is a genuine reason to prefer it, but a deleted dataset, a bad pool import or a lightning strike still takes the whole pool. Keep a second copy elsewhere. For sizing memory for that pool, use the TrueNAS RAM calculator.
Frequently asked questions
How much storage is left with six 20 TB drives in RAID-Z2?
About 80 TB on paper, roughly 72.8 TiB: RAID-Z2 gives two of the six drives to parity, leaving four times 20 TB. ZFS then holds back about one thirty-second as reserve so the pool stays writable when full, so TrueNAS shows around 70.5 TiB in the end.
What is the difference between RAID-Z1, Z2 and Z3?
The number of parity drives. RAID-Z1 survives one drive failing, RAID-Z2 two, RAID-Z3 three. Each parity level costs one whole drive of usable capacity. With six 20 TB drives: Z1 gives 100 TB, Z2 gives 80 TB, Z3 gives 60 TB.
Does ZFS support SHR or mixed drive sizes?
No. ZFS sizes a RAID-Z vdev to the smallest drive, exactly like RAID 5 and 6, and it has no equivalent of Synology's SHR. Mixed drive sizes cost real capacity here — the surplus on your larger drives is simply unused until you replace the smaller ones.
What is the ZFS reserve (slop space)?
ZFS keeps roughly 1/32 of the pool in reserve so it stays writable even when nearly full. That is why the figure TrueNAS reports is a little lower than the raw parity maths suggests. This calculator shows both: the capacity after parity, and the smaller net figure after the reserve.
How many drives should a RAID-Z1 vdev have?
Keep RAID-Z1 to about five drives or fewer. During a resilver, ZFS reads every remaining drive in full, which is exactly when a second failure shows up. Past about six drives, RAID-Z2 is the sensible choice, and for very wide vdevs RAID-Z3.
Does this calculator apply to TrueNAS and Proxmox ZFS?
Yes. RAID-Z capacity maths is the same wherever OpenZFS runs — TrueNAS CORE and SCALE, Proxmox, plain Linux or FreeBSD. Multiple vdevs split the drives into equal groups; each group carries its own parity, so fault tolerance rises and usable capacity falls.
Devin Chua works out which drives, RAM and NVMe cache fit which NAS model at nasdrives.ca, and what the RAID choice means for usable capacity, checked against what is in stock on Amazon.ca.
Data checked by Owen Nakamura, Technical editor (data checking)