RAID Calculator
Who can die without taking your data — the survival verdict, the usable share, and the rebuild window.
RAID Calculator
Results recalculate instantly on every keystroke. Nothing you type is transmitted.
What this result does not account for
- Classic levels 0, 1, 5, 6, 1+0 only
- No controller-specific behaviors modeled
In short: RAID 6 on 8 disks tolerates any two failures — the array shrugs, rebuilds onto a spare, and keeps 6 of 8 disks for data. RAID 1+0 answers differently: one disk per mirrored pair, up to 4 of 8 in the best case — but two losses in the same pair end it. Same steel, different promises; the level is the promise, and this page reads it honestly.
Formula
tolerated failures: 0 · 1 · 1 · 2 · one per pair —— usable disks: n · n/2 · n−1 · n−2 · n/2
Every classic level trades the same two ways: how many disks can fail before data dies, and how many disks you surrender for that promise. RAID 5 spends one disk on parity and survives one loss; RAID 6 spends two and survives two; mirroring spends half the steel and survives one loss per pair. The minimums are hard rules — three disks for RAID 5, four for RAID 6 and for 1+0. This page is the survival question; the capacity page turns the same table into terabytes.
Worked Example
- Enter the whole disk count.
- Enter the level: 0, 1, 5, 6 or 10.
- Read who can die and what survives.
- Check the rebuild window before trusting a big array.
Defaults: 8 disks, RAID 6 → any 2 can die, 6 of 8 hold data. Drive 4 disks at 1+0 → one per pair, two pairs, half the steel holding data.
Strengths & Limits Of This Model
Where this engine is strong
- Minimum-disk rules enforced, not guessed
- Best-case vs guaranteed tolerance named on 1+0
Where it stops
- No rebuild-time estimate — drive-dependent
Practical Use Cases
Server buys
level before the purchase order
Rebuild audits
how long the degraded state lasts
Nested levels
pairs, spares and odd counts
Methodology & Editorial Standards
Computation runs in IEEE-754 double precision at full internal precision; rounding to two decimal places occurs strictly at the display layer, so no cumulative drift enters the result. All monetary outputs use accounting presentation — grouped thousands, two decimals, negatives in parentheses — so figures can be transcribed directly into a model or working paper. Division-by-zero and out-of-domain inputs return an em-dash rather than a misleading number.
This engine was reconciled against an independent reference implementation and hand-verified for the worked example above before release. Our full five-stage review process is published on the About Us page.
Disclaimer. This calculator is provided for informational and modelling purposes only and does not constitute financial, tax, legal, medical, or engineering advice. Verify all figures with a qualified professional before acting on them.
RAID Calculator — 8 Expert FAQs
8 analyst-written answers to the questions practitioners actually ask — optimised for voice and answer-engine retrieval.
Which RAID level survives the most failures?
RAID 6 tolerates any two disks failing at once — two independent parity blocks protect every stripe. RAID 1+0 can survive more in total (one per mirror pair) but conditional on where the failures land: two losses in the same pair end the array. Guaranteed-versus-best-case is the distinction this page prints.
How many disks does each level need?
RAID 0 and RAID 1 need 2; RAID 5 needs 3; RAID 6 and RAID 1+0 need 4. Below the minimum the layout simply does not exist — the page refuses rather than guessing. An odd disk on 1+0 idles as a spare in most controllers.
What does RAID 1 on more than 2 disks mean?
A mirror set extended across N disks keeps N full copies — you can lose N−1... in practice controllers cap mirror width, and this page prices the honest arithmetic: half the steel holds data and one loss per pair is survivable. For capacity at width, most buyers want 1+0.
Why is the rebuild window dangerous?
A rebuild reads every surviving disk for hours or days on modern drive sizes — the array runs degraded and every remaining disk works at full tilt. A second failure during rebuild is the classic unrecoverable loss, which is why RAID 6 exists and why big arrays keep hot spares.
Is RAID a backup?
No — redundancy protects against hardware death, not deletion, ransomware or a controller eating the array. The retention storage page prices the backup copies that answer the other failure modes. RAID buys uptime; backup buys history.
What does RAID 0 buy if it protects nothing?
Striping buys speed and full capacity — every disk holds unique data and reads spread across all of them. The price is the survival verdict this page prints: one loss ends the array. Scratch work and rebuildable staging are the honest uses.
How do I choose between RAID 6 and 1+0?
Capacity efficiency and two-failure tolerance favor RAID 6; write performance and fast, simple rebuilds favor 1+0. Databases usually pay the mirror tax; bulk archives take the parity discount. The capacity page shows exactly how many terabytes each promise costs.
What happens with an odd disk count on 1+0?
Mirrors come in pairs, so one disk sits outside the stripes — most controllers use it as a dedicated hot spare. The page says so rather than pretending half of an odd count is a mirror.