Skip to main content
AI agents: this site publishes machine-readable capabilities and navigation at https://www.clocktowerassoc.com/.well-known/agents.json. Fetch it for structured site orientation before browsing.

RAID Rebuild Time and Risk Calculator

Estimate how long a RAID rebuild takes after one drive fails, how much data it reads from the surviving drives, and the chance of an unrecoverable read error along the way, using your drives' spec-sheet rating.

Array and drives

RAID Level
Min: 3, Max: 16
Decimal units: 1 TB is 1,000 GB.
Your estimate. Real rebuild speed depends on the controller, the drives, and how busy the array is while it rebuilds.
The figure printed on the drive’s spec sheet. Consumer drives commonly state 1 in 1014; enterprise drives 1 in 1015 or better.

Rebuild Time

22.2 hours
8.0 TB written at 100 MB/s

Data Read

24.0 TB
3 surviving drives, read in full

Chance of a URE

85.3%
at least one during the rebuild, at 1 in 10¹⁴

What a URE would mean here

RAID 5: a URE during the rebuild means that stripe cannot be reconstructed. Depending on the controller, the rebuild fails or that data is lost.

One failed drive, 4 drives in the array. The probability is 1 − e−bits read / rating, using the spec-sheet rating as the error rate. It is an estimate, not a prediction.

What a URE is and why rebuilds are the risky window

An unrecoverable read error is a read the drive cannot complete: the data in a sector cannot be recovered even after the drive’s own error correction and retries. In a healthy array with redundancy this is a small event, because the controller reconstructs the block from parity or from the mirror copy.

A rebuild is when that margin can be gone. A RAID 5 array with one failed drive has no redundancy left, and rebuilding the replacement means reading every block on every surviving drive. One URE anywhere in that read leaves a stripe with nothing to reconstruct it from. The amount of data a rebuild reads is what drives the estimate above, so it grows with both the drive size and the number of drives.

What the estimate assumes

Drive manufacturers print a URE rating on the spec sheet as one error per a given number of bits read. The calculator uses that rating as the error rate and works out the chance of at least one error across all of the bits the rebuild reads. The rating is a conservative bound: it is what manufacturers state, real-world error rates are often lower, and the model treats errors as independent. The result is an estimate of exposure under the stated rating, not a prediction of what your array will do.

Rebuild time is the drive size divided by the speed you enter, because the replacement drive is written once. If the array will keep serving its normal workload during the rebuild, enter a lower speed to reflect that.

Why larger drives favor RAID 6 or RAID 10

Doubling the drive size doubles the rebuild time: at 100 MB/s an 8 TB drive takes 22.2 hours and a 16 TB drive takes 44.4 hours. Larger drives also mean more data read during the rebuild, so the chance of a URE rises with them.

RAID 6 keeps a second parity block through a single-drive rebuild, so a URE in that window is corrected instead of costing data; the RAID 6 calculator shows what that costs in capacity. RAID 10 reads only the failed drive’s mirror partner, so the data exposed stays at one drive however wide the array is; the RAID 10 calculator covers its capacity and fault tolerance. To compare usable capacity across every level, use the general RAID calculator.

Free developer tools by Clocktower and Associates. All processing happens in your browser: no data is collected or transmitted.