RAID 6

Jan 18, 2012

RAID 6 aims to increase the reliability of RAID 5 by providing an additional parity block. It uses block-level striping wuth two parity blocks distributed over member disks. Parity can be spread over all disks in the array, as in the case of Reed-Solomon based, or over two disks, as might be found in a vertical XOR. There are also proprietary RAID 6 controllers, like the Adaptec 3805, which uses a custom Horizontal XOR.

Although the read performance of RAID 6 is good, the performance relating to write cycles can be significantly lower than that of RAID 5 volumes and single disk volumes. This is a result of the double parity calculations. The variance in performance is largely due to the capabilities of the RAID 6 controller and the manufacturer’s  storage architecture.

RAID 6 requires one additional disk compared with a RAID 5 array. As such there is a further loss of overall capacity. The usable capacity of a RAID 6 array is (N – 2) x Smin, where N is the total number of disks in the raid array and ‘Smin’ is the capacity of the smallest disk in the array.

Cheadle Data Recovery has experience of rebuilding failed RAID 6 arrays, including those with proprietary parity configuration. This requires the technician working on the array to manually configure and map the parity setup – it is not possible to simply emulate the RAID 6 configuration using off the shelf software.

If you have any data loss issues associated with your RAID 6 array, please contact Cheadle Data Recovery on 0161 408 4857 for a consultation.