Customer Support: 1.866.933.9800

Sales Questions: 1.866.469.6120

What is RAID 6 and how is it different than Intelligent or Accelerated Raid 6?

First let's examine and determine "what exactly is RAID 6?"
RAID 6, originally introduced in 2006, was considered by many to be one of the most significant advancements to modern data storage. As discussed briefly in a prior blog article titled "What RAID Set should I choose for my Dedicated Server?" RAID 6 is block level striping with double distributed parity amongst it's member drives. For every block of data, RAID 6 creates two parity blocks for each block of data. At a minimum, 4 member drives are required to set up a RAID 6 array. With the increasing size of SAN's (Storage Area Network) it is not uncommon to have 16, 24 or even 48 member drives in the array.

What is the main advantage of RAID 6?
In RAID 6 two of the member drives can fail concurrently. This is a major benefit compared to RAID 5 where the tolerance allows only one of the members to fail. This becomes incredibly beneficial especially in larger storage arrays. The reason is simple. In larger arrays there are more member drives, thus incrementally increasing the probability of multiple drives failing at once. Further, in the event of a single drive failure the array will go into a degraded state and the array will automatically begin rebuilding the array to salvage the data lost on the member drive. With many drive members the rebuilding process takes much longer - in some cases even days! If another drive fails, or even blocks on a drive fail during the rebuilding, real data loss will occur in RAID 5. RAID 6 can survive the failure of two of its member drives - this is the key advantage. RAID 6 protects from not one, but two catastrophic drive or block failures. Given the increasing size of drives, especially SATA, the probability of hitting a bad block has dramatically increased and is now a realistic risk that must be addressed by the RAID policy implemented by IT Managers.

Advantages of Intelligent or Accelerated RAID 6?
Intelligent or accelerated RAID 6 uses two unique, independent equations to generate the stored check values that allow the rebuilding of the data when disks or blocks fail. This data is subsequently stored on two of its member disks. The computational process required by the equations to generate the check values is intensive as the computation is literally performed on every byte of data. Instead of using the host's CPU for processing, the computation and generation of the check values is offloaded to an I/O processor on a device which is separate from the general purpose CPU. Hence the computational process is accelerated by bypassing the general purpose CPU for sequential processing and directed to the hardware that is created for the sole purpose of generating the check values used in RAID 6. Often hardware manufacturers such as Dell (in their EqualLogic XVS and ES line) also require the presence of SSD (Solid State) drives to further accommodate the increased write I/O as a result of the accelerated computed check values that subsequently are striped.

The end result is dramatically increased write performance and significantly reduced RAID 6 recovery times. Ultimately, this results in a much longer Mean Time to Data Loss (MTDL) and substantially reduces the probability of suffering an additional failure during RAID reconstruction. For mission critical data and applications, the additional benefits provided by Intelligent or Accelerated RAID 6 are invaluable.