RAID 6: RAID with 2 Disk Fault Tolerance

RAID is an acronym for Redundant array of inexpensive disks and as this full form suggests it is a collection of disk arranged or set up so that it provides redundancy and availability.

There are many types of RAID that are known and used in the industry to provide redundancy and today we are going to talk about RAID 6, which provides 2 disk fault tolerance.

 

What is RAID 6?

 

If you know about RAID 5, it requires 3 minimum disks and has one parity disk which is distributed equally on all the other data disks. The data and parity information are set on the volume so that they are always on different disks.

 

Recommended Article: Difference Between RAID 01 and RAID 10

 

Raid 6 is an extension of RAID 5 and uses two parity blocks for working which are distributed among all the disks in the RAID pool of RAID 6 and thus require a minimum of 4 disks for its operation. With 2 parity blocks, it has the capability to compute data at a time of 2 disk failures.

If it has only 1 disk failure, data will be calculated the same as RAID 5.

RAID 6 write penalty is more than that in RAID 5 and therefore RAID 5 performance is better than RAID 6 and thus widely used in an infrastructure environment. As RAID 6 contains two parity disks, so its restore operation takes a longer time period than RAID 5.

 

RAID 6 Two disk fault tolerance

 

The above diagram shows a RAID 6 with 4 disks and a parity block distributed along all 4 disks. Each disk has 2 parity blocks which are “p” and “,q”. So for RAID 6, we can conclude the below things.

Minimum Disk Required: 4
Storage Efficiency %: (n-2)*100/n

where n= number of disks

Read Performance: Very good for random reads. Good for sequential reads.
Write Performance: Good for small, random writes (has to write penalty).

 

Parity Calculation of RAID 6

Two distinct syndromes need to be computed in order to allow the loss of any 2 drives. One of them, P can be the simple XOR of the data across the stripes, as with RAID 5. A second, independent syndrome is more complex and requires the help of field theory.

Very good explanation of how RAID-6 dual parity calculation works is given at igoro.com and you must check it.


Hope you have got some idea about RAID 6 and its capability to compute data in case of 2 disks failure. If you would like to know difference between RAID 6 and RAID 5, then check this article. RAID 0, RAID 1, RAID 2, RAID 3, RAID 4, RAID 5 and RAID 6 all have their own advantages and disadvantages and should be used accordingly. Let us know if you have any questions through email or comments.

Buy me a coffeeBuy me a coffee

Add Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.