RAID 5 Data Recovery
When a customer comes to use with a RAID 5 query, it is important that our engineer understands the concept of RAID 5, the different versions and in some cases proprietary variations.
A working RAID 5 system will consist of at least 3 drives. The total accessible data are will be the equivalant of the total sum of n-1 times the size of the smallest of the drives in array) so if we build a RAID 5 array that has 4 x 36 Gb drives the size of the array would be 36 x (4-1) which gives 36 x 3 = 108 Gbytes.
The remaining 36 Gb is used to calculate parity this is done using XOR calculations
For RAID 5 there is a parity block that shifts its position in the array for each stripe..
The diagram above shows our 4-drive raid 5 array with the Parity block (p) marked in each stripe (A, B, C, D) and Data blocks marked A1, A2 ... through .. D2, D3
Lets explain this a bit more clearly.
In the above array it shows the first four stripes or rows (A B C and D) There are also 4 Drives (or columns) 1 2 3 4.
Each Row/Column is called a block (A1, A2, etc)
One block in each stripe will contain parity - this is a calculation that is performed on the other blocks that gives a unique value that is stored in the parity block. Using this method if a drive fails, the missing data can be calculated using the parity method to recalculate the missing drive.
Parity Rotation
One of the most important factors to analyse when working with a broken raid array is the parity rotation. There are 4 standard methods and several more advanced methods of rotation.
In our example above, you can see that for the first Row the Partity is on Drive 4
The second row it is on Drive 3
The Third Row is on Drive 2
And the fourth on Drive 1
If I show that as.
01 02 03 PP
04 05 PP 06
07 PP 08 09
PP 10 11 12
We can see the data follows a logical pattern from top to bottom (ignoring Parity blocks) the data from left to right - top to bottom goes 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
Now lets look at another, This one is RAID 5E (or sometimes named RAID 5 Dynamic)
01 02 03 PP
05 06 PP 04
09 PP 07 08
PP 10 11 12
In this set up the data blocks carry on from their previous position. Now reading data blocks only from top left to bottom right gives 01 02 03 05 06 04 09 07 08 10 11 12
So the importance of getting the correct parity rotation is vital in ensuring the correct data is read.
The other most common method is forward parity ..
PP 01 02 03 PP 01 02 03
04 PP 05 06 06 PP 04 05
07 08 PP 09 08 09 PP 07
10 11 12 PP 10 11 12 PP
Then we get to the more complex arrays .. which are delayed parity I will cover these in a separate post.
Use this link to find out more about RAID Data Recovery
More about our Services for Data Recovery
When a customer comes to use with a RAID 5 query, it is important that our engineer understands the concept of RAID 5, the different versions and in some cases proprietary variations.
A working RAID 5 system will consist of at least 3 drives. The total accessible data are will be the equivalant of the total sum of n-1 times the size of the smallest of the drives in array) so if we build a RAID 5 array that has 4 x 36 Gb drives the size of the array would be 36 x (4-1) which gives 36 x 3 = 108 Gbytes.
The remaining 36 Gb is used to calculate parity this is done using XOR calculations
For RAID 5 there is a parity block that shifts its position in the array for each stripe..
The diagram above shows our 4-drive raid 5 array with the Parity block (p) marked in each stripe (A, B, C, D) and Data blocks marked A1, A2 ... through .. D2, D3
Lets explain this a bit more clearly.
In the above array it shows the first four stripes or rows (A B C and D) There are also 4 Drives (or columns) 1 2 3 4.
Each Row/Column is called a block (A1, A2, etc)
One block in each stripe will contain parity - this is a calculation that is performed on the other blocks that gives a unique value that is stored in the parity block. Using this method if a drive fails, the missing data can be calculated using the parity method to recalculate the missing drive.
Parity Rotation
One of the most important factors to analyse when working with a broken raid array is the parity rotation. There are 4 standard methods and several more advanced methods of rotation.
In our example above, you can see that for the first Row the Partity is on Drive 4
The second row it is on Drive 3
The Third Row is on Drive 2
And the fourth on Drive 1
If I show that as.
01 02 03 PP
04 05 PP 06
07 PP 08 09
PP 10 11 12
We can see the data follows a logical pattern from top to bottom (ignoring Parity blocks) the data from left to right - top to bottom goes 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
Now lets look at another, This one is RAID 5E (or sometimes named RAID 5 Dynamic)
01 02 03 PP
05 06 PP 04
09 PP 07 08
PP 10 11 12
In this set up the data blocks carry on from their previous position. Now reading data blocks only from top left to bottom right gives 01 02 03 05 06 04 09 07 08 10 11 12
So the importance of getting the correct parity rotation is vital in ensuring the correct data is read.
The other most common method is forward parity ..
PP 01 02 03 PP 01 02 03
04 PP 05 06 06 PP 04 05
07 08 PP 09 08 09 PP 07
10 11 12 PP 10 11 12 PP
Then we get to the more complex arrays .. which are delayed parity I will cover these in a separate post.
Use this link to find out more about RAID Data Recovery
More about our Services for Data Recovery
Comments
Post a Comment
Please leave a comment on this data recovery blog