RAID 0
RAID 0 divides data into block units and writes then in a dispersed manner across a multiple number of disks. As data is placed on every disk, it is also called "striping".
This process enables high level performance as parallel access to the data on different disks improves speed of retrieval. However no recovery feature is provided if a disk failure occurs.
If one disk fails it effects both reads and writes, and as more disks are added to the array the higher the possibility that a disk failure will occur.
back to top
RAID 1
This level is called "mirroring" as it copies data onto two disk drives simultaneously.
Although there is no enhancement in access speeds, the automatic duplication of the data means there is little likelihood of data loss or system downtime.
RAID 1 provides failure tolerance. If one disk fails the other automatically takes over and continuous operation is maintained.
There is no storage cost performance improvement as duplicating all data means only half the total disk capacity is able for storage.
back to top
RAID 1+0
RAID 1+0 combines RAID 0 and RAID 1.
By configuring both technologies in a single array, both data duplication and improved access speed can be provided.
Although this combination makes installation more expensive compared to other technologies, both reliability and high I/O performance can be guaranteed.
Importantly RAID 1+0 on Fujitsu storage systems provides extra protection. This is because a single disk failure doesn't prevent striping to other disks. This version of the technology should be more correctly called RAID 1+0.
back to top
RAID 2
RAID 2 employs the use of Error Correction Codes (ECC), these are also referred to as Hamming Codes (named after Richard Hamming of Bell Labs, who invented them). They provide the ability to search for and correct errors in data.
Data is divided into bit or byte units, and kept on multiple dedicated data drives.
In practice however, RAID 2 is little used as it is inferior to other RAID levels in terms of both cost and performance.
back to top
RAID 3
With RAID 3, data is divided into bit or byte units and written across multiple dedicated data disk drives.
Parity information is created for each of separate data section and written to a dedicated parity drive.
All disk drives can be accessed in parallel all the time and the data can be transferred in bulk, ensuring high-speed data transfer.
back to top
RAID 4
RAID 4 features data re-creation through a combination of RAID 0 striping and the use of a dedicated parity disk. Data is divided into unit blocks and kept on dedicated data disk drives while parity data is kept on a single dedicated parity disk.
When updating it is necessary to pre-read both the existing and parity data and write updated parity data when the update is complete. This process is referred to as having a "write penalty".
It is usually impossible to implement this process in business environments as the dedicated parity disk becomes a bottleneck during traffic surges and performance suffers.
Special technology called WAFL is employed with Fujitsu ETERNUS NR1000F series. This provides a method of overcoming the bottleneck problem and enables high-speed disk access in high I/O environments.
back to top
RAID 5
RAID 5 is the most popular RAID technology in use today. It uses a technique that avoids the concentration of I/O on a dedicated parity disk, that occurs with RAID 4.
Although RAID 5 divides the data and creates parity information similar to RAID 4, unlike RAID 4 the parity data is written separately across multiple disks.
“Write penalty” still occurs as existing data must be pre-read before update and parity data has to be updated after the data is written.
However, RAID 5 enables multiple write orders to be implemented concurrently because updated parity data is dispersed across the multiple disks. This feature ensures higher performance compared with RAID 4.
back to top
RAID 6
RAID 6 deploys two parity records to different disk drives (double parity) enabling two simultaneous disk drive failures in the same RAID group to be recovered. RAID 6 where parity updates are allocated separately across multiple disks, as well as RAID 5, are able to implement multiple write orders at the same time. This feature ensures higher performance when compared to RAID 4 technology.
back to top |
