RAID is an acronym of Redundant Array of Independent Disks. The RAID controller (Hardware and Software) integrates several individual disks to a virtual large storage with high capacity. The feature of RAID storage are:
Higher speed in read/write: The simultaneously read/write speed of several HDDs is much faster than that of single HDD(ex. RAID 0).
Fault-Tolerant: Keep in normal operation to allow fault-tolerant for one or two damaged HDDs. The HDD can be recovered after rebuild.
Therefore, RAID is a storage system but not a backup solution.
Taking RAID 5 as an example, a basic concept of RAID called EDAP (Extended Data Availability and Protection) which is emphasizing on its Redundant and Fault-Tolerant features. This concept has also been focused by some major suppliers such as LSI, IBM, HP, DELL, Compaq, Adaptec, and Infortrend… etc., including the ability to proceed following steps under“on-line”situation:
RAID Auto Error Detection
RAID Rebuild and Recovery
RAID Hot Spare
RAID Hot Swap
RAID Hot Expanding
The storage principle of RAID is using Bit Striping and Block Striping as two different divisions to save data in each HDD. When the RAID failed, Data Blocks in Data Stripe and Parity Blocks in other HDDs will rebuild data through calculating by XOR.
The picture below graphically shows how RAID 5 is integrated together.
RAID-0 其圖解如下﹕
RAID-1 其圖解如下﹕
RAID-5 其圖解如下﹕
