Separate Disk Redundancy Arrays (RAID) is a public system for high-capacity data storage at the server level. The RAID system uses many small capacity disk drives to store large amounts of data and enhance reliability and redundancy. For a computer, such an array is like a logic unit consisting of a plurality of disk drives.
RAID stores a variety of ways. Some types of RAID emphasizes performance, which emphasizes reliability, fault-tolerant or error correction capabilities. Therefore, the type can be selected according to the task to be completed.
However, all RAID systems commonly characterized - is also the true advantage is "hot exchange" capability: the user can take out a defective drive and insert a new replacement. For most types of RAID, you don't have to interrupt the server or system, you can automatically rebuild data on a failed disk.
RAID is not the only way to protect large amounts of data, but routine backup and mirroring software are slow, and if a drive fails, an interrupt system is often required.
Even if the disk does not cause the server to interrupt, IT staff still need to break away the server to replace the drive. In contrast, RAID uses mirror or parity to rebuild data from the remaining driver, and does not have to interrupt the system.
Level 0, 3 and 5 are three most common RAID implementations.
RAID Level 0 is data segmentation, is the most basic way. On a normal hard drive, the data is stored on the continuous sector of the same disk. RAID 0 uses at least two disk drives and divides data into several blocks from 512 bytes to Duzab, which are alternately written to disk. The first paragraph is written to the magnetic disk 1, and the second paragraph is written to the magnetic disk 2, and so, etc.. When the system reaches the last disk in the array, it is written to the next segment of the disk 1, here is the following.
The split data assigns an I / O load average to all drives. Since the driver can be written or read at the same time, performance is significantly improved. However, it has no data protection capabilities. If a disk is faulty, the data will be lost. RAID 0 does not apply to a critical task environment, but it is ideal for video production and editing or image editing.
RAID LEVEL 3 includes data segmentation, in addition, it also specifies a drive to store parity information. This provides a certain fault-tolerant function, especially in a data-intensive environment or a single user environment, especially for accessing longer continuous records. RAID 3 requires a synchronous spindle drive to prevent shorter performance degradation.
RAID LEVEL 5 is similar to Level 0, but it is not divided into blocks, but splits the bit of each byte to multiple disks. This will increase management costs, however, if a disk is faulty, it can be replaced, and the data can be rebuilt from parity and error correction code. RAID 5 includes all read / write operations. It requires three to five disks to form an array, most suitable for multi-user systems that do not require key features or hardly write operations.
Other uncommon RAID types
RAID LEVEL 1 is a disk mirror-writable to disk 2, and can be read from any one disk. This provides an instant backup, but the number of disk drives needed is most, and performance cannot be improved. RAID 1 provides optimum performance and fault tolerance in multi-user systems, is the most easy implementation configuration, which is most suitable for financial processing, salary, finance, and high availability data environments.
RAID Level 2 is developed for large machines and supercomputers. It corrects data without interruption, but RAID 2 tends to be higher data checksum error correction rate.
RAID Level 4 includes a larger data strip, so you can read records from any drive. Since this type lacks support for multiple simultaneous writes, it is hardly used.
RAID Level 6 has almost no commercial use. It uses a second parity that is assigned to different drives, extending RAID 5. It can withstand multiple drivers at the same time, but performance - especially write operations, and system requires an extremely complex controller. RAID Level 7 has a real-time embedded operating system as a controller, a high-speed bus for cache. It provides fast I / O, but the price is expensive.
RAID Level 10 consists of a data strip array, where each strip is a RAID 1 array of drives. It is the same as the fault tolerance of RAID 1, which requires high performance and redundancy, but does not require high capacity database servers.
RAID Level 53 is the latest type, implementation with the Level 0 data strip array, where each section is a RAID 3 array. Its redundant and fault-tolerant ability with RAID 3. This is beneficial for IT systems that require high data transfer rates, but it is expensive and low efficiency.
