RAID（独立磁盘冗余阵列）The technology is to form multiple ordinary disks into an array to provide external services together.
Its main purpose is to improve the performance of the disk 存储容量, 读写速度and enhance the performance of the disk 可用性和容错能力.
Introduction to various RAID technologies:
RAID 0
Assuming that the server has N disks, RAID 0 means that when data is written to the disk from the memory buffer, 将数据分成N份the data is stored according to the number of disks 同时并发地写入N块磁盘.
Therefore, the overall writing speed and reading speed of RAID 0 are N times that of one disk. However, RAID 0不具有数据备份的功能once one of the N disks is damaged, data integrity will be destroyed.
RAID 1
RAID 1 means that when data is written to the disk, 将一份数据同时写入两块磁盘any disk damage will not cause data loss, and the data can be repaired only by inserting a new disk and duplicating it.
RAID 10
RAID 10 combines the two schemes of RAID 0 and RAID 1 所有磁盘都平均分成两份，数据同时在两份磁盘中写入. It is equivalent to utilizing concurrent read and write of RAID 0 and data backup of RAID 1. But its disadvantage is 磁盘利用率较低that half of the disk capacity is used for backup data.
RAID 3
RAID 3 uses one disk as a dedicated one 奇偶校验磁盘, and it can ensure that data will not be lost when only one disk is damaged 利用其他块磁盘的数据修复. The disadvantage is 做奇偶校验会消耗系统性能that it is easy to cause performance bottlenecks in the system.
RAID 4
The principle of RAID 4 is the same as that of RAID 3. The only difference lies in data partitioning. RAID 3 accesses data on a bit-by-bit basis, while RAID 4 partitions data in units of data blocks.
RAID 5
RAID 5 is similar to RAID 3. The parity data of RAID 5 is not written to a certain disk, but written to all disks in a specific form. In this way, the modification of the checksum data is equally distributed to all the disks, and the data on the failed disk can be rebuilt to the new disk. But if the data has not yet been rebuilt onto the new disks, and one disk fails, all data may be lost.
RAID 6
The principle of RAID 6 is similar to RAID 5, but RAID 6 adds a second independent parity check information block, the data will be striped among all disks, and the parity check blocks of each data block are written to the same on the strip. It rebuilds data on failed disks more quickly.
Comparison of various RAID technologies:
