Raid in Linux

Reference Document http://www.cnblogs.com/ivictor/p/6099807.htmlMaking RAID5 http://blog.51cto.com/11134648/2103384RAID0also known as stripe or striping, it represents the highest storage performance in all RAID levels. RAID 0 improves storage performance by spreading continuous data across multiple disks, so that the system has data requests that can be executed in parallel by multiple disks, each of which performs its own portion of the data request. The parallel operation on this data can make full use of the bus bandwidth and significantly improve the overall disk access performance. How it works: The system sends a logical hard disk (RAID0 disk group) that consists of three disksThe/O data request is converted to 3 operations, each of which corresponds to a physical hard disk. By creating a raid0, the original sequential data request is distributed across all three drives simultaneously. In theory, the parallel operation of the three drives increases the disk read and write speed by 3 times times over the same time. However, due to the influence of bus bandwidth and other factors, the actual increase rate will certainly be lower than the theoretical value, but, the large amount of data parallel transmission and serial transmission comparison, the speed-up effect is obviously undoubtedly. Summary: Disk space usage: -%, so the cost is lowest. Read performance: N*read performance write performance for a single disk: N*Write performance redundancy for a single disk: None, any piece of disk corruption will result in data unavailability. RAID1data redundancy is achieved through disk data mirroring, resulting in mutually backed-up data on paired independent disks. When raw data is busy, data can be read directly from the mirrored copy, so RAID 1 can improve read performance. RAID 1 is the highest unit cost in a disk array, but provides high data security and availability. When a disk fails, the system can automatically switch to read and write on the mirrored disk without having to reorganize the failed data. Here, it is important to note that reading can only be done on a piece of disk, and will not be read in parallel, performance depends on the faster piece of the hard disk. Writing is usually slower than a single disk, although it is parallel write, that is, the write to both disks is simultaneous, but because the data in two hard disks is compared, performance is slower than a single disk. RAID1 data through hard disk data mirroring to achieve redundancy, to protect data security, on two disks to produce mutually backed up data, when the raw data is busy, can directly read from the mirror backup, so RAID1 can provide read performance. RAID1 is the highest unit cost in a hard drive, but provides high data security and availability, and when a hard drive fails, the system can automatically switch to the mirrored hard disk to read/and does not need to reorganize the failed data. Summary: Disk space usage: -%, so the cost is the highest. Read performance: Can only be read on one disk, depending on the disk's faster write performance: two disks to write, although the parallel write, but because of the comparison, so the performance of a single disk slow. Redundancy: As long as there is a disk in any pair of mirrored disks in the system that can be used, even the system can function properly in the event of a problem with half the number of hard drives. RAID5RAID 5 is a compromise between RAID 0 and RAID 1. RAID 5 has a similar data read speed as RAID0, with only one parity information, which is slower than writing to a single disk. At the same time, due to multiple data corresponding to a parity information, RAID5 disk space utilization is higher than RAID 1, the storage cost is relatively low, is currently used more than a solution. How it works: RAID5 stores data and relative parity information on each disk that makes up the RAID5, and the parity information and the corresponding data are stored on separate disks, where any n-The full data is stored on 1 disks, meaning that there is space equivalent to a piece of disk capacity for storing parity information. Therefore, when a disk in RAID5 is damaged, the data integrity is not affected, thus ensuring data security. When the damaged disk is replaced, the raid also automatically rebuilds the data on the disk with the remaining parity information to maintain the RAID5 's high reliability. Do a RAID 5 array all disk capacity must be the same size, when the capacity is not the same, the minimum capacity will prevail. The best drive speed is the same, otherwise it will affect performance, and the available space= Number of disks n1Raid5without a separate parity disk, all the checksum information is distributed across all disks, occupying only one disk capacity. Summary: Disk space utilization: (N-1)/N, which wastes only one disk for parity. Read performance: (n-1)*read performance of a single disk, close to RAID0 read performance. Write performance: Less than the write performance of a single disk (this is not very clear, can not be written in parallel?) Redundancy: Only one disk is allowed to be damaged. RAID10 implementation principle: RAID10 In fact, the structure is very simple, first create 2 independent Raid1, and then the two separate Raid1 composed of a Raid0, when writing data to this logical raid, data is written in an orderly two Raid1. Disk 1 and Disk 2 make up a Raid1, and disk 3 and disk 4 form another Raid1; the two Raid1 form a new Raid0. such as data written on hard disk 1 1,3、5、7, written on hard disk 2 is data 1,3、5、7, the data in the hard disk is 0,2、4、6, the data in hard disk 4 is 0,2、4、6, the data is combined into RAID10 on these four hard drives and features both RAID0 and RAID1. Although the RAID10 solution resulted in a% of disk waste, but it provides 200%and single-disk corrupted data security, and data security can be guaranteed when the disk that is damaged at the same time is not in the same Raid1. If one of the disks in the disk is broken, the entire logical disk will still work. When we need to recover the damaged disk in the RAID10, we just need to replace the new hard disk, according to the principle of RAID10 to do the data recovery, the system can still work while recovering the data. The original data will be restored to the replacement hard drive synchronously. Summary: Disk space utilization: -%. Read performance: N/2*read performance write performance for a single drive: N/2*Write performance redundancy for a single hard drive: There is no problem with a disk in a pair of mirrored disks that can be used. 

Raid in Linux