A detailed description of RAID usage

Concept:

RAID(Redundant array independent disk) Redundant array of independent disks. At the time, in order to replace expensive, high-capacity disks with inexpensive , low-performance, low-capacity disks, the RAID Array was created at the University of California, Berkeley, 1987 . Its main working mechanism is to organize multiple small-capacity disks into a large-capacity disk for use.

RAID Common level:

RAID0:

650) this.width=650; "style=" width:300px;height:168px; "src=" http://s3.51cto.com/wyfs02/M01/7D/40/ Wkiol1bjyjqcenbiaaacgpmft4i194.png "title=" qq20160312154741.png "vspace=" 0 "width=" height= "168" hspace= "0" Border= "0" alt= "wkiol1bjyjqcenbiaaacgpmft4i194.png"/>

RAID0 mechanism: When the data is to be stored on the disk, the average data is divided into n parts, each disk holds The amount of 1/n data, which greatly improves the ability of data writing and reading out the disk, All disks hold this valid data, so the utilization of the disk is 100%; however, since each data is only one copy on the disk, when Any one of the disks in the RAID array is damaged, The rest of the data stored in the disk will be invalid, its fault tolerance is zero, the raid array is only suitable for temporary data, not suitable for storing more important data.

RAID1:

650) this.width=650; "style=" width:300px;height:164px; "src=" http://s3.51cto.com/wyfs02/M02/7D/40/ Wkiol1bjyqixga-xaaabheen17m901.png "title=" qq20160312154934.png "vspace=" 0 "width=" height= "164" hspace= "0" Border= "0" alt= "wkiol1bjyqixga-xaaabheen17m901.png"/>

RAID1 mechanism: When the data to be stored on the disk, all the data will be stored to one disk, and then on the other disk to hold the image of the data, if one of the two disks is damaged, the data will not be affected, if the data is mistakenly deleted, then two disk data will be deleted, it can not replace the data backup function. This mechanism greatly improves the fault tolerance of the disk, but greatly reduces the read and write performance of the data, the disk utilization is 50%, and is the lowest utilization level in All level , mainly used for storing important data to prevent damage to the disk and lead to business termination.

RAID4:

650) this.width=650; "style=" WIDTH:350PX;HEIGHT:169PX; "src=" http://s3.51cto.com/wyfs02/M00/7D/40/ Wkiol1bjyw7juq0naaawlcdbr44868.png "title=" qq20160312155112.png "vspace=" 0 "width=" + height= "169" hspace= "0" Border= "0" alt= "wkiol1bjyw7juq0naaawlcdbr44868.png"/>

Raid4 mechanism: Theraid4 array requires at least three disks, and when the data needs to be stored on disk, the data is equally divided into n points, the first block is stored in the first disk, The second block of data is stored on the second disk, and the third disk holds the checksum of the data for the first disk and the second disk, so that even if one of the disks that holds the data is damaged, it can be used to get the lost data back through the data check code and another piece of data. This mechanism has a certain fault tolerance, the disk utilization is (n-1)/n*100%. However, this kind of mechanism makes the data read and write performance decreases, each read and write data need to operate at least two disks, while the storage of the check code of the disk is too large, causing the disk pressure is too large, the use of the disk is unbalanced.

RAID5 :

650) this.width=650; "style=" WIDTH:350PX;HEIGHT:207PX; "src=" http://s2.51cto.com/wyfs02/M01/7D/42/ Wkiom1bjyvus785xaaaxvpprczo315.png "title=" qq20160312155317.png "vspace=" 0 "width=" + height= "207" hspace= "0" Border= "0" alt= "wkiom1bjyvus785xaaaxvpprczo315.png"/>

Raid5mechanism:RAID5is aRaid4improved version, the mechanism also requires at least three disks, while the data stored in the time will also hold the data check code, however,RAID5for the storage of check code with different disk storage, divided into "left symmetric" storage and "right symmetry" storage, "symmetry" is different from the data check code storage disk location different to divide. To beRAID5"left-symmetric" storage. "Left symmetry" means the first time,1Inventory DataA,2Inventory DataB,3InventoryABThe verification code of the data; the second time,1Inventory DataC,3Inventory DataD,2Inventory DataCDthe third time,2Inventory DataE,3Inventory DataF,1Inventory DataEFchecksum code. This kind of storage mechanism has strong fault tolerance, while the access pressure of each disk is basically consistent, the utilization of disk andRaid4However, the literacy of the mechanism has declined.

RAID10 :

650) this.width=650; "style=" WIDTH:500PX;HEIGHT:207PX; "src=" http://s4.51cto.com/wyfs02/M02/7D/42/ Wkiom1bjy6rbquccaaa6vuouwvg028.png "title=" qq20160312155613.png "vspace=" 0 "width=" "height=" 207 "hspace=" 0 " Border= "0" alt= "wkiom1bjy6rbquccaaa6vuouwvg028.png"/>

Raid10 mechanism: From the disk combination, the disk 221 groups are mirrored at the bottom with raid1 , and then the individual disk groups are raid0 In a way that combines them. From the data storage mode, when the data needs to be stored, the RAID0 mechanism is used to store each data in different disk groups, and then use raid1 to mirror the data stored in the same disk group. This mechanism does not allow two disks in the same disk group to break down.

RAID01:

650) this.width=650; "style=" WIDTH:500PX;HEIGHT:189PX; "src=" http://s2.51cto.com/wyfs02/M02/7D/40/ Wkiol1bjzkrtskwgaaa54dj-bai608.png "title=" qq20160312155735.png "vspace=" 0 "width=" "height=" 189 "hspace=" 0 " Border= "0" alt= "wkiol1bjzkrtskwgaaa54dj-bai608.png"/>

   raid01 mechanism: from the disk combination, first use multiple disks at the bottom of a group to make raid0 raid1 raid1 Mirror incoming data, then save the data and the good image data in different disk groups, and then use the raid0 raid10 raid01 fault tolerance compared to RAID10 weak.

650) this.width=650; "src="/e/u261/themes/default/images/spacer.gif "style=" Background:url ("/e/u261/lang/ Zh-cn/images/localimage.png ") no-repeat center;border:1px solid #ddd;" Width= "701" height= "" alt= "/ >RAID50:

650) this.width=650; "style=" WIDTH:550PX;HEIGHT:217PX; "src=" http://s2.51cto.com/wyfs02/M00/7D/40/ Wkiol1bjzqhhvrwsaabfwiq2w6q878.png "title=" qq20160312155941.png "vspace=" 0 "width=" 550 "height=" 217 "hspace=" 0 " Border= "0" alt= "wkiol1bjzqhhvrwsaabfwiq2w6q878.png"/>

RAID50 mechanism: From the way the disk is organized, use a disk greater than or equal to three to organize the raid5 array together, and then group the multiple disk groups together in a RAID0 array. From the data storage mode, when the data needs to be stored, The data is first divided into n parts, stored in different disk groups, and then used in the disk group RAID5 , and then store the checksum on the data that was deposited. Because each disk group has a check disk, this mechanism does not allow the same disk group to damage two or more than two disks.

This article is from the "11219885" blog, please be sure to keep this source http://11229885.blog.51cto.com/11219885/1750260

A detailed description of RAID usage