CN101620518B

CN101620518B - Method and apparatus for creating redundancy array in disc RAID

Info

Publication number: CN101620518B
Application number: CN2009101629999A
Authority: CN
Inventors: 孙策
Original assignee: Hangzhou H3C Technologies Co Ltd
Current assignee: New H3C Technologies Co Ltd
Priority date: 2009-08-21
Filing date: 2009-08-21
Publication date: 2011-06-29
Anticipated expiration: 2029-08-21
Also published as: CN101620518A

Abstract

The invention discloses a method for creating redundancy array of disc RAID, comprising the following steps: a disc storage controller combines at least two small-volume discs into a virtual disc, creates RAID according to the virtual disc and a large-volume disc, strips the RAID and reads and writes data with the striped RAID. In the invention, when discs of different volumes are available, the small-volume discs are combined together to form the virtual disc for creating the RAID with the virtual disc and the large-volume disc, thereby increasing the use ratio of useful space of discs; besides, the usability of the whole RAID is not affected when two discs in different positions of the virtual disc are damaged.

Description

A kind of creation method of redundancy array in disc RAID and device

Technical field

The present invention relates to technical field of memory, particularly relate to a kind of creation method and device of redundancy array in disc RAID.

Background technology

RAID (Redundant Array of Inexpensive Disks, redundant arrays of inexpensive disks) is the memory technology of a kind of high-performance, high reliability, by a series of independent disks are combined in a different manner, for application terminal or terminal cluster provide in logic disk.Wherein, use the advantage of RAID to comprise: the performance and the safety of data that enlarge disk size, raising disk read-write.

Concrete, the RAID technology has been widely used in the various occasions of data storage, and the RAID technology of long usefulness comprises RAID0, RAID1, RAID5, RAID6, RAID10 etc.; Wherein, RAID0 does not have redundant ability, and RAID1 is not a disk array completely; And RAID5, RAID6 and RAID10 are respectively by the polylith disk (for example, RAID5 and RAID6 are no less than 3, RAID10 is no less than 4) form, each RAID is with the disk write data of mode in array of band, and parity data is left on each disk in the array.

Further, contain 1 check bit on each band of RAID5, support to damage arbitrarily wherein disk, come data reconstruction by the parity check bit on other disks; Contain 2 check bit on each band of RAID6, support to damage arbitrarily wherein two disks, come restore data by the parity check bit on other disks; RAID10 carries out mirror image earlier to disk groups, again disk is carried out band, by its corresponding mirror drive restore data, allows under the diverse location 50% disk failures at most after a disk failures.

As shown in Figure 1, for a kind of be the data storage procedure of example with RAID5, wherein, the RAID5 array is made up of 3 disks, and writes 6 blocks of data that have of RAID5, and at this moment, 6 blocks of data need be divided into 3 bands and write in the disk.Wherein, the data block among the band P (P1, P2 and P3) is a parity data, is calculated by the data XOR in each band respectively, promptly

P 1 = A &CirclePlus; B,

P 2 = C &CirclePlus; D,

P 3 = E &CirclePlus; F;

When supposing that disk 1 damages, for the data A in the band 1, can by

Computing draws.

In the prior art, when creating RAID, RAID is made up of several disk, and each disk exists with individual in RAID, after the RAID establishment is finished, need carry out striping to disk, to guarantee the redundancy of disk, in the striping process, require the size of array must calculate striping process as shown in Figure 2 according to minimum disk space size.In 8 disks, minimum disk space is 300G, is base unit with 300G when striping promptly, when being example to create RAID5, the free space of array is 300* (8-1)=2100G, and it is check bit that 1 disk is wherein arranged, and the disk that promptly can store data is 7.As can be seen, the useful space utilization factor of disk is 2100/3900=53.8%.

In addition, also can form several arrays separately with the identical disk of capacity, 8 disks as shown in Figure 2 can be created array respectively for three groups according to 500G, 300G, 750G, striping process three arrays as shown in Figure 3; Total free space of three arrays is 300*2+500*2+750*1=2350G; As can be seen, the useful space utilization factor of disk is 2100/3900=60.2%.

Based on top analysis, when the low capacity disk existed, when creating RAID, the useful space utilization factor of disk was lower, causes the waste of a large amount of disk spaces in the prior art.

Summary of the invention

The invention provides a kind of creation method and device of redundancy array in disc RAID,, improve the useful space utilization factor of disk, save a large amount of disk spaces with when having the low capacity disk to exist.

In order to achieve the above object, the present invention proposes a kind of creation method of redundancy array in disc RAID, be used for some disks are configured to RAID, comprise in described some disks that at least one large capacity disc and at least two capacity all less than the low capacity disk of aforementioned large capacity disc, said method comprising the steps of:

Disk storage controller synthesizes virtual disks with at least two low capacity disk groups; And the mapping relations of record logical address and physical address, this logical address is the address of the virtual disk after virtual, this physical address is the address of virtual preceding low capacity disk;

Described disk storage controller is created RAID according to described virtual disk and large capacity disc;

Described disk storage controller carries out striping to described RAID, and uses the RAID behind the striping to read and write data;

In the process of carrying out striping, described disk storage controller carries out striping according to the logical address of described virtual disk.

Wherein, described disk storage controller synthesizes two low capacity disk groups after the virtual disk at least, also comprises:

Described disk storage controller writes down virtual disk and forms information in being combined into the low capacity disk of virtual disk, described virtual disk composition information comprises:

The drive of virtual disk; The mapping relations of logical address and physical address; The virtual disk capacity.

Wherein, the RAID behind the described disk storage controller use striping reads and writes data and specifically comprises:

When needs read and write data, described disk storage controller was according to the logical address of virtual disk and the mapping relations of physical address, and the logical address that reads and writes data is obtained the physical address of described virtual disk;

Described disk storage controller reads and writes data by the physical address that uses described virtual disk.

Wherein, when RAID moved, described method also comprised:

The migration disk storage controller obtains the virtual disk of disk and forms information;

Described migration disk storage controller judges according to described virtual disk composition information whether disk is virtual disk;

When disk was virtual disk, described migration disk storage controller joined the mode of disk with virtual disk among the RAID;

When disk was not virtual disk, described migration disk storage controller directly joined disk among the RAID.

Wherein, when disk was virtual disk, described method also comprised:

Described migration disk storage controller judges according to described virtual disk composition information whether described virtual disk is complete;

When described virtual disk was imperfect, described migration disk storage controller selected existing disk to add RAID, kept the part of RAID complete, was having new disk to add the described RAID of fashionable reconstruction;

When described virtual disk is complete but when disk failures was arranged, the disk that described migration disk storage controller is selected not have to damage added RAID, the part of maintenance RAID is complete, is having new disk to add the described RAID of fashionable reconstruction.

A kind of creation apparatus of redundancy array in disc RAID is used for some disks are configured to RAID, comprises at least one large capacity disc and at least two capacity in described some disks all less than the low capacity disk of aforementioned large capacity disc, and described device comprises:

Virtual module is used for the synthetic virtual disk of at least two low capacity disk groups; And the mapping relations of record logical address and physical address, this logical address is the address of the virtual disk after virtual, this physical address is the address of virtual preceding low capacity disk;

Creation module is used for creating RAID according to the virtual disk and the large capacity disc of described virtual module combination;

Processing module is used for the RAID that described creation module is created is carried out striping, and uses the RAID behind the striping to read and write data;

In the process of carrying out striping, carry out striping according to the logical address of described virtual disk.

Wherein, described virtual module also is used for:

The record virtual disk is formed information in being combined into the low capacity disk of virtual disk, and described virtual disk composition information comprises: the drive of virtual disk; The mapping relations of logical address and physical address; The virtual disk capacity.

Wherein, described processing module specifically is used for:

When needs read and write data, according to the logical address of virtual disk and the mapping relations of physical address, and the logical address that reads and writes data was obtained the physical address of described virtual disk; And read and write data by the physical address that uses described virtual disk.

Wherein, when RAID moved, described processing module also was used for:

Obtain the virtual disk of disk and form information; Judge according to described virtual disk composition information whether disk is virtual disk; When disk is virtual disk, the mode of disk with virtual disk joined among the RAID; When disk is not virtual disk, disk is directly joined among the RAID.

Wherein, when disk was virtual disk, described processing module also was used for:

Judge according to described virtual disk composition information whether described virtual disk is complete; When described virtual disk is imperfect, select existing disk to add RAID, keep the part of RAID complete, there is being new disk to add the described RAID of fashionable reconstruction;

When described virtual disk is complete but when disk failures was arranged, the disk of selecting not have to damage added RAID, the part of maintenance RAID is complete, is having new disk to add the described RAID of fashionable reconstruction.

Compared with prior art, the present invention has the following advantages: when the disk of different capabilities, by with the disk combination of low capacity to together, form virtual disk, this virtual disk is participated in the establishment of RAID with large capacity disc, increased the useful space utilization factor of disk; And when two disk failures of diverse location in the virtual disk, also can not influence the availability of whole RAID.

Description of drawings

Fig. 1 is for being the data storage procedure synoptic diagram of example with RAID5 in the prior art;

Fig. 2 is the synoptic diagram of striping process in the prior art;

Fig. 3 is an another kind of striping process synoptic diagram of the prior art;

Fig. 4 is the creation method process flow diagram of a kind of redundancy array in disc RAID of the present invention's proposition;

Fig. 5 is the creation method process flow diagram of a kind of redundancy array in disc RAID of proposition under the application scenarios of the present invention;

Fig. 6 is virtual disk synoptic diagram under the application scenarios of the present invention;

Fig. 7 is under the application scenarios of the present invention inventing the low capacity disk synoptic diagram of large capacity disc;

Fig. 8 creates the synoptic diagram of RAID down for application scenarios of the present invention;

Fig. 9 is for being the synoptic diagram of real physical address with virtual address translation under the application scenarios of the present invention;

Figure 10 is the processing procedure synoptic diagram when under the application scenarios of the present invention disk failures being arranged;

Figure 11 is the creation apparatus structural drawing of a kind of redundancy array in disc RAID of the present invention's proposition.

Embodiment

Basic thought of the present invention is when having the low capacity disk in the disk of forming RAID, by at least two low capacity disks are invented a large capacity disc, and this large capacity disc that invents created RAID with other large capacity discs, thereby reach the purpose of the useful space utilization factor that increases disk.

The creation method of a kind of redundancy array in disc RAID that the present invention proposes, be used for some disks are configured to RAID, comprise in described some disks that at least one large capacity disc and at least two capacity are all less than the low capacity disk of aforementioned large capacity disc, as shown in Figure 4, said method comprising the steps of:

Step S401, disk storage controller will at least two synthetic virtual disks of low capacity disk groups.

Step S402, described disk storage controller is created RAID according to described virtual disk and large capacity disc.

Step S403, described disk storage controller carries out striping to described RAID, and uses the RAID behind the striping to read and write data.

As shown in Figure 5, below in conjunction with a kind of concrete application scenarios, the creation method of the redundancy array in disc RAID that the present invention is proposed is described in detail, the creation method of this redundancy array in disc RAID is applied in when (for example forming disk array, RAID) there are one or more low capacity disks in the disk, several disk spaces can be formed a jumbo disk, and create redundant array, thereby reach the purpose that improves disk useful space utilization factor with other large capacity discs.Wherein, the creation method of above-mentioned redundancy array in disc RAID may further comprise the steps:

Step S501, the disk storage controller disk that capacity is less makes up, and the bigger disk of the capacity that invents, and the bigger disk of capacity after this is virtual is a virtual disk.Wherein, before creating RAID, disk storage controller need invent the bigger disk of capacity by the disk that capacity is less; The less disk of this capacity is that disk storage controller is optional according to actual needs, for example, when the disk space of forming RAID comprises 100G, 200G, 300G and 400G, disk storage controller can all be chosen as the less disk of capacity with 100G, 200G and 300G, and the disk of 400G is chosen as the bigger disk of capacity; Again for example, when the disk space of forming RAID comprised 200G, 300G, 400G and 600G, disk storage controller can all be chosen as the less disk of capacity with 200G, 300G and 400G, and the disk of 600G is chosen as the bigger disk of capacity; When the disk space of forming RAID comprised 200G, 300G, 550G and 600G, disk storage controller can be chosen as the less disk of capacity with 200G and 300G, and the disk of 550G and 600G is chosen as the bigger disk of capacity.For convenience of description, the less disk of capacity all is that example describes with the minidisk in the present invention, and the disk that capacity is bigger is that example describes with big disk all.

In this step, when the disk space of forming RAID comprised 100G, 200G, 300G and 400G, disk storage controller will be according to actual needs invents the virtual disk of 400G with 100G and 300G, and 200G and 200G was invented the virtual disk of 400G; For other disk combination, the corresponding virtual situation similarly no longer encyclopaedizes among the present invention.

Step S502, disk storage controller create disk and form list item, and use this disk to form information such as the mapping relations of list item record logical DOS drives, logical address and physical address, disk size.Wherein, disk storage controller also need mark off a part of space and store the information that writes down in this disk composition list item in each disk, when this disk plugged other disk storage controllers, disk storage controller can read corresponding virtual disk information from this space.

Concrete, after disk storage controller invents virtual disk with minidisk, this logical DOS drives is the drive of the virtual disk after virtual, this logical address is the address of the virtual disk after virtual, this physical address is the address of the minidisk before virtual, and this disk size is the disk size of the virtual disk after virtual.Continuation is that example describes with 8 disks shown in Figure 2, and 8 disks are respectively the disk of 3 300G, the disk of the disk of 3 500G and 2 750G; In step S501, it is minidisk that disk storage controller will be selected the disk of 300G and the disk of 500G, and selecting the disk of 750G is big disk; And the disk of 300G and the disk of 500G are invented virtual disk, virtual disk as shown in Figure 6, this virtual disk arrives the disk of 300G and the disk combination of 500G together; As can be known, the space of this virtual disk will be 800G.

Further, as shown in Figure 7, minidisk is invented the synoptic diagram of big disk (being virtual disk) for the disk storage controller of corresponding diagram 2, wherein, virtual disk 1, virtual disk 2 and virtual disk 3 are the big disk (disk after virtual) of 800G, when illustrating that with situation shown in Figure 7 this disk is formed list item, as shown in table 1.

Table 1

Logical DOS drives	Logical address	Physical address	Disk size
				Virtual disk 1	0-300G 300G-800G	Disk 3 disks 1	800G
Virtual disk 2	0-300G 300G-800G	Disk 4 disks 2	800G
				Virtual disk 3	0-300G 300G-800G	Disk 5 disks 8	800G

As can be seen from Table 1, the disk size of virtual disk 1 is 800G, in the disk of this 800G, the address of 0-300G is the address of disk 3, is the address of disk 1 from 300G to 800G, for the situation of virtual disk 2 and virtual disk 3, similarly, repeat no more among the present invention.Further, disk storage controller also need be at disk 1-disk 5, and the information in the above-mentioned table 1 of record in the disk 8, with when above-mentioned virtual disk is put into other disk storage controllers, other disk storage controllers can read corresponding information, thereby use this virtual disk.

Step S503, disk storage controller is created RAID.Wherein, minidisk is become after the virtual disk, can form RAID, and the disk that the array information that needs write disk is kept at virtual disk is neutralized in other the big disk with other big disk; And do not need to preserve this array information on other disks of virtual disk.When continuation is example with 8 disks shown in Figure 2, the RAID that creates as shown in Figure 8, as can be known, need in disk 3, disk 4, disk 5, disk 6 and disk 7, preserve corresponding array information, and do not need to preserve corresponding array information in disk 1, disk 2 and the disk 8, this array information is the essential information of RAID, includes but not limited to RAID type (RAID0, RAID1, RAID5 etc.); RAID capacity etc.; No longer encyclopaedize among the present invention.

Need to prove, when the RAID shown in Fig. 8 is RAID5, as can be known, the free space of this RAID is 750* (5-1)=3000G, the utilization factor that is the disk useful space is 3000/3900=76.9%, as seen, compares with 60.2% with 53.8% of prior art, by adopting method provided by the present invention, significantly improved the utilization factor of the disk useful space.

Step S504, disk storage controller carries out striping to RAID.Wherein,, and parity data left on the disk in the array because RAID is the disk write data of mode in array with band, promptly RAID create finish after, need carry out the operation of striping to RAID.

Concrete, in the process of striping, for virtual disk, carry out striping at logical address, for application layer, can only see the virtual address (being logical address) of array, also be that virtual address with array is that destination address is operated for data write.In the process that reality is used, for virtual disk, it is that real physical address carries out the data write operation that disk storage controller needs this virtual address translation, wherein, this disk storage controller is to be real physical address with the mapping relations of logical address and physical address with virtual address translation; And repeat no more among the striping process the present invention for big disk; As shown in Figure 9 be the synoptic diagram of real physical address with virtual address translation; Among the present invention, describe for example with corresponding 100 bands in the space of 100G (this quantity is chosen 100 for convenience of description and is example much larger than this); Among the RAID then shown in Figure 8, each disk is with 750 bands of correspondence, when data need store the 10th band into, for virtual disk, can only store on the 10th band of logical address, disk storage controller is when knowing that logical address is the 10th band, and obtaining physical address corresponding is the 10th band (corresponding to virtual disk 1) of disk 3, the 10th band (corresponding to virtual disk 2) of disk 4, the 10th band (corresponding to virtual disk 3) of disk 5; Just with this data storage to the 10th band of disk 3, disk 3 and disk 5; And, do not need the reference address for big disk, directly will get final product on 10 bands of data storage to the.Further, when data need store on the 310th band, disk storage controller is when knowing that logical address is the 310th band, and obtaining physical address corresponding is the 10th band (corresponding to virtual disk 1) of disk 1, the 10th band (corresponding to virtual disk 2) of disk 2, the 10th band (corresponding to virtual disk 3) of disk 8; Just with this data storage to the 10th band of disk 1, disk 2 and disk 8; And, do not need the reference address for big disk, directly will get final product on 310 bands of data storage to the.

As seen, by above-mentioned process, can use the striping to RAID, corresponding, in the present invention, the content in can also his-and-hers watches 1 expands, and the corresponding relation of logical address and physical address is specific on the band, as shown in table 2.

Table 2

Logical DOS drives	Logical address	Physical address	Disk size
				Virtual disk 1	A 0-300 band 300-800 band	0-500 band of the 0-300 of disk 3 band disk 1	800G
Virtual disk 2	A 0-300 band 300-800 band	0-500 band of the 0-300 of disk 4 band disk 2	800G
				Virtual disk 3	A 0-300 band 300-800 band	0-500 band of the 0-300 of disk 5 band disk 8	800G

Need to prove, RAID is being carried out in the process of striping, also need in disk, determine check bit, for example, all need to contain 1 check bit on each band of RAID5, and parity data is to use the XOR computing to draw by the data bit in each band, at this moment, when a disk was damaged, can use the XOR computing by the data bit in the band+parity check bit (was XOR, be used for parity checking, use symbol usually

Expression, its operation rule is:

0 &CirclePlus; 0 = 0,

0 with 0 XOR, and the result is 0;

0 &CirclePlus; 1 = 1,

0 with 1 XOR, and the result is 1;

1 &CirclePlus; 0 = 1,

1 with 0 XOR, and the result is 1;

1 &CirclePlus; 1 = 0,

1 with 1 XOR, and the result is 0;

A &CirclePlus; B = C < = > A &CirclePlus; C = B < = > B &CirclePlus; C = A

) reduce the data of losing, repeat no more among the present invention.

Further, for scheme provided by the present invention, when migration takes place array, also comprise:

Step S505, the migration disk storage controller obtains all disk virtual informations, and creates RAID according to the array information of disk.Wherein, move to a disk storage controller (can be called migration disk storage controller) following time as disk, this migration disk storage controller will read the header information of each disk, when this disk is virtual disk, will carry the disk virtual information in this header information; When this disk is not virtual disk, will not carry the disk virtual information in this header information; Wherein, this disk virtual information is the information in the determined disk composition list item among the above-mentioned steps S502, be that disk storage controller can judge whether disk is virtual disk according to this header information, when disk is virtual disk, judge whether virtual disk is complete; For example, when migration takes place in the described RAID of Fig. 8, the migration disk storage controller is by obtaining the head message on all disks, can know disk 3 and disk 1 composition virtual disk 1, disk 4 and disk 2 are formed virtual disk 2, disk 5 and disk 8 are formed virtual disk 3, disk 6 and disk 7 are not virtual disks, after knowing above-mentioned information, the migration disk storage controller knows that all virtual disks all are complete, when if virtual disk is imperfect, the migration disk storage controller can be created RAID according to actual needs again, or temporarily select existing disk to add array, retaining part array complete finished reconstruction thereby treated to add to after new disk adds again in the array.

Further, as shown in figure 10, when virtual disk when complete, if when finding disk failures is arranged, the migration disk storage controller can be created RAID according to actual needs again, or temporarily select existing disk to add array, retaining part array complete finished reconstruction thereby treated to add to after new disk adds again in the array.In Figure 10, disk 3 and the disk 2 in the virtual disk 2 in the virtual disk 1 damage simultaneously, and RAID can normally move, and after new disk adds, then can rebuild according to array information and virtual disk information.

Need to prove, the migration disk storage controller is to create RAID's according to the array information of disk, as shown in figure 10, in disk 3, disk 4, disk 5, disk 6 and disk 7, preserved array information, for example, when the RAID type is RAID5 in the array information, then move disk storage controller and need create the RAID5 array, do not repeat them here.

Wherein, the step among the present invention can be adjusted according to the actual needs.

The creation apparatus of a kind of redundancy array in disc RAID that the present invention proposes, be used for some disks are configured to RAID, comprise at least one large capacity disc and at least two capacity in described some disks all less than the low capacity disk of aforementioned large capacity disc, as shown in figure 11, described device comprises:

Virtual module 1101 is used for the synthetic virtual disk of at least two low capacity disk groups.

Wherein, at least two minidisks are being combined in the process of virtual disk, described virtual module 1101 also need write down virtual disk and form information in described virtual disk, and described virtual disk composition information comprises: the drive of virtual disk; The mapping relations of logical address and physical address; The virtual disk capacity.

Creation module 1102 is used for creating RAID according to the virtual disk and the large capacity disc of described virtual module combination.

Processing module 1103 is used for the RAID that described creation module is created is carried out striping, and uses the RAID behind the striping to read and write data.

Wherein, when RAID was carried out striping, described processing module 1103 need be carried out striping according to the logical address of described virtual disk; When needs read and write data, according to the logical address of virtual disk and the mapping relations of physical address, and the logical address that reads and writes data was obtained the physical address of described virtual disk; And read and write data by the physical address that uses described virtual disk.

Further, when RAID moved, described processing module 1103 was obtained the virtual disk of disk and is formed information; Judge according to described virtual disk composition information whether disk is virtual disk; When disk is virtual disk, the mode of disk with virtual disk joined among the RAID; When disk is not virtual disk, disk is directly joined among the RAID.

When disk was virtual disk, described processing module 1103 judged according to described virtual disk composition information whether described virtual disk is complete; When described virtual disk is imperfect, select existing disk to add RAID, keep the part of RAID complete, there is being new disk to add the described RAID of fashionable reconstruction; When described virtual disk is complete but when disk failures was arranged, the disk of selecting not have to damage added RAID, the part of maintenance RAID is complete, is having new disk to add the described RAID of fashionable reconstruction.

Wherein, each module of apparatus of the present invention can be integrated in one, and also can separate deployment.Above-mentioned module can be merged into a module, also can further split into a plurality of submodules.

Through the above description of the embodiments, those skilled in the art can be well understood to the present invention and can realize by hardware, also can realize by the mode that software adds necessary general hardware platform.Based on such understanding, technical scheme of the present invention can embody with the form of software product, it (can be CD-ROM that this software product can be stored in a non-volatile memory medium, USB flash disk, portable hard drive etc.) in, comprise some instructions with so that computer equipment (can be personal computer, server, the perhaps network equipment etc.) carry out the described method of each embodiment of the present invention.

It will be appreciated by those skilled in the art that accompanying drawing is the synoptic diagram of a preferred embodiment, module in the accompanying drawing or flow process might not be that enforcement the present invention is necessary.

It will be appreciated by those skilled in the art that the module in the device among the embodiment can be distributed in the device of embodiment according to the embodiment description, also can carry out respective change and be arranged in the one or more devices that are different from present embodiment.The module of the foregoing description can be merged into a module, also can further split into a plurality of submodules.

The invention described above sequence number is not represented the quality of embodiment just to description.

More than disclosed only be several specific embodiment of the present invention, still, the present invention is not limited thereto, any those skilled in the art can think variation all should fall into protection scope of the present invention.

Claims

1. the creation method of a redundancy array in disc RAID, it is characterized in that, be used for some disks are configured to RAID, comprise in described some disks that at least one large capacity disc and at least two capacity all less than the low capacity disk of aforementioned large capacity disc, said method comprising the steps of:

2. the method for claim 1 is characterized in that, described disk storage controller synthesizes two low capacity disk groups after the virtual disk at least, also comprises:

3. method as claimed in claim 1 or 2 is characterized in that, the RAID behind the described disk storage controller use striping reads and writes data and specifically comprises:

4. method as claimed in claim 1 or 2 is characterized in that, when RAID moved, described method also comprised:

5. method as claimed in claim 4 is characterized in that, when disk was virtual disk, described method also comprised:

6. the creation apparatus of a redundancy array in disc RAID, it is characterized in that, be used for some disks are configured to RAID, comprise at least one large capacity disc and at least two capacity in described some disks all less than the low capacity disk of aforementioned large capacity disc, described device comprises:

7. device as claimed in claim 6 is characterized in that, described virtual module also is used for:

8. as claim 6 or 7 described devices, it is characterized in that described processing module specifically is used for:

9. as claim 6 or 7 described devices, it is characterized in that when RAID moved, described processing module also was used for:

10. device as claimed in claim 9 is characterized in that, when disk was virtual disk, described processing module also was used for: