CN104516679A - RAID (Redundant Array Of Independent Disk) data processing method and device - Google Patents

Abstract

The invention provides an RAID (Redundant Array Of Independent Disk) data processing method and device which are applied to the computer system. An RAID comprises n+m member disks, wherein each block of the RAID is equally divided into w sub units. The RAID data processing method includes that: when the quantity of failure member disks is smaller than or equal to m, updating user data of the data sub units needing to be written and updating verification data in a recalculation method verification group by aid of a preset verification algorithm; when the quantity of failure member disks is smaller than or equal to m and the data sub units needing to be read are located on member disks in the normal state, reading corresponding user data from the data sub units, and if the data sub units needing to be read are located on failure member disks, restoring user data in the data sub units according to the preset verification algorithm. Compared with the prior art, the RAID data processing method and device achieve the RAID with the same verification algorithm supporting a plurality of verification disks, and the verification operation process is simple.

Description

A kind of RAID data processing method and device

Technical field

The present invention relates to technical field of data storage, particularly relate to a kind of RAID data processing method and device.

Background technology

Redundant Array of Independent Disks (RAID) (Redundant Array of Independent disks, RAID) be an a kind of disk group of formation that multiple independently physical disk (disk) is combined, therefore can provide the memory property higher than single disk to subscriber's main station, and can data backup function be provided.According to different Method of Data Organizations, conventional RAID comprises the multiple rank such as RAID0, RAID1, RAID5, RAID6, RAID10.Following technology definition is included in RAID.

Band: the data block of the sizes such as continuous print Data Segmentation one-tenth, the data of each data block are written on member's disk of RAID according to RAID algorithm.Data block after the isometric segmentation of continuous data is called band, and data block size is called stripe size.

Stick: the Data distribution8 of each band is on multiple member's disk, and the space size that band takies on each member's disk is called stick.

Data cell: represent the stick for depositing user data.

Verification unit: represent the stick for depositing checking data.

Verification group: in a band, the set be made up of multiple data cell and a verification unit, the checking data in verification unit is calculated by RAID checking algorithm by the user data of data cell in this band.

RAID is synchronous: by RAID checking algorithm, calculates checking data and writes the processing procedure of verification unit.

RAID rebuilds: at RAID member's disk failure or after extracting, by RAID algorithm, calculates failed disk or is pulled out data in disk, and writing reconstruction dish, recovers the process of RAID data redundancy.

In order to avoid single disk failure causes loss of data, storage system uses RAID technique to protect data usually.RAID rank is depended on for the tolerable failed disk number of redundancy RAID, RAID.Can reconstruction be triggered after disk failure, that is, use HotSpare disk to replace failed disk, by RAID reconstruction algorithm the data reconstruction in faulty disk on HotSpare disk, recover the redundancy of RAID.Common redundancy RAID algorithm comprises:

RAID5 checking algorithm: support that 1 disk failures data is not lost, RAID5 active volume=(n-1) × least member disk size, wherein n refers to member's number of disks.

RAID6 checking algorithm: support that 2 disks data that simultaneously break down are not lost, RAID6 active volume=(n-2) × least member disk size, wherein n refers to member's number of disks.

RAID10 checking algorithm: support that the multiple disk of the non-same mirror image pairs data that simultaneously break down are not lost, RAID10 active volume=(n/2) x disk minimum capacity, wherein n refers to member's number of disks.

RAID6 algorithm can ensure, when on a band there is media error in 2 disks simultaneously, and the data of the disk that still can be out of order by the data reconstruction of other disks on this band.The RAID of other ranks also has self advantage and shortcoming, such as RAID10 can support that the multiple disk of the non-same mirror image pairs data that break down are not lost simultaneously, but its capacity utilization only has 1/2, therefore from rate of return on investment, RAID10 is not an ideal chose; And RAID10 cannot solve the situation that same mirror image pairs 2 disks break down, and there is the risk of loss of data simultaneously.RAID5 algorithm capacity utilization is high, but only supports 1 disk failure, and the risk factor of loss of data has relatively high.RAID6 algorithm capacity utilization is between RAID10 and RAID5, but because checking data adopts two kinds of algorithms of different to calculate, the realization of algorithm aspect is quite complicated, in addition, RAID6 does not support that be greater than 2 disks breaks down simultaneously, in the scene higher to reliability requirement, the dirigibility of RAID6 is relative with extendability poor.

Summary of the invention

In view of this, the invention provides a kind of RAID data processing equipment, be applied in computer system, this RAID comprises n+m member's disk, and wherein each stick of this RAID is all divided into w subelement, and in this RAID, each band comprises k according to subelement and j syndrome unit, wherein j=wxm, k=wxn, wherein n, m and w are natural number, and w is greater than 1; Each syndrome unit is under the jurisdiction of a verification group, and each verification group also comprises the data sub-element set of multiple data sub-element composition, and in each band, the data sub-element set of each verification group is different; The position of multiple data sub-element in each band in each verification group in band uniquely corresponds to the position of multiple effective mark in the corresponding a line of the RAID generator matrix preset; This device comprises writing process module and reads processing unit; Wherein:

Writing process module, for the treatment of the write order from user side, when fault member number of disks is less than or equal to m, determines the data sub-element of the needs write that this write order is corresponding; Upgrade the user data in the data sub-element needing write, and use default checking algorithm to upgrade in verification group belonging to this data sub-element all the checking data verified in subelement;

Read processing module, for the treatment of the read command from user side, when fault member number of disks is less than or equal to m, determine the data sub-element of the needs reading that this read command is corresponding, if desired the data sub-element read is positioned on member's disk of normal condition, corresponding user data is read from this data sub-element, if desired the data sub-element read is positioned on fault member disk, data then in recovery group belonging to this data sub-element in current addressable data sub-element and syndrome unit use described default checking algorithm to recover user data in this data sub-element.

Further, the concrete formula using default checking algorithm to upgrade checking data is:

$c_h=d_x\⊕......\⊕d_y$

Wherein in this formula, the subelement at all data places forms a verification group, and h is the natural number being less than or equal to j; X and y is the natural number being less than or equal to k; c _hrepresent in this band the checking data verified in h verification group in subelement; d _x... d _yfor the user data in the multiple data sub-element in this verification group, wherein in data sub-element set the position of each data sub-element in band correspond respectively to h in described RAID generator matrix capable in the position of multiple effective mark; represent XOR.

Further, the generator matrix of described RAID is the matrix of the capable k row of j, the wherein capable i-th column element a of h _hii-th data sub-element in corresponding band, if a _hibe effective mark, represent that in band, i-th data sub-element belongs to h verification group; If a _hifor invalidated identification, represent that i-th data sub-element does not belong to h verification group.

Further, between the generator matrix of described RAID and selected m capable n row Galois Field constant matrices, there are mapping relations, its these mapping relations are: the matrix generated after constant each in this constant matrices being converted to the embedded matrix of w capable w row is this RAID generator matrix, and wherein w is this Galois Field constant matrices Galois Field bit wide.

Further, described constant matrices is full rank, and the anyon matrix of this matrix is also full rank.

Further, wherein said n is more than or equal to m, and n and m sum is less than or equal to the w power of 2.

Further, the concrete formula of described default checking algorithm restoring user data is used to be:

$d_w=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, d _wit is the user data in the data sub-element of inaccessible in a recovery group; Cu...... be the checking data in this recovery group in addressable syndrome unit; Dx...... the user data in addressable data sub-element in this recovery group is represented; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

Further, described RAID recovers the matrix that matrix is the capable s row of j, wherein s=w × (m+n); Described RAID recovers the capable i-th column element a of h in matrix _hii-th subelement in corresponding band, if a _hibe effective mark, represent that in band, i-th subelement belongs to h recovery group; If a _hifor invalidated identification, represent that i-th subelement does not belong to h verification group.

Further, also comprise matrix disposal module, this matrix disposal module is used for generating described RAID according to the RAID extended matrix preset and recovers matrix, and the process of generation comprises: row corresponding with failed disk in RAID extended matrix are carried out diagonalization process and recover matrix to generate described RAID; The diagonalization check matrix that described RAID extended matrix is arranged by RAID generator matrix and the capable j of j docks and forms; In described check matrix, every a line comprises an effectively mark, and this position be effectively identified in check matrix is corresponding with the position of syndrome unit in band that RAID generator matrix docks row corresponding.

Further, also comprise reconstruction process module, wherein this reconstruction process module uses described default checking algorithm to recover checking data, and concrete formula is:

$c_v=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, c _vit is the checking data in the syndrome unit of inaccessible in a recovery group; c _u... be the checking data in this recovery group in addressable syndrome unit; d _x... represent the user data in addressable data sub-element in this recovery group; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

The present invention also provides a kind of RAID data processing method, be applied in computer system, this RAID comprises n+m member's disk, wherein each stick of this RAID is all divided into w subelement, in this RAID, each band comprises k according to subelement and j syndrome unit, wherein j=wxm, k=wxn, wherein n, m and w are natural number, and w is greater than 1; Each syndrome unit is under the jurisdiction of a verification group, and each verification group also comprises the data sub-element set of multiple data sub-element composition, and in each band, the data sub-element set of each verification group is different; The position of multiple data sub-element in each band in each verification group in band uniquely corresponds to the position of multiple effective mark in the corresponding a line of the RAID generator matrix preset; The method comprises the steps:

Steps A, processes the write order from user side, when fault member number of disks is less than or equal to m, determines the data sub-element of the needs write that this write order is corresponding; Upgrade the user data in the data sub-element needing write, and use default checking algorithm to upgrade in verification group belonging to this data sub-element all the checking data verified in subelement;

Step B, process the read command from user side, when fault member number of disks is less than or equal to m, determine the data sub-element of the needs reading that this read command is corresponding, if desired the data sub-element read is positioned on member's disk of normal condition, corresponding user data is read from this data sub-element, if desired the data sub-element read is positioned on fault member disk, data then in recovery group belonging to this data sub-element in current addressable data sub-element and syndrome unit use described default checking algorithm to recover user data in this data sub-element.

Further, the concrete formula using default checking algorithm to upgrade checking data is:

$c_h=d_x\⊕......\⊕d_y$

Wherein in this formula, the subelement at all data places forms a verification group, and h is the natural number being less than or equal to j; X and y is the natural number being less than or equal to k; c _hrepresent in this band the checking data verified in h verification group in subelement; d _x... d _yfor the user data in the multiple data sub-element in this verification group, wherein in data sub-element set the position of each data sub-element in band correspond respectively to h in described RAID generator matrix capable in the position of multiple effective mark; represent XOR.

Further, the generator matrix of described RAID is the matrix of the capable k row of j, the wherein capable i-th column element a of h _hii-th data sub-element in corresponding band, if a _hibe effective mark, represent that in band, i-th data sub-element belongs to h verification group; If a _hifor invalidated identification, represent that i-th data sub-element does not belong to h verification group.

Further, between the generator matrix of described RAID and selected m capable n row Galois Field constant matrices, there are mapping relations, its these mapping relations are: the matrix generated after constant each in this constant matrices being converted to the embedded matrix of w capable w row is this RAID generator matrix, and wherein w is this Galois Field constant matrices Galois Field bit wide.

Further, described constant matrices is full rank, and the anyon matrix of this matrix is also full rank.

Further, wherein said n is more than or equal to m, and n and m sum is less than or equal to the w power of 2.

Further, the concrete formula of described default checking algorithm restoring user data is used to be:

$d_w=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, d _wit is the user data in the data sub-element of inaccessible in a recovery group; Cu...... be the checking data in this recovery group in addressable syndrome unit; Dx...... the user data in addressable data sub-element in this recovery group is represented; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

Further, described RAID recovers the matrix that matrix is the capable s row of j, wherein s=w × (m+n); Described RAID recovers the capable i-th column element a of h in matrix _hii-th subelement in corresponding band, if a _hibe effective mark, represent that in band, i-th subelement belongs to h recovery group; If a _hifor invalidated identification, represent that i-th subelement does not belong to h verification group.

Further, also comprise:

Step C, generate described RAID according to the RAID extended matrix preset and recover matrix, the process of generation comprises: row corresponding with failed disk in RAID extended matrix are carried out diagonalization process and recover matrix to generate described RAID; The diagonalization check matrix that described RAID extended matrix is arranged by RAID generator matrix and the capable j of j docks and forms; In described check matrix, every a line comprises an effectively mark, and this position be effectively identified in check matrix is corresponding with the position of syndrome unit in band that RAID generator matrix docks row corresponding.

Further, also comprise: step D, in process of reconstruction, use described default checking algorithm to recover checking data, and concrete formula be:

$c_v=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, c _vit is the checking data in the syndrome unit of inaccessible in a recovery group; c _u... be the checking data in this recovery group in addressable syndrome unit; d _x... represent the user data in addressable data sub-element in this recovery group; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

In terms of existing technologies, present invention achieves the RAID that same checking algorithm supports multiple check disk, and verification calculating process is relatively simple, does not need complicated computation process.

Accompanying drawing explanation

Fig. 1 is Organization of Data relation schematic diagram in RAID band in prior art.

Fig. 2 is the logical organization schematic diagram of RAID data processing equipment in one embodiment of the present invention.

Fig. 3 is the schematic diagram generating RAID generator matrix in one embodiment of the present invention.

Fig. 4 is data relationship schematic diagram in a verification group in one embodiment of the present invention.

RAID generator matrix schematic diagram corresponding when Fig. 5 is disk failure in one embodiment of the present invention.

Fig. 6 generates the diagonalization process schematic that RAID recovers matrix in one embodiment of the present invention.

Embodiment

For the shortcoming of prior art, the present invention proposes a kind of brand-new RAID implementation, when using single algorithm to support that multiple disk breaks down simultaneously, still can recover data, and to multiple faulty disk without requirement in logic, promote dirigibility and the robustness of RAID use.In realization, relative to using the design of stick in traditional RAID band as data cell or verification unit, stick is further subdivided into less subelement by the present invention, the subelement depositing user data is called data sub-element, in the checking data write syndrome unit that in multiple data sub-element, data XOR obtains, data sub-element and corresponding syndrome unit form a verification group.In the following description, unless otherwise noted, the letter of all expression quantity is natural number.

Please refer to Fig. 2, in a preferred embodiment, the invention provides a kind of enhancement mode RAID data processing equipment be applied in computer system, this RAID comprises n+m member's disk, and wherein each stick of this RAID is all divided into w subelement, and in this RAID, each band comprises k according to subelement and j syndrome unit, wherein j=wxm, k=wxn, wherein n, m and w are natural number, and w is greater than 1; Each syndrome unit is under the jurisdiction of a verification group, and each verification group also comprises the data sub-element set of multiple data sub-element composition, and in each band, the data sub-element set of each verification group is different; This device of position that the position of multiple data sub-element in each band in each verification group in band uniquely corresponds to multiple effective mark in the corresponding a line of the RAID generator matrix preset comprises writing process module and reads processing unit.This RAID treating apparatus comprises: writing process module and read processing module.Described computer system refers to the equipment including several member's disks, and it may be the computing system of arbitrary form, such as workstation, or server, or the special network storage equipment, may be even individual main frame.It should be noted that RAID data processing equipment can use software simulating, also can use hardware implementing, the forms such as such as RAID plug-in card, the mode that even soft or hard can be used to combine realizes.The processing procedure of this device comprises the steps:

Writing process module, for the treatment of the write order from user side, when fault member number of disks is less than or equal to m, determines the data sub-element of the needs write that this write order is corresponding; Upgrade the user data in the data sub-element needing write, and use default checking algorithm to upgrade in verification group belonging to this data sub-element all the checking data verified in subelement;

Read processing module, for the treatment of the read command from user side, when fault member number of disks is less than or equal to m, determine the data sub-element of the needs reading that this read command is corresponding, if desired the data sub-element read is positioned on member's disk of normal condition, corresponding user data is read from this data sub-element, if desired the data sub-element read is positioned on fault member disk, data then in recovery group belonging to this data sub-element in current addressable data sub-element and syndrome unit use described default checking algorithm to recover user data in this data sub-element.

N represents the number of the data disks in a RAID band as previously mentioned, and m represents the verification unit number in a RAID band, i.e. the upper limit of the patient failed disk number of this RAID.Except n and m, also introduce this parameter of Galois Field bit wide w, when being below not particularly illustrated, each parameter characterizing quantity is natural number.The basic constraint condition of the usual demand fulfillment of the above-mentioned parameter in RAID formula (1) and (2):

n≥m （1）

2 ^w≥n+m （2）

In the present invention, will comprise w × m verification group in a band of RAID, each verification group comprises a syndrome unit.In verification group, the corresponding relation of data sub-element and syndrome unit can be expressed by a RAID generator matrix.In Fig. 4, this RAID generator matrix derives from the original constant matrices that a capable n of m arranges, and the element in this original constant matrices is made up of the element selected in Galois Field, and usually meets following 2 conditions:

Condition one: constant matrices is full rank, namely constant matrices is reversible, each provisional capital is linearly incoherent.

Condition two: the anyon matrix of constant matrices is also full rank, namely any submatrix of constant matrices is reversible, each provisional capital is linearly incoherent.

The original constant matrices meeting above-mentioned condition has multiple, developer can select when realizing wherein a kind of matrix as original constant matrices.In the original constant matrices chosen, each constant can be exchanged into a unique corresponding capable w of w and arranges the embedded matrix be made up of " 0 " and " 1 ", obtains matrix and be called RAID generator matrix after all original constant matrices conversions.This RAID generator matrix is the capable k column matrix of j, wherein, j=(w × m), k=(w × n), as shown in formula (3):

$\left(\begin{array}{ccccc}{a}_{11}& a_{12}& a_{13}& \·\·\·& a_{1k}\\ a_{21}& a_{22}& a_{23}& \·\·\·& a_{2k}\\ \·\·\·& \·\·\·& \·\·\·& \·\·\·& \·\·\·\\ a_{j1}& a_{j2}& a_{j3}& \·\·\·& a_{\mathrm{jk}}\end{array}\right)---\left(3\right)$

The capable i-th column element a of h in RAID generator matrix _hii-th data sub-element in corresponding band, if a _hibe effective mark, represent that in band, i-th data sub-element belongs to h verification group; If a _hifor invalidated identification, represent that i-th data sub-element does not belong to h verification group; Wherein a in preferred mode _hivalue be 0 or 1 respectively corresponding invalidated identification and effectively mark.In a band of RAID, all data sub-element are numbered in order, and all syndrome unit are numbered in order.The line number of RAID generator matrix represents the numbering of syndrome unit, and in this row, the element of all numerical value 1 constitutes an effective identified group.After having had RAID generator matrix, just can come to verify the checking data of subelement in calculation check group according to formula (4).

$c_h=d_x\⊕......\⊕d_y---\left(4\right)$

Wherein in this formula (4), the subelement at all data places forms a verification group, and h is the natural number being less than or equal to j; X and y is the natural number being less than or equal to k; c _hrepresent in this band the checking data verified in h verification group in subelement; d _x... d _yfor the user data in the multiple data sub-element in this verification group, wherein in data sub-element set the position of each data sub-element in band correspond respectively to h in described RAID generator matrix capable in the position of multiple effective mark; represent XOR.

Be described for the RAID of 8 dish compositions below.Suppose that this RAID comprises 5 data disks; 3 check disks; Band 3 decile; Now n=5, m=3, w=3, j==m × w=9, k=n × w=15, supposes that original constant matrices is as shown in formula (5):

$\left(\begin{array}{ccccc}1& 5& 2& 7& 4\\ 5& 1& 3& 4& 7\\ 4& 7& 6& 5& 1\end{array}\right)---\left(5\right)$

Please refer to the transformational relation shown in Fig. 3, constant each in original constant matrices is done embedded matrix and transform, this original constant matrices obtains corresponding RAID generator matrix as shown in formula (6) after above-mentioned conversion:

$\left(\begin{array}{ccccc}100& 110& 001& 111& 010\\ 010& 001& 101& 100& 011\\ 001& 100& 010& 110& 101\\ 110& 100& 101& 010& 111\\ 001& 010& 111& 011& 100\\ 100& 001& 011& 101& 110\\ 010& 111& 011& 110& 100\\ 011& 100& 110& 001& 010\\ 101& 110& 111& 100& 001\end{array}\right)---\left(6\right)$

Please refer to shown in Fig. 4, in supposing, de, f represent the user data in data sub-element, and cp, q represent the checking data in syndrome unit.According to above-mentioned RAID generator matrix, the relation in each verification group between checking data and user data can be described by 9 verification groups in formula group (7).

$\begin{array}{c}c\mathrm{0,0}=d\mathrm{0,0}\⊕d\mathrm{1,0}\⊕d\mathrm{1,1}\⊕d\mathrm{2,2}\⊕d\mathrm{3,0}\⊕d\mathrm{3,1}\⊕d\mathrm{3,2}\⊕d\mathrm{4,1}\\ c\mathrm{0,1}=d\mathrm{0,1}\⊕d\mathrm{1,2}\⊕d\mathrm{2,0}\⊕d\mathrm{2,2}\⊕d\mathrm{3,0}\⊕d\mathrm{4,1}\⊕d\mathrm{4,2}\\ c\mathrm{0,2}=d\mathrm{0,2}\⊕d\mathrm{1,0}\⊕d\mathrm{2,1}\⊕d\mathrm{3,0}\⊕d\mathrm{3,1}\⊕d\mathrm{4,0}\⊕d\mathrm{4,2}\\ c\mathrm{1,0}=d\mathrm{0,0}\⊕d\mathrm{0,1}\⊕d\mathrm{1,0}\⊕d\mathrm{2,0}\⊕d\mathrm{2,2}\⊕d\mathrm{3,1}\⊕d\mathrm{4,0}\⊕d\mathrm{4,1}\⊕d\mathrm{4,2}\\ c\mathrm{1,1}=d\mathrm{0,2}\⊕d\mathrm{1,1}\⊕d\mathrm{2,0}\⊕d\mathrm{2,1}\⊕d\mathrm{2,2}\⊕d\mathrm{3,1}\⊕d\mathrm{3,2}\⊕d\mathrm{4,0}\\ c\mathrm{1,2}=d\mathrm{0,0}\⊕d\mathrm{1,2}\⊕d\mathrm{2,1}\⊕d\mathrm{2,2}\⊕d\mathrm{3,0}\⊕d\mathrm{3,2}\⊕d\mathrm{4,0}\⊕d\mathrm{4,1}\\ c\mathrm{2,0}=d\mathrm{0,1}\⊕d\mathrm{1,0}\⊕d\mathrm{1,1}\⊕d\mathrm{1,2}\⊕d\mathrm{2,1}\⊕d\mathrm{2,2}\⊕d\mathrm{3,0}\⊕d\mathrm{3,1}\⊕d\mathrm{4,0}\\ c\mathrm{2,1}=d\mathrm{0,1}\⊕d\mathrm{0,2}\⊕d\mathrm{1,0}\⊕d\mathrm{2,0}\⊕d\mathrm{2,1}\⊕d\mathrm{3,2}\⊕d\mathrm{4,1}\\ c\mathrm{2,2}=d\mathrm{0,0}\⊕d\mathrm{0,2}\⊕d\mathrm{1,0}\⊕d\mathrm{1,1}\⊕d\mathrm{2,0}\⊕d\mathrm{2,1}\⊕d\mathrm{2,2}\⊕d\mathrm{3,0}\⊕d4,2\end{array}---\left(7\right)$

As can be seen from Figure 4, in first verification group, check that the 1st row of RAID generator matrix can be known, 1st, the numerical value of 4,5,9,10,11,12,14 is 1, and these 8 numerical value can be understood as the effective identified group in the RAID generator matrix corresponding with first syndrome unit in the 1st row.So participate in checking data c0 in calculating first verification group in band 1, the user data of 0 is respectively the user data in the 1st, 4,5,9,10,11,12,14 data sub-element, and then can obtain first operation relation in formula group (7).The account form of other checking datas is similar, repeats no more.

In RAID, all disks are all in place and in normal situation, when receiving read command from user's side joint, directly read the user data in corresponding data sub-element.When receiving write order from user's side joint, need according to the first calculation check data of formula (7), can according to step process below:

(1) determine to write the verification group involved by data, determine the data sub-element user data needing in those verification groups to participate in calculating, for the data sub-element without the need to write, the user data in this data sub-element of retaking of a year or grade;

(2) the new checking data of those verification groups is calculated according to formula (7);

(3) data sub-element corresponding for new user data write, checking data is write corresponding syndrome unit.

It should be noted that, in the present invention, the user data upgraded in certain data sub-element may affect multiple verification group, therefore will obtain multiple verification computing formula in formula (7) during calculation check data, calculate the checking data in multiple verification group, and be written to corresponding verification unit.

As previously mentioned, RAID of the present invention supports m check disk, namely 1 arrive m disk failure situation under, user data all can not be lost.Under normal circumstances, read user data and do not relate to checking data; But after disk failure, read write command will relate to checking treatment.When receiving read command from user's side joint, if the data cell that will access of read command may have access to, can direct sense data; If the data cell of read command hit can not be accessed, then need to be recovered by the data of addressable data cell and verification unit.Same reason, if having disk failure in RAID, can have influence on the process of write order equally, and as previously mentioned, write order may relate to the user data in back read data subelement, and this user data cannot may read because of disk failure.No matter be process in read command or write order process, if when the user data in data sub-element cannot read, following manner can be adopted to recover user data.

Continue, with identical example, restoring user data read procedure is described, suppose disk3 fault, correspond to band 1, namely stick D3 loses, and comprises d3, and 0, d3,1 and d3, the user data in 2 three data sub-element.The formula (8) choosing first three the verification group in verification formula group (7) is solved an equation, d3 in equation, and 0, d3,1 and d3,2 is unknown number, and other are given data.

$\begin{array}{c}{d}_{\mathrm{3,0}}\⊕d_{\mathrm{3,1}}\⊕d_{\mathrm{3,2}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{1,0}}\⊕d_{\mathrm{1,1}}\⊕d_{\mathrm{2,2}}\⊕d_{\mathrm{4,1}}\⊕c_{\mathrm{0,0}}\\ d_{\mathrm{3,0}}=d_{0,1}\⊕d_{1,2}\⊕d_{2,0}\⊕d_{\mathrm{2,2}}\⊕d_{\mathrm{4,1}}\⊕d_{\mathrm{4,2}}\⊕c_{0,1}\\ d_{\mathrm{3,0}}\⊕d_{\mathrm{3,1}=}{d}_{0,2}\⊕d_{1,0}\⊕d_{2,1}\⊕d_{4,0}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,2}}\end{array}---\left(8\right)$

Separate formula (8), the 2nd row is added to other two row, and in fact adding here be exactly xor operation, obtains formula (9):

$\begin{array}{c}{d}_{\mathrm{3,1}}\⊕d_{\mathrm{3,2}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,1}}\⊕d_{\mathrm{1,0}}\⊕d_{\mathrm{1,1}}\⊕d_{\mathrm{1,2}}\⊕d_{\mathrm{2,0}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,1}}\\ d_{3,0}=d_{\mathrm{0,1}}\⊕d_{1,2}\⊕d_{2,0}\⊕d_{2,2}\⊕d_{4,1}\⊕d_{4,2}\⊕c_{\mathrm{0,1}}\\ d_{\mathrm{3,1}}=d_{0,1}\⊕d_{0,2}\⊕d_{\mathrm{1,0}}\⊕d_{1,2}\⊕d_{2,0}\⊕d_{2,1}\⊕d_{\mathrm{2,2}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,1}}\⊕c_{0,1}\⊕c_{\mathrm{0,2}}\end{array}---\left(9\right)$

Again the 3rd row in formula (9) is added on the 1st row, obtains formula (10):

$\begin{array}{c}{d}_{\mathrm{3,2}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{1,1}}\⊕d_{\mathrm{2,1}}\⊕d_{\mathrm{2,2}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,1}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,2}}\\ d_{3,0}=d_{\mathrm{0,1}}\⊕d_{1,2}\⊕d_{2,0}\⊕d_{\mathrm{2,2}}\⊕d_{4,1}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,1}}\⊕\\ d_{3,1}=d_{0,1}\⊕d_{\mathrm{0,2}}\⊕d_{1,0}\⊕d_{1,2}\⊕d_{2,0}\⊕d_{2,1}\⊕d_{2,2}\⊕d_{4,0}\⊕d_{4,1}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{0,2}}\end{array}---\left(10\right)$

D3 can be calculated, 0, d3,1 and d3, data original in 2 according to formula (10).

Introduce the date restoring under a kind of extreme case below again, suppose m disk failures in RAID.Continue the example presented above, assumed lost be disk1, disk2, disk3 tri-data disks, correspond to band 1, the data of losing comprise d1, and 0, d1,1, d1,2, d2,0, d2,1, d2,2, d3,0, d3,1, d3,2 totally 9 data cells, according to above-mentioned thinking, change verification group formula can obtain formula (11):

$\begin{array}{c}{d}_{\mathrm{1,0}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,1}}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{1,2}}\⊕c_{\mathrm{2,1}}\\ d_{\mathrm{1,1}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,1}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,0}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{0,2}}\⊕c_{\mathrm{1,0}}\⊕c_{\mathrm{1,2}}\⊕c_{\mathrm{2,1}}\⊕c_{\mathrm{2,2}}\\ d_{\mathrm{1,2}}=d_{\mathrm{0,1}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,1}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,2}}\⊕c_{\mathrm{1,1}}\⊕c_{\mathrm{2,0}}\⊕c_{\mathrm{2,2}}\\ d_{\mathrm{2,0}}=d_{\mathrm{0,1}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{0,2}}\⊕c_{\mathrm{1,1}}\⊕c_{\mathrm{1,2}}\⊕c_{\mathrm{2,2}}\\ d_{\mathrm{2,1}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,1}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,1}}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{1,0}}\⊕c_{\mathrm{1,1}}\⊕c_{\mathrm{2,0}}\⊕c_{\mathrm{2,2}}\\ d_{\mathrm{2,2}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,1}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,1}}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{0,2}}\⊕c_{\mathrm{1,0}}\⊕c_{\mathrm{1,1}}\⊕c_{\mathrm{1,2}}\⊕c_{\mathrm{2,1}}\\ d_{\mathrm{3,0}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,1}}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{0,2}}\⊕c_{\mathrm{1,0}}\⊕c_{\mathrm{1,1}}\⊕c_{\mathrm{2,0}}\⊕c_{\mathrm{2,1}}\\ d_{\mathrm{3,1}}=d_{\mathrm{0,0}}\⊕d_{\mathrm{0,1}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,1}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{1,2}}\⊕c_{\mathrm{2,2}}\\ d_{\mathrm{3,2}}=d_{\mathrm{0,1}}\⊕d_{\mathrm{0,2}}\⊕d_{\mathrm{4,0}}\⊕d_{\mathrm{4,1}}\⊕d_{\mathrm{4,2}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{0,2}}\⊕c_{\mathrm{1,0}}\⊕c_{\mathrm{2,0}}\end{array}---\left(11\right)$

Solve an equation can calculate d1 according to formula (11), 0, d1,1, d1,2, d2,0, d2,1, d2,2, d3,0, d3,1, d3, data original in 2.Can be found out by above two kinds of examples, as long as the quantity of failed disk is less than or equal to m, the present invention can go out need based on same checking algorithm Reverse recovery the user data of recovery, only the quantity of failed disk is larger, in computation process, the calculated amount of backwards calculation is corresponding larger, solve formula more complicated, but the algorithm in solution procedure is still the same, just direction oppositely solves.

Furthermore, the problem of calculation of complex is caused when considering that failed disk is more, the present invention provides a kind of more succinct reset mode in a preferred embodiment, and it remains above-mentioned checking algorithm in essence, just introduces RAID recovery matrix and carrys out simplified operation process.Based on above-mentioned RAID generator matrix, be pre-created a RAID extended matrix.In the present embodiment, said apparatus also comprises matrix disposal module, this matrix disposal module is used for generating described RAID according to the RAID extended matrix preset and recovers matrix, and the process of generation comprises: row corresponding with failed disk in RAID extended matrix are carried out diagonalization process and recover matrix to generate described RAID; The diagonalization check matrix that described RAID extended matrix is arranged by RAID generator matrix and the capable j of j docks and forms; In described check matrix, every a line comprises an effectively mark, and this position be effectively identified in check matrix is corresponding with the position of syndrome unit in band that RAID generator matrix docks row corresponding.

As previously mentioned, this RAID extended matrix to be docked with check matrix by RAID generator matrix and forms.This RAID extended matrix is j × r, wherein r=k+j.A RAID extended matrix is created based on example in Fig. 4, this matrix is example as Fig. 5, and suppose now disk1 and disk2 two disk failures, composition graphs 4 is known, in recovery matrix, in 4-9 row, any one effectively identifies and all points to a data sub-element that cannot access, namely d _1,0, d _1,1, d _1,2, d _2,0, d _2,1, d _2,2these six user data place user data subelements cannot be accessed.In the present embodiment, diagonalization of matrix process is carried out around 4-9 row, final target is that realization matrix has 4-9 row in 6 row all to only have an effectively mark, namely only have the user data in a data sub-element to be unknown data, next once verify computing and can solve and obtain this user data.

Please refer to the diagonalization of matrix process shown in Fig. 6, its target is that 4-9 row are carried out diagonalization process, and namely corresponding with failed disk in RAID extended matrix row carry out diagonalization process.For the 4th row, first choose a line, selected line must have criterion and know on the 4th, can find that the 1st row the 4th data are 1 from Fig. 6, has criterion and knows.Now the 4th row can be understood as and treat zero row, and target allows the 4th to arrange the element value of other row except the 1st row all make zero, and namely becomes invalidated identification.Now with the 1st row respectively with the 3rd, 4,7,8,9 row carry out XOR, the newline gone out with XOR replace respectively original 3rd, 4,7,8,9 row, replaced rear formation first intermediary matrix.Next using the 5th row as treating zero row, then choose a line, require that the 4th column data of selected line is 0, the 5th column data is 1, now chooses behavior the 3rd row, with the 3rd row respectively with the 1st, 4,5,8 row carry out XOR, replace rear formation the second intermediary matrix.Circulation like this is gone down until the 9th row are also made zero, and final formation RAID recovers matrix.As can be seen from Figure 6, RAID recover matrix 4-9 arrange this six row in be present diagonalizable, namely wherein any one row on only have an element to be that 1(effectively identifies).This effectively identifies institute and is expert in (computing row), and only have this element to be 1 in 4-9 element, other are all 0(invalidated identification).So because other arrange the disk that corresponding disk is all normal condition, computing row only have a data sub-element that effectively mark is corresponding cannot access, and other effective data identified in corresponding data sub-element or syndrome unit all can be accessed, therefore can quickly move through XOR algorithm and the user data in this data sub-element is recovered.

Can sum up from above process, user data when disk failure on it in any data sub-element can recover according to formula (12).

$d_w=d_x\⊕......\⊕c_u......---\left(12\right)$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, d _wit is the user data in the data sub-element of inaccessible in a recovery group; Cu...... be the checking data in this recovery group in addressable syndrome unit; Dx...... the user data in addressable data sub-element in this recovery group is represented; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

Please continue to refer to Fig. 6, the 4th element RAID being recovered to the 1st row of matrix is 1, is effective mark, shows the user data d in the 4th data sub-element in band _1,0participate in recovery group verification computing, because it is on failed disk disk1, therefore this user data is exactly the data that will recover.All the other need the data participating in recovery group verification computing to be d successively _0,1, d _3,0, d _3,1, d _3,2, c _0,0, c _0,1, c _1,0, c _2,0, because these data are all on normal disk, be all known, bring in formula 11 and can solve d1,0, please refer to the example of formula 13.

$d_{\mathrm{1,0}}=d_{\mathrm{0,1}}\⊕d_{\mathrm{3,0}}\⊕d_{\mathrm{3,1}}\⊕d_{\mathrm{3,2}}\⊕c_{\mathrm{0,0}}\⊕c_{\mathrm{0,1}}\⊕c_{\mathrm{1,0}}\⊕c_{\mathrm{2,0}}---\left(13\right)$

As can be seen from above description, the position of subelement in band participating in the data place of computing in a recovery group is that the effective identified group recovered in matrix a line by RAID determines.And RAID recovery matrix generates according to RAID extended matrix as previously mentioned.As previously mentioned, the process of generation is that row corresponding with failed disk each in RAID extended matrix are carried out diagonalization process, finally makes RAID recover in matrix, any row to only have a data sub-element that effectively mark is corresponding to be inaccessible.It should be noted that RAID extended matrix can substitute RAID generator matrix in fact, because include RAID generator matrix in RAID extended matrix when realization.In aforementioned calculation check group checking data embodiment in also can use RAID extended matrix, and in fact formula (7) has reflected a RAID extended matrix in essence.

Can be found out by above embodiment, the present invention not only can use single algorithm to support that multiple disk breaks down simultaneously, and this algorithm is fairly simple XOR.In general, invention increases dirigibility and the stalwartness row of RAID, and implementation cost is lower.Comprise eight disks for a RAID, in the scene needing high reliability, n can be 4, and m can be 4, if now in RAID the disk of simultaneous faults be no more than 4, so user data can not be lost, and 4 faulty disks can be any disks, without logical requirements.In the scene needing reliability more balanced with volume ratio, n can be 6, and m can be 2; Suppose that all disk sizes are all identical, the useful capacity so can depositing user data can reach 75% of all disk sizes.In the certain reliability of needs, but in the scene of limited consideration capacity, n can be 7, and m can be 1.For the developer of field of storage, it no longer needs because the difference of RAID rank, and algorithm is much overlapped in exploitation and maintenance, significantly can reduce cost of development.And for user, it has broken the inflexible problem of m quantity in prior art, and in RAID, disk is more, and its flexibility problem is more outstanding, and the present invention there is no the restriction of this respect.

Furthermore, the present invention can further include synchronous processing module and reconstruction process module, for RAID synchronizing characteristics, it is calculation check data write syndrome unit, therefore aforementioned formula can be utilized equally to calculate in each verification group the checking data verified in subelement, the syndrome unit that write is corresponding.Certainly, in the present invention, a band exists multiple verification group, need to calculate respectively the data of multiple verification unit and write, in this band, the checking data of all verification unit writes successfully, just thinks that this band synchronously completes.Rebuild for RAID, can calculate data original in the data cell or verification unit that can not access equally with reference to the process of aforementioned recovery data, write reconstruction is coiled.

In a preferred embodiment, the present invention also comprises synchronous processing module and reconstruction process module.Described synchronous processing module, the checking data calculated in each syndrome unit according to above-mentioned same checking algorithm realizes the synchronous of RAID.Described reconstruction process module is suitable for above-mentioned checking algorithm equally and is recovered by the user data in data sub-element each in failed disk and be written in data sub-element corresponding on HotSpare disk.In process of reconstruction, for the checking data in failed disk; This reconstruction process module can use described checking algorithm will verify date restoring out equally, and its concrete formula is formula (13):

$c_v=d_x\⊕......\⊕c_u......---\left(13\right)$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, c _vit is the checking data in the syndrome unit of inaccessible in a recovery group; c _u... be the checking data in this recovery group in addressable syndrome unit; d _x... represent the user data in addressable data sub-element in this recovery group; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.More specifically realize, its principle is consistent with aforementioned restoring user data process, repeats no more.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.

Claims (20)

1. a RAID data processing equipment, be applied in computer system, this RAID comprises n+m member's disk, wherein each stick of this RAID is all divided into w subelement, in this RAID, each band comprises k according to subelement and j syndrome unit, wherein j=wxm, k=wxn, wherein n, m and w are natural number, and w is greater than 1; Each syndrome unit is under the jurisdiction of a verification group, and each verification group also comprises the data sub-element set of multiple data sub-element composition, and in each band, the data sub-element set of each verification group is different; The position of multiple data sub-element in each band in each verification group in band uniquely corresponds to the position of multiple effective mark in the corresponding a line of the RAID generator matrix preset; This device comprises writing process module and reads processing unit; It is characterized in that:

Writing process module, for the treatment of the write order from user side, when fault member number of disks is less than or equal to m, determines the data sub-element of the needs write that this write order is corresponding; Upgrade the user data in the data sub-element needing write, and use default checking algorithm to upgrade in verification group belonging to this data sub-element all the checking data verified in subelement;

Read processing module, for the treatment of the read command from user side, when fault member number of disks is less than or equal to m, determine the data sub-element of the needs reading that this read command is corresponding, if desired the data sub-element read is positioned on member's disk of normal condition, corresponding user data is read from this data sub-element, if desired the data sub-element read is positioned on fault member disk, data then in recovery group belonging to this data sub-element in current addressable data sub-element and syndrome unit use described default checking algorithm to recover user data in this data sub-element.

2. device as claimed in claim 1, is characterized in that: the concrete formula using default checking algorithm to upgrade checking data is:

$c_h=d_x\⊕......\⊕d_y$

Wherein in this formula, the subelement at all data places forms a verification group, and h is the natural number being less than or equal to j; X and y is the natural number being less than or equal to k; c _hrepresent in this band the checking data verified in h verification group in subelement; d _x... d _yfor the user data in the multiple data sub-element in this verification group, wherein in data sub-element set the position of each data sub-element in band correspond respectively to h in described RAID generator matrix capable in the position of multiple effective mark; represent XOR.

3. device as claimed in claim 2, is characterized in that: the generator matrix of described RAID is the matrix of the capable k row of j, the wherein capable i-th column element a of h _hii-th data sub-element in corresponding band, if a _hibe effective mark, represent that in band, i-th data sub-element belongs to h verification group; If a _hifor invalidated identification, represent that i-th data sub-element does not belong to h verification group.

4. device as claimed in claim 3, it is characterized in that: between the generator matrix of described RAID and selected m capable n row Galois Field constant matrices, there are mapping relations, its these mapping relations are: the matrix generated after constant each in this constant matrices being converted to the embedded matrix of w capable w row is this RAID generator matrix, and wherein w is this Galois Field constant matrices Galois Field bit wide.

5. device as claimed in claim 4, it is characterized in that: described constant matrices is full rank, and the anyon matrix of this matrix is also full rank.

6. device as claimed in claim 1, it is characterized in that: wherein said n is more than or equal to m, n and m sum is less than or equal to the w power of 2.

7. device as claimed in claim 1, is characterized in that: use the concrete formula of described default checking algorithm restoring user data to be:

$d_w=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, d _wit is the user data in the data sub-element of inaccessible in a recovery group; c _u... be the checking data in this recovery group in addressable syndrome unit; d _x... represent the user data in addressable data sub-element in this recovery group; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

8. device as claimed in claim 7, is characterized in that: described RAID recovers the matrix that matrix is the capable s row of j, wherein s=w × (m+n); Described RAID recovers the capable i-th column element a of h in matrix _hii-th subelement in corresponding band, if a _hibe effective mark, represent that in band, i-th subelement belongs to h recovery group; If a _hifor invalidated identification, represent that i-th subelement does not belong to h verification group.

9. device as claimed in claim 8, it is characterized in that: also comprise matrix disposal module, this matrix disposal module is used for generating described RAID according to the RAID extended matrix preset and recovers matrix, and the process of generation comprises: row corresponding with failed disk in RAID extended matrix are carried out diagonalization process and recover matrix to generate described RAID; The diagonalization check matrix that described RAID extended matrix is arranged by RAID generator matrix and the capable j of j docks and forms; In described check matrix, every a line comprises an effectively mark, and this position be effectively identified in check matrix is corresponding with the position of syndrome unit in band that RAID generator matrix docks row corresponding.

10. device as claimed in claim 7, is characterized in that: also comprise reconstruction process module, and wherein this reconstruction process module uses described default checking algorithm to recover checking data, and concrete formula is:

$c_v=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, c _vit is the checking data in the syndrome unit of inaccessible in a recovery group; c _u... be the checking data in this recovery group in addressable syndrome unit; d _x... represent the user data in addressable data sub-element in this recovery group; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

11. 1 kinds of RAID data processing methods, be applied in computer system, this RAID comprises n+m member's disk, wherein each stick of this RAID is all divided into w subelement, in this RAID, each band comprises k according to subelement and j syndrome unit, wherein j=wxm, k=wxn, wherein n, m and w are natural number, and w is greater than 1; Each syndrome unit is under the jurisdiction of a verification group, and each verification group also comprises the data sub-element set of multiple data sub-element composition, and in each band, the data sub-element set of each verification group is different; The position of multiple data sub-element in each band in each verification group in band uniquely corresponds to the position of multiple effective mark in the corresponding a line of the RAID generator matrix preset; It is characterized in that, the method comprises the steps:

Steps A, processes the write order from user side, when fault member number of disks is less than or equal to m, determines the data sub-element of the needs write that this write order is corresponding; Upgrade the user data in the data sub-element needing write, and use default checking algorithm to upgrade in verification group belonging to this data sub-element all the checking data verified in subelement;

Step B, process the read command from user side, when fault member number of disks is less than or equal to m, determine the data sub-element of the needs reading that this read command is corresponding, if desired the data sub-element read is positioned on member's disk of normal condition, corresponding user data is read from this data sub-element, if desired the data sub-element read is positioned on fault member disk, data then in recovery group belonging to this data sub-element in current addressable data sub-element and syndrome unit use described default checking algorithm to recover user data in this data sub-element.

12. methods as claimed in claim 11, is characterized in that: the concrete formula using default checking algorithm to upgrade checking data is:

$c_h=d_x\⊕......\⊕d_y$

Wherein in this formula, the subelement at all data places forms a verification group, and h is the natural number being less than or equal to j; X and y is the natural number being less than or equal to k; c _hrepresent in this band the checking data verified in h verification group in subelement; d _x... d _yfor the user data in the multiple data sub-element in this verification group, wherein in data sub-element set the position of each data sub-element in band correspond respectively to h in described RAID generator matrix capable in the position of multiple effective mark; represent XOR.

13. methods as claimed in claim 12, is characterized in that: the generator matrix of described RAID is the matrix of the capable k row of j, the wherein capable i-th column element a of h _hii-th data sub-element in corresponding band, if a _hibe effective mark, represent that in band, i-th data sub-element belongs to h verification group; If a _hifor invalidated identification, represent that i-th data sub-element does not belong to h verification group.

14. methods as claimed in claim 13, it is characterized in that: between the generator matrix of described RAID and selected m capable n row Galois Field constant matrices, there are mapping relations, its these mapping relations are: the matrix generated after constant each in this constant matrices being converted to the embedded matrix of w capable w row is this RAID generator matrix, and wherein w is this Galois Field constant matrices Galois Field bit wide.

15. methods as claimed in claim 14, it is characterized in that: described constant matrices is full rank, and the anyon matrix of this matrix are also full ranks.

16. methods as claimed in claim 11, it is characterized in that: wherein said n is more than or equal to m, n and m sum is less than or equal to the w power of 2.

17. methods as claimed in claim 11, is characterized in that: use the concrete formula of described default checking algorithm restoring user data to be:

$d_w=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, d _wit is the user data in the data sub-element of inaccessible in a recovery group; Cu...... be the checking data in this recovery group in addressable syndrome unit; Dx...... the user data in addressable data sub-element in this recovery group is represented; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.

18. methods as claimed in claim 17, is characterized in that: described RAID recovers the matrix that matrix is the capable s row of j, wherein s=w × (m+n); Described RAID recovers the capable i-th column element a of h in matrix _hii-th subelement in corresponding band, if a _hibe effective mark, represent that in band, i-th subelement belongs to h recovery group; If a _hifor invalidated identification, represent that i-th subelement does not belong to h verification group.

19. methods as claimed in claim 18, is characterized in that: also comprise:

Step C, generate described RAID according to the RAID extended matrix preset and recover matrix, the process of generation comprises: row corresponding with failed disk in RAID extended matrix are carried out diagonalization process and recover matrix to generate described RAID; The diagonalization check matrix that described RAID extended matrix is arranged by RAID generator matrix and the capable j of j docks and forms; In described check matrix, every a line comprises an effectively mark, and this position be effectively identified in check matrix is corresponding with the position of syndrome unit in band that RAID generator matrix docks row corresponding.

20. methods as claimed in claim 17, is characterized in that: also comprise:

Step D, use described default checking algorithm to recover checking data, and concrete formula is in process of reconstruction:

$c_v=d_x\⊕......\⊕c_u......$

Wherein, u is the natural number being less than or equal to j; X is the natural number being less than or equal to k; In this formula, all data form a recovery group, c _vit is the checking data in the syndrome unit of inaccessible in a recovery group; c _u... be the checking data in this recovery group in addressable syndrome unit; d _x... represent the user data in addressable data sub-element in this recovery group; represent XOR; In a recovery group, the position of each subelement on disk corresponds to the multiple effective mark in RAID recovery matrix in a line.