CN104156276B - It is a kind of to prevent two pieces of RAID methods of disk failures - Google Patents
It is a kind of to prevent two pieces of RAID methods of disk failures Download PDFInfo
- Publication number
- CN104156276B CN104156276B CN201410399100.6A CN201410399100A CN104156276B CN 104156276 B CN104156276 B CN 104156276B CN 201410399100 A CN201410399100 A CN 201410399100A CN 104156276 B CN104156276 B CN 104156276B
- Authority
- CN
- China
- Prior art keywords
- disk
- raid
- data
- new
- pieces
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000012795 verification Methods 0.000 claims abstract description 43
- 238000012937 correction Methods 0.000 claims abstract description 7
- 230000006978 adaptation Effects 0.000 claims abstract description 4
- 230000006378 damage Effects 0.000 claims description 16
- 238000011084 recovery Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000001681 protective effect Effects 0.000 abstract 1
- 206010021703 Indifference Diseases 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Landscapes
- Techniques For Improving Reliability Of Storages (AREA)
Abstract
Prevent two pieces of RAID methods of disk failures the invention discloses a kind of, it is related to RAID reconstruction techniques, propose new RAID structure, including original RAID and new check disk, by new check disk, the method of calibration that Diagonal Parity Check and horizontal parity check are combined, when two pieces of disks are damaged in RAID groups simultaneously, first with new verification computing recovered part data, recover another part data with original verification computing again, and self adaptation checks data, verification error correction is carried out, with the data of Recover from damaging disk；The problems such as preventing data corruption and service pause caused by two pieces of disk failures, is effectively improved the checking feature of RAID, is greatly enhanced the reliability and data protective capability of disk array.
Description
Technical field
It is specifically a kind of to prevent two pieces of RAID methods of disk failures the present invention relates to RAID reconstruction techniques.
Background technology
The construction of mass storage system (MSS) is the most popular and with fastest developing speed field of current computer system, the master of storage system
Partly online storage subsystem.RAID (disk array) is for lifting the efficiency of storage system, the high reliability of data, preventing
Data corruption and service pause are significant.
RAID technique experienced a series of transition and development along with the use process of people, gradually form RAID
The grade RAID of 0, RAID 1, RAID 2, RAID3, RAID 4, RAID 5.These ranks are by RAB (RAID Advisory
Borad) determine, it is different in addition with compound RAID such as RAID 0+1, RAID 1+0, RAID 30, RAID 50 etc.
Manufacturer know-how by optimizing, such as RAID-S, AutoRAID, RAID 6, RAID 7 etc. have been used in respective product.
RAID technique in current practice can only all prevent the damage of monolithic disk, occurred in that in actual production
Much because double plate damages the accident for causing business to pause for a long time.Under the RAID of a multitude of names, they all can not
On the premise of high efficiency data access is ensured, effectively prevent two pieces of disks from damaging the huge loss of data for causing simultaneously,
Once loss of data, can only rely on cold standby technology carries out prolonged data recovery, causes overall system reliability to reduce.
The content of the invention
Part, of the invention to propose a kind of anti-two pieces of RAID methods of disk failures in view of the shortcomings of the prior art.
A kind of technical scheme prevented two pieces of RAID methods of disk failures, solve above-mentioned technical problem use of the present invention
It is as follows：The RAID method for preventing two pieces of disk failures includes：The step of building the verification group of new check disk, and two are damaged
The step of bad disk carries out disk and recovers；The step of verification group of the structure new check disk, includes：According to label original
Data are grouped in RAID, give up the process that other groups are constituted certain group the verification group of new check disk；It is described that two are damaged
Bad disk carries out the step of disk recovers to be included：For two different damage disks, disk voluntarily error correction reparation is carried out.
It is of the present invention to prevent two pieces of RAID methods of disk failures, by new check disk, Diagonal Parity Check and level
The method of calibration that even-odd check is combined, when two pieces of disks are damaged in RAID groups simultaneously, is recovered with new verification computing first
Partial data, then recover another part data with original verification computing, and self adaptation checks data, carries out verification error correction, with
The data of Recover from damaging disk, can effectively improve the reliability of disk array.
Further, the method for calibration being combined using Diagonal Parity Check and horizontal parity check, carries out recovery data
When, Diagonal Parity Check and horizontal parity check is used alternatingly.
Further, the RAID method for preventing two pieces of disk failures proposes a kind of new RAID structure, described new
RAID structure includes original RAID and new check disk.
Further, the data disks of the new RAID structure and the composition of check disk are N+1, then the new check disk is necessary
Constitute the matrix group of (N+1+1) × N.
Further, if two pieces to damage disk be disk in new check disk and one piece of original RAID, new verification is not considered
Disk, the disk in original RAID carries out data recovery by original verification computing.
Further, the new RAID structure includes the original RAID containing disk A, B, C, D, E and new check disk F, if
It is the disk C and disk D in original RAID that two pieces are damaged disk, then the data of label 1 are extensive by new verification computing in disk D
Multiple, the data of label 0 are recovered by original verification computing in disk C.
Further, the new RAID structure includes the original RAID containing disk A, B, C, D, E and new check disk F, if
It is the disk A and disk B in original RAID that two pieces are damaged disk, then the data of label 4 are extensive by new verification computing in disk B
Multiple, the data of label 3 are recovered by original verification computing in disk A.
A kind of RAID method for preventing two pieces of disk failures of the present invention has the advantage that compared with prior art：Institute
The RAID method for stating anti-two pieces of disk failures improves new RAID structure, employs new method of calibration, builds new check disk,
The method of calibration being combined by Diagonal Parity Check and horizontal parity check, it is therefore prevented that data caused by two pieces of disk failures
Destruction and the problems such as service pause, is effectively improved the checking feature of RAID, be greatly enhanced the reliability of disk array with
Data protection ability, and improve the reliability of mass storage system (MSS).
Brief description of the drawings
Accompanying drawing 1 prevents two pieces of schematic diagrames of the RAID method of disk failures for described.
Specific embodiment
To make the object, technical solutions and advantages of the present invention become more apparent, below in conjunction with drawings and Examples,
A kind of RAID method for preventing two pieces of disk failures of the invention is described in detail.
In anti-two pieces of RAID methods of disk failures of the present invention, new RAID structure, the new RAID are constructed
Structure includes original RAID and new check disk, and the data source of the new check disk does not include original all disks, but puts
Abandoning one piece of disk does not carry out verification computing.The RAID method for preventing two pieces of disk failures includes：Build the verification of new check disk
The step of group, and the step of damage disks to two and carry out disk and recover.The step of verification group of the structure new check disk
Including：Data in original RAID are grouped according to label, are given up the verification group that other groups are constituted certain group new check disk；
It is described to damage disks to two and carry out the step of disk recovers and include：For two different damage disks, disk is carried out voluntarily
Error correction is repaired.
It is of the present invention to prevent two pieces of RAID methods of disk failures, by new check disk, Diagonal Parity Check and level
The method of calibration that even-odd check is combined, when two pieces of disks are damaged in RAID simultaneously, first with new verification computing recovery section
Divided data, then recover another part data with original verification computing, can self adaptation check data, and carry out verification error correction,
The reliability of disk array can be effectively improved.
Embodiment：
Below by one embodiment, in anti-two pieces of advantages of the RAID method of disk failures of the present invention and design
Hold, be described in detail.
As shown in Figure 1, new RAID structure includes the original RAID containing disk A, B, C, D, E and new check disk F, this reality
Example anti-two pieces of RAID methods of disk failures are applied, label is first according to and data in original RAID is grouped, in new school
Test on disk and give up 0 group, other groups constitute new verification group.No matter original RAID data group is RAID5 or RAID4, wherein institute
Some disks can be treated unanimously.
Two pieces of RAID methods of disk failures are prevented described in the present embodiment, is voluntarily repaiied for two pieces of different disk failures
Multiple situation is as follows：
(1) if two pieces to damage disk be disk in new check disk and one piece of original RAID
Because data are the check informations of redundancy in new check disk, can not consider, and original RAID can prevent one
The damage of block disk, i.e., the disk in original RAID is recovered by original verification computing, therefore in this case without data
Lose；
(2) if two pieces to damage disk be two pieces of disks in original RAID, this two pieces are damaged disks all comprising new verification
Given up 0 group of disk
If the disk C and disk D in original RAID are damaged, then the data of label 1 can be by new in disk D simultaneously
Verification computing recovers, and the data of label 0 can be recovered by original verification computing in disk C；Once analogize, be successively：Magnetic
The data of label 0 in disk D, the data of label 4 in disk C；The data of label 4 in disk D, the data of label 3 in disk C；Magnetic
The data of label 3 in disk D, the data of label 2 in disk C；In fact, the data of label 2 can directly be recovered in disk C,
Can successively recover since disk C, or recover to centre from two ends；
(3) if two pieces to damage disk be two pieces of disks in original RAID, wherein there is a disk to include label 0
If the disk A and disk B in original RAID are damaged, then the data of label 4 can be by new in disk B simultaneously
Verification computing recovers, and the data of label 3 can be recovered by original verification computing in disk A；The like：Disk B gets the bid
Numbers 3 data, the data of label 2 in disk A；The data of label 2 in disk B, the data of label 1 in disk A；Disk B gets the bid
Number 1 data, the data (can only be in this order) of label 0 in disk A.
Two pieces of RAID methods of disk failures, the data disks of the new RAID structure of proposition and verification are prevented described in the present embodiment
The composition of disk is N+1, can determine that new check disk must be configured into the matrix group of (N+1+1) × N.All magnetic in due to RAID groups
Disk has three labels consistent with other disks, other two label differences, therefore simultaneously indifference between all disks.Similarly, school
The selection for testing group is also indifference, you can to retain the check information of label 0,1,2,3.
(1) according to piezomagnetic principle, a verification group for label can be necessarily obtained, a information is only lost in the group；
Piezomagnetic principle (drawer principle) described here is meant that, if there are n+1 or n+ (n-1) individual element to be put into n set, its
In must at least one set in have two elements；
(2) recovery of data must be that Diagonal Parity Check and horizontal parity check are used alternatingly；
(3) verification computing (XOR) number of times that data are completed is N (N-1) original operation of horizontal+N (N-1) diagonal computings.
Above-mentioned specific embodiment is only specific case of the invention, and scope of patent protection of the invention is included but is not limited to
Above-mentioned specific embodiment, any person of an ordinary skill in the technical field that meet claims of the present invention and any
The appropriate change or replacement done to it, should all fall into scope of patent protection of the invention.
Claims (6)
- It is 1. a kind of to prevent two pieces of RAID methods of disk failures, it is characterised in that, it is described to prevent two pieces of RAID method bags of disk failures Include：The step of building the verification group of new check disk, and the step of damage disks to two and carry out disk and recover；It is described to build newly The step of verification group of check disk, includes：Data in original RAID are grouped according to label, are given up certain group by other groups Constitute the process of the verification group of new check disk；It is described to damage disks to two and carry out the step of disk recovers and include：For difference Two damage disks, disk voluntarily error correction reparation is carried out respectively；It is described to prevent two pieces of RAID methods of disk failures, by new check disk, use Diagonal Parity Check and horizontal parity school The method of calibration being combined is tested, when carrying out recovery data, Diagonal Parity Check and horizontal parity check is used alternatingly；Work as RAID Two pieces of disks are damaged in group simultaneously when, first with new verification computing recovered part data, then recovered with original verification computing Another part data, and self adaptation checks data, carries out verification error correction, carrys out the data of Recover from damaging disk.
- 2. it is according to claim 1 it is a kind of prevent two pieces of RAID methods of disk failures, it is characterised in that, the RAID method is carried A kind of new RAID structure is gone out, the new RAID structure includes original RAID and new check disk.
- 3. it is according to claim 2 it is a kind of prevent two pieces of RAID methods of disk failures, it is characterised in that, the new RAID The data disks of structure and the composition of check disk are N+1, then the new check disk must be configured into the matrix group of (N+1+1) × N.
- 4. it is according to claim 3 it is a kind of prevent two pieces of RAID methods of disk failures, it is characterised in that, if two pieces damage magnetic Disk is the disk in new check disk and one piece of original RAID, then do not consider new check disk, and the disk in original RAID is by original Verification computing carry out data recovery.
- 5. it is according to claim 3 it is a kind of prevent two pieces of RAID methods of disk failures, it is characterised in that, the new RAID Structure includes the original RAID containing disk A, B, C, D, E and new check disk F, if it is the magnetic in original RAID that two pieces are damaged disk Disk C and disk D, then the data of label 1 are recovered by new verification computing in disk D, and the data of label 0 are by original in disk C Verification computing recover.
- 6. it is according to claim 3 it is a kind of prevent two pieces of RAID methods of disk failures, it is characterised in that, the new RAID Structure includes the original RAID containing disk A, B, C, D, E and new check disk F, if it is the magnetic in original RAID that two pieces are damaged disk Disk A and disk B, then the data of label 4 are recovered by new verification computing in disk B, and the data of label 3 are by original in disk A Verification computing recover.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410399100.6A CN104156276B (en) | 2014-08-14 | 2014-08-14 | It is a kind of to prevent two pieces of RAID methods of disk failures |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410399100.6A CN104156276B (en) | 2014-08-14 | 2014-08-14 | It is a kind of to prevent two pieces of RAID methods of disk failures |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104156276A CN104156276A (en) | 2014-11-19 |
| CN104156276B true CN104156276B (en) | 2017-06-09 |
Family
ID=51881781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410399100.6A Active CN104156276B (en) | 2014-08-14 | 2014-08-14 | It is a kind of to prevent two pieces of RAID methods of disk failures |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104156276B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10379952B2 (en) * | 2017-06-16 | 2019-08-13 | Western Digital Technologies, Inc. | Data recovery and regeneration using parity code |
| CN111078472B (en) * | 2019-12-23 | 2023-06-02 | 四川效率源信息安全技术股份有限公司 | Data recovery method after raid6 is in absence of disk |
| CN111813609B (en) * | 2020-07-23 | 2021-10-15 | 深圳大普微电子科技有限公司 | Data recovery method in storage medium, data recovery system and related equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101023412A (en) * | 2003-11-24 | 2007-08-22 | 网络装置公司 | Semi-static parity distribution technique |
| CN101504623A (en) * | 2009-03-20 | 2009-08-12 | 杭州华三通信技术有限公司 | Independent disk redundancy array construction method and device |
| CN102081559A (en) * | 2011-01-11 | 2011-06-01 | 成都市华为赛门铁克科技有限公司 | Data recovery method and device for redundant array of independent disks |
| CN103970487A (en) * | 2014-05-23 | 2014-08-06 | 浙江宇视科技有限公司 | Redundant array of independent disks and implementation method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101625652B (en) * | 2009-08-04 | 2011-06-08 | 成都市华为赛门铁克科技有限公司 | Multi-disk fault-tolerant system and methods for generating check block and recovering data block |
-
2014
- 2014-08-14 CN CN201410399100.6A patent/CN104156276B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101023412A (en) * | 2003-11-24 | 2007-08-22 | 网络装置公司 | Semi-static parity distribution technique |
| CN101504623A (en) * | 2009-03-20 | 2009-08-12 | 杭州华三通信技术有限公司 | Independent disk redundancy array construction method and device |
| CN102081559A (en) * | 2011-01-11 | 2011-06-01 | 成都市华为赛门铁克科技有限公司 | Data recovery method and device for redundant array of independent disks |
| CN103970487A (en) * | 2014-05-23 | 2014-08-06 | 浙江宇视科技有限公司 | Redundant array of independent disks and implementation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104156276A (en) | 2014-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190171524A1 (en) | Isolating the introduction of software defects in a dispersed storage network | |
| US9417963B2 (en) | Enabling efficient recovery from multiple failures together with one latent error in a storage array | |
| US9529670B2 (en) | Storage element polymorphism to reduce performance degradation during error recovery | |
| CN103970481B (en) | The method and apparatus rebuilding memory array | |
| CN101084486B (en) | Method and system for syndrome generation and data recovery | |
| JPS60163138A (en) | Error correction system | |
| CN103778030B (en) | Daily record subsystem wiring method, error tracking method and processor | |
| CN104866244B (en) | The RAID 6I/O dispatching methods that a kind of balance strip band is write | |
| CN102043685A (en) | RAID (redundant array of independent disk) system and data recovery method thereof | |
| US9513993B2 (en) | Stale data detection in marked channel for scrub | |
| CN108228382B (en) | Data recovery method for single-disk fault of EVENODD code | |
| CN103593260B (en) | The guard method of a kind of metadata and device | |
| US20170168896A1 (en) | Raid-6 for storage system employing a hot spare drive | |
| CN104156276B (en) | It is a kind of to prevent two pieces of RAID methods of disk failures | |
| CN101604267A (en) | The framework of Redundant Array of Independent Disks (RAID) | |
| CN104407821B (en) | A kind of method and device for realizing RAID reconstruction | |
| CN105302492A (en) | RAID reconstruction method and device | |
| CN104516679B (en) | A kind of RAID data processing method and processing device | |
| CN109739436A (en) | RAID reconstruction method, storage medium and device | |
| CN105630417B (en) | A kind of RAID5 systems and in the subsequent method for continuing data of RAID5 thrashings | |
| CN109032513A (en) | Based on the RAID framework of SSD and HDD and its backup, method for reconstructing | |
| CN104461373B (en) | A kind of RAID data processing method and processing device | |
| CN105183583A (en) | Method for data reconstruction of disk array, and disk array system | |
| CN104572374B (en) | Processing method, device and the storage device of storage | |
| CN105183590A (en) | Disk array fault tolerance processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |