CN102955673A - RAID5 (Redundant Array of Inexpensive Disc level 5) intelligent regrouping method and device - Google Patents
RAID5 (Redundant Array of Inexpensive Disc level 5) intelligent regrouping method and device Download PDFInfo
- Publication number
- CN102955673A CN102955673A CN2011102503656A CN201110250365A CN102955673A CN 102955673 A CN102955673 A CN 102955673A CN 2011102503656 A CN2011102503656 A CN 2011102503656A CN 201110250365 A CN201110250365 A CN 201110250365A CN 102955673 A CN102955673 A CN 102955673A
- Authority
- CN
- China
- Prior art keywords
- disk
- key
- value pair
- raid5
- mft
- 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.)
- Granted
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an RAID5 (Redundant Array of Inexpensive Disc level 5) intelligent regrouping method and device. The RAID5 intelligent regrouping method comprises the following steps of: searching all MFT (Master File Table) records with a recording character in a continuous area of a first disc; using a middle address section as a standard analysis target according to an address order inside the disk; extracting a 32-bit symbol-free integer at a 0*2C offset position of the standard analysis target, forming a key value pair with disc numbers, and ordering key values in an ascending order; subtracting the previous key value from the key values in each key value pair, and using a differential value with a highest occurrence frequency as a block size; determining a checking plate, and moving the checking plate to different positions, so as to form different combined schemes; and using an accurate combination scheme of a subarea table link and MFT data cluster flow as a final scheme. The RAID5 intelligent regrouping method provided by the invention carries out area analysis based on an MFT and uses a management mechanism of a file system; the parameters of the RAID5 can be calculated by extracting the data of the special area; and the RAID5 intelligent regrouping method has the advantages of rapid regrouping speed, high accuracy, less generated combination scheme and low regrouping cost.
Description
Technical field
The present invention relates to the raid-array technical field, particularly relate to a kind of 5 grades of raid-arrays (RAID5, Redundant Array of Independent DiskLevel5) intelligent recombination method and device.
Background technology
The RAID5 restructuring is the maximum value point that data are recovered industry, mainly for the enterprises and institutions that have huge storage resources, wherein is no lack of the financial industry that involves the interests of the state and the people, and can infer thus, and the consequence that the RAID5 fault causes is catastrophic often.RAID5 focuses on comprising block size, the isoparametric analysis of dish order.
Existing parameters analysis method mainly contains two kinds: the one, and rely on slip-stick artist individual's experience and judgement to analyze, efficient is extremely low, be affected by human factors very large, error rate is higher, regrouping process often will expend the time that a couple of days and even tens of sky do not wait, and therefore, the RAID5 restructuring is charged and be in higher level in all data recovery business, generally all more than ten thousand yuan even 100,000 yuan, greatly increased the system maintenance cost of relevant enterprises and institutions; Another kind is write software with regard to the algorithm that is based on continuous restructuring+constantly checking, carry out the permutation and combination analysis, but this method is in the situation that number of disks is more, the reorganization scheme quantity that produces is (being geometric index with number of disks increases) too much, treatment effeciency even be lower than the Manual analysis method.
Summary of the invention
Technical matters to be solved by this invention provides a kind of RAID5 intelligence recombination method, can solve the existing not high problem for the treatment of effeciency.
The present invention also provides a kind of RAID5 intelligence reconstruction unit, to guarantee said method application in practice.
In order to address the above problem, the invention discloses a kind of RAID5 intelligence recombination method, comprising: search for the MFT record that all have recording feature in the continuum of first disk; Above-mentioned all MFT records by its address sort in disk, will be arranged in the address section of middle MFT record as the standard analysis object; Extract 32 signless integers as key assignments in 0x2C skew place of the standard analysis object of every disk, form a key-value pair＜key assignments, reel number with current disk number 〉; Above-mentioned key-value pair is arranged according to the key assignments ascending order, generated orderly key-value pair formation; Since the 2nd key-value pair, the key assignments in each key-value pair is deducted key assignments in the previous key-value pair, the highest difference of the frequency of occurrences is as block size; Confirm check disk according to number of disks; Based on above-mentioned orderly key-value pair formation, described check disk is moved to diverse location, form n different assembled scheme; Wherein n is the number of disks in the RAID5 disk array; Whether partition table chain and the MFT data manifold flow of verifying above-mentioned every kind of assembled scheme be correct, and the dish order of correct assembled scheme representative is final plan.
Preferably, described recording feature is that the start byte sequence is that 0x460x490x4C0x45, size are the data block of 1024 bytes.
Preferably, described continuum at the first disk method of searching for MFT record that all have specific recording feature is specially: the MFT record that has described recording feature from the starting position search of first disk; After searching first match point, continue the MFT record that search has described recording feature from the end of current MFT record, until described recording feature disappears.
Preferably, the condition of described recording feature disappearance is: continuous 20480 bytes do not have described recording feature data block.
Preferably, describedly confirm that according to number of disks the method for check disk is specially: if RAID5 system disk quantity is odd number, then front 4 bytes of described standard analysis object are that 0 disk is check disk; If RAID5 system disk quantity is even number, then for every disk, carry out following attribute list ergodic process: according to first attribute offset variable location 10H attribute of described standard analysis object, then calculate the address of next attribute according to attribute size, until the traversal attribute list; If location 10H attribute or traversal attribute list can not be successful in the searching process, then this disk namely is check disk, finishes the attribute list ergodic process.
According to another preferred embodiment of the present invention, a kind of RAID5 intelligence reconstruction unit is also disclosed, comprise recording feature search unit, standard analysis object generation unit, key-value pair generation unit, block size computing unit, assembled scheme generation unit and dish order confirmation unit, wherein: described recording feature search unit is for search for the MFT record that all have recording feature in the continuum of first disk; All MFT records that described standard analysis object generation unit is used for described recording feature search unit is obtained will be arranged in the address section of middle MFT record as the standard analysis object by its address sort in disk; Described key-value pair generation unit is used for extracting 32 signless integers as key assignments in 0x2C skew place of the standard analysis object of every disk, form a key-value pair＜key assignments with current disk number, reel number 〉, and, above-mentioned key-value pair is arranged according to the key assignments ascending order, generated orderly key-value pair formation; Described block size computing unit is used for the 2nd key-value pair of the orderly key-value pair formation that generates from described key-value pair generation unit, and the key assignments in each key-value pair is deducted key assignments in the previous key-value pair, and the highest difference of the frequency of occurrences is as block size; Described assembled scheme generation unit is used for confirming check disk according to number of disks, and the orderly key-value pair formation based on described key-value pair generation unit generates moves to diverse location with described check disk, forms n different assembled scheme; Wherein n is the number of disks in the RAID5 disk array; Whether described dish order confirmation unit is correct for partition table chain and the MFT data manifold flow of verifying every kind of assembled scheme that described assembled scheme generation unit generates, and the dish order of correct square combination case representative is final plan.
Preferably, described recording feature search unit specifically comprises the first match point search subelement and continuous space search subelement, and wherein: described the first match point search subelement is used for having from the starting position search of first disk the MFT record of described recording feature; Described continuous space search subelement is used for continuing the MFT record that search has described recording feature from the end that the MFT that described the first match point search subelement obtains records, until continuous 20480 bytes do not have till the described recording feature data block.
Preferably, described assembled scheme generation unit comprises that specifically the first check disk determines that subelement and the second check disk determine subelement, wherein: described the first check disk determines that subelement is used for when RAID5 system disk quantity is odd number, whether front 4 bytes of judging the standard analysis object of every disk are 0, if this disk is check disk; Described the second check disk determines that subelement is used for when RAID5 system disk quantity is even number, determine in the following manner check disk: for every disk, first attribute offset variable location 10H attribute according to described standard analysis object, then calculate the address of next attribute according to attribute size, until the traversal attribute list; If location 10H attribute or traversal attribute list can not be successful in the searching process, then this disk namely is check disk.
Compared with prior art, the present invention has the following advantages:
In the preferred embodiment of the present invention scheme, based on master control file record table (MFT, Master File Table) carries out regional analysis, utilize the administrative mechanism of file system, the precise information of extraction system subregion (such as the NTFS subregion) specific region is directly calculated the RAID5 parameter, has fundamentally abandoned the method that prior art utilizes the experience conjecture to attempt, restructuring speed is fast, accuracy rate is high, can reduce significantly the cost of RAID business reorganization.
Secondly, the assembled scheme quantity that the preferred embodiment of the present invention produces equals the piece number of disk in the RAID5 system, be far smaller than the assembled scheme quantity that existing that produce with the verification algorithm of continuous restructuring+constantly and disk block number are index variation, can greatly alleviate the workload of follow-up selection final plan.RAID5 with 3 disks gives an example, and existing popular continuous restructuring+constantly verification algorithm will produce 6 kinds of assembled schemes, and the assembled scheme that the preferred embodiment of the present invention produces only has 3 kinds; For the RAID5 system of 4 disks, existing constantly restructuring+constantly verification algorithm will produce 24 kinds of assembled schemes, and the assembled scheme that preferred embodiment of the present invention scheme produces only has 4 kinds.The number of disks of RAID5 system is more, and the present invention program's advantage is more obvious, and the omission of assembled scheme is not subjected to the impact of data volume size.
Description of drawings
Fig. 1 is the process flow diagram of RAID5 intelligence recombination method one embodiment of the present invention;
Fig. 2 is the structured flowchart of RAID5 intelligence reconstruction unit one embodiment of the present invention.
Embodiment
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.
Embodiment of the method:
With reference to Fig. 1, show the flow process of RAID5 intelligence recombination method one embodiment of the present invention, specifically may further comprise the steps:
Step S101: search for the MFT record that all have recording feature in the continuum of first disk;
This preferred embodiment is take start byte as 0x46,0x49, and 0x4C, 0x45, size is fixed as the data block of 1024 bytes as recording feature.The method of searching for MFT record that all have recording feature in the continuum is specially: be 0x46 at first disk search start byte, 0x49,0x4C, 0x45, size is fixed as the MFT record of 1024 bytes (hereinafter referred is " recording feature "), after searching first match point, search behavior does not stop.Because the MFT record is deposited continuously, a slice continuum afterwards all can have recording feature, and search procedure should stop when recording feature disappears.
This preferred embodiment recording feature do not occur as the search end condition take continuous 20480 bytes.For example when all finding recording feature in per 1024 bytes in certain zone, begin suddenly all not find recording feature in continuous 20480 byte area, search procedure namely stops so.
Step S102: above-mentioned all MFT records by its address sort in disk, will be arranged in the address section of middle MFT record as the standard analysis object;
All that step S101 is searched have the MFT record of recording feature by its address sort in disk, and wherein being arranged in middle recording feature is the standard analysis object.As, when having searched 50 recording features, be also referred to as the standard analysis object then take the 25th recording feature as the standard analysis object, and with this address section of other disks.
Step S103: extract 32 signless integers as key assignments in 0x2C skew place of the standard analysis object of every disk, form a key-value pair＜key assignments, reel number with current disk number 〉;
Extracting 32 unsigned int variablees as 0x2C skew place in the recording feature of standard analysis object, it is included in the key of dictionary 1, and disk number is included in the corresponding value; Then jump to the identical address place of second disk, still extract 32 unsigned int variablees at the 0x2C place, still it is included in the key of dictionary 1, and disk number is included in the corresponding value; By that analogy until the 0x2C skew at this place, address of every disk all is extracted 32 unsigned int variablees.
Step S104: above-mentioned key-value pair is arranged according to the key assignments ascending order, generated orderly key-value pair formation;
Specific to this preferred embodiment, the key-value pair in the dictionary 1 can be arranged according to the key ascending order.
Step S105: since the 2nd key-value pair, the key assignments in each key-value pair is deducted key assignments in the previous key-value pair, the highest difference of the frequency of occurrences is as block size;
Deduct its poor being kept in the tabulation 1 of value of its previous key element with each the key element in the dictionary 1, the highest value of the frequency of occurrences is exactly block size in the tabulation 1.For example occurred 64,256,128,192 in the dictionary 1 key element, 512 5 values, what the poor frequency of occurrences of adjacent element was the highest after ascending order was arranged is 64, is generally so the size of 1024 byte 1K according to a MFT record, so, the block size of this RAID5 is 64K, namely 128 sectors.
Step S 106: confirm check disk according to number of disks;
From this step, the below's beginning calculating dial order: among the step S104, key-value pair is arranged according to the key assignments ascending order, the order of its intermediate value element is exactly dish order to be verified after the ordering.The dish order that is to say this moment has approached correct, just wherein still comprises a camouflage value of the verification generation of RAID5, and subsequent step mainly is which is judged is the camouflage value.
If number of disks is an odd number among the RAID5, we only need to judge whether front 4 bytes of the standard analysis object of which disk are 0, if be 0 to be a check block so, when front disk is regarded as check disk;
If number of disks is an even number among the RAID5, need to be according to first attribute offset variable location 10H attribute in the standard analysis object (being still a MFT record), then calculate the address of next attribute according to attribute size, until traversal attribute list, if location 10H attribute or traversal attribute list can not be successful in the searching process, search so attribute work knot and adopt and can prove that it is a check block (because not being correct MFT record), the disk of its representative is called check disk.
Step S107: based on above-mentioned orderly key-value pair formation, described check disk is moved to diverse location, form n different assembled scheme; Wherein n is the number of disks in the RAID5 disk array;
After finding out check disk, which position this comes on owing to not knowing check disk, can only go successively examination.All check disk is inserted in the diverse location of dictionary 1 value element at every turn, form different assembled schemes, as: the RAID5 of three dishes can have following several scheme (P represents check disk):
P-1-2
1-P-2
1-2-P
Step S108: verify whether the partition table chain of above-mentioned every kind of assembled scheme and MFT data manifold flow be correct, and the dish order of correct assembled scheme representative is final plan.
Whether the parameter such as partition table chain, MFT data manifold flow item of verifying every kind of scheme is correct, and correct scheme is final plan
For aforesaid each embodiment of the method, simple in order to describe, therefore it all is expressed as a series of combination of actions, but those skilled in the art should know, the present invention is not subjected to the restriction of described sequence of movement, because according to the present invention, some step can adopt other orders or carry out simultaneously secondly, those skilled in the art also should know, said method embodiment all belongs to preferred embodiment, and related action and module might not be that the present invention is necessary.
Device embodiment:
With reference to Fig. 2, show the structured flowchart of RAID5 intelligence reconstruction unit one embodiment of the present invention, comprise recording feature search unit 21, standard analysis object generation unit 22, key-value pair generation unit 23, block size computing unit 24, assembled scheme generation unit 25 and dish order confirmation unit 26, wherein:
Recording feature search unit 21: for search for the MFT record that all have recording feature in the continuum of first disk;
Wherein, recording feature search unit 21 specifically comprises the first match point search subelement 211 and continuous space search subelement 212, and the first match point search subelement 211 wherein is used for having from the starting position search of first disk the MFT record of recording feature; Continuous space search subelement 212 is used for continuing the MFT record that search has recording feature from the end that the MFT that the first match point search subelement 211 obtains records, until continuous 20480 bytes do not have till the described recording feature data block.
Standard analysis object generation unit 22: all MFT records that are used for recording feature search unit 21 is obtained will be arranged in the address section of middle MFT record as the standard analysis object by its address sort in disk;
Key-value pair generation unit 23: be used for extracting 32 signless integers as key assignments in 0x2C skew place of the standard analysis object of every disk, form a key-value pair＜key assignments, reel number with current disk number 〉; And, above-mentioned key-value pair is arranged according to the key assignments ascending order, generate orderly key-value pair formation;
Block size computing unit 24: be used for the 2nd key-value pair of the orderly key-value pair formation that generates from key-value pair generation unit 23, the key assignments in each key-value pair is deducted the highest difference of the key assignments frequency of occurrences in the previous key-value pair as block size;
Assembled scheme generation unit 25: be used for confirming check disk according to number of disks; And the orderly key-value pair formation based on key-value pair generation unit 23 generates moves to diverse location with described check disk, forms n different assembled scheme; Wherein n is the number of disks in the RAID5 disk array;
Wherein, assembled scheme generation unit 25 comprises that specifically the first check disk determines that subelement 251 and the second check disk determine subelement 252, the first check disk determines that subelement 251 is used for when RAID5 system disk quantity is odd number, whether front 4 bytes of judging the standard analysis object of every disk are 0, if this disk is check disk; The second check disk determines that subelement 252 is used for when RAID5 system disk quantity is even number, determine in the following manner check disk: for every disk, first attribute offset variable location 10H attribute according to the standard analysis object, then calculate the address of next attribute according to attribute size, until the traversal attribute list; If location 10H attribute or traversal attribute list can not be successful in the searching process, then this disk namely is check disk.
Dish order confirmation unit 26: whether correct for partition table chain and the MFT data manifold flow of verifying every kind of assembled scheme that assembled scheme generation unit 25 generates, the dish order of correct assembled scheme representative is final plan.
Need to prove, said apparatus embodiment belongs to preferred embodiment, and related unit and module might not be that the present invention is necessary.
Each embodiment in this instructions all adopts the mode of going forward one by one to describe, and what each embodiment stressed is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For device embodiment of the present invention because itself and embodiment of the method basic simlarity, so describe fairly simple, relevant part gets final product referring to the part explanation of embodiment of the method.
Above a kind of RAID5 intelligence recombination method provided by the present invention and device are described in detail, have used specific case herein the explanation that principle of the present invention and embodiment have carried out setting forth above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (8)
1. a RAID5 intelligence recombination method is characterized in that, comprising:
Search for the MFT record that all have recording feature in the continuum of first disk;
Above-mentioned all MFT records by its address sort in disk, will be arranged in the address section of middle MFT record as the standard analysis object;
Extract 32 signless integers as key assignments in 0x2C skew place of the standard analysis object of every disk, form a key-value pair＜key assignments, reel number with current disk number 〉;
Above-mentioned key-value pair is arranged according to the key assignments ascending order, generated orderly key-value pair formation;
Since the 2nd key-value pair, the key assignments in each key-value pair is deducted key assignments in the previous key-value pair, the highest difference of the frequency of occurrences is as block size;
Confirm check disk according to number of disks;
Based on above-mentioned orderly key-value pair formation, described check disk is moved to diverse location, form n different assembled scheme; Wherein n is the number of disks in the RAID5 disk array;
Whether partition table chain and the MFT data manifold flow of verifying above-mentioned every kind of assembled scheme be correct, and the dish order of correct assembled scheme representative is final plan.
2. the method for claim 1 is characterized in that, described recording feature is that the start byte sequence is that 0x460x490x4C0x45, size are the data block of 1024 bytes.
3. the method for claim 1 is characterized in that, the method that all MFT records with specific recording feature are searched in described continuum at first disk is specially:
The MFT record that has described recording feature from the starting position search of first disk;
After searching first match point, continue the MFT record that search has described recording feature from the end of current MFT record, until described recording feature disappears.
4. method as claimed in claim 3 is characterized in that, the condition that described recording feature disappears is: continuous 20480 bytes do not have described recording feature data block.
5. the method for claim 1 is characterized in that, described method according to number of disks affirmation check disk is specially:
If RAID5 system disk quantity is odd number, then front 4 bytes of described standard analysis object are that 0 disk is check disk;
If RAID5 system disk quantity is even number, then for every disk, carry out following attribute list ergodic process: according to first attribute offset variable location 10H attribute of described standard analysis object, then calculate the address of next attribute according to attribute size, until the traversal attribute list; If location 10H attribute or traversal attribute list can not be successful in the searching process, then this disk namely is check disk, finishes the attribute list ergodic process.
6. a RAID5 intelligence reconstruction unit is characterized in that, comprises recording feature search unit, standard analysis object generation unit, key-value pair generation unit, block size computing unit, assembled scheme generation unit and dish order confirmation unit, wherein:
Described recording feature search unit is for search for the MFT record that all have recording feature in the continuum of first disk;
All MFT records that described standard analysis object generation unit is used for described recording feature search unit is obtained will be arranged in the address section of middle MFT record as the standard analysis object by its address sort in disk;
Described key-value pair generation unit is used for extracting 32 signless integers as key assignments in 0x2C skew place of the standard analysis object of every disk, forms a key-value pair＜key assignments, reel number with current disk number 〉; And, above-mentioned key-value pair is arranged according to the key assignments ascending order, generate orderly key-value pair formation;
Described block size computing unit is used for the 2nd key-value pair of the orderly key-value pair formation that generates from described key-value pair generation unit, and the key assignments in each key-value pair is deducted key assignments in the previous key-value pair, and the highest difference of the frequency of occurrences is as block size;
Described assembled scheme generation unit be used for according to number of disks confirm check disk and, the orderly key-value pair formation based on described key-value pair generation unit generates moves to diverse location with described check disk, forms n different assembled scheme; Wherein n is the number of disks in the RAID5 disk array;
Whether described dish order confirmation unit is correct for partition table chain and the MFT data manifold flow of verifying every kind of assembled scheme that described assembled scheme generation unit generates, and the dish order of correct assembled scheme representative is final plan.
7. device as claimed in claim 6 is characterized in that, described recording feature search unit specifically comprises the first match point search subelement and continuous space search subelement, wherein:
Described the first match point search subelement is used for having from the starting position search of first disk the MFT record of described recording feature;
Described continuous space search subelement is used for continuing the MFT record that search has described recording feature from the end that the MFT that described the first match point search subelement obtains records, until continuous 20480 bytes do not have till the described recording feature data block.
8. device as claimed in claim 6 is characterized in that, described assembled scheme generation unit comprises that specifically the first check disk determines that subelement and the second check disk determine subelement, wherein:
Described the first check disk determines that subelement is used for when RAID5 system disk quantity is odd number, judges whether front 4 bytes of the standard analysis object of every disk are 0, if this disk is check disk;
Described the second check disk determines that subelement is used for when RAID5 system disk quantity is even number, determine in the following manner check disk: for every disk, first attribute offset variable location 10H attribute according to described standard analysis object, then calculate the address of next attribute according to attribute size, until the traversal attribute list; If location 10H attribute or traversal attribute list can not be successful in the searching process, then this disk namely is check disk.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110250365.6A CN102955673B (en) | 2011-08-29 | 2011-08-29 | RAID5 intelligence recombination method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110250365.6A CN102955673B (en) | 2011-08-29 | 2011-08-29 | RAID5 intelligence recombination method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102955673A true CN102955673A (en) | 2013-03-06 |
| CN102955673B CN102955673B (en) | 2015-08-12 |
Family
ID=47764519
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110250365.6A Active CN102955673B (en) | 2011-08-29 | 2011-08-29 | RAID5 intelligence recombination method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102955673B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104794017A (en) * | 2015-04-27 | 2015-07-22 | 四川效率源信息安全技术有限责任公司 | Data recovery method for RAID5 with odd number of disks |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040078519A1 (en) * | 2002-10-17 | 2004-04-22 | Yuhyeon Bak | Data reorganization method in a RAID striping system |
| CN1670682A (en) * | 2004-03-15 | 2005-09-21 | 华为技术有限公司 | A data reintegration method |
| CN101571820A (en) * | 2009-05-27 | 2009-11-04 | 成都市华为赛门铁克科技有限公司 | Redundancy protected method, device and system |
-
2011
- 2011-08-29 CN CN201110250365.6A patent/CN102955673B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040078519A1 (en) * | 2002-10-17 | 2004-04-22 | Yuhyeon Bak | Data reorganization method in a RAID striping system |
| CN1670682A (en) * | 2004-03-15 | 2005-09-21 | 华为技术有限公司 | A data reintegration method |
| CN101571820A (en) * | 2009-05-27 | 2009-11-04 | 成都市华为赛门铁克科技有限公司 | Redundancy protected method, device and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104794017A (en) * | 2015-04-27 | 2015-07-22 | 四川效率源信息安全技术有限责任公司 | Data recovery method for RAID5 with odd number of disks |
| CN104794017B (en) * | 2015-04-27 | 2018-05-04 | 四川效率源信息安全技术股份有限公司 | A kind of data reconstruction method for odd number disk RAID5 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102955673B (en) | 2015-08-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI702844B (en) | Method, device, apparatus, and storage medium of generating features of user | |
| CN104731896B (en) | A kind of data processing method and system | |
| CN110413227B (en) | Method and system for predicting remaining service life of hard disk device on line | |
| CN105631003B (en) | Support intelligent index construct, inquiry and the maintaining method of mass data classified statistic | |
| CN109710572B (en) | HBase-based file fragmentation method | |
| CN104142984A (en) | Video fingerprint retrieval method based on coarse and fine granularity | |
| CN106897322A (en) | The access method and device of a kind of database and file system | |
| CN102831222A (en) | Differential compression method based on data de-duplication | |
| US20130311642A1 (en) | Automated discovery of template patterns based on received server requests | |
| CN104636401B (en) | A kind of method and device of SCADA system data rewind | |
| CN105426410A (en) | Data acquisition system and analytic method for same | |
| CN104484131B (en) | The data processing equipment of multiple disks server and corresponding processing method | |
| CN110471820A (en) | A kind of cloud storage system disk failure prediction technique based on Recognition with Recurrent Neural Network | |
| CN106446091A (en) | Preprocessing method and device for multi-source time series data | |
| CN105095247A (en) | Symbolic data analysis method and system | |
| WO2022120995A1 (en) | Device computing power evaluation method and system based on pow consensus mechanism | |
| CN104991741B (en) | A kind of situation adaptation power network big data storage method based on key-value model | |
| CN106897342A (en) | A kind of data verification method and equipment | |
| CN104778123A (en) | Method and device for detecting system performance | |
| CN104199754A (en) | Production failure analysis system | |
| US8134935B2 (en) | Transaction topology discovery using constraints | |
| CN106354587A (en) | Mirror image server and method for exporting mirror image files of virtual machine | |
| CN107818106B (en) | Big data offline calculation data quality verification method and device | |
| CN105447519A (en) | Model detection method based on feature selection | |
| CN113342418B (en) | Distributed machine learning task unloading method based on block chain |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |