CN102567147A - Hard disk array controller based on double hard disk redundancy - Google Patents
Hard disk array controller based on double hard disk redundancy Download PDFInfo
- Publication number
- CN102567147A CN102567147A CN2011104609867A CN201110460986A CN102567147A CN 102567147 A CN102567147 A CN 102567147A CN 2011104609867 A CN2011104609867 A CN 2011104609867A CN 201110460986 A CN201110460986 A CN 201110460986A CN 102567147 A CN102567147 A CN 102567147A
- Authority
- CN
- China
- Prior art keywords
- circleplus
- data
- hard disk
- circletimes
- disk array
- 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 discloses a hard disk array controller based on double hard disk redundancy, which comprises a first data buffer area, a second data buffer area, a third data buffer area, a fourth data buffer area, a data playback verification module, a redundant data generation module, a damaged hard disk reconstruction module, a data gate, a hard disk control interface and a hard disk array. The storage and reading rate of the hard disk array controller can be still kept unchanged when an optional hard disk or two optional hard disks in the hard disk array are in failure, the hard disk array controller can still realize the functions of recording while playing back and playing back while reconstructing when some hard disks in the array are in failure, and safety of the hard disk array is greatly enhanced.
Description
Technical field
The present invention relates to a kind of based on the redundant hard disk array controller of two hard disks.
Background technology
In high speed hard-disk array data field of storage; Along with improving constantly of monolithic disk read-write speed and capacity; The memory rate of hard disk array and memory bandwidth are also in continuous lifting, and two-forty, big bandwidth, high capacity have become hard disk array field of data storage main flow development trend.But thing followed hard disk array data security problem also highlights, and in many instances, people are to the concern of data safety problem even surpassed the concern to the data memory rate.Particularly in space industry; Data storage and data security problem are particularly important; Human and material resources and financial resources that every execution one subtask is spent are all quite huge, so the data message that obtains in the task process is also very precious, if can not get off the appropriate preservation of these information securities; Loss will be difficult to estimate, even can cause the failure of whole task.In hard disk array, add the redundant data dish; When having hard disk to lose efficacy in the array; Can the inefficacy hard disc data be recovered out to become the main means that improve the hard disk array security according to redundant information; But in a single day occur that the hard disk corruptions situation is arranged in the array in the existing commercial hard disk array storage solution, DISK to Image array readwrite performance will receive very big influence, even can't reach the minimum memory property requirement of executing the task.
Summary of the invention
Technology of the present invention is dealt with problems and is: to the deficiency of prior art, provide a kind of based on the redundant hard disk array controller of two hard disks.Still can keep storage and reading rate constant under the situation that adopts the present invention can be in hard disk array any one or two hard disks to lose efficacy; And when having hard disk to lose efficacy in the array, still can realize the function that limit writing edge playback, playback limit, limit rebuild, hard disk array security simultaneously is greatly improved.
Technical solution of the present invention is:
A kind of based on the redundant hard disk array controller of two hard disks; Comprise: data buffer one, data buffer two, data buffer three, data buffer four, data readback verification module, redundant data generation module, damage hard disk rebuilding module, data strobe, hard disk control interface and hard disk array; Wherein, When record data, after cushioning, a pair of input front end data in data buffer output to the redundant data generation module; The redundant data generation module generates check information according to the front end data of input, and outputs to data buffer three through data strobe; Three pairs of data buffers are stored front end data and check information in the hard disk array into through the hard disk control interface; When playback of data, after cushioning, the record data that hard disk array is exported in data buffer two output to data readback verification module; Data readback verification module does not have in hard disk array when damaging hard disk, through data buffer four record data is directly exported.
Said redundant data generation module; By band front end data is carried out checking treatment and generate and the corresponding checking data of each band, and store into respectively in the data disks and check disk in the hard disk array by the order of packet in the band checking data with front end data and generation;
Said data readback verification module; When there is the degrading data dish in hard disk array; Read by band in packet and the check disk of data disks stored in proper working order in the hard disk array and band corresponding check data; Through recovering to handle the packet that obtains degrading data dish stored in the band, by the record order of redundant data generation module the data recorded bag is exported then; Said recovery processing and checking treatment are that contrary is handled;
Said damage hard disk rebuilding module; When there is the degrading data dish in hard disk array; Read by band in packet and the check disk of data disks stored of operate as normal in the hard disk array and band corresponding check data; Handle the packet that obtains degradation dish stored in the band through recovery, and the packet that recovers is stored in the initiate data disks of hard disk array.
Said redundant data generation module carries out checking treatment through following formula to front end data and generates checking data:
Wherein, P
kAnd Q
kBe checking data according to k band; N is the hard disk sum in the hard disk array.
Said data readback verification module or damage hard disk rebuilding module recover to handle the data of recovering in the degrading data dish through following formula to front end data:
Wherein k is the band numbering, and n is the hard disk numbering, and i, j are degradation hard disk numbering.
The present invention compared with prior art has following advantage:
(1) the present invention is based on the redundant hard disk array high-speed real-time calibration technology of two hard disks; Situation to the use of hard disk array degradation; Data are rebuild function and data in real time playback function afterwards to be separated respectively by damaging hard disk rebuilding module and the completion of data readback verification module; Damaging the hard disk rebuilding module mainly carries out when having degradation (inefficacy) data disks in the hard disk array; Data recorded is rebuild; Can be in non task in system utilizes the packet and the corresponding check data of storing in other data disks that the data of degraded fail data dish are recovered out during the period; Data readback verification module is mainly carried out the data readback function when data disks lost efficacy; Be in when having one or two blocks of data dishes to lose efficacy in task period and the array through recovering to handle in system the data in real time of fail data dish is recovered and front end is given in playback, therefore guaranteed when having one or two blocks of data dishes to lose efficacy, still can carried out the storage of data and read and reduction of speed not according to original speed by record data;
(2) data readback verification module is collected strip data among the present invention when data readback; When no hard disk lost efficacy; Export effective data packets to front end and shield the checking data bag simultaneously; When having data disks to lose efficacy, also can export effective data packets and shield the checking data bag simultaneously according to checking algorithm, realize the complete playback of band valid data.
(3) but the present invention makes full use of the characteristics of FPGA device concurrent operation; Propose to be fit to the hard disk array high-speed real-time calibration technology that FPGA realizes targetedly based on two hard disks redundancies; Adopt the framework of multimode parallel processing; This framework has guaranteed the real-time of data processing; Data in the time of can supporting that based on the redundant checking algorithm of two hard disks maximum two hard disks lost efficacy simultaneously in the array are recovered, relative monolithic hard disk redundant array, and the checking algorithm of two hard disks redundancies is from improved the reliability of system to a great extent.
Description of drawings
Fig. 1 is a synoptic diagram of the present invention;
Fig. 2 is a hard disk array strip data storage organization synoptic diagram of the present invention;
Fig. 3 is a P check disk redundant data generating algorithm structural representation of the present invention;
Fig. 4 is a Q check disk redundant data generating algorithm structural representation of the present invention;
Fig. 5 takes advantage of the modular algorithm structural representation for mould of the present invention;
Fig. 6 takes advantage of the employed table of logarithm of module for mould of the present invention;
Fig. 7 takes advantage of the employed antilogarithm table of module for mould of the present invention.
Embodiment
Just combine accompanying drawing that the present invention is done further introduction below.
Be illustrated in figure 1 as and the present invention includes: data buffer one, data buffer two, data buffer three, data buffer four, data readback verification module, redundant data generation module, damage hard disk rebuilding module, data strobe, hard disk control interface and hard disk array
The front end data source, data buffer one, the redundant data generation module, data strobe, data buffer three, the hard disk control interface, hard disk array is according to data flow direction composition data record branch road.Wherein, the front end data source is for needing the data inlet of storage, exchange data using packet switch framework receiving front-end packet.
Hard disk array, the hard disk control interface, data buffer two, data readback verification module, data buffer four plays back to front-end interface, by data flow composition data playback branch road.
Hard disk array, the hard disk control interface, the hard disk rebuilding module is damaged in data buffer two, data strobe, data buffer three, the hard disk control interface, hard disk array is rebuild branch road by the data flow composition data.
Each can cushion the data capacity of a stripe size four data buffer zones, and data packet number is a hard disk number in the hard disk array in the band, and redundancy check data generates that recovery all is that unit is operated with the band with data in real time.
The redundant data generation module adopts based on the redundant data generating algorithm of two hard disks, raw data is sent in order to the back-end, and generate redundancy check information simultaneously and be inserted between raw data, and this module operating rate is not influenced by hard disk array degradation (inefficacys).Data check algorithm use mould based on two hard disks are redundant is taken advantage of the checking treatment algorithm that combines with XOR, but fast parallel arithmetic goes out P check disk and Q check disk redundant data information.
Generate the checking treatment algorithmic formula of checking data according to following the present invention:
Wherein
representes XOR;
symbolic representation modular multiplication; Modular multiplication priority is higher than XOR priority; 0~11 represents 12 data disks in the hard disk array.
The first capable STRIP0 was an example during then band was arranged among Fig. 2, and the checking data of generation is distinguished as follows:
Data readback verification module adopts based on the redundant data check algorithm of two hard disks; When one or two hard disk corruptions are arranged in the hard disk array; Can come out according to the packet recovery that valid data are real-time in the residue health data dish; And original valid data are played back to front-end interface, this module operating rate does not receive the hard disk array degradation impacts.The redundant recovery Processing Algorithm of two hard disks is the algorithm for inversion of the redundant checking treatment algorithm of two hard disks, but fast parallel arithmetic goes out the data of corrupt data dish, and is as follows.
Wherein k is the band numbering, and n is the hard disk numbering, and i, j are degradation hard disk numbering.
Be example with the first row STRIP0 during band is arranged among Fig. 2 equally, data disks 8 and number are arranged in array
When damaging simultaneously, recover D according to 11 liang of blocks of data dishes of dish
(0,8)And D
(0,11)Method following:
P wherein
(0)The checking data of storing on the expression P check disk, Q
(0)The checking data of storing on the expression Q check disk, because two blocks of data hard disks damage the situation that comprises monolithic data hard disk corruptions simultaneously, so the restoration methods of monolithic data disks is contained in the following formula, getting wherein any expression formula can be restored.
Damaging the hard disk rebuilding module adopts based on the redundant data check algorithm of two hard disks; In one or 2 blocks of data adjustment debit bad times, arranged in hard disk array; Can corrupt data dish data in real time be recovered out according to the data in the residue health data dish; And data write in the new data dish of replacing after will recovering, and this module operating rate does not receive the hard disk array degradation impacts.It is identical with the recovery Processing Algorithm formula of data readback verification module to recover the Processing Algorithm formula.
The effect of data strobe module is that redundant data generation module and the data of damaging the output of hard disk rebuilding module are made gating; When array is in logging mode; The data of selecting the output of redundant data generation module are to rear module; When array was in the hard disk reconstruct mode, the data of selecting to damage the output of hard disk rebuilding module were to rear module.
Hard disk control interface module is accomplished the read-write operation function to hard disk array monolithic hard disk.Adopt the parallel processing framework, can accomplish reading while write of polylith hard disk simultaneously.
Fig. 2 is a hard disk array strip data storage organization synoptic diagram of the present invention; The hard disk array that this synoptic diagram is formed with 14 hard disks is an example; Operate simultaneously to improve data storage speed for realizing many hard disks; Data write in the hard disk array with the band mode, and the packet that writes each blocks of data dish of hard disk array and check disk at synchronization is formed a data band, comprises 12 effective data packets, 1 P check disk packet and 1 Q check disk packet in the same data band.Also be that unit recovers out with data when carrying out data restore operation with the band.
According to above-mentioned checking treatment algorithmic formula, it is as shown in Figure 3 that the present invention generates the checking treatment structure of redundant data in the P check disk.P check disk redundant data is the result of the mutual XOR of each effective data packets in the band; Its generation module adopts circulation XOR mode; Be made up of an XOR gate and two twoport ram, first effective data packets is stored to earlier among the twoport ram A, when second packet gets into, reads the first bag data of having stored among the twoport ram A simultaneously; Through being stored among the twoport ram B behind the XOR gate XOR; Read data among the twoport ram B simultaneously when the three guarantees data get into, cross XOR gate and carry out being stored among the twoport ram A behind the XOR, the result who obtains until last packet that so circulates is P check disk redundant data.
In like manner; According to recovering Processing Algorithm; Fig. 4 generates the checking treatment structural representation of Q check disk redundant data for the present invention; Similar with P check disk generation method, Q check disk data generation module also adopts circulation XOR framework, and different is the step that before XOR gate, will carry out modular multiplication in data earlier.
Fig. 5 takes advantage of the modular algorithm structural representation for mould of the present invention, and common multiplication can produce data carry problem after data multiply each other, and the characteristics that mould is taken advantage of can not produce data carry problem after to be two numbers multiply each other.It is to patrol in China (Galois) territory at character used in proper names and in rendering some foreign names to do multiplication that mould is taken advantage of, and does mould between the 8bit bit wide data and takes advantage of, and is equivalent at GF (2
8) do multiplication in the territory, there are following characteristics in this territory:
It is a number field, comprises 256 elements, promptly 0,1,2 ..., 255}, if with 16 systems represent wherein element promptly 00,01,02 ..., FF}.
It is a closed domain, and the result that the territory interior element performs the four operations must drop in the territory, is similar to the periodicity of trigonometric function.
All computings all are man-to-man mappings, i.e. the inverse operation of all computings obtains unique result.
As field element being converted into 2 systems, its each position can be regarded polynomial expression a as
7X
7+ a
6X
6+ a
5X
5+ a
4X
4+ a
3X
3+ a
2X
2+ a
1X+a
0Coefficient, and polynomial expression X
8+ X
4+ X
3+ X
2+ 1 is called the primitive polynomial in this territory, and the root of primitive polynomial is called primitive element, and each of primitive element time power can generate all elements in the territory.
Therefore the method that two digital-to-analogues are taken advantage of in this territory is equal to two data are taken the logarithm respectively and sues for peace, and summed result is the process of negate logarithm again.It asks the method for logarithm and inverse logarithm to adopt lookup table mode to realize that Fig. 6 is 8bit data table of logarithm, and Fig. 7 is 8bit data antilogarithm table.Two tables are stored in two rom storeies, will import data, obtain the result and be checking result as rom memory read address.
The unspecified part of the present invention belongs to general knowledge as well known to those skilled in the art.
Claims (2)
1. one kind based on the redundant hard disk array controller of two hard disks; Comprise: data buffer one, data buffer two, data buffer three, data buffer four, data readback verification module, redundant data generation module, damage hard disk rebuilding module, data strobe, hard disk control interface and hard disk array; Wherein, When record data, after cushioning, a pair of input front end data in data buffer output to the redundant data generation module; The redundant data generation module generates check information according to the front end data of input, and outputs to data buffer three through data strobe; Three pairs of data buffers are stored front end data and check information in the hard disk array into through the hard disk control interface; When playback of data, after cushioning, the record data that hard disk array is exported in data buffer two output to data readback verification module; Data readback verification module does not have in hard disk array when damaging hard disk, through data buffer four record data is directly exported; It is characterized in that:
Said redundant data generation module; By band front end data is carried out checking treatment and generate and the corresponding checking data of each band, and store into respectively in the data disks and check disk in the hard disk array by the order of packet in the band checking data with front end data and generation;
Said data readback verification module; When there is the degrading data dish in hard disk array; Read by band in packet and the check disk of data disks stored in proper working order in the hard disk array and band corresponding check data; Through recovering to handle the packet that obtains degrading data dish stored in the band, by the record order of redundant data generation module the data recorded bag is exported then; Said recovery processing and checking treatment are that contrary is handled;
Said damage hard disk rebuilding module; When there is the degrading data dish in hard disk array; Read by band in packet and the check disk of data disks stored of operate as normal in the hard disk array and band corresponding check data; Handle the packet that obtains degradation dish stored in the band through recovery, and the packet that recovers is stored in the initiate data disks of hard disk array.
2. as claimed in claim 1 a kind of based on the redundant hard disk array controller of two hard disks, it is characterized in that: said redundant data generation module carries out checking treatment through following formula to front end data and generates checking data:
Wherein, P
kAnd Q
kBe checking data according to k band; N is the hard disk sum in the hard disk array
Said data readback verification module or damage hard disk rebuilding module recover to handle the data of recovering in the degrading data dish through following formula to front end data:
Wherein k is the band numbering, and n is the hard disk numbering, and i, j are degradation hard disk numbering.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110460986.7A CN102567147B (en) | 2011-12-29 | 2011-12-29 | Hard disk array controller based on double hard disk redundancy |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110460986.7A CN102567147B (en) | 2011-12-29 | 2011-12-29 | Hard disk array controller based on double hard disk redundancy |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102567147A true CN102567147A (en) | 2012-07-11 |
CN102567147B CN102567147B (en) | 2015-05-27 |
Family
ID=46412625
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110460986.7A Expired - Fee Related CN102567147B (en) | 2011-12-29 | 2011-12-29 | Hard disk array controller based on double hard disk redundancy |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102567147B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103970690A (en) * | 2014-05-19 | 2014-08-06 | 浪潮电子信息产业股份有限公司 | High-performance high-fault-tolerance storage design method and device based on channel binding |
CN110688263A (en) * | 2019-09-30 | 2020-01-14 | 中国工程物理研究院计算机应用研究所 | FPGA-based hard disk automatic switching device and application method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1467636A (en) * | 2003-05-10 | 2004-01-14 | 北京亚细亚智业科技有限公司 | Redundant data protecting method for portable storage media |
CN101387975A (en) * | 2008-10-20 | 2009-03-18 | 中科院成都信息技术有限公司 | Magnetic disc array system |
-
2011
- 2011-12-29 CN CN201110460986.7A patent/CN102567147B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1467636A (en) * | 2003-05-10 | 2004-01-14 | 北京亚细亚智业科技有限公司 | Redundant data protecting method for portable storage media |
CN101387975A (en) * | 2008-10-20 | 2009-03-18 | 中科院成都信息技术有限公司 | Magnetic disc array system |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103970690A (en) * | 2014-05-19 | 2014-08-06 | 浪潮电子信息产业股份有限公司 | High-performance high-fault-tolerance storage design method and device based on channel binding |
CN110688263A (en) * | 2019-09-30 | 2020-01-14 | 中国工程物理研究院计算机应用研究所 | FPGA-based hard disk automatic switching device and application method |
Also Published As
Publication number | Publication date |
---|---|
CN102567147B (en) | 2015-05-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104035830B (en) | A kind of data reconstruction method and device | |
CN108733503B (en) | Storage node, distributed storage system and method for storing data | |
US9798620B2 (en) | Systems and methods for non-blocking solid-state memory | |
US11017112B2 (en) | Distributed network for storing a redundant array of independent blockchain blocks | |
CN103718162B (en) | Method and apparatus for flexible RAID in SSD | |
KR101298827B1 (en) | Improved error correction in a solid state disk | |
CN101093434B (en) | Method of improving input and output performance of raid system using matrix stripe cache | |
CN101084486B (en) | Method and system for syndrome generation and data recovery | |
WO2005066761A3 (en) | Method, system, and program for managing parity raid data reconstruction | |
US20070294565A1 (en) | Simplified parity disk generation in a redundant array of inexpensive disks | |
CN103034458B (en) | Method and the device of Redundant Array of Independent Disks (RAID) is realized in solid state hard disc | |
CN104461781A (en) | Data block reconstruction method based on erasure codes | |
CN102207895B (en) | Data reconstruction method and device of redundant array of independent disk (RAID) | |
CN104809032A (en) | Method and system for service-aware parity placement in a storage system | |
WO2012051931A1 (en) | Method, device and system for reconstructing redundant array of inexpensive disks | |
CN103488432B (en) | Hybrid disk array, deferred write verification method for hybrid disk array, and data recovery method for hybrid disk array | |
TW201040953A (en) | Rewriting codeword objects to magnetic data tape upon detection of an error | |
JPH02501869A (en) | Stored data error correction method | |
CN101251812A (en) | Method for cluster system data fault tolerance | |
CN102567147B (en) | Hard disk array controller based on double hard disk redundancy | |
CN105630417B (en) | A kind of RAID5 systems and in the subsequent method for continuing data of RAID5 thrashings | |
JP4988418B2 (en) | Method, program and data storage device for reading / writing data from / to multiple layers of rotating storage medium | |
CN104572374B (en) | Processing method, device and the storage device of storage | |
TW201124836A (en) | Raid system | |
US8176390B2 (en) | Single XOR operation WEAVER reconstruction of a failed drive of a RAID |
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 | ||
C41 | Transfer of patent application or patent right or utility model | ||
TR01 | Transfer of patent right |
Effective date of registration: 20161101 Address after: 100176, C38 building, building 10, No. 1, Spark Road, Fengtai District Science City, Beijing Patentee after: Beijing Long March Yutong measurement and Control Communication Technology Co Ltd Address before: 100076 Beijing, box 9200, box 74 Patentee before: Beijing Institute of Telemetry Technology |
|
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150527 Termination date: 20201229 |
|
CF01 | Termination of patent right due to non-payment of annual fee |