CN103049222A - RAID5 (redundant array of independent disk 5) write IO optimization processing method - Google Patents
RAID5 (redundant array of independent disk 5) write IO optimization processing method Download PDFInfo
- Publication number
- CN103049222A CN103049222A CN2012105823792A CN201210582379A CN103049222A CN 103049222 A CN103049222 A CN 103049222A CN 2012105823792 A CN2012105823792 A CN 2012105823792A CN 201210582379 A CN201210582379 A CN 201210582379A CN 103049222 A CN103049222 A CN 103049222A
- Authority
- CN
- China
- Prior art keywords
- address
- write
- data
- mapping table
- raid5
- 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
- Techniques For Improving Reliability Of Storages (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Disclosed is an RAID5 (redundant array of independent disk 5) write IO optimization processing method. The method includes following steps: 1), storage system creation; 2), data reading and writing operation; and 3), failure data block recovery and has the advantage of capability of effectively lowering write amplification effect of RAID level disk arrays. For conventional magnetic disks, sequential write performance is much higher than random write performance. The address mapping method enables random write operation to be converted into sequential write operation, write performance of the magnetic disk is effectively brought into play, IO operation needed for computation checking is lowered, and response capability of an RAID system is evidently improved, and power consumption of the system is lowered. In addition, an address mapping module can flexibly choose arrangement positions and the number of checking blocks, so that the RAID5 write IO optimization processing method is not limited to be applicable to disk arrays of RAID5 level and can be popularized for realization of RAID4 and RAID6 levels.
Description
Technical field
The present invention relates to the computer data field of storage, what be specifically related to a kind of RAID5 writes the IO optimized treatment method.
Background technology
Disk array RAID(Redundant Access Independent Disks, raid-array) technology is widely used in the data storage as a kind of proven technique.The RAID technology has 0 to 6 seven kind of basic RAID rank from RAID, and different RAID ranks are representing different memory properties, data security and carrying cost.RAID comprises the hard disk of N+P storage data; it is take band as the basis; after the file data piecemeal, write in the data disks; to in other data disks with this data block same disc in all data blocks of position calculate their parity values (xor operation); the final message block that is formed by parity values (check block) that forms; and with its hard disk that writes correspondence position, this is called as data protection.Its benefit has provided fault-tolerant ability, and namely any one or a plurality of (less than P) break down in this N+P disk, can not cause the problem of obliterated data.Because the symmetry of parity checking can with the content of N remaining disk, calculate the correct data content that leaves in the failed disk, although this calculating is more time-consuming.Wherein the verification deposit data is on a fixing hard disk among the RAID4, and this hard disk becomes the bottleneck of whole RAID4 system.RAID5 evenly spreads to check information in the hard disks all in the system for solving this bottleneck, and RAID6 then deposits two parts of checking datas, thereby can tolerate that two disk failure are arranged in the RAID system simultaneously.
The security of RAID system improving is to sacrifice performance as cost.In a typical RAID5 system, as shown in Figure 1, if revise the data of A1, new data is A1 ＇, then need legacy data A1 101 and old verification P (A) 102 are read out, and calculate verification and the P (A023) (the XOR verification is symmetrical) of A0+A2+A3 with them, then use P (A023) and A1 ＇ to calculate new verification P (A) ＇, at last A1 ＇ is write 101, P (A) ＇ and write 102.A write operation can derive twice read operation that hence one can see that and twice write operation (this be called write enlarge-effect), this greatly reduces the effective performance of system.By that analogy, RAID6 has higher security than RAID5, but effective performance is lower.For traditional disk, the address of IO operation is if random, and then its effective performance can reduce hundreds of times.
Chinese patent CN101566924A " a kind of data read-write method of disk array ", by in internal memory, creating the write operation frequency of a disk band state bitmap recording data blocks, for the high disk band of write operation frequency, postpone to upgrade proof mass, to reach the purpose of accelerating disk array read-write treatment effeciency.Chinese patent CN101702113A " write operation method and device ", by setting up the tabulation of write operation requests in internal memory, the buffer memory write operation requests if there is parallel processing during at least two write operation requests on the band, is write enlarge-effect with minimizing.Chinese patent CN101526886 " a kind of data manipulation method of Redundant Array of Independent Disks (RAID) and device " dynamically adjusts stripe size according to the block size of data writing, to improve the reading and writing data performance of RAID.Chinese patent CN102184079 " a kind of write performance optimization method of RAID5 rank disk array " is the performance that the RAID5 disk array promotes system when writing that load requests is poly-to be sent out by suspending method that check bit upgrades.Chinese patent CN102270102 " a kind of optimization method of RAID6 disk array write performance ", the daily record dish is set outside the RAID6 disc array structure, log recording corresponding to data block that is used for record RAID6 disk array, select writing mode (reading rewriting and reconstruct writes) for write request according to the frequency that is updated data block in the band, to reach the purpose of optimizing write performance.Article " based on the RAID5 small letter performance optimization of disk Exclusive-or engine ", the Exclusive-or engine that utilizes part scsi disk and fibre disk to realize, the XOR of new data and legacy data calculates and realizes in disk, and main frame is finished write operation, thereby purpose optimized in the small letter that reaches RAID5.
US Patent No. 6704837 B2 " METHOD AND APPARATUS FOR INCREASING RAID WRITE PERFORMANCE BY MAINTAINING A FULL TRACK WRITE COUNTER " are arranged again, proposition writes counter for band of each magnetic track maintenance for write operation, analyze the Counter Value of the last operation magnetic track of write request, judge whether data writing is whole band, if it is utilize the strip data in the buffer memory to calculate the verification that makes new advances, if not then according to reading old verification, read legacy data, calculate new verification, write the classic method of new data and carry out.The method that this United States Patent (USP) provides is read the number of times of legacy data, old verification by reducing when the calculation check piece system, reduced to a certain extent system overhead, but when the data of write operation were not a band, the caused enlarge-effect of writing of write operation did not change.
Above patent can be found out, existing work is to pass through diverse ways, as postponing to write data, dynamically adjusting the method such as stripe size and optimize write performance, although write performance is had certain optimization, but all fundamentally do not solve the caused enlarge-effect of writing of RAID tolerant fail algorithm write operation, and the risk that exists data to recover is as postponing to write loss of data etc. after the System Sudden power down.And dynamically adjust the methods such as stripe size, increase daily record dish, and then need extra system overhead, increased system burden.
As seen, also there is defective in prior art, haves much room for improvement and develops.
Summary of the invention
What purpose of the present invention proposed a kind of RAID5 writes the IO optimized treatment method, the address mapping method of data block when adopting a kind of write operation of having optimized, the RAID controller is that data block is set up address mapping table, what main frame LBA address was no longer fixing corresponds on the hard disk, but the LBA address of main frame is rearranged according to particular algorithm, thereby the write operation of random address is changed into the write operation of sequence address, reach and need not to read former data block and former check block when writing data block, significantly reduce the IO operand of check disk when writing data by the method, effectively eliminate the enlarge-effect of writing of check disk, solve the existing not high problem of RAID5 rank disk array write performance, optimize the disk array effective performance.
The present invention
OneKind of RAID5 writes the IO optimized treatment method
,Comprise the steps:
1), storage system makes up;
2), data read-write operation;
3), the fail data piece reclaims.
Described RAID system constructing:
(1.1) provide the RAID system:
Described RAID system: comprise RAID controller, storage medium array;
Described storage medium array can be drive bar, solid-state disk row or flash memory row etc.;
Described RAID controller includes address mapping module, is logic function and the device of storage address mapping table, can be DRAM or solid-state disk etc., and mapping table is preserved the map information of each LBA address and PBA address.
(1.2) address mapping table mainly comprises following field information:
I. " logical address " field (LBA address), the logical address of recording data blocks;
II. " physical address " field (PBA address), the physical address of recording data blocks;
III. " valid data " field, this field comprises three states, and the stored data block of the corresponding physical address of " effectively " state representation is latest data; This physical address data piece of engineering noise state representation lost efficacy, and was recyclable; This physical address space of " free time " state representation reclaims;
IV. " check block " field records this data block and whether is the proof test value in the RAID5 disk array;
V. " check block protection " field, whether the data block number that records the protection of this proof test value is less than the data block of a band, this field is mainly used in preserving when disk array generation power down, and whether this band normally finishes write operation;
(1.3) address mapping table is preserved the map information of each LBA address and PBA address, therefore before storage system is used, need to carry out initialization to mapping table, during initialization, " physical address ", " check block " field etc. are set to the initial value of system configuration, and all the other fields are made as effectively.
(2) processing of read-write IO operation
This storage system is identical with general RAID5 disk array, and the proof test value of each band is evenly distributed on each hard disk or the storage medium.
Of the present invention to read the IO operating process simpler, only needs search map entry corresponding to LBA address in address mapping table, obtains PBA address information corresponding to LBA address, then to the hard disk reading out data of the PBA address of correspondence.
Write in the IO operation in processing, the RAID controller is that data block is set up address mapping table and the logical address of main frame is rearranged according to specific process, the write operation of random address is converted to the write operation of sequence address, need not to read former data block and former check block, and need not the backwards calculation check information.Its key step is as follows:
(2.1) when main frame sent the data block write request of arbitrary address, system deposited this data block in buffer memory, and buffer memory comprises N data block and 1 check block take band as unit; Simultaneously the proof test value in this data block and the buffer memory is calculated new verification and cover former proof test value and carry out buffer memory.
Whether the LBA address of (2.2) searching this data block has respective record in address mapping table, if corresponding record is arranged, turn step (2.3), otherwise turn (2.4);
(2.3) illustrate that this writes IO is rewriting operation, in the corresponding record of this data block in address mapping table, it is invalid that " valid data " field all is designated.
(2.4) record that the address mapping table write pointer variable of safeguarding according to the RAID administration module points to, obtain the new PBA address that this data block will write, the LBA address of this data block is write " logical address " field, and this data block write hard disk corresponding to physical address and revise new PBA address corresponding " valid data " be set to effectively.
(2.5) judge the whether tail end of corresponding this band of write pointer variable, if so, turn step (2.6), otherwise turn step (2.7);
(2.6) according to the value of check block address variable, send the check block write order to the hard disk of respective physical address, and address mapping table write pointer variable is sequentially increased progressively, and proof test value is set to 0 in the buffer memory;
(2.7) judge that whether current address mapping table write pointer points to the physical address record that the check block of this band should be deposited, and if so, turns step (2.9);
(2.8) the write pointer variable points to next PBA address, and this write operation finishes;
(2.9) with current PBA address caching to check block address variable, and address mapping table write pointer variable sequentially increased progressively, after finishing, this write operation finishes.
(3) the fail data piece reclaims
Address mapping table be recorded as the continuous data block in PBA address, for guaranteeing in the storage system a certain amount of idle physical space is always arranged, the RAID controller regularly carries out the fail data piece and reclaims (Garbage Collection:GC), the so-called recovery, namely revising information corresponding to PBA address in mapping table gets final product, the RAID administration module is safeguarded " address recovery " pointer (GC pointer), this pointer is used in reference to the address mapping table record that carries out the address reclaimer operation, its address mapping table end to end between the circulation.Its treatment scheme is as follows:
(3.1) in the threshold values of setting, if so, then system is current does not carry out the address reclaimer operation to the idle physical space number percent of judging system, otherwise turns step (3.2).The address is reclaimed threshold values and can be chosen between 0.2 ~ 0.8 according to the application demand of system.
(3.2) judge that whether system carries out the address reclaimer operation for the first time, if so, turns step (3.3), otherwise turns step (3.4);
(3.3) initialization address reclaims pointer, with pointed address mapping table head.
(3.4) check " valid data " field of address mapping table, if judge to be invalid data, then this field is rewritten as idle condition, and turns step (3.6).Otherwise turn step (3.5);
(3.5) the current address mapping table is recorded as valid data, then calls and writes IO operation steps (2), and this data block is write current PBA address as new data, and " valid data " field that should record is set to idle condition.
(3.6) next bar address mapping table record of pointed is reclaimed in the address, determines whether to show tail, if so, turns step (3.3), otherwise turns step (3.2).
The present invention
OneThe advantage of writing the IO optimized treatment method of planting RAID5 can effectively reduce the enlarge-effect of writing of RAID rank disk array.Traditional disk, the sequential write performance is far longer than random write performance, address mapping method provided by the invention changes into the sequential write operation with random write operation, effectively brought into play the write performance of disk, reduced the required IO operation of calculation check, significantly improve the responding ability of RAID system, reduced simultaneously system power dissipation.Simultaneously, the present invention is applied in the solid-state memory system, and the RAID controller has been realized the correlation functions such as address mapping table, so that the design of the controller of solid-state disk is simplified.At last, address mapping module is selection check piece position and the quantity of disposing flexibly, therefore the invention is not restricted to the application of other disk array of RAID5 level, can be generalized to RAID4, other realization of RAID6 level.
Description of drawings
Fig. 1 is the topological structure schematic diagram of existing RAID5 system;
Fig. 2 is the configuration diagram of system of the present invention;
Fig. 3 is address mapping module work schematic diagram provided by the present invention;
Fig. 4 is the read operation process flow diagram in the system provided by the present invention;
Fig. 5 is the write operation address mapping schematic diagram in the system provided by the present invention;
Fig. 6 is the write operation process flow diagram in the system provided by the present invention;
Fig. 7 is the address recovery process figure in the system provided by the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is carried out clear, complete description, described embodiment is the present invention's part example, rather than whole embodiment.Based on embodiments of the invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that obtains under the creative work prerequisite.
The invention provides the RAID system as shown in Figure 2, comprise RAID controller, storage medium array etc., the storage medium array can be drive bar, solid-state disk row or flash memory row etc.The address mapping module of RAID controller is responsible for the mapping of LBA address and PBA address, for data block is set up address mapping table.Address mapping table is preserved the relevant information of each LBA address and PBA address date piece, essential record corresponding logical address, the physical address of this data block, the status informations such as valid data, check block, check block protection.As shown in Figure 3, the mapping table module of RAID controller is responsible for the mapping of LBA address and PBA address.In the PBA address space, be always write operation and keep certain space 311, write operation always is written in the idle space 311, and the current band that is writing is 310, and after current band 310 used, order switched to next idle band.
Step 1: initialization address mapping table
Before storage system is used, need to carry out initialization to mapping table.Before the RAID system used, address mapping module was carried out initialization to address mapping table, and " physical address ", " check block " field etc. are set to the initial value of system configuration, and all the other fields are made as effectively.
Step 2: the read-write operation of RAID system
Of the present invention to read the IO operating process simpler, only needs search map entry corresponding to LBA address in address mapping table, obtains PBA address information corresponding to LBA address, then to the hard disk reading out data of the PBA address of correspondence.As shown in Figure 4, step 401, the RAID controller is received the read request from host side, and response; Step 402 is searched address mapping table, finds PBA address corresponding to LBA; Step 403 sends the PBA read command to hard disk.Step 404, return data is to host side.Step 405, read operation finishes.
Write operation more complicated of the present invention, because random write operation will be mapped as continuous write operation, all write operations need to be mapped in the new continuous space and go, be illustrated in figure 5 as in the N+1 RAID5 system, herein N be 4 only for the signal, be the corresponding check blocks of per 4 data blocks, check block is evenly distributed in 5 hard disks.When host side sent the write request of arbitrary address, mapping table module always was mapped to it on continuous physical address.Concrete steps are as follows, for example main frame sends LBA1 address write request 501, system deposits this data block buffer memory in and calculates new proof test value, the RAID administration module is searched the corresponding record clauses and subclauses according to the LBA1 address in mapping table, it is invalid that " valid data " fields of these records is made as, and according to address mapping module write pointer variable the LBA1 address is mapped to the PBA14 address, upgrade the address mapping table clauses and subclauses of PBA14 and send write order 505 that data block is write in hard disk corresponding to PBA14 address, write pointer increases 1 and points to next bar address mapping table record.According to above-mentioned steps, process successively LBA6 address write request 502, LBA8 address write request 503, LBAf address write request 504, be mapped to respectively 506,507,508, can find out, the write operation after mapped is continuous at physical address.In the RAID controller, cache module is used for data and the check information that buffer memory writes recently, in above-mentioned example, after 501 data arrive, be buffered the module buffer memory, after 502 data arrive, with 501 data and 502 data calculation check and buffer memorys, after 503 data arrive, continue calculation check with the verification of buffer memory, and buffer memory, after 504 data arrive, continue calculation check with the verification of buffer memory, and buffer memory, N is 4 in this example system, after therefore per 4 data blocks arrive, the check block that it is corresponding writes physical address PA5, and buffer memory only need be preserved data and the check information of nearest band.The write operation steps flow chart as shown in Figure 6.
Step 3: fail data piece reclaimer operation
Can see from the above process of writing, may take a plurality of PBA address to the repeatedly write operation of same LBA address, can cause reusing of physical space like this.For a certain amount of idle physical space always being arranged for write operation in the assurance system, the RAID controller can regularly carry out the address reclaimer operation according to setting value.Reclaim the address is the sequential scanning address mapping table, valid data are read out and calls the write operation flow process as new data, and directly reclaim the PBA address that invalid data is corresponding, is about to " valid data " and is made as the free time.The write operation of back can distribute address space from these idle PBA addresses like this.In actual implementation procedure, affect normal write operation for preventing reclaimer operation, usually reserve a part of physical space and reclaim use to the address specially.
Can find out, in the removal process, also all be sequence address to the read-write operation of hard disk.For hard disk, the effective performance of order read-write can reach more than 100 times of random read-write performance.Address reclaimer operation flow process as shown in Figure 7.
Claims (7)
1. that plants RAID5 writes the IO optimized treatment method
,It is characterized in that
:Comprise the steps: 1) the storage system structure; 2) data read-write operation; 3) the fail data piece reclaims.
A kind of RAID5 as claimed in claim 1 write the IO optimized treatment method, it is characterized in that
:Described RAID system constructing: comprising: (1.1) provide the RAID system: (1.2) address mapping table mainly comprises field information: (1.3) address mapping table is preserved the map information of each LBA address and PBA address, therefore before storage system is used, need to carry out initialization to mapping table, during initialization, " physical address ", " check block " field etc. are set to the initial value of system configuration, and all the other fields are made as effectively.
3. a kind of RAID5's as claimed in claim 1 writes the IO optimized treatment method
,It is characterized in that: the processing of the described IO of writing operation comprises: when (2.1) sent the data block write request of arbitrary address when main frame, system deposited this data block in buffer memory, and buffer memory comprises N data block and 1 check block take band as unit; Simultaneously the proof test value in this data block and the buffer memory is calculated new verification and cover former proof test value and carry out buffer memory; Whether the LBA address of (2.2) searching this data block has respective record in address mapping table, if corresponding record is arranged, turn step (2.3), otherwise turn (2.4); (2.3) illustrate that this writes IO is rewriting operation, in the corresponding record of this data block in address mapping table, it is invalid that " valid data " field all is designated; (2.4) record that the address mapping table write pointer variable of safeguarding according to the RAID administration module points to, obtain the new PBA address that this data block will write, the LBA address of this data block is write " logical address " field, and this data block write hard disk corresponding to physical address and revise new PBA address corresponding " valid data " be set to effectively; (2.5) judge the whether tail end of corresponding this band of write pointer variable, if so, turn step (2.6), otherwise turn step (2.7); (2.6) according to the value of check block address variable, send the check block write order to the hard disk of respective physical address, and address mapping table write pointer variable is sequentially increased progressively, and proof test value is set to 0 in the buffer memory; (2.7) judge that whether current address mapping table write pointer points to the physical address record that the check block of this band should be deposited, and if so, turns step (2.9); (2.8) address mapping table write pointer variable sequentially increases progressively the next PBA of sensing address, and this write operation finishes; (2.9) with current PBA address caching to check block address variable, and address mapping table write pointer variable sequentially increased progressively, after finishing, this write operation finishes.
4. a kind of RAID5's as claimed in claim 1 writes the IO optimized treatment method
, it is characterized in that:Described fail data piece reclaims, comprise: (3.1) judge that the idle physical space number percent of system is whether in the threshold values of setting, if, the current address reclaimer operation that do not carry out of system then, otherwise turn step (3.2), the address is reclaimed threshold values and can be chosen between 0.2 ~ 0.8 according to the application demand of system; (3.2) judge that whether system carries out the address reclaimer operation for the first time, if so, turns step (3.3), otherwise turns step (3.4); (3.3) initialization address reclaims pointer, with pointed address mapping table head; (3.4) check " valid data " field of address mapping table, if judge to be invalid data, then this field is rewritten as idle condition, and turns step (3.6), otherwise turn step (3.5); (3.5) the current address mapping table is recorded as valid data, then calls and writes IO operation steps (2), and this data block is write current PBA address as new data, and " valid data " field that should record is set to idle condition; (3.6) next bar address mapping table record of pointed is reclaimed in the address, determines whether to show tail, if so, turns step (3.3), otherwise turns step (3.2).
5. a kind of RAID5's as claimed in claim 2 writes the IO optimized treatment method
,It is characterized in that
:Described RAID system: comprise RAID controller, storage medium array.
6. a kind of RAID5's as claimed in claim 2 writes the IO optimized treatment method
,It is characterized in that: it is as follows that (1.2) address mapping table mainly comprises field information
:" logical address " field (LBA address), the logical address LBA address of recording data blocks; " physical address " field (PBA address), the physical address of recording data blocks storage; " valid data " field, this field comprise three states, and the stored data block of the corresponding physical address of " effectively " state representation is latest data; This physical address data piece of engineering noise state representation lost efficacy, and was recyclable; This physical address space of " free time " state representation reclaims; " check block " field records this data block and whether is the proof test value in the RAID5 disk array; " check block protection " field records the data block number of this proof test value protection whether less than the data block of a band, and this field is mainly used in preserving when disk array generation power down, and whether this band normally finishes write operation.
7. a kind of RAID5's as claimed in claim 5 writes the IO optimized treatment method
,It is characterized in that: described storage medium array can be drive bar, solid-state disk row or flash memory row etc.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210582379.2A CN103049222B (en) | 2012-12-28 | 2012-12-28 | A kind of RAID5 writes IO optimized treatment method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210582379.2A CN103049222B (en) | 2012-12-28 | 2012-12-28 | A kind of RAID5 writes IO optimized treatment method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103049222A true CN103049222A (en) | 2013-04-17 |
| CN103049222B CN103049222B (en) | 2016-05-25 |
Family
ID=48061875
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210582379.2A Active CN103049222B (en) | 2012-12-28 | 2012-12-28 | A kind of RAID5 writes IO optimized treatment method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103049222B (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488583A (en) * | 2013-09-09 | 2014-01-01 | 华中科技大学 | High-performance reliable solid-state disk realizing method |
| CN104407813A (en) * | 2014-11-20 | 2015-03-11 | 上海宝存信息科技有限公司 | RAID (Redundant Array of Independent Disk) system and method based on solid-state storage medium |
| CN104598345A (en) * | 2015-02-09 | 2015-05-06 | 浪潮电子信息产业股份有限公司 | Method for pretesting Oracle system IO performance by using ORION |
| CN104765575A (en) * | 2015-04-23 | 2015-07-08 | 成都博元时代软件有限公司 | Information storage processing method |
| CN104778100A (en) * | 2015-04-23 | 2015-07-15 | 成都博元时代软件有限公司 | Safe data backup method |
| CN105138288A (en) * | 2015-08-19 | 2015-12-09 | 浪潮(北京)电子信息产业有限公司 | Storage method and storage device for realizing RAID5 function, and main control chip system |
| CN105426130A (en) * | 2015-11-19 | 2016-03-23 | 四川神琥科技有限公司 | Rapid mail handling method |
| WO2016082550A1 (en) * | 2014-11-28 | 2016-06-02 | 华为技术有限公司 | Method and apparatus for flushing cache data |
| CN106293543A (en) * | 2016-08-25 | 2017-01-04 | 中国科学院计算技术研究所 | A kind of based on incidentally reclaim watt record a RAID write ordered method and device |
| CN106528441A (en) * | 2016-10-26 | 2017-03-22 | 珠海格力电器股份有限公司 | Data processing method and device of simulate EEPROM and electronic equipment |
| CN106649132A (en) * | 2016-12-29 | 2017-05-10 | 记忆科技(深圳)有限公司 | Solid-state drive junk recovery method |
| CN106886368A (en) * | 2016-12-30 | 2017-06-23 | 北京同有飞骥科技股份有限公司 | A kind of block device writes IO shapings and multi-controller synchronization system and synchronous method |
| WO2017140262A1 (en) * | 2016-02-18 | 2017-08-24 | 华为技术有限公司 | Data updating technique |
| CN107402890A (en) * | 2017-07-24 | 2017-11-28 | 华中科技大学 | A kind of data processing method and system based on Solid-state disc array and caching |
| CN107562646A (en) * | 2017-08-28 | 2018-01-09 | 记忆科技(深圳)有限公司 | A kind of method for lifting solid-state storage garbage reclamation performance |
| CN107577551A (en) * | 2017-09-06 | 2018-01-12 | 郑州云海信息技术有限公司 | A kind of solid state disk write fail processing method and system |
| CN107870744A (en) * | 2017-10-27 | 2018-04-03 | 上海新储集成电路有限公司 | The hybrid hard disk array storage system and method for a kind of asynchronous mirror image |
| CN109491593A (en) * | 2018-09-28 | 2019-03-19 | 方信息科技（上海）有限公司 | A kind of data storage management system and method |
| CN110399310A (en) * | 2018-04-18 | 2019-11-01 | 杭州宏杉科技股份有限公司 | A kind of recovery method and device of memory space |
| CN111125033A (en) * | 2018-10-31 | 2020-05-08 | 深信服科技股份有限公司 | Space recovery method and system based on full flash memory array |
| CN111338559A (en) * | 2018-12-19 | 2020-06-26 | 深圳市茁壮网络股份有限公司 | Data processing method, user module, information management module and system |
| CN112214175A (en) * | 2020-10-21 | 2021-01-12 | 重庆紫光华山智安科技有限公司 | Data processing method, data processing device, data node and storage medium |
| CN113204554A (en) * | 2021-05-11 | 2021-08-03 | 深圳市杉岩数据技术有限公司 | Method and device for realizing sparse writing of object storage system and electronic equipment |
| CN114327270A (en) * | 2021-12-23 | 2022-04-12 | 郑州云海信息技术有限公司 | Request processing method, device, equipment and readable storage medium |
| CN114415968A (en) * | 2022-01-26 | 2022-04-29 | 南京晓庄学院 | Storage system and data writing method thereof |
| CN115129265A (en) * | 2022-09-01 | 2022-09-30 | 苏州浪潮智能科技有限公司 | Partitioned caching method, device, equipment and readable medium for redundant array of independent disks |
| CN115793985A (en) * | 2023-01-09 | 2023-03-14 | 苏州浪潮智能科技有限公司 | Safe storage method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6266648B1 (en) * | 1996-06-25 | 2001-07-24 | Baker, Iii Bernard R. | Benefits tracking and correlation system for use with third-party enabling organizations |
| CN101539842A (en) * | 2009-04-27 | 2009-09-23 | 杭州华三通信技术有限公司 | Data wiring method of disk array system and disk array system |
| CN101916228A (en) * | 2010-08-17 | 2010-12-15 | 中国人民解放军国防科学技术大学 | Flash translation layer (FTL) with data compression function and implementation method |
-
2012
- 2012-12-28 CN CN201210582379.2A patent/CN103049222B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6266648B1 (en) * | 1996-06-25 | 2001-07-24 | Baker, Iii Bernard R. | Benefits tracking and correlation system for use with third-party enabling organizations |
| CN101539842A (en) * | 2009-04-27 | 2009-09-23 | 杭州华三通信技术有限公司 | Data wiring method of disk array system and disk array system |
| CN101916228A (en) * | 2010-08-17 | 2010-12-15 | 中国人民解放军国防科学技术大学 | Flash translation layer (FTL) with data compression function and implementation method |
Cited By (47)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488583B (en) * | 2013-09-09 | 2016-08-10 | 华中科技大学 | The solid-state disk implementation method that a kind of high-performance is highly reliable |
| CN103488583A (en) * | 2013-09-09 | 2014-01-01 | 华中科技大学 | High-performance reliable solid-state disk realizing method |
| CN104407813A (en) * | 2014-11-20 | 2015-03-11 | 上海宝存信息科技有限公司 | RAID (Redundant Array of Independent Disk) system and method based on solid-state storage medium |
| CN104407813B (en) * | 2014-11-20 | 2019-02-19 | 上海宝存信息科技有限公司 | A kind of RAID system and method based on solid storage medium |
| US10353614B2 (en) | 2014-11-20 | 2019-07-16 | Shannon Systems Ltd. | Raid system and method based on solid-state storage medium |
| WO2016082550A1 (en) * | 2014-11-28 | 2016-06-02 | 华为技术有限公司 | Method and apparatus for flushing cache data |
| CN104598345B (en) * | 2015-02-09 | 2017-12-19 | 浪潮电子信息产业股份有限公司 | A kind of method using ORION pretest Oracle system IO performances |
| CN104598345A (en) * | 2015-02-09 | 2015-05-06 | 浪潮电子信息产业股份有限公司 | Method for pretesting Oracle system IO performance by using ORION |
| CN104765575A (en) * | 2015-04-23 | 2015-07-08 | 成都博元时代软件有限公司 | Information storage processing method |
| CN104778100A (en) * | 2015-04-23 | 2015-07-15 | 成都博元时代软件有限公司 | Safe data backup method |
| CN104765575B (en) * | 2015-04-23 | 2017-09-15 | 成都博元时代软件有限公司 | information storage processing method |
| CN105138288A (en) * | 2015-08-19 | 2015-12-09 | 浪潮(北京)电子信息产业有限公司 | Storage method and storage device for realizing RAID5 function, and main control chip system |
| CN105426130A (en) * | 2015-11-19 | 2016-03-23 | 四川神琥科技有限公司 | Rapid mail handling method |
| CN105426130B (en) * | 2015-11-19 | 2018-06-29 | 四川神琥科技有限公司 | Mail immediate processing method |
| US11422703B2 (en) | 2016-02-18 | 2022-08-23 | Huawei Technologies Co., Ltd. | Data updating technology |
| WO2017140262A1 (en) * | 2016-02-18 | 2017-08-24 | 华为技术有限公司 | Data updating technique |
| US11698728B2 (en) | 2016-02-18 | 2023-07-11 | Huawei Technologies Co., Ltd. | Data updating technology |
| US10725662B2 (en) | 2016-02-18 | 2020-07-28 | Huawei Technologies Co., Ltd. | Data updating technology |
| CN106293543B (en) * | 2016-08-25 | 2019-03-12 | 中国科学院计算技术研究所 | It is a kind of that ordered method and device is write based on watt record RAID incidentally recycled |
| CN106293543A (en) * | 2016-08-25 | 2017-01-04 | 中国科学院计算技术研究所 | A kind of based on incidentally reclaim watt record a RAID write ordered method and device |
| CN106528441A (en) * | 2016-10-26 | 2017-03-22 | 珠海格力电器股份有限公司 | Data processing method and device of simulate EEPROM and electronic equipment |
| CN106649132A (en) * | 2016-12-29 | 2017-05-10 | 记忆科技(深圳)有限公司 | Solid-state drive junk recovery method |
| CN106886368A (en) * | 2016-12-30 | 2017-06-23 | 北京同有飞骥科技股份有限公司 | A kind of block device writes IO shapings and multi-controller synchronization system and synchronous method |
| CN106886368B (en) * | 2016-12-30 | 2019-08-16 | 北京同有飞骥科技股份有限公司 | A kind of block device writes IO shaping and multi-controller synchronization system and synchronous method |
| CN107402890A (en) * | 2017-07-24 | 2017-11-28 | 华中科技大学 | A kind of data processing method and system based on Solid-state disc array and caching |
| CN107402890B (en) * | 2017-07-24 | 2020-10-02 | 华中科技大学 | Data processing method and system based on solid-state disk array and cache |
| CN107562646A (en) * | 2017-08-28 | 2018-01-09 | 记忆科技(深圳)有限公司 | A kind of method for lifting solid-state storage garbage reclamation performance |
| CN107577551A (en) * | 2017-09-06 | 2018-01-12 | 郑州云海信息技术有限公司 | A kind of solid state disk write fail processing method and system |
| CN107870744A (en) * | 2017-10-27 | 2018-04-03 | 上海新储集成电路有限公司 | The hybrid hard disk array storage system and method for a kind of asynchronous mirror image |
| CN110399310A (en) * | 2018-04-18 | 2019-11-01 | 杭州宏杉科技股份有限公司 | A kind of recovery method and device of memory space |
| CN110399310B (en) * | 2018-04-18 | 2021-08-31 | 杭州宏杉科技股份有限公司 | Method and device for recovering storage space |
| CN109491593B (en) * | 2018-09-28 | 2020-11-17 | 方一信息科技（上海）有限公司 | Data storage management system and method |
| CN109491593A (en) * | 2018-09-28 | 2019-03-19 | 方信息科技（上海）有限公司 | A kind of data storage management system and method |
| CN111125033A (en) * | 2018-10-31 | 2020-05-08 | 深信服科技股份有限公司 | Space recovery method and system based on full flash memory array |
| CN111125033B (en) * | 2018-10-31 | 2024-04-09 | 深信服科技股份有限公司 | Space recycling method and system based on full flash memory array |
| CN111338559A (en) * | 2018-12-19 | 2020-06-26 | 深圳市茁壮网络股份有限公司 | Data processing method, user module, information management module and system |
| CN111338559B (en) * | 2018-12-19 | 2023-11-10 | 深圳市茁壮网络股份有限公司 | Data processing method, user module, information management module and system |
| CN112214175A (en) * | 2020-10-21 | 2021-01-12 | 重庆紫光华山智安科技有限公司 | Data processing method, data processing device, data node and storage medium |
| CN113204554B (en) * | 2021-05-11 | 2023-03-31 | 深圳市杉岩数据技术有限公司 | Method and device for realizing sparse writing of object storage system and electronic equipment |
| CN113204554A (en) * | 2021-05-11 | 2021-08-03 | 深圳市杉岩数据技术有限公司 | Method and device for realizing sparse writing of object storage system and electronic equipment |
| CN114327270A (en) * | 2021-12-23 | 2022-04-12 | 郑州云海信息技术有限公司 | Request processing method, device, equipment and readable storage medium |
| CN114327270B (en) * | 2021-12-23 | 2023-09-15 | 郑州云海信息技术有限公司 | Request processing method, device, equipment and readable storage medium |
| CN114415968A (en) * | 2022-01-26 | 2022-04-29 | 南京晓庄学院 | Storage system and data writing method thereof |
| CN114415968B (en) * | 2022-01-26 | 2023-04-07 | 南京晓庄学院 | Storage system and data writing method thereof |
| CN115129265A (en) * | 2022-09-01 | 2022-09-30 | 苏州浪潮智能科技有限公司 | Partitioned caching method, device, equipment and readable medium for redundant array of independent disks |
| WO2024046213A1 (en) * | 2022-09-01 | 2024-03-07 | 苏州元脑智能科技有限公司 | Method and apparatus for caching blocks of redundant array of independent disks, and device and readable medium |
| CN115793985A (en) * | 2023-01-09 | 2023-03-14 | 苏州浪潮智能科技有限公司 | Safe storage method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103049222B (en) | 2016-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103049222B (en) | A kind of RAID5 writes IO optimized treatment method | |
| US5309451A (en) | Data and parity prefetching for redundant arrays of disk drives | |
| US7856528B1 (en) | Method and apparatus for protecting data using variable size page stripes in a FLASH-based storage system | |
| CN101566924B (en) | Data read-write method of disk array | |
| CN104471546B (en) | Effective redundant arrays of inexpensive disks skill and technique for improving solid state hard disc reliability | |
| US6523087B2 (en) | Utilizing parity caching and parity logging while closing the RAID5 write hole | |
| KR100827677B1 (en) | A method for improving I/O performance of RAID system using a matrix stripe cache | |
| US9292228B2 (en) | Selective raid protection for cache memory | |
| CN101236482B (en) | Method for processing data under degrading state and independent redundancy magnetic disc array system | |
| KR101678868B1 (en) | Apparatus for flash address translation apparatus and method thereof | |
| EP0586117A1 (en) | Optimized stripe detection for redundant arrays of disk drives | |
| US20110213919A1 (en) | FLASH-based Memory System with Static or Variable Length Page Stripes Including Data Protection Information and Auxiliary Protection Stripes | |
| CN102799533B (en) | Method and apparatus for shielding damaged sector of disk | |
| CN101436149B (en) | Method for rebuilding data of magnetic disk array | |
| CN103870352B (en) | Method and system for data storage and reconstruction | |
| JPH04230512A (en) | Method and apparatus for updating record for dasd array | |
| US10564865B2 (en) | Lockless parity management in a distributed data storage system | |
| CN108958656B (en) | Dynamic stripe system design method based on RAID5 solid state disk array | |
| JP2013156977A (en) | Elastic cache of redundant cache data | |
| CN105353991A (en) | Disk array reconstruction optimization method and device | |
| US9106260B2 (en) | Parity data management for a memory architecture | |
| CN102270102A (en) | Method for optimizing writing performance of RAID6 (Redundant Array of Independent Disks) disk array | |
| WO2011019794A2 (en) | Method and apparatus for addressing actual or predicted failures in a flash-based storage system | |
| US11379326B2 (en) | Data access method, apparatus and computer program product | |
| CN107402890A (en) | A kind of data processing method and system based on Solid-state disc array and caching |
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 |