CN103645997B - A kind of method and system of data protection - Google Patents
A kind of method and system of data protection Download PDFInfo
- Publication number
- CN103645997B CN103645997B CN201310733189.0A CN201310733189A CN103645997B CN 103645997 B CN103645997 B CN 103645997B CN 201310733189 A CN201310733189 A CN 201310733189A CN 103645997 B CN103645997 B CN 103645997B
- Authority
- CN
- China
- Prior art keywords
- disk
- read
- write
- redundant array
- mistake
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention discloses the method and system of a kind of data protection, wherein, said method comprising the steps of: S1, to redundant array of inexpensive disk occurring disk read-write mistake catch, obtain the area information of medium read-write error；S2, according to described area information, by normal for Data Migration therein to medium region；S3, re-starts mapping at logical layer volume.Use such scheme, reconstruction time is fast, without HotSpare disk, the problem finding medium read-write error in time, achieve the fast data recovery of redundant array of inexpensive disk, and decrease the movement replacing it rear magnetic head, reduce the performance impact that bad block is replaced, improve the robustness of redundant array of inexpensive disk, there is the highest market using value.
Description
Technical field
The present invention relates to the data protection of redundancy magnetic disk group, in particular, a kind of data protection
Method and system.
Background technology
Existing storage system, has high requirement to the reliability of data, and the storage of data is situated between
Matter is but unlikely to be the most reliably, once media damage, then will produce damage difficult to the appraisal
Lose.Although now with redundancy magnetic disk group (RAID, Redundant Arrays of Inexpensive Disks)
Reduce the damnous probability of media damage, but the function that existing redundant array data are recovered is but
There is a following limitation:
One, reconstruction time is long, and present reconstruction is all that unit is rebuild by one piece of dish, with
The capacity dish now is increasing, and reconstruction time is more and more longer, and the reliability of system can not be guaranteed；
Two, need one piece of HotSpare disk, otherwise cannot realize rebuilding, thus need to waste some
Memory space.
Summary of the invention
The technical problem to be solved is to provide the method and system of a kind of new data protection.
Technical scheme is as follows: a kind of method of data protection, and it comprises the following steps: S1,
To redundant array of inexpensive disk occurring disk read-write mistake catch, obtain the region of medium read-write error
Information；S2, according to described area information, by normal for Data Migration therein to medium region；S3,
Mapping is re-started at logical layer volume.
Preferably, in described method, step S1 specifically includes following steps: in redundant array of inexpensive disk
Occur retrying during disk read-write mistake, it is judged that whether disk read-write mistake can repair, and otherwise reports
Medium read-write error and area information thereof.
Preferably, in described method, in step S2, described Data Migration presses redundant array of inexpensive disk band
Alignment.
Preferably, in described method, step S2 specifically includes following steps: according to described area information,
Align by redundant array of inexpensive disk band, read and according to pre-configured strategy, will appear from disk read-write wrong
The data of the wrong redundant array of inexpensive disk band at place, move on normal redundant array of inexpensive disk band by mistake.
Preferably, in described method, described area information includes logical block addresses and length.
Preferably, in described method, step S3 specifically includes following steps: revise institute at logical layer volume
State the logical mappings of mistake redundant array of inexpensive disk band place physical address.
Preferably, in described method, after step S3, also perform following steps: described logic reflected
In the amendment write metadata penetrated.
Preferably, in described method, before step S1, also execution step S0: pre-configured described strategy.
Preferably, in described method, in step S0, after described strategy is included in each virtual disk
Arranging a replacement region, the bad block for corresponding virtual disk is replaced.
The another technical scheme of the present invention is as follows: the system of a kind of data protection, including Logical Disk pipe crimping
Reason layer, wherein, described Logical Disk volume management layer Provisioning Policy configuration module, disk read-write error trapping
Module, Data Migration module and modified module；Described strategy configuration module and described Data Migration module
Connect, for pre-configured strategy and store；Described disk read-write error trapping module is moved with described data
Shifting formwork block connects, for occurring retrying during disk read-write mistake in redundant array of inexpensive disk, it is judged that
When disk read-write mistake repeats to occur, report medium read-write error and area information thereof；Described data are moved
Shifting formwork block is also connected with described modified module, for according to described area information, by redundant array of inexpensive disk
Band aligns, and reads and according to pre-configured strategy, and the mistake that will appear from disk read-write mistake place is superfluous
The data of residual magnetism disk array band, move on normal redundant array of inexpensive disk band；Described modified module
For reflecting in the logic of logical layer volume amendment described mistake redundant array of inexpensive disk band place physical address
Penetrate.
Using such scheme, the present invention finds media error in time, triggers active data transfer behavior,
There is the highest market using value.
Accompanying drawing explanation
Fig. 1 is the functional realiey schematic diagram of one embodiment of the present of invention；
Fig. 2 is that the replacement region of one embodiment of the present of invention arranges schematic diagram；
Fig. 3 is that the LVM of one embodiment of the present of invention monitors schematic diagram；
Fig. 4 is the mapping structure schematic diagram of prior art；
Fig. 5 is the mapping structure schematic diagram of one embodiment of the present of invention.
Detailed description of the invention
For the ease of understanding the present invention, below in conjunction with the accompanying drawings and specific embodiment, the present invention is carried out more
Detailed description.Accompanying drawing gives the preferred embodiment of the present invention.But, the present invention can be with
Many different forms realize, however it is not limited to the embodiment described by this specification.On the contrary, carry
Purpose for these embodiments is to make the understanding to the disclosure more thorough comprehensively.
It should be noted that when element is referred to as " being fixed on " another element, and it can directly exist
On another element or element placed in the middle can also be there is.When an element is considered as " connection "
Another element, it can be directly to another element or may be simultaneously present centering elements.
Term " vertical ", " level ", "left", "right" and the similar table that this specification is used
State for illustrative purposes only.
Unless otherwise defined, this specification is used all of technology and scientific terminology with belong to this
The implication that bright those skilled in the art are generally understood that is identical.Saying in the present invention in this specification
Term used in bright book is intended merely to describe the purpose of specific embodiment, is not intended to limit this
Invention.The term "and/or" that this specification is used includes one or more relevant Listed Items
Arbitrary and all of combination.
One embodiment of the present of invention is, a kind of method of data protection, and it comprises the following steps: S1,
To redundant array of inexpensive disk occurring disk read-write mistake catch, obtain the region of medium read-write error
Information；S2, according to described area information, by normal for Data Migration therein to medium region；S3,
Mapping is re-started at logical layer volume.Preferably, in described method, in step S2, described data are moved
Move and press the alignment of redundant array of inexpensive disk band.Wherein, use LVM(Logical Volume Manager,
Logical Disk volume management) mode manages described redundant array of inexpensive disk；LVM be built upon hard disk and subregion it
On a logical layer, improve disk partition management motility.LVM does several underlying device
Become physical volume, physical volume group, it is thus possible to format, can be extended and shrink, for storing data.
Its process is PD(block device) → PV(physical volume) → VG(physical volume group) → LV(logic
Subregion).By using existing LVM pattern, all physical disks and subregion, no matter theirs is big
Little and distribution mode how, is all conceptualized as single storage (single storage) source.
Such as, subregion can be become virtual disk (VD, a Virtual by LVM with disk aggregation
Disk), thus a unified large space is formed with little memory space.This virtual disk is referred to as volume
Group (volume group).Therefore the file system bigger than maximum disk can be set up, it is also possible to
In disk pool, add disk and subregion, existing file system is extended online, and for example, use
One 160GB disk replaces two 80GB disks, without allowing system off-line, it is not required that
Manual transfer data between disk；When memory space exceedes required amount of space, remove from pond
Disk, thus reduce file system.
Such as, said method comprising the steps of: S1, redundant array of inexpensive disk occurs disk read-write
Retry during mistake, it is judged that whether disk read-write mistake can repair, and otherwise reports medium read-write error
And area information；Wherein, the purpose retried is to judge these zone errors whether unrepairable, because
Some areas of dielectric retry can recover normal after writing or retry reading, does not affect subsequent use；Such as,
Step S1 judging, whether disk read-write mistake can repair, is not deal with, or only report upper strata
Or manager.S2, according to described area information, by normal for Data Migration therein to medium region；
S3, re-starts mapping at logical layer volume.Such as, step S1 obtains disk read-write mistake letter in real time
Breath or obtain in real time disk read-write information and also judge whether disk read-write mistake, be, retry with
Judge whether disk read-write mistake repeats to occur.And for example, identical file or onesize file are used
Retry, to judge whether disk read-write mistake repeats to occur.And for example.Employing timing mechanism judges
Whether there is disk read-write mistake.
And for example, said method comprising the steps of: S1, redundant array of inexpensive disk occurs disk read-write
Retry during mistake, it is judged that whether disk read-write mistake repeats to occur, and is, report medium read-write mistake
Miss and area information；S2, according to described area information, aligns by redundant array of inexpensive disk band, reads
Take and according to pre-configured strategy, will appear from the wrong redundant array of inexpensive disk bar at disk read-write mistake place
The data of band, move on normal redundant array of inexpensive disk band；S3, re-starts at logical layer volume and reflects
Penetrate.Preferably, in described method, described area information includes logical block addresses and length.According to
The IO and LBA(Logical Block therein of the medium read-write error reported
Address/Addressing, logical block addresses/addressing) and LENGTH(length), carry out by battle array
Row band aligns, and according to strategy, will appear from disk read-write mistake place RAID stripe (strip RAID)
Data, by configuration strategy, move on the RAID stripe at other normal band places.Use
RAID stripe technology, leaves in data on the subregion of two or more hard disk, and these data have one
Being partly on a hard disk, second half is on another hard disk, and such data are same from two hard disks
Time read out, such that it is able to improve the speed of disk read-write.Such as, weight in logical layer rolls up a layer
Newly map；And for example, after remapping, transmission reports notice.
Preferably, in described method, step S3 specifically includes following steps: revise institute at logical layer volume
State the logical mappings of mistake redundant array of inexpensive disk band place physical address.Preferably, in described method,
After step S3, also perform following steps: the amendment of described logical mappings is write metadata
(Metadata), in, metadata is used to describe the data of data, as minimum a kind of data unit,
Metadata can be that data illustrate its element or attribute, such as title, size, data type etc., or
Its structure, such as length, field, data row etc., or its related data, be such as positioned at where, as
What contact, owner etc..So contributing to misregistration information, after system reboot, new reflects
Penetrate and still come into force.Preferably, in step S3, further comprise the steps of: reconstruction RAID.Preferably, step
In rapid S3, further comprise the steps of: according to the amended described logical mappings in metadata, complete RAID
Reconstruction.
Preferably, in described method, before step S1, also execution step S0: pre-configured described strategy.
Such as, strategy is arranged according to the object of data protection, and and for example, strategy sets according to the rank of data protection
Put.Preferably, described strategy setting replaces region；Preferably, in described method, in step S0,
Described strategy arranges a replacement region after being included in each virtual disk, for corresponding virtual disk
Bad block is replaced.Such as, refer to Fig. 2, a replacement region is set after virtual disk V0, then
Set virtual disk V1 again；And for example, arranging a replacement region after each virtual disk, its capacity is
The 0.1%-0.2% of previous virtual disk capacity, or, replacing field capacity is previous virtual disk capacity
0.05%-0.08% with a rear virtual disk capacity sum；And for example, the size in this replacement region is dynamically adjusted
Whole, according to the residual capacity of previous virtual disk or virtual according to after LVM distribution during system start-up
The residual capacity of disk adjusts, and such as, is dynamically adjusted to the 1%-2% of residual capacity.Preferably, institute
State strategy also include distribution or adjust redundant array of inexpensive disk band.So, by redundant array
When there is disk read-write mistake, catch in time, thus know the place, region that media error occurs,
Then triggering active data transfer behavior, the area data that will appear from disk read-write mistake moves to new
Medium normal region, after Data Migration completes, re-starts mapping in logic volume layer.So, after
Digital independent write be all to use new normal band, it is achieved thereby that redundant array of inexpensive disk is quick
Recover.
It is conjointly employed in any of the above-described related embodiment, it is preferred that present invention also offers a kind of data
The system of protection, including Logical Disk volume management layer, wherein, described Logical Disk volume management layer Provisioning Policy
Configuration module, disk read-write error trapping module, Data Migration module and modified module.That is, existing
Some LVM increase following several functional module: strategy configures module, disk read-write error trapping mould
The modified module that block, Data Migration module and logical volume map.It realizes as it is shown in figure 1, raid0
Disk0, disk1 and disk2 are written and read operation, find disk generation read-write error (Disk io err)
Time, carry out vdisk(VD) map amendment, then perform bad block data and migrate.
Described strategy configuration module is connected with described Data Migration module, for pre-configured strategy and store；
Described disk read-write error trapping module is connected with described Data Migration module, in redundancy magnetic disk battle array
Row occur retry during disk read-write mistake, it is judged that when disk read-write mistake repeats to occur, report
Medium read-write error and area information thereof；Described Data Migration module is also connected with described modified module,
For according to described area information, aliging by redundant array of inexpensive disk band, read and according to pre-configured
Strategy, will appear from the data of the wrong redundant array of inexpensive disk band at disk read-write mistake place, moves to
On normal redundant array of inexpensive disk band；Described modified module is superfluous for revising described mistake at logical layer volume
The logical mappings of residual magnetism disk array band place physical address.
Strategy configuration module, for configuring the strategy of replacement.In order to optimize performance, replace it at bad block
The movement of rear minimizing magnetic disc head, retains certain space after each VD, as in figure 2 it is shown,
The bad block that this space is exclusively used in VD is replaced.The size replacing region can calculate according to configuration strategy.
Such as: calculate in proportion, a VD size is 1T, then retain 1/1000 space of VD size
For replacing, i.e. 1G.
Disk read-write error trapping module, after there is disk read-write mistake, retries, if
Remain as disk read-write mistake, then by the read-write IO(input/output of this medium read-write error, defeated
Entering output, also referred to as " read-write ") information is to being reported LVM.Preferably, disk read-write error trapping mould
Block finds disk read-write mistake in time, then retries.
LVM monitoring module, i.e. Data Migration module, according to the IO of the medium read-write error reported, root
According to LBA and LENGTH of IO, carry out aliging by array stripe, according to strategy, will appear from disk
The data of read-write error place RAID stripe, by the strategy of configuration, move to other normal band institutes
RAID stripe on.As it is shown on figure 3, LVM monitoring module knows DISK medium read-write error
Lba and len of IO, then finds one according to strategy to RAID and replaces band, and at Data Migration
During LOCK live this band, then revise mapping address.
Modified module, after Data Migration completes, revises the logical mappings of this band place physical address.
The most original is mapped as an extent(expansion area, is also called " section ") corresponding one piece of physical areas
Territory, as shown in Figure 4.The most a lot of file system all have employed extent and substitute block(block, again
Being referred to as in " block district ", usual magnetic head once can read a block) carry out hyperdisk.Extent is just
Being some continuous print block, an extent is defined plus length by the block initiateed, such as,
Section is by logical block side-play amount (logical offset), length (length) and physical address (physical
Address) tlv triple formed describes.Wherein may be calculated thing by logical block side-play amount and length
Reason address.Physical volume (PV, physical volume) refers to fdisk or logically divides with disk
District has the equipment of said function, such as RAID, be LVM substantially store logical block, but with substantially
Physical storage medium different, physical storage medium such as subregion, disk etc., PV includes and LVM
Relevant management parameters.As it is shown in figure 5, originally point to and be replaced region after mapping amendment
Extent1 is modified to point to replace region.After mapping amendment completes, the amendment write this mapped is first
In data, in order to after restarting, new mapping still comes into force.
Preferably, this system also sets up transmission module in mistake, and it is with disk read-write error trapping module even
Connect, for occur disk read-write mistake time, report medium read-write error and area information thereof to upper strata,
Such as, manager.Preferably, this system also sets up reconstruction module, for map amendment before or
After person, rebuild RAID.
Further, embodiments of the invention also include, each technical characteristic of the various embodiments described above, phase
The method and system of the data protection that combination is formed mutually, replaces by achieving bad block at VD logical layer,
Can be combined with each other with the reconstruction etc. of RAID, promote further RAID vigorousness (robustness,
Also known as robustness)；If disk is not enough, it is also possible to is used alone and protects data；Replace by band alignment
Region, thus after replacing it, do not affect performance；Also by configuration strategy, reserve and replace requisite space,
And headspace is all after each VD, thus reduce the movement replacing it rear magnetic head, lower bad
The performance impact that block is replaced.
The technique effect of the various embodiments described above includes that reconstruction time is fast, it is not necessary to HotSpare disk, finds in time to be situated between
The problem of matter read-write error, it is achieved that the fast data recovery of redundant array of inexpensive disk, and decrease and replace
The movement of magnetic head after alternatively, reduces the performance impact that bad block is replaced, improves redundant array of inexpensive disk
Robustness, has the highest market using value.
It should be noted that above-mentioned each technical characteristic continues to be mutually combined, formed the most enumerated above
Various embodiments, are accordingly to be regarded as the scope that description of the invention is recorded；Further, to ordinary skill
For personnel, can be improved according to the above description or be converted, and all these modifications and variations are all
The protection domain of claims of the present invention should be belonged to.
Claims (7)
1. the method for a data protection, it is characterised in that comprise the following steps:
S1, to occurring disk read-write mistake to catch in redundant array of inexpensive disk, obtains medium read-write mistake
Area information by mistake；
S2, according to described area information, by normal for Data Migration therein to medium region；
S3, re-starts mapping at logical layer volume；
Wherein, step S1 specifically includes following steps: occur that in redundant array of inexpensive disk disk read-write is wrong
Mistake and retry, it is judged that whether disk read-write mistake can repair, otherwise report medium read-write error and
Its area information；In step S2, described Data Migration presses the alignment of redundant array of inexpensive disk band；Step S2
Specifically include following steps: according to described area information, align by redundant array of inexpensive disk band, read
And according to pre-configured strategy, will appear from the wrong redundant array of inexpensive disk band at disk read-write mistake place
Data, move on normal redundant array of inexpensive disk band.
Method the most according to claim 1, it is characterised in that described area information includes logic
Block address and length.
Method the most according to claim 2, it is characterised in that step S3 specifically includes following
Step: the logic at logical layer volume amendment described mistake redundant array of inexpensive disk band place physical address is reflected
Penetrate.
The most according to claim 3, method, it is characterised in that after step S3, also perform
Following steps: the amendment of described logical mappings is write in metadata.
The most according to claim 4, method, it is characterised in that before step S1, also perform
Step S0: pre-configured described strategy.
Method the most according to claim 5, it is characterised in that in step S0, described strategy
Arranging a replacement region after being included in each virtual disk, the bad block for corresponding virtual disk is replaced.
7. a system for data protection, including Logical Disk volume management layer, it is characterised in that described
Logical Disk volume management layer Provisioning Policy configuration module, disk read-write error trapping module, Data Migration mould
Block and modified module；
Described strategy configuration module is connected with described Data Migration module, for pre-configured strategy and store；
Described disk read-write error trapping module is connected with described Data Migration module, at redundancy magnetic
Disk array occurs retry during disk read-write mistake, it is judged that when disk read-write mistake repeats to occur,
Report medium read-write error and area information thereof；
Described Data Migration module is also connected with described modified module, is used for according to described area information,
Align by redundant array of inexpensive disk band, read and according to pre-configured strategy, will appear from disk read-write wrong
The data of the wrong redundant array of inexpensive disk band at place, move on normal redundant array of inexpensive disk band by mistake；
Described modified module is at logical layer volume amendment described mistake redundant array of inexpensive disk band place thing
The logical mappings of reason address.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310733189.0A CN103645997B (en) | 2013-12-26 | 2013-12-26 | A kind of method and system of data protection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310733189.0A CN103645997B (en) | 2013-12-26 | 2013-12-26 | A kind of method and system of data protection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103645997A CN103645997A (en) | 2014-03-19 |
| CN103645997B true CN103645997B (en) | 2016-12-07 |
Family
ID=50251218
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310733189.0A Active CN103645997B (en) | 2013-12-26 | 2013-12-26 | A kind of method and system of data protection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103645997B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108228081B (en) | 2016-12-21 | 2021-03-02 | 伊姆西Ip控股有限责任公司 | Method and apparatus for data protection |
| CN109240841B (en) * | 2018-07-02 | 2021-12-14 | 广东睿江云计算股份有限公司 | Fault discovery and isolation method and device for combined file system |
| CN111124266B (en) * | 2018-10-31 | 2023-09-29 | 伊姆西Ip控股有限责任公司 | Data management method, apparatus and computer program product |
| CN109445715B (en) * | 2018-11-09 | 2022-02-18 | 郑州云海信息技术有限公司 | Bad block processing method and device for combined block |
| CN111913647B (en) * | 2019-05-08 | 2022-10-11 | 华为技术有限公司 | Wear leveling method and device for storage equipment and related equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1597674B1 (en) * | 2003-02-10 | 2008-04-09 | Netezza Corporation | Rapid regeneration of failed disk sector in a distributed database system |
| CN102799533A (en) * | 2012-07-10 | 2012-11-28 | 浙江宇视科技有限公司 | Method and apparatus for shielding damaged sector of disk |
| CN102981930A (en) * | 2012-11-15 | 2013-03-20 | 浪潮电子信息产业股份有限公司 | Automatic restoration method for disk array multi-level data |
| CN103064804A (en) * | 2012-12-13 | 2013-04-24 | 华为技术有限公司 | Method and device for access control of disk data |
| CN103389918A (en) * | 2013-07-24 | 2013-11-13 | 北京鲸鲨软件科技有限公司 | Repair method for false fault in RAID (Redundant Array of Independent Disks) system |
-
2013
- 2013-12-26 CN CN201310733189.0A patent/CN103645997B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1597674B1 (en) * | 2003-02-10 | 2008-04-09 | Netezza Corporation | Rapid regeneration of failed disk sector in a distributed database system |
| CN102799533A (en) * | 2012-07-10 | 2012-11-28 | 浙江宇视科技有限公司 | Method and apparatus for shielding damaged sector of disk |
| CN102981930A (en) * | 2012-11-15 | 2013-03-20 | 浪潮电子信息产业股份有限公司 | Automatic restoration method for disk array multi-level data |
| CN103064804A (en) * | 2012-12-13 | 2013-04-24 | 华为技术有限公司 | Method and device for access control of disk data |
| CN103389918A (en) * | 2013-07-24 | 2013-11-13 | 北京鲸鲨软件科技有限公司 | Repair method for false fault in RAID (Redundant Array of Independent Disks) system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103645997A (en) | 2014-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8239714B2 (en) | Apparatus, system, and method for bad block remapping | |
| CN101916173B (en) | RAID (Redundant Array of Independent Disks) based data reading and writing method and system thereof | |
| CN103645997B (en) | A kind of method and system of data protection | |
| US9311006B2 (en) | Table journaling in flash storage devices | |
| JP4815825B2 (en) | Disk array device and method for reconstructing the same | |
| US8127182B2 (en) | Storage utilization to improve reliability using impending failure triggers | |
| CN104503706B (en) | A kind of data storage and read method based on disk array | |
| US20110264949A1 (en) | Disk array | |
| CN104461391A (en) | Method and system for managing and processing metadata of storage equipment | |
| CN104778018B (en) | Wide band disk array and storage method based on asymmetric hybrid magnetic disk mirroring | |
| CN103577121A (en) | High-reliability linear file access method based on nand flash | |
| CN101609420A (en) | Realize method and the redundant arrays of inexpensive disks and the controller thereof of rebuilding of disc redundant array | |
| CN102508733B (en) | A kind of data processing method based on disk array and disk array manager | |
| CN101567211A (en) | Method for improving usability of disk and disk array controller | |
| CN101504625A (en) | Method for implementing independent disk redundancy array, solid state disk and electronic equipment | |
| CN104166601B (en) | The backup method and device of a kind of data storage | |
| CN104142872A (en) | Quick reestablishment method for RAID (redundan array of inexpensive disk) | |
| CN102968361A (en) | RAID (Redundant Array of Independent Disk) data self-repairing method | |
| CN103678025B (en) | A kind of disk failure processing method in disk array | |
| CN103019623A (en) | Memory disc processing method and device | |
| US11093339B2 (en) | Storage utilizing a distributed cache chain and a checkpoint drive in response to a data drive corruption | |
| CN107239233A (en) | A kind of disk array and data reconstruction method and storage device based on it | |
| CN106095330A (en) | A kind of storage method and apparatus of metadata | |
| US7133965B2 (en) | Raid storage device | |
| CN102609370B (en) | Solid-state disk and method for preventing data loss thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | 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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190114 Address after: 518000 Fuyuan Industrial Building, 138 Freedom Road, Baoan 47 District, Baoan District, Shenzhen City, Guangdong Province, 721-2, 7th floor Patentee after: Shenzhen Feiteng Xin'an Technology Co., Ltd. Address before: 518000 South Pass No. 2 Zhiheng Strategic Emerging Industrial Park, Nanshan District, Shenzhen City, Guangdong Province, 30 buildings and 5 floors Patentee before: Shenzhen Data Fault Tolerance System Co., Ltd. |