CN106095330A - A kind of storage method and apparatus of metadata - Google Patents
A kind of storage method and apparatus of metadata Download PDFInfo
- Publication number
- CN106095330A CN106095330A CN201610375013.6A CN201610375013A CN106095330A CN 106095330 A CN106095330 A CN 106095330A CN 201610375013 A CN201610375013 A CN 201610375013A CN 106095330 A CN106095330 A CN 106095330A
- Authority
- CN
- China
- Prior art keywords
- disk
- metadata
- memory area
- raid
- mapping relations
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0608—Saving storage space on storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0644—Management of space entities, e.g. partitions, extents, pools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
- G06F3/0689—Disk arrays, e.g. RAID, JBOD
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides the storage method and apparatus of a kind of metadata, is applied to include in the RAID of N number of member's disk, and the redundancy of RAID is that M, N are more than M；The method includes: safeguard the mapping relations of member's disk and memory area；Wherein, for each member's disk, from N number of member's disk, choose L member's disk, and from each member's disk of L member's disk, choose a memory area, and safeguard member's disk and the mapping relations of L the memory area chosen；L is less than or equal to N more than M, L, and is divided into L memory area in the metadata space of each member's disk；When needing the metadata storing member's disk, utilize described mapping relations, inquire about L the memory area that this member's disk is corresponding, and store metadata in this L memory area.By the technical scheme of the application, can save the memory space of disk, improve space availability ratio, improve operating efficiency and the operating characteristics of metadata, renewal speed efficiency is high.
Description
Technical field
The application relates to technical field of memory, particularly relates to the storage method and apparatus of a kind of metadata.
Background technology
RAID (Redundant Array of Independent Disks, RAID) be one will be many
The physical disk of block independence, differently combines, and forms a disk group, thus provides than single physical disk more
High storage performance, and the technology of data backup is provided.According to different Method of Data Organizations, conventional RAID can include
RAID0, RAID1, RAID5, RAID6, RAID10 etc..Wherein, the physical disk of composition RAID is referred to as member's disk, and this member
The space of disk includes metadata space and data space, and data space is used for storing real data, and metadata space is used
In storage metadata.Metadata includes such as, RAID is by which member's disk for managing RAID and the data of member's disk thereof
Composition, the state of RAID, the organizational information of RAID, the reconstruction information etc. of member's disk.
Each member's disk of RAID has the metadata of self, and is stored in the metadata space of this member disk.Example
As, the metadata 1 of member's disk 1 is stored in the metadata space of this member disk 1, and the metadata 2 of member's disk 2 is stored in this
The metadata space of member's disk 2, the metadata 3 of member's disk 3 is stored in the metadata space of this member disk 3, with this type of
Push away.Further, during the read-write of RAID, member's disk likely goes offline, therefore to ensure first number of member's disk
According to not losing, generally the metadata of each member's disk is stored on all member's disks.As it is shown in figure 1, will become
The metadata 1 of member's disk 1 is sequentially stored in the metadata space of member's disk 1-member's disk 8, by the metadata of member's disk 2
The 2 metadata spaces being sequentially stored in member's disk 1-member's disk 8, by that analogy.
But, if the quantity of member's disk is a lot, then need the metadata space at each member's disk to store greatly
The metadata of amount, such as, when the quantity of member's disk is 64, the metadata space of the most each member's disk stores 64 one-tenth
The metadata of member's disk.So, the disk space that metadata takies is excessive, causes the waste of disk space.In addition, it is desirable to will be every
The metadata of individual member's disk stores the metadata space of each member's disk successively, and the operation causing more new metadata can not
Concurrently carrying out, renewal speed efficiency is low.
Summary of the invention
The application provides a kind of storage method of metadata, is applied to include the independent disk redundancy battle array of N number of member's disk
In row RAID, the redundancy of described RAID is that M, N are more than M；Described method includes:
Safeguard the mapping relations of member's disk and memory area；Wherein, for each member's disk, from N number of member's disk
In choose L member's disk, and from each member's disk of L member's disk, choose a memory area, and safeguard described
The mapping relations of member's disk and L the memory area chosen；L is less than or equal to N more than M, L, and in the unit of each member's disk
Data space is divided into L memory area；
When needing the metadata storing member's disk, utilize described mapping relations, inquire about L corresponding to this member's disk
Memory area, and described metadata is stored in this L memory area.
The determination mode of described L specifically includes: determine that redundancy M that backup quantity L is described RAID of metadata adds 1.
The described process choosing L member's disk from N number of member's disk, specifically includes: for each member's disk, from
N number of member's disk is chosen described member's disk, and chooses L-1 member's disk from remaining N-1 member's disk.
The described process choosing L-1 member's disk from remaining N-1 member's disk, specifically includes: from remaining
In N-1 member's disk, choose L-1 the member disk adjacent with described member's disk；Wherein, adjacent with described member's disk
L-1 member's disk refer to: in member's drive bar table space, if located in the member's disk number after described member's disk
Amount is not less than L-1, and the most described L-1 member's disk includes being positioned at described member's disk continuous L-1 member's disk below；
If located in the member's number of disks after described member's disk less than L-1, the most described L-1 member's disk includes being positioned at institute
Stating member's disk all member's disks below, remaining member's disk is the first member from described member's drive bar table space
Continuous member's disk that disk starts.
The described process choosing a memory area from each member's disk of L member's disk, specifically includes: from
First member's disk to l-th member's disk sequentially, chooses first memory area successively to l-th memory area.
The described process that described metadata is stored in this L memory area, specifically includes: store at least two at needs
During the metadata of member's disk, if L memory area corresponding to described at least two member's disk is in different member's magnetic
Dish, the most concurrently performs to be stored in the metadata of described at least two member's disk the process of L memory area of correspondence.
Described method also includes:
When needing the metadata reading member's disk, utilize described mapping relations, inquire about L corresponding to this member's disk
Memory area, and from this L memory area memory area, read metadata.
Described method also includes:
When N number of member's disk there being member's disk occur abnormal, if increasing newcomer's disk in described RAID, then
Utilize described mapping relations, obtain from member's disk of no exceptions and first number of record in abnormal member's disk occurs
According to, and the metadata of acquisition is stored in the memory area of described newcomer's disk.
The application provides the storage device of a kind of metadata, is applied to include the independent disk redundancy battle array of N number of member's disk
In row RAID, the redundancy of described RAID is that M, N are more than M；Described device includes:
Maintenance module, for safeguarding the mapping relations of member's disk and memory area；Wherein, safeguarding member's disk and depositing
During the mapping relations in storage area territory, for each member's disk, from N number of member's disk, choose L member's disk, and from
Each member's disk of L member's disk is chosen a memory area, and safeguards described member's disk and L the storage chosen
The mapping relations in region；L less than or equal to N more than M, L, and is divided into L in the metadata space of each member's disk and deposits
Storage area territory；
Processing module, for when needing the metadata storing member's disk, utilizing described mapping relations, inquire about this member
L the memory area that disk is corresponding, and described metadata is stored in this L memory area.
Described maintenance module, specifically for during determining described L, determines that backup quantity L of metadata is described
Redundancy M of RAID adds 1；During choosing L member's disk from N number of member's disk, for each member's disk, from
Described N number of member's disk is chosen described member's disk, and chooses L-1 member's disk from remaining N-1 member's disk；
During choosing L-1 member's disk from remaining N-1 member's disk, from remaining N-1 member's disk, choosing
Take L-1 the member disk adjacent with described member's disk；One is being chosen from each member's disk of described L member's disk
During individual memory area, from first member's disk to l-th member's disk sequentially, choose first successively and deposit
Storage area territory is to l-th memory area；Wherein, L-1 adjacent with described member's disk member's disk refers to: at member's drive bar
In table space, if located in the member's number of disks after described member's disk not less than L-1, the most described L-1 member's magnetic
Dish includes being positioned at described member's disk continuous L-1 member's disk below, if located in the member after described member's disk
Number of disks is less than L-1, and the most described L-1 member's disk includes being positioned at described member's disk all member's disks below,
Remaining member's disk is continuous member's disk that the first member's disk from described member's drive bar table space starts；
Described processing module, specifically for, during described metadata is stored in this L memory area, working as needs
During the metadata of storage at least two member's disk, if L memory area corresponding to described at least two member's disk is in
Different member's disks, the most concurrently performs to be stored in the metadata of described at least two member's disk L memory block of correspondence
The process in territory；
Described processing module, is additionally operable to, when needing the metadata reading member's disk, utilize described mapping relations, inquiry
L the memory area that this member's disk is corresponding, and from this L memory area memory area, read metadata；With/
Or, described processing module, it is additionally operable to when described N number of member's disk there being member's disk occur abnormal, if at described RAID
During middle increase newcomer's disk, then utilize described mapping relations, obtain from member's disk of no exceptions and exception occurs
The metadata of record in member's disk, and the metadata of acquisition is stored in the memory area of described newcomer's disk.
Based on technique scheme, in the embodiment of the present application, can only store metadata in depositing of a small amount of member's disk
In storage area territory rather than store metadata in the memory area of all member's disks, thus avoid the magnetic that metadata takies
Disk space is excessive, it is to avoid the waste of disk space, saves the memory space of disk, improves space availability ratio.Furthermore, it is possible to it is concurrent
Perform to be stored in the metadata of at least two member's disk the process of memory area, improve operating efficiency and the operation of metadata
Performance, renewal speed efficiency is high.
Accompanying drawing explanation
In order to the embodiment of the present application or technical scheme of the prior art are clearly described, below will be to the application
In embodiment or description of the prior art, the required accompanying drawing used is briefly described, it should be apparent that, in describing below
Accompanying drawing is only some embodiments described in the application, for those of ordinary skill in the art, it is also possible to according to these
Accompanying drawing obtains other accompanying drawing.
Fig. 1 is the schematic diagram that the metadata of each member's disk is stored in all member's disks；
Fig. 2 is the flow chart of the storage method of the metadata in a kind of embodiment of the application；
Fig. 3 is the structural representation of the metadata storage in a kind of embodiment of the application；
Fig. 4 and Fig. 5 is that the member's disk in a kind of embodiment of the application occurs abnormal schematic diagram；
Fig. 6 is the hardware structure diagram of the storage device in a kind of embodiment of the application；
Fig. 7 is the structure chart of the storage device of metadata in a kind of embodiment of the application.
Detailed description of the invention
At term used in this application merely for the sake of describing the purpose of specific embodiment, and unrestricted the application.This Shen
Please with claims used in " a kind of ", " described " and " being somebody's turn to do " of singulative be also intended to include most form, unless
Context clearly shows that other implication.It is also understood that term "and/or" used herein refers to comprise one or more
Any or all of the project of listing being associated may combination.
Although should be appreciated that in the application possible employing term first, second, third, etc. to describe various information, but this
A little information should not necessarily be limited by these terms.These terms are only used for same type of information is distinguished from each other out.Such as, without departing from
In the case of the application scope, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as
One information.Depend on linguistic context, additionally, the word used " if " can be construed to " ... time " or " when ... "
Or " in response to determining ".
For problems of the prior art, the embodiment of the present application proposes a kind of storage method of metadata, the party
Method can apply to include in the RAID of N number of member's disk, and the redundancy of RAID is M.In one example, the redundancy of RAID
Degree refers to when RAID normally reads and writes, it is allowed to the maximum quantity of abnormal member's disk occurs, if the redundancy of RAID5 is 1,
The redundancy of RAID6 is 2.Based on this, when the quantity that abnormal member's disk occurs is less than or equal to the redundancy of RAID, RAID
Still can normally read and write.In order to realize the process that RAID normally reads and writes, then quantity N of member's disk redundancy M more than RAID.
In one example, the space of member's disk includes metadata space and data space, and data space is used for storing
Real data, and metadata space is used for storing metadata.In the subsequent process of the embodiment of the present application, this metadata includes
For managing RAID and the data of member's disk thereof, such as, RAID which member's disk is made up of, the state of RAID, RAID
Organizational information, the reconstruction information etc. of member's disk.
In one example, each member's disk of RAID has the metadata of self, and such as, it is right that member's disk 1 has
The metadata 1 answered, member's disk 2 has the metadata 2 etc. of correspondence, by that analogy.
Under above-mentioned application scenarios, as in figure 2 it is shown, the storage method of this metadata may comprise steps of:
Step 201, safeguards the mapping relations of member's disk and memory area；Wherein, for each member's disk, from N number of
Member's disk is chosen L member's disk, and from each member's disk of L member's disk, chooses a memory area, and
The mapping relations of L the memory area safeguarding member's disk and choose；L is less than or equal to N more than M, L, and at each member's disk
Metadata space in be divided into L memory area.
Step 202, when needing the metadata storing member's disk, utilizes above-mentioned mapping relations, inquires about this member's disk
L corresponding memory area, and this metadata is stored in this L memory area.
For step 201, in one example, the determination mode of L specifically can include but not limited to: determines metadata
Backup quantity L is that redundancy M of RAID adds 1.Certainly, if redundancy M that backup quantity L of metadata is more than RAID, and little
In quantity N equal to member's disk, it is not limiting as the concrete numerical value of L, such as the redundancy that backup quantity L is RAID of metadata
Degree M adds 2, adds 3, adds 4.
Wherein, in order to ensure that RAID can normally read and write, then need to record the metadata of all member's disks.Therefore, for
Realizing when the redundancy occurring the quantity of abnormal member's disk less than or equal to RAID, RAID still can normally read and write,
For the metadata of each member's disk, then, in addition to there is abnormal member's disk, at least also need to member's magnetic
In dish, record has this metadata, this way it is secured that record the metadata of all member's disks in RAID, the most normally reads
Write.Such as, the metadata 1 of member's disk 1 records the metadata space at member's disk 1, member's disk 2 and member's disk 3, false
If the redundancy of RAID is 2, then when member's disk 1 breaks down, due to only one of which member's disk failures, and RAID
Redundancy be 2, therefore RAID can normally read and write, and exists in the metadata space of now member's disk 2 and member's disk 3
Metadata 1, it may therefore be assured that the normal read-write of RAID.When member's disk 1 and member's disk 2 break down, due to only
Two member's disk failures, and the redundancy of RAID is 2, therefore RAID can normally read and write, and now member's disk 3
Metadata space exists metadata 1, it may therefore be assured that the normal read-write of RAID.When member's disk 1, member's disk 2 with become
When member's disk 3 breaks down, owing to there being three member's disk failures, and the redundancy of RAID is 2, and therefore RAID cannot be just
Often read-write, the most just need not ensure at least also have record in member's disk to have metadata 1.
In above process, it is ensured that the reason of the normal read-write of RAID is: when data are written to fault member's disk 1
Certain block of data space time, due to member's disk 1 fault, therefore can write data into other by redundant arithmetic
Certain block of the data space of member's disk, and in the data space of member's disk 1, the data mistake of this block.Treat as
Member disk 1 fault recovery time, RAID it is to be appreciated that member's disk 1 data space in, the data of this block be mistake information,
It is thus desirable to record this information in metadata 1, and by storing metadata in the metadata space of multiple member's disks
1 so that RAID it is known that member's disk 1 data space in, the data of this block be mistake information, the most normally read and write.
Based on above-mentioned analysis, then backup quantity L of metadata needs redundancy M more than RAID.Additionally, in order to save magnetic
The memory space of dish, improves space availability ratio, then backup quantity L of metadata is the smaller the better.Therefore based on above-mentioned two strategy,
Then backup quantity L of metadata can be that redundancy M of RAID adds 1.
In one example, need in the metadata space of each member's disk, mark off L memory area, such as,
When L is 4, then can mark off first memory area, second memory block in the metadata space of each member's disk
Territory, the 3rd memory area, the 4th memory area etc..Wherein, when dividing memory area, the size of the most each memory area
Need the size more than or equal to metadata.
In one example, from N number of member's disk, choose the process of L member's disk, specifically can include but not limit
In following manner: for each member's disk, from N number of member's disk, choose this member's disk, and from remaining N-1 member
Disk is chosen L-1 member's disk.Such as, when N is 8, and L is 4, for member's disk 1, then choose this member's disk 1, and
3 member's disks are chosen, as chosen member's disk 3, member's disk 5, member's disk 7 from remaining 7 member's disks.When
So, it would however also be possible to employ alternate manner chooses member's disk, as member's disk 1, then choose member's disk 2, member's disk 4,
Member's disk 6, member's disk 8 etc..In the embodiment of the present application, repeat no more for choosing the mode of member's disk, as long as can
Select L different member's disk, from N number of member's disk, choose this member's disk with above-mentioned, and from remaining N-1
Illustrate as a example by individual member's disk is chosen L-1 member's disk.
In one example, from remaining N-1 member's disk, choose the process of L-1 member's disk, the most permissible
Include but not limited to following manner: from remaining N-1 member's disk, choose L-1 the member magnetic adjacent with member's disk
Dish.Wherein, L-1 adjacent with member's disk member's disk refers to: in member's drive bar table space, if located in member's magnetic
Dish member's number of disks below is not less than L-1, then L-1 member's disk includes being positioned at member's disk continuous L-1 below
Individual member's disk；If located in the member's number of disks after member's disk less than L-1, then L-1 member's disk includes position
All member's disks after member's disk, remaining member's disk is the first member's disk from member's drive bar table space
The continuous member's disk started.It is of course also possible to use alternate manner to choose L-1 member's disk, as member's disk 1,
It not to choose member's disk 2, member's disk 3, member's disk 4, but choose member's disk 2, member's disk 5, member's disk 8
Deng.In the embodiment of the present application, repeat no more for choosing the mode of L-1 member's disk, as long as L-1 difference can be selected
Member's disk, with above-mentioned from remaining N-1 member's disk, choose L-1 the member magnetic adjacent with member's disk
Illustrate as a example by dish.
In one example, from each member's disk of L member's disk, the process of a memory area is chosen, specifically
Following manner can be included but not limited to: from first member's disk to l-th member's disk sequentially, choose successively
One memory area is to l-th memory area.Such as, choose first memory area from first member's disk, from second
Second memory area chosen by member's disk, by that analogy, chooses l-th memory area from l-th member's disk.Certainly, also
Alternate manner can be used to choose memory area, as chosen second memory area from first member's disk, from second one-tenth
The 3rd memory area chosen by member's disk, by that analogy, chooses first memory area from l-th member's disk.The application is real
Execute in example, repeat no more for choosing the mode of a memory area from each member's disk, as long as can be from each member
Disk is chosen a memory area, with above-mentioned from first member's disk to l-th member's disk sequentially, depends on
Secondary choose first memory area and illustrate to l-th memory area.
Below in conjunction with the application scenarios shown in Fig. 3, said process is described in detail.In figure 3, N number of member's disk divides
Not Wei member's disk 1, member's disk 2, member's disk 3, member's disk 4, member's disk 5, member's disk 6, member's disk 7, become
Member's disk 8, additionally, redundancy M of RAID is 3.
During determining backup quantity L of metadata, it may be determined that L be M+1, i.e. L be 4.
At member's disk 1, member's disk 2, member's disk 3, member's disk 4, member's disk 5, member's disk 6, member's magnetic
Dish 7, member's disk 8 metadata space in be respectively divided out 4 memory areas, i.e. first memory area, second storage
Region, the 3rd memory area, the 4th memory area etc..
During choosing L member's disk from N number of member's disk, for member's disk 1,4 members of selection
Disk can comprise member's disk 1, it is also possible to does not comprise member's disk 1, as a example by comprising member's disk 1.For remaining 3
Member's disk, can be from member's disk 2, member's disk 3, member's disk 4, member's disk 5, member's disk 6, member's disk 7, one-tenth
Member disk 8 in, randomly choose 3 member's disks, it is also possible to choose 3 adjacent with member's disk 1 member's disk, with choose with
As a example by 3 member's disks that member's disk 1 is adjacent.For other member's disk, process similar with the process of member's disk 1.
In member's drive bar table space, comprise successively member's disk 1, member's disk 2, member's disk 3, member's disk 4,
Member's disk 5, member's disk 6, member's disk 7, member's disk 8.For member's disk 1, it is positioned at member's disk 1 member below
Number of disks is 7, and not less than 3, therefore for member's disk 1,4 member's disks of selection are member's disk 1, member's magnetic
Dish 2, member's disk 3, member's disk 4.In like manner, for member's disk 2,4 member's disks of selection are member's disk 2, member
Disk 3, member's disk 4, member's disk 5.For member's disk 3,4 member's disks of selection are member's disk 3, member's disk
4, member's disk 5, member's disk 6.For member's disk 4,4 member's disks of selection are member's disk 4, member's disk 5, become
Member disk 6, member's disk 7.For member's disk 5,4 member's disks of selection are member's disk 5, member's disk 6, member's magnetic
Dish 7, member's disk 8.For member's disk 6, being positioned at member's disk 6 member's number of disks below is 2, less than 3, therefore
For member's disk 6,4 member's disks of selection are member's disk 6, member's disk 7, member's disk 8, member's disk 1, member
Disk 7 and member's disk 8 are member's disk all member's disks below, and member's disk 1 is from member's drive bar table space
Member's disk of starting of first member's disk.For member's disk 7,4 member's disks of selection are member's disk 7, member
Disk 8, member's disk 1, member's disk 2, member's disk 8 is member's disk all member's disks below, and member's disk 1 He
Member's disk 2 is continuous member's disk that the first member's disk from member's drive bar table space starts.For member's disk 8,
4 the member's disks selected are member's disk 8, member's disk 1, member's disk 2, member's disk 3, member's disk 1, member's disk
2 and member's disk 3 be continuous member's disk that the first member's disk from member's drive bar table space starts.
During choosing a memory area from each member's disk of L member's disk, for member's disk
1,4 member's disks of selection are respectively member's disk 1, member's disk 2, member's disk 3, member's disk 4, are therefore member's magnetic
The memory area that dish 1 is chosen can be any memory area of member's disk 1, member's disk 2, member's disk 3, member's disk 4,
Such as first memory area of member's disk 1, first memory area of member's disk 2, first memory block of member's disk 3
Territory, first memory area of member's disk 4.In one example, the memory area selected for member's disk 1 can be:
First memory area of member's disk 1, second memory area of member's disk 2, the 3rd memory block of member's disk 3
Territory, the 4th memory area of member's disk 4.For other member's disk, choose the mode of memory area and for member's magnetic
The mode that memory area chosen by dish 1 is identical, does not repeats them here.
Based on above-mentioned analysis, for each member's disk, laterally correspondence self and L-1 member's magnetic of rear adjacent
Dish, be longitudinally corresponding in turn to first memory area, second memory area, by that analogy, l-th memory area.Therefore, member
The mapping relations of disk and memory area can be as it is shown on figure 3, first memory area of member's disk 1 and member's disk 1, become
Member's second memory area of disk 2, the 3rd memory area of member's disk 3, the 4th memory area of member's disk 4
Mapping relations；Member's disk 2 and first memory area of member's disk 2, second memory area, member of member's disk 3
3rd memory area of disk 4, the mapping relations of the 4th memory area of member's disk 5；By that analogy, member's disk 8
With first memory area of member's disk 8, second memory area of member's disk 1, the 3rd memory block of member's disk 2
Territory, the mapping relations of the 4th memory area of member's disk 3.
For step 202, when needing the metadata storing member's disk 1, it is possible to use above-mentioned mapping relations, inquire
First memory area of L memory area of this member's disk 1 correspondence, i.e. member's disk 1, second of member's disk 2 deposit
Storage area territory, the 3rd memory area of member's disk 3, the 4th memory area of member's disk 4, therefore, it can member's magnetic
The metadata of dish 1 is respectively stored in second memory area, member of first memory area of member's disk 1, member's disk 2
3rd memory area of disk 3, the 4th memory area of member's disk 4.In like manner, other member's disk is stored needing
During metadata, its storage mode is similar with the storage mode of the metadata of member's disk 1, does not repeats them here.
In one example, store metadata in the process of L memory area, specifically can include but not limited to as follows
Mode: when needing the metadata storing at least two member's disk, if L corresponding to described at least two member's disk is deposited
Storage area territory is in different member's disks, then it is right concurrently to perform to be stored in the metadata of described at least two member's disk
The process of L the memory area answered.
Wherein, in the ablation process of metadata, can control the member's disk not conflicted to be carried out concurrent operations simultaneously,
When writing the metadata of certain block member's disk, other of identical member disk the most corresponding with this member's disk can be simultaneously written
The metadata of member's disk.Such as, for member's disk 1, corresponding member's disk 1, member's disk 2, member's disk 3, member's magnetic
Dish 4, and for member's disk 5, corresponding member's disk 5, member's disk 6, member's disk 7, member's disk 8, therefore, member's disk
4 memory areas that 4 memory areas of 1 correspondence are corresponding with member's disk 5, are in different member's disks, two member's magnetic
All memory areas that dish is corresponding overlap, and therefore, the metadata of member's disk 1 are being respectively stored in member's disk 1
First memory area, second memory area of member's disk 2, the 3rd memory area, member's disk of member's disk 3
During 4th memory area of 4, the metadata of member's disk 5 can be respectively stored in the of member's disk 5 simultaneously
One memory area, second memory area of member's disk 6, the 3rd memory area of member's disk 7, member's disk 8
4th memory area.Furthermore, it is possible to concurrently perform the metadata of member's disk 2 and member's disk 6 is stored in depositing of correspondence
The process in storage area territory, can concurrently perform the metadata of member's disk 3 and member's disk 7 is stored in corresponding memory area
Process, can concurrently perform to be stored in the metadata of member's disk 4 and member's disk 8 process of the memory area of correspondence.And
Other is combined, owing to all memory areas overlap, therefore cannot concurrently perform, such as member's disk 1 and member's magnetic
The memory area of dish 4, overlaps on member's disk 4.
Based on technique scheme, in the embodiment of the present application, can only store metadata in depositing of a small amount of member's disk
In storage area territory (such as redundancy M+1 the memory area of RAID) rather than store metadata in the storage of all member's disks
In region, thus the disk space avoiding metadata to take is excessive, it is to avoid the waste of disk space, and the storage saving disk is empty
Between, improve space availability ratio.Furthermore, it is possible to concurrently perform the metadata of at least two member's disk is stored in memory area
Process, improves operating efficiency and the operating characteristics of metadata, and renewal speed is in hgher efficiency, provided the user more smooth
Business experience.
In the embodiment of the present invention, for the reading process of metadata, then when needing the metadata reading member's disk, can
To utilize the mapping relations of member's disk and memory area, inquire about L the memory area that this member's disk is corresponding, and from this L
One memory area of memory area reads metadata.
Such as, when needing the metadata reading member's disk 1, it is possible to use member's disk closes with the mapping of memory area
System, L the memory area inquiring member's disk 1 is: first memory area of member's disk 1, second of member's disk 2
Memory area, the 3rd memory area of member's disk 3, the 4th memory area of member's disk 4.Based on this, then from arbitrarily
One memory area reads the metadata of member's disk 1, as can be to read member's magnetic in the first of member's disk 1 memory area
The metadata of dish 1, or, the metadata of member's disk 1 can be read from the second of member's disk 2 memory area, or,
The metadata of member's disk 1 can be read from the 3rd of member's disk 3 the memory area, or, can be from member's disk 4
4th memory area reads the metadata of member's disk 1.
In one example, for the reading process of metadata, needing to read the metadata of at least two member's disk
Time, can concurrently read the metadata of at least two member's disk, as long as concurrently not reading at least on same member's disk
The metadata of two member's disks.Such as, when needing the metadata reading member's disk 1 and member's disk 2, Ke Yicong
First memory area of member's disk 1 reads the metadata of member's disk 1, and deposits from first of member's disk 2 simultaneously
Storage area territory is read the metadata of member's disk 2, and on member's disk 2, does not read member's disk 1 and member's disk 2 simultaneously
Metadata.
In one example, for the reading process of metadata, when reading metadata, it is possible to specify need to read unit's number
According to first member's disk (member's disk that i.e. metadata is corresponding), directly read first memory area of this member's disk
In metadata.If reading successfully, then can terminate the reading process of metadata.If reading unsuccessfully, then can read tight
Metadata in second memory area of adjacent first member's disk thereafter.If reading successfully, then can terminate unit's number
According to reading process.If reading unsuccessfully, then can read the 3rd memory area of next-door neighbour's second member's disk thereafter
In metadata；By that analogy, read correct metadata always, and terminate the reading process of metadata.
Such as, when reading the metadata of member's disk 1, member's disk corresponding to metadata is member's disk 1, therefore needs
First member's disk of metadata to be read is member's disk 1.Based on this, directly read first storage of member's disk 1
Metadata in region.If reading successfully, then terminate the reading process of metadata.If reading unsuccessfully, then read member's magnetic
Metadata in second memory area of dish 2.If reading successfully, then terminate the reading process of metadata.Lose if read
Lose, then read the metadata in the 3rd memory area of member's disk 3.If reading successfully, then terminate the reading of metadata
Process.If reading unsuccessfully, then read the metadata in the 4th memory area of member's disk 4；By that analogy, read always
To correct metadata, and terminate the reading process of metadata.
In the embodiment of the present application, for the recovery process of metadata, abnormal when N number of member's disk there being member's disk occur
Time, if increasing newcomer's disk in RAID, then can utilize the mapping relations of member's disk and memory area, never occur
Abnormal member's disk gets the metadata of record in abnormal member's disk occurs, and the metadata of acquisition is stored in
In the memory area of described newcomer's disk.
Wherein, member's disk occurs extremely can include but not limited to: member's disk goes offline, member's disk generation medium wrong
By mistake etc..
Wherein, when member's disk occurs abnormal, the metadata of this member's magnetic disc recording is inoperable, but is not sending out
On raw abnormal member's disk, metadata (metadata i.e. backed up) yet suffers from, and the unit of abnormal member's disk now occurs
Data still can normally be read and write, and as shown in Figure 4, is that 3 pieces of member's disks occur abnormal schematic diagram.Although member's disk 3, member
There is exception in disk 4, member's disk 5, but, the metadata of member's disk 3, remain stored on member's disk 6, can be based on
The metadata of member's disk 3 read and write by member's disk 6.Additionally, the metadata of member's disk 4, remain stored in member's disk 6 and become
On member's disk 7, the metadata of member's disk 4 can be read and write based on member's disk 6 and member's disk 7.First number of member's disk 5
According to, remain stored on member's disk 6, member's disk 7 and member's disk 8, can be based on member's disk 6, member's disk 7 with become
The metadata of member's disk 5 read and write by member's disk 8.
In one example, owing to the metadata of member's disk 3 is only stored on member's disk 6, therefore metadata is only
Portion, it is very risky that this metadata is lost.Based on this, when there being member's disk to occur abnormal, if increasing new in RAID
Member's disk, i.e. uses newcomer's disk to replace and abnormal member's disk occurs, carry out metadata recovery, then can utilize member
Disk and the mapping relations of memory area, get from member's disk of no exceptions and occur to remember in abnormal member's disk
The metadata of record, and the metadata of acquisition is stored in the memory area of described newcomer's disk.As it is shown in figure 5, when using
Newcomer's disk is replaced when there is abnormal member's disk 3, then read member's disk from the 4th of member's disk 6 the memory area
The metadata of 3, is written to first memory area of newcomer's disk by the metadata of member's disk 3；From the of member's disk 2
One memory area reads the metadata of member's disk 2, and the metadata of member's disk 2 is written to second of newcomer's disk
Memory area；First memory area from member's disk 1 reads the metadata of member's disk 1, by the metadata of member's disk 1
It is written to the 3rd memory area of newcomer's disk；First memory area from member's disk 8 reads the unit of member's disk 8
Data, are written to the 4th memory area of newcomer's disk by the metadata of member's disk 8.
Conceive based on the application as said method, the embodiment of the present application additionally provides the storage dress of a kind of metadata
Putting, the storage device of this metadata is applied on a storage device.The storage device of this metadata can be realized by software, it is possible to
Realize in the way of by hardware or software and hardware combining.As a example by implemented in software, as the device on a logical meaning, it is
By the processor of the storage device at its place, computer program instructions corresponding in reading non-volatile storage is formed.
For hardware view, as shown in Figure 6, the one of the storage device at the storage device place of the metadata proposed for the application is hard
Part structure chart, in addition to the processor shown in Fig. 6, nonvolatile memory, storage device can also include other hardware, as negative
Duty processes the forwarding chip of message, network interface, internal memory etc.；From the point of view of from hardware configuration, this storage device is it is also possible that be distributed
Formula equipment, potentially includes multiple interface card, in order to carry out the extension of Message processing at hardware view.
As it is shown in fig. 7, the structure chart storing device of a kind of metadata proposed for the application, can apply to include N
In the RAID RAID of individual member's disk, wherein, the redundancy of described RAID is M, and described N is more than described M；Institute
The storage device stating metadata specifically includes:
Maintenance module 11, for safeguarding the mapping relations of member's disk and memory area；Wherein, safeguard member's disk with
During the mapping relations of memory area, for each member's disk, from N number of member's disk, choose L member's disk, and
From each member's disk of L member's disk, choose a memory area, and safeguard that described member's disk is deposited with the L chosen
The mapping relations in storage area territory；L is less than or equal to N more than M, L, and is divided into L in the metadata space of each member's disk
Memory area；
Processing module 12, for when needing the metadata storing member's disk, utilizing described mapping relations, inquire about this one-tenth
L the memory area that member's disk is corresponding, is stored in described metadata in this L memory area.
Described maintenance module 11, specifically for during determining described L, determines that backup quantity L of metadata is institute
Redundancy M stating RAID adds 1；During choosing L member's disk from N number of member's disk, for each member's disk,
From described N number of member's disk, choose described member's disk, and choose L-1 member's magnetic from remaining N-1 member's disk
Dish；During choosing L-1 member's disk from remaining N-1 member's disk, from remaining N-1 member's disk
In, choose L-1 the member disk adjacent with described member's disk；From each member's disk of described L member's disk
During choosing a memory area, from first member's disk to l-th member's disk sequentially, choose successively
One memory area is to l-th memory area；Wherein, L-1 adjacent with described member's disk member's disk refers to: member
In drive bar table space, if located in the member's number of disks after described member's disk not less than L-1, the most described L-1
Member's disk includes being positioned at described member's disk continuous L-1 member's disk below, after described member's disk
Member's number of disks less than L-1, the most described L-1 member's disk includes being positioned at described member's disk all members below
Disk, remaining member's disk is continuous member's disk that the first member's disk from described member's drive bar table space starts；
Described processing module 12, specifically for during being stored in this L memory area by described metadata, when needing
During the metadata of at least two member's disk to be stored, if at L memory area corresponding to described at least two member's disk
At different member's disks, the most concurrently perform to be stored in the metadata of described at least two member's disk L storage of correspondence
The process in region；
Described processing module 12, is additionally operable to, when needing the metadata reading member's disk, utilize described mapping relations, look into
Ask L the memory area that this member's disk is corresponding, and from this L memory area memory area, read metadata；With/
Or, described processing module 12, it is additionally operable to when described N number of member's disk there being member's disk occur abnormal, if described
When RAID increases newcomer's disk, then utilize described mapping relations, from member's disk of no exceptions, obtain generation different
The metadata of record in normal member's disk, and the metadata of acquisition is stored in the memory area of described newcomer's disk.
Wherein, the modules of the application device can be integrated in one, it is also possible to separates and disposes.Above-mentioned module can be closed
And be a module, it is also possible to it is further split into multiple submodule.
Through the above description of the embodiments, those skilled in the art is it can be understood that can be by the application
Software adds the mode of required general hardware platform and realizes, naturally it is also possible to by hardware, but a lot of in the case of the former is more
Good embodiment.Based on such understanding, prior art is contributed by the technical scheme of the application the most in other words
Part can embody with the form of software product, and this computer software product is stored in a storage medium, if including
Dry instruction is with so that a computer equipment (can be personal computer, server, or the network equipment etc.) performs this Shen
Please method described in each embodiment.It will be appreciated by those skilled in the art that accompanying drawing is the schematic diagram of a preferred embodiment,
Module or flow process in accompanying drawing are not necessarily implemented necessary to the application.
It will be appreciated by those skilled in the art that the module in the device in embodiment can describe according to embodiment to carry out point
It is distributed in the device of embodiment, it is also possible to carry out respective change and be disposed other than in one or more devices of the present embodiment.On
The module stating embodiment can merge into a module, it is possible to is further split into multiple submodule.Above-mentioned the embodiment of the present application
Sequence number, just to describing, does not represent the quality of embodiment.
The several specific embodiments being only the application disclosed above, but, the application is not limited to this, any ability
What the technical staff in territory can think change all should fall into the protection domain of the application.
Claims (10)
1. the storage method of a metadata, it is characterised in that be applied to include the RAID of N number of member's disk
In RAID, the redundancy of described RAID is that M, N are more than M；Described method includes:
Safeguard the mapping relations of member's disk and memory area；Wherein, for each member's disk, select from N number of member's disk
Take L member's disk, and from each member's disk of L member's disk, choose a memory area, and safeguard described member
The mapping relations of disk and L the memory area chosen；L is less than or equal to N more than M, L, and in the metadata of each member's disk
Space is divided into L memory area；
When needing the metadata storing member's disk, utilize described mapping relations, inquire about L the storage that this member's disk is corresponding
Region, and described metadata is stored in this L memory area.
Method the most according to claim 1, it is characterised in that the determination mode of described L specifically includes: determine metadata
Backup quantity L is that redundancy M of described RAID adds 1.
Method the most according to claim 1, it is characterised in that described choose L member's disk from N number of member's disk
Process, specifically includes: for each member's disk, chooses described member's disk from N number of member's disk, and from remaining N-1
Individual member's disk is chosen L-1 member's disk.
Method the most according to claim 3, it is characterised in that the described L-1 of choosing from remaining N-1 member's disk
The process of member's disk, specifically includes: from remaining N-1 member's disk, chooses the L-1 adjacent with described member's disk
Member's disk；Wherein, L-1 the member disk adjacent with described member's disk refers to: in member's drive bar table space, if
It is positioned at described member's disk member's number of disks below not less than L-1, described in the most described L-1 member's disk includes being positioned at
Member's disk continuous L-1 member's disk below；If located in the member's number of disks after described member's disk less than L-1
Individual, the most described L-1 member's disk includes being positioned at described member's disk all member's disks below, and remaining member's disk is
The continuous member's disk started from first member's disk of described member's drive bar table space.
Method the most according to claim 3, it is characterised in that described choosing from each member's disk of L member's disk
Take the process of a memory area, specifically include: from first member's disk to l-th member's disk sequentially, successively
Choose first memory area to l-th memory area.
Method the most according to claim 1, it is characterised in that described described metadata is stored in this L memory area
Process, specifically include: need store at least two member's disk metadata time, if described at least two member's disk
L corresponding memory area is in different member's disks, the most concurrently performs the metadata of described at least two member's disk
It is stored in the process of L memory area of correspondence.
Method the most according to claim 1, it is characterised in that described method also includes:
When needing the metadata reading member's disk, utilize described mapping relations, inquire about L the storage that this member's disk is corresponding
Region, and from this L memory area memory area, read metadata.
Method the most according to claim 1, it is characterised in that described method also includes:
When N number of member's disk there being member's disk occur abnormal, if increasing newcomer's disk in described RAID, then utilize
Described mapping relations, obtain from member's disk of no exceptions and the metadata of record in abnormal member's disk occur, and
The metadata of acquisition is stored in the memory area of described newcomer's disk.
9. the storage device of a metadata, it is characterised in that be applied to include the RAID of N number of member's disk
In RAID, the redundancy of described RAID is that M, N are more than M；Described device includes:
Maintenance module, for safeguarding the mapping relations of member's disk and memory area；Wherein, member's disk and memory block are being safeguarded
During the mapping relations in territory, for each member's disk, from N number of member's disk, choose L member's disk, and from L
Each member's disk of member's disk is chosen a memory area, and safeguards described member's disk and L the memory block chosen
The mapping relations in territory；L is less than or equal to N more than M, L, and is divided into L storage in the metadata space of each member's disk
Region；
Processing module, for when needing the metadata storing member's disk, utilizing described mapping relations, inquire about this member's disk
L corresponding memory area, and described metadata is stored in this L memory area.
Device the most according to claim 9, it is characterised in that
Described maintenance module, specifically for during determining described L, determines that backup quantity L of metadata is described RAID
Redundancy M add 1；During choosing L member's disk from N number of member's disk, for each member's disk, from described
N number of member's disk is chosen described member's disk, and chooses L-1 member's disk from remaining N-1 member's disk；From
During L-1 member's disk chosen by remaining N-1 member's disk, from remaining N-1 member's disk, choose with
L-1 member's disk that described member's disk is adjacent；Deposit choosing one from each member's disk of described L member's disk
During storage area territory, from first member's disk to l-th member's disk sequentially, choose first memory block successively
Territory is to l-th memory area；Wherein, L-1 adjacent with described member's disk member's disk refers to: empty in member's disk list
In between, if located in the member's number of disks after described member's disk not less than L-1, the most described L-1 member's disk bag
Include continuous L-1 the member's disk being positioned at described member's disk below, if located in the member's disk after described member's disk
Quantity is less than L-1, and the most described L-1 member's disk includes being positioned at described member's disk all member's disks below, remaining
Member's disk be continuous member's disk that the first member's disk from described member's drive bar table space starts；
Described processing module, specifically for during being stored in this L memory area by described metadata, when needs store
During the metadata of at least two member's disk, if L memory area corresponding to described at least two member's disk is in difference
Member's disk, the most concurrently perform to be stored in the metadata of described at least two member's disk L memory area of correspondence
Process；
Described processing module, is additionally operable to, when needing the metadata reading member's disk, utilize described mapping relations, inquire about this one-tenth
L the memory area that member's disk is corresponding, and from this L memory area memory area, read metadata；And/or, institute
State processing module, be additionally operable to when described N number of member's disk there being member's disk occur abnormal, if increased in described RAID
During newcomer's disk, then utilize described mapping relations, obtain from member's disk of no exceptions and abnormal member's magnetic occurs
The metadata of record in dish, and the metadata of acquisition is stored in the memory area of described newcomer's disk.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610375013.6A CN106095330B (en) | 2016-05-30 | 2016-05-30 | A kind of storage method and device of metadata |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610375013.6A CN106095330B (en) | 2016-05-30 | 2016-05-30 | A kind of storage method and device of metadata |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106095330A true CN106095330A (en) | 2016-11-09 |
| CN106095330B CN106095330B (en) | 2019-04-16 |
Family
ID=57230886
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610375013.6A Active CN106095330B (en) | 2016-05-30 | 2016-05-30 | A kind of storage method and device of metadata |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106095330B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107301106A (en) * | 2017-06-28 | 2017-10-27 | 郑州云海信息技术有限公司 | The restoration methods and device of a kind of RAID system failure |
| CN107315659A (en) * | 2017-06-27 | 2017-11-03 | 郑州云海信息技术有限公司 | The redundancy backup method and device of a kind of metadata |
| CN109002266A (en) * | 2018-08-29 | 2018-12-14 | 郑州云海信息技术有限公司 | A method of promoting metadata reliability in traditional RAID group |
| CN109445694A (en) * | 2018-10-19 | 2019-03-08 | 郑州云海信息技术有限公司 | A kind of distributed memory system separated from meta-data method and apparatus |
| CN112269540A (en) * | 2020-10-16 | 2021-01-26 | 苏州浪潮智能科技有限公司 | Data storage method, system, equipment and computer readable storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080189482A1 (en) * | 2004-12-23 | 2008-08-07 | International Business Machines Corporation | Storage System With Multiple Copy Targeting |
| CN102508620A (en) * | 2011-12-01 | 2012-06-20 | 浪潮电子信息产业股份有限公司 | Method for processing RAID5 (Redundant Array of Independent Disks) bad sector |
| CN103226519A (en) * | 2012-01-31 | 2013-07-31 | Lsi公司 | Elastic cache of redundant cache data |
| CN104268097A (en) * | 2014-10-13 | 2015-01-07 | 浪潮(北京)电子信息产业有限公司 | Metadata processing method and system |
| US20150095695A1 (en) * | 2011-09-12 | 2015-04-02 | Microsoft Corporation | Volatile memory representation of nonvolatile storage device set |
| CN104956309A (en) * | 2013-01-25 | 2015-09-30 | 株式会社日立制作所 | Storage system and data management method |
| CN104991747A (en) * | 2015-07-30 | 2015-10-21 | 湖南亿谷科技发展股份有限公司 | Method and system for data management |
| CN105556520A (en) * | 2013-09-21 | 2016-05-04 | 甲骨文国际公司 | Mirroring, in memory, data from disk to improve query performance |
-
2016
- 2016-05-30 CN CN201610375013.6A patent/CN106095330B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080189482A1 (en) * | 2004-12-23 | 2008-08-07 | International Business Machines Corporation | Storage System With Multiple Copy Targeting |
| US20150095695A1 (en) * | 2011-09-12 | 2015-04-02 | Microsoft Corporation | Volatile memory representation of nonvolatile storage device set |
| CN102508620A (en) * | 2011-12-01 | 2012-06-20 | 浪潮电子信息产业股份有限公司 | Method for processing RAID5 (Redundant Array of Independent Disks) bad sector |
| CN103226519A (en) * | 2012-01-31 | 2013-07-31 | Lsi公司 | Elastic cache of redundant cache data |
| CN104956309A (en) * | 2013-01-25 | 2015-09-30 | 株式会社日立制作所 | Storage system and data management method |
| CN105556520A (en) * | 2013-09-21 | 2016-05-04 | 甲骨文国际公司 | Mirroring, in memory, data from disk to improve query performance |
| CN104268097A (en) * | 2014-10-13 | 2015-01-07 | 浪潮(北京)电子信息产业有限公司 | Metadata processing method and system |
| CN104991747A (en) * | 2015-07-30 | 2015-10-21 | 湖南亿谷科技发展股份有限公司 | Method and system for data management |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107315659A (en) * | 2017-06-27 | 2017-11-03 | 郑州云海信息技术有限公司 | The redundancy backup method and device of a kind of metadata |
| CN107315659B (en) * | 2017-06-27 | 2020-09-18 | 苏州浪潮智能科技有限公司 | Metadata redundancy backup method and device |
| CN107301106A (en) * | 2017-06-28 | 2017-10-27 | 郑州云海信息技术有限公司 | The restoration methods and device of a kind of RAID system failure |
| CN109002266A (en) * | 2018-08-29 | 2018-12-14 | 郑州云海信息技术有限公司 | A method of promoting metadata reliability in traditional RAID group |
| CN109445694A (en) * | 2018-10-19 | 2019-03-08 | 郑州云海信息技术有限公司 | A kind of distributed memory system separated from meta-data method and apparatus |
| CN109445694B (en) * | 2018-10-19 | 2022-02-18 | 郑州云海信息技术有限公司 | Metadata separation method and device for distributed storage system |
| CN112269540A (en) * | 2020-10-16 | 2021-01-26 | 苏州浪潮智能科技有限公司 | Data storage method, system, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106095330B (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101916173B (en) | RAID (Redundant Array of Independent Disks) based data reading and writing method and system thereof | |
| US8307159B2 (en) | System and method for providing performance-enhanced rebuild of a solid-state drive (SSD) in a solid-state drive hard disk drive (SSD HDD) redundant array of inexpensive disks 1 (RAID 1) pair | |
| TWI450087B (en) | Data storage method for a plurality of raid systems and data storage system thereof | |
| CN106095330B (en) | A kind of storage method and device of metadata | |
| US7330932B2 (en) | Disk array with spare logic drive created from space physical drives | |
| US20070088990A1 (en) | System and method for reduction of rebuild time in raid systems through implementation of striped hot spare drives | |
| CN100524237C (en) | Computer hard disk data continuous time point storage and recovery method | |
| CN101436151A (en) | Data real time backup method and system based on file system | |
| JP4441929B2 (en) | Disk device and hot swap method | |
| CN103064765A (en) | Method and device for data recovery and cluster storage system | |
| CN104375784B (en) | A kind of method and apparatus for reducing virtual disk management complexity | |
| CN104035830A (en) | Method and device for recovering data | |
| CN102508733A (en) | Disk array based data processing method and disk array manager | |
| CN104778018A (en) | Broad-strip disk array based on asymmetric hybrid type disk image and storage method of broad-strip disk array | |
| CN102177496A (en) | System and method for transferring data between different RAID data storage types for current data and replay data | |
| JP4261532B2 (en) | Logical disk management method and virtualization apparatus | |
| CN103019623A (en) | Memory disc processing method and device | |
| CN103678025B (en) | A kind of disk failure processing method in disk array | |
| US11093339B2 (en) | Storage utilizing a distributed cache chain and a checkpoint drive in response to a data drive corruption | |
| CN103645997B (en) | A kind of method and system of data protection | |
| US20060259812A1 (en) | Data protection method | |
| CN107239233A (en) | A kind of disk array and data reconstruction method and storage device based on it | |
| JP2006079274A (en) | File management device, network system, file management method and program | |
| US7600151B2 (en) | RAID capacity expansion interruption recovery handling method and system | |
| CN102262657A (en) | Method and system for storing multimedia data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Hangzhou City, Zhejiang Province, Binjiang District West Street 310053 Qianmo Road No. 482 building eleventh layer to the twelfth layer A Applicant after: Hangzhou Sequoia Polytron Technologies Inc Address before: Hangzhou City, Zhejiang Province, Binjiang District West Street 310053 Qianmo Road No. 482 building eleventh layer to the twelfth layer A Applicant before: Hangzhou MacroSAN Technologies Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |