CN104267913A - Storage method and system allowing dynamic asynchronous RAID level adjustment - Google Patents

Storage method and system allowing dynamic asynchronous RAID level adjustment Download PDF

Info

Publication number
CN104267913A
CN104267913A CN201410559584.6A CN201410559584A CN104267913A CN 104267913 A CN104267913 A CN 104267913A CN 201410559584 A CN201410559584 A CN 201410559584A CN 104267913 A CN104267913 A CN 104267913A
Authority
CN
China
Prior art keywords
block
disk
band
data
piece
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
Application number
CN201410559584.6A
Other languages
Chinese (zh)
Other versions
CN104267913B (en
Inventor
张宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beiya Kangcheng (Beijing) Technology Co.,Ltd.
Original Assignee
BEIJING BEIYA SHIDAI TECHNOLOGY Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by BEIJING BEIYA SHIDAI TECHNOLOGY Co Ltd filed Critical BEIJING BEIYA SHIDAI TECHNOLOGY Co Ltd
Priority to CN201410559584.6A priority Critical patent/CN104267913B/en
Publication of CN104267913A publication Critical patent/CN104267913A/en
Application granted granted Critical
Publication of CN104267913B publication Critical patent/CN104267913B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0602Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/0608Saving storage space on storage systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0602Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/062Securing storage systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0629Configuration or reconfiguration of storage systems
    • G06F3/0631Configuration or reconfiguration of storage systems by allocating resources to storage systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0638Organizing or formatting or addressing of data
    • G06F3/0644Management of space entities, e.g. partitions, extents, pools

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)
  • Techniques For Improving Reliability Of Storages (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)

Abstract

The invention provides a storage method and system allowing dynamic asynchronous RAID level adjustment. The method includes first generating an initial virtual disk in an online mode and then increasing or reducing the redundancy through the backstage mode when the IO is free. Due to the fact that the redundancy increasing or reducing operation is conducted at the backstage, and IO resource in the online mode is not occupied, the asynchronous mode is achieved. In addition, the RAID levels of block strips are different, and dynamic adjustment can be conducted, so that the redundancy level adjustment flexibility of the storage system is improved, and the storage requirements of the user can be met. According to the storage method and system, the free IO and the free space are fully utilized under the premise that the data access speed is not affected, the safety level is increased accordingly, and the storage method and system is applicable to large-scale application occasions applied to cloud storage and centralized storage.

Description

The storage means of a kind of dynamic asynchronous adjustment RAID rank and storage system
Technical field
The invention belongs to technical field of data storage, be specifically related to storage means and the storage system of a kind of dynamic asynchronous adjustment RAID rank.
Background technology
RAID (Redundant Arrays of Independent Disks, Redundant Array of Independent Disks (RAID)) technology be widely used in data store various occasions, RAID technique is divided into several different rank, can provide different speed, security and cost performance respectively.Select suitable RAID rank can meet the requirement of user to storage system availability, performance and capacity according to actual conditions.
Conventional RAID rank comprises following several: RAID0, RAID4 and RAID5 etc.
Wherein, RAID 0 adopts banding technique, for the disk array be made up of multiple physical disk, each physical disk is carried out dividing processing, obtains multiple disk block thus; Then, the disk block of different physical disk is formed a band; In data storage procedure, first continuous print data are carried out segmentation by the quantity of the physical disk in disk array, obtain the data block of formed objects, again in units of band, each data block is walked abreast on the different disk block of the different bands be simultaneously written in array, due to parallel read/write can be carried out to data, there is the advantage that data access speed is fast; But because RAID 0 does not have redundant ability, at least need two physical disks, as long as damaging appears in one of them physical disk, then all data that whole disk array can be caused to preserve are all unavailable.
RAID 4 adopts banding technique too, and it is chosen one piece of physical disk and is specifically designed to storage checking data, is called check disk; Data then fragmented storage in other physical disks.On same band, the change of other data all will revise checking data.Therefore, when damaging appears in a certain piece of physical disk, only need damage physical disk to be replaced by new physical disk, and then according to the checking data of check disk, in new physics disk, rebuild the data on damage dish.Utilize independent check disk protected data, improve data redundancy storage capacity, thus improve the security of data storage.
RAID 5 and RAID4 distinguish and are, RAID 5 is not using independent physical disk as check disk, but verification data interaction is stored on each physical disk.Therefore, IO performance is more balanced, and handling capacity and disk life-span all can best configuration.Something in common is, any one physical disk damages, and all do not affect valid data, the data of disappearance generate out by other data blocks on same band.
RAID 4, RAID 5 etc. are had to the storage system of check bit, main employing stores data and checking data stores synchronized structure, that is: for a band, only have and checking data is write, just can think that this band is write, the advantage of this kind of storage organization is: when the disasters such as sudden power-off occur, can ensure the stability of data, Controller gain variations is comparatively simple simultaneously.But along with cloud stores and the large-scale application of centralize storage; the storage organization be made up of some groups of single RAID usually can cause large-area applications to be paralysed because of one group of RAID failure; be exaggerated data risk; at this moment; increase the quantity of redundancy magnetic disk simply; or performance sharply declines, or space waste is surprising.So needing to find Best Point in space availability ratio and access performance, how under the prerequisite not affecting data access speed, make good use of idle IO and the free space of storage, thus promote level of security, is that storage products needs the key issue that solves instantly.
Summary of the invention
For the defect that prior art exists, the invention provides storage means and the storage system of a kind of dynamic asynchronous adjustment RAID rank, can effectively solve the problem.
The technical solution used in the present invention is as follows:
The present invention also provides a kind of dynamic asynchronous adjustment RAID storage means of rank, comprises the following steps:
S1, for the disk array be made up of multiple physical disk, carries out dividing processing by each physical disk, obtains multiple disk block thus;
Be numbered disk block described in each, this numbering is constituted jointly by physical disk PN and disk block BN, is designated as PN:BN;
Then, definition storage pool, described storage pool has disk block allocation table and block band redundance label table; Wherein, described piece of band redundance label table is for recording the redundance of each piece of band;
By all number after disk block stored in defined described storage pool;
S2, in described storage pool, define the virtual disk be made up of the capable m row of a n disk block, all disk blocks of every a line form a disk block band, referred to as block band; And arrange described piece of band and meet following 2 conditions:
In condition 1, each piece of band, the disk block with same physical disk number PN only allows to occur once;
The disk block number needs that condition 2, each piece of band have meets the minimum disk block quantity set when creating described virtual disk;
In addition, when creating described virtual disk, the data redundancy rank of virtual disk described in definition or default definition, this data redundancy rank is determined by following 4 parameters:
The disk block quantity of the storage authentic data in a, each piece of band: m-h; Wherein, the disk block of authentic data will be stored referred to as data block;
The disk block quantity of the storage proof test value in b, each piece of band: h; Wherein, the disk block of proof test value will be stored referred to as check block;
The check block quantity strengthened is allowed: M in c, each piece of band;
D, virtual disk level of security: Vs;
S3, if by sequentially having n original block band from top to bottom, be designated as respectively: original block band 1, original block band 2 ... original block band n;
If any one original block band i forms by m disk block, by order from left to right, be designated as respectively: disk block i-1, disk block i-2.., disk block i-m; Further, m disk block of described original block band is online disk block; In m the disk block that original block band described in each has, configuration h is for storing the check block of proof test value, and other m-h disk block is the data block for storing authentic data; Wherein, 0≤h < m, h are natural number;
S4, when carrying out write operation, RAID controller receives the data of virtual disk to be written, according to the offset address of data in virtual disk and the amount of capacity of data, defined by the structure of virtual disk, by described data sectional, then calculate the block address of data block and the block address of check block of the write of the needs of the data after segmentation; Then, the data after the data block address write segmentation obtained after calculating, the check block address write checking data obtained after calculating, returns and is write as merit;
When carrying out read operation, after the space address of described virtual disk and size conversion, being undertaken reading, gathering by parallel thread, getting final product sense data;
S4, specifically comprises:
S4.1, the idle condition of RAID controller monitoring I/O module, when monitoring I/O module and being idle, performs S4.2;
S4.2, described virtual disk increases the operation of block band redundance:
If met the following conditions simultaneously:
The free disk number of blocks of current described storage pool exceedes setting value; Further, there is first piece of band collection that check block quantity is less than maximum quantity M;
Then: described virtual disk is based on priority redundancy strategy, each block band according to priority concentrated described first piece of band respectively in descending order increases the redundancy process of a rank, that is: for obtained block band increases a check block, and based on default checking algorithm, in increased check block, write checking data; Upgrade the block band redundance in described piece of band redundance label table; Repeat this step, constantly redundance is increased to each block band, change the RAID rank of each block band;
Described virtual disk reduces the operation of block band redundance:
If the free disk number of blocks of current described storage pool is lower than setting value, then according to priority respectively a check block is reclaimed to each block band from low to high, reduce the redundancy rank of block band; Upgrade the block band redundance mark in described piece of band redundance label table; Then, identifying at the described disk block allocation table of storage pool the described check block reclaimed is free state, makes it can be used as data block and uses.
Preferably, in S2, h is integer, and h>=0 and h<=5.
Preferably, in S2, M is integer and 0<=M<=5.
Preferably, described priority redundancy strategy comprises:
Based on predefined virtual disk level of security Vs, virtual disk level of security is higher, then show that the importance degree of virtual disk is higher, then its priority level is higher;
The importance degree of the concrete data that each block band stores, that is: the concrete data stored are more important, and its priority level is higher;
Each block band finally performs the time of storage operation, that is: finally perform the time storing operation less from the time interval of current time, its priority level is higher.
Preferably, in S4.2, described default checking algorithm is the multi head linear equation group principle based on Reed-solomon algorithm design, that is: when this solution of equations is unique, can increase multistage verification.
The present invention also provides a kind of dynamic asynchronous adjustment RAID storage system of rank, comprising:
Segmentation module, for the disk array be made up of multiple physical disk, carries out dividing processing by each physical disk, obtains multiple disk block thus;
Numbering module, for being numbered disk block described in each, this numbering is constituted jointly by physical disk PN and disk block BN, is designated as PN:BN;
Storage pool definition module, for defining storage pool, described storage pool has disk block allocation table and block band redundance label table; Wherein, described piece of band redundance label table is for recording the redundance of each piece of band;
Virtual disk definition module, for defining the virtual disk be made up of the capable m row of a n disk block in described storage pool, all disk blocks of every a line form a disk block band, referred to as block band; And arrange described piece of band and meet following 2 conditions:
In condition 1, each piece of band, the disk block with same physical disk number PN only allows to occur once;
The disk block number needs that condition 2, each piece of band have meets the minimum disk block quantity set when creating described virtual disk;
The also data redundancy rank of virtual disk described in definition or default definition, this data redundancy rank is determined by following 4 parameters:
The disk block quantity of the storage authentic data in a, each piece of band: m-h; Wherein, the disk block of authentic data will be stored referred to as data block;
The disk block quantity of the storage proof test value in b, each piece of band: h; Wherein, the disk block of proof test value will be stored referred to as check block;
The check block quantity strengthened is allowed: M in c, each piece of band;
D, virtual disk level of security: Vs;
Data block quantity initial value arranges module with check block quantity initial value, forms by m disk block for any one original block band i, by order from left to right, is designated as respectively: disk block i-1, disk block i-2.., disk block i-m; Further, m disk block of described original block band is online disk block; In m the disk block that original block band described in each has, configuration h is for storing the check block of proof test value, and other m-h disk block is the data block for storing authentic data; Wherein, 0≤h < m, h are natural number;
IO monitoring module, for monitoring the idle condition of I/O module;
Block band redundance increases module, during for monitoring the I/O module free time when described IO monitoring module, and meets the following conditions: the free disk number of blocks of current described storage pool exceedes setting value; Further, there is first piece of band collection that check block quantity is less than maximum quantity M;
Then: based on priority redundancy strategy, each block band according to priority concentrated described first piece of band respectively in descending order increases the redundancy process of a rank, that is: for obtained block band increases a check block, and based on default checking algorithm, in increased check block, write checking data; Upgrade the block band redundance in described piece of band redundance label table; Repeat this step, constantly redundance is increased to each block band, change the RAID rank of each block band;
Block band redundance reduces module, during for monitoring the I/O module free time when described IO monitoring module, and the free disk number of blocks of current described storage pool is lower than setting value, then according to priority respectively a check block is reclaimed to each block band from low to high, reduce the redundancy rank of block band; Upgrade the block band redundance mark in described piece of band redundance label table; Then, identifying at the described disk block allocation table of storage pool the described check block reclaimed is free state, makes it can be used as data block and uses.
Beneficial effect of the present invention is as follows:
The storage means of dynamic provided by the invention asynchronous adjustment RAID rank and storage system, adopt background mode to increase or reduce the redundance of each block band, under having the prerequisite not affecting online data storage speed, also utilize idle IO and free space, thus lifting level of security, be more suitable for the large-scale application occasion being applied to cloud storage and centralize storage.In addition, in the present invention, the RAID rank of each piece of band can carry out dynamic conditioning, and the RAID rank of each block band can not be identical, improves the dirigibility of storage system redundancy rank adjusting, more can meet user's storage demand.
Accompanying drawing explanation
Fig. 1 is the principle schematic of the storage means of dynamic provided by the invention asynchronous adjustment RAID rank;
Fig. 2 is the structural representation of the storage system of dynamic provided by the invention asynchronous adjustment RAID rank.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in detail:
The invention provides the storage means of a kind of dynamic asynchronous adjustment RAID rank, as shown in Figure 1, be the principle schematic for illustration of storage means of the present invention, comprise the following steps:
S1, for the disk array be made up of multiple physical disk, carries out dividing processing by each physical disk, obtains multiple disk block thus;
Be numbered disk block described in each, this numbering is constituted jointly by physical disk PN and disk block BN, is designated as PN:BN;
With reference to figure 1, have 6 physical disks, each physical disk is all split into 4 disk blocks, therefore, for the 1st physical disk, physical disk PN is PD_A, disk block BN is respectively 0,1,2 and 3, and therefore, disk block numbering is respectively: PD_A0, PD_A1, PD_A2 and PD_A3; Equally, for the 2nd physical disk, physical disk PN is PD_B, and disk block BN is respectively 0,1,2 and 3, and therefore, disk block numbering is respectively: PD_B0, PD_B1, PD_B2 and PD_B3; For 3-6 physical disk, physical disk PN is respectively PD_C, PD_D, PD_E, PD_F, and each disk block BN is 0,1,2 and 3, and concrete disk block numbering repeats no more.
Then, definition storage pool, described storage pool has disk block allocation table and block band redundance label table; Wherein, described piece of band redundance label table is for recording the redundance of each piece of band;
By all number after disk block stored in defined described storage pool;
S2, in described storage pool, define the virtual disk be made up of the capable m row of a n disk block, all disk blocks of every a line form a disk block band, referred to as block band; And arrange described piece of band and meet following 2 conditions:
In condition 1, each piece of band, the disk block with same physical disk number PN only allows to occur once;
That is, in the set of disk blocks numbering (PN:BN) all in each piece of band, different PN only allows to occur once, does not namely allow the different disk block of identical physical disk to occur in each piece of band.Such as, as Fig. 1, in block band 1, have 5 disk blocks, its numbering is respectively: PD_A2, PD_B0, PD_C3, PD_D2 and PD_E3, comes from 5 different physical disks respectively; Again such as, for these 4 disk blocks of PD_A0, PD_A1, PD_A2 and PD_A3, all come from physical disk PD_A, therefore, in these 4 disk blocks, when any one disk block appears at band 1, other 3 disk blocks all do not allow to appear in band 1.
The disk block number needs that condition 2, each piece of band have meets the minimum disk block quantity set when creating described virtual disk;
In addition, when creating described virtual disk, the data redundancy rank of virtual disk described in definition or default definition, this data redundancy rank is determined by following 4 parameters:
The disk block quantity of the storage authentic data in a, each piece of band: m-h; Wherein, the disk block of authentic data will be stored referred to as data block;
The disk block quantity of the storage proof test value in b, each piece of band: h; Wherein, the disk block of proof test value will be stored referred to as check block;
The check block quantity strengthened is allowed: M in c, each piece of band;
D, virtual disk level of security: Vs;
In practical application, h is integer, and h>=0 and h<=5, i.e. h=0,1,2,3,4 or 5.RAID rank of the prior art, the maximal value of the check block quantity that each piece of band comprises is 3, in the present invention, and the maximal value of the check block quantity that each piece of band comprises is 5, increases a redundance than prior art.M is integer and 0<=M<=5, i.e. M=0,1,2,3,4 or 5, that is, according to actual needs, each piece of band both can not comprise check block quantity, was RAID0 rank, also can support at most 5 check blocks, that is, the redundance variation range of each piece of band is very large.
S3, if by sequentially having n original block band from top to bottom, be designated as respectively: original block band 1, original block band 2 ... original block band n;
If any one original block band i forms by m disk block, by order from left to right, be designated as respectively: disk block i-1, disk block i-2.., disk block i-m; Further, m disk block of described original block band is online disk block; In m the disk block that original block band described in each has, configuration h is for storing the check block of proof test value, and other m-h disk block is the data block for storing authentic data; Wherein, 0≤h < m, h are natural number;
With reference to figure 1, during initialization, symbiosis becomes 4 original block bands, is respectively original block band 1, original block band 2, original block band 3 and original block band 4; Further, when initial, the data block quantity that each original block band comprises is all equal, and such as, each original block band includes 4 data blocks; Then, each original block band all can not comprise check block, is 4+0 online data pattern; Also can include 1 check block, be 4+1 online data pattern, analogize, also can be designed as 4+2 online data pattern or 4+3 online data pattern.Time initial, the disk block that each original block band comprises is online disk block, in addition, the redundance (the check block quantity namely comprised) of each original block band is arranged according to the actual requirements, set check block quantity is fewer, and redundance is lower, but space availability ratio is higher; On the contrary, set check block quantity is more, and redundance is more, but space availability ratio is lower.Therefore, according to demand, between space availability ratio and redundance, the best quantity that check block is set is found.
S4, when carrying out write operation, RAID controller receives the data of virtual disk to be written, according to the offset address of data in virtual disk and the amount of capacity of data, defined by the structure of virtual disk, by described data sectional, then calculate the block address of data block and the block address of check block of the write of the needs of the data after segmentation; Then, the data after the data block address write segmentation obtained after calculating, the check block address write checking data obtained after calculating, returns and is write as merit;
When carrying out read operation, after the space address of described virtual disk and size conversion, being undertaken reading, gathering by parallel thread, getting final product sense data;
S4, specifically comprises:
S4.1, the idle condition of RAID controller monitoring I/O module, when monitoring I/O module and being idle, performs S4.2;
S4.2, described virtual disk increases the operation of block band redundance:
If met the following conditions simultaneously:
The free disk number of blocks of current described storage pool exceedes setting value; Further, there is first piece of band collection that check block quantity is less than maximum quantity M;
Then: described virtual disk is based on priority redundancy strategy, each block band according to priority concentrated described first piece of band respectively in descending order increases the redundancy process of a rank, that is: for obtained block band increases a check block, and based on default checking algorithm, in increased check block, write checking data; Upgrade the block band redundance in described piece of band redundance label table; Repeat this step, constantly redundance is increased to each block band, change the RAID rank of each block band; By this step, because according to priority sequencing increases the redundance of each block band, therefore, at a time, the redundance of each block band is not identical, realizes thus dynamically changing each other effect of block band redundant level.
Wherein, priority redundancy strategy includes but not limited to following several strategy: (1), based on predefined virtual disk level of security Vs, virtual disk level of security is higher, then show that the importance degree of virtual disk is higher, then its priority level is higher; (2) importance degree of concrete data that stores of each block band, that is: the concrete data stored are more important, and its priority level is higher; (3) each block band finally performs the time of storage operation, that is: finally perform the time storing operation less from the time interval of current time, its priority level is higher.
Wherein, default checking algorithm can be the multi head linear equation group principle based on Reed-solomon algorithm design, that is: when this solution of equations is unique, can increase multistage verification.
Described virtual disk reduces the operation of block band redundance:
If the free disk number of blocks of current described storage pool is lower than setting value, then according to priority respectively a check block is reclaimed to each block band from low to high, reduce the redundancy rank of block band; Upgrade the block band redundance mark in described piece of band redundance label table; Then, identifying at the described disk block allocation table of storage pool the described check block reclaimed is free state, makes it can be used as data block and uses.
By this step, because according to priority sequencing reduces the redundance of each block band, therefore, at a time, the redundance of each block band is not identical, realizes thus dynamically changing each other effect of block band redundant level.
Such as, when initial, generate 4+1 online data pattern, that is, initial each block band generated is RAID1 rank, comprises 4 data blocks and 1 check block, carries out real-time online protection, ensure that data reading speed by low-level RAID; Then, when the idle and available free space of IO, 1 redundance is dynamically increased to each piece of band, namely increases by 1 check block, thus increase the level of security of storage system; If storage space allows, or else disconnected increase check block, often increase a check block, namely adds a level of security of storage system, and the redundancy rank that different band is supported is different; When IO is idle and free space is inadequate, dynamically redundance is reduced to each piece of band, thus release free space.
As can be seen here, in the present invention, first initial virtual disk is generated by line model, then, when IO is idle, adopts background mode to increase or reduce redundance, because the operation increasing or reduce redundance is carried out on backstage, not needing IO resource when taking line model, therefore, is a kind of asynchronous mode; In addition, in the present invention, the RAID rank of each piece of band can be all not identical, and can dynamic conditioning be carried out, allow user according to actual needs, preferentially certain specific block band is increased redundance, thus improve the dirigibility of storage system redundancy rank adjusting, more can meet user's storage demand.Therefore, the storage means of dynamic provided by the invention asynchronous adjustment RAID rank and storage system, under the prerequisite not affecting data access speed, make full use of idle IO and free space, thus lifting level of security, be more suitable for the large-scale application occasion being applied to cloud storage and centralize storage.
As shown in Figure 2, the present invention also provides a kind of dynamic asynchronous adjustment RAID storage system of rank, comprising:
Segmentation module, for the disk array be made up of multiple physical disk, carries out dividing processing by each physical disk, obtains multiple disk block thus;
Numbering module, for being numbered disk block described in each, this numbering is constituted jointly by physical disk PN and disk block BN, is designated as PN:BN;
Storage pool definition module, for defining storage pool, described storage pool has disk block allocation table and block band redundance label table; Wherein, described piece of band redundance label table is for recording the redundance of each piece of band;
Virtual disk definition module, for defining the virtual disk be made up of the capable m row of a n disk block in described storage pool, all disk blocks of every a line form a disk block band, referred to as block band; And arrange described piece of band and meet following 2 conditions:
In condition 1, each piece of band, the disk block with same physical disk number PN only allows to occur once;
The disk block number needs that condition 2, each piece of band have meets the minimum disk block quantity set when creating described virtual disk;
The also data redundancy rank of virtual disk described in definition or default definition, this data redundancy rank is determined by following 4 parameters:
The disk block quantity of the storage authentic data in a, each piece of band: m-h; Wherein, the disk block of authentic data will be stored referred to as data block;
The disk block quantity of the storage proof test value in b, each piece of band: h; Wherein, the disk block of proof test value will be stored referred to as check block;
The check block quantity strengthened is allowed: M in c, each piece of band;
D, virtual disk level of security: Vs;
Data block quantity initial value arranges module with check block quantity initial value, forms by m disk block for any one original block band i, by order from left to right, is designated as respectively: disk block i-1, disk block i-2.., disk block i-m; Further, m disk block of described original block band is online disk block; In m the disk block that original block band described in each has, configuration h is for storing the check block of proof test value, and other m-h disk block is the data block for storing authentic data; Wherein, 0≤h < m, h are natural number;
IO monitoring module, for monitoring the idle condition of I/O module;
Block band redundance increases module, during for monitoring the I/O module free time when described IO monitoring module, and meets the following conditions: the free disk number of blocks of current described storage pool exceedes setting value; Further, there is first piece of band collection that check block quantity is less than maximum quantity M;
Then: based on priority redundancy strategy, each block band according to priority concentrated described first piece of band respectively in descending order increases the redundancy process of a rank, that is: for obtained block band increases a check block, and based on default checking algorithm, in increased check block, write checking data; Upgrade the block band redundance in described piece of band redundance label table; Repeat this step, constantly redundance is increased to each block band, change the RAID rank of each block band;
Block band redundance reduces module, during for monitoring the I/O module free time when described IO monitoring module, and the free disk number of blocks of current described storage pool is lower than setting value, then according to priority respectively a check block is reclaimed to each block band from low to high, reduce the redundancy rank of block band; Upgrade the block band redundance mark in described piece of band redundance label table; Then, identifying at the described disk block allocation table of storage pool the described check block reclaimed is free state, makes it can be used as data block and uses.
In sum, the storage means of dynamic provided by the invention asynchronous adjustment RAID rank and storage system, space availability ratio and access performance find Best Point, under the prerequisite not affecting data access speed, make full use of the idle IO and free space that store, thus promote level of security.
The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should look protection scope of the present invention.

Claims (6)

1. a storage means for dynamic asynchronous adjustment RAID rank, is characterized in that, comprise the following steps:
S1, for the disk array be made up of multiple physical disk, carries out dividing processing by each physical disk, obtains multiple disk block thus;
Be numbered disk block described in each, this numbering is constituted jointly by physical disk PN and disk block BN, is designated as PN:BN;
Then, definition storage pool, described storage pool has disk block allocation table and block band redundance label table; Wherein, described piece of band redundance label table is for recording the redundance of each piece of band;
By all number after disk block stored in defined described storage pool;
S2, in described storage pool, define the virtual disk be made up of the capable m row of a n disk block, all disk blocks of every a line form a disk block band, referred to as block band; And arrange described piece of band and meet following 2 conditions:
In condition 1, each piece of band, the disk block with same physical disk number PN only allows to occur once;
The disk block number needs that condition 2, each piece of band have meets the minimum disk block quantity set when creating described virtual disk;
In addition, when creating described virtual disk, the data redundancy rank of virtual disk described in definition or default definition, this data redundancy rank is determined by following 4 parameters:
The disk block quantity of the storage authentic data in a, each piece of band: m-h; Wherein, the disk block of authentic data will be stored referred to as data block;
The disk block quantity of the storage proof test value in b, each piece of band: h; Wherein, the disk block of proof test value will be stored referred to as check block;
The check block quantity strengthened is allowed: M in c, each piece of band;
D, virtual disk level of security: Vs;
S3, if by sequentially having n original block band from top to bottom, be designated as respectively: original block band 1, original block band 2 ... original block band n;
If any one original block band i forms by m disk block, by order from left to right, be designated as respectively: disk block i-1, disk block i-2.., disk block i-m; Further, m disk block of described original block band is online disk block; In m the disk block that original block band described in each has, configuration h is for storing the check block of proof test value, and other m-h disk block is the data block for storing authentic data; Wherein, 0≤h < m, h are natural number;
S4, when carrying out write operation, RAID controller receives the data of virtual disk to be written, according to the offset address of data in virtual disk and the amount of capacity of data, defined by the structure of virtual disk, by described data sectional, then calculate the block address of data block and the block address of check block of the write of the needs of the data after segmentation; Then, the data after the data block address write segmentation obtained after calculating, the check block address write checking data obtained after calculating, returns and is write as merit;
When carrying out read operation, after the space address of described virtual disk and size conversion, being undertaken reading, gathering by parallel thread, getting final product sense data;
S4, specifically comprises:
S4.1, the idle condition of RAID controller monitoring I/O module, when monitoring I/O module and being idle, performs S4.2;
S4.2, described virtual disk increases the operation of block band redundance:
If met the following conditions simultaneously:
The free disk number of blocks of current described storage pool exceedes setting value; Further, there is first piece of band collection that check block quantity is less than maximum quantity M;
Then: described virtual disk is based on priority redundancy strategy, each block band according to priority concentrated described first piece of band respectively in descending order increases the redundancy process of a rank, that is: for obtained block band increases a check block, and based on default checking algorithm, in increased check block, write checking data; Upgrade the block band redundance in described piece of band redundance label table; Repeat this step, constantly redundance is increased to each block band, change the RAID rank of each block band;
Described virtual disk reduces the operation of block band redundance:
If the free disk number of blocks of current described storage pool is lower than setting value, then according to priority respectively a check block is reclaimed to each block band from low to high, reduce the redundancy rank of block band; Upgrade the block band redundance mark in described piece of band redundance label table; Then, identifying at the described disk block allocation table of storage pool the described check block reclaimed is free state, makes it can be used as data block and uses.
2. the storage means of dynamic according to claim 1 asynchronous adjustment RAID rank, it is characterized in that, in S2, h is integer, and h>=0 and h<=5.
3. the storage means of dynamic according to claim 1 asynchronous adjustment RAID rank, it is characterized in that, in S2, M is integer and 0<=M<=5.
4. the storage means of dynamic according to claim 1 asynchronous adjustment RAID rank, it is characterized in that, described priority redundancy strategy comprises:
Based on predefined virtual disk level of security Vs, virtual disk level of security is higher, then show that the importance degree of virtual disk is higher, then its priority level is higher;
The importance degree of the concrete data that each block band stores, that is: the concrete data stored are more important, and its priority level is higher;
Each block band finally performs the time of storage operation, that is: finally perform the time storing operation less from the time interval of current time, its priority level is higher.
5. the storage means of dynamic according to claim 1 asynchronous adjustment RAID rank, it is characterized in that, in S4.2, described default checking algorithm is the multi head linear equation group principle based on Reed-solomon algorithm design, that is: when this solution of equations is unique, multistage verification can be increased.
6. a storage system for dynamic asynchronous adjustment RAID rank, is characterized in that, comprising:
Segmentation module, for the disk array be made up of multiple physical disk, carries out dividing processing by each physical disk, obtains multiple disk block thus;
Numbering module, for being numbered disk block described in each, this numbering is constituted jointly by physical disk PN and disk block BN, is designated as PN:BN;
Storage pool definition module, for defining storage pool, described storage pool has disk block allocation table and block band redundance label table; Wherein, described piece of band redundance label table is for recording the redundance of each piece of band;
Virtual disk definition module, for defining the virtual disk be made up of the capable m row of a n disk block in described storage pool, all disk blocks of every a line form a disk block band, referred to as block band; And arrange described piece of band and meet following 2 conditions:
In condition 1, each piece of band, the disk block with same physical disk number PN only allows to occur once;
The disk block number needs that condition 2, each piece of band have meets the minimum disk block quantity set when creating described virtual disk;
The also data redundancy rank of virtual disk described in definition or default definition, this data redundancy rank is determined by following 4 parameters:
The disk block quantity of the storage authentic data in a, each piece of band: m-h; Wherein, the disk block of authentic data will be stored referred to as data block;
The disk block quantity of the storage proof test value in b, each piece of band: h; Wherein, the disk block of proof test value will be stored referred to as check block;
The check block quantity strengthened is allowed: M in c, each piece of band;
D, virtual disk level of security: Vs;
Data block quantity initial value arranges module with check block quantity initial value, forms by m disk block for any one original block band i, by order from left to right, is designated as respectively: disk block i-1, disk block i-2.., disk block i-m; Further, m disk block of described original block band is online disk block; In m the disk block that original block band described in each has, configuration h is for storing the check block of proof test value, and other m-h disk block is the data block for storing authentic data; Wherein, 0≤h < m, h are natural number;
IO monitoring module, for monitoring the idle condition of I/O module;
Block band redundance increases module, during for monitoring the I/O module free time when described IO monitoring module, and meets the following conditions: the free disk number of blocks of current described storage pool exceedes setting value; Further, there is first piece of band collection that check block quantity is less than maximum quantity M;
Then: based on priority redundancy strategy, each block band according to priority concentrated described first piece of band respectively in descending order increases the redundancy process of a rank, that is: for obtained block band increases a check block, and based on default checking algorithm, in increased check block, write checking data; Upgrade the block band redundance in described piece of band redundance label table; Repeat this step, constantly redundance is increased to each block band, change the RAID rank of each block band;
Block band redundance reduces module, during for monitoring the I/O module free time when described IO monitoring module, and the free disk number of blocks of current described storage pool is lower than setting value, then according to priority respectively a check block is reclaimed to each block band from low to high, reduce the redundancy rank of block band; Upgrade the block band redundance mark in described piece of band redundance label table; Then, identifying at the described disk block allocation table of storage pool the described check block reclaimed is free state, makes it can be used as data block and uses.
CN201410559584.6A 2014-10-20 2014-10-20 It is a kind of can dynamic asynchronous adjust RAID storage method and storage system Active CN104267913B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410559584.6A CN104267913B (en) 2014-10-20 2014-10-20 It is a kind of can dynamic asynchronous adjust RAID storage method and storage system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410559584.6A CN104267913B (en) 2014-10-20 2014-10-20 It is a kind of can dynamic asynchronous adjust RAID storage method and storage system

Publications (2)

Publication Number Publication Date
CN104267913A true CN104267913A (en) 2015-01-07
CN104267913B CN104267913B (en) 2017-06-16

Family

ID=52159439

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410559584.6A Active CN104267913B (en) 2014-10-20 2014-10-20 It is a kind of can dynamic asynchronous adjust RAID storage method and storage system

Country Status (1)

Country Link
CN (1) CN104267913B (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9697079B2 (en) 2015-07-13 2017-07-04 International Business Machines Corporation Protecting data integrity in de-duplicated storage environments in combination with software defined native raid
WO2017157037A1 (en) * 2016-03-17 2017-09-21 华为技术有限公司 Method and device for data striping
CN107197324A (en) * 2017-05-15 2017-09-22 郑州云海信息技术有限公司 A kind of video error tolerance method and device of distributed file system
US9846538B2 (en) 2015-12-07 2017-12-19 International Business Machines Corporation Data integrity and acceleration in compressed storage environments in combination with software defined native RAID
CN108255429A (en) * 2018-01-10 2018-07-06 郑州云海信息技术有限公司 A kind of write operation control method, system, device and computer readable storage medium
CN110059062A (en) * 2019-04-22 2019-07-26 苏州浪潮智能科技有限公司 A kind of band setting method, system and the associated component of file system
CN110413203A (en) * 2018-04-28 2019-11-05 伊姆西Ip控股有限责任公司 For managing the method, equipment and computer program product of storage system
WO2020019267A1 (en) * 2018-07-26 2020-01-30 华为技术有限公司 Data processing method and device
CN111522501A (en) * 2019-02-01 2020-08-11 杭州海康威视数字技术股份有限公司 Disk array space dividing method and device, electronic equipment and storage medium
CN114415968A (en) * 2022-01-26 2022-04-29 南京晓庄学院 Storage system and data writing method thereof
WO2023184911A1 (en) * 2022-03-30 2023-10-05 苏州浪潮智能科技有限公司 Data storage method and apparatus, and device and non-volatile readable storage medium
CN117112455A (en) * 2023-10-24 2023-11-24 苏州元脑智能科技有限公司 Space address conversion method and device, computer equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5857112A (en) * 1992-09-09 1999-01-05 Hashemi; Ebrahim System for achieving enhanced performance and data availability in a unified redundant array of disk drives by using user defined partitioning and level of redundancy
US20070143541A1 (en) * 2005-12-19 2007-06-21 Lsi Logic Corporation Methods and structure for improved migration of raid logical volumes
CN101587425A (en) * 2009-06-16 2009-11-25 杭州华三通信技术有限公司 A kind of method and device that increases magnetic disc redundant array redundancy
CN102110154A (en) * 2011-02-21 2011-06-29 中国科学院计算技术研究所 File redundancy storage method in cluster file system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5857112A (en) * 1992-09-09 1999-01-05 Hashemi; Ebrahim System for achieving enhanced performance and data availability in a unified redundant array of disk drives by using user defined partitioning and level of redundancy
US20070143541A1 (en) * 2005-12-19 2007-06-21 Lsi Logic Corporation Methods and structure for improved migration of raid logical volumes
CN101587425A (en) * 2009-06-16 2009-11-25 杭州华三通信技术有限公司 A kind of method and device that increases magnetic disc redundant array redundancy
CN102110154A (en) * 2011-02-21 2011-06-29 中国科学院计算技术研究所 File redundancy storage method in cluster file system

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9697079B2 (en) 2015-07-13 2017-07-04 International Business Machines Corporation Protecting data integrity in de-duplicated storage environments in combination with software defined native raid
US9846538B2 (en) 2015-12-07 2017-12-19 International Business Machines Corporation Data integrity and acceleration in compressed storage environments in combination with software defined native RAID
US10152235B2 (en) 2015-12-07 2018-12-11 International Business Machines Corporation Data integrity and acceleration in compressed storage environments in combination with software defined native RAID
US10572157B2 (en) 2015-12-07 2020-02-25 International Business Machines Corporation Data integrity and acceleration in compressed storage environments in combination with software defined native RAID
CN107203559B (en) * 2016-03-17 2021-01-01 华为技术有限公司 Method and device for dividing data strips
WO2017157037A1 (en) * 2016-03-17 2017-09-21 华为技术有限公司 Method and device for data striping
CN107203559A (en) * 2016-03-17 2017-09-26 华为技术有限公司 A kind of method and apparatus for dividing data strip
CN107197324A (en) * 2017-05-15 2017-09-22 郑州云海信息技术有限公司 A kind of video error tolerance method and device of distributed file system
CN108255429A (en) * 2018-01-10 2018-07-06 郑州云海信息技术有限公司 A kind of write operation control method, system, device and computer readable storage medium
CN110413203B (en) * 2018-04-28 2023-07-18 伊姆西Ip控股有限责任公司 Method, apparatus and computer readable medium for managing a storage system
CN110413203A (en) * 2018-04-28 2019-11-05 伊姆西Ip控股有限责任公司 For managing the method, equipment and computer program product of storage system
US11593000B2 (en) 2018-07-26 2023-02-28 Huawei Technologies Co., Ltd. Data processing method and apparatus
CN112513804A (en) * 2018-07-26 2021-03-16 华为技术有限公司 Data processing method and device
CN112513804B (en) * 2018-07-26 2022-08-19 华为技术有限公司 Data processing method and device
WO2020019267A1 (en) * 2018-07-26 2020-01-30 华为技术有限公司 Data processing method and device
CN111522501A (en) * 2019-02-01 2020-08-11 杭州海康威视数字技术股份有限公司 Disk array space dividing method and device, electronic equipment and storage medium
CN111522501B (en) * 2019-02-01 2022-01-21 杭州海康威视数字技术股份有限公司 Disk array space dividing method and device, electronic equipment and storage medium
CN110059062B (en) * 2019-04-22 2020-05-29 苏州浪潮智能科技有限公司 Stripe setting method and system of file system and related components
CN110059062A (en) * 2019-04-22 2019-07-26 苏州浪潮智能科技有限公司 A kind of band setting method, system and the associated component of file system
CN114415968A (en) * 2022-01-26 2022-04-29 南京晓庄学院 Storage system and data writing method thereof
CN114415968B (en) * 2022-01-26 2023-04-07 南京晓庄学院 Storage system and data writing method thereof
WO2023184911A1 (en) * 2022-03-30 2023-10-05 苏州浪潮智能科技有限公司 Data storage method and apparatus, and device and non-volatile readable storage medium
CN117112455A (en) * 2023-10-24 2023-11-24 苏州元脑智能科技有限公司 Space address conversion method and device, computer equipment and storage medium
CN117112455B (en) * 2023-10-24 2024-02-06 苏州元脑智能科技有限公司 Space address conversion method and device, computer equipment and storage medium

Also Published As

Publication number Publication date
CN104267913B (en) 2017-06-16

Similar Documents

Publication Publication Date Title
CN104267913A (en) Storage method and system allowing dynamic asynchronous RAID level adjustment
CN102880428B (en) The creation method of distributed Redundant Array of Independent Disks (RAID) and device
CN103793182B (en) System and method for expansible storage protection
CN101504623B (en) Independent disk redundancy array construction method and device
CN101587425B (en) Method and device for increasing magnetic disc redundant array redundancy
CN103049222A (en) RAID5 (redundant array of independent disk 5) write IO optimization processing method
US20150019808A1 (en) Hybrid storage control system and method
CN102122235A (en) RAID4 (redundant array of independent disks) system and data reading and writing method thereof
US9063869B2 (en) Method and system for storing and rebuilding data
CN102326141A (en) Processing method and apparatus for raid configuration information and raid controller
US20150089328A1 (en) Flex Erasure Coding of Controllers of Primary Hard Disk Drives Controller
CN104347122A (en) Accessing and memorizing method and accessing and memorizing device of message type DRAM (Dynamic Random Access Memory) module
CN103034458A (en) Method and device for realizing redundant array of independent disks in solid-state drive
CN101604227B (en) Method and equipment for data storage
CN102799533A (en) Method and apparatus for shielding damaged sector of disk
CN103019893A (en) Multi-disk fault-tolerant two-dimensional hybrid disk RAID4 system architecture and read-write method thereof
WO2023184911A1 (en) Data storage method and apparatus, and device and non-volatile readable storage medium
CN102799543B (en) On the storage medium of dynamic change, dispersion stores data and restoration methods
CN101901115B (en) Method for constructing redundant array of inexpensive disks (RAID) 6 level
CN103699340B (en) A kind of request processing method and equipment
US20170185478A1 (en) Memory controller, storage apparatus, information processing system, and memory controller control method
CN101571820B (en) Redundancy protected method, device and system
CN104516679A (en) RAID (Redundant Array Of Independent Disk) data processing method and device
US20150081969A1 (en) Storage apparatus and control method thereof, and recording medium
EP2889751B1 (en) Method and apparatus for determining redundant array of independent disks

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
ASS Succession or assignment of patent right

Owner name: BEIJING BEIYA CHENXING TECHNOLOGY CO., LTD.

Free format text: FORMER OWNER: BEIJING BEIYA SHIDAI TECHNOLOGY CO., LTD.

Effective date: 20150908

C41 Transfer of patent application or patent right or utility model
TA01 Transfer of patent application right

Effective date of registration: 20150908

Address after: 100000 Beijing city Haidian District Guangyuan Gate No. 5 Building 1, 7 layers of 701-426

Applicant after: BEIJING BEIYA SHIDAI TECHNOLOGY CO.,LTD.

Address before: 100080, 11, Zhongguancun Avenue, Beijing, Haidian District, 8, 879

Applicant before: BEIJING BEIYA SHIDAI TECHNOLOGY Co.,Ltd.

GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20170714

Address after: 100094, new building, building 2, 7 Hui Feng Road, Beijing, Haidian District, 209

Patentee after: North Asia Security Technology (Beijing) Co.,Ltd.

Address before: 100000 Beijing city Haidian District Guangyuan Gate No. 5 Building 1, 7 layers of 701-426

Patentee before: BEIJING BEIYA SHIDAI TECHNOLOGY CO.,LTD.

TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20220713

Address after: 100095 room 231, second floor, building 16, Chuangke town community supporting commercial building, Wenquan Town, Haidian District, Beijing

Patentee after: Beiya Kangcheng (Beijing) Technology Co.,Ltd.

Address before: 100094 209, 2 / F, new materials venture building, No.7 Fenghui Middle Road, Haidian District, Beijing

Patentee before: North Asia Security Technology (Beijing) Co.,Ltd.

TR01 Transfer of patent right