CN109558066A

CN109558066A - Restore the method and apparatus of metadata in storage system

Info

Publication number: CN109558066A
Application number: CN201710880549.8A
Authority: CN
Inventors: 方协云; 吴丹; 魏明昌; 朱小兵
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2017-09-26
Filing date: 2017-09-26
Publication date: 2019-04-02
Anticipated expiration: 2037-09-26
Also published as: CN109558066B

Abstract

This application provides the method and apparatus for restoring metadata in a kind of storage system, the storage system includes first memory block group CKG and the 2nd CKG, first CKG includes at least one band, 2nd CKG includes at least one band, this method comprises: obtaining the first data of the first band in the first CKG and the second data of the second strip in the 2nd CKG；If first data are identical with second data, according to effective columns of effective columns of first band and second strip, and the first CKG state and the 2nd CKG state, select data in first data and second data, which includes writing full state and not writing full state.The method and apparatus for restoring metadata in the storage system of the embodiment of the present application, can be improved the reliability of data.

Description

Restore the method and apparatus of metadata in storage system

Technical field

This application involves computer field of storage, and more particularly, to recovery metadata in a kind of storage system Method and apparatus.

Background technique

In storage system, in addition to the data of storage user on disk, it is also necessary to which the metadata of storage management user data is believed Breath.In order to accelerate the reconstruction speed of the metadata after power-on and power-off, metadata generallys use Idea of Classification organization and administration.Usually come It says, one piece of smaller memory space can be divided on disk and is used to store top layer metadata.The reliability of top layer metadata is to be The basis for reliability of uniting.Bottom metadata uses independent hard disk redundancy array (Redundant Array of Independent Disks, RAID) mode stores.

Independent hard disk redundancy array RAID is a kind of the independent hard disk of muti-piece (physical hard disk) to be combined in different ways Get up to be formed the technology of a hard disk groups (logic hard disk), RAID is capable of providing storage performance more higher than single hard disk and data Backup performance.Data in RAID are with band (Stripe) formal distribution, and band can be understood as in hard disk array two Or the relevant piecemeal in one group of position on multiple subregions, the piecemeal are alternatively referred to as stripe cell (Stripe Unit, SU), i.e. band It is made of multiple stripe cells.It include stripe cell for storing data and check number for storing data in one band According to stripe cell.

In the recovery operation of metadata, for there are identical two parts of data, the prior art is one number of random selection According to the reliability of the data selected in this way is lower.For example, if some data is stored in different CKG bands simultaneously, wherein one Effective columns of CKG is 3 column where part data, and effective columns of CKG only has 1 column where another data, using random selection Mode may choose the data that effective columns only has 1 column, in this way after carrying out system loads, if 1 column data where Disk break down, then will lead to loss of data, reliability is lower.Therefore, need to propose a kind of scheme improves data can By property.

Summary of the invention

The application provides the method and apparatus for restoring metadata in a kind of storage system, can be improved the reliability of data, To promote the reliability of whole system.

In a first aspect, providing a kind of method for restoring metadata in storage system, which is characterized in that the storage system Including first memory block group CKG and the 2nd CKG, the first CKG includes at least one band, and the 2nd CKG includes at least One band, which comprises

Obtain the first band in the first CKG the first data and the second strip in the 2nd CKG second Data；

If first data are identical with second data, according to effective columns of first band and described second The state of the state and the 2nd CKG of effective columns of band and the first CKG, in first data and described Data are selected in second data, the state is including writing full state and not writing full state.

In the embodiment of the present application, when the first data are identical with the second data, equipment having according to the first band is stored The state for imitating effective columns of columns and second strip and the state of the first CKG and the 2nd CKG, in the first data and second Data are selected in data, can select the higher data of reliability, to promote the reliability of storage system.

In some possible implementations, the state of the first CKG is identical as the state of the 2nd CKG, described Effective columns of first band and effective columns of the second strip are different；

Wherein, according to effective columns and described first of effective columns of first band and the second strip The state of the state of CKG and the 2nd CKG selects data in first data and second data, comprising:

In first data and second data, the maximum data of effective columns of corresponding band are selected.

Therefore, when CKG state is identical, the higher data of reliability can be selected according to effective columns of band.

In some possible implementations, the state of the first CKG is different from the state of the 2nd CKG, described Effective columns of first band and effective columns of the second strip are identical；

In first data and second data, the state of the CKG where selecting is to write the data of full state.

Therefore, when effective columns of band is identical, the higher data of reliability can be selected according to the state of CKG.

In some possible implementations, according to the effective of effective columns of first band and the second strip The state of the state and the 2nd CKG of columns and the first CKG, in first data and second data Select data, comprising:

According to the state of effective columns of first band and the first CKG, it is corresponding to calculate first data Weight；

According to the state of effective columns of the second strip and the 2nd CKG, it is corresponding to calculate second data Weight；

Compare the weight of first data and the weight of second data, and in first data and described second The maximum data of weight are selected in data.

Therefore, the corresponding weight of data can be calculated, in order to select according to effective columns of band and the state of CKG The bigger data of weight are selected, to obtain the higher data of reliability.

In some possible implementations, calculates the corresponding weight of first data or calculate second data pair The weight answered, comprising:

If the state of the CKG where the corresponding band of data is to write full state, weight is calculated according to the following formula:

W=N*a+b,

Wherein, if the state of the CKG where the corresponding band of data is not write full state, weight is calculated according to the following formula:

W=N*a；

Alternatively, calculating weight according to the following formula if the state of the CKG where the corresponding band of data is to write full state:

W=N*a,

W=N*a-b,

Wherein, the W indicates that the corresponding weight of data, the N indicate effective columns of the corresponding band of data, a It is the factor of effective columns, the b is state factor, and a and the b are the constant greater than 0, and a > b.

It is therefore possible to use above-mentioned weight computing formula, calculates the weight of data, in order to the number for selecting weight bigger According to obtain the higher data of reliability.

Second aspect provides the device for restoring metadata in a kind of storage system, for execute above-mentioned first aspect or Method in any possible implementation of first aspect.Specifically, the device include for execute above-mentioned first aspect or The module of method in any possible implementation of first aspect.

The third aspect provides the device for restoring metadata in a kind of storage system, which includes processor, memory And transceiver.Processor is connect with memory and transceiver.For storing instruction, processor is used to execute the instruction memory, Transceiver with other network elements for being communicated under the control of a processor.The processor executes the instruction of memory storage When, the execution is so that the processor executes the method in any possible implementation of first aspect or first aspect.

Fourth aspect provides a kind of computer readable storage medium, which has program, It is above-mentioned in a first aspect, and its any in various implementations that the program executes the device for restoring metadata in storage system The method of kind management network slice.

5th aspect, provides a kind of storage system, which includes any in the third aspect or the third aspect Device in the possible implementation of kind；Alternatively, the storage system includes any possible reality in fourth aspect or fourth aspect Device in existing mode.

6th aspect, provides a kind of computer program product comprising instruction, when run on a computer, so that Computer executes the method in above-mentioned first aspect or its any possible implementation.

Detailed description of the invention

Fig. 1 is a schematic diagram for storing equipment.

Fig. 2 is a schematic diagram of the layout in space on disk.

Fig. 3 is the schematic diagram according to an example of the module restored for metadata of the embodiment of the present application.

Fig. 4 is the flow chart according to an example for sweeping disk of the embodiment of the present application.

Fig. 5 is the schematic flow chart for restoring the method for metadata in the storage system according to the embodiment of the present application.

Fig. 6 is the schematic diagram according to an example of the embodiment of the present application.

Fig. 7 is the schematic diagram according to another example of the embodiment of the present application.

Fig. 8 is the schematic block diagram for restoring the device of metadata in the storage system according to the embodiment of the present application.

Fig. 9 is the structural block diagram for restoring the device of metadata in the storage system according to the embodiment of the present application.

Specific embodiment

Below in conjunction with attached drawing, the technical solution in the application is described.

The technical solution of the embodiment of the present application can be applied to the storage equipment using band (Stripe) as way to manage Or storage system, for example, storage array or SSD itself or folded of solid state hard disk (Solid State Disk, SSD) composition The storage array etc. of tile style magnetic recording (Shingled Magneting Recording, SMR) disk composition.

Stripe is a kind of method that multiple hard disks (being then multiple storage mediums in single disc) are merged into a volume, Stripe can be understood as in hard disk array (or hard disk) the relevant piecemeal in one group of position, the piecemeal on two or more subregions It can be described as stripe cell (Stripe Unit, SU), i.e. Stripe is made of multiple SU.Stripe is to carry out item to memory space With change management as a result, the specific understanding that may refer to those skilled in the art to Stripe.SU with the block in hard disk (Block) there is corresponding relationship, illustratively, as shown in Figure 1, Stripe 0 includes 3 SU, this 3 SU are corresponded respectively in Fig. 1 The data block (i.e. Block) in memory block (Chunk, CK) 1 in 102 signified a line, data block in memory block 2 ..., hard Verification data block in disk n；Stripe 1 includes 3 SU, this 3 SU correspond respectively to the storage in Fig. 1 in 104 signified a line The verification data block in data block, memory block 2 in block 1 ..., the data block in hard disk n；Stripe 3 includes 3 SU, this 3 A SU corresponds respectively in Fig. 1 the data block in the memory block 1 in 106 signified a line, the data block in memory block 2 ..., hard Verification data block in disk n；Stripe 3 includes 3 SU, this 3 SU correspond respectively to the storage in Fig. 1 in 108 signified a line Verification data block in block 1, the data block in memory block 2 ..., the data block in hard disk n.In Fig. 1, a memory block group (Chunk Group, CKG) is made of n CK.

Non-limiting as example, storage equipment is, for example, SSD storage array shown in Fig. 1, such as shown in Fig. 1 Memory block 1, memory block 2 ... memory block n is SSD.It in order to facilitate understanding and describes, is applied to below with the embodiment of the present application It is illustrated for storage equipment based on SSD, but those skilled in the art can be very according to the introduction of the embodiment of the present application It is expressly understood, the method for the embodiment of the present application can be applied equally to other storages using Stripe as way to manage and set Standby, such application is each fallen in the protection scope of the application.

SSD is a kind of Novel storage system using flash memory as storage medium, and there is no the machinery as magnetic head for flash memory The expense of component, random access mode and sequential access mode be it is comparable, therefore, SSD than conventional hard have it is bigger Performance advantage.Flash memory, which has, first wipes the characteristic that could be written again, for example, when will more new data in a flash memory when, in data Before old value is not wiped free of, be cannot physical address write-in data where the old value in data be newly worth, that is, first will need The old value of data is wiped, just can physical address write-in data where the old value in data new value.But there are also a for flash memory Characteristic is, reading and writing operate as unit of page, erasing operation with block (including multiple pages) for unit.In other words, Yao Shixian is in number According to old value where physical address write-in data new value, need to where the old value of data entire block (usually 64K or Data on 128K) are wiped, and the new value of data then could be written in the block of erased state.But firstly, erasing The cost of block is very huge, because while the block in flash memory can be wiped repeatedly, but each piece of erasing times It is limited, the service life of flash memory is calculated by erasing times, such as multi-level cell memory (Multi-Level Cell, MLC) The service life of flash memory, the service life of single-order storage unit (Single-Level Cell, SLC) flash memory was then generally at 1000-10000 times At 100,000 times or so.Secondly, needing before wiping the block first will be effective if further including valid data in the block to be wiped Data are read, and new block address is then written, this process is known as data-moving, and data-moving can generate write-in amplification (Write Amplification, WA), that is, the data volume being actually written into, which will be greater than, needs to be written the data volume of SSD, and write-in amplification is one The undesirable phenomenon of kind.In order to write data to user's vacating space, need to carry out garbage reclamation (Garbage Collection, GC), i.e., all valid data present in the band are moved in other idle bands, then recycles the band.And During moving, the case where not wiping out with regard to powered-off fault also there are old piece, alternatively, there are where old piece before old piece of erasing The case where hard disk in CKG where some CK is pulled out, these situations be likely to result in individual data be stored in it is multiple and different CKG in.For these situations, the prior art is usually to randomly choose a data, and reliability is lower.

As shown in Figure 1, including controller and SSD connected to the controller (assuming that Fig. 1 by the storage array that SSD is formed Shown in hard disk be SSD).Wherein, controller is used to SSD being divided into multiple Stripe (Stripe as shown in Figure 1 0, Stripe1, Stripe2 and Stripe3), Stripe is reused as host, and logical unit number (Logic Unit is provided Number, LUN).LUN can be regarded as an accessible logic hard disk of host, also referred to as logical space unit, It is the space that read-write may be implemented in user.As shown in Figure 1, a Stripe includes multiple SU, each SU has block (Block) It is corresponding to it.It should be understood that SU can be written in corresponding data when host carries out write operation in LUN, can accordingly be written corresponding Block in.When carrying out garbage reclamation to Stripe, corresponding piece can be also recycled, that is, wipes corresponding piece, and discharges its storage Space.

Specifically, controller shown in Fig. 1 is, for example, RAID controller, then Stripe 0 shown in FIG. 1, Stripe1, Stripe2 and Stripe3 is the band set up based on RAID technique.

It should be understood that the embodiment of the present application can be applied to the storage system using RAID1 technology, it also can be applied to other Scene based on RAID technique, for example, the technologies such as RAID6, RAID10, RAID1E, RAID-TP.

Within the storage system, on hard disk other than the data of storage user, it is also necessary to first number of storage management user data It is believed that breath.The storage of metadata can be using the scene based on RAID technique shown in above-mentioned Fig. 1.Metadata generallys use classification and thinks Want to organize, first layer meta-data region is the metadata information of data field, the first layer meta-data region of second layer meta-data region Metadata information ..., n-th layer meta-data region is the metadata information of (n-1)th layer of meta-data region, so that top layer (n-th Layer) metadata data volume it is few as far as possible, to quickly reconstruct the root of metadata after power-on and power-off, provide read-write for user Business.

Within the storage system, one piece of smaller memory space can be divided on usual disk to be used to store top layer metadata, pushed up The recovery of layer metadata is usually to be restored by the way of sweeping disk.Fig. 2 shows the schematic layout patterns in space on disk.Such as Fig. 2 institute Show, every piece of hard disk partition is top layer meta-data region (area meta) and data field (area data), wherein the area meta is for storing top layer Metadata, the area data for storing data with non-top layer metadata.The process for powering on scan rebuilding top layer metadata is exactly to sweep The above-mentioned area meta is retouched, in order to identify valid data, restores top layer metadata.

During restoring metadata, storing may include the software module restored for metadata in equipment.Fig. 3 is shown The schematic diagram of one example of the module restored for metadata.As shown in figure 3, restoring in process in metadata, storage pool For controlling metadata process, block management (Block Manager, BM) module is used for management (Pool Manger, PMGR) module Designated memory space is scanned, region (ZONE) module is used for the CK, Metadata Service (Metadata that tissue scanning comes out Service, MDS) module is for being responsible for organization and administration and verification of metadata etc., block device management (Block Device Management, BDM) module drives for the service of connection devices and hardware, it is responsible for the management of hard disk.Above-mentioned modules can combine Using or be used alone, to realize the technical solution of the embodiment of the present application.Above-mentioned modules can be deployed in storage equipment, It can also be individually present, this is not construed as limiting.

Here in conjunction with the flow chart in Fig. 4 to the invention relates to disk process of sweeping be described.Fig. 4 shows root According to the flow chart of an example 400 for sweeping disk of the embodiment of the present application.For example, as shown in figure 4, the example 400 includes:

401, judge whether designated memory space scans completion.

Specifically, the BM module of the storage equipment in Fig. 3 can deposit specified according to the granule size of memory block CK Memory block CK in storage space is scanned, and judges whether the memory space scans completion.If judging, designated memory space is swept Completion is retouched, then stops scanning, designated memory space scanning does not complete if judging, goes to step 402.

402, read the data of next memory block CK.

Specifically, if judging, designated memory space scanning is not completed, and can continue to execute scanning motion, that is, it is empty to read storage Between in next memory block CK data.

403, judge whether to be valid data.

Specifically, the MDS module of the storage equipment in Fig. 3 can verify whether the data in memory block CK are significant figures According to.If the data in memory block CK are not valid data, go to 401 and continue to scan on, if the data in memory block CK have Data are imitated, then go to 404.

404 and other memory block CK scanned form memory block group CKG.

It is formed specifically, the ZONE module of the storage equipment in Fig. 3 can will scan the effective memory block CK come Memory block group CKG.Wherein, the state of memory block group CKG, which can be, writes full or does not write full.For ease of description, full shape will be write The memory block group of state claims less than CKG, and the memory block group for not writing full state is known as less than CKG.

Optionally, memory block group CKG may be full CKG or less than CKG.

Metadata is described here in conjunction with Fig. 4 to restore to sweep disk process in process.Wherein, the purpose of scanning is to obtain effective CKG, and the grain data (the minimum memory access unit of metadata) in CKG will be lost to metadata according to the state of CKG Management module, in order to which subsequent metadata management module can select reliably according to the state of CKG and the columns of valid data The higher data of property.

The embodiment of the present application is applied in the scene that metadata is restored.Usually, metadata is restored to establish in top layer member On the basis of data are restored, i.e., only top layer metadata is completely restored after completing, and could successively restore the metadata of other layers. The embodiment of the present application, which provides, to be restored the method for metadata and can choose when carrying out metadata recovery in a kind of storage system The higher data of reliability, to improve the reliability of whole system.

Fig. 5 shows the method 500 for restoring metadata in the storage system according to the embodiment of the present application.The storage system Including first memory block group CKG and the 2nd CKG, the first CKG includes at least one band, and the 2nd CKG includes at least One band.The method 500 can be applied to scene described in above-mentioned Fig. 1.Optionally, the method 500 can be by storing Equipment stores the metadata management module in equipment to realize, the metadata management module can be deployed in Fig. 1 In RAID controller, alternatively, the metadata management module is independently arranged, this is not construed as limiting.For ease of description, under Face will be described for storing equipment.As shown in figure 5, the method 500 includes:

S510 obtains the first data of the first band in the first CKG and the second number of the second strip in the 2nd CKG According to.

Optionally, storage equipment is by carrying out validity check to the data in the first band, when data check passes through Above-mentioned first data can be obtained.Wherein, first data are the first bands being stored in the first CKG in a manner of RAID. For example, the first CKG can organize data using RAID1 backup mode.

Optionally, storage equipment is by carrying out validity check to the data in second strip, when data check passes through Above-mentioned second data can be obtained.Wherein, second data are the second strips being stored in the 2nd CKG in a manner of RAID. For example, the 2nd CKG can organize data using RAID1 backup mode.

It should be understood that above-mentioned first band or second strip refer to single band.Data in single band can be with prepare more part Mode store, wherein the minimum memory access unit of data can be grain data.

Optionally, the first data or the second data are potentially stored in the band in full CKG, it is also possible to are stored in less than In band in CKG.It is swept in the band of full CKG or less than CKG obtained in disk process for example, being stored in earlier figures 4.

Optionally, there is corresponding effective columns in above-mentioned first band or second strip.For example, right in Fig. 1 In band 1, in n memory block, if after the data check stored in memory block 1 and memory block 2 being effective, memory block 1 It can be understood as effective columns of band 1 with memory block 2.

Optionally, storage equipment can the data content to above-mentioned first data and above-mentioned second data data content into Row compares, if the data content of the first data and the data content of the second data are identical, can execute subsequently selected step, Data are selected in the first data and the second data.If the data content of the first data and the data content of the second data are not Together, then the data information of the first data and the data information of the second data are recorded.That is, storage equipment is also possible to get The data not occurred before did not occurred the information of data before then can first saving this, in order to subsequent use.

It should be understood that being to be illustrated with the first data and the second data instance here, not to the guarantor of the embodiment of the present application It protects range and constitutes restriction.

It should also be understood that be illustrated by taking the first CKG and the 2nd CKG in storage system as an example here, in practice may be used There are more CKG, this is not construed as limiting.

S520, if first data are identical with second data, according to effective columns of first band and The state of the state and the 2nd CKG of effective columns of two bands and the first CKG, in first data and institute It states and selects data in the second data, the state is including writing full state and not writing full state.

Optionally, store equipment judge the first data data content and the second data content it is identical when, can be with Comprehensively consider following factor: (1) effective columns of band；(2) the corresponding state of CKG where band, in order in the first number According to in the second data, the highest data of reliability are selected.Further, after selected data, metadata management module can To save or record the information of selected data, selected data are put into metadata organization management.

For example, metadata management module can be according to the selected data of record during rebuilding top layer metadata Information, carry out the recovery operation of data.

It should be understood that in the embodiment of the present application, introducing number " first ", " second " ... is intended merely to distinguish different pairs As for example, different " CKG " is distinguished, alternatively, different " band " is distinguished, not to the protection scope structure of the embodiment of the present application At restriction.

It should be noted that above-mentioned only carry out so that identical data possibly are present at the first band and second strip as an example Illustrate, the protection scope of the embodiment of the present application is not constituted and limited.In specific implementation, identical data possibly are present at In N number of band (N is greater than or equal to 2 integer), in this case, the technical solution of the embodiment of the present application is still applicable 's.That is, those skilled in the art can be according to the effective columns and CKG in the embodiment of the present application based on band State determines the scheme of weight, and the maximum data of weight are selected in N number of band.

The embodiment based on above-mentioned two factor selection pending data is described in detail respectively below.

Optionally, as one embodiment, the state of the first CKG is identical as the state of the 2nd CKG, and described Effective columns of one band and effective columns of the second strip are different；

Wherein, S520 includes:

That is, if the state of above-mentioned first CKG and the state of the 2nd CKG are identical, for example, being to write full shape State CKG is not write full state CKG, then effective columns of band need to be considered by storing equipment only.Then, metadata management Module is in the first data and the second data, the most data of corresponding effective columns in the band where selecting.Due to having Effect columns is more, and the reliability of the data stored in band is higher.In this way, storage is set in the identical situation of state of CKG It is standby can be according to effective columns corresponding in each band, to select the data of better reliability.

Optionally, as one embodiment, the state of the first CKG is different from the state of the 2nd CKG, and described Effective columns of one band and effective columns of the second strip are identical；

Wherein, S520 includes:

That is, if effective columns of above-mentioned first band and effective columns of second strip be it is identical, deposit Storage equipment need to only consider the state of memory block group CKG.Equipment is stored in the first data and the second data, where selecting Memory block group CKG be the data for writing full state.Since the data for writing the memory block group CKG of full state are more complete, full state is write Memory block group CKG in the data reliability that stores can be higher.For example, the 2nd CKG is not write if the first CKG is to write full state Full state, then storage equipment selects the first data.In this way, in the identical situation of effective columns of band, it can be according to every The state of memory block group CKG where a band, to select the data of better reliability.

Optionally, S520 includes:

That is, can effective columns and CKG based on band state computation data weight, in order to right to choose It is worth maximum data.Wherein, the weight can be understood as the weighted value for describing data reliability.

Further, the embodiment of the present application proposes the weight that following formula calculate data, and following formula are used to be based on item The weight of the state computation data of effective columns and CKG of band.Here above-mentioned first data and above-mentioned second data can adopt Corresponding weight is calculated with following formula.

For example, if the state of the CKG where the corresponding band of data is to write full state, following formula meter can be used Calculate weight:

W=N*a+b (1.1),

Accordingly, if the state of the CKG where the corresponding band of data is not write full state, following formula meter can be used Calculate weight:

W=N*a (1.2)；

Alternatively, if the state of the CKG where the corresponding band of data is to write full state power can be calculated using following formula Value:

W=N*a (1.3),

W=N*a-b (1.4)；

Wherein, the W indicates that the corresponding weight of data, the N indicate effective columns of the corresponding band of data, a It is the factor of effective columns, the b is the state factor of memory block group CKG, and a and the b are the constant greater than 0, and a >b。

Here, it in the identical situation of effective columns or in the identical situation of state of CKG, can continue to use above-mentioned Formula 1.1-1.4 calculates the weight of data, and therefore not to repeat here.

Therefore, the embodiment of the present application, which introduces, calculates weight computing formula, and it is corresponding to calculate data based on weight computing formula Weight, and by comparing the biggish data of Weight selected weight of data, to improve the reliability of data.

For the ease of it will be understood by those skilled in the art that, retouched below with reference to the process of the specific implementation in Fig. 6 and Fig. 7 It states.In the example of Fig. 6 and Fig. 7 below, CKG is to carry out storing data in such a way that RAID1 writes more parts.

For example, Fig. 6 shows the process flow diagram of full CKG.As shown in fig. 6, the process 600 includes:

601, read the first row CK data in full CKG.

It should be understood that be illustrated with the data instance for reading first row CK here, this is not construed as limiting.

For full CKG, since current CKG is to organize data with the backup mode of RAID1, so each arrange CK's Data are the same, therefore can read the data of a wherein column CK at random, for example, the BM module of storage equipment can be read completely The data of first row CK in CKG.

602, judge whether all first row CK data complete by processing.If all processing is completed, terminate；If not all processing It completes, goes to 603.

Here, the ZONE module for storing equipment needs to be traversed for each data in first row CK data, until first row CK All data in data all handle completion.

603, judge whether the individual data in first row CK data verifies and passes through.If verification passes through, equipment is stored ZONE module sends the data to the metadata management module of storage equipment, and goes to 604；If verification does not pass through, go to The process flow (hereafter will do it introduction) of less than CKG.

Here, it needs to carry out validity check to individual data.If verification failure, needs the stream according to less than CKG Journey processing；If verified successfully, following 604 can be continued to execute.

604, judge whether identical data occur.If there are identical data, 605 are skipped to.

For example, storage equipment in metadata management module judge whether to occur it is identical as the data that above-mentioned verification passes through Data then skip to 605 if there is the identical data of the data that pass through with above-mentioned verification, compare the corresponding power of identical data Value.Wherein, the calculation method of the weight may refer to the formula of previously described full CKG, not repeat here.If above-mentioned verification By data before do not occurred, then go to 606, it can first retain the information of the data not occurred, in order to It is subsequent to occur being compared when identical data.

605, the biggish data of weight are selected, and go to 606.

For example, there is Data1 in the band 1 in CKG1 and the band in CKG2 2, then can use described previously Weight formula, the weight of the state computation Data1 of effective columns and CKG1 based on band 1, and based on the effective of band 2 The weight of the state computation Data1 of columns and CKG2, and the biggish Data1 of weight is selected, and record the information of data, in order to Management.In the process of Fig. 7, the calculating of the weight of data needs to calculate according to the previously described formula for writing full CKG.

606, it is put into metadata organization management.

Specifically, the information of data of above-mentioned maximum weight can be saved by storing the metadata management module in equipment, with Convenient for being used when subsequent recovery metadata.Metadata management module in storage equipment, which also can recorde, does not occur identical data Information judges to use when there is identical data in order to subsequent.For example, if the metadata management module in storage equipment judges Above-mentioned first data are different with the data content of the second data, then record the information of the first data and the information of the second data.

In another example Fig. 7 shows the process flow diagram of less than CKG.As shown in fig. 7, the process 700 includes:

701, judge whether the data in CKG in all bands handle completion.If completing, terminate；If not completing, turn To 702.

Here, the ZONE module for storing equipment needs successively to traverse the data of each band in less than CKG, until each The data processing of band is completed.

702, for each band in CKG, read corresponding offset data in each column CK.

That is, needing to read the more parts of data stored in the band for each band.

703, in the same band, judge whether the data content for the more parts of data that verification passes through is identical.

Since current CKG is to organize data with the backup mode of RAID1, so more parts of data in a band It is identical.Here, every part of data for needing to be successively read in same band, if the data content of the more parts of data is identical, 704 are then gone to, which can be put into metadata organization management；If the data content of the more parts of data is different, go to 707.For example, n parts, i.e. n parts of Data1 can be stored in the same band for data Data1, judging n parts of Data1's When data content is identical, Data1 is put into metadata organization management by this.

704, judge whether identical data occur.Identical data if it exists, then go to 705；It is identical if it does not exist Data then go to 706.

For example, storage equipment in metadata management module after obtaining Data1, judge whether occurred before with The identical data of Data1.If there is mistake, then 705 weights for comparing data are gone to；If do not occurred identical as Data1 Data, then Data1 is put into metadata organization management.

705, the data for selecting weight big, and go to 706.

For example, the metadata management module in storage equipment occurred and Data1 phase before judgement after obtaining Data1 Same data select the biggish data of weight then the corresponding weight of Data1 can be calculated.For example, the band in CKG1 There is Data1 in the band 2 in 1 and CKG2, then previously described weight formula, effective column based on band 1 can be used The weights of several and CKG1 state computation Data1, and the state computation Data1 of effective columns based on band 2 and CKG2 Weight, and select the biggish Data1 of weight.In the process of Fig. 7, the calculating of the weight of data need according to it is previously described not The formula for writing full CKG calculates.

706, it is put into metadata organization management.

Specifically, the information of the biggish data of weight can be saved by storing the metadata management module in equipment, in order to Follow-up management.Metadata management module in storage equipment also can recorde the information for not occurring identical data, in order to subsequent Judge to use when there is identical data.For example, if judging, there is no data identical with Data1, then can in 704 To record the information of Data1.

707, skip data.

When reading every part of data in same band, if there is the different data of data content, then need to skip This part of data, do not deal with.

It should be understood that above-mentioned combination Fig. 6 and Fig. 7 describes the specific implementation flow according to the embodiment of the present application, not to this Apply for that the protection scope of embodiment is constituted to limit.

It should be understood that the example of Fig. 6 and Fig. 7 is intended merely to facilitate those skilled in the art understand that the embodiment of the present application, and The embodiment of the present application have to be limited to the concrete scene illustrated.Those skilled in the art are according to the example of Fig. 6 and Fig. 7, it is clear that can To carry out the modification or variation of various equivalences, such modification or variation are also fallen into the range of the embodiment of the present application.

Foregoing describe the methods for restoring metadata in the storage system according to the embodiment of the present application, and basis is described below Restore the device of metadata in the storage system of the embodiment of the present application.

Fig. 8 shows the schematic block diagram for restoring the device 800 of metadata in the storage system according to the embodiment of the present application. Optionally, the storage system includes first memory block group CKG and the 2nd CKG, and the first CKG includes at least one band, 2nd CKG includes at least one band, and described device 800 is used to execute the method or step of previous embodiment.Optionally, Modules in described device 800, which can be through software, to be realized.Optionally, described device 800 can be deployed in logical With in computer equipment.As shown in figure 8, described device 800 includes:

Obtain module 810, the Article 2 in the first data and the 2nd CKG for obtaining the first band in the first CKG Second data of band；

Selecting module 820, if identical with second data for first data, according to having for first band The state for imitating effective columns of columns and second strip and the state of the first CKG and the 2nd CKG, described the Data are selected in one data and second data, the state is including writing full state and not writing full state.

Optionally, the state of the first CKG is identical as the state of the 2nd CKG, effective column of first band Number is different from effective columns of the second strip；

Wherein, the selecting module 820 is specifically used for:

Optionally, the state of the first CKG is different from the state of the 2nd CKG, effective column of first band Number is identical as effective columns of the second strip；

Wherein, the selecting module 820 is specifically used for:

Optionally, the selecting module 820 includes:

Computing unit, for according to effective columns of first band and the state of the first CKG, calculating described the The corresponding weight of one data；The state of the effective columns and the 2nd CKG according to the second strip is also used to, described in calculating The corresponding weight of second data；

Comparing unit, for the weight of first data and the weight of second data, and described first The maximum data of weight are selected in data and second data.

Optionally, the computing unit is specifically used for:

W=N*a+b,

W=N*a；

W=N*a,

W=N*a-b,

Restore metadata in aforementioned storage system it should be understood that can be used for executing according to the device 800 of the embodiment of the present application Step in method, and above and other management operation and/or function of the modules and/or unit in device 800 point Not in order to realize the corresponding steps of aforementioned each method, therefore the beneficial effect in preceding method embodiment also may be implemented, is Succinct, details are not described herein.

Fig. 9 is the structural frames for restoring the device 900 of metadata in the storage system provided according to another embodiment of the application Figure.For example, described device 900 is common hardware computer equipment, or storage equipment.Restore in storage system shown in Fig. 9 The device 900 of metadata includes: processor 901, memory 902 and transceiver 903.

It is communicated with each other between processor 901, memory 902 and transceiver 903 by internal connecting path, transmitting control And/or data-signal.In a possible design, processor 901, memory 902 and transceiver 903 can pass through chip reality It is existing.The memory 902 can store program code, and the program code that processor 901 calls memory 902 to store is somebody's turn to do with realizing Restore the corresponding function of the device of metadata in storage system.

The processor 901 is for executing following steps:

Obtain the first data of the first band in the first CKG and the second data of the second strip in the 2nd CKG；If institute State that the first data are identical with second data, according to effective columns of effective columns of first band and second strip, And the first CKG state and the 2nd CKG state, select number in first data and second data According to the state is including writing full state and not writing full state.

It is understood that although not showing that, which can also include other devices, such as input unit, defeated Device, battery etc. out.

In a possible design, in some embodiments, memory 902 be can store for executing in preceding method Store the instruction for the method that equipment executes.Processor 901 can execute the instruction stored in memory 902 and combine other hardware (such as transceiver 903) completes the step of storage equipment executes in preceding method, and specific work process and beneficial effect can join See the description in preceding method embodiment.

The embodiment of the present application also provides a kind of storage systems, including the device 900 in above-described embodiment.

The method that above-mentioned the embodiment of the present application discloses can be applied in processor, or be realized by processor.Processor It may be a kind of IC chip, the processing capacity with signal.During realization, each step of the above method can lead to The instruction of the integrated logic circuit or software form of crossing the hardware in processor is completed.Above-mentioned processor can be general place Manage device, digital signal processor (digital signal processor, DSP), specific integrated circuit (application Specific integrated circuit, ASIC), ready-made programmable gate array (field programmable gate Array, FPGA) perhaps other programmable logic device discrete gates or transistor logic, discrete hardware components may be used also To be System on Chip/SoC (system on chip, SoC), can also be central processing unit (central processor unit, CPU), it can also be network processing unit (network processor, NP), can also be digital signal processing circuit (digital signal processor, DSP) can also be microcontroller (micro controller unit, MCU), also It can be programmable controller (programmable logic device, PLD) or other integrated chips.May be implemented or Execute disclosed each method, step and the logic diagram in the embodiment of the present application.General processor can be microprocessor or The processor is also possible to any conventional processor etc..The step of method in conjunction with disclosed in the embodiment of the present application, can be direct Be presented as that hardware decoding processor executes completion, or in decoding processor hardware and software module combination execute completion. Software module can be located at random access memory (random access memory, RAM), flash memory, read-only memory The abilities such as (read-only memory, ROM), programmable read only memory or electrically erasable programmable memory, register In the storage medium of domain maturation.The storage medium is located at memory, and processor reads the instruction in memory, complete in conjunction with its hardware The step of at the above method.

It should also be understood that magnitude of the sequence numbers of the above procedures are not meant to execute in the various embodiments of the application Sequence it is successive, the execution of each process sequence should be determined by its function and internal logic, the reality without coping with the embodiment of the present application It applies process and constitutes any restriction.

It should also be understood that the terms "and/or", only a kind of incidence relation for describing affiliated partner, expression can be with There are three kinds of relationships, for example, A and/or B, can indicate: individualism A exists simultaneously A and B, these three feelings of individualism B Condition.In addition, character "/" herein, typicallys represent the relationship that forward-backward correlation object is a kind of "or".

Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed Scope of the present application.

It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description, The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.

In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit It closes or communicates to connect, can be electrical property, mechanical or other forms.

The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme 's.

It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit It is that each unit physically exists alone, can also be integrated in one unit with two or more units.

It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps. And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), arbitrary access are deposited The various media that can store program code such as reservoir (Random Access Memory, RAM), magnetic or disk.

The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain Lid is within the scope of protection of this application.Therefore, the protection scope of the application should be based on the protection scope of the described claims.

Claims

1. restoring the method for metadata in a kind of storage system, which is characterized in that the storage system includes first memory block group CKG and the 2nd CKG, the first CKG includes at least one band, and the 2nd CKG includes at least one band, the method Include:

Obtain the first data of the first band in the first CKG and the second data of the second strip in the 2nd CKG；

If first data are identical with second data, according to effective columns of first band and the second strip Effective columns and the first CKG state and the 2nd CKG state, in first data and described second Data are selected in data, the state is including writing full state and not writing full state.

2. the method according to claim 1, wherein the state of the state of the first CKG and the 2nd CKG Identical, effective columns of first band and effective columns of the second strip are different；

Wherein, described according to effective columns of first band and effective columns of the second strip and described first The state of the state of CKG and the 2nd CKG selects data in first data and second data, comprising:

3. the method according to claim 1, wherein the state of the state of the first CKG and the 2nd CKG Difference, effective columns of first band and effective columns of the second strip are identical；

4. according to the method in any one of claims 1 to 3, which is characterized in that the having according to first band The state for imitating effective columns of columns and the second strip and the state of the first CKG and the 2nd CKG, in institute It states in the first data and second data and selects data, comprising:

According to the state of effective columns of first band and the first CKG, the corresponding weight of first data is calculated；

According to the state of effective columns of the second strip and the 2nd CKG, the corresponding weight of second data is calculated；

Compare the weight of first data and the weight of second data, and in first data and second data The maximum data of middle selection weight.

5. according to the method described in claim 4, it is characterized in that, described calculate the corresponding weight of first data or calculating The corresponding weight of second data, comprising:

W=N*a+b,

W=N*a；

W=N*a,

W=N*a-b,

Wherein, the W indicates that the corresponding weight of data, the N indicate effective columns of the corresponding band of data, and a is that have The factor of columns is imitated, the b is state factor, and a and the b are the constant greater than 0, and a > b.

6. restoring the device of metadata in a kind of storage system, which is characterized in that the storage system includes first memory block group CKG and the 2nd CKG, the first CKG includes at least one band, and the 2nd CKG includes at least one band, described device Include:

Obtain module, the Article 2 in the first data and the 2nd CKG for obtaining the first band in the first CKG Second data of band；

Selecting module, if identical with second data for first data, according to effective columns of first band With the state of effective columns of the second strip and the state of the first CKG and the 2nd CKG, described first Data are selected in data and second data, the state is including writing full state and not writing full state.

7. device according to claim 6, which is characterized in that the state of the first CKG and the state of the 2nd CKG Identical, effective columns of first band and effective columns of the second strip are different；

Wherein, the selecting module is specifically used for:

8. device according to claim 6, which is characterized in that the state of the first CKG and the state of the 2nd CKG Difference, effective columns of first band and effective columns of the second strip are identical；

Wherein, the selecting module is specifically used for:

9. the device according to any one of claim 6 to 8, which is characterized in that the selecting module includes:

Computing unit, for calculating first number according to effective columns of first band and the state of the first CKG According to corresponding weight；It is also used to the state of the effective columns and the 2nd CKG according to the second strip, calculates described second The corresponding weight of data；

Comparing unit, for the weight of first data and the weight of second data, and in first data With the selection maximum data of weight in second data.

10. device according to claim 9, which is characterized in that the computing unit is specifically used for:

W=N*a+b,

W=N*a；

W=N*a,

W=N*a-b,