CN109697136A

CN109697136A - A kind of data reconstruction method, apparatus and system

Info

Publication number: CN109697136A
Application number: CN201710985172.2A
Authority: CN
Inventors: 舒治
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2017-10-20
Filing date: 2017-10-20
Publication date: 2019-04-30
Anticipated expiration: 2037-10-20
Also published as: CN109697136B

Abstract

The embodiment of the present invention provides a kind of data reconstruction method, apparatus and system.First server receives the metadata for the first data fragmentation that second server is sent, and the first instruction information that the metadata of the first data fragmentation includes is the mark or storage location of Backup Data fragment of first data fragmentation in the first server.First server confirms that the first server does not save the first data fragmentation in the Backup Data fragment of first server according to the metadata of the first data fragmentation, Backup Data fragment of first data fragmentation in second server is requested to second server, saves the Backup Data fragment of the first data fragmentation.Due to meta-data preservation the first data fragmentation first server Backup Data fragment mark or first server Backup Data fragment storage location data volume it is small relative to the data volume of cryptographic Hash.Therefore, to transmit Internet resources used in a large amount of metadata from second server few, improves data recovering efficiency.

Description

A kind of data reconstruction method, apparatus and system

Technical field

The present embodiments relate to data to restore field more particularly to a kind of data reconstruction method, apparatus and system.

Background technique

Existing data backup mode includes that local backup and remote backup relative to local backup pass through remote backup Improve the reliability of Backup Data.When restoring local data, user is according to local Backup Data or the backup number of distal end Factually show the recovery of local data.To improve the utilization rate that resource is locally stored, local expired backup locally can be constantly deleted Data.

There are two types of traditional data reset modes, and first way reads full backup number from distal end for local server According to then saving full backup data to local.The second way, which is that local server confirmation is local, whether there is Backup Data, Distal end is not spent for Backup Data existing for local to read, and for the Backup Data being locally not present, needs to read from distal end. Before restoring data by the second way, Backup Data is distally divided into several data fragmentations with local can pre-save In each data fragmentation and each data fragmentation cryptographic Hash.Local expired Backup Data can be constantly deleted due to local, So the local cryptographic Hash that may save part or total data fragment.When restoring data, if local search is protected to local Include the cryptographic Hash of data fragmentation to be restored in the cryptographic Hash deposited, then read the data fragmentation of backup from local, if local It does not include the cryptographic Hash of data fragmentation to be restored in the cryptographic Hash of preservation, then it is local that the data fragmentation backed up is read from distal end.

Data are restored using the first data reset mode, are needed through network transmission full backup data, when recovery passes Defeated data volume is big, expends Internet resources.Data are restored using second of data reset mode, need in advance to draw Backup Data It is divided into several data fragmentations, to the cryptographic Hash for saving each data fragmentation after the progress Hash operation processing of each data fragmentation, behaviour Make complicated, it is larger to expend the byte space that the cryptographic Hash of process resource and data fragmentation occupies, and expends local and distal end deposit Store up resource.In addition, when restoring data using second of data reset mode, it is local to need to cache from the backup number distally read According to all data fragmentations cryptographic Hash, need to occupy Internet resources and local cache resources.

Summary of the invention

The invention discloses a kind of data reconstruction method, apparatus and system, the data reconstruction method does not have to every number Hash operation processing is carried out according to fragment, it is easy to operate, improve data backup efficiency.

In a first aspect, providing a kind of data reconstruction method, executed by first server.The first server sends first number According to acquisition request, the metadata acquisition request includes Backup Data mark, and the Backup Data includes the first data fragmentation.So Afterwards, the first server receives the metadata for first data fragmentation that second server is sent.Wherein, first number Metadata according to fragment includes the first instruction information, and the first instruction information is first data fragmentation in first clothes The mark or storage location for the Backup Data fragment being engaged in device.The first server is according to first number of first data fragmentation According to confirming whether the first server preserves first data fragmentation in the Backup Data fragment of the first server, When confirming that the first server does not save Backup Data fragment of first data fragmentation in the first server, The first server sends the first data fragmentation acquisition request, and the first data fragmentation acquisition request includes described the The metadata of one data fragmentation.The first data fragmentation acquisition request is used to indicate the second server and searches described first Backup Data fragment of the data fragmentation in the second server.The first server receives the second server and sends Backup Data fragment of first data fragmentation in the second server, save first data fragmentation described Backup Data fragment in second server.

In data reconstruction method provided in an embodiment of the present invention, what it is due to meta-data preservation is first data fragmentation The Backup Data fragment of first server mark or first data fragmentation first server Backup Data fragment Storage location, mark or first data fragmentation of first data fragmentation in the Backup Data fragment of first server It is small relative to the data volume of cryptographic Hash in the data volume of the storage location of the Backup Data fragment of first server.Therefore, relatively A large amount of cryptographic Hash is transmitted from second server 120 in existing, a large amount of metadata is transmitted from second server 120 and is used Internet resources it is few, propagation delay time is short, improves data recovering efficiency.Moreover, first server receives in the embodiment of the present invention To after a large amount of metadata, a large amount of spatial cache storage is expended in the first server relative to existing first server It is few to store cache resources used in a large amount of metadata for a large amount of cryptographic Hash.

Based in a first aspect, in the first possible implementation, the first server sends metadata acquisition and asks Before asking, the method also includes: the first server saves the copy of first data fragmentation as first number According to fragment the first server Backup Data fragment；The first server generates first number of first data fragmentation According to；The first server sends the metadata of first data fragmentation and first data fragmentation to the second service Device.

Based in the possible implementation of the first of first aspect or first aspect, in second of possible implementation In, it is described whether the first server preserves according to the metadata of first data fragmentation confirmation first server Backup Data fragment of first data fragmentation in the first server, comprising: the first server is according to second clothes The metadata for first data fragmentation that business device is sent, confirms whether the first server preserves first data point The metadata of piece, wherein the first server is by confirming that the first server does not save first data fragmentation Metadata, confirm that the first server does not save the metadata of first data fragmentation in the first server Backup Data fragment.

It is possible at the third in the first or second of possible implementation based on first aspect or first aspect In implementation, further includes: the first server receives data modification request, and the data modification request is used to indicate described It is target data fragment that first server, which modifies first data fragmentation,.The first server modifies first data point After piece is the target data fragment, the copy of the target data fragment is saved as the target data fragment described the The Backup Data fragment of one server, and the metadata of the target data fragment is generated, delete first data fragmentation Metadata, the metadata of the target data fragment includes Target indication information, the Target indication information be the target Mark or storage location of the data fragmentation in the Backup Data fragment of the first server.Described in the first server is sent Target data fragment, the target data fragment metadata to the second server.

Based on the first of first aspect or first aspect into the third any possible implementation, at the 4th kind In possible implementation, the metadata of first data fragmentation further includes first data fragmentation in the Backup Data In distributing position information, the first server sends before metadata acquisition request, further includes: the first server connects Data recovery request is received, the data recovery request includes that the Backup Data mark and target disk mark, the data are extensive Multiple request, which is used to indicate, restores the Backup Data to the target disk.The first server saves first data Fragment, comprising: distributing position information of the first server according to first data fragmentation in the Backup Data, it will First data fragmentation is saved to the target disk.

Second aspect, provides a kind of data reconstruction method, and the method is executed by first server.The first server Metadata acquisition request is sent, the metadata acquisition request includes Backup Data mark, and the Backup Data includes the second number According to fragment.Then, the first server receives the metadata for second data fragmentation that second server is sent.Described The metadata of two data fragmentations includes second indication information, and the second indication information is second data fragmentation described the The mark or storage location of the Backup Data fragment of one server.The first server is according to the member of second data fragmentation Whether first server described in data validation preserves second data fragmentation in the Backup Data point of the first server Piece.Second data fragmentation is preserved in the Backup Data fragment of the first server in the confirmation first server When, the first server second data fragmentation according to the metadata lookup of second data fragmentation is in first clothes The Backup Data fragment of business device.Then.The first server saves second data fragmentation in the first service The Backup Data fragment of device.

In data reconstruction method provided in an embodiment of the present invention, what it is due to meta-data preservation is second data fragmentation The Backup Data fragment of first server mark or second data fragmentation first server Backup Data fragment Storage location, mark or second data fragmentation of second data fragmentation in the Backup Data fragment of first server It is small relative to the data volume of cryptographic Hash in the data volume of the storage location of the Backup Data fragment of first server.Therefore, relatively A large amount of cryptographic Hash is transmitted from second server in existing, transmits network used in a large amount of metadata from second server Resource is few, and propagation delay time is short, improves data recovering efficiency.Moreover, first server receives largely in the embodiment of the present invention Metadata after, expend that a large amount of spatial cache storage of the first server is a large amount of to breathe out relative to existing first server Uncommon value, it is few to store cache resources used in a large amount of metadata.

Based on second aspect, in the first possible implementation, the first server sends metadata acquisition and asks Before asking, the method also includes: the first server saves the copy of second data fragmentation as second number According to fragment the first server Backup Data fragment.The first server generates first number of second data fragmentation The metadata of second data fragmentation and second data fragmentation is sent to the second service according to, the first server Device.

Based on the possible implementation of the first of second aspect or second aspect, in second of possible implementation In, it is described whether the first server preserves according to the metadata of second data fragmentation confirmation first server Backup Data fragment of second data fragmentation in the first server, comprising: the first server is according to second clothes The metadata for second data fragmentation that business device is sent, confirms whether the first server preserves second data point The metadata of piece, wherein the first server is by confirming that the first server preserves second data fragmentation Metadata confirms that the first server preserves backup of the metadata in the first server of second data fragmentation Data fragmentation.

The first or second of possible implementation based on second aspect or second aspect, in the third possible reality In existing mode, the metadata of second data fragmentation further includes distribution of second data fragmentation in the Backup Data Location information, the first server are sent before metadata acquisition request, further includes: it is extensive that the first server receives data Multiple request, the data recovery request include that the Backup Data mark and target disk mark, the data recovery request are used The Backup Data is restored to the target disk in instruction.The first server saves second data fragmentation, packet It includes: distributing position information of the first server according to second data fragmentation in the Backup Data, by described Two data fragmentations are saved to the target disk.

It is any possible into the third possible implementation based on the first of second aspect or second aspect Implementation, in the fourth possible implementation, if the second indication information is second data fragmentation in institute The storage location of the Backup Data fragment of first server is stated, then the first server is according to the member of second data fragmentation Backup Data fragment of second data fragmentation described in data search in the first server, comprising: the first server root According to second data fragmentation in the storage location of the Backup Data fragment of the first server, second data point are searched Piece takes in second data fragmentation that the storage location of the Backup Data fragment of the first server is directed toward described first The Backup Data fragment of business device.

It is any possible into the 4th kind of possible implementation based on the first of second aspect or second aspect Implementation, in a fifth possible implementation, if the second indication information is second data fragmentation in institute The mark of the Backup Data fragment of first server is stated, the first server pre-saves second data fragmentation described The mark of the Backup Data fragment of first server and second data fragmentation divide in the Backup Data of the first server The corresponding relationship of the storage location of piece.The first server is second according to the metadata lookup of second data fragmentation Backup Data fragment of the data fragmentation in the first server, comprising: the first server is believed according to second instruction Breath searches the mark of Backup Data fragment of second data fragmentation in the first server；The first server root According to second data fragmentation in the mark of the Backup Data fragment of the first server, searches second data fragmentation and exist The storage location of the Backup Data fragment of the first server；The first server is according to second data fragmentation in institute The storage location of the Backup Data fragment of first server is stated, searches second data fragmentation in the standby of the first server Backup Data fragment of second data fragmentation that the storage location of part data fragmentation is directed toward in the first server.

The third aspect, provides a kind of data reconstruction method, and the method is executed by second server.The second server The metadata acquisition request that first server is sent is received, the metadata acquisition request includes Backup Data mark, described standby Part data include the first data fragmentation.The second server first data fragmentation according to the Backup Data identifier lookup Metadata.The second server sends the metadata of first data fragmentation to the first server, and described first The metadata of data fragmentation includes the first instruction information, and the first instruction information is first data fragmentation described first The mark or storage location of the Backup Data fragment of server.The first server is not protected in the confirmation first server First data fragmentation is deposited in the Backup Data fragment of the first server, the first data fragmentation acquisition is sent and asks It asks to the second server, the first data fragmentation acquisition request includes the metadata of first data fragmentation.It is described Second server receives the first data fragmentation acquisition request that the first server is sent.The second server according to The first data fragmentation acquisition request inquires first data fragmentation Backup Data in the second server point Piece.The second server sends Backup Data fragment of first data fragmentation in the second server to described One server.

Based on the third aspect, in the first possible implementation, the second server receives first server hair Before the metadata acquisition request sent, the method also includes: the second server receives what the first server was sent The metadata of first data fragmentation and first data fragmentation is to the second server.The second server saves Backup Data fragment of the copy of first data fragmentation as first data fragmentation in the second server, with And the second server saves the metadata of first data fragmentation.

Based on the possible implementation of the first of the third aspect or the third aspect, in second of possible implementation In, the method also includes: the second server receives target data fragment, the target data that the first server is sent The metadata of fragment.The second server saves the metadata of the target data fragment and the target data fragment, institute After target data fragment is stated to modify after first server reception data modification request to first data fragmentation Target data fragment.It is mesh that the data modification request, which is used to indicate the first server and modifies first data fragmentation, Data fragmentation is marked, the metadata of the target data fragment is the copy that the first server saves the target data fragment The metadata of the target data fragment of Shi Shengcheng.The metadata of the target data fragment includes Target indication information, institute State the mark or storage that Target indication information is Backup Data fragment of the target data fragment in the first server Position.The copy for the target data fragment that the first server saves is as the target data fragment described first Backup Data fragment in server.

Fourth aspect, provides a kind of data reconstruction method, and the method is executed by second server.The second server The metadata acquisition request that first server is sent is received, the metadata acquisition request includes Backup Data mark, described standby Part data include the second data fragmentation.The second server second data fragmentation according to the Backup Data identifier lookup Metadata.The second server sends the metadata of second data fragmentation to the first server, and described second The metadata of data fragmentation includes second indication information.The second indication information is second data fragmentation described first The mark or storage location of the Backup Data fragment of server.The first server is confirming first server preservation institute The second data fragmentation is stated in the Backup Data fragment of the first server, is looked into according to the metadata of second data fragmentation It looks for second data fragmentation in the Backup Data fragment of the first server, and saves second data fragmentation in institute State the Backup Data fragment of the first server.

Based on fourth aspect, in the first possible implementation, the second server receives first server hair Before the metadata acquisition request sent, the method also includes: the second server receives what the first server was sent The metadata of second data fragmentation and second data fragmentation is to the second server.The second server saves Backup Data fragment of the copy of second data fragmentation as second data fragmentation in the second server, with And the second server saves the metadata of second data fragmentation.

5th aspect, provides a kind of local server, comprising: can for executing first aspect or any of first aspect It is able to achieve each unit of the data reconstruction method in mode, the unit can also pass through hardware by hardware realization Execute corresponding software realization.

6th aspect, provides a kind of local server, comprising: can for executing second aspect or any of second aspect It is able to achieve each unit of the data reconstruction method in mode, the unit can also pass through hardware by hardware realization Execute corresponding software realization.

7th aspect, provides a kind of local server, including processor, memory and interface, and the interface is used for and the Two server communications, the memory include instruction for storing computer program code, the computer program code, work as institute When stating processor execution described instruction, the local server executes any side of being able to achieve of first aspect or first aspect Data reconstruction method in formula.

Eighth aspect, provides a kind of local server, including processor, memory and interface, and the interface is used for and the Two server communications, the memory include instruction for storing computer program code, the computer program code, work as institute When stating processor execution described instruction, the local server executes any side of being able to achieve of second aspect or second aspect Data reconstruction method in formula.

9th aspect, provides a kind of remote server, comprising: can for executing the third aspect or any of the third aspect It is able to achieve each unit of the data reconstruction method in mode, the unit can also pass through hardware by hardware realization Execute corresponding software realization.

Tenth aspect, provides a kind of remote server, comprising: can for executing fourth aspect or any of fourth aspect It is able to achieve each unit of the data reconstruction method in mode, the unit can also pass through hardware by hardware realization Execute corresponding software realization.

On the one hand tenth, provides a kind of remote server, including processor, memory and interface, the interface be used for First server communication, the memory include instruction for storing computer program code, the computer program code, when When the processor executes described instruction, the remote server is used to execute any possibility of the third aspect or the third aspect Data reconstruction method in implementation.

12nd aspect, provides a kind of remote server, including processor, memory and interface, the interface be used for First server communication, the memory include instruction for storing computer program code, the computer program code, when When the processor executes described instruction, the remote server is used to execute any possibility of fourth aspect or fourth aspect Data reconstruction method in implementation.

13rd aspect provides a kind of data recovery system, including first server and second server, first clothes Business device is the local server in any possible implementation that the 5th aspect or the 7th aspect provide, the second server The remote server in any possible implementation provided for the 9th aspect or the tenth one side.

Fourteenth aspect provides a kind of data recovery system, including first server and second server, first clothes Business device is the local server in any possible implementation that the 6th aspect or eighth aspect provide, the second server The remote server in any possible implementation provided for the tenth aspect or the 12nd aspect.

Detailed description of the invention

In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment Attached drawing be briefly described.

Fig. 1 is a kind of configuration diagram of standby system provided in an embodiment of the present invention；

Fig. 2 is a kind of existing schematic diagram of data backup process；

Fig. 3 is a kind of data back up method schematic diagram provided in an embodiment of the present invention；

Fig. 4 is a kind of structural schematic diagram of first server 110 provided in an embodiment of the present invention；

Fig. 5 is a kind of structural schematic diagram of second server 120 provided in an embodiment of the present invention；

Fig. 6 is a kind of flow chart of data reconstruction method provided in an embodiment of the present invention；

Fig. 7 is the flow chart of another data reconstruction method provided in an embodiment of the present invention；

Fig. 8 a is a kind of flow chart of magnetic disk of virtual machine data provided in an embodiment of the present invention full backup for the first time；

Fig. 8 b is the modified incremental backup of data fragmentation in a kind of magnetic disk of virtual machine data provided in an embodiment of the present invention Flow chart；

Fig. 8 c is the flow chart of another data reconstruction method provided in an embodiment of the present invention；

Fig. 9 a is a kind of 110 full backup schematic diagram of first server provided in an embodiment of the present invention；

Fig. 9 b is a kind of 120 full backup schematic diagram of second server provided in an embodiment of the present invention；

Figure 10 a is a kind of 110 incremental backup schematic diagram of first server provided in an embodiment of the present invention；

Figure 10 b is a kind of local incremental backup schematic diagram deleted after stale data provided in an embodiment of the present invention；

Figure 10 c is a kind of 120 incremental backup schematic diagram of second server provided in an embodiment of the present invention；

Figure 11 is a kind of structure chart of local server provided in an embodiment of the present invention；

Figure 12 is the structure chart of another local server provided in an embodiment of the present invention.

Specific embodiment

Below in conjunction with attached drawing, technical scheme in the embodiment of the invention is clearly and completely described.

Referring to Figure 1, Fig. 1 is a kind of configuration diagram of standby system provided in an embodiment of the present invention.The backup system System 100 realizes the backup of local backup and second server 120, and standby system 100 includes first server 110 and second service Device 120, first server 110 and second server 120 pass through network connection.In embodiments of the present invention, the first service Device 110 is used for local backup, and the second server 120 is used for remote backup.

User is communicated by client and standby system 100.

Client can be physical server or various types of terminal devices.The terminal device packet of the embodiment of the present invention Include tablet computer, laptop, mobile internet device, palm PC, desktop computer, mobile phone or other product form Terminal device.

Client is also possible to software module, such as can be the software module run on the physical devices or operate in Virtual machine on physical server.

A kind of process of data backup of the introduction about the construction of standby system 100 first as shown in connection with fig. 1.

In embodiments of the present invention, Backup Data is illustrated with magnetic disk of virtual machine data instance.

Fig. 2 is referred to, Fig. 2 is a kind of existing schematic diagram of data backup process.As shown in Fig. 2, standby system 100 is real When existing data backup, the backup of local magnetic disk of virtual machine data is realized by first server 110 and second server 120. Specifically, magnetic disk of virtual machine data are divided into multiple data fragmentations by first server 110, every number is calculated by hash algorithm According to the cryptographic Hash of fragment, Hash table is generated, Hash table includes the cryptographic Hash for obtain after Hash operation to data fragmentation.First After server 110 calculates the cryptographic Hash of each data fragmentation by hash algorithm, the copy of each data fragmentation and every can be saved The cryptographic Hash of a data fragmentation, the local backup as magnetic disk of virtual machine data.

In data back up method shown in Fig. 2, after first server 110 calculates the cryptographic Hash of each data fragmentation, Cryptographic Hash to the second server 120 for sending each data fragmentation and each data fragmentation in magnetic disk of virtual machine data is protected It deposits, using the backup of the second server 120 as magnetic disk of virtual machine data.

Relative to data back up method shown in Fig. 2, another data back up method and data back up method shown in Fig. 2 It is similar, and the partition virtual machines data in magnetic disk unlike data back up method shown in Fig. 2 is multiple data fragmentations, Yi Jitong The executing subject for crossing the cryptographic Hash that hash algorithm calculates data fragmentation is different, i.e., is divided by second server 120 from first service The magnetic disk of virtual machine data that device 110 is sent, and the Kazakhstan of each data fragmentation is calculated by second server 120 by hash algorithm Uncommon value, generates Hash table.After second server 120 calculates the cryptographic Hash of each data fragmentation by hash algorithm, it can save every The cryptographic Hash of the copy of a data fragmentation and each data fragmentation, second server 120 as magnetic disk of virtual machine data it is standby Part.After second server 120 calculates the cryptographic Hash of each data fragmentation by hash algorithm, each data fragmentation and every can be sent The cryptographic Hash of a data fragmentation is to first server 110.First server 110 can save the copy of each data fragmentation and each The cryptographic Hash of data fragmentation, the local backup as magnetic disk of virtual machine data.

Data back up method shown in Fig. 2 or another data backup similar with data back up method shown in Fig. 2 In method, any server required in the first server 110 and second server 120 in standby system 100 divides void Quasi- machine data in magnetic disk is complicated for operation to carry out Hash operation to each data fragmentation by hash algorithm after multiple data fragmentations, Expend process resource.After the cryptographic Hash for calculating each data fragmentation, first server 110 and second server 120 distribute respectively Memory space storage occupies the biggish cryptographic Hash of byte space, consumes the storage resource of local and second server 120.

Relative to above-mentioned data back up method shown in Fig. 2 or another kind similar with data back up method shown in Fig. 2 Data back up method, the embodiment of the invention provides a kind of different data back up method.Fig. 3 is referred to, Fig. 3 is that the present invention is real A kind of data back up method schematic diagram of example offer is provided.As shown in figure 3, in data back up method provided in an embodiment of the present invention, After first server 110 receives the data backup requests of client transmission, stand-by program is called to draw magnetic disk of virtual machine data It is divided into several data fragmentations.Secondly, the first server 110 saves the copy of each data fragmentation as the data The local backup of fragment, i.e., as the data fragmentation in local Backup Data fragment.Then, the first server 110 One metadata is generated to each data fragmentation.Wherein, the metadata of the data fragmentation includes instruction information, the finger Show information be the data fragmentation in the mark of local Backup Data fragment or the data fragmentation in local Backup Data The storage location of fragment.Then, the first server 110 sends the metadata of each data fragmentation to second server 120. The second server 120 saves the copy of each data fragmentation as the second server 120 of the data fragmentation Backup, i.e., as the data fragmentation second server 120 Backup Data fragment.Such as shown in Fig. 3, magnetic disk of virtual machine number According to including the first data fragmentation and the second data fragmentation.In the embodiment of the present invention, meta-data preservation is that the data fragmentation exists The mark of local Backup Data fragment or the data fragmentation local Backup Data fragment storage location, it is occupied Memory space is small.Relative in backup method shown in Fig. 2 or similar with data back up method shown in Fig. 2, it is local and Second server 120 requires to save cryptographic Hash, and the byte space that a cryptographic Hash occupies is at least 40 bytes.In virtual machine The data volume of data in magnetic disk is big, and in the case that the quantity of the data fragmentation marked off is more, local and second server 120 needs to divide The cryptographic Hash of each data fragmentation is saved with more memory spaces.Therefore, if using it is shown in Fig. 2 or with it is shown in Fig. 2 The similar backup method backup virtual machine data in magnetic disk of data back up method consumes depositing for local and second server 120 Store up resource.In addition, after being divided into data fragmentation to magnetic disk of virtual machine data, not having to each data point in the embodiment of the present invention Piece carries out Hash operation processing, easy to operate, avoids the process resource for expending first server 110, improves data backup effect Rate.

Below with reference to the structure of standby system 100 shown in FIG. 1, data backup schematic diagram shown in Fig. 2 and shown in Fig. 3 The data back up method of the embodiment of the present invention introduces the process of data recovery.

The data back up method that schematic diagram based on above-mentioned data backup shown in Fig. 2 provides, the first server 110 After backup virtual machine data in magnetic disk, for the memory space for saving the first server 110, the first server 110 can determine Phase delete the expired Backup Data fragment in the first server 110 or data fragmentation modification before it is described first clothes The Backup Data fragment of business device 110, while will be deleted the cryptographic Hash of data fragmentation expired in the backup of first server 110 Or the cryptographic Hash of the data fragmentation before modification.Therefore, before restoring data, for first clothes of magnetic disk of virtual machine data Be engaged in device 110 in backup, the first server 110 may only save partial data fragment in the first server 110 backup or the backup in the first server 110 without saving any data fragmentation.

So being based on data back up method shown in Fig. 2, subsequent when restoring data, first server 110 can be received The data recovery request that client is sent.Then, first server 110 obtains magnetic disk of virtual machine data from second server 120 In each data fragmentation cryptographic Hash.If the first server 110 inquires the Hash that the first server 110 saves Include the cryptographic Hash of data fragmentation in table, then data fragmentation is read from the first server 110, if the first service Do not include the cryptographic Hash of data fragmentation in the Hash table that device 110 saves, then reads data fragmentation from second server 120.Due to The data volume of each cryptographic Hash is big, if after receiving the data recovery request that client is sent, transmitted from second server 120 A large amount of cryptographic Hash can expend Internet resources, and propagation delay time is long, and data recovering efficiency is low.Moreover, the first server After 110 receive a large amount of cryptographic Hash of the transmission of second server 120, it is desirable to provide enough spatial cache cachings are a large amount of to be breathed out Uncommon value, expends the cache resources of the first server 110.Based on backup side similar with data back up method shown in Fig. 2 Method there is a problem of same when restoring data.

For the deficiency of existing data reconstruction method, the embodiment of the present invention provides a kind of different data reconstruction method, Data reconstruction method provided in an embodiment of the present invention is realized based on data back up method shown in Fig. 3.Implement in the present invention In the data reconstruction method that example provides, after first server 110 receives the data recovery request of client transmission, first number is sent According to acquisition request to second server 120.Wherein, data recovery request includes Backup Data mark and target disk mark.Such as Fruit is to restore magnetic disk of virtual machine data instance explanation, then Backup Data is identified as magnetic disk of virtual machine Data Identification.And metadata Acquisition request includes magnetic disk of virtual machine Data Identification.Data recovery request instruction restores magnetic disk of virtual machine data to target disk. Magnetic disk of virtual machine data include several data fragmentations, for example including the first data fragmentation shown in Fig. 3 or the second data fragmentation.Institute State the metadata that first server 110 receives each data fragmentation that second server 120 is sent.Wherein, the data fragmentation Metadata include instruction information, it is described instruction information be the data fragmentation the first server 110 Backup Data The storage location of the mark of fragment or the data fragmentation in the Backup Data fragment of the first server 110.Then, described First server 110 confirms whether the first server 110 preserves the data fragmentation in the first server 110 Backup Data fragment does not save the data fragmentation in the first server 110 in the confirmation first server 110 When Backup Data fragment, the first server 110 sends the data fragmentation acquisition request to second server 120, with from Second server 120 obtains the Backup Data fragment that the first server 110 does not save.Specifically, the first service Whether device 110 can preserve the metadata confirmation described first of the data fragmentation by the determination first server 110 Whether server 110 preserves the data fragmentation in the Backup Data fragment of the first server 110.Specifically, described First server 110 confirms described the by confirming that the first server 110 does not save the metadata of the data fragmentation One server 110 does not preserve the data fragmentation in the Backup Data fragment of the first server 110.First clothes When business device 110 confirms that the first server 110 does not save the metadata of the data fragmentation, sent out to second server 120 Send the data fragmentation acquisition request.Specifically, data fragmentation acquisition request includes the metadata of data fragmentation.Second server After 120 receive data fragmentation acquisition request, data fragmentation is in second server according to the metadata lookup of data fragmentation 120 Backup Data fragment, find second server 120 preservation the data fragmentation Backup Data fragment after, send The Backup Data fragment of the data fragmentation is to the first server 110.The first server 110 receives second server After the Backup Data fragment of 120 data fragmentations sent, the Backup Data fragment of the data fragmentation is saved to target disk.Separately Outside, if the first server 110 confirms that the first server 110 preserves the data fragmentation in the first service The Backup Data fragment of device 110 or the metadata of the data fragmentation, then the first server 110 is according to the data fragmentation Metadata obtain the data fragmentation in the Backup Data fragment of the first server 110.

In data reconstruction method provided in an embodiment of the present invention, what it is by meta-data preservation is the data fragmentation in institute State the Backup Data fragment of first server 110 mark or the data fragmentation the first server 110 backup number According to the storage location of fragment, mark or the number of the data fragmentation in the Backup Data fragment of the first server 110 According to fragment the storage location of the Backup Data fragment of the first server 110 data volume of the data volume relative to cryptographic Hash It is small.A large amount of cryptographic Hash is transmitted from second server 120 accordingly, with respect to existing, is transmitted largely from second server 120 Internet resources used in metadata are few, and propagation delay time is short, improve data recovering efficiency.Moreover, in the embodiment of the present invention, institute It states after first server 110 receives a large amount of metadata, to expend the first server 110 largely slow relative to existing It deposits space and stores a large amount of cryptographic Hash, it is few to store cache resources used in a large amount of metadata.

The knot of first server 110 in the embodiment of the present invention is described below based on the structure of standby system shown in FIG. 1 Structure.

Fig. 4 is referred to, Fig. 4 is a kind of structural schematic diagram of first server 110 provided in an embodiment of the present invention.Such as Fig. 4 Shown, first server 110 includes controller 410 and storage array 420.Controller 410 and storage array 420 connect.

The structure of first server 110 below with reference to the structure of standby system shown in FIG. 1, and based on shown in Fig. 4 The effect of each component part of first server 110 is described.

Storage array 420, for saving the Backup Data for needing to back up in backup procedure, e.g. magnetic disk of virtual machine data Backup Data, i.e. the backup in the first server 110 of magnetic disk of virtual machine data can claim in embodiments of the present invention For the metadata of each data fragmentation in the local backup or magnetic disk of virtual machine data of the first server 110.Virtual machine The metadata of each data fragmentation also can store in controller 410 in data in magnetic disk, for every in magnetic disk of virtual machine data The storage location of the metadata of a data fragmentation, however it is not limited to implementation provided in an embodiment of the present invention.

In first server 110, controller 410, for receiving the data recovery request of client transmission, to virtual machine Data in magnetic disk is restored.It include that magnetic disk of virtual machine Data Identification and target disk identify in data recovery request.Data are restored Request instruction restores magnetic disk of virtual machine data to target disk.Magnetic disk of virtual machine data include several data fragmentations.

Target disk identifies the Chinese character such as can be " creation data " or " business datum ", is also possible to virtual The disk of machine identifies, and is also possible to letter, number or other symbols, is also possible to the combination of letter, number or other symbols.Tool Body realization is not limited by the present embodiment.

Target disk is used for the magnetic disk of virtual machine of storage virtual machine data in magnetic disk before can be, be also possible to other magnetic Disk.Target disk can be deployed in storage array 420 or in first server 110, can also be deployed in other physical equipments In, e.g. other storage equipment, others storage equipment can be other storage arrays for being different from storage array 420.

Controller 410 is also used to send metadata acquisition request to second server 120.Metadata acquisition request includes Magnetic disk of virtual machine Data Identification.

Second server 120 pre-saves the corresponding of the metadata of magnetic disk of virtual machine Data Identification and data fragmentation and closes System.The metadata of the data fragmentation includes instruction information, and the instruction information is the data fragmentation in the first service The mark of the Backup Data fragment of device 110 or the data fragmentation are deposited in the Backup Data fragment of the first server 110 Storage space is set.

Second server 120, after receiving metadata acquisition request, for obtaining number according to magnetic disk of virtual machine Data Identification According to the metadata of fragment, the metadata of data fragmentation is sent to first server 110.

Controller 410 is also used to receive the metadata of the data fragmentation of the transmission of second server 120.

Controller 410 receives in magnetic disk of virtual machine data after the metadata of each data fragmentation, is also used to confirm described Whether first server 110 preserves the metadata of the data fragmentation, does not save institute in the confirmation first server 110 When stating the metadata of data fragmentation, the data fragmentation acquisition request is sent to second server 120.Data fragmentation acquisition request Metadata including data fragmentation.

Second server 120, fragment acquisition request, divides according to the metadata lookup data of data fragmentation for receiving data Piece, find second server 120 preservation the data fragmentation Backup Data fragment after, send the standby of the data fragmentation Part data fragmentation is to first server 110.

Controller 410 is also used to receive the Backup Data fragment of the data fragmentation of the transmission of second server 120, saves institute The Backup Data fragment of data fragmentation is stated to target disk.

When confirming that the first server 110 does not save the metadata of data fragmentation the above description is that controller 410, How controller 410 realizes the method that data are restored, and the first server 110 is described below and preserves the data fragmentation When metadata, how controller 410 realizes the method that data are restored.

If the first server 110 preserves the metadata of the data fragmentation, controller 410 receives data After the metadata of fragment, when confirming that the first server 110 preserves the metadata of the data fragmentation, it is also used to according to institute The metadata for stating data fragmentation obtains the Backup Data fragment of the first server 110.For example, controller 410 can be from depositing It stores up and obtains data fragmentation in array 420 in the Backup Data fragment of the first server 110.

As shown in figure 4, controller 410 includes interface 411 and interface 412, interface 411 is used for and client communication.Interface 412 with storage array 420 for communicating.Interface 411, for receiving the data recovery request of client transmission.

Controller 410 further includes control module 430, and control module 430 is used for according to the received client of the interface 411 The data recovery request of transmission restores magnetic disk of virtual machine data.Specifically, including virtual machine magnetic in data recovery request Disk Data Identification and target disk mark.Data recovery request instruction restores magnetic disk of virtual machine data to target disk.Virtual machine Data in magnetic disk includes several data fragmentations.

Control module 430 is also used to send metadata acquisition request to second server 120.Metadata acquisition request packet Include magnetic disk of virtual machine Data Identification.

Control module 430 is also used to receive the metadata of the data fragmentation of the transmission of second server 120.The data point The metadata of piece includes instruction information, and the instruction information is backup number of the data fragmentation in the first server 110 According to fragment mark or the data fragmentation the Backup Data fragment of the first server 110 storage location.

Control module 430 receives in magnetic disk of virtual machine data after the metadata of each data fragmentation, is also used to confirm institute The metadata whether first server 110 preserves the data fragmentation is stated, is not saved in the confirmation first server 110 When the metadata of the data fragmentation, the data fragmentation acquisition request is sent to second server 120.Data fragmentation acquisition is asked Seek the metadata including data fragmentation.

Control module 430 is also used to receive the data fragmentation of the transmission of second server 120, saves the data fragmentation extremely Target disk.

The above description is that control modules 430 to confirm that the first server 110 does not save the metadata of data fragmentation When, how control module 430 realizes the method that data are restored, and the first server 110 is described below and preserves the data When the metadata of fragment, how control module 430 realizes the method that data are restored.

If the first server 110 preserves the metadata of the data fragmentation, control module 430 receives number After the metadata of fragment, when confirming that the first server 110 preserves the metadata of the data fragmentation, it is also used to root The Backup Data fragment of the first server 110 is obtained according to the metadata of the data fragmentation.For example, control module 430 can To obtain the Backup Data fragment of the first server 110 from storage array 420.

As shown in figure 4, control module 430 includes processor 413 and memory 414.Processor 413 and memory 414 connect It connects.

Processor 413 restores magnetic disk of virtual machine data for receiving the data recovery request of client transmission. Specifically, including that magnetic disk of virtual machine Data Identification and target disk identify in data recovery request.Data recovery request instruction is extensive Magnetic disk of virtual machine data are answered to target disk.Magnetic disk of virtual machine data include several data fragmentations.

Processor 413 is also used to send metadata acquisition request to second server 120.Metadata acquisition request includes Magnetic disk of virtual machine Data Identification.

Processor 413 is also used to receive the metadata of the data fragmentation of the transmission of second server 120.The data fragmentation Metadata include instruction information, it is described instruction information be the data fragmentation the first server 110 Backup Data The storage location of the mark of fragment or the data fragmentation in the Backup Data fragment of the first server 110.

Processor 413 receives in magnetic disk of virtual machine data after the metadata of each data fragmentation, is also used to confirm described Whether first server 110 preserves the metadata of the data fragmentation, does not save institute in the confirmation first server 110 When stating the metadata of data fragmentation, the data fragmentation acquisition request is sent to second server 120.Data fragmentation acquisition request Metadata including data fragmentation.

Processor 413 is also used to receive the data fragmentation of the transmission of second server 120, saves the data fragmentation to mesh Mark disk.

When confirming that the first server 110 does not save the metadata of data fragmentation the above description is that processor 413, How processor 413 realizes the method that data are restored, and the first server 110 is described below and preserves the data fragmentation When metadata, how processor 413 realizes the method that data are restored.

If the first server 110 preserves the metadata of the data fragmentation, processor 413 receives data After the metadata of fragment, when confirming that the first server 110 preserves the metadata of the data fragmentation, it is also used to basis The metadata of the data fragmentation obtains the Backup Data fragment of the first server 110.For example, processor 413 can be from The Backup Data fragment of the first server 110 is obtained in storage array 420.

Processor 413 can be any calculating device, can be general central processor (CPU), microprocessor, may be programmed Controller, application-specific integrated circuit (application-specific integrated circuit, ASIC) or one Or it is multiple for controlling the integrated circuit of above scheme program execution.Processor 413 is stored in memory by operation or execution Software program and/or application module in 414, and the data being stored in memory 414 are called, execute first server 110 various functions and processing data, to carry out integral monitoring to first server 110.In the concrete realization, as one kind Embodiment, processor 413 may include one or more CPU.

Memory 414, for temporarily storing the information sent from client or second server 120.Memory 414 is also used In storage software program (such as data recovery procedures and stand-by program shown in Fig. 3) and application module.Processor 413 passes through Operation is stored in the software program and application module of memory 414, thereby executing the various function application of first server 110 And data processing.

Memory 414 mainly includes storing program area and storage data area.Wherein, storing program area can store operation system Application program needed for system, at least one function, for example send response message etc.；Storage data area can store application program and wait locating The data of reason.Memory 414 may include volatile memory (Volatile Memory), for example, random access memory (Random-Access Memory, RAM)；The memory 414 also may include nonvolatile memory (non-volatile Memory), for example, read-only memory (Read-Only Memory, ROM), flash memory (Flash Memory), hard disk (Hard Disk Drive, HDD), solid state hard disk (Solid-State Drive, SSD), magnetic disk storage medium or other deposit Storage equipment or can be used in carry or store have instruction or data structure form desired program code and can be by net Any other medium of network equipment access, but not limited to this.

The knot of second server 120 in the embodiment of the present invention is described below based on the structure of standby system shown in FIG. 1 Structure.

Fig. 5 is referred to, Fig. 5 is a kind of structural schematic diagram of second server 120 provided in an embodiment of the present invention.Such as Fig. 5 Shown, second server 120 includes controller 510 and storage array 520.Controller 510 and storage array 520 connect.

Below with reference to the structure of standby system shown in FIG. 1, and the structure based on second server 120 shown in fig. 5 The effect of each component part of second server 120 is described.

After the metadata that second server 120 receives Backup Data and Backup Data, storage array 520, for protecting Deposit the Backup Data for needing to back up in backup procedure, the e.g. Backup Data of magnetic disk of virtual machine data, i.e. magnetic disk of virtual machine number According to second server 120 backup or magnetic disk of virtual machine data in each data fragmentation metadata.Magnetic disk of virtual machine number The metadata of each data fragmentation also can store in controller 510 in, for each data in magnetic disk of virtual machine data The storage location of the metadata of fragment, however it is not limited to implementation provided in an embodiment of the present invention.

In second server 120, controller 510, the metadata acquisition for receiving the transmission of first server 110 is asked It asks, the metadata acquisition request includes magnetic disk of virtual machine Data Identification.Magnetic disk of virtual machine data include by the virtual machine magnetic Several data fragmentations after the division of disk data.

Storage array 520 pre-saves the correspondence of the metadata of magnetic disk of virtual machine Data Identification and each data fragmentation Relationship.The metadata of the data fragmentation includes instruction information, and the instruction information is the data fragmentation in first clothes Be engaged in device 110 Backup Data fragment mark or the data fragmentation the first server 110 Backup Data fragment Storage location.

Controller 510 is also used to search the metadata of each data fragmentation according to magnetic disk of virtual machine Data Identification, sends The metadata of each data fragmentation is to first server 110.

First server 110 receives in magnetic disk of virtual machine data after the metadata of each data fragmentation, confirms described the Whether one server 110 preserves the metadata of the data fragmentation, is confirming that it is described that the first server 110 does not save When the metadata of data fragmentation, the data fragmentation acquisition request is sent to second server 120.Data fragmentation acquisition request packet Include the metadata of data fragmentation.

Controller 510 is also used to receive the data fragmentation acquisition request of the transmission of first server 110, according to data fragmentation Metadata search data fragmentation from storage array 520, find the data fragmentation in the backup number of second server 120 After fragment, the data fragmentation is sent in the Backup Data fragment of second server 120 to first server 110.

First server 110 receives Backup Data of the data fragmentation in second server 120 of the transmission of second server 120 After fragment, the data fragmentation is saved in the Backup Data fragment of second server 120 to target disk.

As shown in figure 5, controller 510 includes interface 511 and interface 512, interface 511 is used for and first server 110 is logical Letter.Interface 512 is used for and storage array 520 communicates.Interface 511, the metadata for receiving the transmission of first server 110 obtain Request.

Controller 510 further includes control module 530, control module 530, for according to the interface 511 received first The metadata acquisition request that server 110 is sent, according to magnetic disk of virtual machine Data Identification, by interface 512 from storage array 520 The middle metadata for searching each data fragmentation sends the metadata of each data fragmentation to first server 110.

Control module 530 is also used to receive the data fragmentation acquisition request of the transmission of first server 110, according to data point The metadata of piece searches data fragmentation from storage array 520, after finding the data fragmentation, sends the data fragmentation extremely First server 110.

As shown in figure 5, control module 530 includes processor 513 and memory 514.Processor 513 and memory 514 connect It connects.

Processor 513, the metadata acquisition request for being sent according to the received first server 110 of the interface 511, According to magnetic disk of virtual machine Data Identification, the metadata of each data fragmentation is searched from storage array 520, sends each data point The metadata of piece is to first server 110.

Processor 513 is also used to receive the data fragmentation acquisition request of the transmission of first server 110, according to data fragmentation Metadata search data fragmentation from storage array 520, find the data fragmentation in the backup number of second server 120 After fragment, the data fragmentation is sent in the Backup Data fragment of second server 120 to first server 110.

Processor 513 can be any calculating device, can be general central processor (CPU), microprocessor, may be programmed Controller, application-specific integrated circuit (application-specific integrated circuit, ASIC) or one Or it is multiple for controlling the integrated circuit of above scheme program execution.Processor 513 is stored in memory by operation or execution Software program and/or application module in 514, and the data being stored in memory 514 are called, execute first server 110 various functions and processing data, to carry out integral monitoring to first server 110.In the concrete realization, as one kind Embodiment, processor 513 may include one or more CPU.

Memory 514, for temporarily storing the information sent from client or second server 120.Memory 514 is also used In storage software program (such as searching metadata program and lookup data fragmentation program) and application module.Processor 513 is logical Software program and application module that operation is stored in memory 514 are crossed, is answered thereby executing the various functions of first server 110 With and data processing.

Memory 514 mainly includes storing program area and storage data area.Wherein, storing program area can store operation system Application program needed for system, at least one function, for example send response message etc.；Storage data area can store application program and wait locating The data of reason.Memory 514 may include volatile memory (Volatile Memory), for example, random access memory (Random-Access Memory, RAM)；The memory also may include nonvolatile memory (non-volatile Memory), for example, read-only memory (Read-Only Memory, ROM), flash memory (Flash Memory), hard disk (Hard Disk Drive, HDD), solid state hard disk (Solid-State Drive, SSD), magnetic disk storage medium or other deposit Storage equipment or can be used in carry or store have instruction or data structure form desired program code and can be by net Any other medium of network equipment access, but not limited to this.

Below with reference to the structure of standby system 100 shown in FIG. 1, the flow diagram of Backup Data shown in Fig. 3, Fig. 4 Shown in first server 110 structure and second server shown in fig. 5 120 the introduction about the construction embodiment of the present invention provide Data restore method.

Fig. 6 is referred to, Fig. 6 is a kind of flow chart of data reconstruction method provided in an embodiment of the present invention.As shown in fig. 6, In data reconstruction method provided in this embodiment, step S610, S640, S650 and S680 are by Fig. 1 or first service shown in Fig. 4 Device 110 executes, specifically can be by the control module 430 or control module shown in fig. 5 430 in first server 110 Reason device 413 executes and step S620, S630, S660 and S670 are executed by Fig. 1 or second server shown in fig. 5 120, has Body can be executed by the processor 513 in the control module 530 or control module shown in fig. 6 530 in second server 120. Below based on the flow chart of data reconstruction method shown in fig. 6, it is described in detail in magnetic disk of virtual machine data backup to described After one server 110 and second server 120, the first server 110 deletes expired data fragmentation or data fragmentation In the case where after modification, if the first server 110 does not save the metadata of data fragmentation to be restored, how to restore The method of magnetic disk of virtual machine data.The corresponding data fragmentation example to be restored of the unsaved metadata of the first server 110 It such as can be the first data fragmentation in magnetic disk of virtual machine data shown in Fig. 3.As shown in fig. 6, the method specifically include as Lower step.

S610, the first server 110 send metadata acquisition request.The metadata acquisition request includes virtual machine Data in magnetic disk mark.Magnetic disk of virtual machine data include several data fragmentations, such as a copy of it data fragmentation can be shown in Fig. 3 The first data fragmentation.

If necessary to restore the data in magnetic disk of virtual machine, magnetic disk of virtual machine can be identified by magnetic disk of virtual machine mark In magnetic disk of virtual machine data, virtual machine magnetic can also be identified by client magnetic disk of virtual machine Data Identification predetermined Magnetic disk of virtual machine data in disk.The present embodiment, which is merely exemplary, illustrates that Backup Data identifies, and does not identify to Backup Data Or other realizations of magnetic disk of virtual machine Data Identification have any restriction.

Magnetic disk of virtual machine mark or magnetic disk of virtual machine Data Identification for example can be " creation data " or " business datum " Equal Chinese characters are also possible to the disk mark of virtual machine, are also possible to letter, number or other symbols, be also possible to letter, The combination of number or other symbols.Specific implementation is not limited by the present embodiment.

Magnetic disk of virtual machine can be deployed in storage array 420 or in first server 110, can also be deployed in other objects It manages in equipment, e.g. other storage equipment, storage equipment can be other storage arrays for being different from storage array 420.

S620, second server 120 obtain the metadata of each data fragmentation according to magnetic disk of virtual machine Data Identification.

Second server 120 pre-saves the corresponding of the metadata of magnetic disk of virtual machine Data Identification and data fragmentation and closes System.

Second server 120 can save the metadata of each data fragmentation by metadata table.The data fragmentation Metadata includes instruction information, and the instruction information is Backup Data point of the data fragmentation in the first server 110 The storage location of the mark of piece or the data fragmentation in the Backup Data fragment of the first server 110.

A kind of optional implementation of metadata is that the metadata of each data fragmentation includes data fragmentation described The Backup Data fragment of the mark of the Backup Data fragment of one server 110 or data fragmentation in the first server 110 Storage location.

Optionally, another implementation of the metadata of each data fragmentation is the metadata packet of each data fragmentation Distributing position information and data fragmentation of the data fragmentation in Backup Data is included to divide in the Backup Data of the first server 110 The corresponding relationship of the mark of piece.Distributing position information of the data fragmentation in Backup Data is for describing the data fragmentation in institute State the distributing position in Backup Data.For example, based on the first data fragmentation and second in magnetic disk of virtual machine data shown in Fig. 3 The metadata table that data fragmentation is realized is as shown in table 1 below.

Table 1

Optionally, another implementation of the metadata of each data fragmentation is the metadata packet of each data fragmentation Distributing position information and data fragmentation of the data fragmentation in Backup Data is included to divide in the Backup Data of the first server 110 The corresponding relationship of the storage location of piece.Distributing position information of the data fragmentation in Backup Data is for describing the data fragmentation Distributing position in the Backup Data.

Optionally, the optional implementation of another kind of metadata is, metadata includes the data fragmentation described the Corresponding relationship mark before one server 110 backs up and after the first server 110 backup.Corresponding relationship mark can be word Female, number or other symbols, are also possible to the combination of letter, number or other symbols.Specific implementation is not by the limit of the present embodiment System.Corresponding relationship identify unique identification magnetic disk of virtual machine data in data fragmentation before the first server 110 backs up and Corresponding relationship after the backup of first server 110.For example, data fragmentation can be found in institute by corresponding relationship mark State the correspondence saved between address saved after address and the backup of the first server 110 before first server 110 backs up Relationship.Corresponding relationship mark based on the metadata that the first data fragmentation in magnetic disk of virtual machine data shown in Fig. 3 is realized, example It such as can be index0, the first data fragmentation in magnetic disk of virtual machine data can be found in first clothes by index0 Corresponding relationship before business device 110 backs up and after the first server 110 backup.For another example, it is based on magnetic disk of virtual machine shown in Fig. 3 The corresponding relationship mark for the metadata that the second data fragmentation in data is realized can be index1, can be searched by index1 To the second data fragmentation in magnetic disk of virtual machine data before the first server 110 backs up and the first server 110 Corresponding relationship after backup.Pass through preservation Address Confirmation backup of each data fragmentation before the first server 110 backup Distributing position information of each data fragmentation in the Backup Data in data.

S630, second server 120 send the metadata of data fragmentation to first server 110.

Second server 120 sends the metadata of each data fragmentation in the Backup Data to first server 110.

S640, the first server 110 confirm whether the first server 110 preserves the member of the data fragmentation Data.When confirming that the first server 110 does not save the metadata of the data fragmentation, step S650 is executed.For example, The first server 110 does not save the metadata of the first data fragmentation shown in Fig. 3, then confirms the first server 110 do not save first data fragmentation in the Backup Data fragment of the first server 110, then execute step S650。

First server 110 pre-saves the metadata corresponding relationship of magnetic disk of virtual machine Data Identification and data fragmentation.

First server 110 inquires the data point that the first server 110 saves according to magnetic disk of virtual machine Data Identification The metadata of piece.

The specific implementation for the metadata that first server 110 pre-saves can refer in above-mentioned steps S620, the The specific implementation for the metadata that two servers 120 save, details are not described herein.Optionally, if metadata includes described Corresponding relationship mark of the data fragmentation before the first server 110 backs up and after the first server 110 backup, then the One server 110 can be looked into after the metadata for getting each data fragmentation by the corresponding relationship mark that metadata includes Find the guarantor that saves address and the first server 110 backup after of the data fragmentation before the first server 110 backup Deposit the corresponding relationship between address, and then the preservation address by each data fragmentation before the first server 110 backup Confirm distributing position information of each data fragmentation in the Backup Data in Backup Data.

S650, the first server 110 send data fragmentation acquisition request.The data fragmentation acquisition request includes number According to the metadata of fragment.

The data fragmentation acquisition request is used to indicate second server 120 and searches data fragmentation in second server 120 Backup Data fragment.

S660, second server 120 are according to the metadata lookup data fragmentation of data fragmentation in the standby of second server 120 Part data fragmentation.

Optionally, the metadata and data fragmentation that second server 120 pre-saves data fragmentation are in second server The corresponding relationship of the storage location of 120 Backup Data fragment, then second server 120 is receiving data fragmentation acquisition request Afterwards, according to the metadata lookup data fragmentation of data fragmentation the Backup Data fragment of second server 120 storage location, so Data fragmentation is searched in second server in the storage location of the Backup Data fragment of second server 120 according to data fragmentation afterwards Backup Data fragment of the data fragmentation that the storage location of 120 Backup Data fragment is directed toward in second server 120.Second clothes Business device 120 is to receive and save the data fragmentation and data of the Backup Data of the transmission of first server 110 in second server 120 After the metadata of fragment, save data fragmentation metadata and data fragmentation second server 120 Backup Data fragment The corresponding relationship of storage location.

S670, second server 120 send data fragmentation second server 120 Backup Data fragment to first service Device 110.

S680, the first server 110 save the data fragmentation in the Backup Data fragment of second server 120.

Specifically, the first server 110 saves the data fragmentation in the Backup Data fragment of second server 120 To target disk.

Optionally, the metadata of the data fragmentation further includes distributing position letter of the data fragmentation in Backup Data Breath, distributing position information of the data fragmentation in Backup Data is for describing the data fragmentation in the Backup Data Distributing position, the first server 110 saves the data fragmentation, comprising: the first server 110 is according to the number The data fragmentation is saved according to distributing position of the fragment in the Backup Data, so that the number after Backup Data recovery Do not change according to the distributing position of fragment and the distributing position before backup.

Above-mentioned data reconstruction method shown in fig. 6 is described in detail the first server 110 and deletes expired or number In the case where the data fragmentation of the backup in the first server 110 before being modified according to fragment, the first server 110 The Data Recovery Process when metadata of data fragmentation is not saved.

Below with reference to the structure of standby system 100 shown in FIG. 1, the schematic diagram of Backup Data shown in Fig. 3, shown in Fig. 4 First server 110 structure and second server shown in fig. 5 120 structure, introduce provided in an embodiment of the present invention In data reconstruction method, if the first server 110 preserves first number of another data fragmentation in magnetic disk of virtual machine data According to how carrying out the process of data recovery.The metadata that the first server 110 is preserved is corresponding to be restored described another One data fragmentation for example can be the second data fragmentation in magnetic disk of virtual machine data shown in Fig. 3.

Specific process refers to Fig. 7, and Fig. 7 is the process of another data reconstruction method provided in an embodiment of the present invention Figure.As shown in fig. 7, step S710, S740, S750 and S760 are by Fig. 1 or Fig. 4 in data reconstruction method provided in this embodiment Shown in first server 110 execute, specifically can be by the control module 430 or control shown in fig. 5 in first server 110 Processor 413 in module 430 executes and step S720, S730 is executed by Fig. 1 or second server shown in fig. 5 120, It can specifically be held by the processor 513 in the control module 530 or control module shown in fig. 6 530 in second server 120 Row.Below based on the flow chart of data reconstruction method shown in Fig. 7, it is described in detail in magnetic disk of virtual machine data backup to described After first server 110 and second server 120, the first server 110 deletes expired data fragmentation or data point It is how extensive if the first server 110 preserves the metadata of data fragmentation to be restored in the case where after piece modification The method of multiple magnetic disk of virtual machine data, the first server 110 are preserved the data fragmentation to be restored and for example be can be The second data fragmentation in magnetic disk of virtual machine data shown in Fig. 3.As shown in fig. 7, the method specifically comprises the following steps.

S710, the first server 110 send metadata acquisition request.The metadata acquisition request includes virtual machine Data in magnetic disk mark.Magnetic disk of virtual machine data include several data fragmentations, such as a copy of it data fragmentation can be shown in Fig. 3 The second data fragmentation.

Magnetic disk of virtual machine Data Identification can be with reference to the associated description in step S610 shown in fig. 6, and detail is no longer It repeats at this.

S720, second server 120 obtain the metadata of data fragmentation according to magnetic disk of virtual machine Data Identification.

In this step S720, second server 120 obtain data fragmentation metadata and metadata content it is specific Realize details, can be with reference to the associated description in step S620 shown in fig. 6, detail is no longer repeated at this.

S730, second server 120 send the metadata of data fragmentation to first server 110.

S740, the first server 110 confirm whether the first server 110 preserves first number of data fragmentation According to.When confirming that the first server 110 preserves the metadata of the data fragmentation, step S750 is executed.For example, described First server 110 saves the metadata of the second data fragmentation shown in Fig. 3, then confirms that the first server 110 saves Then Backup Data fragment of first data fragmentation in the first server 110 executes step S750.

In this step S740, first server 110 confirms whether the first server 110 preserves the data fragmentation Metadata specific implementation details, can be with reference to the associated description in step S640 shown in fig. 6, detail is no longer at this It repeats.

S750, the first server 110 are obtained according to the metadata of the data fragmentation in the first server 110 The data fragmentation of middle backup.

If the metadata of the data fragmentation includes Backup Data of the data fragmentation in the first server 110 The mark of fragment, i.e. instruction information are mark of the data fragmentation in the Backup Data fragment of the first server 110, institute The data fragmentation can be pre-saved in the mark of the Backup Data fragment of the first server 110 by stating first server 110 Know the corresponding relationship with the data fragmentation in the storage location of the Backup Data fragment of the first server 110.Therefore, institute It states first server 110 and obtains the number backed up in the first server 110 according to the metadata of the data fragmentation It is that the first server 110 is according to data in the metadata of the data fragmentation point according to a kind of specific implementation of fragment Piece data fragmentation described in the identifier lookup of the Backup Data fragment of the first server 110 is in the first server 110 Backup Data fragment storage location.Then, the first server 110 is according to the data fragmentation in the first service Storage location after the backup of device 110 is searched storage location of the data fragmentation after the first server 110 backup and is directed toward The data fragmentation the backup in the first server 110.

If the metadata of the data fragmentation includes Backup Data of the data fragmentation in the first server 110 The storage location of fragment, i.e. instruction information are that the data fragmentation is deposited in the Backup Data fragment of the first server 110 Storage space is set, then the first server 110 is obtained according to the metadata of the data fragmentation and backed up in the first server 110 Another specific implementation of the data fragmentation be that the first server 110 searches institute according to the instruction information State taking described first for the data fragmentation that storage location of the data fragmentation after the first server 110 backup is directed toward The backup of business device 110.

If the metadata of the data fragmentation includes the data fragmentation before the first server 110 backs up and institute State the corresponding relationship mark after first server 110 backs up.Such as the corresponding relationship is identified as the data fragmentation described The corresponding pass saved between address saved after address and the backup of the first server 110 before the backup of first server 110 System.The first server 110 is according to the corresponding relationship identifier lookup to the data fragmentation in the first server 110 Preservation address after backup passes through preservation address search of the data fragmentation after the first server 110 backup to institute The data fragmentation that saves address be directed toward of the data fragmentation after the first server 110 backup is stated, the data fragmentation is in institute Stating the preservation address after first server 110 backs up is Backup Data fragment of the data fragmentation in the first server 110 Storage location.

S760, first server 110 save the data fragmentation.

Specifically, the first server 110 saves the data fragmentation to target disk.

The metadata of the data fragmentation further includes position of the data fragmentation before the first server 110 backup Information, location information of the data fragmentation before the first server 110 backup is for describing the data fragmentation in institute The distributing position in Backup Data is stated, the first server 110 saves the data fragmentation, comprising: the first server 110 save the data fragmentation according to distributing position of the data fragmentation in the Backup Data, so that the backup number Distributing position according to data fragmentation described after recovery does not change.

Below with reference to the structure of standby system 100 shown in FIG. 1, the schematic diagram of Backup Data shown in Fig. 3, shown in Fig. 4 First server 110 structure and second server shown in fig. 5 120 introduction about the construction number provided in an embodiment of the present invention According to another implementation of the method for recovery.

Fig. 8 a- Fig. 8 c is referred in data reconstruction method provided in this embodiment, to essentially describe as shown by figures 8 a-8 c Modify data fragmentation after how the process of Backup Data, and based on modification data fragmentation realize data backup after how to realize number According to the process of recovery.Wherein, Fig. 8 a essentially describes the process of magnetic disk of virtual machine data full backup for the first time, and Fig. 8 b is mainly described In magnetic disk of virtual machine data after the modification of any data fragment incremental backup process, Fig. 8 c essentially describes based on modification data The process that data are restored how is realized after the incremental backup of fragment realization data.Referring first to Fig. 8 a, wherein step S801 is extremely S806 is executed by Fig. 1 or first server shown in Fig. 4 110, specifically can by first server 110 control module 430 or Processor 413 in person's control module 430 shown in fig. 5 executes and step S807 is by Fig. 1 or second service shown in fig. 5 Device 120 executes, specifically can be by the control module 530 or control module shown in fig. 6 530 in second server 120 Device 513 is managed to execute.The method specifically comprises the following steps.

S801, the request of 110 backup reception data of first server.

First server 110 receives the data backup requests that client is sent.Data backup requests include Backup Data mark Know.

If illustrating Backup Data with magnetic disk of virtual machine data instance, data backup requests include magnetic disk of virtual machine data Mark.

S802, first server 110 search magnetic disk of virtual machine data according to the magnetic disk of virtual machine Data Identification.

It is several data fragmentations that S803, first server 110, which divide the magnetic disk of virtual machine data,.

Such as it is shown in Fig. 3, magnetic disk of virtual machine data include the first data fragmentation and the second data fragmentation.

S804, first server 110 save the copy of each data fragmentation, the first server as data fragmentation 110 backups.

S805, first server 110 generate the metadata of each data fragmentation.The metadata of data fragmentation includes instruction letter Breath, mark or the number of the instruction information for Backup Data fragment of the data fragmentation in the first server 110 According to fragment the Backup Data fragment of the first server 110 storage location.Such as first data fragmentation metadata packet The first instruction information is included, the first instruction information is first data fragmentation after the first server 110 backup The storage location of Data Identification or first data fragmentation after the first server 110 backup.For example, the second data point The metadata of piece includes second indication information, and the second indication information is second data fragmentation in the first server The storage location of Data Identification or second data fragmentation after the first server 110 backup after 110 backups.

Step S801 to step S805 describes the backup stream in the first server 110 of magnetic disk of virtual machine data Journey.Assuming that this time described first server 110 that the same day that first server 110 receives the data backup requests is done Backup be backup of the magnetic disk of virtual machine data in first day first server 110, then the current first server After 110 Backup end, such as Fig. 9 a of the backup instances in the first server 110 of first day Day1 of magnetic disk of virtual machine data It is shown.If magnetic disk of virtual machine data include n parts of data fragmentations, magnetic disk of virtual machine data in the first server 110 Backup include n part data fragmentations copy, metadata has n a.

S806, first server 110 send the metadata and every number of magnetic disk of virtual machine Data Identification, each data fragmentation According to fragment to second server 120.

S807, second server 120 save the corresponding relationship of magnetic disk of virtual machine Data Identification and each data fragmentation, and Save the corresponding relationship of the metadata of magnetic disk of virtual machine Data Identification and each data fragmentation.

Step S806 and step S807 describes magnetic disk of virtual machine data after the completion of the first server 110 backup The process of first day Day1 backup is realized in second server 120.The second server of first day Day1 of magnetic disk of virtual machine data 120 backup instances are as shown in figure 9b.Second server 120 receives and saves the member of each data fragmentation and each data fragmentation Data, and save the corresponding relationship of magnetic disk of virtual machine Data Identification and each data fragmentation.And second server 120 can also The corresponding relationship of the metadata of magnetic disk of virtual machine Data Identification and each data fragmentation is saved, but is not shown in figure 9b virtual Machine data in magnetic disk mark.

Secondly referring to Fig. 8 b, Fig. 8 b essentially describe based on after the full backup of magnetic disk of virtual machine data shown in Fig. 8 a such as How the modification of any data fragment realizes the process of incremental backup in fruit magnetic disk of virtual machine data.Wherein step S810 to S814 by Fig. 1 or first server shown in Fig. 4 110 execute, specifically can by first server 110 control module 430 or Fig. 5 institute Processor 413 in the control module 430 shown executes and step S815 is held by Fig. 1 or second server shown in fig. 5 120 Row, specifically can be by the processor 513 in the control module 530 or control module shown in fig. 6 530 in second server 120 It executes.The method specifically comprises the following steps.

Any number in magnetic disk of virtual machine data is modified after S810, the reception data modification request of the first server 110 It is target data fragment according to fragment, and saves the copy of target data fragment, as target data fragment in first clothes The backup of business device 110.The data modification request, which is used to indicate the first server 110, to be modified first data fragmentation and is Target data fragment

Such as based on magnetic disk of virtual machine data shown in Fig. 3, in this step S810, the first server 110 is modified The first data fragmentation in magnetic disk of virtual machine data is after third data fragmentation, first server 110 can save third data point The copy of piece, the backup in the first server 110 as third data fragmentation.

S811, first server 110 generate the metadata of target data fragment.The metadata of target data fragment includes referring to Show that information, the instruction information are Data Identification or institute of the target data fragment after the first server 110 backs up State storage location of the target data fragment after the first server 110 backup.Such as based on magnetic disk of virtual machine shown in Fig. 3 Data, in this step S820, first server 110 can generate the metadata of third data fragmentation.The member of third data fragmentation Data include third instruction information, and the third instruction information is that the third data fragmentation is standby in the first server 110 The storage location of Data Identification or the third data fragmentation after the first server 110 backup after part.

It is modified described that step S810 and step S811 describes the arbitrary data fragment in magnetic disk of virtual machine data The backup process of first server 110.

After arbitrary data fragment modification in magnetic disk of virtual machine data, first server 110 can be in any data fragmentation The copy that modified data fragmentation is backed up after modification, as modified data fragmentation in the standby of the first server 110 Part.

First server 110 can also daily back up magnetic disk of virtual machine data, at first day to magnetic disk of virtual machine After data carry out full backup, to magnetic disk of virtual machine data after the later data backup requests for receiving client transmission daily In have modification data fragmentation carry out incremental backup.Specifically, being carried out to the data fragmentation for having modification in magnetic disk of virtual machine data When backup, only confirmation has the data fragmentation of modification, and the copy of its modified data fragmentation is backed up to the data fragmentation for having modification, As modified data fragmentation the first server 110 backup.Client can receive the number of user's input daily It is pre-set according to backup request or client and periodically sends data backup requests daily to first server 110.

Based on the backup instances of first day Day1 of magnetic disk of virtual machine data shown in Fig. 9 a, if the first data fragmentation changed For third data fragmentation, then first server 110 may refer to shown in Figure 10 a the case where second day Day2 is backed up.First number After being revised as third data fragmentation according to fragment, the metadata of third data fragmentation is metadata 1 '.

S812, first server 110 delete data fragmentation and mistake expired in the backup of the first server 110 The metadata of the data fragmentation of phase.

Specific illustrative, if the arbitrary data fragment in magnetic disk of virtual machine data has modification, newest modified number The backup of the first server 110 according to fragment is effective, the first server 110 of the data fragmentation before newest modification Backup is expired data fragmentation.

Based on modified shown in Figure 10 a the first data fragmentation be third data fragmentation after the first server 110 second The backup of its Day2, the expired data fragmentation that first server 110 will be deleted 110 first days Day1 of first server exist The backup of the first server 110, i.e. the first data fragmentation are in the backup of the first server 110, i.e. the first data point Piece copy.And first server 110 can delete the metadata 1 of the first data fragmentation.

S813, data fragmentation and expired data fragmentation expired in the backup of the first server 110 are deleted After metadata, first server 110 is by the data fragmentation that do not delete in the backup of the first server 110 and modified The backup in the first server 110 of data fragmentation merges, as each in modified magnetic disk of virtual machine data Data fragmentation the first server 110 backup, and by the metadata for the data fragmentation that do not delete and modified The metadata of data fragmentation merges, the metadata as each data fragmentation in modified magnetic disk of virtual machine data.

Based on modified shown in Figure 10 a the first data fragmentation be third data fragmentation after the first server 110 second The backup of its Day2, after executing step S812 and step S813, the backup merging and metadata of the first server 110 Situation after merging is as shown in fig. lob.Be left in the backup of the first server 110 unmodified data fragmentation copy and The copy of the copy of modified data fragmentation, specific modified data fragmentation is the copy of third data fragmentation.And First server 110 preserves the metadata of unmodified data fragmentation and the metadata of modified data fragmentation, after modification Data fragmentation metadata be third data fragmentation metadata 1 '.

S814, first server 110 send each data in target data fragment and modified magnetic disk of virtual machine data The metadata of fragment is to second server 120.

The metadata of each data fragmentation includes the metadata of target data fragment in modified magnetic disk of virtual machine data, But not including that any data fragment be revised as target data fragment before any data fragment metadata.

The case where 120 second days Day2 backup target data fragmentations of second server, is as shown in figure l0c.Second day Day2, The metadata of each data fragmentation includes the metadata of third data fragmentation in modified magnetic disk of virtual machine data, but does not wrap Include the metadata of first data fragmentation.

S815, second server 120 save the second server 120 of magnetic disk of virtual machine Data Identification and target data fragment Backup corresponding relationship, and save each data in magnetic disk of virtual machine Data Identification and modified magnetic disk of virtual machine data The corresponding relationship of the metadata of fragment.

Referring then to Fig. 8 c, Fig. 8 c, which is essentially described, realizes that the increment of data is standby based on modification data fragmentation shown in Fig. 8 b How process that data restore is realized after the method for part.Wherein step S820, S821 and step S830 to S832, Yi Jibu Rapid S835 is executed by Fig. 1 or first server shown in Fig. 4 110, specifically can be by the control module 430 in first server 110 Or processor 413 in control module 430 shown in fig. 5 execute and step S822, S823, S833 and S834 by Fig. 1 or Second server 120 shown in fig. 5 executes, specifically can be by control module 530 in second server 120 or shown in fig. 6 Processor 513 in control module 530 executes.The method specifically comprises the following steps.

S820, the first server 110 receive data recovery request.

The first server 110 receives the data recovery request that client is sent.In this step S820, data are restored Request includes that magnetic disk of virtual machine Data Identification and target disk identify, and further includes recovery time.Recovery time indicates virtual machine magnetic The backup time of disk data.

For example, recovery time in the present embodiment by taking the backup time of magnetic disk of virtual machine data is second day Day2 as an example into Row explanation.

S821, the first server 110 send metadata acquisition request to second server 120.The metadata obtains Taking request includes magnetic disk of virtual machine Data Identification and recovery time.Magnetic disk of virtual machine data include several data fragmentations, such as its Middle portion data fragmentation can be the first data fragmentation shown in Fig. 3.

S822, second server 120 are according to magnetic disk of virtual machine Data Identification, when acquisition backup time is the recovery time The metadata for each data fragmentation that second server 120 saves.

Specific illustrative, recovery time is second day Day2, then second server 120 is according to magnetic disk of virtual machine data mark Know, obtains the metadata for each data fragmentation that second server 120 saves when backup time is second day Day2.

The case where Day2 was backed up in 120 second days for second server shown in Figure 10 c, magnetic disk of virtual machine data second day The metadata that Day2 is saved includes the metadata 1 ' of third data fragmentation, further includes first number of other data fragmentations that do not modify According to, the metadata of other data fragmentations that do not modify, such as the second data fragmentation each data fragmentation into the n-th data fragmentation Metadata, i.e. each metadata of the metadata 2 into metadata n.

When S823, the transmission backup time of second server 120 are the recovery time, the second server 120 is saved Each data fragmentation metadata to first server 110.

When S830, the inquiry backup time of first server 110 are the recovery time, first server 110 is saved every The metadata of a data fragmentation.

The case where Day2 was backed up in 110 second days for first server shown in Figure 10 a, magnetic disk of virtual machine data second day The metadata that Day2 is saved includes the metadata 1 ' of third data fragmentation, further includes first number of other data fragmentations that do not modify According to, the metadata of other data fragmentations that do not modify, such as the second data fragmentation each data fragmentation into the n-th data fragmentation Metadata, i.e. each metadata of the metadata 2 into metadata n.

When S831, first server 110 according to backup time are the recovery time, the second server 120 and the The metadata for each data fragmentation that one server 110 saves respectively determines different metadata and identical metadata.

If it is determined that there is different metadata, it is determined that the different metadata gone out are that the first server 110 does not have Have the metadata of preservation, then show data fragmentation corresponding with the metadata not in the backup of the first server 110, It needs to obtain data fragmentation corresponding with the metadata from second server 120 in the backup of second server 120, that is, executes Step S832 to S835 obtains data fragmentation corresponding with the metadata in the backup of second server 120.

If it is determined that part metadata is identical, for identical metadata, without obtaining data from second server 120 Fragment searches the corresponding data fragmentation of each same metadata from the first server 110.

Illustratively, if recovery time is first day Day1, first server 110 is described the according to backup time When one day Day1, the metadata for each data fragmentation that the second server 120 and first server 110 save respectively, really Fixed different metadata and identical metadata.Based on modified shown in Figure 10 b the first data fragmentation be third data fragmentation after 110 situation of the first server of expired data fragmentation is deleted in the backup procedure of the first server 110, and The backup instances of second server 120 shown in Figure 10 c, when recovery time is first day Day1,120 He of second server In the metadata for each data fragmentation that first server 110 saves respectively, each metadata of the metadata 2 into metadata n What is saved in the first server 110 and second server 120 is metadata 1 that is identical, saving in second server 120 It is not identical with the metadata 1 ' that is saved in first server 110.Therefore, the metadata 1 saved in second server 120 is described The metadata that first server 110 does not save needs to obtain data corresponding with the metadata 1 from second server 120 Fragment obtains data point corresponding with the metadata 1 in the backup of second server 120, i.e. execution step S832 to S835 Backup of the piece in second server 120.Have in first server 110 for each metadata of the metadata 2 into metadata n It saves, first server 110 can search each metadata into metadata n point of metadata 2 from the first server 110 Not corresponding data fragmentation.For example, corresponding second data fragmentation of metadata 2, first server 110 can be from first clothes Business device 110, which is searched, to be obtained.

If it is determined that not different metadata, then show that the first server 110 preserves each metadata difference Corresponding data fragmentation is then not necessarily to obtain data fragmentation from second server 120 in the backup of the first server 110, from The first server 110 searches the corresponding data fragmentation of each metadata.

Illustratively, if recovery time is second day Day2, first server 110 is described the according to backup time When two days Day2, the metadata for each data fragmentation that the second server 120 and first server 110 save respectively, really Fixed not different metadata.Therefore, first server 110 can search each metadata point from the first server 110 Not corresponding data fragmentation.

S832, the first server 110 send data fragmentation acquisition request.The data fragmentation acquisition request includes number According to the metadata of fragment, the data fragmentation is identified as the mark of the corresponding data fragmentation of different metadata.

The data fragmentation acquisition request is used to indicate second server 120 according to the metadata lookup of the data fragmentation Data fragmentation.

S833, the second server 120 corresponding data of metadata different according to the metadata lookup of data fragmentation Fragment.

S834, second server 120 send the corresponding data fragmentation of the different metadata to first server 110.

S835, the first server 110 save the corresponding data fragmentation of the different metadata.

Specifically, the first server 110 saves the corresponding data fragmentation of the different metadata to target disk.

1, Figure 11 is a kind of structure chart of local server provided in an embodiment of the present invention, local server referring to Figure 1 1100 include: transmission unit 1110, receiving unit 1111 and processing unit 1112.Processing unit 1112 and transmission unit 1110 and Receiving unit 1111 is separately connected.The effect of each unit is described below in detail.

Transmission unit 1110 is obtained for after local server 1100 receives data recovery request, sending metadata Request, the metadata acquisition request include Backup Data mark, and the Backup Data includes the first data fragmentation.Processing unit 1112 indicate that transmission unit 1110 sends metadata acquisition request after local server 1100 receives data recovery request.

In the present embodiment, the function that transmission unit 1110 sends metadata acquisition request can be by shown in Fig. 4 Interface 411 in controller 410 in first server 110 realizes that transmission unit 1110 sends the tool of metadata acquisition request Body realizes details, can refer to the detail of step S610 shown in fig. 6, repeat no more herein.

Receiving unit 1111, the metadata of first data fragmentation for receiving the transmission of second server 120 are described The metadata of first data fragmentation includes the first instruction information, and the first instruction information is first data fragmentation described The mark or storage location of Backup Data fragment in local server.

In the present embodiment, receiving unit 1111 receives first data fragmentation that second server 120 is sent The function of metadata can be realized by the interface 411 in the controller 410 in first server 110 shown in Fig. 4, be received single Member 1111 receives the specific implementation details of the metadata for first data fragmentation that second server 120 is sent, and can refer to The detail of step S630 shown in fig. 6, repeats no more herein.

Processing unit 1112, for confirming in local server whether save according to the metadata of first data fragmentation There is first data fragmentation in the Backup Data fragment of local server.

In the present embodiment, processing unit 1112 confirms local server according to the metadata of first data fragmentation In whether preserve first data fragmentation the function of the Backup Data fragment of local server can be by shown in Fig. 4 Controller 410 in first server 110 or in which processor 413 realize that processing unit 1112 is according to first data First data fragmentation whether is preserved in the Backup Data of local server in the metadata confirmation local server of fragment The specific implementation details of fragment can refer to the detail of step S640 shown in fig. 6, repeat no more herein.

The transmission unit 1110 is also used to not save first data fragmentation in local in confirmation local server When the Backup Data fragment of server, the first data fragmentation acquisition request, the first data fragmentation acquisition request are sent Metadata including first data fragmentation.

In the present embodiment, the function that transmission unit 1110 sends the first data fragmentation acquisition request can pass through Interface 411 in controller 410 in first server 110 shown in Fig. 4 realizes, transmission unit 1110 sends described the The specific implementation details of one data fragmentation acquisition request can refer to the detail of step S650 shown in fig. 6, herein not It repeats again.

The receiving unit 1111 is also used to receive first data fragmentation that the second server 120 is sent and exists Backup Data fragment in the second server 120.

In the present embodiment, receiving unit 1111 receives first data point that the second server 120 is sent Piece can be by first server 110 shown in Fig. 4 in the function of the Backup Data fragment in the second server 120 Interface 411 in controller 410 realizes that receiving unit 1111 receives first data that the second server 120 is sent The specific implementation details of Backup Data fragment of the fragment in the second server 120 can refer to step shown in fig. 6 The detail of S670, repeats no more herein.

It is standby in the second server 120 to be also used to save first data fragmentation for the processing unit 1112 Part data fragmentation.

In the present embodiment, processing unit 1112 saves first data fragmentation in the second server 120 Backup Data fragment function can by controller 410 in first server 110 shown in Fig. 4 or in which processor 413 realize, processing unit 1112 saves Backup Data fragment of first data fragmentation in the second server 120 Details is implemented, the detail of step S680 shown in fig. 6 can be referred to, repeated no more herein.

As an alternative embodiment, the processing unit 1112, is also used to save first data fragmentation Copy, in local Backup Data fragment, generates the metadata of first data fragmentation as first data fragmentation.

The transmission unit 1110 is also used to send the metadata of first data fragmentation and first data fragmentation To the second server 120.

As an alternative embodiment, the processing unit 1112, is also used to be sent out according to the second server 120 The metadata of first data fragmentation sent, confirms whether the local server preserves the member of first data fragmentation Data, wherein the processing unit 1112 does not save the metadata of first data fragmentation by confirmation local server, Confirmation local server does not save the metadata of first data fragmentation in local Backup Data fragment.

As an alternative embodiment, the receiving unit 1111, is also used to receive data modification request, the number Being used to indicate the local server to modify first data fragmentation according to modification request is target data fragment.The processing is single Member 1112 is also used to modify first data fragmentation to save the target data fragment after the target data fragment Copy, in the Backup Data fragment of local server, and generates the target data fragment as the target data fragment Metadata deletes the metadata of first data fragmentation, and the metadata of the target data fragment includes Target indication information, The Target indication information is mark or storage location of the target data fragment in the Backup Data fragment of local server. The transmission unit 1110 is also used to send the metadata of the target data fragment, the target data fragment to described Two servers 120.

As an alternative embodiment, the metadata of first data fragmentation further includes first data fragmentation Distributing position information in the Backup Data.The receiving unit 1111 is also used to receive data recovery request, the number It include the Backup Data mark and target disk mark according to recovery request, the data recovery request is used to indicate will be described standby Part data are restored to the target disk.The processing unit 1112 is also used to according to first data fragmentation described standby Distributing position information in part data, first data fragmentation is saved to the target disk.

2, Figure 12 is the structure chart of another local server provided in an embodiment of the present invention referring to Figure 1.Local service Device 1200 includes: transmission unit 1210, receiving unit 1211 and processing unit 1212.Processing unit 1212 and transmission unit 1210 It is separately connected with receiving unit 1211.The effect of each unit is described below in detail.

Transmission unit 1210 is obtained for after local server 1200 receives data recovery request, sending metadata Request, the metadata acquisition request include Backup Data mark, and the Backup Data includes the second data fragmentation.Processing unit 1212 indicate that transmission unit 1210 sends metadata acquisition request after local server 1200 receives data recovery request.

In the present embodiment, the function that transmission unit 1210 sends metadata acquisition request can be by shown in Fig. 4 Interface 411 in controller 410 in first server 110 realizes that transmission unit 1210 sends the tool of metadata acquisition request Body realizes details, can refer to the detail of step S710 shown in Fig. 7, repeat no more herein.

Receiving unit 1211, the metadata of second data fragmentation for receiving the transmission of second server 120 are described The metadata of second data fragmentation includes second indication information, and the second indication information is second data fragmentation described The mark or storage location of the Backup Data fragment of local server.

In the present embodiment, receiving unit 1211 receives second data point that the second server 120 is sent The function of the metadata of piece can be realized by the interface 411 in the controller 410 in first server 110 shown in Fig. 4, be connect The specific implementation details that unit 1211 receives the metadata for second data fragmentation that second server 120 is sent is received, it can be with With reference to the detail of step S730 shown in fig. 6, repeat no more herein.

Processing unit 1212, for confirming whether the local server is protected according to the metadata of second data fragmentation There is second data fragmentation in the Backup Data fragment of the local server, preserved in the confirmation local server Second data fragmentation is in the Backup Data fragment of the local server, according to the metadata of second data fragmentation Second data fragmentation is searched in the Backup Data fragment of the local server, saves second data fragmentation described The Backup Data fragment of local server.

In the present embodiment, processing unit 1212 is according to the metadata of the second data fragmentation confirmation local clothes Whether business device preserves second data fragmentation in the Backup Data fragment of the local server, is confirming the local clothes Business device preserves second data fragmentation in the Backup Data fragment of the local server, according to second data point Second data fragmentation described in the metadata lookup of piece saves second data in the Backup Data fragment of the local server Fragment can pass through the control in first server 110 shown in Fig. 4 in the function of the Backup Data fragment of the local server Device 410 processed or in which processor 413 realize, processing unit 1212 according to the metadata of second data fragmentation confirm described in Whether local server preserves second data fragmentation in the Backup Data fragment of the local server, described in confirmation Local server preserves second data fragmentation in the Backup Data fragment of the local server, according to described second Second data fragmentation described in the metadata lookup of data fragmentation saves described the in the Backup Data fragment of the local server Two data fragmentations can refer to step shown in Fig. 7 in the specific implementation details of the Backup Data fragment of the local server The detail of S740, S750, S760, repeat no more herein.

As a kind of optional implementation, the processing unit 1212 is also used to save second data fragmentation Copy, in the Backup Data fragment of the local server, generates second data fragmentation as second data fragmentation Metadata.The transmission unit 1210 is also used to send the metadata of second data fragmentation and second data fragmentation To the second server 120.

As a kind of optional implementation, the processing unit 1212 is also used to be sent out according to the second server 120 The metadata of second data fragmentation sent, confirms whether the local server preserves the member of second data fragmentation Data, wherein the processing unit 1212 is by confirming that the local server preserves first number of second data fragmentation According to confirming that the local server preserves Backup Data of the metadata in the local server of second data fragmentation Fragment.

As a kind of optional implementation, the metadata of second data fragmentation further includes second data fragmentation Distributing position information in the Backup Data.The receiving unit 1211 is also used to receive data recovery request, the number It include the Backup Data mark and target disk mark according to recovery request, the data recovery request is used to indicate will be described standby Part data are restored to the target disk.The processing unit 1212 is also used to according to second data fragmentation described standby Distributing position information in part data, second data fragmentation is saved to the target disk.

As a kind of optional implementation, if the second indication information is second data fragmentation at described The storage location of the Backup Data fragment of ground server, the processing unit 1212 are also used to according to second data fragmentation In the storage location of the Backup Data fragment of the local server, second data fragmentation is searched in the local server Backup Data fragment storage location be directed toward second data fragmentation the local server Backup Data fragment.

As a kind of optional implementation, if the second indication information is second data fragmentation at described The mark of the Backup Data fragment of ground server, the local server pre-save second data fragmentation in the local The Backup Data fragment of the mark of the Backup Data fragment of server and second data fragmentation in the local server The corresponding relationship of storage location.The processing unit 1212 is also used to search second number according to the second indication information According to fragment the Backup Data fragment of the local server mark, according to second data fragmentation in the local service The mark of the Backup Data fragment of device, the Backup Data fragment for searching second data fragmentation in the local server are deposited Storage space is set, according to second data fragmentation the Backup Data fragment of the local server storage location, described in lookup Second data fragmentation exists in second data fragmentation that the storage location of the Backup Data fragment of the local server is directed toward The Backup Data fragment of the local server.

The embodiment of the present invention also provides a kind of local server, including processor and memory and interface, and the interface is used It is communicated in second server 120, for storing computer program code, the computer program code includes the memory Instruction, when the processor executes described instruction, the local server is executed first in data reconstruction method described in Fig. 6 The step of server 110 executes, specific implementation details no longer repeat herein.

The embodiment of the present invention also provides a kind of local server, including processor, memory and interface, and the interface is used for It is communicated with second server 120, the memory includes referring to for storing computer program code, the computer program code It enables, when the processor executes described instruction, the local server executes the first clothes in data reconstruction method described in Fig. 7 The step of business device 110 executes, specific implementation details no longer repeat herein.

The embodiment of the present invention also provides a kind of remote server, comprising: for executing in data reconstruction method shown in fig. 6 Second server 120 execute each step each unit, the unit can be by hardware realization, can also be by hard Part executes corresponding software realization, and specific implementation details no longer repeats herein.

The embodiment of the present invention also provides a kind of remote server, comprising: for executing in data reconstruction method shown in Fig. 7 Second server 120 execute each step each unit, the unit can be by hardware realization, can also be by hard Part executes corresponding software realization, and specific implementation details no longer repeats herein.

The embodiment of the present invention also provides a kind of remote server, including processor, memory and interface, and the interface is used for It is communicated with first server 110, the memory includes referring to for storing computer program code, the computer program code It enables, when the processor executes described instruction, the remote server executes the second clothes in data reconstruction method shown in fig. 6 The step of business device 120 executes, specific implementation details no longer repeat herein.

The embodiment of the present invention also provides a kind of remote server, including processor, memory and interface, and the interface is used for It is communicated with first server 110, the memory includes referring to for storing computer program code, the computer program code It enables, when the processor executes described instruction, the remote server executes the second clothes in data reconstruction method shown in Fig. 7 The step of business device 120 executes, specific implementation details no longer repeat herein.

The embodiment of the present invention provides a kind of data reconstruction method, apparatus and system.First server 110 receives second service The metadata for the first data fragmentation that device 120 is sent, the first instruction information that the metadata of the first data fragmentation includes is first The mark or storage location of Backup Data fragment of the data fragmentation in the first server 110.110 basis of first server The metadata of first data fragmentation confirms that the first server 110 does not save the first data fragmentation in first server 110 Backup Data fragment when, request Backup Data of first data fragmentation in second server 120 point to second server 120 Piece saves the Backup Data fragment of the first data fragmentation.Since the first data fragmentation of meta-data preservation is in first server 110 Backup Data fragment mark or first server 110 Backup Data fragment storage location data volume relative to Hash The data volume of value is small.Therefore, few, propagation delay time of transmitting Internet resources used in a large amount of metadata from second server 120 It is short, improve data recovering efficiency.

In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can be with It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the module It divides, only a kind of logical function partition, there may be another division manner when realization, such as multiple module or components can be with In conjunction with or be desirably integrated into another system, or some features can be ignored or not executed.Another point, it is shown or discussed Mutual coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING of device or module or Communication connection can be electrical property, mechanical or other forms.

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

It, can also be in addition, each functional module in each embodiment of the present invention can integrate in a processing module It is that modules physically exist alone, can also be integrated in two or more modules in a module.Above-mentioned integrated mould Block both can take the form of hardware realization, can also realize in the form of hardware adds software function module.

The above-mentioned integrated module realized in the form of software function module, can store and computer-readable deposit at one In storage media.Above-mentioned software function module is stored in a storage medium, including some instructions are used so that a computer Equipment (can be personal computer, server or the network equipment etc.) executes the portion of each embodiment the method for the present invention Step by step.And storage medium above-mentioned includes: mobile hard disk, read-only memory, random access memory, magnetic or disk etc. The various media that can store program code.

Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features； And these are modified or replaceed, the protection model for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution It encloses.

Claims

1. a kind of data reconstruction method, which is characterized in that the method is executed by first server, comprising:

The first server sends metadata acquisition request, and the metadata acquisition request includes Backup Data mark, described Backup Data includes the first data fragmentation；

The first server receives the metadata for first data fragmentation that second server is sent, first data point The metadata of piece includes the first instruction information, and the first instruction information is first data fragmentation in the first server In Backup Data fragment mark or storage location；

The first server confirms whether the first server saves according to the metadata of first data fragmentation The first data fragmentation is stated in the Backup Data fragment of the first server；

First data fragmentation is not saved in the Backup Data point of the first server in the confirmation first server When piece, the first server sends the first data fragmentation acquisition request, and the first data fragmentation acquisition request includes The metadata of first data fragmentation；

The first server receives first data fragmentation of the second server transmission in the second server Backup Data fragment；

The first server saves Backup Data fragment of first data fragmentation in the second server.

2. the method as described in claim 1, which is characterized in that before the first server sends metadata acquisition request, The method also includes:

The copy that the first server saves first data fragmentation takes as first data fragmentation described first The Backup Data fragment of business device；

The first server generates the metadata of first data fragmentation；

The first server sends the metadata of first data fragmentation and first data fragmentation to second clothes Business device.

3. method according to claim 1 or 2, which is characterized in that the first server is according to first data fragmentation Metadata confirm whether the first server preserves first data fragmentation in the backup number of the first server According to fragment, comprising:

The metadata for first data fragmentation that the first server is sent according to the second server confirms described the Whether one server preserves the metadata of first data fragmentation, wherein the first server is by confirming described the One server does not save the metadata of first data fragmentation, confirms that the first server does not save first number According to fragment metadata the first server Backup Data fragment.

4. method a method according to any one of claims 1-3, which is characterized in that further include:

The first server receives data modification request, and the data modification request is used to indicate the first server modification First data fragmentation is target data fragment；

The first server modifies first data fragmentation to save the target data point after the target data fragment The copy of piece, in the Backup Data fragment of the first server, and generates the number of targets as the target data fragment According to the metadata of fragment, the metadata of first data fragmentation is deleted, the metadata of the target data fragment includes target Indicate that information, the Target indication information are mark of the target data fragment in the Backup Data fragment of the first server Knowledge or storage location；

The first server sends the metadata of the target data fragment, the target data fragment to the second service Device.

5. the method as described in claim 1-4 is any, which is characterized in that the metadata of first data fragmentation further includes institute Distributing position information of first data fragmentation in the Backup Data is stated, the first server sends metadata acquisition request Before, further includes:

The first server receives data recovery request, and the data recovery request includes the Backup Data mark and target Disk mark, the data recovery request, which is used to indicate, restores the Backup Data to the target disk；

The first server saves first data fragmentation, comprising:

Distributing position information of the first server according to first data fragmentation in the Backup Data, by described One data fragmentation is saved to the target disk.

6. a kind of data reconstruction method, which is characterized in that the method is executed by first server, comprising:

The first server sends metadata acquisition request, and the metadata acquisition request includes Backup Data mark, described Backup Data includes the second data fragmentation；

The first server receives the metadata for second data fragmentation that second server is sent, second data point The metadata of piece includes second indication information, and the second indication information is second data fragmentation in the first server Backup Data fragment mark or storage location；

The first server confirms whether the first server saves according to the metadata of second data fragmentation The second data fragmentation is stated in the Backup Data fragment of the first server；

Second data fragmentation is preserved in the Backup Data fragment of the first server in the confirmation first server When, the first server second data fragmentation according to the metadata lookup of second data fragmentation is in first clothes The Backup Data fragment of business device；

The first server saves second data fragmentation in the Backup Data fragment of the first server.

7. method as claimed in claim 6, which is characterized in that before the first server sends metadata acquisition request, The method also includes:

The copy that the first server saves second data fragmentation takes as second data fragmentation described first The Backup Data fragment of business device；

The first server generates the metadata of second data fragmentation；

The first server sends the metadata of second data fragmentation and second data fragmentation to second clothes Business device.

8. method according to claim 6 or 7, which is characterized in that the first server is according to second data fragmentation Metadata confirm whether the first server preserves second data fragmentation in the backup number of the first server According to fragment, comprising:

The metadata for second data fragmentation that the first server is sent according to the second server confirms described the Whether one server preserves the metadata of second data fragmentation, wherein the first server is by confirming described the One server preserves the metadata of second data fragmentation, confirms that the first server preserves second data point Backup Data fragment of the metadata of piece in the first server.

9. such as method as claimed in claim 6 to 8, which is characterized in that the metadata of second data fragmentation further includes institute Distributing position information of second data fragmentation in the Backup Data is stated, the first server sends metadata acquisition request Before, further includes:

The first server saves second data fragmentation, comprising:

Distributing position information of the first server according to second data fragmentation in the Backup Data, by described Two data fragmentations are saved to the target disk.

10. the method as described in claim 6-9 is any, which is characterized in that if the second indication information is described second Data fragmentation is in the storage location of the Backup Data fragment of the first server, then the first server is according to described second Backup Data fragment of second data fragmentation described in the metadata lookup of data fragmentation in the first server, comprising:

The first server is according to second data fragmentation in the storage position of the Backup Data fragment of the first server It sets, searches the storage location of Backup Data fragment of second data fragmentation in the first server is directed toward described second Backup Data fragment of the data fragmentation in the first server.

11. the method as described in claim 6-9 is any, which is characterized in that if the second indication information is described second For data fragmentation in the mark of the Backup Data fragment of the first server, it is several that the first server pre-saves described second According to fragment in the mark of the Backup Data fragment of the first server and second data fragmentation in the first server Backup Data fragment storage location corresponding relationship；

The first server second data fragmentation according to the metadata lookup of second data fragmentation is described first The Backup Data fragment of server, comprising:

The first server searches second data fragmentation in the first server according to the second indication information The mark of Backup Data fragment；

The first server, in the mark of the Backup Data fragment of the first server, is looked into according to second data fragmentation Look for second data fragmentation in the storage location of the Backup Data fragment of the first server；

12. a kind of server characterized by comprising

Transmission unit, for sending metadata acquisition request, the metadata acquisition request includes Backup Data mark, described standby Part data include the first data fragmentation；

Receiving unit, the metadata of first data fragmentation for receiving second server transmission, first data point The metadata of piece includes the first instruction information, and the first instruction information is first data fragmentation in the server The mark or storage location of Backup Data fragment；

Processing unit, for confirming whether preserve described the in the server according to the metadata of first data fragmentation Backup Data fragment of one data fragmentation in the server；

The transmission unit is also used to not save first data fragmentation in the server in the confirmation server When Backup Data fragment, the first data fragmentation acquisition request is sent, the first data fragmentation acquisition request includes described The metadata of first data fragmentation；

The receiving unit is also used to receive first data fragmentation of the second server transmission in the second service Backup Data fragment in device；

The processing unit is also used to save Backup Data fragment of first data fragmentation in the second server.

13. server as claimed in claim 12, which is characterized in that

The processing unit is also used to save the copy of first data fragmentation as first data fragmentation local Backup Data fragment generates the metadata of first data fragmentation；

The transmission unit is also used to send the metadata of first data fragmentation and first data fragmentation to described Two servers.

14. server as described in claim 12 or 13, which is characterized in that

The processing unit, the metadata for first data fragmentation for being also used to be sent according to the second server, confirmation Whether the server preserves the metadata of first data fragmentation, wherein the processing unit is by confirming the clothes Business device does not save the metadata of first data fragmentation, confirms that the server does not save first data fragmentation Metadata is in local Backup Data fragment.

15. the server as described in claim 12-14 is any, which is characterized in that

The receiving unit, is also used to receive data modification request, and the data modification request is used to indicate the server and repairs Changing first data fragmentation is target data fragment；

The processing unit is also used to modify first data fragmentation to save the target after the target data fragment The copy of data fragmentation, in the Backup Data fragment of the server, and generates the target as the target data fragment The metadata of data fragmentation deletes the metadata of first data fragmentation, and the metadata of the target data fragment includes mesh Mark instruction information, the Target indication information are mark of the target data fragment in the Backup Data fragment of the server Or storage location；

The transmission unit is also used to send the metadata of the target data fragment, the target data fragment to described Two servers.

16. the server as described in claim 12-15 is any, which is characterized in that the metadata of first data fragmentation is also Distributing position information including first data fragmentation in the Backup Data；

The receiving unit, is also used to receive data recovery request, and the data recovery request includes the Backup Data mark It is identified with target disk, the data recovery request, which is used to indicate, restores the Backup Data to the target disk；

The processing unit is also used to the distributing position information according to first data fragmentation in the Backup Data, will First data fragmentation is saved to the target disk.

17. a kind of server characterized by comprising

Transmission unit, for sending metadata acquisition request, the metadata acquisition request includes Backup Data mark, described standby Part data include the second data fragmentation；

Receiving unit, the metadata of second data fragmentation for receiving second server transmission, second data point The metadata of piece includes second indication information, and the second indication information is second data fragmentation in the standby of the server The mark or storage location of part data fragmentation；

Processing unit, for confirming whether the server preserves described second according to the metadata of second data fragmentation Data fragmentation preserves second data fragmentation in institute in the confirmation server in the Backup Data fragment of the server When stating the Backup Data fragment of server, according to the second data fragmentation described in the metadata lookup of second data fragmentation in institute The Backup Data fragment of server is stated, saves second data fragmentation in the Backup Data fragment of the server.

18. server as claimed in claim 17, which is characterized in that

The processing unit is also used to save the copy of second data fragmentation as second data fragmentation in the clothes The Backup Data fragment of business device, generates the metadata of second data fragmentation；

The transmission unit is also used to send the metadata of second data fragmentation and second data fragmentation to described Two servers.

19. the server as described in claim 17 or 18, which is characterized in that

The processing unit, the metadata for second data fragmentation for being also used to be sent according to the second server, confirmation Whether the server preserves the metadata of second data fragmentation, wherein the processing unit is by confirming the clothes Business device preserves the metadata of second data fragmentation, confirms that the server preserves first number of second data fragmentation According to the Backup Data fragment in the server.

20. the server as described in claim 17-19 is any, which is characterized in that the metadata of second data fragmentation is also Distributing position information including second data fragmentation in the Backup Data；

The processing unit is also used to the distributing position information according to second data fragmentation in the Backup Data, will Second data fragmentation is saved to the target disk.

21. the server as described in claim 17-20 is any, which is characterized in that if the second indication information is described Second data fragmentation the Backup Data fragment of the server storage location,

The processing unit is also used to according to second data fragmentation in the storage position of the Backup Data fragment of the server It sets, searches second data that the storage location of Backup Data fragment of second data fragmentation in the server is directed toward Backup Data fragment of the fragment in the server.

22. the server as described in claim 17-20 is any, which is characterized in that if the second indication information is described For second data fragmentation in the mark of the Backup Data fragment of the server, the server pre-saves second data point Piece divides in the mark of the Backup Data fragment of the server and second data fragmentation in the Backup Data of the server The corresponding relationship of the storage location of piece；

The processing unit is also used to search second data fragmentation in the server according to the second indication information The mark of Backup Data fragment looked into according to second data fragmentation in the mark of the Backup Data fragment of the server It looks for second data fragmentation in the storage location of the Backup Data fragment of the server, is existed according to second data fragmentation The storage location of the Backup Data fragment of the server searches second data fragmentation in the Backup Data of the server Backup Data fragment of second data fragmentation that the storage location of fragment is directed toward in the server.

23. a kind of server, which is characterized in that including processor, memory and interface, the interface is used for and second server Communication, the memory is for storing computer program code, and the computer program code includes instruction, when the processor When executing described instruction, the server executes the data reconstruction method as described in any one of claim 1-5.

24. a kind of server, which is characterized in that including processor, memory and interface, the interface is used for and second server Communication, the memory is for storing computer program code, and the computer program code includes instruction, when the processor When executing described instruction, the server executes the data reconstruction method as described in any one of claim 6-11.

25. a kind of data recovery system, which is characterized in that including first server and second server, the first server For server described in any one of claim 12-16 or 23.

26. a kind of data recovery system, which is characterized in that including first server and second server, the first server For server described in any one of claim 17-22 or 24.