WO2015055033A1

WO2015055033A1 - Metadata protection method and device

Info

Publication number: WO2015055033A1
Application number: PCT/CN2014/083034
Authority: WO
Inventors: 雷延钊
Original assignee: 华为技术有限公司
Priority date: 2013-10-17
Filing date: 2014-07-25
Publication date: 2015-04-23
Also published as: CN103593260B; CN103593260A

Abstract

The present invention relates to the field of computers. Provided are a metadata protection method and device, providing metadata with an additional protection mechanism and improving security. The specific solution is: firstly, respectively storing the metadata belonging to data of different users into a corresponding first-class storage unit group, the first-class storage unit group being in a first redundant array of independent disks (RAID) storage mode; then selecting at least two storage unit groups from the first-class storage unit group, and forming a third storage unit group according to the at least two selected storage unit groups, the third storage unit group storing at least two metadata belonging to data of different users, the third storage unit group belonging to a second-class storage unit group, and the second-class storage unit group being in a second RAID storage mode; and generating and storing a first parity check code for the at least two metadata belonging to data of different users stored in the third storage unit group so as to protect the metadata.

Description

Method and device for protecting metadata

The present application claims priority to Chinese Patent Application No. 201310487474.9, entitled "A Method and Apparatus for Protecting Metadata", filed on October 17, 2013, the entire contents of which is incorporated herein by reference. in.

Technical field

The present invention relates to the field of computers, and in particular, to a method and an apparatus for protecting metadata.

Background technique

In the storage domain, the data of the user is distributed on the logical hard disk. The logical hard disk is composed of multiple independent physical hard disks or hard disk blocks. The hard disk block refers to the physical hard disk group according to a certain granularity (a number of physical hard disk combinations). Together, the hard disk space implements block virtualization, and the physical addresses of the virtualized hard disk blocks are continuous. The metadata stored in the logical hard disk is generated during the user service configuration and running process, and is used to express key information such as spatial information, mapping relationship, and configuration information of the user stored data. Metadata Once the compromised logical hard drive is not working properly, metadata protection is critical. Redundant Array of Independent Disks (RAID) data storage method is to store data on a storage unit group composed of a plurality of independent physical hard disks or hard disk blocks, by storing the same or different data in Data backup or redundancy protection is implemented on multiple physical hard disks or hard disk blocks.

RAID1 mirrors the data of one hard disk block to another hard disk block. That is to say, when data is written to one hard disk block, an image file is generated on another idle hard disk block, which maximizes the performance without affecting performance. Assuring the reliability and repairability of the system, as long as at least one of the hard disk blocks in any pair of mirrored hard disk blocks in the system can be used, even when half of the number of hard disk blocks have problems, the system can operate normally, when a hard disk block When it fails, the system ignores the hard disk block and uses the remaining mirrored hard disk block to read and write data, which has good disk redundancy. In the prior art, the metadata is backed up and protected by the data protection method of the established RAID1, and the protection mode is relatively simple. When one or more hard disk blocks fail, the correct data may be indistinguishable or the data recovery may exist. There is a risk that it takes a long time to read the correct metadata information even if the data can be recovered.

Summary of the invention

Embodiments of the present invention provide a method and an apparatus for protecting metadata, which solves the problem that a single protection mode may cause a failure to distinguish between correct data or a risk of data recovery.

To achieve the above objective, the embodiment of the present invention uses the following technical solution: In a first aspect, a method for protecting metadata is provided, where the method includes:

The metadata belonging to different user data are respectively stored in the corresponding first type of storage unit group, and the first type of storage unit group uses the first RAID storage mode;

Selecting at least two storage unit groups in the first type of storage unit group, and forming a third storage unit group according to the at least two storage unit groups, wherein the third storage unit group stores at least two belonging to different users. The metadata of the data, the third storage unit group belongs to the second type of storage unit group, and the second type of storage unit group uses the second RAID storage mode;

Generating a first parity code for the at least two metadata belonging to different user data stored by the third storage unit group and storing. With reference to the first aspect, in a first possible implementation manner, metadata belonging to different user data in the metadata is separately stored in a corresponding first type of storage unit group, where the first type of storage unit group The first type of RAID storage method includes: storing metadata belonging to different user data in the metadata in each physical storage unit or hard disk block in the corresponding first type of storage unit group; A type of storage unit group includes at least two physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing physical storage unit groups according to a preset granularity. With reference to the first possible implementation of the first aspect, in a second possible implementation, at least two storage unit groups are selected in the first type of storage unit group, and according to the at least two storage units Forming the third storage unit group includes: selecting one physical storage unit or hard disk block in the first category;

L disk blocks constitute the third storage unit group;

The third storage unit group includes at least three physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing physical storage unit groups according to a preset granularity. With reference to the first aspect, in a third possible implementation, generating, by the third storage unit group, the at least two metadata belonging to different user data, the first parity code and storing further includes:

Generating a second parity code for the first parity code and at least one other parity code and storing the second parity code;

The at least one other parity code is a parity code other than the third metadata of the second type of memory cell group stored in the different user data.

With reference to the first aspect, in a fourth possible implementation manner, the first RAID storage mode is RAID1, and the second RAID storage mode is RAID5. A second aspect provides a protection device for metadata, where the device includes: a storage module, configured to acquire metadata, and store metadata belonging to different user data in the metadata in a corresponding first class In the storage unit group, the first type of storage unit group uses a first type of storage mode, and the protection module is configured to select at least two storage unit groups in the first type of storage unit group, and Forming, according to the at least two storage unit groups, a third storage unit group, wherein the third storage unit group stores at least two metadata belonging to different user data, and the third storage unit group belongs to the second type storage unit group The second type of storage unit group uses the second RAID storage mode; the parity code module is configured to generate, for the third storage unit group, the at least two metadata belonging to different user data. A parity code is stored and stored.

With reference to the second aspect, in a first possible implementation manner, the storage module is specifically configured to: store metadata belonging to different user data in the metadata in a corresponding first type of storage unit group Each of the physical storage units or the hard disk block; the first type of storage unit group includes at least two physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing physical storage unit groups according to a preset granularity.

In conjunction with the first possible implementation of the second aspect, in a second possible implementation, the protection module is specifically configured to:

In the first category - selecting a physical storage unit or a hard disk block;

L disk blocks constitute the third storage unit group;

The second type of storage unit group includes at least three physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing physical storage unit groups according to a preset granularity.

With reference to the second aspect, in a third possible implementation, the parity code module is further configured to: generate a second parity for the first parity code and the at least one other parity code Code and store;

The at least one other parity code is other than the third storage unit group except the third storage unit group.

Parity code of metadata for different user data < With reference to the second aspect, in a fourth possible implementation manner, the first RAID storage mode is RAID1, and the second RAID storage mode is RAID5. An embodiment of the present invention provides a method and an apparatus for protecting metadata, first acquiring metadata, and storing metadata belonging to different user data in the metadata in a corresponding first type of storage unit group. The first type of storage unit group uses the first RAID storage mode, and then selects at least two storage unit groups in the first type of storage unit group, and forms a third storage unit according to the at least two storage unit groups. The third storage unit group stores at least two metadata belonging to different user data, the third storage unit group belongs to the second type storage unit group, and the second type storage unit group uses the second type. And storing, by the RAID storage mode, the first parity code for the at least two metadata belonging to different user data stored by the third storage unit group and storing the first parity code. This adds a layer of protection to the metadata, which reduces the risk of data recovery and improves security.

BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings to be used in the embodiments or the description of the prior art will be briefly described below, and obviously, in the following description The drawings are only some of the embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative work. 1 is a schematic flowchart 1 of a method for protecting metadata according to an embodiment of the present invention; FIG. 2 is a schematic flowchart 2 of a method for protecting metadata according to an embodiment of the present invention; FIG. 4 is a schematic diagram 2 of a method for protecting a metadata according to an embodiment of the present invention; FIG. 5 is a diagram of protection of metadata according to an embodiment of the present invention; FIG. 6 is a schematic structural diagram 2 of a metadata protection device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The technical solutions in the embodiments of the present invention will be clarified in the following with reference to the accompanying drawings in the embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS It is apparent that the described embodiments are only a part of the embodiments of the invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without departing from the inventive scope are the scope of the present invention. An embodiment of the present invention provides a method for protecting metadata. As shown in FIG. 1, the method includes:

101. The metadata belonging to different user data in the metadata is respectively stored in the corresponding first type of storage unit group, and the first type of storage unit group uses the first type of RAID storage manner.

102. Select at least two storage unit groups in the first type of storage unit group, and form a third storage unit group according to the at least two storage unit groups, where the third storage unit group stores at least two metadata belonging to different user data. The third storage unit group belongs to the second type storage unit group, and the second type storage unit group uses the second RAID storage method.

103. Generate a first parity code for at least two metadata belonging to different user data stored in the third storage unit group and store the first parity code. The embodiment of the present invention provides a method for protecting metadata, which first stores metadata belonging to different user data in a corresponding first type of storage unit group, and the first type of storage unit uses the first RAID storage method. And then selecting at least two storage unit groups in the first type of storage unit group, and forming a third storage unit group according to the selected at least two storage unit groups, the third storage unit group storing at least two belonging to different user data. Metadata, the third storage unit group belongs to the second type of storage unit group, the second type of storage unit group uses the second RAID storage mode, and then the third storage unit group stores at least two metadata belonging to different user data. A first parity code is generated and stored. This adds a layer of protection to the metadata and improves security. An embodiment of the present invention further provides a method for protecting metadata, as shown in FIG. 2, the method includes:

201. The metadata belonging to different user data are respectively stored in the corresponding first type of storage unit group, and the first type of storage unit group uses the first type of RAID storage manner. The hard disk block (Chunk) is a hard disk space obtained by dividing the hard disk space according to a certain granularity, and the physical address of the hard disk in the hard disk block is continuous. Taking the first type of storage unit group as an example, if the first type of storage unit group is composed of physical storage units, the first type of storage unit group includes at least two physical storage units, and if the first type of storage unit group is composed of a hard disk The first type of storage unit group includes at least two hard disk blocks. In addition, a logical space composed of a plurality of hard disk blocks according to a specified RAID type is called a hard disk block group (Chunk Group), and therefore, One type of storage unit group is called the first type of Chunk Group, and the second type of storage unit group is also applicable. The physical storage unit may be a physical hard disk, such as a disk or an SSD (Solid State Disk), or may be a flash memory, or may be a compact disk or other physical storage. Therefore, the metadata belonging to different user data are respectively stored in the corresponding first type of storage unit group, and the specific metadata may be: storing each metadata belonging to different user data in each physical medium of the corresponding first type storage unit group. The storage unit or the hard disk block; the first type of storage unit group includes at least two physical storage units or hard disk blocks. The hard disk block is obtained by dividing a physical storage unit group according to a preset granularity. The first RAID storage mode mentioned in the embodiment of the present invention may be RAID1. Taking a physical hard disk as an example, the physical storage unit group may be composed of multiple large-capacity physical hard disks, for example, four 500G physical technologies. The hard disks form a physical storage unit group. Among them, RAID1 is a technology for mirroring data of one physical storage unit or hard disk block to another physical storage unit or hard disk block, that is, data is idle while being written to one physical storage unit or hard disk block. An image file is generated on a physical storage unit or a hard disk block to ensure the reliability and repairability of the system to the utmost extent without affecting performance. As long as there is at least one physical storage in any pair of mirrored physical storage units or hard disk blocks in the system. The unit or hard disk block can be used, and the system can operate normally even when there are problems in half of the physical storage unit or the hard disk block. When a physical storage unit or hard disk block fails, the system ignores the physical storage unit or the hard disk block. Turn instead Use the remaining mirrored physical storage unit or hard disk block to read and write data, with good disk redundancy.

202. Select at least two storage unit groups in the first type of storage unit group, and form a third storage unit group according to the at least two storage unit groups. Example ')· born, can pass:

Preferably, a physical storage unit or a hard disk block is selected from each of the at least two storage unit groups;

Forming a third storage unit group according to the selected physical storage unit or hard disk block in each storage unit group; the third storage unit group belongs to the second type storage unit group, and the third storage unit group stores at least two belonging to different user data. Metadata, the second type of storage unit group uses the second RAID storage method. The second type of storage unit group includes at least three physical storage units or hard disk blocks, and the hard disk block is obtained by dividing a physical storage unit group according to a preset granularity, so the third storage unit group also includes at least three physical storage units. Or hard disk block. Illustratively, the second RAID storage method mentioned in the embodiment of the present invention may be RAID 5. The RAID 5 is a physical storage unit (or a hard disk block) in which the data is stored, and a physical storage unit (or a hard disk block) in which the corresponding parity code is located is formed into a storage unit group. technology. When the data of one physical storage unit (or hard disk block) of RAID5 is damaged, the remaining data and the corresponding parity code can be used to recover the damaged data.

The RAID5 data storage method requires at least three physical storage units or hard disk blocks, and stores the data and the corresponding parity code on each physical storage unit or hard disk block that constitutes RAID5, and the odd-numbered n-1 (n is composed of RAID5). The physical storage unit or the number of hard disk blocks) the physical storage unit or the hard disk block stores complete data, that is, equivalent to one physical storage unit or hard disk block capacity The amount of space is used to store the parity code. Therefore, when a physical storage unit or hard disk block of RAID5 is damaged, the integrity of the data is not affected, thereby ensuring data security. When a damaged physical storage unit or hard disk block is replaced, RAID automatically uses the remaining parity code to reconstruct the data on the physical storage unit or hard disk block, maintaining high reliability of data storage.

203. Generate a first parity code for at least two metadata belonging to different user data stored in the third storage unit group and store the first parity code. Illustratively, the parity code may be a vertical parity code, a horizontal parity code, or a horizontal vertical parity code.

204. Generate a second parity code for the first parity code and the at least one other parity code and store the second parity code. Specifically, the at least one other parity code is a parity code other than the third metadata belonging to different user data in the second type of storage unit group. In order to enable a person skilled in the art to more clearly understand the technical solutions provided by the embodiments of the present invention, another method for protecting metadata provided by the embodiments of the present invention is described in detail by using specific examples: 3, the black and white part of the figure is the first memory cell group and the second memory cell group using the first type of storage mode RAID1, wherein the first memory cell group is composed of hard disk blocks A, B, C composition, the same metadata 1 is stored on the hard disk blocks A, B, and C, and the second storage unit group is composed of hard disk blocks 0, E, and Z, and the same metadata 2 is stored on the hard disk blocks D, E, and Z. Then, the hard disk blocks A and D are selected in the first storage unit group and the second storage unit group, and the metadata 1 and the metadata 2 stored on the hard disk blocks A and D are XORed to obtain the metadata 1 and the metadata 2 Parity code 1, the obtained parity code 1 is stored on the hard disk block Z, the parity code 1 belongs to the first parity code, and the hard disk blocks A, D, Z are stored in the storage mode of the RAID 5 to form the third storage. Unit group, the third storage unit group belongs to the Type memory cell groups. Thereby adding a layer of protection to the metadata, which is better Protection metadata.

Example 2, as shown in FIG. 4, the hard disk blocks G, H, and I form a fourth storage unit group, the hard disk blocks J, K, and L form a fifth storage unit group, and the hard disk blocks 0, P, and Q form a sixth storage unit group. The hard disk blocks R, S, and T constitute a seventh storage unit group, and the fourth, fifth, sixth, and seventh storage unit groups respectively store metadata 3, 4, 5, 6, fourth, fifth, sixth, and seventh storage unit groups. The RAID 1 storage mode is used and both belong to the first type of storage unit group, and the parity code 2 of the metadata 3 and the metadata 4 and the parity code of the metadata 5 and the metadata 6 are obtained according to the method in the first example. 4. Then, the parity code 2 and the parity code 3 are XORed to obtain the parity code 2 of the parity code 2 and the parity code 3, and the parity code 2 and the parity code 3 belong to the a parity check code, the parity check code 4 belongs to the second parity check code, the hard disk blocks 0, M, and J constitute the eighth storage unit group, and the hard disk blocks 0, R, and N constitute the ninth storage unit group, and the hard disk block M , N, U form the tenth storage unit group, the eighth, ninth, tenth storage unit group In RAID5 storage composition, eighth, ninth, tenth memory cell group belong to the second memory cell group. This also protects the parity code, further enhancing the protection of the metadata. It should also be noted here that the hard disk block M storing the parity code of the metadata 3 and the metadata 4 in this example is not specifically used for storing the parity code, and may be stored by a hard disk block storing other metadata. The parity code, for example, may be the hard disk block P storing the metadata 5. This is because, for the RAID5 storage mode, the parity code of the metadata and the metadata are separately stored on different hard disk blocks, and different metadata and corresponding different parity codes are separately stored and stored in different hard disk blocks. In the above, as long as the metadata and the parity code corresponding to the metadata are not stored on the same hard disk block, the effect of the storage is equivalent to storing the parity code separately by using one hard disk block.

It is also worth mentioning that the parity codes in the above examples 1 and 2 can be used to verify whether the metadata stored on the hard disk block is correct. For example, in the example 1, it is assumed that the metadata 1 stored on the hard disk block A is incorrectly verified by the check calculation, then the correct metadata 1 is re-stored on the hard disk block A, and if the hard disk block A is damaged, the new healthy state is good. The hard disk block replaces hard disk block A and stores the correct metadata 1 on the new hard disk block. The parity codes in the above examples 1 and 2 can also be used to recover corrupted metadata. For example, in Example 2, it is assumed that the metadata 3 stored on the hard disk blocks G, H, and I is all wrong and stored. If the parity code 2 stored on the hard disk block M is also incorrect, the parity can be calculated by redundancy recovery according to the parity code 3 stored on the hard disk block N and the parity code 4 stored on the hard disk block U. The code 2 is stored on the hard disk block M. If the hard disk block M is damaged, the hard disk block M is replaced with a new healthy hard disk block and the correct parity code 2 is stored on the new hard disk block, and then according to the parity table. The verification code 2 and the metadata 4 stored on the hard disk blocks K, J, and L are calculated by redundancy recovery to obtain the correct metadata 3 and correspondingly stored on the hard disk blocks G, H, and I, if the hard disk blocks G, H, I If there is damage, replace the damaged one with a new healthy hard disk block and store the correct metadata 3 on the new hard disk block. In this way, a layer of protection is added to the metadata to better protect the metadata, and it is easier to recover lost data in the event of an accident.

Finally, it should be pointed out that in the prior art, metadata is generally stored in the form of RAID1 in four storage units, and if there are N different metadata, 4N storage units are needed, but the root storage unit is This saves storage space and reduces redundant space. For example, in Example 1, N=2, each metadata requires 4 storage units to protect data, and the method provided by use case 1 only needs 3N+1=7 storage units, if using the prior art method, 4N = 8 memory cells, therefore, Nl = 1 memory cells are reduced compared to the prior art, and it can be seen that the redundancy space is reduced with the same redundancy.

The embodiment of the present invention provides a method for protecting metadata, which first stores metadata belonging to different user data in a corresponding first type of storage unit group, and the first type of storage unit uses the first RAID storage method. And then selecting at least two storage unit groups in the first type of storage unit group, and forming a third storage unit group according to the selected at least two storage unit groups, the third storage unit group having at least two elements belonging to different user data. Data, the third storage unit group belongs to the second type of storage unit group, and the second type of storage unit group uses the second RAID storage mode, and then generates at least two metadata belonging to different user data stored in the third storage unit group. The first parity code is stored. This adds a layer of protection to the metadata, improves security, and reduces redundancy with the same redundancy. The embodiment of the present invention further provides a metadata protection device 00. As shown in FIG. 5, the protection device includes:

The storage module 10 is configured to store metadata belonging to different user data in a corresponding first type of storage unit group, and the first type of storage unit group uses a first type of RAID storage manner. Optionally, the storage module 10 is specifically configured to: store metadata belonging to different user data on each physical storage unit or hard disk block in the corresponding first type of storage unit group; the first type of storage unit group includes At least two physical storage units or hard disk blocks obtained by dividing physical storage unit groups according to a preset granularity.

The protection module 20 is configured to select at least two storage unit groups in the first type of storage unit group, and form a third storage unit group according to the at least two storage unit groups, where the third storage unit group stores at least two belonging to different users. The metadata of the data, the third storage unit group belongs to the second type storage unit group, and the second type storage unit group uses the second RAID storage method. Optionally, the protection module 20 can be specifically configured to: a physical storage unit or a hard disk block;

Forming a third storage unit group according to a physical storage unit or a hard disk block selected in each storage unit group;

The third storage unit group includes at least three physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing physical storage unit groups according to a preset granularity.

The parity code module 30 is configured to generate and store a first parity code for at least two metadata belonging to different user data stored by the third storage unit group. Specifically, the module is further configured to generate and store a second parity code for the first parity code and the at least one other parity code. The at least one other parity code is a parity code other than the third metadata stored in different user data in the second type of memory cell group.

Exemplarily, the above parity code may be a vertical parity code, a horizontal parity code or a horizontal vertical parity code. Embodiments of the present invention provide a metadata protection apparatus, which first stores metadata belonging to different user data in a corresponding first type of storage unit group, and the first type of storage unit group uses the first type.

RAID storage mode, then selecting at least two storage unit groups in the first type of storage unit group, and forming a third storage unit group according to the selected at least two storage unit groups, the third storage unit group having at least two belonging to different users Metadata of the data, the third storage unit group belongs to the second type of storage unit group, the second type of storage unit group uses the second RAID storage mode, and then the third storage unit group stores at least two belonging to different user data. The metadata generates a first parity code and stores it. This adds a layer of protection to the metadata and improves security. Embodiments of the present invention also provide a metadata protection device 90. As shown in FIG. 6, the device 90 includes: a bus 94; and a processor 91, a memory 92, and an interface 93 connected to the bus 94, wherein the interface 93 For communication; the memory 92 is for storing instructions, and the processor 91 is configured to execute the instructions for:

The metadata belonging to different user data are respectively stored in the corresponding first type of storage unit group, and the first type of storage unit group uses the first independent redundant disk array RAID storage mode; At least two storage unit groups, and forming a third storage unit group according to at least two storage unit groups, the third storage unit group storing at least two metadata belonging to different user data, and the third storage unit group belongs to the second type storage The second group of storage unit groups uses the second RAID storage mode; the first parity code is generated and stored for at least two metadata belonging to different user data stored in the third storage unit group. Optionally, the executing the instruction by the processor 91 may be specifically used to:

Storing metadata belonging to different user data in the metadata on each hard disk or hard disk block in the corresponding first type of storage unit group; the first type of storage unit group includes at least two physical storage units Or a hard disk block, where the hard disk block is obtained by dividing a physical storage unit group according to a preset granularity.

Optionally, the executing the instruction by the processor 91 may also be specifically used to:

a physical storage unit or hard disk block;

The third storage unit group is formed according to the physical storage unit or the hard disk block selected in each storage unit group; the third storage unit group includes at least three physical storage units or hard disk blocks, and the hard disk block divides the physical according to the preset granularity. The storage unit group got.

Optionally, the processor 91 executes the instruction to: generate a second parity code for the first parity code and the at least one other parity code and store the second parity code.

The at least one other parity code is a parity code of at least two metadata belonging to different user data stored in the second type of storage unit group except the third storage unit group. Optionally, the first RAID storage mode may be RAID1, and the second RAID storage mode may be RAID5. The embodiment of the present invention provides a metadata protection device, which first stores metadata belonging to different user data in a corresponding first type of storage unit group, and the first type of storage unit group uses a first RAID storage method. And then selecting at least two storage unit groups in the first type of storage unit group, and forming a third storage unit group according to the selected at least two storage unit groups, the third storage unit group having at least two elements belonging to different user data. Data, the third storage unit group belongs to the second type of storage unit group, the second The class storage unit group uses the second RAID storage mode, and then generates and stores the first parity code for at least two metadata belonging to different user data stored in the third storage unit group. This adds a layer of protection to the metadata and improves security. The character "/" in this article generally indicates that the contextual object is an "or" relationship.

The device embodiments described above are merely illustrative. For example, the division of modules is only a logical function division, and the actual implementation may have another division manner. Alternatively, the modules shown or discussed may be connected to each other through some interface, and may be in electrical, mechanical or other form. The individual modules may or may not be physically separate and may or may not be physical units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiment scheme. In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may be physically included separately, or two or more modules may be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of hardware plus software function modules. The above-described integrated modules implemented in the form of software functional units can be stored in a computer readable storage medium. The above software function modules are stored in a storage medium and include a number of instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform part of the steps of the method of the various embodiments of the present invention. The foregoing storage medium includes: a USB flash drive, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and the like can store program codes. Medium. The above is only the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the present invention. It should be covered by the scope of the present invention. Therefore, the scope of the invention should be determined by the scope of the appended claims.

Claims

Rights request

1. A metadata protection method, characterized by including:

Among them, the first type of storage unit group adopts the first independent redundant disk array RAID storage method; at least two storage unit groups are selected from the first type of storage unit group, and according to the at least two storage units The unit group forms a third storage unit group, the third storage unit group stores at least two metadata belonging to different user data, the third storage unit group belongs to a second type of storage unit group, and the second type of storage The unit group adopts the second RAID storage method; generating and storing a first parity check code for the at least two metadata belonging to different user data stored in the third storage unit group.

2. The method according to claim 1, characterized in that: metadata belonging to different user data in the metadata are stored in corresponding first-type storage unit groups, and the first-type storage unit group Using the first RAID storage method includes: storing the metadata belonging to different user data in the metadata on each hard disk or hard disk block in the corresponding first-type storage unit group;

The first type of storage unit group includes at least two physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing the physical storage unit group according to a preset granularity.

3. The method according to claim 2, characterized in that, selecting at least two memory unit groups from the first type of memory unit group, and forming a third memory unit group based on the at least two memory unit groups includes: :

Select a physical storage unit or hard disk block;

Three storage unit groups; The third storage unit group includes at least three physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing the physical storage unit group according to a preset granularity.

4. The method according to claim 1, wherein the method further comprises: generating and storing a second parity check code for the first parity check code and at least one other parity check code; The at least one other parity check code is a parity check code of the third type of stored data in the second type of storage unit group.

5. The method according to claim 1, characterized in that the first RAID storage method is RAID1, and the second RAID storage method is RAID5.

6. A metadata protection device, characterized in that it includes: a storage module, configured to store metadata belonging to different user data in corresponding first-type storage unit groups, said first-type storage unit group Adopt the first independent redundant disk array RAID storage method; a protection module, used to select at least two storage unit groups from the first type of storage unit group, and form a third storage unit group based on the at least two storage unit groups Storage unit group, the third storage unit group stores at least two metadata belonging to different user data, the third storage unit group belongs to the second type of storage unit group, the second type of storage unit group adopts the third Two RAID storage methods; A parity check code module, configured to generate and store a first parity check code for the at least two metadata belonging to different user data stored in the third storage unit group.

7. The device according to claim 6, wherein the storage module is specifically configured to: store metadata belonging to different user data in the metadata in corresponding first-type storage unit groups. On each physical storage unit or hard drive block;

The first type of storage unit group includes at least two physical storage units or hard disk blocks, and the hard disk Disk blocks are obtained by dividing physical storage unit groups according to preset granularity.

8. The device according to claim 7, characterized in that the protection module is specifically used for:

Select a physical storage unit or hard disk block;

Three storage unit groups;

The third storage unit group includes at least three physical storage units or hard disk blocks, and the hard disk blocks are obtained by dividing the physical storage unit group according to a preset granularity.

9. The device according to claim 6, wherein the parity check code module is further configured to: generate a second parity check code for the first parity check code and at least one other parity check code and storage;

The at least one other parity check code is at least two stored by other storage unit groups using the second RAID storage method except the third storage unit group in the second type of storage unit group and belonging to different users. The parity code of the data's metadata.

10. The device according to claim 6, wherein the first RAID storage method is RAID1, and the second RAID storage method is RAID5.