CN112558876A

CN112558876A - LUN destroying method and storage equipment

Info

Publication number: CN112558876A
Application number: CN202011476459.0A
Authority: CN
Inventors: 上官应兰
Original assignee: Macrosan Technologies Co Ltd
Current assignee: Macrosan Technologies Co Ltd
Priority date: 2020-12-14
Filing date: 2020-12-14
Publication date: 2021-03-26
Anticipated expiration: 2040-12-14
Also published as: CN112558876B

Abstract

The application provides a LUN destroying method and storage equipment. The storage device comprises a LUN and a RAID for storing data corresponding to the LUN. In the application, the LUN directly issues the destroying command to the RAID, and the destroying data does not need to be written into the cache and then is written into the RAID, so that the cache resources can be saved, and the destroying efficiency is improved. In addition, when the RAID executes the destruction command, if the destruction of the entire band is involved, RAID redundancy processing is not executed on the band for this destruction, so that RAID computing resources can be saved, and the destruction efficiency is improved.

Description

LUN destroying method and storage equipment

Technical Field

The present application relates to the field of storage technologies, and in particular, to a LUN destruction method and a storage device.

Background

The Logical Unit Number (LUN) destruction means that when user data in a LUN is no longer used, the existing data is covered by writing destruction data to ensure that the existing user data cannot be restored, thereby ensuring the security of the user data.

The existing destroying process is the same as the data writing process, namely, the LUN constructs a writing command containing the destroying data needing to be written into the appointed address field; the cache allocates a cache block for the destroyed data corresponding to the write command, and sends the destroyed data to an Independent Redundant Array of Independent Disks (English: Redundant Array of Independent Disks, abbreviation: RAID) through a background disk-flushing command; and the RAID calculates redundant data according to a RAID algorithm, and then writes the destroyed data and the corresponding redundant data into each disk.

As can be seen from the above destruction process: on the one hand, LUN destruction requires a large amount of resources. For example, occupying cache resources; carrying the destroyed data in the write command according to the standard of writing user data can consume a large amount of bandwidth resources; RAID calculation for destroying data consumes RAID calculation resources and the like. On the other hand, LUN destruction requires a complete data write process, and therefore LUN destruction efficiency is not high.

Disclosure of Invention

In view of this, the present application provides an LUN destruction method and a storage device, so as to reduce resource consumption and improve destruction efficiency.

In order to achieve the purpose of the application, the application provides the following technical scheme:

in a first aspect, the present application provides a LUN destruction method, which is applied to a storage device, where the storage device includes at least one LUN and at least one RAID, where data of a target LUN corresponding to the at least one LUN is stored in at least one target RAID, the target LUN is divided into a plurality of logical spaces, and the logical spaces of the target LUN correspond to physical spaces in the target RAID one to one and are equal in size, where the method includes:

the following processing is performed for each logical space:

the target LUN issues a destruction command to a first target RAID, wherein the first target RAID is a target RAID to which a target physical space corresponding to a current logical space belongs, and the destruction command comprises an address range of the target physical space and preset first-length destruction data;

the first target RAID determines at least one target stripe involved in the target physical space according to the address range of the target physical space;

the following destruction process is performed for each target band:

if the target stripe all falls into the target physical space, the first target RAID writes the destroyed data into the target stripe, and does not execute RAID redundancy processing on the target stripe aiming at the write operation;

and if the target stripe does not fall into the target physical space, the first target RAID writes the destroyed data into the address field of the target stripe falling into the target physical space, and executes RAID redundancy processing on the target stripe.

Optionally, the first length is smaller than the target physical space size.

Optionally, the destruction command further includes destruction times, and the destruction command includes destruction data of a first length, which is the same as the destruction times;

the executing of the destruction process for each target band includes:

the first target RAID executes the destruction number of times destruction processing for each target stripe, wherein each time destruction processing uses destruction data of a first length corresponding to the destruction.

Optionally, the first target RAID includes a plurality of member disks, and the writing of the destroyed data to the target stripe by the first target RAID includes:

the first target RAID determines a target member disk to which each target stripe block needing to be written with destroyed data in the target stripe belongs;

and for each target member disk, the first target RAID issues a write command to the target member disk, the write command includes destruction data of a second length, the destruction data of the second length is generated according to the destruction data of the first length, and the second length is the size of an address field in the target stripe block corresponding to the target member disk, to which the destruction data is to be written.

Optionally, the storage device further includes a cache, where the cache is divided into a plurality of cache blocks, and before the target LUN issues the destroy command to the first target RAID, the method further includes:

the target LUN does not issue new disk writing commands any more, and waits for the return results of all issued disk writing commands;

and after receiving the return result, recovering all cache blocks corresponding to the target LUN.

In a second aspect, the present application provides a storage device, where the storage device includes at least one LUN and at least one RAID, where data of a target LUN corresponding to the at least one LUN is stored in the at least one target RAID, the target LUN is divided into a plurality of logical spaces, and the logical spaces of the target LUN correspond to physical spaces in the target RAID one to one and are equal in size;

the target LUN is used for issuing a destruction command to a first target RAID aiming at each logic interval, wherein the first target RAID is a target RAID to which a target physical space corresponding to the current logic space belongs, and the destruction command comprises an address range of the target physical space and preset first-length destruction data;

the first target RAID is used for determining at least one target stripe involved in the target physical space according to the address range of the target physical space; for each target stripe, if all the target stripes fall into the target physical space, writing the destroyed data into the target stripe, and not executing RAID redundancy processing on the target stripe aiming at the writing operation; and if the target stripe does not fall into the target physical space, writing the destroyed data into the address field of the target stripe falling into the target physical space, and executing RAID redundancy processing on the target stripe.

Optionally, the first length is smaller than the target physical space size.

the first target RAID is specifically configured to execute the destruction number destruction processing for each target stripe, where each destruction processing uses destruction data of a corresponding first length destroyed by the time.

Optionally, the first target RAID includes a plurality of member disks;

the first target RAID is specifically configured to determine a target member disk to which each target stripe block in which destroyed data needs to be written belongs in the target stripe; and issuing a write command to each target member disk, wherein the write command comprises destruction data with a second length, the destruction data with the second length is generated according to the destruction data with the first length, and the second length is the size of an address field, in which the destruction data is to be written, in a target bar block corresponding to the target member disk.

Optionally, the storage device further includes a cache, and the cache is divided into a plurality of cache blocks;

the target LUN is also used for not issuing a new disk writing command before issuing the destruction command to the first target RAID and waiting for the return results of all issued disk writing commands; and after receiving the return result, recovering all cache blocks corresponding to the target LUN.

As can be seen from the above description, in the embodiment of the present application, the LUN directly issues the destroy command to the RAID without writing the destroy data into the cache and then flushing the disk to the RAID, so that cache resources can be saved, and the destroy efficiency is improved. In addition, when the RAID executes the destruction command, if the destruction of the entire band is involved, RAID redundancy processing is not executed on the band for this destruction, so that RAID computing resources can be saved, and the destruction efficiency is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a flowchart illustrating a LUN destruction method according to an embodiment of the present application;

fig. 2 is a flow chart illustrating an implementation of writing destroyed data to a target stripe by a first target RAID according to an embodiment of the present application.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application.

The terminology used in the embodiments of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the embodiments of the present application. As used in the embodiments of the present application, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present application to describe various information, the information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, the negotiation information may also be referred to as second information, and similarly, the second information may also be referred to as negotiation information without departing from the scope of the embodiments of the present application. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

In order to make the objects, technical solutions and advantages of the embodiments of the present application more clear, the embodiments of the present application are described in detail below with reference to the accompanying drawings and specific embodiments:

referring to fig. 1, a flowchart of a LUN destruction method according to an embodiment of the present application is shown. The flow is applied to a storage device.

The storage device includes at least one RAID. Each RAID is made up of multiple member disks, providing higher storage performance and data redundancy protection than a single disk.

The storage device also includes at least one LUN. Each LUN is divided into a plurality of logical spaces, and each logical space corresponds to a physical space of the same size in RAID. The LUN converts an access request of a user for the logical space into a read-write command for a corresponding physical space by maintaining a mapping relationship between the logical space and the physical space, so as to realize the use of storage resources by the user.

Referring to table 1, an example of a mapping relationship between a logical space of a LUN and a physical space of a RAID is shown in the embodiment of the present application.

TABLE 1

When the user data in the LUN is no longer used, the destruction process shown in fig. 1 may be executed, and the process includes the following steps:

step 101, a target LUN issues a destroy command to a first target RAID, where the destroy command includes an address range of a target physical space and a preset first length of destroy data.

Here, the target LUN may be any LUN that needs to be destroyed in the storage device; and the RAID storing the data corresponding to the target LUN is called a target RAID. It is understood that the reference to target LUN and target RAID is a name for convenience of distinguishing and is not intended to be limiting.

For example, LUN1 is the target LUN to be destroyed, and the RAID-A, RAID-B, RAID-C shown in Table 1 is the target RAID of LUN 1.

The embodiment of the application executes the following processing to each logic space in the target LUN:

and the target LUN issues a destruction command to a target RAID to which the physical space corresponding to the current logic space to be processed belongs. Here, the target RAID to which the physical space corresponding to the current logical space belongs is referred to as a first target RAID. It is understood that the reference to the first target RAID is a name for convenience of distinction and is not intended to be limiting.

For example, if the current logical space to be processed is the 1 st logical space of LUN1 (the starting address is 0, and the space size is 1GB), LUN1 determines that the RAID to which the logical space corresponds to the physical space is RAID-a according to the space mapping table (table 1), and issues a destroy command to RAID-a.

The destroy command comprises an address range of the target physical space and destroy data with a first length. Here, the target physical space is a physical space corresponding to the current logical space to be processed. It is to be understood that the reference to the target physical space, the first length, is a name for convenience of distinction and is not intended to be limiting.

Still take the 1 st logical space (with the start address being 0 and the space size being 1GB) of the LUN1 as an example, the logical space corresponds to a physical space with the start address being xxxx and the space size being 1GB in the RAID-a, and the destroy command issued by the LUN1 to the RAID-a may include the start address (xxxx), the space size (1GB) and the destroy data of the first length of the physical space.

Here, it should be noted that the first length may be smaller than the size of the target physical space. For example, the first length is 1 byte. As an example, the one-byte destruction data may be 00000000, 11111111, 01010101, or the like.

It can be seen that compared with the standard write command (the data volume carried by the write command is consistent with the size of the address field to be written), the data volume carried by the destroy command is obviously reduced, and therefore, the bandwidth consumption in the transmission process can be effectively reduced.

In step 102, the first target RAID determines at least one target stripe involved in the target physical space according to the address range of the target physical space.

Each member disk in the RAID may be divided into a plurality of stripes according to a preset size. The same position bar in each member disk constitutes a strip.

The first target RAID may determine, based on the address range of the target physical space, the respective stripes involved in the target physical space. Each slice involved in the target physical space is referred to herein as a target slice. It is to be understood that the reference to target bands is a name for ease of distinction and is not intended to be limiting.

Here, it should be noted that the target stripes referred to by the target physical space include a part of the target stripes falling into the target physical space (for example, if the starting address of the target physical space is located in the middle of a stripe, the stripe is a target stripe corresponding to the target physical space) and all target stripes falling into the target physical space (that is, the whole stripe falls into the target physical space).

The first target RAID performs the following destruction processing for each target stripe:

step 103, if all the target stripes fall into the target physical space, the first target RAID writes the destroyed data into the target stripes, and does not perform RAID redundancy processing on the target stripes for the write operation.

And if the whole target band falls into the target physical space, which indicates that all the data in the target band are the data of the target LUN, and all the data need to be destroyed, writing the destroyed data into the whole target band. For example, if the destroy command carries one byte of destroy data, and the size of the target stripe is 512KB, 512K pieces of destroy data are written into the target stripe.

Because the destroyed data is written into the whole target stripe, that is, no valid user data exists in the target stripe, the RAID redundancy processing may not be performed on the target stripe for this write operation, so as to improve the destruction efficiency.

And 104, if the target stripes do not fall into the target physical space, writing the destroyed data into the address field of the target stripe falling into the target physical space by the first target RAID, and executing RAID redundancy processing on the target stripe.

If the target stripe does not fall into the target physical space in the whole, it means that only part of the data of the target stripe (the data of the target stripe falling into the address field of the target physical space) is the data of the target LUN, then only the destroyed data is written into the address field of the target stripe falling into the target physical space. For example, the target stripe size is 512KB, and only the second half stripe (256KB) falls into the target physical space, then the destroyed data is written into the last 256KB address field, for example, 256K destroyed data with a size of one byte is written.

Since valid user data still exists in the target stripe, this write operation also needs to perform RAID redundancy processing on the target stripe.

Thus, the flow shown in fig. 1 is completed.

As can be seen from the process shown in fig. 1, in the embodiment of the present application, the LUN directly issues the destroy command to the RAID without writing the destroy data into the cache and then flushing the disk to the RAID, so that cache resources can be effectively saved, and the destroy efficiency is improved. In addition, when the RAID executes the destruction command, if the destruction of the entire band is involved, RAID redundancy processing is not executed on the band for this destruction, so that RAID computing resources can be saved, and the destruction efficiency is improved.

It should be added that, in the embodiment of the present application, a maximum destruction range of the destruction command may be preset, for example, 2 GB. When the size of the target physical space is larger than the preset maximum destruction range, the target physical space can be divided into a plurality of sub-target physical spaces, and a destruction command is issued for each sub-target physical space.

For example, if the size of the physical space in the 2 nd mapping relationship in table 1 is 5GB, the 5GB physical space can be split into 3 sub-physical spaces with "the starting address is yyyy, the space size is 2 GB", "the starting address is yyyy +2GB, the space size is 2 GB", and "the starting address is yyyy +4GB, and the space size is 1 GB", and the destroy command is issued for each sub-physical space. That is, the destroy command includes the address range of the sub-physical space and the destroy data of the first length. The subsequent processing flows are the same, and are not described herein again.

As an embodiment, in order to ensure the destruction effect, the destruction order further includes the number of times of destruction. The first target RAID directly executes the destruction number times destruction processing for the target stripe (step 103 or step 104). For example, if the number of times of destruction is 5, the first target RAID directly executes 5 times of destruction processing on the target stripe.

Because the destruction commands directly carry the destruction times, compared with issuing a plurality of single destruction commands, the network bandwidth can be effectively saved. In addition, if the member disk of the RAID is a mechanical disk, multiple times of destruction are directly executed on the same position, and the problem of low destruction efficiency caused by repeated addressing based on a plurality of single destruction commands can be effectively solved.

Further, the destruction command may include the same number of destruction data of the first length as the number of times of destruction. For example, if the number of times of destruction is 3 and the first length is one byte, the destruction command may include three destruction data, i.e., "00000000", "11111111" and "01010101", as an example.

The first target RAID may perform destruction processing on the target stripe by using different destruction data at each destruction, for example, first write "00000000", second write "11111111", and third write "01010101", so as to achieve a better destruction effect.

The following describes a process in which the first target RAID writes the destroyed data to the target stripe. Referring to fig. 2, an implementation flow of writing the destroyed data to the target stripe by the first target RAID is shown in the embodiment of the present application.

As shown in fig. 2, the process may include the following steps:

step 201, the first target RAID determines a target member disk to which each target stripe block in which the destroyed data needs to be written belongs in the target stripe.

As previously mentioned, part or all of the data in the target strip may need to be destroyed. Therefore, it is necessary to determine which target blocks of the target stripe are located by the data to be destroyed, and the target member disk to which each target block belongs, so that the first target RAID destroys the data to be destroyed in the target block corresponding to the target member disk.

The target bar and the target member disk are named for convenience of distinction and are not intended to be limiting.

Step 202, for each target member disk, the first target RAID issues a write command to the target member disk.

The write command includes a second length of destruction data. Wherein the destruction data of the second length is generated based on the destruction data of the first length.

Here, the second length is the length of the address field of the target member disk corresponding to the target stripe block to which the destroyed data is to be written. It is to be understood that the reference to the second length is merely a designation for convenience of distinction and is not intended to be limiting.

For example, the destruction data carried by the destruction command is "00000000", and the length of the destruction data is 1 byte. The length of the data to be destroyed (or, alternatively, the data to be written to destroy) in the target stripe is 256KB, then the write command should carry 256KB of the length of the data to be destroyed (256K "00000000").

This is determined by the write command format of the disk (the amount of data carried is equal to the size of the area to be written). That is to say, when the first target RAID executes destruction, the first target RAID needs to convert the destruction command (carrying a small amount of destruction data) into a write command format supported by the target member disk and issue a write command to the target member disk, so as to implement data destruction.

Thus, the flow shown in fig. 2 is completed.

Writing the destroyed data into the disk is realized through the process shown in fig. 2, and data destruction is completed.

Furthermore, it should be added that before the destruction process shown in fig. 1 is executed, the front-end IO may be suspended, and the front-end server is no longer allowed to access the target LUN. Meanwhile, the target LUN does not issue new disk write commands any more, and waits for the return results of all issued disk write commands, so as to avoid processing exception of the issued commands.

After receiving the returned result, all cache blocks corresponding to the target LUN may be recycled. This is because the data in the target LUN is meaningless (needs to be destroyed), and therefore, the data does not need to be flushed to the disk, and the destruction efficiency can be further improved.

The method provided by the embodiment of the present application is described above, and the storage device provided by the embodiment of the present application is described below:

the storage equipment comprises at least one LUN and at least one RAID, wherein data of a target LUN corresponding to the at least one LUN is stored in the at least one target RAID, the target LUN is divided into a plurality of logical spaces, and the logical spaces of the target LUN correspond to physical spaces in the target RAID one to one and are equal in size;

For one embodiment, the first length is less than the target physical space size.

As an embodiment, the destruction command further includes destruction times, and the destruction command includes destruction data of a first length with the same number as the destruction times;

For one embodiment, the first target RAID comprises a plurality of member disks;

As an embodiment, the storage device further comprises a cache, the cache being divided into a plurality of cache blocks;

The above description is only a preferred embodiment of the present application, and should not be taken as limiting the present application, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present application shall be included in the scope of the present application.

Claims

1. A LUN destroying method is applied to a storage device, the storage device comprises at least one LUN and at least one RAID, data of a target LUN in the at least one LUN is stored in at least one target RAID, the target LUN is divided into a plurality of logical spaces, the logical spaces of the target LUN correspond to physical spaces in the target RAID one to one and are equal in size, and the method comprises the following steps:

the following processing is performed for each logical space:

the following destruction process is performed for each target band:

2. The method of claim 1, wherein the first length is less than the target physical space size.

3. The method of claim 1, wherein the destruction command further comprises a number of destruction times, the destruction command comprising a same number of destruction data of a first length as the number of destruction times;

the executing of the destruction process for each target band includes:

4. The method of claim 1, wherein the first target RAID comprises a plurality of member disks, the first target RAID writing destruction data to a target stripe comprising:

5. The method of claim 1, wherein the storage device further comprises a cache, the cache divided into a plurality of cache blocks, the target LUN prior to issuing the destroy command to the first target RAID, the method further comprising:

6. The storage device is characterized by comprising at least one LUN and at least one RAID (redundant array of independent disks), wherein data of a target LUN in the at least one LUN is stored in the at least one target RAID, the target LUN is divided into a plurality of logical spaces, and the logical spaces of the target LUN correspond to physical spaces in the target RAID one to one and are equal in size;

7. The apparatus of claim 6, wherein the first length is less than the target physical space size.

8. The device of claim 6, wherein the destruction command further comprises a number of destruction times, the destruction command comprising a same number of destruction data of a first length as the number of destruction times;

9. The apparatus of claim 6, wherein the first target RAID comprises a plurality of member disks;

10. The device of claim 6, wherein the storage device further comprises a cache, the cache divided into a plurality of cache blocks;