WO2020024933A1

WO2020024933A1 - Data writing method and server

Info

Publication number: WO2020024933A1
Application number: PCT/CN2019/098338
Authority: WO
Inventors: 陈伟; 汪渭春
Original assignee: 杭州海康威视系统技术有限公司
Priority date: 2018-07-31
Filing date: 2019-07-30
Publication date: 2020-02-06
Also published as: CN110795386B; CN110795386A

Abstract

The present application provides a data writing method and a server. In the present application, metadata is distinguished from general data, the distinguished metadata is permanently stored in a cache area, and the distinguished general data is written into a block device, so that a case where mixed flashing of the metadata and the general data causes frequent movement of a magnetic head between a general data storage position and a metadata storage position of the block device can be avoided, thereby improving the read-write performance of data.

Description

Data writing method and server

Technical field

The present application relates to the field of computer technology, and in particular, to a data writing method and a server.

Background technique

A block device is a device that can read a certain length of data from any location on the device, including hard disks, disks, USB flash drives, and flash memory. The operating system uses a part of the system memory as a cache area. When receiving the data to be stored, the operating system can first save the data to be stored in the cache area, and then write some data in the cache area according to the existing elimination algorithm. Block device saved.

The software organization responsible for managing and storing file information in the operating system is called a file system, and the data in the file system includes general data and metadata. Among them, general data refers to the actual data in ordinary files, and metadata is system data used to describe the characteristics of a file, such as access rights, file owner, and distribution information of file data blocks (such as Inode). Ideally, storing metadata and some popular general data in the cache area can speed up the read and write operations of block devices.

However, since the block device is generally used to mount the file system, when the block device has been formatted as a file system, there is no file attribute information after the data reaches the block device, and the operating system cannot accurately distinguish the received data. Data is also metadata, which may cause important metadata to be eliminated from the cache.

Summary of the invention

In view of this, the present application provides a data writing method and a server, which are used to distinguish metadata from general data, and write the distinguished metadata into a cache area to optimize the read and write performance of a block device.

According to a first aspect of the embodiments of the present application, a data writing method is provided. The method is applied to a server and includes: receiving first to-be-stored data sent by a client and the first to-be-stored data in a block device. Write location; comparing the write location of the first to-be-stored data with the area location information of the first file system that has been generated in the server; the first file system has been generated for the server One of a plurality of file systems, each of the file system's area location information is generated according to the storage space size of the block device, and each of the file system's area location information includes the file system's metadata area and general data Location information of the area; when it is determined that the writing position of the first to-be-stored data is located in the metadata area of the first file system, writing the first to-be-stored data into a cache area of the server.

The second aspect of the embodiments of the present application provides a server, which has a function of implementing the method provided in the first aspect. The functions may be implemented by hardware, and may also be implemented by hardware executing corresponding software. The hardware or software includes one or more modules or units corresponding to the functions described above.

In an implementation manner, the server may include: a receiving unit, configured to receive the first to-be-stored data and the writing position of the first to-be-stored data sent by the client in a block device; a comparison unit, Comparing the writing position of the first to-be-stored data with the area location information of the first file system that has been generated in the server; the first file system is a plurality of files that have been generated by the server One of the file systems, the area location information of each of the file systems is generated according to the storage space size of the block device, and the area location information of each of the file systems includes the location of a metadata area and a general data area of the file system Information; a write data unit, configured to write the first data to be stored when the comparison unit determines that a write position of the first data to be stored is located in a metadata area of the first file system A cache area of the server.

In another implementation manner, the server includes a processor, a communication interface, a memory, and a bus, where the processor, the communication interface, and the memory are connected to each other through a bus; the memory stores machine-readable instructions, and the processor The method provided by the first aspect of the embodiments of the present application is executed by calling the machine-readable instructions.

In the embodiment of the present application, metadata is distinguished from general data, and the distinguished metadata is stored in the buffer area for a long time. The distinguished general data is written into the block device according to the method, which can avoid the mixed writing of metadata and general data, which leads to the magnetic head. Frequent movement between general data storage locations and metadata storage locations of block devices improves data read and write performance.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of an architecture provided by an embodiment of the present application;

2 is a flowchart of a method provided by an embodiment of the present application;

3 is a schematic diagram of partitioning a block device in an EXT2 system according to an embodiment of the present application;

4 is a schematic diagram of a logical structure of a block group in an EXT2 system according to an embodiment of the present application;

FIG. 5a, FIG. 5b, FIG. 5c, and FIG. 5d are schematic diagrams of an anti-check operation provided by an embodiment of the present application;

FIG. 6 is a module block diagram of a server provided in an embodiment of the present application; FIG.

FIG. 7 is a hardware structural diagram of a server provided in an embodiment of the present application.

detailed description

In the following, some terms used in the examples of the present application will be explained first.

Metadata: includes the data structure that manages the entire file system, the data structure that manages the allocation of space, the attributes and pointers of files, the attributes and contents of directories. Taking the EXT2 (second extended filesystem) file system as an example, superblocks, block group descriptors, data block bitmaps, inode bitmaps, inode tables, and directory entry pages all belong to the metadata of the file system.

General data: refers to the content of the file. Taking the EXT2 file system as an example, the general data of the file system is the file content stored in the data area.

Metadata area (matadata area): Refers to a contiguous space of cache addresses allocated in the cache. It is used to cache superblocks, block group descriptors, inode bitmaps, data block bitmaps, and inode information created during copy directories and compressed file decompression operations. Cache refers to a part of space divided from system memory.

General data area (data area): Refers to a contiguous space of cache addresses allocated in the cache. It is used to cache data information created during the copy directory and compressed file decompression operation.

The technical solution of the present application will be described below with reference to the accompanying drawings and embodiments of the specification.

Please refer to FIG. 1, which is an overall architecture diagram of an embodiment of the present application. The embodiment of the present application includes a client and a server. The server is connected to multiple block devices, and the server is connected between the client and the block device. The connection between the server and the client and between the server and the block device can be It is a network connection or an electrical connection. The server can set up a cache area for each of these block devices, and create a virtual device corresponding to the block device on the client. The process of the client saving data to the virtual device is equivalent to sending a write instruction to the server, and the server writes the data carried by the write instruction to the buffer area or block device. The process of the client extracting data from the virtual device is equivalent to Send a read instruction to the server, and the server extracts the corresponding data from the block device or the buffer area according to the access address of the read instruction and returns it to the client. In order to improve the read and write performance of block devices, this application mainly improves the process of writing data, which is manifested in the server's separation and identification of the data to be stored, distinguishing between metadata and general data, and making the distinguished metadata permanent. Save to buffer area.

Referring to FIG. 2, it is a schematic flow chart of the operation of the data writing method provided by the present application. For each client, when the server receives the data to be stored from the client for the first time, perform the following steps:

Step 201: The server receives the first to-be-stored data and the writing position of the first to-be-stored data sent by the client in the block device.

In the embodiment of the present application, the writing position of the first to-be-stored data in the block device is indicated by the following information: the starting writing position of the first to-be-stored data in the block device and the data size of the first to-be-stored data. By combining the start writing position and the data size of the first data to be stored, the start writing position and the end writing position of the first data to be stored in the block device may be determined.

Step 202: The server compares the writing position of the first to-be-stored data with the regional location information of the first file system generated in the server. The first file system is a plurality of file systems generated in the server. First, the area location information of each file system is generated according to the storage space size of the block device. The area location information of each file system includes the location information of the metadata area and the general data area of the file system.

Before step 201, the client applies for a block device from the server in advance. The size of the storage space of the block device that can be used by the client can be requested by the client or allocated uniformly by the server. After the server accepts the client's application for the use of the block device, it determines the block device for use by the client among the multiple block devices connected to the server, and creates a virtual device on the client. Creating a virtual device on the client is equivalent to mounting a block device on the client. This virtual device can be regarded as a mapping of the block device. For example, the client formatting the virtual device is equivalent to sending formatting instructions to the server. The server formats the block device allocated to the client; the client saves the data to the virtual device, which is equivalent to sending a write instruction to the server, and the server writes the data carried by the write instruction to the cache area or Block device.

If the client wants to use the block device as a file, it needs to format the block device into a file system. The design idea of this application is to judge the type of data by comparing the location information of the block device formatted with the write position of the data. If the data write location is in the metadata area of the formatted block device, the data is considered as metadata; if the data write location is in the general data area of the formatted block device, the data to be considered is General data.

However, in general, the server is not sure which file system the client uses to format the block device it uses. Therefore, in the embodiment of the present application, the server can set up a file system template library. File system structure, such as EXT2 (second extended filesystem), EXT3 (third extended filesystem), XFS, FAT (File Allocation Table), NTFS (New Technology File System) and so on. After the client determines the storage space size of the block device to be used, the server can use the storage space size as an input to the file system template library, and use the file system template library to establish a set that matches the size of the storage space. The regional location information of each file system; the size of the input storage space is different, and the regional location information of each file system is also different. For each client, the server can establish a set of area location information of the file system corresponding to the storage space size of the block device used by the client.

The following uses the EXT2 file system as an example to describe the area location information of the file system. Referring to FIG. 3, when a block device is formatted as an EXT2 file system, the block device is divided into a boot block and multiple block groups; the boot block refers to the boot sector of the file system. Can be used to install the boot program; each block group has its own metadata area and general data area, and the internal logical results of each block group are shown in Figure 4, including the general data area and the metadata area.

The logical structure of the metadata area and the general data area in the block group can be seen from FIG. 4. Generally, the structure of the data area is relatively single and consists of many blocks of the same size. The metadata area includes multiple sub-metadata areas.For example, the metadata area of the EXT2 file system consists of a Super Block, a GDT (group descriptor), an inode table, an inode bitmap, and data. A block bitmap is composed of five sub-metadata areas.

Step 203: When the server determines that the write position of the first data to be stored is located in the metadata area of the first file system, the server writes the first data to be stored into the cache area of the server.

In this embodiment, the server compares the writing position of the first to-be-stored data with the area position information of multiple file systems that have been generated in the server in sequence. When compared to the first file system, the first to-be-stored data is found. The start writing position and the ending writing position of the data are located in the same sub-metadata area of the first file system, and it is determined that the writing position of the first to-be-stored data is located in the metadata area of the first file system. At this time, the server determines that the first file system is a file system type that the block device used by the client may adopt, and determines that the first data to be stored may be metadata. After the first file system is found, the server can stop the comparison and directly write the first data to be stored into the cache area.

In another example, if no file system is found until the last file system is compared: the write position of the first to-be-stored data is located in the metadata area of the file system, the server can determine the first The stored data is general data. Next, the server can process the first to-be-stored data according to a set general data processing strategy. For example, the first to-be-stored data can be written directly to the block device, or the first to-be-stored data can be written to the cache area first. When the available space in the cache area is less than a certain value, the general data in the cache area is eliminated. A specific implementation method is as follows: When it is determined that the write position of the first data to be stored is located in multiple generated file systems In the general data area of each file system, determine the first to-be-stored data as general data, and determine whether the available space in the cache area of this server is larger than the occupied space of the first to-be-stored data; if so, The stored data is written into the cache area of this server; if not, according to the preset elimination algorithm, such as FIFO (First In, First Out) algorithm, LRU (Least, Recently Used) algorithm, LFU (Least Frequently Used, least frequently used) algorithm, etc., determine the obsolete data in the general data stored in the cache area of this server; according to the obsolete data, In position, selecting at least one data having a writing position adjacent from the buffer; writes a block of the present server apparatus together with at least one data out of the data having adjacent writing position.

Here, by aggregating a plurality of data with adjacent writing positions into a large data written to the block device at one time, compared with writing a plurality of data into the block device in batches, it is possible to reduce the back and forth of the magnetic head on the magnetic disk. Swing, as well as reducing the time it takes to write data to a block device. Similarly, when it is necessary to write the data in the buffer area that is considered as metadata into the block device, the aggregation method can also be used to write the metadata with adjacent write positions to the block device together.

It should be noted that, in the embodiment of the present application, after the data in the cache area is written to the block device, the written data also needs to be deleted from the cache area; when the data in the block device is written to the cache area, there is no need to When the delete operation is performed, the corresponding data remains in the block device.

The above describes the data writing process performed for each client when the server receives the data to be stored from the client for the first time. After that, when the server receives the data to be stored from the client again, the server can perform steps 201-203 again. Or, in another embodiment, in order to further improve the read and write performance of the block device, when the server performs steps 201-203 once and has determined the type of file system that the block device used by the client may adopt, the server When receiving the data to be stored subsequently sent by the same client, such as the second data to be stored, the following process may be performed:

The server receives the second to-be-stored data and the writing position of the second to-be-stored data sent by the client in the block device; the writing position of the second to-be-stored data in the block device includes the second to-be-stored data in the block device Start writing position and ending writing position.

The server compares the writing position of the second data to be stored with the area position information of the first file system. The first file system is the file system type that the block device used by the client determined in step 203 may adopt. Here, the write position of the second to-be-stored data is directly compared with the area position information of the first file system. Compare the write position of the second to-be-stored data with the regional position information of the multiple file systems corresponding to the client that has been generated.

In one case, if the start writing position and the ending writing position of the second data to be stored are located in the same sub-metadata area of the first file system, it is determined that the writing position of the second data to be stored is located in the first file system. In the metadata area. At this time, the server still judges the first file system as the type of file system that the block device used by the client may use, and determines that the second data to be stored may be metadata, and directly writes the second data to be stored in the cache. Area.

In another case, if the start writing position and the end writing position of the second data to be stored are in the general data area of the first file system, the server still determines that the first file system is a block used by the client. The file system type that the device may use, and determines that the second data to be stored may be general data. Next, the server can process the second data to be stored according to a set general data processing policy, and the specific processing process is not described in detail.

In another case, if the start writing position and the end writing position of the second data to be stored are in different sub-metadata areas of the first file system, or the start writing position of the second data to be stored is located in the first file The metadata area of the system and the end write location of the second to-be-stored data are outside the metadata area of the first file system, indicating that the first file system is not the file system type used by the block device used by the client. The server needs to re-compare the writing position of the second to-be-stored data with the regional position information of other file systems (except the first file system) in the multiple file systems that have been generated.

As an implementation manner, if no file system is found to satisfy: the start writing position and the end writing position of the second data to be stored are located in a metadata area of the file system, it is determined that the second data to be stored is normal data.

Conversely, if it is determined that the start writing position and the ending writing position of the second to-be-stored data are located in the same sub-metadata area of the second file system, at this time, the second file system may be determined to be a block used by the client. The file system type that the device may adopt, and determines that the second data to be stored may be metadata. After the second file system is found, the server can stop the comparison and directly write the second data to be stored into the cache area of the server.

Further, after determining that the second file system is a file system type that the client may use as the block device, the server may also perform a back-check operation to transfer the data previously misjudged as metadata to the block device, and Data that was previously misjudged as general data is rewritten into the cache area, as follows:

Referring to FIG. 5a, if the range [x1, y1] of the metadata area of the first file system includes the range [x2, y2] of the metadata area of the second file system, the server can save the start in the cache area of the server The data at the write position and / or the end write position are in [x1, x2], and the data at the start write position and / or the end write position are in [y2, y1] are written to the block device of this server.

Referring to FIG. 5b, if the metadata area range [x1, y1] of the first file system and the metadata area range [x2, y2] of the second file system intersect and x1> x2, y1> y2, the server can set Data stored in the start and / or end write locations of the server's buffer area in [y2, y1] is written to the block device of the server, and the start of write is stored in the block device of the server The data at the input position and the end write position are both in [x2, x1] and written into the cache area of this server.

Referring to FIG. 5c, if the metadata area range [x1, y1] of the first file system and the metadata area range [x2, y2] of the second file system intersect and x1 <x2, y1 <y2, the server can set Data stored in the start and / or end write locations of the server's cache area in [x1, x2] is written to the block device of the server, and the start of write is stored in the block device of the server The data at the entry position and the end writing position are located in [y1, y2] and written into the cache area of this server.

Referring to FIG. 5d, if the metadata area range [x2, y2] of the second file system includes the metadata area range [x1, y1] of the first file system, the server can save the start of the block device in the server The data where the write position and the end write position are both in [x2, x1], and the data where the start write position and the end write position are both in [y1, y2] are written into the cache area of this server.

So far, the description of the method provided in this application is completed.

After the block device is formatted as a file system, general data and metadata are usually stored separately in the block device. In the embodiment of the present application, the metadata is distinguished from the general data, and the distinguished metadata is stored in the cache area for a long time. By writing the distinguished general data to the block device, it is possible to avoid the mixed writing of metadata and general data, which causes frequent movement of the magnetic head between the general data storage location and the metadata storage location of the block device, and improves data read and write. performance.

Further, in the embodiment of the present application, by aggregating a plurality of data having adjacent write positions into a large data and writing it to the block device at one time, compared with writing a plurality of data into the block device in batches, the magnetic head is more reduced. The back-and-forth wobble and reduce the time it takes to write data to the block device.

The device provided in the embodiments of the present application is described below.

Referring to FIG. 6, FIG. 6 is a functional module diagram of a server provided in an embodiment of the present application. As shown in Figure 6, the server includes:

The receiving unit 601 is configured to receive first data to be stored and a writing position of the first data to be stored in a block device sent by a client;

A comparison unit 602 is configured to compare the writing position of the first to-be-stored data with the regional position information of the first file system generated in the server; the first file system is the service One of a plurality of file systems that have been generated in the client, each of the file system's area location information is generated according to the storage space size of the block device, and each of the file system's area location information includes metadata of the file system Location information of the data area and general data area;

The data writing unit 603 is configured to write the first data to be stored when the comparison unit 602 determines that the writing position of the first data to be stored is located in a metadata area of the first file system. The cache area of this server.

In one embodiment, the metadata area of each file system includes multiple sub-metadata areas, and the writing position of the first data to be stored in the block device includes the first data to be stored in the block device. The starting writing position and the ending writing position; the comparison unit 602 is configured to determine that the starting writing position and the ending writing position of the first to-be-stored data are located in the same sub-unit of the first file system When in the data area, it is determined that the writing position of the first data to be stored is located in the metadata area of the first file system.

In one implementation manner, the receiving unit 601 is further configured to receive the second to-be-stored data and the writing position of the second to-be-stored data in the block device sent by the client; the second The writing position of the data to be stored in the block device includes a start writing position and an end writing position of the second data to be stored in the block device;

The comparison unit 602 is further configured to compare the writing position of the second data to be stored with the area position information of the first file system; if the starting writing position of the second data to be stored is And the end writing position are located in different sub-metadata areas of the first file system, or the start writing position of the second to-be-stored data is located in the metadata area of the first file system and the second The end writing position of the data to be stored is outside the metadata area of the first file system, and then the writing position of the second data to be stored and the area position of the second file system in the plurality of file systems Information comparison

The writing data unit 603 is further configured to: when the comparing unit 602 determines that the start writing position and the ending writing position of the second to-be-stored data are located in a same sub-metadata area of the second file system , Writing the second to-be-stored data into a cache area of the server.

In one embodiment, the write data unit 603 is further configured to: if the metadata area range [x1, y1] of the first file system includes the metadata area range [x2, y2], the data stored in the start write position and / or the end write position in [x1, x2] stored in the cache area of this server, and the start write position and / Or the end write data located in [y2, y1] is written to the block device of this server.

In one implementation manner, the write data unit 603 is further configured to: if the range of the metadata area [x1, y1] of the first file system and the range of the metadata area [x2, y2] if there is an intersection and x1> x2, y1> y2, then write the data stored in the cache area of the server and / or the end write location in [y2, y1] to the block of the server The device, and the data stored in the start writing position and the end writing position stored in the block device of the server are both in [x2, x1] and written to the buffer area of the server.

In one implementation manner, the write data unit 603 is further configured to: if the range of the metadata area [x1, y1] of the first file system and the range of the metadata area [x2, y2] if there is an intersection and x1 <x2, y1 <y2, the data stored in the start and / or end write locations in the cache area of this server is written to the block of this server The device, as well as the data stored in the block device of the server at the beginning and end of writing, are located in the buffer area of the server.

In one embodiment, the write data unit 603 is further configured to: if the metadata area range [x2, y2] of the second file system includes the metadata area range [x1, y1], the data stored in the start block and end write positions of the server block device are both located in [x2, x1], and the start write position and end are stored in the block device of this server end The data whose write positions are located in [y1, y2] are written into the cache area of this server.

In one implementation manner, the write data unit 603 is further configured to, when it is determined that the write position of the first data to be stored is located in a general data area of each file system of the multiple file systems, Determine that the first data to be stored is general data, and determine whether the available space in the cache area of the server is larger than the occupied space of the first data to be stored; if so, write the first data to be stored in the service The cache area on the server side; if not, the elimination data is determined from the general data stored in the cache area of the server side according to a preset elimination algorithm; according to the write position of the elimination data, an adjacent area is selected from the cache area on the server side Write at least one data at a location; write the obsolete data together with the at least one data with an adjacent write location into a block device on the server side.

It should be noted that the division of the units in the embodiments of the present application is schematic, and is only a logical function division. There may be another division manner in actual implementation. The functional units in the embodiments of the present application may be integrated into one processing unit, or each of the units may exist separately physically, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional unit.

For details about the implementation process of the functions and functions of the units in the foregoing device, see the implementation process of the corresponding steps in the foregoing method for details, and details are not described herein again.

So far, the description of the device shown in FIG. 6 is completed.

Correspondingly, the embodiment of the present application also provides a hardware structure of the server. Referring to FIG. 7, FIG. 7 is a schematic diagram of a hardware structure of a server provided in an embodiment of the present application. The server includes a processor 701, a communication interface 702, a memory 703, and a bus 704. Among them, the processor 701, the communication interface 702, and the memory 703 Connected to each other via a bus.

A communication interface 702, configured to communicate between the server and the client;

A memory 703, configured to store machine-readable instructions;

The processor 701 executes the method shown in FIG. 2 by calling the machine-readable instructions.

An embodiment of the present application further provides a machine-readable storage medium that stores machine-readable instructions. When the machine-readable instructions are called and executed by a processor, the machine-executable instructions cause The processor implements the method shown in FIG. 2.

The above are only preferred embodiments of this application, and are not intended to limit this application. Any modification, equivalent replacement, or improvement made within the spirit and principles of this application shall be included in this application Within the scope of protection.

Claims

A data writing method applied to a server includes:

Receiving the first to-be-stored data and the writing position of the first to-be-stored data in the block device sent by the client;

Comparing the writing position of the first to-be-stored data with the regional position information of the first file system that has been generated in the server; the first file system is a plurality of files that have been generated in the server One of the file systems, the area location information of each of the file systems is generated according to the storage space size of the block device, and the area location information of each of the file systems includes the location of a metadata area and a general data area of the file system information;

When it is determined that the writing position of the first to-be-stored data is located in a metadata area of the first file system, the first to-be-stored data is written into a cache area of the server.
The method according to claim 1, wherein:

The metadata area of each of the file systems includes a plurality of sub-metadata areas,

The writing position of the first data to be stored in the block device includes a start writing position and an end writing position of the first to be stored data in the block device;

Determining that the writing position of the first to-be-stored data is located in a metadata area of the first file system includes:

If the start writing position and the ending writing position of the first to-be-stored data are located in the same sub-metadata area of the first file system, determine that the writing position of the first to-be-stored data is located in the first A file system's metadata area.
The method of claim 2, further comprising:

Receiving the second to-be-stored data and the writing position of the second to-be-stored data in the block device sent by the client; the writing position of the second to-be-stored data in the block device includes the The start writing position and the end writing position of the second to-be-stored data in the block device;

Comparing the writing position of the second data to be stored with the area position information of the first file system;

If the start writing position and the ending writing position of the second to-be-stored data are located in different sub-metadata areas of the first file system, or the start writing position of the second to-be-stored data is located in the first A file system's metadata area and the end writing position of the second to-be-stored data is outside the metadata area of the first file system, then writing the second to-be-stored data's writing position with the Comparing the location information of the second file system in the multiple file systems;

When it is determined that the start writing position and the ending writing position of the second to-be-stored data are located in the same sub-metadata area of the second file system, write the second to-be-stored data into a cache of the server Area.
The method according to claim 3, wherein after determining that the start writing position and the ending writing position of the second to-be-stored data are located in the same sub-metadata area of the second file system, the method The method also includes:

If the range [x1, y1] of the metadata area of the first file system includes the range [x2, y2] of the metadata area of the second file system, start writing to the cache area of the server Data whose position and / or end write position is in [x1, x2], and data write start and / or end write positions stored in the cache area of the server are written in [y2, y1] Into the block device.
The method according to claim 3, wherein after determining that the start writing position and the ending writing position of the second to-be-stored data are located in the same sub-metadata area of the second file system, the method The method also includes:

If the metadata area range [x1, y1] of the first file system and the metadata area range [x2, y2] of the second file system intersect and x1> x2, y1> y2, then

Writing data in the start writing position and / or ending writing position in [y2, y1] stored in the cache area of the server to the block device, and

Write the data stored in the block device at the start writing position and the end writing position in [x2, x1] into the cache area of the server.
The method according to claim 3, wherein after determining that the start writing position and the ending writing position of the second to-be-stored data are located in the same sub-metadata area of the second file system, the method The method also includes:

If the metadata area range [x1, y1] of the first file system and the metadata area range [x2, y2] of the second file system intersect and x1 <x2, y1 <y2, then

Writing data in the start writing position and / or the end writing position in [x1, x2] stored in the cache area of the server to the block device, and

Write the data stored in the block device at the start writing position and the end writing position in [y1, y2] into the cache area of the server.
The method according to claim 3, wherein after determining that the start writing position and the ending writing position of the second to-be-stored data are located in the same sub-metadata area of the second file system, the method The method also includes:

If the range [x2, y2] of the metadata area of the second file system includes the range [x1, y1] of the metadata area of the first file system, the start writing position and The data whose end writing position is located in [x2, x1], and the data whose start writing position and end writing position are stored in [y1, y2] stored in the block device are written into the cache of the server Area.
The method of claim 1, further comprising:

When it is determined that the writing position of the first to-be-stored data is located in a general data area of each of the file systems in the plurality of file systems,

Determining that the first data to be stored is general data,

Determining whether the available space in the cache area of the server is larger than the occupied space of the first data to be stored;

If the available space of the cache area is larger than the occupied space of the first data to be stored, writing the first data to be stored into the cache area of the server;

If the available space of the buffer area is less than or equal to the occupied space of the first data to be stored, then

Determining elimination data from general data stored in a cache area of the server according to a preset elimination algorithm;

Selecting at least one data with an adjacent write position from a cache area of the server according to the write position of the obsolete data; and

Write the obsolete data together with the at least one data into the block device.
The method of claim 1, further comprising:

Establish a file system template library;

Sending the size of the storage space of the block device as an input parameter into the file system template library;

Use the file system template library to generate region location information of the multiple file systems that match the storage space size of the block device.
A server, including:

A receiving unit, configured to receive the first to-be-stored data sent by the client and a writing position of the first to-be-stored data in the block device;

A comparison unit, configured to compare the writing position of the first to-be-stored data with the area position information of the first file system that has been generated in the server; the first file system is the server One of a plurality of file systems that have been generated in the system, each of the file system's area location information is generated according to the storage space size of the block device, and each of the file system's area location information includes a metadata area of the file system And the location information of the general data area;

A data writing unit, configured to write the first data to be stored when the comparison unit determines that a writing position of the first data to be stored is located in a metadata area of the first file system; The cache area of the server.
A server includes a processor, a communication interface, a memory, and a bus, wherein the processor, the communication interface, and the memory are connected to each other through the bus;

Machine-readable instructions are stored in the memory, and the processor executes the method according to any one of claims 1 to 9 by calling the machine-readable instructions.