CN106502587B - Hard disk data management method and hard disk control device - Google Patents

Abstract

The embodiment of the invention discloses a hard disk data management method and a hard disk control device, which are used for efficiently managing fragments on the hard disk. The embodiment of the present invention is applied to a hard disk control device including a hard disk. The hard disk includes a data area and a log area. The method includes: writing data to the cache device; After the modification and release of the set number of times, the hard disk can generate data with a preset number of fragments; if the data is not hot data, allocate data area space for the data in the data area, and write the data into the data area space; if the data is hot data, then Allocate the log area space for the data in the log area, and write the data into the log area space. By storing different types of data in different areas on the hard disk and managing them in different ways, the efficiency of fragment management on the hard disk can be improved, and the efficient management of hard disk fragments in the log area can reduce the generation of hard disk fragments.

Description

Hard disk data management method and hard disk control device

technical field

The invention relates to the field of data processing, in particular to a hard disk data management method and a hard disk control device.

Background technique

For ordinary mechanical hard disks, because they rely on mechanically rotating the hard disk and moving the magnetic head to position the reading and writing position, sequential reading and writing of the hard disk is the most ideal reading and writing model. If the hard disk space is fragmented, continuous space cannot be allocated when writing data, resulting in severe head vibration. The main time of data transmission is spent on positioning tracks and sectors, leaving little time for data transmission. Because the data of the file is relatively discrete, the efficiency of reading these files is also low.

Therefore, most hard disk file systems are trying their best to avoid creating a large amount of fragmented space, but fragmentation is still unavoidable.

For example, using the COW mechanism can take advantage of the hard disk sequential write. When you want to modify and write a piece of data, instead of directly overwriting the old version of the data, you read the old version of the data. After the modification, write to a new location, aggregate the written data, and write them to the hard disk sequentially. , release the old version of the data. Because the location of the data has changed, it is necessary to modify the pointer to the index block of the previous layer of the data, so as to recurse to the topmost layer. This will release a large amount of data, resulting in a large amount of fragmentation on the hard disk.

Contents of the invention

Embodiments of the present invention provide a hard disk data management method and a hard disk control device for efficient management of fragments on a hard disk.

The first aspect of the present invention provides a hard disk data management method. The method is applied to a hard disk control device including a hard disk. The hard disk includes a data area and a log area. The method includes:

The hard disk control device writes data to the cache device, and the cache device can be, for example, a storage device different from the hard disk such as a memory, a flash card, a solid state disk, and then, the hard disk control device judges whether the data is hot data, wherein the hot data is stored in After the preset number of modifications and releases on the hard disk, the hard disk can generate a preset number of fragmented data. By judging the written data on the cache device, the type of the data is determined, so as to execute different processing methods for different data.

When writing data to the hard disk, if the data is not hot data, allocate data area space for the data in the data area, and write the data into the data area space; if the data is hot data, allocate log area for the data in the log area space, write data into the log area space.

The hard disk data management method of the first aspect of the present invention divides the data to be written into the hard disk into hot data and non-hot data, hot data is likely to cause fragmentation of the hard disk, and the hot data is stored in the log area and managed in a log manner, Even if the data in the log area is frequently modified to cause hard disk fragments, it is convenient to recover and manage these fragments, and save non-hot data in the data area, the release of non-hot data will not easily cause hard disk fragments, and the data area does not need to be hard disk fragments Manage the allocation of excessive resources. Therefore, by saving different types of data in different areas on the hard disk and managing them in different ways, the efficiency of fragmentation management on the hard disk can be improved, and the fragmentation on the hard disk can be effectively managed to avoid or reduce Hard disk fragmentation occurs.

In combination with the first aspect, in the first possible implementation manner, the cache device is a memory, and after the log area space is allocated for the data in the log area, the first possible implementation manner further includes: establishing a relationship between the data and the log area space Mapping relations. That is, after the hard disk control device allocates log area space for the data in the log area, it establishes a mapping relationship on the cache device, and records the corresponding relationship between the data and the log area space allocated to it, and records the data in the log area through the mapping relationship. Storage conditions, so that the mapping relationship can be used to manage the data in the log area and the data elimination operation on the cache device. Wherein, in the first possible implementation manner, the cache device is a memory, but the cache device may also be other situations.

In combination with the first possible implementation of the first aspect, the mapping relationship between data and log area space is established in the second possible implementation, including: establishing multiple target data and the log area space to which multiple target data are allocated The mapping relationship of , where the target data belongs to hot data;

Writing data into the log area space includes: combining multiple write operations of multiple target data into one transaction; writing all target data of the transaction into the log area space. And when the write operation of one of the target data of the transaction fails, the write operations of other target data of the transaction fail. A plurality of target data refers to at least two target data, and correspondingly, a plurality of write operations refers to at least two write operations. In this way, the concept of transactions in the database domain is introduced, and multiple hot data are used as a unit to operate, such as establishing a mapping relationship with multiple hot data belonging to the same transaction, and performing write operations to the log area together with the write operation of all hot data of the transaction write operation. This can improve the efficiency of data processing.

In combination with the second possible implementation manner of the first aspect, a third possible implementation manner further includes: caching data belonging to hot data in memory. Cache hotspot data in memory, for example, when writing hotspot data to log area, keep these data in memory or read hotspot data from log area before writing data to memory, and cache In the memory, in this way, when data is subsequently written to the content, the data can be directly modified in the memory, and the data is migrated in the memory, which reduces the generation of fragments on the hard disk, and can organize the data in the log area according to the migration situation .

In combination with the third possible implementation of the first aspect, before writing all target data of the transaction into the log area space in the fourth possible implementation, the fourth possible implementation also includes:

Establish a data linked list based on multiple target data, wherein the data linked list is used to manage the target data, and the target data managed by the data linked list is the same as the target data of the transaction; then, the target data is managed according to the data linked list, and the target data is managed according to the data linked list Data management includes: after the second data link list is established, when the second target data managed by the second data link list is modified from the first target data managed by the pre-established first data link list, the first data link list is released. Management of the first target data; deleting information of the first target data on the first mapping relationship corresponding to the first data linked list. In this way, in the memory, data migration between different transactions can be managed through the data linked list.

The way to manage the data on the memory according to the data link list can be: under the preset release condition, according to the establishment sequence of the data link list from first to last, find the data that has not been released from the data link list; release the target data link list on the memory. The managed data is released, and the target mapping relationship corresponding to the target data linked list is reserved in memory. The capacity of the memory management data can be expanded by releasing the data on the target data linked list. The system can read the corresponding data from the log area by querying the target mapping relationship.

In combination with the fourth possible implementation of the first aspect, in the sixth possible implementation, under the preset release condition, according to the establishment sequence of the data linked list from first to last, search for data that has not been released from the data linked list, including : When the memory reaches the first preset water level, according to the establishment order of the data link list from first to last, search for the data that has not been released from the data link list; in addition, this method also includes the second stage of memory data elimination, that is, release on the memory After the target data link list has not released the managed data, the method also includes: when the memory reaches the second preset water level, reading the data pointed to by the target mapping relationship from the log area; writing the data pointed to by the target mapping relationship into the data area; Delete the target mapping relationship in memory. Through the two-stage memory data elimination mechanism, the management capacity of memory for data can be expanded. In the second stage of memory data elimination, the data pointed to by the target mapping relationship at this time is inactive data, and the possibility of being modified is low. These data can be migrated from the log area to the data area for storage, which will not increase the fragmentation of the data area too much.

In combination with the fourth possible implementation manner of the first aspect, in the sixth possible implementation manner, the method further includes: assigning a transaction number to the data linked list corresponding to the transaction according to an increment rule according to the writing sequence of the transaction. By assigning a transaction number to the data link list, the data link list can be managed according to the transaction number, which improves the management efficiency. For example, starting from the data linked list with the smallest current transaction number, search for the data that has not been released from the data linked list according to the order of the transaction number from small to large, so that you can find the unreleased data linked list according to the order in which the data linked list is established from first to last managed data.

In combination with the fourth possible implementation manner of the first aspect, in a seventh possible implementation manner, the method further includes: under preset recycling conditions, performing a step of relocating log area data. For example, the steps to perform data relocation in the log area include:

Find the mapping relationship;

According to the information recorded in the mapping relationship, it is judged whether the data on the first log area corresponding to the mapping relationship has been migrated;

If the data on the first log area has not been migrated, then determine the space utilization rate on the first log area according to the information recorded in the mapping relationship;

When the space utilization rate of the first log area is less than the preset utilization threshold, the data in the first log area is migrated to the second log area, and the mapping relationship corresponding to the relocated data is updated, where the second log area is A free log area or a log area that has been used when recycling a log area;

When the total space water level of the current log area reaches the preset space threshold, the step of relocating the data in the log area is stopped; otherwise, the step of relocating the data in the log area is continued.

In combination with the seventh possible implementation of the first aspect, in the eighth possible implementation, the method further includes: assigning a transaction number to the data linked list corresponding to the transaction according to the increment rule according to the write order of the transaction, and assigning a transaction number according to the transaction number Manage the data linked list to improve management efficiency. For example, starting from the data linked list with the smallest current transaction number, the mapping relationship corresponding to the data linked list can be searched according to the sequence of transaction numbers from small to large, so as to realize the search for the mapping relationship.

In combination with the seventh possible implementation of the first aspect, the preset recovery conditions in the ninth possible implementation include timer timeout, completion of the recovery operation on memory data, and the total water level in the log area reaching the preset water level threshold at least one of the .

Combined with the first possible implementation of the first aspect, in the tenth possible implementation, there are many ways to cache data belonging to hot data in memory, for example, after judging whether the data is hot data, if the data is For hot data, keep the data in the memory; or, before writing data to the cache device, read data from the log area to the cache of the cache device.

In combination with the first aspect or any of the second to tenth possible implementations of the first aspect, in the eleventh possible implementation, the hotspot data includes data and/or hotspots whose data size is smaller than a preset data threshold Data includes metadata. The preset data threshold can be, for example, 128KB or other space size, and the specific value can be adjusted according to the business type. If the data size of the data is smaller than the preset data threshold, the frequent release of the data may cause a large amount of data to be generated on the hard disk. debris. The metadata includes data management data, such as an indirect block for storing data addresses, and a metadata block for storing object management structures. Metadata can also cause massive fragmentation of the hard drive. These hot data should be screened out to save in the log area.

In combination with the first aspect or any of the second to tenth possible implementations of the first aspect, in the twelfth possible implementation, allocate log area space for data in the log area, and write data into the log area space , including: allocating log area space for data in the log area in sequence, and sequentially appending data to the log area space. In this way, the sequential reading and writing of data in the log area can be realized, so that there is no metadata overhead when relocating the data in the log area. The sorting overhead is relatively small, effectively ensuring the stability of system performance.

In combination with the first aspect or any of the second to tenth possible implementation manners of the first aspect, in the thirteenth possible implementation manner, the method further includes: when the space utilization rate of the data area is greater than the preset data area When the threshold is used, the currently free log area is converted into a data area; when the space utilization rate of the log area is greater than the preset log area utilization threshold, the data area converted from the idle log area is converted into a log area. In this way, the log area and the data area are transformed into each other to adapt to changes in system capacity. It can flexibly adapt to specific usage scenarios and improve the efficiency of hard disk usage.

In combination with the first aspect or any of the second to tenth possible implementation manners of the first aspect, in the fourteenth possible implementation manner, the hard disk further includes a super block, each log section is provided with identification information, and the super block It is used to record the identification information of the modified log area after the log area is modified. The log area is further managed through the super block. For example, if the system is powered off or recovers from a crash, the hard disk control device can search for the modified log area in time according to the information recorded in the super block.

With reference to the first aspect or any of the second to tenth possible implementation manners of the first aspect, in a fifteenth possible implementation manner, the log area and the data area are arranged alternately on the hard disk. In this way, the data in the log area and the data area can be set closer.

In combination with the first aspect or any of the second to tenth possible implementation manners of the first aspect, in the sixteenth possible implementation manner, the hard disk further includes a block group, and the block group includes a preset number of log areas and data area, the log area and data area of the block are set continuously, and the use of the data area and log area in the block can be adjusted through the block. For example, after the free target data area is determined according to the management information of the block group, the The target data area allocates the data area space; thus after the data is written into the data area space, the method also includes: generating target metadata according to the data and the data area space; writing the target metadata to the cache device, and the hard disk control device determines that the metadata is After hot data, determine the target block group to which the target data area belongs; determine the available log area of the target block group; write the target metadata into the available log area. In this way, the position of the metadata on the hard disk can be close to the storage position of the data corresponding to the metadata on the hard disk, which facilitates reading and writing of data.

In combination with any of the second to tenth possible implementations of the first aspect, in the seventeenth possible implementation, the method further includes: allocating log area space for the mapping relationship in the log area, and then, mapping The relationship is written to the log area space allocated by the mapping relationship. That is, the mapping relationship is also stored in the log area, so that the mapping relationship is reliably stored on the hard disk.

The second aspect of the present invention provides a hard disk control device, the hard disk control device includes a hard disk, the hard disk includes a data area and a log area, and the hard disk control device has the functions of the hard disk control device in the above method. This function may be realized by hardware, and may also be realized by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the above functions.

In a possible implementation manner, the hard disk control device includes:

A write unit, used to write data to the cache device;

The cache manager is used to determine whether the data is hot data, wherein the hot data is data that can cause the hard disk to generate a preset number of fragments after being modified and released for a preset number of times after being stored on the hard disk;

The data manager is used to allocate data area space for the data in the data area if the data is not hot data, and write the data into the data area space;

The log manager is used to allocate log area space for the data in the log area if the data is hot data, and write the data into the log area space.

In another possible implementation manner, the hard disk control device includes:

processor;

The processor performs the following actions: writing data to the cache device;

The processor performs the following actions: judging whether the data is hot data, wherein the hot data is data that can cause the hard disk to generate a preset number of fragments after being modified and released for a preset number of times after being stored on the hard disk;

The processor performs the following actions: if the data is not hot data, allocate data area space for the data in the data area, and write the data into the data area space;

The processor performs the following actions: if the data is hot data, allocate log area space for the data in the log area, and write the data into the log area space.

In a third aspect, an embodiment of the present application provides a computer storage medium, where the computer storage medium stores a program code, and the program code is used to instruct execution of the method in the first aspect above.

It can be seen from the above technical solutions that the embodiments of the present invention have the following advantages:

On a hard disk control device including a hard disk, the hard disk includes a data area and a log area. After writing data to the cache device, the hard disk control device judges whether the data is hot data. If the data is not hot data, the data is in the data The data area is allocated in the data area space, and the data is written into the data area space; if the data is hot data, the log area space is allocated for the data in the log area, and the data is written into the log area space. In this way, the data to be written into the hard disk is divided into hot data and non-hot data. Hot data is data that can cause the hard disk to generate a preset number of fragments after being stored on the hard disk after a preset number of modifications and releases. Hot data is easy to cause The hard disk generates fragments, save the hotspot data in the log area, and manage it in the form of logs. Even if the data on the log area is frequently modified to generate hard disk fragments, it is convenient to recover and manage these fragments, and save the non-hot data in the data area , the release of non-hot data is not easy to cause fragmentation of the hard disk, and the data area does not need to allocate too many resources for the management of hard disk fragmentation. Therefore, by storing different types of data in different areas on the hard disk and managing them in different ways, it can improve The efficiency of fragment management on the hard disk and the efficient management of hard disk fragments in the log area can reduce the generation of hard disk fragments.

Description of drawings

Fig. 1 is a logical view of an object on the log area provided by the embodiment of the present invention;

Fig. 2 is a flow chart of a hard disk data management method shown in an embodiment of the present invention;

FIG. 3 is a schematic diagram of data migration in memory involved in the embodiment shown in FIG. 2;

FIG. 4 is a schematic diagram of data cached in memory involved in the embodiment shown in FIG. 2;

5 is a schematic structural diagram of a hard disk control device provided by another embodiment of the present invention;

FIG. 6 is a schematic structural view of the recovery unit of the hard disk control device shown in FIG. 5;

FIG. 7 is a schematic diagram of a hardware structure of a hard disk control device provided by another embodiment of the present invention.

Detailed ways

Embodiments of the present invention provide a hard disk data management method and a hard disk control device for efficient management of fragments on the hard disk.

1. The implementation environment involved in the hard disk data management method of the embodiment of the present invention

A kind of hard disk data management system of the embodiment of the present invention, this hard disk data management system comprises hard disk, internal memory, and this internal memory can be used as cache device, and this hard disk is divided into data area Date zone and log area Journal zone, wherein this log area is with log way to manage the data on it.

Before the hard disk data management system writes data to the hard disk, write the data to the memory that is used as a cache device first. If the hard disk data management system judges that the data is hot data, the hot data is stored on the hard disk and then preset After the number of modifications and releases, the hard disk will generate a preset number of fragmented data. After these hot data are modified in the data area of the hard disk, the hard disk will generate a large number of fragments. Therefore, after data is written into the memory, if the data is not hot data, a data area space is allocated for the data in the data area, and then the data is written into the data area space. If the data is hot data, allocate log area space for the data in the log area, and write the data into the log area space.

Wherein, the data area space is the storage space on the data area, which may be part of the space on one data area, or all the space on one data area. The log area space is the storage space on the data area, which can be part of a log area or the entire space of a log area.

In this way, the data to be written into the hard disk is divided into hot data and non-hot data. Hot data is data that can cause the hard disk to generate a preset number of fragments after being stored on the hard disk after a preset number of modifications and releases. Hot data is easy to cause The hard disk generates fragments, save the hotspot data in the log area, and manage it in the form of logs. Even if the data on the log area is frequently modified to generate hard disk fragments, it is convenient to recover and manage these fragments, and save the non-hot data in the data area , the release of non-hot data is not easy to cause fragmentation of the hard disk, and the data area does not need to allocate too many resources for the management of hard disk fragmentation. Therefore, by storing different types of data in different areas on the hard disk and managing them in different ways, it can improve The fragmentation management efficiency on the hard disk can effectively manage the fragments on the hard disk, and the effect of avoiding hard disk fragmentation can also be achieved through the management of the log area.

The log area and the data area of the hard disk can be set in many ways, which will be described in detail below as one of the implementation ways.

The hard disk is divided into two types of areas, the data area and the log area, and the space size of the data area and the log area is not specifically limited in this embodiment of the present invention, for example, it may be 256M. The data area and the log area can be set alternately, as shown in Table 1, which is an example of a hard disk space layout, and the hard disk is divided into a super block, a data area and a log area. Optionally, 0.1% of the log area set is used as a fixed fixed log area, which is distributed evenly on the hard disk. This type of log area can only be used as a log area, while other log areas are in the hard disk space. When it is insufficient, it can be converted into a data area.

Table I

The data area and the log area have multiple setting modes on the hard disk, and the above-mentioned data area and the log area are arranged alternately, which is only one of them. The present invention does not specifically limit this. area, the log area is continuously set in another area of the hard disk, or multiple data areas are continuously set as data area groups, multiple log areas are continuously set as log area groups, and then data area groups and log area groups are set alternately, and so on.

The data area can store non-hot data, such as writing data larger than 128KB directly into the data area. Wherein, in the data area, after data is written into the data area, metadata for managing the data will be generated. After the metadata can be written into the cache device, space is allocated in the log area for storage.

The log area can save hot data, for example, save data and metadata less than 128KB in the log area. In some embodiments, the hot data can be stored in the log area in an additional manner. In some embodiments, It is also possible to sequentially assign an identification information ID to the log area. In the log area, data is managed in log mode.

In some embodiments, in the log area, io processing is performed in a sequentially appended write mode. When a data block needs to be written into the log area, space is allocated from the end of the last write in the log area. When the log area cannot accommodate a data block, re-select a log area with the largest free space for additional writing.

The layout of the log area is shown in Table 2, where Journal ctrl is the identification information ID, Map is the mapping relationship, and Record is data, such as data and metadata less than 128KB.

Table II

As shown in Table 1, in some embodiments, the hard disk of the hard disk control device is also provided with a super block. After the data is written into the log area, the log area is modified, and the super block will record the sign of the modified log area. Information ID. For example, when a batch of write operations to the log area is combined into a transaction, after the data of a transaction is saved to the hard disk, the identification information ID of the log area modified by this transaction can be recorded in the corresponding bitmap of the super block.

For the space size of the super block, the embodiment of the present invention does not specifically limit it, and it can be adjusted according to the specific conditions of the device. For example, a 4T disk has 4T/256M/2=8192 log areas, and a 1024B is required in the super block to record The overall usage of the log area.

The layout of the super block can be shown in Table 3. Among them, Super Blkctrk is used to record the overall management information, such as the used capacity of the hard disk, the total capacity, the total free capacity, the number of log areas, and the number of data areas. Journalbitmap is used to record the transaction numbers that have been processed. The capacity of Super blkctrk and Journalbitmap can be 4K respectively

Table three

Super blkctrk Journal bitmap

In order to make the data allocated to the data area and the log area close together, in some embodiments, multiple log areas and data areas may be combined into a block, and the log areas and data areas on the block are set continuously. Each block group has a space management object, and the use of hard disk space in this block group is managed in the form of a bitmap file bitmap. For example, 16 log areas and data areas set consecutively can be combined into one block group.

Table 4 and Table 5 show the relationship among the chunk, data area and log area. Table 4 shows the layout of the hard disk in units of blocks, and Table 5 shows the layout of block 1 in Table 4.

Table four

Table five

As shown in Table 2, the mapping relationship Map is also stored on the log area, and the mapping relationship can be used to record the corresponding relationship between the data on the log area and the log area space to which the data is allocated, as shown in Figure 1. A logical view of the hard disk object is shown, and the mapping relationship is illustrated according to the figure.

As shown in Figure 1, it shows an object on the log area. An object can be divided into multiple levels, and the lowest level level is 0, which corresponds to the data block of the object. Above level0 is an indirect block with level 1. The top layer is the block where the object management structure is located, and the level is 2. In the case of many data blocks, one indirect block cannot store the address pointers of so many data blocks. At this time, multiple indirect blocks are needed, and the number of layers of objects also increases. Blocks of the same layer are numbered from left to right, for example, the blkid of the data block at the bottom layer is numbered 0, 1, 2, and 3 in sequence.

When modifying a data block in a transaction, the relationship between the data and the log area space allocated to the data needs to be recorded in a mapping relationship. For example, if a transaction creates the object in Figure 1, then the information in Table 6 needs to be recorded in the mapping relationship. In Table 6, the information types of data recorded in each column in the mapping relationship are objsetid, objid, levelid, blkid, journalid, offset, and size in sequence.

Among them, objsetid refers to the object set ID, objid refers to the object ID, levelid refers to the number of layers where the data block is located, blkid refers to the number of layers where the data block is located, and the serial number from left to right, journalid refers to the id of the log area where the data block is written, offset refers to the relative offset of the data block written to the log area, and size refers to the size of the data block written.

It can be understood that the information types of the mapping relation records may include all the above information types, may also include some of the above information types, and may also include more other information types, which are not specifically limited in this embodiment of the present invention.

Table six

It can be understood that, in some embodiments, the cache device can be replaced by other devices, such as Nvdimm, Flash card, SSD (Solid State Drives, Solid State Drives) and so on. It can be understood that, in some embodiments, the hard disk may not include a log area, but the hotspot data may be saved on the cache device, which is not specifically limited in this embodiment of the present invention.

It can be understood that the hard disk control apparatus in the embodiment of the present invention can be used in equipment such as computers and servers, which is not specifically limited in the embodiment of the present invention.

Fig. 2 is a flowchart of a hard disk data management method according to an exemplary embodiment. The method is applied to a hard disk control device, and the hard disk control device includes a hard disk, and the hard disk includes a data area and a log area. Combining the first part of the above description, that is, the implementation environment involved in the hard disk data management method of the embodiment of the present invention, taking the perspective of the hard disk control device executing the method provided by the embodiment of the present invention as an example, see FIG. 2 , the embodiment of the present invention provides The method flow includes:

Step 201: write data to memory;

Before the device writes data to the hard disk, the hard disk control device first writes the data into the cache device for management before writing to the hard disk, for example, using the cache manager of the hard disk control device to manage the data.

The cache device may be a memory, or a cache device such as a flash memory, which is not specifically limited in this embodiment of the present invention.

In the embodiment of the present invention, the cache device is used as a memory for description. The data written into the internal memory includes all external data of the hard disk control device and metadata generated when data is written to the data area of the hard disk.

Among them, writing data to the memory includes modifying and creating writing. Modifying the data written to the memory means modifying the data on the memory. New data overwrites the modified data, and creating a write means writing to the memory. Entering new data, the original data of the new data is not cached in the memory.

Step 202: Determine whether the data is hot data, if the data is not hot data, execute step 203, if the data is hot data, execute step 204.

After writing data into the internal memory, the hard disk control device judges whether the data is hot data, such as through the cache manager module of the hard disk control device. After the hard disk control device judges whether the data to be written into the hard disk is hot data, it executes different processing methods according to the judgment result.

Wherein, the hot data is the data stored on the hard disk that can cause the hard disk to generate a preset number of fragments after a preset number of modifications and releases. For example, in this embodiment, the hot data may refer to data whose data size is smaller than a preset data threshold, and/or the hot data may also refer to metadata.

The data whose data size is smaller than the preset data threshold is small data, and the frequent modification and release of this small data will easily cause hard disk fragmentation, while the data whose data size is larger than a certain preset data threshold will not generate a large amount of data even if it is frequently modified on the hard disk. Hard disk fragmentation. Wherein, the setting of the preset data threshold is related to the business model, for example, it can be set to 64KB, 128KB, etc.

The metadata is a data block that records the object management structure and records the address of the data block. When modifying the data block on the hard disk data area, because the data area generally uses the COW mechanism, that is, when modifying and writing a piece of data, the old version of the data is not directly overwritten, but the old version of the data is read. After modification, write to a new location and release the old version of the data. Because the location of the data has changed, it is necessary to modify the pointer to the index block of the upper layer of the data, that is, to modify the metadata, and the modified new metadata is allocated to the new space, while the modified old metadata needs to be Released, so recursively to the top. In this way, a large amount of data will be released due to the modification of metadata, and fragmentation will occur at the position of the released data on the hard disk, thereby accelerating the process of hard disk fragmentation.

Therefore, in the embodiment of the present invention, the data and/or metadata whose data size is smaller than the preset data threshold are classified as hot data. These data are likely to cause fragmentation of the hard disk and need to be managed accordingly.

Step 203: Allocate data area space for the data in the data area, and write the data into the data area space.

Data that is judged not to be hot data can be stored in the data area on the hard disk because it is not likely to cause fragmentation on the hard disk. Wherein, the data area is an area for storing data on the hard disk, and the COW mechanism can be used in this area to manage the data thereon. The data area may be an area on the hard disk with a preset space size, for example, the preset space size may be 256M. For the description of the data area, please refer to the description of the data area in the above-mentioned implementation environment part.

For example, when the size of a piece of data is greater than 128KB, the cache manager judges that the data is not hot data, then the data zone manager module of the hard disk control device can allocate data zone space on the data zone for the data, and then write the data into The allocated data area space.

Wherein, after data is written into the data area, metadata will be generated. The metadata is used to record the address of the data, so as to facilitate the management of the data. For example, when organizing multiple blocks of data, a multi-level index is generally used, that is, an index block is allocated on the upper layer of the data block, and its content is to record the address of the data block. Through this index block, multiple data blocks can be spliced. as a logically continuous object. This index block is a type of metadata. After data is written into the data area to generate metadata, the metadata can be written into the memory, and the above step 201 is executed. The cache manager can determine that the metadata is hot data and save it in the log area and memory.

It can be understood that, in some embodiments, the hard disk of the hard disk control device also includes block groups, and the block groups include a preset number of log areas and data areas, and the log areas and data areas of the block groups are set continuously. In the control device, the specific implementation method of allocating data area space for data in the data area is to select an appropriate block group, such as a block group with less space usage or a block group with more free data areas, and then allocate The data area space of the data area is allocated on the block group. After allocating space, it is necessary to record the usage of the space, that is, after finding and allocating free data blocks in the space management bitmap of the block group, modify the management structure of the block group. Later, the metadata can be written to the log area belonging to the block group according to the space usage recorded by the block group. If the block group has no free log area, you need to select an adjacent log area to write the metadata. In order to make the location of the metadata and the data pointed to by the metadata close to each other on the hard disk.

Wherein, for the block group, reference may be made to the corresponding description of the block group in the above-mentioned implementation environment section.

It can be understood that, in addition to the data that is judged as non-hot data, the data written to the data area may also include the data eliminated from the memory due to insufficient space. For example, in the second stage of subsequent memory reclamation, the data pointed to by the mapping relationship Migrate from the log area to the data area. The migrated data can also be written into the data area space after the data area space is allocated by the data space manager.

Step 204: Allocate log area space for data in the log area.

There is also a log area on the hard disk, and the data in the log area is managed in a log mode in the log area.

When the data in step 201 is judged as hot data, and allocate log area space on the log area for the data, for example, the data is passed to the log manager, and the log manager allocates the log area on the log area for the hot data space so that the hotspot data is stored in the log area.

In the embodiment of the present invention, in order to manage the data in the log area more conveniently, the data belonging to the hot data can be allocated space in the log area according to the order in which they are written into the log area. Of course, in other embodiments, the log area space may not be allocated sequentially for hotspot data, which is not specifically limited in this embodiment of the present invention.

The data belonging to the hot data is stored in the log area. When the hot data on the memory is lost due to power failure, the data on the log area can be read to the memory for the device to perform operations, and the memory and the log area Used in conjunction with saving the hotspot data in the log area, the data volume of the hotspot data that can be managed by the memory can be expanded.

The log area may be an area with a certain size on the hard disk, for example, it may be 256M. There may be multiple log areas and data areas on the hard disk, and identification information may be assigned to the log areas according to the sequence set on the hard disk. For the specific setting method of the log area, refer to the corresponding description of the log area in the above-mentioned implementation environment section.

There are many ways to set the log area and the data area on the hard disk, for example, the log area and the data area are alternately set on the hard disk, as shown in Table 1 above. Of course, the data area and the log area can also be set in other ways, which is not specifically limited in the embodiment of the present invention. For details, refer to the corresponding description of setting the data area and the log area in the above-mentioned implementation environment section.

Step 205: Establish a mapping relationship between multiple target data and log area spaces allocated to the multiple target data.

Among them, the target data belongs to hot data.

After writing multiple data into the memory, after judging whether it is hot data, you may get multiple data belonging to the hot data in the memory. The hard disk control device determines a plurality of target data belonging to the hot data on the internal memory, so as to manage multiple target data together and improve processing efficiency. The log area space is allocated to the target data in the log area, and the hard disk control device establishes a mapping relationship between the target data and the log area space allocated to the target data. These target data belong to hot data, that is, the target data includes metadata and/or data whose data size is smaller than a preset data threshold.

All data written to the log area needs to record the mapping relationship between the data and the log area space, and record the data on the log area through the mapping relationship, so as to manage the hot data on the memory or the data on the log area according to the mapping relationship For example, the device reads the corresponding data stored in the log area according to the index of the mapping relationship cached in the memory, or reclaims the log area according to the data information recorded in the mapping relationship, so as to perform fragmentation management on the log area.

The information type of the mapping relation record can include information such as objsetid, objid, levelid, blkid, journalid, offset, size, etc.

For more information about the mapping relationship, refer to the corresponding description in the above-mentioned implementation environment part.

Step 206: Cache the mapping relationship in memory.

After the mapping relationship is established, it is saved in memory to prepare for subsequent operations.

In some embodiments, the mapping relationship can also be stored in the log area, and when it needs to be stored in the memory, the mapping relationship can be read from the log area to be cached in the memory. Of course, in some embodiments, the mapping relationship can be stored in the memory and the log area at the same time.

Step 207: Combine multiple write operations of the multiple target data into one transaction.

After determining a plurality of target data, it is necessary to perform write operations on the multiple target data to write into the log area. Hard disk operation. A transaction is a name for a combination of multiple write operations, not an execution of a write operation. A plurality of target data refers to at least two target data, and correspondingly, a plurality of write operations refers to at least two write operations.

In order to improve writing efficiency and ensure writing reliability, when the device writes data to the log area, it generally does not perform only one write operation, but performs multiple writes of multiple data in one operation of writing data to the log area. operate. These multiple target data belonging to the same batch of write operations are either all written successfully or all written to the hard disk. Multiple write operations of multiple target data of the same batch are combined into one transaction.

It can be understood that the execution sequence of step 207 and step 205 is not specifically limited in this embodiment of the present invention. That is, write operations of multiple target data are combined into one transaction, and a mapping relationship can be established according to these target data, so that one transaction corresponds to one mapping relationship.

In some embodiments, the method of the embodiment of the present invention further includes establishing a data link list based on the multiple target data, that is, forming a data link list based on a series of metadata in a transaction and data less than a preset threshold, the The data linked list is used to manage the target data. When forming the mapping relationship, the mapping relationship can be established according to the data on the data link list and the space allocated for these data in the log area.

Step 208: Write all target data of the transaction into the log area space.

The hard disk control device writes all target data of the transaction into the log area space. After multiple target data are combined into one transaction, when the write operation of one target data of the transaction fails, the write operation of other target data of the transaction fails. Only if every write operation of the target data is executed successfully can the write operation of the transaction succeed.

After the mapping relationship is established, log area space is allocated for the mapping relationship through the log area manager, and data belonging to hot data is also allocated log area space, so that both the mapping relationship and the target data can be written into the log to which they are allocated. area space to save the mapping relationship and target data in the log area.

After the mapping relationship is saved in the log area, the system can read the mapping relationship to the memory, so that the memory can reacquire the mapping relationship, which is especially useful for re-working after the memory is powered off. Of course, in some embodiments , the mapping relationship may not be stored in the log area, so there is no need to allocate space for the mapping relationship in the log area, which can also achieve the effect of reducing hard disk fragmentation in the data area, which is not specifically limited in the embodiment of the present invention.

In the embodiment of the present invention, there is no specific limitation on the order in which the mapping relationship and the target data of the transaction are written into the log area space.

In some embodiments, the log area space is allocated for the data in the log area, and the specific method of writing the data into the log area space is to allocate the log area space for the data in the log area in order, and write the data sequentially The log area space. Sequential allocation of space and sequential appending of data means allocating space or writing data in sequence in the storage space of the log area.

By allocating the log area space for the data in the log area in sequence, and adding the data to the log area space sequentially, the data in the log area can be read and written sequentially when managing the data in the log area, improving the efficiency of data management , and determine the data on the log area according to the order, and the mapping relationship records the corresponding relationship between the data stored in the log area and the log area space of these data, using the mapping relationship can replace the role of metadata, no need to use metadata to determine Data on the log area, so there is no additional metadata management overhead. In addition, when data is written in the log area in a sequential appending manner, the mapping relationship can be conveniently recorded for the storage space information of the data in the log area.

In an embodiment of the present invention, the method of the embodiment of the present invention further includes: if the data is hot data, caching the data in the memory, that is, writing all target data of the transaction into the log area space, and caching the target data of the transaction In the memory, that is, after judging whether the data is hot data, if the data is hot data, then keep the data in the memory, so that in the subsequent operation process, if the data is written to the memory, if the write operation is cached For modifying and writing the target data in the internal memory, the data can be directly modified in the internal memory to perform the above-mentioned steps of managing the target data according to the data link list. In this way, the data is retained in the memory so that the data can be migrated in the memory. Since the hot data is likely to cause fragmentation of the hard disk, caching these data in the memory instead of storing them in the data area can avoid the data being damaged in the data area. Data migration creates fragments.

In some embodiments, if the data is hot data, the step of caching the data in the memory may not be performed, but before step 201, the data is read from the log area to the memory for caching, that is, writing to the caching device Before entering data, read data from the log area to the cache device cache. In this way, the data can be modified directly in the memory, and the above-mentioned steps of managing the target data according to the data linked list can be completed. Therefore, the data is retained in the memory, so that the data is migrated in the memory, and fragmentation in the data area due to the migration of the data is avoided.

In some embodiments, the hard disk control device further includes a super block. When each log area is provided with identification information, the super block can be used to record the identification information of the modified log area after the log area is modified. Therefore, when the memory is powered off, the data on the corresponding log area can be read according to the identification information of the log area recorded by the super block, and then the read data on the log area can be cached in the memory to continue to execute subsequent data operate.

For example, after all target data and mapping relations of a transaction are written into the log area space, when all write operations of a transaction are completed, the identification information of the modified log area is recorded in the bitmap of the super block.

It can be understood that, in some embodiments, the hard disk further includes a block group, the block group includes a preset number of log areas and data areas, and the log areas and data areas of the block are set continuously. At this time, as described above, the data Allocating data area space in the data area may specifically include: determining a free target data area according to block group management information; and allocating data area space in the target data area for data.

Therefore, after data is written into the data zone space, target metadata is generated according to the data and the data zone space, and the target metadata is used to record the address of the data on the data zone, so as to facilitate management and query of the data.

After the metadata is generated, write the target metadata to the memory; after judging that the metadata is hot data, determine the target block group to which the target data area belongs; determine the available log area of the target block group, that is, find the available log area on the target block group log area, and then write the target metadata to the available log area. If not, find the nearest available log area near the target data area or target block group.

Wherein, when writing data to the log area is in units of transactions, the above-mentioned method can be used to allocate log area space for multiple target data of the transaction, and the metadata of the target data is allocated close to the data pointed to by the metadata.

In this way, after the metadata and the data pointed to by the metadata are stored on the hard disk according to the above method, the storage location of the metadata is close to the storage location of the data pointed to by the metadata, thereby reducing the moving distance of the magnetic head and improving the reading and writing of the hard disk. efficiency.

In some embodiments of the present invention, multiple target data belonging to the same transaction on the memory are managed in the form of a linked list. As mentioned above, after the method of the embodiment of the present invention determines a plurality of target data, a data linked list can be established according to the multiple target data, and the data linked list is used to manage the target data. That is, after the data and metadata of each transaction are submitted to the hard disk, the data in the corresponding log area is managed by a linked list method.

Because the writing operations of these target data are combined into a transaction, a transaction corresponds to a data linked list, and the target data managed by the data linked list is the same as the target data of the transaction. And a mapping relationship is established according to these target data, so that a data link table corresponds to a mapping relationship, and the mapping relationship records the storage situation of the data managed by the data link list in the log area.

The target data can be managed according to the data linked list. The specific management method is as follows:

According to the above-mentioned execution steps, after determining a plurality of first target data belonging to the hotspot data in the memory, a first data link list is established according to these first target data, and these first target data belong to the first transaction, that is, these first target data When data is written into the log area, it is written into the log area according to the same write batch. As long as one of the first target data fails to be written, the other data of the first transaction will fail to be written. These first target data also establish a first mapping relationship. The first mapping relationship is cached in memory. For example, the first data link list manages the first target data A1, the first target data B1, the first target data C1, and the first target data D1 in the internal memory. The corresponding first mapping relationship records the relationship between the first target data A1, the first target data B1, the first target data C1, the first target data D1 and the log area space to which these data are allocated in the log area. In the subsequent process, the hard disk control device establishes a second data link list according to a plurality of second target data, the plurality of second target data belongs to hot data, and belongs to the second transaction at the same time, and a second mapping is established according to the plurality of second target data For example, the data linked list can manage the second object data A2, the second object data E1, the second object data F1, and the second object data D2.

When the second target data managed by the second data linked list is modified by the first target data managed by the pre-established first data linked list, release the management of the first target data on the first data linked list; The first target data information is deleted from the first mapping relationship corresponding to the linked list. In this way, the migration of the target data on the data linked list and the mapping relationship is realized. For example, when the second target data A2 managed by the second data link list is obtained by modifying the first target data A1 of the first data link list, because the first target data A1 on the internal memory has been modified to the second target data A2, the second target data A2 A target data A1 is no longer needed, so the management of the first data link list on the first target data A1 can be released. Correspondingly, the information about the first target data A1 recorded in the first mapping can also be deleted. The information of the first target data A1 recorded due to the first mapping relationship is deleted. When the log area is recycled, according to the first mapping relationship, it can be judged that there is fragmentation in the log area corresponding to the first target data A1 , and when migrating and merging the data on the log area, because the information of the first target data A1 cannot be read from the first mapping relationship, it is not necessary to relocate the first target data A1 on the log area to a new log area, That is, the first target data A1 in the log area can be released.

FIG. 3 is a schematic diagram of data migration in memory, wherein each transaction includes a plurality of target data, and in FIG. 3 only some of the data blocks are marked with labels. As shown in Figure 3, due to the continuous generation of transactions, the subsequent transactions modify the data managed in the first transaction due to the locality of io. For example, the second transaction modifies the first target data A1 to obtain the second target data A2, the fifth transaction modifies the first target data B1 to obtain the fifth target data B2, and the fourth transaction modifies the first target data C1 to obtain the fourth target data C2, the second transaction modifies the first target data D1 to obtain the second target data D2, and the third transaction modifies the second target data D3 to obtain the third target data D3. In this way, according to the corresponding first transaction and the first data linked list, it can be known that all the data of the first transaction are migrated to the following transactions, and correspondingly, the log area can release all the target data of the first transaction. As shown in Table 7, the target data of the first transaction is stored in the log area 1. After all the data in the first transaction are migrated to the subsequent transactions, all the data blocks in the log area 1 are migrated to the subsequent transactions. In other log areas, that is, other log areas store the new version of data corresponding to the data in log area 1. At this time, log area 1 can be released as an empty log area when the log area is recycled. Migration of other target data is similar. When the fifth transaction is executed, the actual cache situation in the memory is shown in FIG. 4 .

Table 7

In order to more conveniently manage the target data according to the data linked list, in some embodiments, in the method of the embodiment of the present invention, it also includes assigning a transaction number to the data linked list corresponding to the transaction according to the incremental rule according to the writing order of the transaction, and the transaction The write order refers to the order in which different transactions are written to the log area. The transaction number assigned to the data linked list corresponding to the transaction written first is smaller, and the transaction number of the data linked list corresponding to the transaction written to the log area is increased by one. unit, so that each data linked list can have a corresponding identification, and these identifications also have an increasing law. Therefore, according to the transaction number, the establishment sequence of the data linked list on the memory can be determined conveniently.

For example, first establish the data linked list of the first transaction, and write the first transaction into the log area first, thereby assigning transaction number 1 to the data linked list of the first transaction, and then build the data linked list of the second transaction according to a plurality of second target data , the data of the second transaction will be written into the log area after the first transaction, thereby assigning transaction number 2 to the second data linked list corresponding to the second transaction, similarly, assigning transaction number 3 to the third data linked list, and so on .

The above is part of the content of managing data according to the data link list. The data belonging to the hot data is written into the log area. After caching the data belonging to the hot data in the memory or reading the data from the log area, the read data is cached in the memory, so that the data can be paired in the memory according to the data link list. Manage the data, let the hot data migrate on the memory, reduce the migration of the data on the hard disk, and reduce the fragmentation of the data caused by the hard disk.

The data is cached in the memory, and the data can also be read directly from the memory when reading the data, avoiding reading the data in the log area.

And by managing the data in the memory according to the data link list, the mapping relationship can be established and changed, and the data and fragments in the log area can be sorted according to the mapping relationship, so as to avoid or reduce the fragmentation of the hard disk and improve fragmentation processing. s efficiency. In order to make full use of the memory space, the hot data cached in the memory can be released, and the memory space can be reclaimed, so that the memory can cache other more data. Therefore, the method of the embodiment of the present invention can reclaim the memory space under the preset release condition. For example, cache eviction is triggered when the recycling watermark of system memory is reached. An exemplary recycling method is shown below, which is divided into two stages.

The first stage

When the memory reaches the first preset water level, start from the data link list with the smallest current transaction number, and search for the unmanaged data in the data link list according to the sequence of transaction numbers from small to large. Then, for the found target data linked list, release the unmanaged data of the target data linked list in the memory, and keep the target mapping relationship corresponding to the target data linked list in the memory.

Because the data is managed through the data link list, the data managed by the data link list established earlier may be eliminated in the memory due to the modification of the data written later. These eliminated data have been released from the data link list . For this, reference may be made to the above description of the management method of the data link list. Reclaim the memory, that is, find the data that has not been eliminated in the data linked list, and release the data that has not been eliminated in the memory. The data linked list to which the data released during memory recovery belongs may be called the target data linked list. The mapping relationship corresponding to the target data linked list is still kept in the memory. The memory reclamation in the first stage does not stop until the memory space reaches the preset stop water level.

After the above steps, the data less than or equal to the preset data threshold and the metadata generated by writing the data are written to the log area and cached in the memory at the same time. For these hot data in the log area and memory, as above, manage them in the order of transactions, so that when the system runs for a period of time and the cache reaches the first preset water level, the background cache recovery thread will be triggered from small to large transaction numbers Memory space is reclaimed in order. For example, starting from the data linked list with the smallest current transaction number, the unmanaged data in the data linked list is searched according to the order of the transaction number from small to large, and the data in the memory is released from the first established data linked list.

The data that is not eliminated in the memory of the data link list established earlier is less active data, and the device is less likely to read and modify it, so that these data can be released from the memory, so that the data read of the device Efficiency is less affected. Since the target mapping relationship is stored in the memory, when the device wants to read data, if it cannot be read from the memory, it will query according to the mapping relationship retained in the memory. If the data to be queried is determined according to the target mapping relationship, then The corresponding data can be read from the log area according to the target mapping relationship. In this way, the memory can manage more hot data.

It can be understood that when the memory reaches the first preset water level, it is only one of the preset release conditions, and the search for unmanaged data in the data link list can also be triggered under other preset release conditions, such as when the set timer expires, etc. This embodiment of the present invention does not specifically limit it.

second stage

After the memory is recycled in the first stage, when the memory reaches the second preset water level, read the data pointed to by the target mapping relationship from the log area; write the data pointed to by the target mapping relationship into the data area; delete the target mapping relationship on the memory .

In order to further fully utilize the memory space, the memory space can be reclaimed for the second time in the second stage. Wherein, the second preset water level is the water level that triggers the memory to execute the second stage of recovery. The second preset water level may be the same as the first preset water level, or may be different from the first preset water level, which is not specifically limited in this embodiment of the present invention. The embodiment of the present invention does not specifically limit the value, for example, it can be flexibly set according to the actual memory capacity and service type.

After the data pointed to by the target mapping relationship is read from the log area, the data pointed to by the target mapping relationship can be written into the data area. When the memory reaches the second preset water level, the memory has been running for a certain period of time, and more and more mapping relationships are cached in the memory. When the cache water level is finally reached, it is necessary to read the data pointed to by these mapping relationships and write into the data area.

When the internal memory reaches the second preset water level, due to the implementation of the above-mentioned management method of the data linked list, the possibility of modifying the data of the data linked list established earlier is less likely to be modified, because if the management method on the data linked list If the data is modified by the data written later, the data is migrated from the data linked list to the corresponding data linked list of the subsequent transaction. Thereby, the remaining data of the data link list established earlier can be judged as inactive data. These inactive data are less likely to be modified, so they will cause fewer hard disk fragments due to modification, and the target can be mapped to The data pointed to by the relationship is written into the data area. For example, the data is read from the log area according to the target mapping relationship, and then the data area manager allocates space for the data area on the data area, and then writes the data into the allocated data area.

After the execution of the above steps, when the device wants to read the data on the hard disk, for example, the data that has been written in the log area is accessed, the hard disk control device can first search in the memory, and if it is hit in the memory, it can directly Return that data. After the first stage of memory reclamation, some hot data is deleted from the memory, and the corresponding mapping relationship is retained, so if the data to be accessed is not found in the memory, it can be found in the mapping relationship of the memory cache , then the data to be accessed can be directly read in the log area corresponding to the mapping relationship. If the data to be searched cannot be found in the mapping relationship, search and read the data in the data area.

It can be understood that, in the embodiment of writing data to the log area in the manner of sequentially appending writes, after the data linked lists of multiple consecutive transactions are eliminated, for example, after the above-mentioned second phase of memory recovery is performed, the deleted mapping The log area corresponding to the relationship is also completely released and can be used as an empty log area again. The specific method of releasing the log area and reclaiming the log area will be described below.

In some embodiments, after the above method is executed, the embodiment of the present invention further includes the operation of reclaiming the log area. Thereby reducing fragmentation on the log area and making full use of the space in the log area.

The recovery of the log area can be based on the above-mentioned method, and the recovery of the log area will be described in an embodiment in which data is written to the log area in a sequentially appended manner. That is, step 208 is to sequentially write the target data of the transaction into the log area space, that is, the data on the log area is stored according to the order of the transactions,

The specific recovery method of the log area, for example, may include the following steps:

A1: Under the preset recycling conditions, perform the steps of data relocation in the log area;

Wherein, the preset recovery condition includes at least one of timer timeout, completion of the recovery operation on memory data, and total water level of the log area reaching a preset water level threshold. Completion of the memory data recovery operation means that when the above-mentioned memory recovery is completed, the log area data relocation step is triggered after each stage is completed, that is, the log recovery thread is triggered to execute the log area recovery process.

A2: When the total space water level of the current log area reaches the preset space threshold, stop executing the steps of log area data relocation, otherwise continue to execute the steps of log area data relocation.

Among them, the steps to perform log area data relocation include:

B1: Starting from the data linked list with the smallest current transaction number, search for the mapping relationship corresponding to the data linked list according to the order of transaction numbers from small to large.

After the above-mentioned method executes the step of assigning transaction numbers to the data linked list corresponding to the transaction according to the increasing rule according to the writing order of the transactions, the hard disk control device searches for the mapping relationship corresponding to the data linked list in the order of transaction numbers from small to large, because the mapping The relationship records the relationship between the data and the space allocated to the data in the log area, so that the distribution of the data blocks in the mapping relationship in the log area can be analyzed to know how much data in the corresponding log area has not been migrated. The data that is not migrated is located in the corresponding log area.

And because when the transaction is advancing, it will switch to the next log area after using one log area, so the data of continuous transactions will be written to the continuous log area. By analyzing the mapping relationship according to the sequence of transaction numbers from small to large, the data storage situation on the continuous log area can be obtained.

B2: According to the information recorded in the mapping relationship, it is judged whether the data on the first log area corresponding to the mapping relationship has been migrated;

When managing the data on the memory according to the data link list, the corresponding mapping relationship should also be modified for the data migrated on the memory. The data corresponding to the log area has a new version in other log areas, or after the second stage of memory reclamation, the data on the corresponding log area has been moved to the data area. At this time, it is judged that the mapping relationship corresponds to The data in the log area has been migrated.

B3: If the data migration in the first log area is completed, reclaim the first log area;

If the data in the first log area has been migrated, the first log area is recovered. In the embodiment where the usage of the log area is recorded by the super block, the information corresponding to the first log area in the super block can be cleared at this time.

B4: If the data on the first log area has not been migrated, then determine the space utilization rate on the first log area according to the information recorded in the mapping relationship;

Since the mapping relationship records the relationship between the data and the space allocated to the data in the log area, the space utilization rate of the first log area can be analyzed according to the information recorded in the mapping relationship.

B5: When the space utilization rate of the first log area is less than the preset utilization threshold, migrate the data in the first log area to the second log area, and update the mapping relationship corresponding to the relocated data.

The second log area is an idle log area or a log area that was used when the log area was recovered, for example, a log area that was used when the log area was recovered last time. The preset utilization threshold can be set according to specific usage conditions. For example, when the log area is used more, but the utilization rate of each log area is low, the preset utilization threshold can be lowered. For example, the preset utilization threshold may be set to 50%, etc., which is not specifically limited in this embodiment of the present invention. Updating the mapping relationship corresponding to the relocated data may be updating the corresponding relationship between the data and the new log area space in time on the mapping relationship related to the data after the data is relocated in the log area.

Through the above log area recovery method, the recovery of the log area can be realized, the debris on the log area can be reduced, and the space of the log area can be fully utilized. Since the log area is read and written sequentially, it corresponds to the order of the transaction numbers of the data link list from small to large, so that the corresponding mapping relationship can be queried according to the transaction number of the data link list, and the data can be relocated. At this time, metadata is not required to record The location of the data in the log area, thereby reducing the overhead of metadata.

In the method of the embodiment of the present invention, on the hard disk control device including a hard disk, the hard disk includes a data area and a log area. After writing data to the cache device, the hard disk control device judges whether the data is hot data, if the data is not hot data , allocate data area space for the data in the data area, and write the data into the data area space; if the data is hot data, allocate log area space for the data in the log area, and write the data into the log area space. In this way, the data to be written into the hard disk is divided into hot data and non-hot data. Hot data is data that can cause the hard disk to generate a preset number of fragments after being stored on the hard disk after a preset number of modifications and releases. Hot data is easy to cause The hard disk generates fragments, save the hotspot data in the log area, and manage it in the form of logs. Even if the data on the log area is frequently modified to generate hard disk fragments, it is convenient to recover and manage these fragments, and save the non-hot data in the data area , the release of non-hot data is not easy to cause fragmentation of the hard disk, and the data area does not need to allocate too many resources for the management of hard disk fragmentation. Therefore, by storing different types of data in different areas on the hard disk and managing them in different ways, it can improve The efficiency of fragment management on the hard disk and the efficient management of hard disk fragments in the log area can reduce the generation of hard disk fragments.

The data in the log area is migrated due to the locality of hot data. In fact, when the log area is sorted, less data is read, so the data management efficiency of the log area is higher. However, storing hot data in the log area, because the hot data is prone to fragmentation, the fragments are concentrated in the log area, which further improves the efficiency of defragmentation.

When data is written to the log area in the form of sequential append writing, only sequential reading and writing of data is required in the log area, and there is no metadata overhead for relocating data, and the efficiency of defragmentation in the log area is high. By sorting the data in the log area, hard disk fragmentation can be effectively avoided or reduced.

In some embodiments, data may not be written to the journal area in a sequential appending manner. When the space allocation of the journal area is known, a bitmap file bitmap may be used to manage the journal zone.

For example, each bit corresponds to a fixed block size, such as 4K, then a 256M journal zone needs 8K to manage, and the bitmap needs to be modified every time the journal zone is written.

It can be understood that in the embodiment of the present invention, the data area and the log area can be hierarchical storage, that is, the data area and the log area are at different levels, and when the log area is recycled, the hot data is migrated to a lower-level storage layer .

It can be understood that in some embodiments, the mapping relationship can be searched in other ways, such as randomly querying the mapping relationship, and then analyzing the storage space of the corresponding log area according to the found mapping relationship, and then relocating the data in the log area. At this time, it is not necessary to write the target data of the transaction into the log area space in the manner of sequential appending, and the specific writing method is not limited. However, such a method may not reflect the full storage space of the log area due to the mapping relationship, resulting in an unsatisfactory recycling effect of the log area.

It can be understood that in the embodiments of the present invention that include multiple log areas and multiple data areas, in order to make full use of the data areas and log areas, thereby making full use of hard disk space, the method in the embodiment of the present invention also includes data areas and multiple data areas. The conversion step of the log area, for example, when the space utilization rate of the data area is greater than the preset data area utilization threshold, convert the currently free log area into a data area; when the space utilization rate of the log area is greater than the preset log area utilization threshold When the value is set, the data area converted from the free log area will be converted into the log area.

For example, when the hard disk is initialized, it is pre-set that half of the hard disk space is the log area, and the rest is the data area. After the system has been running for a period of time, the utilization rate of the data area is high, and the idle log area is searched for in the increasing order of the identification information of the log area, and converted into a data area. After conversion to a data area, in an embodiment including a block group, it is managed using the space management object of the block group. Set the state of the log area to the data area, record it in the management structure of the block group, and write it to disk for storage. When the data on the data area is deleted, the hard disk space is released, and the space of a data area converted from the log area is completely released. At this time, the state of the data area is switched to the log area and recorded in the management structure of the block group , write to disk to save.

Fig. 5 is a schematic structural diagram of a hard disk control device shown according to an exemplary embodiment, the hard disk control device includes a hard disk, the hard disk includes a data area and a log area, and the hard disk control device is used to implement the embodiment corresponding to Fig. The function performed by the hard disk control unit. Referring to Fig. 5, the hard disk control device includes:

A writing unit 501, configured to write data into the cache device;

The cache manager 502 is used to determine whether the data is hot data, wherein the hot data is data that can cause the hard disk to generate a preset number of fragments after being modified and released for a preset number of times after being stored on the hard disk;

The data manager 503 is used to allocate data area space in the data area for the data if the data is not hot data, and write the data into the data area space;

The log manager 504 is configured to allocate log area space for the data in the log area if the data is hot data, and write the data into the log area space.

Optionally, the cache device is a memory, and the hard disk control device also includes:

A mapping relationship establishment unit 505, configured to establish a mapping relationship between data and log area space in memory;

Optionally, the hard disk control device also includes:

The caching unit 506 is configured to cache data belonging to hot data in memory.

Optionally,

The mapping relationship establishment unit 505 is further configured to establish a mapping relationship between multiple target data and the log area space to which the multiple target data are allocated, wherein the target data belongs to hot data;

The log manager 504 is also used to combine multiple write operations of multiple target data into a transaction and write all target data of the transaction into the log area space. When the write operation of one of the target data of the transaction fails, the transaction's Write operations performed on other target data failed.

Optionally,

The hard disk control unit also includes:

A linked list establishment unit 509, configured to establish a data linked list according to a plurality of target data, wherein the data linked list is used to manage the target data, and the target data managed by the data linked list is the same as the target data of the transaction;

A linked list management unit 510, configured to manage the target data according to the data linked list;

The search unit 511 is used to search for the data that has not been released from the data link list according to the order in which the data link list was first established under the preset release condition;

The memory management unit 512 is used to release the unmanaged data of the target data linked list in the memory, and retain the target mapping relationship corresponding to the target data linked list in the memory;

Wherein, the linked list management unit 510 is also used for:

After the second data link list is established, when the second target data managed by the second data link list is modified by the first target data managed by the pre-established first data link list, the first target data is released on the first data link list. Management; delete the information of the first target data on the first mapping relationship corresponding to the first data linked list

Optionally,

The search unit 511 is also used to search for data that has not been released from the management of the data link list according to the order in which the data link list is established from first to last when the memory reaches the first preset water level;

The hard disk control unit also includes:

A reading unit 523, configured to read the data pointed to by the target mapping relationship from the log area when the memory reaches the second preset water level;

A mapping data writing unit 513, configured to write the data pointed to by the target mapping relationship into the data area;

The deleting unit 514 is configured to delete the target mapping relationship in memory.

Optionally,

The hard disk control unit also includes:

A transaction number allocation unit 515, configured to allocate a transaction number for the data linked list corresponding to the transaction according to the incremental rule according to the writing sequence of the transaction;

The search unit 511 is further configured to start from the data link list with the smallest current transaction number, and search for data that has not been released from the data link list according to the order of transaction numbers from small to large.

Optionally,

The hard disk control unit also includes:

A recovery unit 516, configured to perform the step of relocating data in the log area under preset recovery conditions;

As shown in FIG. 6, in the step of performing log area data relocation, the recovery unit 516 includes:

Recycling lookup module 517, used to look up the mapping relationship;

Recycling judging module 518, used to judge whether the data on the first log area corresponding to the mapping relationship has been migrated according to the information recorded in the mapping relationship;

The recovery determination module 519 is used to determine the space utilization rate on the first log area according to the information recorded in the mapping relationship if the data on the first log area has not been migrated;

A recovery execution module 520, configured to migrate the data in the first log area to the second log area when the space utilization rate of the first log area is less than the preset utilization threshold, and update the mapping relationship corresponding to the relocated data , wherein the second log area is a free log area or a log area used when recycling the log area;

The recycling module 521 is configured to stop executing the step of relocating data in the log area when the total space water level in the current log area reaches a preset space threshold, otherwise continue to execute the step of relocating data in the log area.

Optionally,

The hard disk control unit also includes:

A transaction number allocation unit 515, configured to allocate a transaction number for the data linked list corresponding to the transaction according to the incremental rule according to the writing sequence of the transaction;

The recovery search module 517 is also used to start from the data linked list with the smallest current transaction number, and search for the mapping relationship corresponding to the data linked list according to the order of transaction numbers from small to large;

Optionally, the preset recovery condition includes at least one of timer timeout, completion of recovery operation on memory data, and total water level of the log area reaching a preset water level threshold.

Optionally,

The cache unit 506 is also used to cache the data in the memory if the data is hot data; or read the data from the log area to the cache of the cache device.

Optionally, the hotspot data includes data whose data size is smaller than a preset data threshold and/or the hotspot data includes metadata.

Optionally,

The log manager 504 is further configured to sequentially allocate log area space for data in the log area, and sequentially write data into the log area space.

Optionally,

The hard disk control unit also includes:

The data area conversion unit 522 is used to convert the currently idle log area into a data area when the space utilization rate of the data area is greater than the preset data area utilization threshold;

The log area conversion unit 524 is configured to convert the data area converted from the idle log area into a log area when the space utilization rate of the log area is greater than a preset log area utilization threshold.

Optionally,

The hard disk also includes a super block, each log area is provided with identification information, and the super block is used to record the identification information of the modified log area after the log area is modified.

Optionally, the log area and the data area are set alternately on the hard disk.

Optionally,

The hard disk also includes a block group. The block group includes a preset number of log areas and data areas. The log area and data area of the block are set continuously.

Data manager 503, comprising:

A free area determining module 525, configured to determine a free target data area according to the management information of the block group;

An allocation module 508, configured to allocate data area space for data in the target data area;

The hard disk control unit also includes:

A metadata generation module 507, configured to generate target metadata according to the data and data zone space;

The writing unit 501 is also used to write target metadata to the cache device;

After the cache manager determines that the metadata is hot data, the log manager 504 is further configured to determine the target block group to which the target data area belongs; determine the available log area of the target block group; and write the target metadata into the available log area.

Optionally,

The log manager 504 is further configured to allocate log area space on the log area for the mapping relationship; and write the mapping relationship into the log area space allocated to the mapping relationship.

To sum up, on a hard disk control device including a hard disk, the hard disk includes a data area and a log area. After the writing unit 501 writes data to the cache device, the cache manager 502 judges whether the data is hot data. If the data If it is not hot data, then the data manager 503 allocates a data area space for the data in the data area, and writes the data into the data area space; if the data is hot data, then the log manager 504 allocates a log area for the data in the log area space, write the data into the log area space. In this way, the data to be written into the hard disk is divided into hot data and non-hot data. Hot data is data that can cause the hard disk to generate a preset number of fragments after being stored on the hard disk after a preset number of modifications and releases. Hot data is easy to cause The hard disk generates fragments, save the hotspot data in the log area, and manage it in the form of logs. Even if the data on the log area is frequently modified to generate hard disk fragments, it is convenient to recover and manage these fragments, and save the non-hot data in the data area , the release of non-hot data is not easy to cause fragmentation of the hard disk, and the data area does not need to allocate too many resources for the management of hard disk fragmentation. Therefore, by storing different types of data in different areas on the hard disk and managing them in different ways, it can improve The efficiency of fragment management on the hard disk and the efficient management of hard disk fragments in the log area can reduce the generation of hard disk fragments.

FIG. 7 is a schematic diagram of a hardware structure of a hard disk control device according to another embodiment of the present invention. The hard disk control device includes a processor CPU 701 , a cache device 703 , a hard disk 702 , a hard disk controller 705 and a bus 704 . The hard disk 702 includes a data area and a log area, and in some embodiments, the cache device may be, for example, a memory.

The steps performed by the hard disk control device in the above embodiments may be based on the structure of the hard disk control device shown in FIG. 7 .

The processor 701 executes a program, so that the hard disk control device executes the method of the above-mentioned hard disk data management method, examples of various optional designs are as follows.

The processor 701 executes the program, so that the hard disk control device has the following functions: write data to the cache device; judge whether the data is hot data, wherein the hot data is stored on the hard disk and can be modified and released after a preset number of times to make the hard disk Generate data with a preset number of fragments; if the data is not hot data, allocate data area space for the data in the data area, and write the data into the data area space; if the data is hot data, allocate log area space for the data in the log area, Write data to the log area space.

In an optional design, the cache device is a memory, and the processor 701 executes a program so that the hard disk control device has the following function: after allocating log area space for data in the log area, establish a mapping relationship between data and log area space.

An optional design, the processor 701 executes a program, so that the hard disk control device has the following functions: establish a mapping relationship between multiple target data and the log area space to which the multiple target data is allocated, wherein the target data belongs to hot data; Combine multiple write operations of one target data into one transaction; write all the target data of the transaction into the log area space, and when the write operation of one of the target data of the transaction fails, the write operations of the other target data of the transaction fail.

In an optional design, the processor 701 executes a program, so that the hard disk control device has the following function: cache data belonging to hot data in memory.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions: before writing all the target data of the transaction into the log area space, a data link list is established according to a plurality of target data, wherein the data link list is used to manage The target data, the target data managed by the data linked list is the same as the target data of the transaction; the target data is managed according to the data linked list; under the preset release condition, according to the establishment order of the data linked list from first to last, search for the unmanaged data linked list Data; release the data that is not managed by the target data linked list on the memory, and retain the target mapping relationship corresponding to the target data linked list in the memory; wherein, the target data is managed according to the data linked list, including: After the second data linked list is established, when When the second target data managed by the second data link list is modified by the first target data managed by the pre-established first data link list, the management of the first target data is released on the first data link list; Information about deleting the first target data on the corresponding first mapping relationship

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions: when the memory reaches the first preset water level, according to the order in which the data link list is established from first to last, search for data that has not been released from the data link list ;

After releasing the unmanaged data of the target data linked list on the memory, when the memory reaches the second preset water level, read the data pointed to by the target mapping relationship from the log area;

Write the data pointed to by the target mapping relationship into the data area;

Delete the target mapping relationship in memory.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

According to the writing order of the transaction, the transaction number is assigned to the data linked list corresponding to the transaction according to the increment rule;

Starting from the data linked list with the smallest current transaction number, search for the unmanaged data in the data linked list according to the order of transaction numbers from small to large.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions: under the preset recovery condition, execute the step of relocating the data in the log area;

Perform the steps of log area data migration, including:

Find the mapping relationship;

According to the information recorded in the mapping relationship, it is judged whether the data on the first log area corresponding to the mapping relationship has been migrated;

If the data on the first log area has not been migrated, then determine the space utilization rate on the first log area according to the information recorded in the mapping relationship;

When the space utilization rate of the first log area is less than the preset utilization threshold, the data in the first log area is migrated to the second log area, and the mapping relationship corresponding to the relocated data is updated, where the second log area is A free log area or a log area that has been used when recycling a log area;

When the total space water level of the current log area reaches the preset space threshold, the step of relocating the data in the log area is stopped; otherwise, the step of relocating the data in the log area is continued.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

According to the writing order of the transaction, assign the transaction number to the data linked list corresponding to the transaction according to the increment rule; start from the data linked list with the smallest current transaction number, and search for the mapping relationship corresponding to the data linked list according to the order of the transaction number from small to large;

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

The preset recovery conditions include at least one of timer timeout, completion of the recovery operation on memory data, and total water level of the log area reaching a preset water level threshold.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

After judging whether the data is hot data, if the data is hot data, cache the data in memory; or,

Before writing data to the cache device, read data from the log area to the cache device cache.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

The hot data includes data whose data size is smaller than a preset data threshold and/or the hot data includes metadata.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

Allocate the log area space for the data in the log area in sequence, and write the data sequentially into the log area space.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

When the space utilization rate of the data area is greater than the preset data area utilization threshold, the currently idle log area is converted into a data area;

When the space utilization rate of the log area is greater than the preset log area utilization threshold, the data area converted from the idle log area is converted into the log area.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

The hard disk also includes a super block, each log area is provided with identification information, and the super block is used to record the identification information of the modified log area after the log area is modified.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions:

The log area and data area are set alternately on the hard disk.

An optional design, the processor 701 executes the program, so that the hard disk control device has the following functions: the hard disk also includes a block group, the block group includes a preset number of log areas and data areas, and the log area and data area of the block are set continuously , determine the free target data area according to the management information of the block group; allocate data area space for the data in the target data area;

After writing the data into the data area space, generate the target metadata according to the data and the data area space; write the target metadata to the cache device; after judging that the metadata is hot data, determine the target block group to which the target data area belongs; determine the target Available log area for block group; write target metadata to available log area.

In an optional design, the processor 701 executes a program so that the hard disk control device has the following functions: allocate log area space for the mapping relationship in the log area; write the mapping relationship into the log area space allocated to the mapping relationship.

To sum up, on a hard disk control device including a hard disk, the hard disk includes a data area and a log area. After the processor 701 writes data to the cache device, the processor 701 judges whether the data is hot data. If the data If it is not hot data, then the processor 701 allocates data area space for the data in the data area, and writes the data into the data area space; if the data is hot data, then the processor 701 allocates a log area for the data in the log area space, write the data into the log area space. In this way, the data to be written into the hard disk is divided into hot data and non-hot data. Hot data is data that can cause the hard disk to generate a preset number of fragments after being stored on the hard disk after a preset number of modifications and releases. Hot data is easy to cause The hard disk generates fragments, save the hotspot data in the log area, and manage it in the form of logs. Even if the data on the log area is frequently modified to generate hard disk fragments, it is convenient to recover and manage these fragments, and save the non-hot data in the data area , the release of non-hot data is not easy to cause fragmentation of the hard disk, and the data area does not need to allocate too many resources for the management of hard disk fragmentation. Therefore, by storing different types of data in different areas on the hard disk and managing them in different ways, it can improve The efficiency of fragment management on the hard disk and the efficient management of hard disk fragments in the log area can reduce the generation of hard disk fragments.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-OnlyMemory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes.

As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still understand the foregoing The technical solutions recorded in each embodiment are modified, or some of the technical features are replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (24)

1. A hard disk data management method, characterized in that, the method is applied to a hard disk control device comprising a hard disk, and the hard disk comprises a data area and a log area, and the method comprises: Write data to the cache device; Judging whether the data is hot data, wherein the hot data is data that can cause the hard disk to generate a preset number of fragments after a preset number of modifications and releases after being stored on the hard disk; If the data is not hot data, then allocate a data area space in the data area for the data, and write the data into the data area space; If the data is hot data, allocate a log area space for the data in the log area, and write the data into the log area space; The method also includes: When the space utilization rate of the data area is greater than the preset data area utilization threshold, the currently idle log area is converted into a data area; When the space utilization rate of the log area is greater than the preset log area utilization threshold, the data area converted from the idle log area is converted into the log area. 2. The method according to claim 1, wherein the cache device is a memory, After the log area space is allocated for the data in the log area, the method further includes: A mapping relationship between the data and the log area space is established. 3. The method of claim 2, wherein, The establishment of the mapping relationship between the data and the log area space includes: Establishing a mapping relationship between multiple target data and the log area space to which the multiple target data are allocated, wherein the target data belongs to hot data; The writing the data into the log area space includes: Combining multiple write operations of the multiple target data into one transaction; All target data of the transaction is written into the log area space, and when the write operation of one of the target data of the transaction fails, the write operation of other target data of the transaction fails. 4. method according to claim 3, is characterized in that, described method also comprises: The data belonging to the hotspot data is cached on the memory. 5. The method of claim 4, wherein, Before writing all target data of the transaction into the log area space, the method also includes: Establishing a data link list according to the plurality of target data, wherein the data link list is used to manage the target data, and the target data managed by the data link list is the same as the target data of the transaction; managing the target data according to the data linked list; Under the preset release condition, according to the establishment sequence of the data link list from first to last, search for the data that has not been released from the management of the data link list; releasing the unmanaged data of the target data linked list in the internal memory, and retaining the target mapping relationship corresponding to the target data linked list in the internal memory; Wherein, the managing the target data according to the data link list includes: After the second data link list is established, when the second target data managed by the second data link list is obtained by modifying the first target data managed by the pre-established first data link list, release all data from the first data link list. management of the first target data; deleting the information of the first target data on the first mapping relationship corresponding to the first data linked list. 6. The method of claim 5, wherein, Under the preset release condition, according to the establishment order of the data link list from first to last, searching for the data that has not been released from the management of the data link list includes: When the internal memory reaches the first preset water level, according to the order in which the data linked list is established from first to last, search for data that has not been released from management in the data linked list; After releasing the unmanaged data of the target data linked list on the memory, the method further includes: When the internal memory reaches a second preset water level, read the data pointed to by the target mapping relationship from the log area; writing the data pointed to by the target mapping relationship into the data area; Delete the target mapping relationship on the memory. 7. The method of claim 5, wherein, The method also includes: Under the preset recycling conditions, execute the steps of log area data relocation; The steps of performing log area data relocation include: Find the mapping relationship; According to the information recorded in the mapping relationship, it is judged whether the data on the first log area corresponding to the mapping relationship has been migrated; If the data on the first log area has not been migrated, then determine the space utilization rate on the first log area according to the information recorded in the mapping relationship; When the space utilization rate of the first log area is less than the preset utilization threshold, migrate the data in the first log area to the second log area, and update the mapping relationship corresponding to the relocated data, Wherein the second log area is an idle log area or a log area used when reclaiming the log area; When the total space water level of the current log area reaches the preset space threshold, the step of relocating the data in the log area is stopped; otherwise, the step of relocating the data in the log area is continued. 8. The method of claim 7, wherein, The method also includes: Assigning a transaction number to the data linked list corresponding to the transaction according to the order of writing of the transaction according to an increment rule; In the step of performing log area data relocation, the searching for the mapping relationship includes: Starting from the data link list with the smallest current transaction number, the mapping relationship corresponding to the data link list is searched according to the order of the transaction numbers from small to large. 9. The method of claim 4, wherein, The caching of the data belonging to the hotspot data on the memory includes: After determining whether the data is hot data, if the data is hot data, then retain the data in the memory; or, Before the data is written into the cache device, data is read from the log area and cached by the cache device. 10. The method according to any one of claims 1 to 9, wherein the hotspot data includes data whose data size is smaller than a preset data threshold and/or the hotspot data includes metadata. 11. The method according to any one of claims 1 to 9, characterized in that, The allocating log area space for the data in the log area, and writing the data into the log area space includes: A log area space is allocated in sequence for the data in the log area, and the data is sequentially appended to the log area space. 12. The method according to any one of claims 2 to 9, wherein The method also includes: Allocating log area space on the log area for the mapping relationship; Writing the mapping relationship into the log area space to which the mapping relationship is allocated. 13. A hard disk control device, characterized in that the hard disk control device comprises: A write unit, used to write data to the buffer; A cache manager for judging whether the data is hot data, where the hot data is data that can cause the hard disk to generate a preset number of fragments after a preset number of modifications and releases after being stored on the hard disk ; The data manager is used to allocate data area space for the data in the data area of the hard disk if the data is not hot data, and write the data into the data area space; The log manager is used to allocate log area space for the data in the log area of the hard disk if the data is hot data, and write the data into the log area space; The hard disk control device also includes: A data area conversion unit, configured to convert the currently idle log area into a data area when the space utilization rate of the data area is greater than the preset data area utilization threshold; The log area conversion unit is configured to convert the data area converted from the idle log area into the log area when the space utilization rate of the log area is greater than the preset log area utilization threshold. 14. The hard disk control device according to claim 13, wherein the cache device is a memory, and the hard disk control device further comprises: A mapping relationship establishing unit, configured to establish a mapping relationship between the data and the log area space. 15. The hard disk control device according to claim 14, characterized in that, The mapping relationship establishment unit is further configured to establish a mapping relationship between multiple target data and the log area space to which the multiple target data are allocated, wherein the target data belongs to hot data; The log manager is also used to combine multiple write operations of the multiple target data into one transaction; write all target data of the transaction into the log area space, when one of the target data of the transaction When the execution of the write operation fails, the write operations performed by other target data of the transaction fail. 16. The hard disk control device according to claim 15, wherein the hard disk control device further comprises: A cache unit, configured to cache data belonging to the hotspot data on the memory. 17. The hard disk control device according to claim 16, characterized in that, The hard disk control device also includes: A linked list establishment unit, configured to establish a data linked list according to the plurality of target data, wherein the data linked list is used to manage the target data, and the target data managed by the data linked list is the same as the target data of the transaction; a linked list management unit, configured to manage the target data according to the data linked list; A search unit, configured to search for data that has not been released from management in the data link list according to the order in which the data link list was first established under a preset release condition; A memory management unit, configured to release the unmanaged data of the target data linked list in the memory, and retain the target mapping relationship corresponding to the target data linked list in the memory; Wherein, the linked list management unit is also used for: After the second data link list is established, when the second target data managed by the second data link list is obtained by modifying the first target data managed by the pre-established first data link list, release all data from the first data link list. management of the first target data; deleting the information of the first target data on the first mapping relationship corresponding to the first data linked list. 18. The hard disk control device according to claim 17, wherein: The search unit is further configured to search for data that has not been released from management in the data link list according to the order in which the data link list is established from first to last when the memory reaches a first preset water level; The hard disk control device also includes: A reading unit, configured to read the data pointed to by the target mapping relationship from the log area when the memory reaches a second preset water level; a mapping data writing unit, configured to write the data pointed to by the target mapping relationship into the data area; A deletion unit, configured to delete the target mapping relationship on the memory. 19. The hard disk control device according to claim 17, wherein: The hard disk control device also includes: The recovery unit is used to perform the steps of data relocation in the log area under preset recovery conditions; In the step of performing log area data relocation, the recycling unit includes: Recycling a search module, configured to search for the mapping relationship; A recycling judging module, configured to judge whether the data on the first log area corresponding to the mapping relationship has been migrated according to information recorded in the mapping relationship; A recovery determination module, configured to determine the space utilization rate of the first log area according to the information recorded in the mapping relationship if the data on the first log area has not been migrated; A recovery execution module, configured to migrate the data in the first log area to the second log area when the space utilization rate of the first log area is less than a preset utilization threshold, and update the The mapping relationship corresponding to the data, wherein the second log area is an idle log area or a log area used when recycling the log area; The recovery module is used to stop executing the step of data relocation in the log area when the total space water level in the current log area reaches the preset space threshold, otherwise continue to execute the step of relocating the data in the log area. 20. The hard disk control device according to claim 19, wherein: The hard disk control device also includes: A transaction number allocation unit, configured to allocate a transaction number for the data linked list corresponding to the transaction according to an increment rule according to the writing sequence of the transaction; In the step of performing log area data relocation, the searching for the mapping relationship includes: Starting from the data link list with the smallest current transaction number, the mapping relationship corresponding to the data link list is searched according to the order of the transaction numbers from small to large. 21. The hard disk control device according to claim 16, characterized in that, The buffer unit is further configured to retain the data in the memory if the data is hot data; or read data from the log area to the buffer device for buffering. 22. The hard disk control device according to any one of claims 13 to 21, wherein the hot data includes data whose data size is smaller than a preset data threshold and/or the hot data includes metadata. 23. The hard disk control device according to any one of claims 13 to 21, characterized in that, The log manager is further configured to sequentially allocate log area space for the data in the log area, and sequentially write the data into the log area space. 24. The hard disk control device according to any one of claims 15 to 21, characterized in that, The log manager is further configured to allocate log area space for the mapping relationship in the log area; and write the mapping relationship into the log area space allocated to the mapping relationship.