WO2020192710A1

WO2020192710A1 - Method for processing garbage based on lsm database, solid state hard disk, and storage apparatus

Info

Publication number: WO2020192710A1
Application number: PCT/CN2020/081281
Authority: WO
Inventors: 刘绍全; 陈祥; 李卫军; 杨亚飞
Original assignee: 深圳大普微电子科技有限公司
Priority date: 2019-03-28
Filing date: 2020-03-26
Publication date: 2020-10-01
Also published as: CN110007860A

Abstract

Disclosed in the present application are a method for processing garbage based on an LSM database, a solid state hard disk, and a storage apparatus. The method comprises: in a solid state hard disk, creating a log corresponding to data to be written; writing the data to be written to the corresponding log; when a condition for executing a merge operation is met, executing a merge operation on the logs in the solid state hard disk; and deleting the merged logs and labelling the super block in which the deleted logs were located as an invalid data state. By means of the present method, the present application can flexibly perform garbage processing on the data of an LSM database stored in the solid state hard disk, reducing write amplification.

Description

Garbage processing method based on LSM database, solid state hard disk and storage device

Technical field

This application relates to the field of storage technology, and in particular to a method for garbage disposal based on an LSM database, a solid state drive, and a storage device.

Background technique

The LSM (Log Structured Merge Trees) database engine converts random write IO for operating solid-state drives into sequential write IO, which improves the write performance of the database and provides relatively good read performance.

The write operation of the LSM database engine specifically includes: when there is a write operation, the written data is first written into the buffer of the memory, the sequence of the written data is recorded in the memory through a specific data structure, and the written data is additionally written To the logfile of the solid-state hard drive for recovery if necessary. The written data in the memory is flushed to the solid-state hard disk regularly or in a fixed size to store multiple ordered sstfile files on the solid-state hard disk.

Among them, with more and more write operations, there are more and more sstfile files accumulated on the solid state drive. These files are not writable and in order. The LSM database will periodically perform compaction on the sstfile files, and the merge is completed and deleted. Sstfile file, reduce the number of files.

As we all know, in the field of fixed hard disk technology, garbage disposal is an important part of the solid state hard disk firmware design, and it is also the main factor affecting the stable performance of the solid state hard disk. Among them, the particle characteristics of the flash memory (NAND) in the solid state drive are: multiple physical blocks (Block) form the flash memory, and the physical blocks must be erased before data can be written. If there are some valid data pages (physical pages storing user data) in a physical block before erasing, in order not to lose user data, the data in the valid physical pages must be read out and written to another physical block before Erase the physical block, which is garbage disposal.

When a solid state drive is used as the storage medium of the LSM database, the data merging operation performed by the LSM database engine, that is, the garbage processing performed cannot be perceived by the solid state drive, which causes the LSM database engine to perform the merging operation, that is, garbage processing. The solid-state drive will also perform garbage processing on the same data again, which will cause repeated data movement and increase write amplification.

Summary of the invention

The present application provides a method for garbage processing based on an LSM database, a solid state drive, and a storage device, which can flexibly perform garbage processing on data in the LSM database and reduce write amplification.

In order to solve the above technical problems, a technical solution adopted by this application is to provide a garbage disposal method based on the LSM database. The method includes: The data is the sstfile file or logfile file in the LSM database; write the data to be written into the corresponding log; perform the merge operation on the logs in the solid state drive under the condition of performing the merge operation; delete the merged log and mark the The super block where the deleted log is located is in a data invalid state, where the super block is a physical address space allocated in the solid state disk for storing each log.

Wherein, the method further includes: obtaining the number of super blocks in the idle state in the solid state disk; obtaining the access strength of the solid state disk; detecting whether the number and access strength of the super blocks in the idle state meet a predetermined condition; wherein, when the detection result is When the number of super blocks in the idle state and the access intensity meet the predetermined conditions, the conditions for performing the merge operation are satisfied.

Among them, the steps of performing a merge operation on each log in the solid state drive include: obtaining the number of logs stored in each super block and the amount of data corresponding to each log; obtaining the corresponding data of each super block according to the number of logs and the amount of data corresponding to each log Effective data volume; merge the super block with the smallest effective data volume.

Wherein, the step of creating a log corresponding to the data to be written in the solid state drive includes: obtaining a group identifier corresponding to the data to be written, wherein each group identifier corresponds to a super block, and each super block includes a first predetermined There are multiple first logic blocks of different sizes, all super blocks correspond to an entry parameter list, and the first logic block in each super block corresponds to an entry parameter in the entry parameter list; dynamic in the super block corresponding to the group identifier The first logical block that is free for storing logs is allocated.

Wherein, the step of writing the data to be written into the corresponding log further includes: dividing the log after writing the data into a plurality of second logical blocks of a second predetermined size;

Find the entry parameter corresponding to the free first logic block in the super block; write the content in the second logic block into the corresponding free first logic block; modify the entry parameter corresponding to the first logic block that has completed the write operation It is the log ID of the log and the logical block ID of the second logical block that has been written.

Wherein, the method further includes: using the log identifier and the logical block identifier written in the entry parameters as the hash key value, obtaining the hash value according to the hash key value; obtaining the index value of the hash bucket according to the hash value; and detecting the index value Whether the corresponding hash bucket is empty; if the detection result is that the hash bucket is empty, the subscript of the entry parameter is saved in the hash bucket corresponding to the index value; if the detection result is that the hash bucket is not empty, then Save the subscript of the entry parameter in the corresponding hash link.

Wherein, before the step of creating a log corresponding to the data to be written in the solid state disk, the method further includes: obtaining a group identifier in the LSM database; creating corresponding super blocks for different group identifiers in the solid state disk, where the LSM database The sstfile files in the same layer belong to the same group ID.

Wherein, the method further includes: dividing each super block into a plurality of first logical blocks of a first predetermined size; and creating an entry parameter list corresponding to all super blocks according to the sum of the number of first logical blocks in each super block.

In order to solve the above technical problems, another technical solution adopted in this application is to provide a solid-state hard disk, which includes a processor and a storage controller coupled to the processor, wherein the storage controller stores a storage device for implementing any of the above A program instruction of a method for garbage disposal based on the LSM database; the processor is used to execute program instructions stored by the storage controller to process garbage on the solid state drive; wherein the LSM database interacts with the solid state drive through SPDK.

In order to solve the above technical problems, another technical solution adopted in this application is to provide a storage device that stores program files that can implement any of the above methods.

The beneficial effects of this application are: the method for garbage disposal based on the LSM database, the solid state hard disk and the storage device of this application create a log corresponding to the data to be written in the solid state hard disk; write the data to be written into the corresponding log ; Under the condition of performing the merge operation, perform the merge operation on the logs in the solid state hard disk; delete the merged log and mark the super block where the deleted log is in a data invalid state. Through the above method, the present application can flexibly perform garbage processing on the data of the LSM database stored in the solid state hard disk, and reduce write amplification.

Description of the drawings

FIG. 1 is a schematic flowchart of a garbage processing method based on an LSM database according to a first embodiment of the present application;

2 is a schematic flowchart of a garbage processing method based on an LSM database according to a second embodiment of the present application;

FIG. 3 is a schematic diagram of the structure of a super block according to an embodiment of the present application;

FIG. 4 is a schematic diagram of the relationship between a super block, a log, and an entry parameter list in an embodiment of the present application;

FIG. 5 is a schematic diagram of the relationship among the hash bucket, the entry parameter list, and the hash link in an embodiment of the present application;

FIG. 6 is a schematic structural diagram of a solid-state hard disk according to an embodiment of the present application;

FIG. 7 is a schematic diagram of the link between the LSM database and the solid state drive shown in FIG. 6;

FIG. 8 is a schematic structural diagram of a storage device according to an embodiment of the present application.

detailed description

The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

The terms "first", "second", and "third" in this application are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second", and "third" may explicitly or implicitly include at least one of the features. In the description of this application, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of this application are only used to explain the relative positional relationship between the components in a specific posture (as shown in the drawings) If the specific posture changes, the directional indication will change accordingly. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes unlisted steps or units, or optionally also includes Other steps or units inherent to these processes, methods, products or equipment.

Reference to "embodiments" herein means that a specific feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art clearly and implicitly understand that the embodiments described herein can be combined with other embodiments.

Fig. 1 is a flowchart of a method for garbage processing based on an LSM database according to a first embodiment of the present application. It should be noted that if there is substantially the same result, the method of the present application is not limited to the sequence of the process shown in FIG. 1. As shown in Figure 1, the method includes steps:

Step S101: Create a log corresponding to the data to be written in the solid state hard disk.

In this embodiment, all interfaces provided by the solid state drive are carried by the nvme protocol, that is, the nvme command is customized on the nvme protocol to implement the corresponding interface; in order to provide higher performance, the LSM database uses SPDK to directly send the corresponding nvme command and Solid state drives interact. Among them, the nvme commands include commands such as creating log CreatLog, appending to writing log AppendLog, reading log ReadLog, and deleting log DeleteLog.

In step S101, the LSM database uses the create log command CreatLog to create a log corresponding to the data to be written in the solid state disk, where the data to be written is an sstfile file or a logfile file in the LSM database.

Step S102: Write the data to be written into the corresponding log.

In step S102, the LSM database uses the append log command AppendLog to write the data to be written into the corresponding log, and store it in the super block of the solid state disk.

Step S103: Perform a merge operation on the logs in the solid state disk under the condition that the merge operation is performed.

In step S103, when the solid-state hard disk meets the trigger condition for executing the merge operation, the merge operation is performed on the logs in the solid-state hard disk. In other words, the time for performing the merge operation on the logs in the solid state drive can be flexibly configured.

Among them, in the process of merging, the LSM database uses the read log command ReadLog to read out the valid data in the log to be merged and transfer it to another effective physical address space.

Step S104: Delete the merged log and identify the super block where the deleted log is in a data invalid state, where the super block is a physical address space allocated in the solid state disk for storing each log.

In step S104, the LSM database uses the delete log command DeleteLog to delete the merged log, and at the same time mark the super block where the log is located as a data invalid state, so as to achieve the purpose of quickly releasing the super block.

FIG. 2 is a schematic flowchart of a method for garbage processing based on flash memory according to a second embodiment of the present application. It should be noted that if there is substantially the same result, the method of the present application is not limited to the sequence of the process shown in FIG. 2. As shown in Figure 2, the method includes the following steps:

Step S201: Acquire group identifiers in the LSM database, and create corresponding super blocks in the solid state disk for groups corresponding to different group identifiers.

In step S201, the group identifier in the LSM database corresponds to the layer identifier where the sstfile file in the LSM database is located, that is, the sstfile file of the same layer in the LSM database belongs to the same group identifier. The solid state disk creates corresponding super blocks for groups corresponding to different group identifiers, where the super block is a physical address space allocated for storing data in the solid state disk.

Step S202: Divide each super block into a plurality of first logical blocks of a first predetermined size.

In step S202, each super block is divided into a plurality of first logical blocks of a first predetermined size, for example, 1 MB, so as to use the first logical blocks in the super block to sequentially store the data to be written.

FIG. 3 is a schematic structural diagram of a super block according to an embodiment of the application. As shown in Fig. 3, the super block includes a plurality of first logical blocks BLOCKN (N=0, 1, 2,...). Among them, the first logical block in the super block is dynamically allocated. After the allocated first logical block completes the storage operation, if the storage of all the data to be written is not completed, continue to allocate a new first logical block to Continue to store the data to be written.

Step S203: Create an entry parameter list corresponding to each super block according to the sum of the number of first logical blocks in each super block.

In step S203, the space for storing the entry table of the entry parameter list is requested according to the number of the first logical blocks provided by the super block to create the entry table of the entry parameter list.

Specifically, assuming that the size of all super blocks is 1TB, the size of the first logical block is 1MB, and each entry parameter entry in the entry parameter list is 4 bytes, the space required for the entry parameter list is 4MB.

Among them, the entry parameter entry does not need to save the identifier of the super block and the logical identifier of the first logical block, that is, which is the first logical block in the super block, which can be listed in the entry parameter list through the entry parameter entry The subscript in the entry table is the corresponding calculation of the position of the entry parameter entry in the entry parameter list entry table, that is, each first logical block corresponds to the entry parameter entry one-to-one.

Step S204: Create a log corresponding to the data to be written in the solid state hard disk.

In step S204, the step of creating a log corresponding to the data to be written in the solid state disk includes: obtaining the group identifier corresponding to the data to be written, and dynamically allocating free space for storing the log in the super block corresponding to the group identifier The first logical block.

When the LSM database creates a log log, it needs to know the group ID of the log log, where the group ID of the log log is the group ID of the data to be written. In addition, when the log log is created, the space allocation of the corresponding log log is allocated on the super block created by the group corresponding to the group identifier.

Step S205: Write the data to be written into the corresponding log.

In step S205, the step of writing the data to be written into the corresponding log includes: writing the data to be written into the log log and dividing the log log into a plurality of second logical blocks of a second predetermined size BlockN(N =0,1,2...); find the entry parameter entry corresponding to the free first logical block; write the content in the second logical block into the corresponding free first logical block; modify the completed write operation The entry parameter entry corresponding to the first logical block is the log identifier of the log log and the logical block identifier of the second logical block that has been written (as shown in FIG. 4).

That is to say, after the data to be written is stored as a log in the first logical block of the super block of the solid state hard disk, the information of the entry parameter corresponding to the first logical block will be updated at the same time.

Preferably, the second logical block and the first logical block have the same size, for example, the same 1MB.

Preferably, the step of writing the data to be written into the corresponding log further includes: using the log ID and the logical block ID block id of the entry parameter as the hash key value hash key, and the hash key value is based on the hash key value. Get the hash value hash value; get the index value index of the hash bucket bucket according to the hash value hash value; check whether the hash bucket bucket corresponding to the index value index is empty; if the check result is that the hash bucket bucket is empty, then Save the subscript of the entry parameter entry in the hash bucket bucket corresponding to the index value; if the detection result is that the hash bucket bucket is not empty, save the entry subscript of the entry parameter entry in the corresponding hash link hashlink.

It needs to be emphasized that if the detection result is that the hash bucket bucket is not empty, that is to say, it proves that a hash conflict has occurred currently. At this time, you need to find the value stored in the hash bucket bucket, that is, the index of the entry parameter The subscript of the corresponding hash link hashlink, where the subscript of the hash link hashlink corresponds to the subscript of the entry parameter entry one-to-one, and then the hash link corresponding to the subscript saves the next entry parameter Mark. If the hash link corresponding to the subscript also has a value, it proves that there is still a conflict. You need to continue to find the next next based on the value saved in the hash link hashlink corresponding to the subscript, that is, the subscript of the entry parameter entry The target hash link hashlink until there is no conflict.

For example, as shown in Figure 5, the hash bucket bucket is not empty, and its saved value is 2, then look for the hash link hashlink with subscript 2 and find that its saved value is 4, then continue to look for the subscript The hash link hashlink is 4, and the saved value is found to be 6, and the hash link hashlink with the subscript 6 is continued to search, and when it is found to be empty, the subscript of the entry parameter entry to be saved can be stored.

Step S206: Perform a merge operation on the logs in the solid state hard disk when the conditions for performing the merge operation are met.

In step S206, the operation of judging whether the condition for performing the merge operation is met is specifically: obtaining the number of free super blocks in the solid state disk; obtaining the access strength of the solid state disk; detecting the number and access of the super blocks in the idle state Whether the strength meets a predetermined condition; wherein, when the detection result is that the number of super blocks in an idle state and the access strength meet the predetermined condition, the condition for performing the merge operation is satisfied. Among them, the access intensity of the solid state drive is the business load of the solid state drive, which can also be understood as the window time, that is, the number of accesses to the solid state drive within a unit time.

Specifically, when the detection result is that the number of super blocks in the idle state is less than the first predetermined threshold and the access intensity of the solid state disk is less than the second predetermined threshold, the merge operation is performed. That is to say, when the number of super blocks in the idle state is relatively small and the access intensity of the solid state disk is relatively small, the merge operation is performed. At this time, the influence of the merge operation on the host's access to the solid state disk can be avoided. Conversely, when the detection result is that the number of super blocks in the idle state is greater than or equal to the first predetermined threshold or the access intensity of the solid state disk is greater than or equal to the second predetermined threshold, the merge operation is not performed to ensure the quality of the host's access to the solid state disk.

The steps of performing a merge operation on the logs in the solid state drive include: obtaining the number of logs stored in each super block and the amount of data corresponding to each log, which can be obtained by traversing the entry parameter entry; according to the number of logs and the amount of data corresponding to each log Obtain the effective data volume corresponding to each super block; merge the super block with the smallest effective data volume. That is to say, the LSM database can select the appropriate time to merge the super block with the smallest amount of effective data in the solid state disk according to the number of idle super blocks inside the solid state disk and the current access intensity of the solid state disk.

Among them, in the merging operation of the super block, the data in the super block needs to be read, and the operation of reading the data is specifically: obtaining the corresponding log ID and logical block ID block id according to the logical address lba of the read IO; Log ID and logical block ID block id are used as the hash key value hash key to calculate the hash value hash value; generate the index value of the hash bucket bucket according to the hash data hash key; according to the index value index of the hash bucket bucket Get the subscript of the entry parameter entry; determine whether the log id stored in the entry parameter entry, the logical block id block id and the log id of the IO read this time, and the logical block id block id are equal, if they are equal, the entry parameter entry is used Calculate the identifier of the corresponding super block and the logical block identifier of the first logical block, and read the data in the first logical block; if they are not equal, look up the entry index of the entry parameter saved in the hashlink, Compare again until it matches.

Step S207: Delete the merged log and mark the super block where the deleted log is located in a data invalid state.

In step S207, each log in the merged super block is deleted and the super block is marked as a data invalid state, thereby releasing the super block to achieve the purpose of garbage disposal.

Please refer to FIG. 6, which is a schematic structural diagram of a solid state drive according to an embodiment of the application. As shown in FIG. 6, the solid state drive 10 includes a processor 11 and a storage controller 12 coupled to the processor.

The storage controller 12 stores program instructions for implementing the LSM database-based garbage processing method described in any of the above embodiments.

The processor 12 is configured to execute program instructions stored by the storage controller 12 to perform garbage processing on the solid state hard disk.

The processor 11 may also be referred to as a CPU (Central Processing Unit, central processing unit). The processor 11 may be an integrated circuit chip with signal processing capabilities. The processor 11 may also be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component . The general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.

Please refer to FIG. 7 together, which is a schematic diagram of the link between the LSM database and the solid state drive shown in FIG. 6. As shown in FIG. 7, the LSM database 20 interacts with the solid-state hard disk through SPDK30 for 10 lines, where SPDK30 is a storage performance development kit.

Specifically, all interfaces provided by the solid state drive 10 are carried by the nvme protocol, that is, the nvme command is customized on the nvme protocol to implement the corresponding interface; in order to provide higher performance, the LSM database 20 uses the SPDK30 to directly send the corresponding nvme command and The solid state drive 10 interacts. Among them, the nvme commands include commands such as creating log CreatLog, appending to writing log AppendLog, reading log ReadLog, and deleting log DeleteLog.

To put it another way, the solid state drive 10 is a solid state drive 10 provided in the form of a log, and the interaction between the LSM database 20 and the solid state drive 10 is realized by adding a special nvme command.

Refer to FIG. 8, which is a schematic structural diagram of a storage device according to an embodiment of the application. The storage device in the embodiment of the present application stores a program file 21 that can implement all the above methods. The program file 21 can be stored in the storage device in the form of a software product, and includes several instructions to enable a computer device (which can It is a personal computer, a server, or a network device, etc.) or a processor (processor) that executes all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage devices include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code , Or terminal devices such as computers, servers, mobile phones, and tablets.

The beneficial effects of this application are: the method for garbage disposal based on the LSM database, the solid state hard disk and the storage device of this application create a log corresponding to the data to be written in the solid state hard disk; write the data to be written into the corresponding log ; Under the condition of performing the merge operation, perform the merge operation on the logs in the solid state hard disk; delete the merged log and mark the super block where the deleted log is in a data invalid state. Through the above method, the present application can avoid the LSM database engine and the solid state hard disk in the prior art from performing repeated garbage processing on the same data. Further, the present application can flexibly perform garbage processing on the data of the LSM database stored in the solid state hard disk, and reduce write amplification.

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

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

The above are only implementations of this application, and do not limit the scope of this application. Any equivalent structure or equivalent process transformation made by using the description and drawings of this application, or directly or indirectly applied to other related technologies In the same way, all fields are included in the scope of patent protection of this application.

Claims

A garbage processing method based on LSM database, characterized in that the method includes:

Creating a log corresponding to the data to be written in the solid state drive, where the data to be written is an sstfile file or a logfile file in the LSM database;

Write the data to be written into the corresponding log;

Performing a merging operation on the logs in the solid-state hard disk when the conditions for performing the merge operation are met;

The merged log is deleted and the super block where the deleted log is located is in a data invalid state, where the super block is a physical address space allocated in the solid-state hard disk for storing each log.
The method of claim 1, wherein the method further comprises:

Acquiring the number of the super blocks in the idle state in the solid state hard disk;

Obtaining the access intensity of the solid state hard disk;

Detecting whether the number of the super blocks in the idle state and the access intensity meet a predetermined condition;

Wherein, when the detection result is that the number of the super blocks in the idle state and the access intensity satisfy a predetermined condition, the condition for performing the merge operation is satisfied.
The method according to claim 1 or 2, wherein the step of performing a merge operation on the logs in the solid state drive comprises:

Acquiring the number of the logs stored in each super block and the amount of data corresponding to each of the logs;

Acquiring the effective data amount corresponding to each super block according to the number of the logs and the data amount corresponding to each of the logs;

The super block with the smallest amount of effective data is merged.
The method according to claim 1, wherein the step of creating a log corresponding to the data to be written in the solid state hard disk comprises:

Obtain the group identifier corresponding to the data to be written, wherein each of the group identifiers corresponds to one of the super blocks, each of the super blocks includes a plurality of first logical blocks of a first predetermined size, and all the super blocks Corresponding to an entry parameter list, the first logical block in each super block corresponds to an entry parameter in the entry parameter list;

The free first logical block for storing the log is dynamically allocated in the super block corresponding to the group identifier.
The method according to claim 4, wherein the step of writing the data to be written into the corresponding log comprises:

Writing the data to be written into the log and dividing the log into a plurality of second logical blocks of a second predetermined size;

Searching for the entry parameter corresponding to the first logical block that is free in the super block;

Write the content in the second logical block into the corresponding free first logical block;

Modifying the entry parameters corresponding to the first logical block for which the writing operation has been completed is the log identifier of the log and the logical block identifier of the second logical block that has been written.
The method according to claim 5, wherein the step of writing the data to be written into the corresponding log further comprises:

Using the log identifier and the logical block identifier written in the entry parameter as a hash key value, and obtaining a hash value according to the hash key value;

Obtaining the index value of the hash bucket according to the hash value;

Detecting whether the hash bucket corresponding to the index value is empty;

If the detection result is that the hash bucket is empty, save the index of the entry parameter in the hash bucket corresponding to the index value;

If the detection result is that the hash bucket is not empty, the subscript of the entry parameter is stored in the corresponding hash link.
The method according to claim 1, characterized in that, before the step of creating a log corresponding to the data to be written in the solid state hard disk, the method further comprises:

Acquiring a group identifier in the LSM database;

In the solid state hard disk, corresponding super blocks are created for groups corresponding to different group identifiers, wherein the sstfile files of the same layer in the LSM database belong to the same group identifier.
The method according to claim 7, wherein the method further comprises:

Dividing each of the super blocks into a plurality of first logical blocks of a first predetermined size;

Create an entry parameter list corresponding to all the super blocks according to the sum of the number of the first logical blocks in each of the super blocks.
A solid-state hard disk, characterized in that, the solid-state hard disk comprises a processor and a storage controller coupled to the processor, wherein:

The storage controller stores program instructions for implementing the method for garbage processing based on the LSM database according to any one of claims 1-8;

The processor is configured to execute the program instructions stored by the storage controller to perform garbage processing on the solid-state hard disk;

Wherein, the LSM database interacts with the solid state hard disk through SPDK.
A storage device, characterized in that it stores a program file capable of implementing the method according to any one of claims 1-8.