WO2019057000A1

WO2019057000A1 - Log writing method, apparatus and system

Info

Publication number: WO2019057000A1
Application number: PCT/CN2018/105904
Authority: WO
Inventors: 曹伟
Original assignee: 阿里巴巴集团控股有限公司
Priority date: 2017-09-21
Filing date: 2018-09-17
Publication date: 2019-03-28
Also published as: CN109542329B; CN109542329A

Abstract

A log writing method, apparatus and system. In this method, in the premise that the size of a physical sector in a log disk is greater than or equal to the sum of the size of log meta information and the size of a reference data block, when a log entry to be written is written into the log disk, the log entry to be written can be spliced to form a memory block by means of address alignment and data splicing, so that the log entry to be written can be written into the physical sector of the log disk in less IO operations, the required number of IO operations for writing the log entry into the log disk is reduced, and the execution efficiency for writing the log entry into the log disk is improved.

Description

Log writing method, device and system

The present application claims the priority of the Japanese Patent Application No. PCT Application No. No. No. No. No. No. No

Technical field

The present application relates to the field of storage technologies, and in particular, to a log writing method, apparatus, and system.

Background technique

Write Ahead Log (WAL) is a technology that guarantees atomicity and durability in the storage system, and provides a good foundation for the recovery mechanism of the storage system.

In a storage system using WAL, before writing data to the database, the data is written to the log file on the log disk, and then the data is persisted to the database, so that when the storage system is down, the WAL can be passed. The log is restored to the state before the downtime to avoid data loss.

In the above process, the efficiency of writing log data to the log disk is relatively low, and a new scheme is needed to improve the efficiency of writing the log.

Summary of the invention

Aspects of the present application provide a log writing method, apparatus, and system for improving the efficiency of writing a log.

The embodiment of the present application provides a log writing method, including:

Obtaining a log entry to be written, where the log entry to be written includes log meta information and log data to be written, where the size of the log data to be written is M times the size of the reference data block, and M is a positive integer;

Performing address alignment on the log entry to be written according to the size of the physical sector in the log disk to determine a starting address of the physical sector for storing the log entry to be written, the physical sector in the log disk The size is greater than or equal to the sum of the size of the log meta information and the size of the reference data block;

Performing data splicing on the log meta information and the to-be-written log data to form N memory blocks, N<=M, and N is a positive integer;

Writing the N memory blocks into the N physical sectors from the start address in the log disk.

The embodiment of the present application further provides a storage management device, including: a memory, a processor, and a communication component;

The memory is configured to store a program;

The processor is coupled to the memory for executing the program for:

Writing, by the communication component, the N memory blocks into N physical sectors in the log disk starting from the start address;

The communication component is configured to write the N memory blocks into N physical sectors in the log disk starting from the start address.

The embodiment of the present application further provides a log-based storage system, including the storage management device, the log disk, and the database provided by the foregoing embodiment; the storage management device is connected to the log disk and the database.

In the embodiment of the present application, when the size of the physical sector in the log disk is greater than or equal to the sum of the size of the log meta information and the size of the reference data block, when the log entry to be written is written to the log disk, the address alignment is combined. And the data splicing can splicing the log items to be written into a memory block, so that the log items to be written can be written into the physical sector of the log disk by less IO operations, and the log entries are written to the log disk. The number of IO operations required increases the efficiency of writing log entries to the log disk.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the present application, and are intended to be a part of this application. In the drawing:

FIG. 1 is a schematic structural diagram of a log-based storage system according to an exemplary embodiment of the present application;

FIG. 2 is a schematic diagram of an alignment state of an address alignment example according to an exemplary embodiment of the present application; FIG.

FIG. 3 is a schematic flowchart of a log writing method according to another exemplary embodiment of the present application;

4a is a schematic structural diagram of a log writing apparatus according to still another exemplary embodiment of the present application;

FIG. 4b is a schematic structural diagram of a storage management device according to another exemplary embodiment of the present application.

Detailed ways

The technical solutions of the present application will be clearly and completely described in the following with reference to the specific embodiments of the present application and the corresponding drawings. It is apparent that the described embodiments are only a part of the embodiments of the present application, and not all of them. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without departing from the inventive scope are the scope of the present application.

The solution of the prior art is that the efficiency of writing log data to the log disk is relatively low. The embodiment of the present application provides a solution. The basic principle is that the size of the physical sector in the log disk is greater than or equal to the size of the log metadata. Based on the sum of the size of the reference data block, when writing the log entry to be written to the log disk, the address alignment and data splicing can be used to splicing the log items to be written into memory blocks, so that fewer IO operations can be performed. The number of IO operations required to write log entries to the log disk is reduced in the physical sectors to be written to the log disk, which improves the efficiency of writing log entries to the log disk.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic structural diagram of a log-based storage system 100 according to an exemplary embodiment of the present application. As shown in FIG. 1, the storage system 100 includes a storage management device 10, a log disk 20, and a database 30.

The storage management device 10 is primarily responsible for the storage control logic of the storage system 100, such as providing read, write, and query functions of the database 30, and the like. The storage management device 10 can be any device having processing capabilities, such as a server, a desktop computer, a personal computer, a mobile phone, a tablet, and the like. The server can be a regular server, a cloud server, a cloud host, a virtual center, and the like. The implementation structure of the storage management device 10 may include a processor, a system bus, and at least one physical storage medium such as a hard disk, a memory, and the like.

Database 30 organizes, stores, and manages data in storage system 100 primarily in accordance with a certain data structure. The database 30 can be various types of databases, which is not limited in this embodiment. For example, from a database developer's perspective, database 30 can be Oracle, DB2, Sybase, MS SQL Server, Informax, MySQL, and the like. From the perspective of data organization structure, the database 30 may be a network database, a relational database, a hierarchical database, an object-oriented database, or the like.

The storage system 100 provided in this embodiment provides a storage function to a user. The user of storage system 100 may be a computer, virtual machine, various types of applications, terminal devices, and users, etc. that have access and/or usage rights to storage system 100. The user of the storage system 10 can have one or more. The user may send an operation request to the storage management device 10 in the storage system 100 during use of the storage system 100. The storage management device 10 receives the operation request sent by the user, and operates on the data at the corresponding address in the database 30 according to the database address in the operation request, for example, the data can be read and returned to the user, or the data is modified, or written. Enter new data and so on. Among them, how the storage management device 10 operates on the data at the corresponding address in the database 30 depends on the type of operation request.

In the storage system 100, in order to achieve atomicity and durability, the operations in the storage system 100 are recorded using WAL technology. The log disk 20 is mainly used to store WAL logs, and the WAL log mainly records database operations in the storage system 100. Based on this, the storage management device 10 can record the operation request of the user to the log disk 20 in the process of operating the data at the corresponding address in the database 30 according to the operation request of the user. Depending on the type of operation request, the storage management device 10 may choose to record the operation request in the log disk 20, or may choose not to record the operation request.

In the present embodiment, the user's operation request is divided into a database update request and a non-update request based on whether the data in the database 30 is changed as a standard. The database update request may change the data in the database 30, for example, may be a write request, a modification request, a delete request, etc.; the non-update request does not change the data in the database 30, and may be, for example, a query request, a read request, or the like.

For the non-update request, the storage management device 10 receives the non-update request sent by the user, reads the corresponding data from the database 30 according to the database address in the non-update request, and returns it to the user. Optionally, the storage management device 10 may select to record the non-update request to the log disk 20, and details are not described herein.

For the database update request, the storage management device 10 needs to record the database update request to the log file, and persist the log file to the log disk 20; then write the data involved in the database update request into the dirty page of the memory, and adopt asynchronous The method persists the dirty pages in the memory into the database 30; after persisting the data to the database 30, a response message to the successful write is returned to the user.

If the storage system 100 is down before the storage management device 10 persists the data to the database 30, the dirty pages in the memory are lost. Since the data update request has been persisted to the log disk 20, when the storage system 100 is restarted, the log disk 20 can be scanned, and database update requests that have not been persisted into the database 30 are obtained from the log disk 20, and these database updates are updated. The request is played back into the dirty page of the memory, and then the dirty pages in the memory are persisted into the database 30 in an asynchronous manner. It can be seen that the storage system 100 of the embodiment adopts the WAL technology and provides a good foundation for the recovery mechanism of the storage system 100.

In the above process, the storage management device 10 needs to record a database update operation or a non-update operation to the log disk 20. Alternatively, the storage management device 10 may generate a log entry to be written according to a database update operation or a non-update operation, and then write the log entry to be written into the log disk 20. Taking the database update operation as an example, the storage management device 10 can receive the database update request sent by the user of the storage system 10, generate a log entry to be written according to the database update request, and record the log entry to be written to the log disk 20. Among them, the database update request mainly includes updating the operation control information and the data block to be written into the database. The update operation control information mainly includes information such as a database address, a length of a data block to be written into the database, and a check code of the data block. Based on this, the storage management device 10 can use the update operation control information and the data block to be written into the database as the log meta information and the log data to be written, respectively, to generate a log entry to be written. Generally, the data block to be written into the database is generally an integer multiple of the reference data block size, denoted as M times, and M is a positive integer; accordingly, the size of the log data to be written is also M times the size of the reference data block.

The reference data block size refers to the size of the data block that can be read and written by a single read/write operation supported by the storage medium under normal circumstances. The size of the data block is related to the development of the storage technology, and different values can be taken in different periods. The reference data block size is equivalent to the size of the conventional physical sector in the current development stage of the storage technology, that is, the amount of data that the storage medium allows reading and writing operations to read and write in a normal case, for example, 512 bytes, 4096 bytes, and the like.

It can be seen from the above definition of the reference data block that the size of the reference data block is 4 KB, the size of the physical sector in the log disk 20 is 4 KB, the size of the log data to be written is M*4 KB, and the size of the log meta information is 64. Byte, because the total size of the log entry to be written is M*4096 bytes (ie, 4 KB) + 64 bytes, which exceeds the size of M physical sectors, so the storage control device 10 writes the log entry to be written. The log disk 20 needs to perform M+1 IO operations, and the log meta information and the log data to be written are respectively written into the M+1 physical sectors in the log disk 20. It can be seen that, due to the existence of the log meta-information, when the storage management device 10 writes a log entry to the log disk 20, a write operation across the physical sector is generated once, and the execution efficiency of the write log is low.

In the storage system 100 of the embodiment, the log disk 20 adopts an unconventional disk, and the size of the physical sector in the log disk 20 is greater than or equal to the size of the log meta information and the size of the reference data block. Sum. For example, assuming that the reference data block size is 4 KB and the log meta information size is 64 bytes, the size of the physical sector in the log disk 20 may be 4160 bytes, 4224 bytes, or the like. Alternatively, the log disk 20 may be a disk supporting a variable sector size, such as a Non-Volatile Memory Express (NVMe) disk, but is not limited thereto. NVMe disks are disks that support the NVMe interface standard.

Based on the support of the log disk 20 in the storage space, and in conjunction with the new log writing process provided by the storage management device 10, the log items to be written can be spliced into memory blocks so that the IO operations can be written to by less IO operations. The log entry is written to the physical sector of the log disk 20, which reduces the number of IO operations across physical sectors required to write log entries to the log disk, and improves the efficiency of writing log entries to the log disk. The process of the storage management device 10 writing a log entry to the log disk 20 is as follows: On the basis that the size of the physical sector in the log disk 20 is greater than or equal to the sum of the size of the log meta information and the size of the reference data block:

The storage management device 10 can obtain a log entry to be written according to an operation request of the user, where the log to be written includes log meta information and log data to be written, and the size of the log data to be written is the size of the reference data block. Times. Thereafter, the storage management device 10 performs address alignment on the one hand according to the size of the physical sector in the log disk 20 to determine the starting address of the physical sector in the log disk 20 for storing the log entry to be written; On the other hand, the log metadata information and the log data to be written are spliced to form N memory blocks, and N memory blocks are written into the N physical sectors starting from the above starting address in the log disk 20, and completed. Write operation of log entries. Where N<=M and N is a positive integer. The memory block here refers to a data block that can be written to one physical sector at a time, and the size of the memory block is smaller than or equal to the size of the physical sector in the log disk 20.

It should be noted that, in this embodiment, the order of address alignment and data splicing is not limited. For example, the storage management device 10 may perform address alignment and data splicing in parallel, or perform address aligning and then perform data splicing, or may perform data splicing before performing address alignment.

Since the physical sector of the log disk is greater than or equal to the sum of the size of the log meta-information and the size of the reference data block, it is possible to splicing the log items to be written into no more than M data blocks through data splicing, thereby occupying a smaller number. The physical sector does not need to occupy M+1 physical sectors, which reduces the number of IO operations across physical sectors required to write log entries to the log disk, and improves the execution of writing log entries to the log disk. effectiveness.

In some exemplary embodiments, it is assumed that the size of the physical sector in the log disk 20 is K bytes, and K is a positive integer. Based on this, after obtaining the log entry to be written, the storage management device 10 can round up the end address of the last write operation in the log disk 20 by K bytes to obtain an address information, that is, the log disk 20 for storage. The starting address of the physical sector to be written to the log entry. For example, suppose the reference data block size is 4 KB (that is, 4096 bytes), the size of the log meta information is 64 bytes, and the size K of the physical sector in the log disk 20 is 4160 bytes, and the last write operation in the log disk 20 The end address is 4150 bytes, and the 4150 bytes are rounded up according to 4160 bytes, and an address, that is, the 4160th byte, can be obtained, which means that the storage can be written from the 4160th byte in the log disk 20. Enter the log entry.

In some exemplary embodiments, when the storage management device 10 performs data splicing on the log meta information and the log data to be written, N memory blocks may be spliced at one time; or, N memory blocks may be spliced each time. Part of the memory block, after a number of stitching, finally get N memory blocks. For example, one memory block can be spliced each time, and N memory blocks are obtained after N times of splicing. Correspondingly, when N memory blocks are written into N physical sectors, N memory blocks may be written into N physical sectors after N times of writing operations after splicing out N memory blocks; or Alternatively, each time a memory block is spliced, a write operation is performed, and the spliced memory block is written into the corresponding physical sector, so that N memory blocks can be written to N physics after N write operations. Sector.

In some exemplary embodiments, when the storage management device 10 performs data splicing on the log meta information and the log data to be written, the log meta information and the log data to be written may be spliced through a memory buffer to form N A memory block whose size is equal to the size of N physical sectors. A memory block corresponds to one physical sector. A memory block is a contiguous piece of data in a memory buffer. Optionally, the memory buffer may be a continuous area in the memory, but is not limited thereto.

The splicing manner of splicing the log meta-information and the data to be written into the log data into the N-segment continuous data through the memory buffer is applicable to the embodiment of the present application.

For example, in an exemplary embodiment, the storage management device 10 may split the log meta information and the log data to be written into N shares, select a continuous area in the memory buffer, and cache one of the data each time. Obtain a memory block in the contiguous area, and write the obtained memory block into the corresponding physical sector, so after N operations, N memory blocks can be spliced and N memory blocks are written into N physics. In the sector.

For another example, in another exemplary embodiment, the storage management device 10 may split the log meta information and the log data to be written into N shares, and cache the N pieces of data into N consecutive areas of the memory buffer. N memory blocks are obtained, and a continuous piece of data is stored in each successive area, and the size of each continuous area is less than or equal to the size of one physical sector.

Optionally, an embodiment of data splicing is given from the perspective of spliced content. In this embodiment, the storage management device 10 may cache the log meta information and the header data or the tail data to be written in the log data into a continuous area of the memory buffer to obtain a memory block; the header data or the tail data. The size is the same as the size of the reference data block; then, in the unit of the reference data block size, the other data to be written into the log data is separately cached into other consecutive areas of the memory buffer to obtain N-1 memory blocks, where N=M. Other continuous areas are N-1 continuous areas. In this embodiment, the log meta information and the header data or the tail data to be written in the log data are buffered into one memory block, and other data to be written in the log data can be buffered in units of the reference data block size. M-1 memory blocks, and finally get M memory blocks, and then the log items to be written are written into the log disk 20 by M times of IO operations across physical sectors, which is reduced compared to M+1 IO operations. An IO operation is beneficial to improve the execution efficiency of the write log.

It is worth noting that, in addition to the reference data block size, other data to be written into the log data is cached as M-1 memory blocks, and other data to be written in the log data may be cached to be smaller than M-1 memory blocks, the specific number depends on the size of the physical sector.

Optionally, an implementation of two data splicing is given from the perspective of memory block continuity. In one embodiment, the storage management device 10 may splicing the log meta information and the log data to be written in a continuous manner in a memory buffer to obtain N memory blocks. In another implementation manner, the storage management device 10 may splicing the log meta information and the log data to be written in an interval in a memory buffer to obtain N memory blocks.

In an embodiment in which N memory blocks are spliced in a continuous manner, the memory blocks are sequentially adjacent to each other with no gap therebetween. In this embodiment, if the size of the memory buffer is greater than the sum of the sizes of the N memory blocks, the extra storage space may be located at the beginning and/or the end of the memory buffer.

The embodiment in which N memory blocks are spliced in an interval manner is applicable to a case where the size of the memory buffer is larger than the sum of the sizes of the N memory blocks. In this embodiment, there is at least one byte gap between two adjacent memory blocks, through which the different memory blocks can be easily distinguished.

In the storage system 100 provided by the embodiment of the present application, the size of the physical sector of the log disk 20 is required to be greater than or equal to the sum of the size of the log meta information and the size of the reference data block. In some exemplary embodiments, a disk whose physical sector size is greater than or equal to the sum of the size of the log meta information and the size of the reference data block may be directly selected as the log disk 20. In other exemplary embodiments, a disk supporting a variable sector size may be selected as the log disk 20.

In an embodiment in which a disk supporting a variable sector size is selected as the log disk 20, at the beginning, the log disk 20 can be implemented as a regular disk whose physical sector size is equal to the reference data block size. Based on this, before the log entry is written to the log disk 20, the size of the physical sector in the log disk can be adjusted to be greater than or equal to the sum of the size of the log meta information and the size of the reference data block.

In an exemplary embodiment, the lower limit value of the size of the physical sector in the log disk 20 may be calculated according to the size of the meta information and the reference data block size; at least two physical sector sizes supported from the log disk 20. In the middle, a size greater than or equal to the lower limit value is selected as the target value; the size of the physical sector in the log disk 20 is adjusted from the current value by the target value.

For example, assuming that the reference data block size is 4 KB (ie, 4096 bytes), the log meta information size is 64 bytes, and the log disk 20 supports sector sizes such as 4096 bytes, 4104 bytes, 4160 bytes, and 4224 bytes. Assuming that the size of the physical sector in the log disk 20 is 4096 bytes at the beginning, the lower limit of the size of the physical sector is calculated according to the reference data block size of 4 KB (ie, 4096 bytes) and the size of the log meta information of 64 bytes. The value is 4096 bytes + 64 bytes = 4160 bytes, and then the sector size greater than or equal to 4160 bytes is selected from sector sizes such as 4096 bytes, 4104 bytes, 4160 bytes, and 4224 bytes, for example, Select 4160 bytes as the target value, and then adjust the size of the physical sector in the log disk 20 from 4096 bytes to 4160 bytes, so that the size of the physical sector satisfies the size of the log meta information and the size of the reference data block. with.

Further, when the size of the physical sector in the log disk 20 is adjusted from the current value by the current value, the log disk may be re-formatted according to the target value in response to the formatting request to obtain a size greater than or equal to the log meta information. The physical sector that is the sum of the size of the reference data block; alternatively, the size of the physical sector in the log disk 20 can be adjusted from the current value by the disk tool or the command line. The embodiment of the present application is not limited to this. Any implementation manner in which the size of the physical sector in the log disk 20 can be adjusted from the current value is adapted to the embodiment of the present application.

To facilitate the understanding of the technical solution of the embodiment of the present application, the process of writing the log to the log disk 20 by the storage management device 10 is illustrated in the following with reference to the address alignment process shown in FIG. 2 .

As shown in FIG. 2, taking the reference data block size as 4 KB (that is, 4096 bytes) as an example, the log disk 20 shown in the left side of FIG. 2 includes physical sectors denoted as A1, B1, and C1, and each physical sector. The size of the log is 4096 bytes, and the size of the log meta information to be written to the log entry is LogEntryHeaderSize, and the size of the log data to be written is 1 times of the reference data block, that is, 4096 bytes. As shown in FIG. 2, if the log entry to be written is written to the log disk 20 shown on the left side of FIG. 2, at least two IO operations are required. In this embodiment, the storage management device 10 can adjust the size of the physical sector of the log disk 20 from 4096 to 4096 bytes and 4104 supported by the log disk 20 by a Variable Sector Size (VSS) technology. A value greater than or equal to (LogEntryHeaderSize+4096) in sector sizes such as bytes, 4160 bytes, and 4224 bytes. In the embodiment shown in FIG. 2, taking LogEntryHeaderSize=64 bytes as an example, if 4096+64=4160 bytes, 4160 bytes or 4224 bytes can be selected, and the embodiment shown in FIG. 2 selects 4160 bytes as example. The log disk 20 after adjusting the size of the physical sector is as shown on the right side of FIG. 2, and includes three physical sectors denoted as A2, B2, and C2.

It should be noted that the log disk 20 shown on the right side of FIG. 2 seems to be larger than the log disk 20 shown on the left side. This is only for convenience of illustration, and the total storage space of the log disk 20 on the left and right sides has not changed. Only the size of the physical sector has changed.

For the convenience of description, the size of the adjusted physical sector is recorded as SectorSize. To ensure that each log entry can be aligned according to SectorSize at the starting position in the log disk 20, the storage management device 10 performs address alignment on the log entry to be written. The alignment result is shown in FIG. 2, and the log entry to be written is aligned with the physical sector B2 in the log disk 20. When the storage management device 10 writes the log entry to be written to the log disk 20, the log meta information to be written to the log entry and the log data to be written are first spliced in a memory buffer to obtain a size of 4160 bytes. The memory block, and the spliced memory block is written once in the physical sector B2 aligned by SectorSize. It can be seen that, based on the log disk 20 shown on the right side of FIG. 2, if the log entry to be written is written to the log disk 20 shown on the right side of FIG. 2, only one IO operation is required, since the number of IO operations is reduced. Can improve the efficiency of writing logs.

FIG. 3 is a schematic flowchart diagram of a log writing method according to another exemplary embodiment of the present application. The method can be applied to the storage system shown in FIG. 1, but is not limited thereto. The log writing method provided in this embodiment can be used in any application scenario involving a write log operation. As shown in FIG. 3, the method includes:

301. Obtain a log entry to be written. The log entry to be written includes log meta information and log data to be written. The size of the log data to be written is M times the size of the reference data block, and M is a positive integer.

302. Perform address alignment on the log entry according to the size of the physical sector in the log disk to determine a starting address of a physical sector for storing the log entry to be written, where the size of the physical sector is greater than or It is equal to the sum of the size of the log meta information and the size of the reference data block.

303. Perform data splicing on the log meta information and the log data to be written to form N memory blocks, where N<=M, and N is a positive integer.

304. Write N memory blocks into the log disk in N physical sectors starting from the above starting address.

In this embodiment, the log disk adopts an unconventional disk. The unconventional mainly means that the size of the physical sector in the log disk is greater than or equal to the sum of the size of the log meta information and the size of the reference data block. The reference data block size refers to the size of the data block that can be read and written by a single read/write operation supported by the storage medium. The size of the data block is related to the development of the storage technology, and different values can be taken in different periods. The reference data block size is equivalent to the size of the conventional physical sector in the current development stage of the storage technology, that is, the amount of data that the storage medium allows reading and writing operations to read and write in a normal case, for example, 512 bytes, 4096 bytes, and the like.

Based on the above, when a log entry needs to be written to the log disk, the log entry to be written can be obtained. The log entry to be written includes log meta information and log data to be written, and the size of the log data to be written is M times the size of the reference data block. Then, on the one hand, according to the size of the physical sector in the log disk, the log entry is to be aligned to determine the starting address of the physical sector in the log disk for storing the log entry to be written; The information and the log data to be written are spliced to form N memory blocks, N <= M, and N is a positive integer. After that, N memory blocks are written into the log disk in the N physical sectors starting from the above starting address, and the log entry is completed. The memory block here refers to a data block that can be written to one physical sector at a time, and the size of the memory block is less than or equal to the size of the physical sector in the log disk.

Since the physical sector of the log disk is greater than or equal to the sum of the size of the log meta-information and the size of the reference data block, it is possible to splicing the log items to be written into no more than M data blocks through data splicing, thereby occupying a smaller number. Physical sectors, instead of M+1, reduce the number of IO operations across physical sectors required to write log entries to the log disk, improving the efficiency of writing log entries to the log disk.

Depending on the application scenario, the way to get log entries to be written varies. For example, in the WAL log-based storage system, a database update request may be received, where the database update request includes update operation control information and a data block to be written into the database; and the update operation control information and the data block to be written into the database are respectively taken as Log meta information and log data to be written to generate log entries to be written. In addition to this method, the log entry to be written may also be obtained in other manners. For example, the log entry to be written may be obtained directly from the log generating device.

In some exemplary embodiments, assume that the size of the physical sector in the log disk is K bytes and K is a positive integer. Based on this, after obtaining the log entry to be written, the end address of the last write operation in the log disk can be rounded up by K bytes to obtain an address information, that is, the log disk is used to store the log entry to be written. The starting address of the physical sector. For example, suppose the base block size is 4KB (that is, 4096 bytes), the size of the log meta information is 64 bytes, and the size K of the physical sector in the log disk is 4160 bytes, and the end address of the last write operation in the log disk. For 4150 bytes, the 4150 bytes are rounded up by 4160 bytes, and an address, that is, 4160 bytes, can be obtained, which means that the log entry to be written can be stored starting from the 4160th byte in the log disk. .

In some exemplary embodiments, when data splicing is performed on the log meta information and the log data to be written, N memory blocks may be spliced at one time; or, part of the N memory blocks may be spliced each time. Block, after multiple stitching, finally get N memory blocks. For example, one memory block can be spliced each time, and N memory blocks are obtained after N times of splicing. Correspondingly, when N memory blocks are written into N physical sectors, N memory blocks may be written into N physical sectors after N times of writing operations after splicing out N memory blocks; or Alternatively, each time a memory block is spliced, a write operation is performed, and the spliced memory block is written into the corresponding physical sector, so that N memory blocks can be written to N physics after N write operations. Sector.

In some exemplary embodiments, when data splicing is performed on the log meta information and the log data to be written, the log meta information and the log data to be written may be spliced through a memory buffer to form N memory blocks. The size of the memory buffer is equal to the size of N physical sectors. A memory block corresponds to one physical sector. A memory block is a contiguous piece of data in a memory buffer. Optionally, the memory buffer may be a continuous area in the memory, but is not limited thereto.

For example, in an exemplary embodiment, the log meta information and the log data to be written may be split into N shares, and a continuous area is selected in the memory buffer, and one of the data is buffered to the continuous area each time. A memory block is obtained, and the obtained memory block is written into the corresponding physical sector, so after N operations, N memory blocks can be spliced and N memory blocks are written into N physical sectors.

For another example, in another exemplary embodiment, the log meta information and the log data to be written may be split into N shares, and the N pieces of data are separately cached into N consecutive areas of the memory buffer to obtain N. A memory block in which a continuous piece of data is stored in each successive area, and the size of each successive area is less than or equal to the size of one physical sector.

Optionally, an embodiment of data splicing is given from the perspective of spliced content. In this embodiment, the log meta information and the header data or the tail data to be written in the log data may be buffered into a continuous area of the memory buffer to obtain a memory block; the size of the header data or the tail data and the reference data. The block size is the same; then, in the unit of the reference data block size, the other data to be written into the log data is separately cached into other consecutive areas of the memory buffer to obtain N-1 memory blocks, where N=M. Other continuous areas are M-1 continuous areas. In this embodiment, the log meta information and the header data or the tail data to be written in the log data are buffered into one memory block, and other data to be written in the log data can be buffered in units of the reference data block size. M-1 memory blocks, and finally get M memory blocks, and then the log items to be written are written into the log disk 20 by M times of IO operations across physical sectors, which is reduced compared to M+1 IO operations. An IO operation is beneficial to improve the execution efficiency of the write log.

Optionally, an implementation of two data splicing is given from the perspective of memory block continuity. In an embodiment, the log meta information and the log data to be written may be spliced in a continuous manner in a memory buffer to obtain N memory blocks. In another implementation manner, the log meta information and the log data to be written may be spliced in an interval in a memory buffer to obtain N memory blocks.

In the above method embodiment, the size of the physical sector of the log disk is required to be greater than or equal to the sum of the size of the log meta information and the size of the reference data block. In some exemplary embodiments, a disk whose physical sector size is greater than or equal to the sum of the log metadata information and the reference data block size may be directly selected as the log disk. In other exemplary embodiments, a disk supporting a variable sector size may be selected as the log disk.

In an embodiment where a variable sector size disk is selected as the log disk, at the beginning, the log disk can be implemented as a regular disk whose physical sector size is equal to the reference data block size. Based on this, before the log entry is written to the log disk, the method further includes the step of adjusting the size of the physical sector in the log disk to be greater than or equal to the sum of the size of the log meta information and the size of the reference data block.

Optionally, the lower limit value of the size of the physical sector in the log disk may be calculated according to the size of the meta information and the reference data block size; and the at least two physical sector sizes supported by the log disk are selected to be greater than or equal to The size of the lower limit is used as the target value; the size of the physical sector in the log disk is adjusted from the current value by the target value.

Further, when the size of the physical sector in the log disk is adjusted from the current value by the current value, the log disk may be re-formatted according to the target value in response to the formatting request to obtain a size greater than or equal to the size of the log meta information. The physical sector of the sum of the reference block sizes; alternatively, the size of the physical sector in the log disk can be adjusted from the current value by the disk tool or the command line. The embodiment of the present application does not limit this. Any implementation manner in which the size of the physical sector in the log disk can be adjusted from the current value is adapted to the embodiment of the present application.

It should be noted that the execution bodies of the steps of the method provided by the foregoing embodiments may all be the same device, or the method may also be performed by different devices. For example, the execution body of steps 301 to 304 may be device A; for example, the execution bodies of steps 301 and 302 may be device A, the execution bodies of

steps

303 and 304 may be device B, and the like.

In addition, some of the processes described in the above-described embodiments and the accompanying drawings include a plurality of operations occurring in a specific order, but it should be clearly understood that the operations may be performed in the order in which they are presented or executed in parallel. The serial number of the operation, such as 301, 302, etc., is only used to distinguish the different operations, and the serial number itself does not represent any execution order. Additionally, these processes may include more or fewer operations, and these operations may be performed sequentially or in parallel. It should be noted that the descriptions of “first” and “second” in this document are used to distinguish different messages, devices, modules, etc., and do not represent the order, nor the “first” and “second”. It is a different type.

FIG. 4 is a schematic structural diagram of a log writing apparatus according to still another exemplary embodiment of the present application. As shown in FIG. 4a, the log writing device includes an acquisition module 41, an alignment module 42, a splicing module 43, and a writing module 44.

The obtaining module 41 is configured to obtain a log entry to be written, where the log entry to be written includes log meta information and log data to be written, and the size of the log data to be written is M times the size of the reference data block, and M is a positive integer.

The aligning module 42 is configured to perform address alignment on the log entry to be written according to the size of the physical sector in the log disk to determine a starting address of a physical sector for storing the log entry to be written, and a physical sector in the log disk The size is greater than or equal to the sum of the size of the log meta information and the size of the reference data block.

The splicing module 43 is configured to perform data splicing on the log meta information and the log data to be written to form N memory blocks, N<=M, and N is a positive integer.

The writing module 44 is configured to write N memory blocks into the N physical sectors in the log disk starting from the above starting address.

In some exemplary embodiments, the aligning module 42 is specifically configured to: round up the last address of the last write operation in the log disk by K bytes to obtain the above starting address, and the size of the physical sector in the log disk is K. byte.

In some exemplary embodiments, the splicing module 43 is specifically configured to: splicing out N memory blocks at a time; or, splicing out a portion of the N memory blocks at a time, and finally obtaining N after multiple splicing Memory block. For example, one memory block can be spliced each time, and N memory blocks are obtained after N times of splicing. Correspondingly, the writing module 44 can write N memory blocks into N physical sectors after N times of writing operations after the splicing module 43 splices out N memory blocks; or, each splicing module 43 can be spliced out. A memory block performs a write operation, and the memory blocks spliced out by the splicing module 43 are written into the corresponding physical sectors, so that N memory blocks can be written into N physical sectors after N write operations.

In some exemplary embodiments, the splicing module 43 is specifically configured to: splice log metadata information and log data to be written through a memory buffer to form the N memory blocks, where the size of the memory buffer is equal to The size of the N physical sectors.

Further, the splicing module 43 is specifically configured to cache the log metadata information and the header data or the tail data to be written into the log data into a continuous area of the memory buffer to obtain a memory block, or the header data or The size of the tail data is the reference data block size; in the unit of the reference data block size, the other data to be written into the log data is separately buffered into other consecutive areas of the memory buffer to obtain N-1 memory blocks. N=M.

Further, the splicing module 43 is specifically configured to: in a memory buffer, splicing the log meta information and the log data to be written in a continuous manner to obtain N memory blocks; or, in the memory buffer, at intervals The method combines the log meta information and the log data to be written to obtain N memory blocks.

In some exemplary embodiments, the obtaining module 41 is specifically configured to: receive a database update request, the database update request includes updating operation control information and a data block to be written into the database; and updating the operation control information and the data block to be written into the database The log metadata information and the log data to be written are respectively generated to generate a log entry to be written.

In some exemplary embodiments, the log writing device further includes: an adjustment module. The adjustment module is configured to adjust the size of the physical sector in the log disk to be greater than or equal to the sum of the size of the log meta information and the reference data block size before the alignment module 42 performs the address alignment operation.

In some exemplary embodiments, the adjusting module is specifically configured to: calculate a lower limit value of a size of a physical sector in the log disk according to a size of the meta information and a reference data block size; and at least two physical fans supported by the log disk In the area size, a size greater than or equal to the lower limit value is selected as the target value; the size of the physical sector in the log disk is adjusted from the current value by the target value.

In some exemplary embodiments, when the size of the physical sector in the log disk is adjusted from the current value by the target value, the adjustment module is specifically configured to: in response to the formatting request, re-format the log disk according to the target value, to Obtain a physical sector whose size is greater than or equal to the sum of the size of the log meta information and the size of the reference data block; or, by using the disk tool or the command line, the size of the physical sector in the log disk is adjusted from the current value by the target value. The embodiment of the present application does not limit this. Any implementation manner in which the size of the physical sector in the log disk can be adjusted from the current value is adapted to the embodiment of the present application.

The log writing device provided in this embodiment is based on the sum of the size of the physical sector in the log disk being greater than or equal to the size of the log metadata information and the size of the reference data block, and when writing the log entry to be written to the log disk, Combining address alignment and data splicing can splicing the log items to be written into memory blocks, so that the log items to be written can be written into the physical sectors of the log disk by less IO operations, and writing to the log disk is reduced. The number of IO operations required for log entries improves the efficiency of writing log entries to the log disk.

The internal function and structure of the log writing device are described above, as shown in FIG. 4b. In practice, the log writing device can be implemented as a storage management device, including: a memory 51, a processor 52, and a communication component 53.

The memory 51 can be configured to store other various data to support operations on the storage management device. Examples of such data include instructions for any application or method operating on a storage management device, phone book data, messages, pictures, videos, and the like.

The memory 51 can be implemented by any type of volatile or non-volatile storage medium or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable. Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Disk or Optical Disk.

The processor 52 is coupled to the memory 51 for executing a program in the memory 51 for:

Obtaining a log entry to be written, the log entry to be written includes log meta information and log data to be written, and the size of the log data to be written is M times the size of the reference data block, and M is a positive integer;

The write log entry is address aligned according to the size of the physical sector in the log disk to determine the starting address of the physical sector for storing the log entry to be written. The size of the physical sector in the log disk is greater than or equal to the log. The sum of the size of the meta information and the size of the reference data block;

Performing data splicing on the log meta information and the log data to be written to form N memory blocks, N<=M, and N is a positive integer;

N memory blocks are written by the communication component 53 into N physical sectors starting from the above starting address in the log disk.

Correspondingly, the communication component 53 is configured to write N memory blocks into the N physical sectors starting from the above starting address in the log disk.

In some exemplary embodiments, when performing address alignment, the processor 52 is specifically configured to: round up the end address of the last write operation in the log disk by K bytes to obtain the above starting address, and physical in the log disk. The size of the sector is K bytes.

In some exemplary embodiments, when performing data splicing, the processor 52 is specifically configured to: splicing out N memory blocks at a time; or splicing out a part of the memory blocks in each of the N memory blocks, and splicing multiple times. Finally get N memory blocks. For example, one memory block can be spliced each time, and N memory blocks are obtained after N times of splicing. Correspondingly, when the N memory blocks are written into the N physical sectors, the processor 52 is specifically configured to: after splicing out the N memory blocks, write N memory blocks to N physical physics after N times of writing operations. Sector; or, each time a memory block is spliced, a write operation is performed, and the spliced memory block is written into the corresponding physical sector, so that N memory blocks can be written to N after N write operations. Physical sectors.

In some exemplary embodiments, when performing data splicing, the processor 52 is specifically configured to: splice log metadata information and log data to be written through a memory buffer to form the N memory blocks, where the memory The size of the buffer is equal to the size of the N physical sectors.

In some exemplary embodiments, when splicing a memory block, the processor 52 is specifically configured to: cache log metadata and header data or tail data to be written into the log data into a continuous area of the memory buffer to obtain a memory block, the size of the header data or the tail data is a reference data block size; in the unit of the reference data block size, the other data to be written into the log data is separately cached in other consecutive areas of the memory buffer to obtain N-1 memory blocks, at this time N=M.

In some exemplary embodiments, when the memory block is spliced, the processor 52 is specifically configured to: in a memory buffer, splicing the log meta information and the log data to be written in a continuous manner to obtain N memory blocks; or In the memory buffer, the log meta information and the log data to be written are spliced in an interval manner to obtain N memory blocks.

In some exemplary embodiments, when acquiring the log item to be written, the processor 52 is specifically configured to: receive a database update request, where the database update request includes updating operation control information and a data block to be written into the database; and controlling the update operation The information and the data block to be written into the database are respectively used as log meta information and log data to be written to generate a log entry to be written.

In some exemplary embodiments, the processor 52 is further configured to adjust the size of the physical sector in the log disk to be greater than or equal to the sum of the size of the log meta information and the reference data block size before performing the address alignment operation.

In some exemplary embodiments, when adjusting the size of the physical sector in the log disk, the processor 52 is specifically configured to: calculate a lower limit of the size of the physical sector in the log disk according to the size of the meta information and the reference data block size. Value; from the at least two physical sector sizes supported by the log disk, select a size greater than or equal to the lower limit as the target value; adjust the size of the physical sector in the log disk from the current value to the target value.

In some exemplary embodiments, when the size of the physical sector in the log disk is adjusted by the current value from the current value, the processor 52 is specifically configured to: re-format the log disk according to the target value in response to the formatting request, A physical sector having a size greater than or equal to the size of the log meta-information and the size of the reference data block is obtained; or the size of the physical sector in the log disk may be adjusted from the current value by the disk tool or the command line. The embodiment of the present application does not limit this. Any implementation manner in which the size of the physical sector in the log disk can be adjusted from the current value is adapted to the embodiment of the present application.

Further, as shown in FIG. 4b, the storage management device further includes: a display 54, a power supply component 55, an audio component 56, and the like. Only some of the components are schematically illustrated in Figure 4b, and it is not meant that the storage management device includes only the components shown in Figure 4b.

The communication component 53, can be configured to facilitate wired or wireless communication between the device to which the communication component belongs and other devices. The device to which the communication component belongs can access a wireless network based on a communication standard such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast associated information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component further includes a near field communication (NFC) module to facilitate short range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

The display 54, which may include a screen, may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensor may sense not only the boundary of the touch or sliding action, but also the duration and pressure associated with the touch or slide operation.

The power supply assembly 55 provides power to various components of the device to which the power supply component belongs. The power components can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the devices to which the power components belong.

The audio component 56 is configured to output and/or input an audio signal. For example, the audio component includes a microphone (MIC) that is configured to receive an external audio signal when the device to which the audio component belongs is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in a memory or transmitted via a communication component. In some embodiments, the audio component further includes a speaker for outputting an audio signal.

The storage management device provided in this embodiment combines the size of the physical sector in the log disk to be greater than or equal to the size of the log metadata information and the size of the reference data block, and writes the log entry to be written to the log disk. Address alignment and data splicing can splicing the log items to be written into memory blocks, so that the log items to be written can be written into the physical sectors of the log disk by less IO operations, and the log is written to the log disk. The number of IO operations required by the item improves the efficiency of writing log entries to the log disk.

Correspondingly, the embodiment of the present application further provides a computer readable storage medium storing a computer program, and the computer program can be implemented by a computer to implement the method steps or functions in the foregoing embodiments, and details are not described herein.

Those skilled in the art will appreciate that embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory. Memory is an example of a computer readable medium.

Computer readable media includes both permanent and non-persistent, removable and non-removable media. Information storage can be implemented by any method or technology. The information can be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary storage of computer readable media, such as modulated data signals and carrier waves.

It is also to be understood that the terms "comprises" or "comprising" or "comprising" or any other variations are intended to encompass a non-exclusive inclusion, such that a process, method, article, Other elements not explicitly listed, or elements that are inherent to such a process, method, commodity, or equipment. An element defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device including the element.

The above description is only an embodiment of the present application and is not intended to limit the application. Various changes and modifications can be made to the present application by those skilled in the art. Any modifications, equivalents, improvements, etc. made within the spirit and scope of the present application are intended to be included within the scope of the appended claims.

Claims

A log writing method, comprising:

Obtaining a log entry to be written, where the log entry to be written includes log meta information and log data to be written, where the size of the log data to be written is M times the size of the reference data block, and M is a positive integer;

Performing address alignment on the log entry to be written according to the size of the physical sector in the log disk to determine a starting address of the physical sector for storing the log entry to be written, the physical sector in the log disk The size is greater than or equal to the sum of the size of the log meta information and the size of the reference data block;

Performing data splicing on the log meta information and the to-be-written log data to form N memory blocks, N<=M, and N is a positive integer;

Writing the N memory blocks into the N physical sectors from the start address in the log disk.
The method according to claim 1, wherein the log entry to be written is address aligned according to the size of a physical sector in the log disk to determine a physics for storing the log entry to be written. The starting address of the sector, including:

The end address of the last write operation in the log disk is rounded up by K bytes to obtain the start address, and the size of the physical sector in the log disk is K bytes, and K is a positive integer.
The method according to claim 1, wherein the data splicing of the log meta information and the log data to be written to form N memory blocks comprises:

The log metadata information and the to-be-written log data are spliced by a memory buffer to form the N memory blocks, and the size of the memory buffer is equal to the size of the N physical sectors.
The method according to claim 3, wherein the splicing the log meta information and the log data to be written by a memory buffer to form the N memory blocks comprises:

Cacheing the log meta information and the header data or the tail data in the log data to be cached into a contiguous area of the memory buffer to obtain a memory block, where the size of the header data or the tail data is The reference data block size;

All other data in the log data to be written are respectively buffered into other consecutive areas of the memory buffer in units of the reference data block size to obtain N-1 memory blocks, N=M.
The method according to claim 3, wherein the splicing the log meta information and the log data to be written by a memory buffer to form the N memory blocks comprises:

In the memory buffer, splicing the log meta information and the to-be-written log data in a continuous manner to obtain N memory blocks; or

In the memory buffer, the log meta information and the log data to be written are spliced in an interval manner to obtain N memory blocks.
The method according to claim 1, wherein the obtaining a log entry to be written comprises:

Receiving a database update request, the database update request including update operation control information and a data block to be written into the database;

The update operation control information and the data block to be written into the database are respectively used as the log meta information and the log data to be written to generate the log entry to be written.
The method according to any one of claims 1-6, wherein before the address alignment of the log entry to be written is performed according to the size of the physical sector in the log disk, the method further includes:

The size of the physical sector in the log disk is adjusted to be greater than or equal to the sum of the size of the log meta information and the size of the reference data block.
The method according to claim 7, wherein the adjusting the size of the physical sector in the log disk to be greater than or equal to the sum of the size of the log meta information and the size of the reference data block comprises:

Calculating a lower limit value of a size of a physical sector in the log disk according to the size of the meta information and the reference data block size;

Selecting, from the at least two physical sector sizes supported by the log disk, a size greater than or equal to the lower limit value as a target value;

The target value is adjusted from the current value by the size of the physical sector in the log disk.
A storage management device, comprising: a memory, a processor, and a communication component;

The memory is configured to store a program;

The processor is coupled to the memory for executing the program for:

Obtaining a log entry to be written, where the log entry to be written includes log meta information and log data to be written, where the size of the log data to be written is M times the size of the reference data block, and M is a positive integer;

Performing address alignment on the log entry to be written according to the size of the physical sector in the log disk to determine a starting address of the physical sector for storing the log entry to be written, the physical sector in the log disk The size is greater than or equal to the sum of the size of the log meta information and the size of the reference data block;

Performing data splicing on the log meta information and the to-be-written log data to form N memory blocks, N<=M, and N is a positive integer;

Writing, by the communication component, the N memory blocks into N physical sectors in the log disk starting from the start address;

The communication component is configured to write the N memory blocks into N physical sectors in the log disk starting from the start address.
The storage management device according to claim 9, wherein when the processor performs address alignment, the processor is specifically configured to:

The end address of the last write operation in the log disk is rounded up by K bytes to obtain the start address, and the size of the physical sector in the log disk is K bytes, and K is a positive integer.
The storage management device according to claim 9, wherein the processor is specifically configured to:

The log metadata information and the to-be-written log data are spliced by a memory buffer to form the N memory blocks, and the size of the memory buffer is equal to the size of the N physical sectors.
The storage management device according to claim 11, wherein when the processor forms the N memory blocks, the processor is specifically configured to:

Cacheing the log meta information and the header data or the tail data in the log data to be cached into a contiguous area of the memory buffer to obtain a memory block, where the size of the header data or the tail data is The reference data block size;

All other data in the log data to be written are respectively buffered into other consecutive areas of the memory buffer in units of the reference data block size to obtain N-1 memory blocks, N=M.
The storage management device according to claim 11, wherein when the processor forms the N memory blocks, the processor is specifically configured to:

In the memory buffer, splicing the log meta information and the to-be-written log data in a continuous manner to obtain N memory blocks; or

In the memory buffer, the log meta information and the log data to be written are spliced in an interval manner to obtain N memory blocks.
The storage management device according to any one of claims 9 to 13, wherein the processor is further configured to: adjust a size of a physical sector in the log disk to be greater than or equal to the log meta information. The sum of the size and the size of the reference data block.
The storage management device according to claim 14, wherein the processor is specifically configured to:

Calculating a lower limit value of a size of a physical sector in the log disk according to the size of the meta information and the reference data block size;

Selecting, from the at least two physical sector sizes supported by the log disk, a size greater than or equal to the lower limit value as a target value;

The target value is adjusted from the current value by the size of the physical sector in the log disk.
A log-based storage system, comprising: the storage management device, the log disk, and the database according to any one of claims 9-15; wherein the storage management device is connected to the log disk and the database.