CN106776790A - Concurrent master-slave synchronisation method and device based on token - Google Patents

Abstract

The invention discloses a token-based concurrent master-slave synchronization method and device. The method includes: receiving a data synchronization task from the main database; setting a token for the data synchronization task according to the order in which the data synchronization task is received; distributing the data synchronization task set with the token to one thread among multiple threads, and the Threads perform data synchronization tasks based on tokens. Utilizing the above scheme of the present invention, multiple threads in the slave database can be used to execute part of the data synchronization tasks in parallel, which improves the data synchronization speed and overcomes the significant out-of-sync data between the master database and the slave database caused by the single-threaded execution of data synchronization tasks in the slave database The problem is to limit the execution sequence of data synchronization tasks by setting tokens, and execute data synchronization tasks serially based on tokens to ensure data consistency with the master database.

Description

Token-based concurrent master-slave synchronization method and device

technical field

The invention relates to the technical field of the Internet, in particular to a token-based concurrent master-slave synchronization method and device.

Background technique

A distributed data storage system is to disperse and store data on multiple independent devices. That is to say, for a piece of data, it is stored on N databases, usually in one master database and multiple slave databases. Among them, the slave The data in the database is obtained by synchronizing with the main database.

The data synchronization process between the master database and the slave database in the prior art is specifically, after the master database receives the client’s request to write data, writes the data into the master database, and records the write data operation to the log file of the master database middle. Subsequently, the master database sends a data synchronization task to the slave database, and the slave database executes a write data operation of writing data into the slave database according to the data synchronization task, and records the write data operation into a log file of the slave database. During the data synchronization process, only one thread of the slave database is used to process the data synchronization task sent by the master database. This thread receives the data synchronization task from the main database, and needs to serially execute the data writing operation and log writing operation involved in the data synchronization task. In the case of too many data synchronization tasks in a short period of time, the thread cannot process the data synchronization tasks in a timely manner, resulting in the problem of low synchronization efficiency of the slave database, resulting in data inconsistency between the master database and the slave database.

Contents of the invention

In view of the above problems, the present invention is proposed to provide a token-based concurrent master-slave synchronization method and a corresponding token-based concurrent master-slave synchronization device that overcome the above problems or at least partially solve the above problems.

According to one aspect of the present invention, a token-based concurrent master-slave synchronization method is provided, comprising:

Receive data synchronization tasks from the main database;

Set tokens for data synchronization tasks in the order in which they are received;

The data synchronization task set with the token is distributed to one of the multiple threads, and the thread executes the data synchronization task according to the token.

According to another aspect of the present invention, a token-based concurrent master-slave synchronization device is provided, which includes:

A receiving module adapted to receive data synchronization tasks from the master database;

The setting module is suitable for setting tokens for the data synchronization tasks according to the order in which the data synchronization tasks are received;

A distribution module, suitable for distributing the data synchronization task provided with the token to a parallel processing module in a plurality of parallel processing modules;

Multiple parallel processing modules adapted to perform data synchronization tasks based on tokens.

According to the solution provided by the present invention, the data synchronization task from the master database is received, a token is set for the data synchronization task according to the order in which the data synchronization task is received, and the data synchronization task with the token is distributed to one of the multiple threads, The data synchronization task is performed by this thread according to the token. Utilizing the above scheme of the present invention, multiple threads in the slave database can be used to execute part of the data synchronization tasks in parallel, which improves the data synchronization speed and overcomes the significant out-of-sync between the data in the master database and the slave database caused by the single-threaded execution of the data synchronization tasks in the slave database The problem is to limit the execution sequence of data synchronization tasks by setting tokens, and execute data synchronization tasks serially based on tokens to ensure data consistency with the master database.

The above description is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable , the specific embodiments of the present invention are enumerated below.

Description of drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiment. The drawings are only for the purpose of illustrating a preferred embodiment and are not to be considered as limiting the invention. Also throughout the drawings, the same reference numerals are used to designate the same parts. In the attached picture:

Fig. 1 shows a schematic flow diagram of a token-based concurrent master-slave synchronization method according to an embodiment of the present invention;

Fig. 2 shows a schematic flow diagram of a token-based concurrent master-slave synchronization method according to another embodiment of the present invention;

Fig. 3 shows a schematic structural diagram of a token-based concurrent master-slave synchronization device according to an embodiment of the present invention.

detailed description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

Fig. 1 shows a schematic flowchart of a token-based concurrent master-slave synchronization method according to an embodiment of the present invention. As shown in Figure 1, the method includes the following steps:

Step S100, receiving a data synchronization task from the master database.

A distributed data storage system is to disperse and store data on multiple independent devices. That is to say, for a piece of data, it is stored in N databases, usually in one master database and multiple slave databases. The data is obtained by synchronizing with the master database, that is to say, after the data is written into the master database, the master database needs to synchronize the written data to the slave database. Therefore, the master database sends a data synchronization task to the slave database, and the slave database receives Data synchronization tasks from the master database.

The data synchronization task is generated according to the data written into the master database, that is to say, a corresponding data synchronization task will be generated for any data written into the master database.

Step S101, according to the order in which the data synchronization tasks are received, tokens are set for the data synchronization tasks.

Wherein, the tokens mark the execution order of the data synchronization tasks, and only after the data synchronization tasks corresponding to the tokens in the earlier order are executed, can the data synchronization tasks corresponding to the tokens in the lower order be executed. For example, according to the order of receiving data synchronization tasks, the tokens of the data synchronization tasks are set to 1, 2, 3, and 4 respectively. In this way, the command can be executed only after the data synchronization task corresponding to token 1 is executed. For the data synchronization task corresponding to card 2, if the data synchronization task corresponding to token 1 has not been executed, the data synchronization task corresponding to token 2 cannot be executed. Similarly, the data corresponding to tokens 3 and 4 will be executed by analogy. Synchronize tasks. In the embodiment of the present invention, a token is set for the data synchronization task, which ensures that the data synchronized to the slave database and the data written to the master database will not be inconsistent during the data synchronization process.

Step S102, distributing the data synchronization task set with the token to one thread among multiple threads, and the thread executes the data synchronization task according to the token.

Specifically, after the token is set for the data synchronization task, the data synchronization task with the token can be distributed to one thread among multiple threads, and the thread executes the data synchronization task according to the token. For data synchronization tasks of different data Can be distributed to different threads for processing.

According to the method provided by the above-mentioned embodiments of the present invention, the data synchronization task from the main database is received, the token is set for the data synchronization task according to the order in which the data synchronization task is received, and the data synchronization task with the token is distributed to multiple threads A thread that performs data synchronization tasks based on the token. Utilizing the above scheme of the present invention, multiple threads in the slave database can be used to execute part of the data synchronization tasks in parallel, which improves the data synchronization speed and overcomes the significant out-of-sync between the data in the master database and the slave database caused by the single-threaded execution of the data synchronization tasks in the slave database The problem is to limit the execution sequence of data synchronization tasks by setting tokens, and execute data synchronization tasks serially based on tokens to ensure data consistency with the master database.

Fig. 2 shows a schematic flowchart of a token-based concurrent master-slave synchronization method according to another embodiment of the present invention. As shown in Figure 2, the method includes the following steps:

Step S200, receiving a data synchronization task from the master database.

The data synchronization task is generated according to the data written into the master database, that is to say, a corresponding data synchronization task will be generated for any data written into the master database.

For example, if the user writes data Set a=1; Set b=2; Set c=1; Set a=4 to the master database through the client, the master database will send 4 data synchronization tasks Set a=1 to the slave database respectively ; Set b=2; Set c=1; Set a=4, that is, the data synchronization task received from the master database from the database is: Set a=1; Set b=2; Set c=1; Set a=4.

Step S201, according to the order in which the data synchronization tasks are received, tokens are set for the data synchronization tasks.

Among them, the master database can send data synchronization tasks to the slave database in real time, that is to say, the slave database can continuously receive the data synchronization tasks sent by the master database, and the master database sends data to the slave database in the order in which the data is written to the master database Synchronization tasks, therefore, when setting tokens for data synchronization tasks, tokens also need to be set in the order in which data synchronization tasks are received. For example, the master database sequentially sends 4 data synchronization tasks to the slave database, namely: Set a=1; Set b=2; Set c=1; Set a=4, then, the data synchronization tasks received from the database The sequence is: Set a=1; Set b=2; Set c=1; Set a=4, so that the slave database can set tokens for data synchronization tasks in the order in which they are received, for example, set 4 The tokens for data synchronization tasks are 1, 2, 3, 4.

Step S202, selecting a thread from multiple threads according to the load balancing strategy.

Among them, the load balancing strategy refers to the business load pressure balance of multiple threads, and the business load pressure of some threads is very high, while the business load pressure of some threads is small, and the data synchronization task is distributed according to the business load of the threads. .

After setting the token for the data synchronization task, the data synchronization task needs to be distributed to the thread, and the thread performs the data synchronization task. There are multiple threads in the database that can perform the data synchronization task. Here, a suitable one needs to be selected from multiple threads. Threads are used to perform data synchronization tasks. The basis for selecting threads is the load balancing strategy. The purpose is to balance the business load pressure among each thread, so that the business load pressure of each thread is basically balanced.

Step S203, distributing the data synchronization task set with the token to the selected thread.

After selecting a thread from multiple threads according to the load balancing strategy, the data synchronization task set in the token is distributed to the selected thread, and the thread performs the data synchronization task.

Wherein, the data synchronization task includes: a write data operation of writing data into the slave database, and a write log operation of recording the write data operation into a log file. The log file is used to record data write operation records, so that when data loss occurs in the slave database, the data in the slave database can be easily restored according to the write data operation records in the log file. Therefore, the write data operation records required to be recorded in the log file are executed in the order in which the data synchronization tasks are received.

There can be multiple data synchronization tasks received from the database. That is to say, multiple threads can be selected to perform data synchronization tasks according to the load balancing strategy. Among them, the data writing operations performed between each thread are parallel operations; The write log operation is a serial operation. For different data, each thread can execute the write data operation of writing data to the slave database in parallel without any impact on the data written to the slave database. However, the data synchronization task for the same data is distributed to the same thread, that is, That is to say, for the same data, the write data operation of writing data to the slave database should also be executed serially by threads, the purpose is to prevent distribution to different threads for processing, resulting in inconsistency between the data of the slave database and the data of the master data .

In addition, each thread defines functions that can perform parallel operations and serial operations. For example, the function that performs parallel operations is function1, and the function that performs serial operations is function2. That is to say, write data to the slave database. The write data operation can be performed by function1, and the write log operation that records the write data operation into the log file can be performed by function2. After the thread receives the data synchronization task, it will perform the data synchronization task by function1 and function2.

In step S204, the thread executes a data write operation for writing data into the slave database.

For example, for the data synchronization task: Set a=1; Set b=2; Set c=1; Set a=4, the data synchronization task Set a=1; Set a=4 can be distributed to threads according to the load balancing strategy 1, the data synchronization task Set b=2 is distributed to thread 2, and the data synchronization task Set c=1 is distributed to thread 3, wherein, for the write data operation of writing data from the database in the data synchronization task, the first thread 1, 2 and 3 can be executed in parallel, but the write log operation of recording the write data operation into the log file needs to be executed serially by the first threads 1, 2, and 3. When threads 1, 2, and 3 perform the write log operation of recording the write data operation into the log file, they need to execute according to the token. Specifically, the following method steps can be used to perform the write data operation recorded into the log file. Write log operations.

Step S205, judging whether the log writing operation of the data synchronization task corresponding to the token ranked one place before the token is completed, if yes, execute step S206; if not, execute step S207.

Wherein, the arrangement order of the tokens is consistent with the order of receiving data synchronization tasks.

Specifically, before function2 executes the write log operation of recording the write data operation into the log file, it also needs to judge whether the write log operation of the data synchronization task corresponding to the token that is ranked one place before the token is completed. The function function2 The code can realize the judgment function. Here, it is judged whether the log writing operation of the data synchronization task corresponding to the token ranked in the first order of the token is completed. The main purpose is to ensure that the log writing operation is executed in the order of receiving the data synchronization task. It is ensured that when problems such as data loss in the slave database occur, the data in the slave database can be easily restored according to the write data operation records in the log file.

For example, for the data synchronization task from the database: Set a=1; Set b=2; Set c=1; Set a=4, set tokens to 1, 2, 3, 4 respectively. When thread 2 executes and records Set b=2 in the log file, it needs to judge whether the data synchronization task corresponding to token 1 is executed. Similarly, when thread 3 executes and records Set c=1 in the log file, it needs to Determine whether the data synchronization tasks corresponding to tokens 1 and 2 have been executed. Similarly, when thread 1 executes and records Set a=4 in the log file, it needs to determine whether the data synchronization tasks corresponding to tokens 1, 2, and 3 are executed. Finish.

Step S206, executing a write log operation for recording the write data operation into a log file.

If it is determined that the log write operation of the data synchronization task corresponding to the token ranked one bit before the token is completed, the thread may execute the log write operation for recording the write data operation into the log file.

Step S207, continue to wait.

If it is judged that the log writing operation of the data synchronization task corresponding to the token that ranks one place before the token has not been completed, you need to continue to wait for the data synchronization task corresponding to the token that ranks one place before the token. After the log write operation is completed, the log write operation corresponding to the token that records the write data operation into the log file is executed.

According to the method provided by the above-mentioned embodiments of the present invention, there are multiple threads in the slave database that can be used to execute part of the data synchronization tasks in parallel, which improves the data synchronization speed and overcomes the gap between the master database and the slave database caused by the single-threaded execution of the data synchronization tasks in the slave database. If the data is significantly out of sync, set a token for the data synchronization task in the order in which the data synchronization task is received, and select a thread from multiple threads according to the load balancing strategy, and distribute the data synchronization task with the token to the selected Threads can ensure that the load and business pressure of each thread is balanced, and there will be no situation where some threads have too much load and business pressure, and some threads have too little load pressure, and the data synchronization corresponding to the token whose order is ranked before the token is judged Whether the log writing operation of the task is completed, if so, execute the log writing operation used to record the write data operation into the log file; if not, continue to wait, so that the log writing operation recorded to the log file is synchronized according to the received data The order of tasks is executed strictly serially, which ensures that in the event of data loss in the slave database, the data in the slave database can be easily restored according to the write data operation records in the log file.

Fig. 3 shows a schematic structural diagram of a token-based concurrent master-slave synchronization device according to an embodiment of the present invention. As shown in FIG. 3 , the device includes: a receiving module 300 , a setting module 310 , a distribution module 320 , and multiple parallel processing modules 330 .

The receiving module 300 is adapted to receive data synchronization tasks from the master database.

The setting module 310 is adapted to set tokens for the data synchronization tasks according to the order in which the data synchronization tasks are received.

The distribution module 320 is adapted to distribute the data synchronization task set with the token to one parallel processing module in the plurality of parallel processing modules.

Multiple parallel processing modules 330 are adapted to execute data synchronization tasks according to tokens.

Optionally, the distribution module 320 is further adapted to: select a parallel processing module from multiple parallel processing modules according to a load balancing strategy; and distribute the data synchronization task with the token set to the selected parallel processing module.

Optionally, the data synchronization task includes: a write data operation of writing data into the slave database, and a write log operation of recording the write data operation into a log file.

Optionally, the data writing operation performed between the parallel processing modules is a parallel operation; the log writing operation performed between the parallel processing modules is a serial operation.

Optionally, the parallel processing module 330 is further adapted to: judging whether the log writing operation of the data synchronization task corresponding to the token preceding the token is completed; The sequence is consistent; if yes, execute the write log operation for recording the write data operation into the log file; if not, continue to wait.

Optionally, the distribution module is further adapted to: distribute data synchronization tasks for the same data to the same thread.

According to the device provided by the above-mentioned embodiments of the present invention, there are multiple threads in the slave database that can be used to execute part of the data synchronization tasks in parallel, which improves the data synchronization speed and overcomes the gap between the master database and the slave database caused by the single-threaded execution of the data synchronization tasks in the slave database. If the data is significantly out of sync, set a token for the data synchronization task in the order in which the data synchronization task is received, and select a thread from multiple threads according to the load balancing strategy, and distribute the data synchronization task with the token to the selected Thread, judging whether the write log operation of the data synchronization task corresponding to the token preceding the token in order is completed, if so, execute the write log operation for recording the write data operation into the log file; if not, Then continue to wait, ensuring that the write log operation recorded to the log file is strictly serially executed in accordance with the order of the received data synchronization tasks, ensuring that when data loss occurs from the database, etc., it can be conveniently based on the write data in the log file Operation records restore data from the database.

The algorithms and displays presented herein are not inherently related to any particular computer, virtual system, or other device. Various generic systems can also be used with the teachings based on this. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not specific to any particular programming language. It should be understood that various programming languages can be used to implement the content of the present invention described herein, and the above description of specific languages is for disclosing the best mode of the present invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure the understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, in order to streamline this disclosure and to facilitate an understanding of one or more of the various inventive aspects, various features of the invention are sometimes grouped together in a single embodiment, figure, or its description. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art can understand that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. Modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings) and any method or method so disclosed may be used in any combination, except that at least some of such features and/or processes or units are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will understand that although some embodiments described herein include some features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention. and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of some or all of the components in the token-based concurrent master-slave synchronization device according to the embodiment of the present invention. Full functionality. The present invention can also be implemented as an apparatus or an apparatus program (for example, a computer program and a computer program product) for performing a part or all of the methods described herein. Such a program for realizing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet site, or provided on a carrier signal, or provided in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. does not indicate any order. These words can be interpreted as names.

The present invention discloses:

A1, a token-based concurrent master-slave synchronization method, which includes:

Receive data synchronization tasks from the main database;

According to the order in which the data synchronization tasks are received, a token is set for the data synchronization tasks;

Distributing the data synchronization task set with the token to one of the multiple threads, and the thread executes the data synchronization task according to the token.

A2, according to the token-based concurrent master-slave synchronization method described in A1, wherein the distribution of the data synchronization task provided with the token to a thread in a plurality of threads further includes:

Select a thread from multiple threads according to the load balancing strategy;

Distributes data synchronization tasks with tokens set to the selected threads.

A3. The token-based concurrent master-slave synchronization method according to A1, wherein the data synchronization task includes:

A write data operation of writing data into the slave database, and a write log operation of recording the write data operation into a log file.

A4. The token-based concurrent master-slave synchronization method according to A3, wherein the data writing operation executed between threads is a parallel operation; the log writing operation executed between threads is a serial operation.

A5. According to the token-based concurrent master-slave synchronization method described in A1, wherein the execution of the data synchronization task by the thread according to the token further includes:

Judging whether the write log operation of the data synchronization task corresponding to the token preceding the token is completed; wherein, the sequence of the tokens is consistent with the order in which the data synchronization task is received;

If so, then perform a write log operation for recording the write data operation into a log file;

If not, continue to wait.

A6. The token-based concurrent master-slave synchronization method according to A1, wherein the data synchronization tasks for the same data are distributed to the same thread.

B7, a token-based concurrent master-slave synchronization device, which includes:

A receiving module adapted to receive data synchronization tasks from the master database;

The setting module is adapted to set tokens for the data synchronization tasks according to the order in which the data synchronization tasks are received;

A distribution module, suitable for distributing the data synchronization task provided with the token to a parallel processing module in a plurality of parallel processing modules;

A plurality of parallel processing modules are adapted to execute the data synchronization task according to the token.

B8. The token-based concurrent master-slave synchronization device according to B7, wherein the distribution module is further adapted to: select a parallel processing module from multiple parallel processing modules according to a load balancing strategy; Data synchronization tasks are distributed to selected parallel processing modules.

B9, according to the token-based concurrent master-slave synchronization device described in B7, wherein the data synchronization task includes:

A write data operation of writing data into the slave database, and a write log operation of recording the write data operation into a log file.

B10, according to the token-based concurrent master-slave synchronization device described in B9, wherein, the described write data operation performed between each parallel processing module is a parallel operation; the described write log operation performed between each parallel processing module is Serial operation.

B11. The token-based concurrent master-slave synchronization device according to B7, wherein the parallel processing module is further adapted to:

Judging whether the write log operation of the data synchronization task corresponding to the token preceding the token is completed; wherein, the sequence of the tokens is consistent with the order in which the data synchronization task is received;

If so, then perform a write log operation for recording the write data operation into a log file;

If not, continue to wait.

B12. The token-based concurrent master-slave synchronization device according to B7, wherein the distribution module is further adapted to: distribute data synchronization tasks for the same data to the same thread.

Claims (10)

1. A token-based concurrent master-slave synchronization method, comprising: Receive data synchronization tasks from the main database; According to the order in which the data synchronization tasks are received, a token is set for the data synchronization tasks; Distributing the data synchronization task set with the token to one of the multiple threads, and the thread executes the data synchronization task according to the token. 2. The concurrent master-slave synchronization method based on tokens according to claim 1, wherein said distributing the data synchronization task provided with tokens to a thread in a plurality of threads further comprises: Select a thread from multiple threads according to the load balancing strategy; Distributes data synchronization tasks with tokens set to the selected threads. 3. The token-based concurrent master-slave synchronization method according to claim 1, wherein the data synchronization task comprises: A write data operation of writing data into the slave database, and a write log operation of recording the write data operation into a log file. 4. the token-based concurrent master-slave synchronization method according to claim 3, wherein, the write data operation performed between each thread is a parallel operation; the described write log operation performed between each thread is a serial operate. 5. The token-based concurrent master-slave synchronization method according to claim 1, wherein said execution of the data synchronization task by the thread according to the token further comprises: Judging whether the write log operation of the data synchronization task corresponding to the token preceding the token is completed; wherein, the sequence of the tokens is consistent with the order in which the data synchronization task is received; If so, then perform a write log operation for recording the write data operation into a log file; If not, continue to wait. 6. The token-based concurrent master-slave synchronization method according to claim 1, wherein data synchronization tasks for the same data are distributed to the same thread. 7. A token-based concurrent master-slave synchronization device comprising: A receiving module adapted to receive data synchronization tasks from the master database; The setting module is adapted to set tokens for the data synchronization tasks according to the order in which the data synchronization tasks are received; A distribution module, suitable for distributing the data synchronization task provided with the token to a parallel processing module in a plurality of parallel processing modules; A plurality of parallel processing modules are adapted to execute the data synchronization task according to the token. 8. The token-based concurrent master-slave synchronization device according to claim 7, wherein the distribution module is further adapted to: select a parallel processing module from a plurality of parallel processing modules according to a load balancing strategy; The card data synchronization task is distributed to the selected parallel processing modules. 9. The token-based concurrent master-slave synchronization device according to claim 7, wherein the data synchronization task comprises: A write data operation of writing data into the slave database, and a write log operation of recording the write data operation into a log file. 10. The token-based concurrent master-slave synchronization device according to claim 9, wherein the write data operation performed between each parallel processing module is a parallel operation; the write log performed between each parallel processing module The operation is a serial operation.