CN1756108A - Master/backup system data synchronizing method - Google Patents

Abstract

A method for synchronizing data between master and backup systems, the method comprising: the master system divides the data to be synchronized into several data blocks according to the characteristics of the data to be synchronized; volume synchronization; otherwise, the main system initiates batch synchronization according to the trigger conditions (such as timing synchronization period, system reset, abnormal communication interruption of the main and standby systems, etc. that may cause data inconsistency); batch synchronization and incremental synchronization are prioritized according to the divided data blocks The data blocks with high priority are synchronized first. This method reduces the amount of unnecessary synchronization data in the data synchronization of the main and backup systems, thereby reducing the amount of data synchronization between the main and backup systems; at the same time, the method of synchronizing data with high priority improves the reliability of the system compared with the general method sex.

Description

Data Synchronization Method of Active/Standby System

technical field

The invention relates to a data synchronization method, in particular to a data synchronization method between main and standby systems.

Background technique

Data communication systems usually require the system to have high reliability, and redundancy is a basic method to improve reliability. In the redundant system, 1+1 backup or N+1 backup is generally used for the core system. Once the working main system fails, the backup system can quickly take over the work of the main system and become the main one. so that the entire system can continue to work normally. In the active/standby switchover system, many operations such as dynamic data and configuration data may be involved. To ensure that the switchover between the active and standby systems will not affect the normal operation of the system, it must involve the relationship between the active and standby systems. data synchronization problem.

In order to improve system reliability, the key systems in the system are usually protected by active and standby hot backup mechanisms. Under normal working conditions, the main system is in working condition and completes the necessary functions of the system. The backup system keeps real-time data synchronization with the main system. When the main system has problems and cannot work, the backup system can completely replace the main system in time. work to improve system reliability.

In the prior art, the data to be synchronized is compressed in the primary system, and then decompressed after being transmitted to the backup system using a sliding window protocol to ensure data synchronization between the primary and backup systems.

Can analyze its existence following deficiency from prior art:

The use of compression algorithm increases the computational complexity of the system. At the same time, the synchronization method in this scheme needs to synchronize all the data every time, regardless of whether the data to be synchronized already exists in the backup system, so when a large amount of data in the main and backup systems When they are already consistent, a lot of unnecessary synchronization data will be transmitted during periodic synchronization, thereby increasing the communication burden of the active and standby systems; at the same time, if the active and standby systems are switched during the The compression method is used to transmit, resulting in incomplete compressed data received by the backup system, which cannot be decompressed correctly. At this time, the backup system cannot protect the main system, thereby reducing the reliability of the system.

Contents of the invention

The problem solved by the present invention is to provide a method for synchronizing the data of the main and backup systems, which avoids synchronizing unnecessary synchronization data; at the same time, it can ensure the availability and completeness of the data in the backup system to the greatest extent, thereby improving the overall system efficiency. reliability.

In order to solve the above problems, the present invention provides a method for synchronizing the data of the main and backup systems, the method comprising:

A. The main system divides the data to be synchronized into several data blocks;

B. Determine whether the content of the data block has changed, and if so, initiate incremental synchronization; otherwise, initiate batch synchronization according to the trigger condition.

Wherein, the process of dividing the data to be synchronized into several data blocks specifically includes:

21), according to the data characteristics of the data to be synchronized in the system, the data category and the operating speed of the system to determine the size of the data block;

22). Divide the data to be synchronized into several data blocks according to the size of the data blocks.

Wherein, the process of the incremental synchronization specifically includes: synchronizing the data block to the backup system; ending the incremental synchronization process;

Wherein, the process of batch synchronization specifically includes:

31), the backup system uploads the data block to the main system;

32), the master system judges whether the content of the data block transmitted by the backup system is consistent with the content of the data block corresponding to the master system, if inconsistent, then synchronize the data block to the backup system, end the batch synchronization process, and continue to perform step B; if consistent, Then the batch synchronization process is ended, and step B is continued.

Wherein, the process of dividing the data to be synchronized into several data blocks further includes: selecting corresponding types of check codes for the several data blocks according to the size of the data blocks and the operation speed of the system.

Wherein, between the steps A and B, it further includes: the master and backup systems calculate their corresponding check codes according to the data of each data block.

Wherein, the process of batch synchronization specifically includes:

41), the standby system reports the calculated check code to the main system;

42), the main system judges whether the received check code is the same as the check code calculated by itself, if not, then synchronize the data block to the backup system, end the batch synchronization process, and continue to execute step B; if they are the same, then End the batch sync process and proceed to step B.

Wherein, between step A and step B includes: prioritizing the data blocks to be synchronized according to the importance of the data.

Wherein, the process of synchronizing the data block to the backup system in the process of incremental synchronization and in step 32) is specifically: synchronizing the data block to the backup system successively according to the priority of the data block from high to low system.

Wherein, the process of synchronizing the data blocks to the backup system in step 42) is specifically: synchronizing the data blocks to the backup system successively according to the priority of the data blocks from high to low.

Wherein, the trigger conditions include: cycle synchronization, system reset, and communication interruption of the master and backup systems.

Compared with the prior art, the present invention has the following advantages: due to the adoption of data blocks, the unnecessary amount of synchronous data in the data synchronization of the main and backup systems is reduced, thereby reducing the communication burden of the main and backup systems. In addition, by dividing the data into blocks and judging whether the data blocks are consistent through the check code, the unnecessary synchronization data volume in the data synchronization process of the main and backup systems is further reduced, thereby further reducing the communication burden of the main and backup systems; At the same time, by dividing the data into blocks and checking whether the data blocks are consistent with a check code and dividing the priority, the present invention improves the reliability of the system compared with the general method.

Description of drawings

Fig. 1 is the flowchart of the first embodiment of the present invention;

Fig. 2 is the flowchart of incremental synchronization in the first embodiment of the present invention;

Fig. 3 is the flow chart of batch synchronization in the first embodiment of the present invention (do not adopt verification code);

Fig. 4 is the flowchart of batch synchronization in the second embodiment of the present invention (using check code);

Fig. 5 is a schematic diagram of dividing and prioritizing data blocks in the active system and synchronizing them to the standby system;

Fig. 6 is the flowchart of incremental synchronization in the third embodiment of the present invention;

Fig. 7 is the flow chart of batch synchronization in the third embodiment of the present invention (not adopting check code and synchronizing by data block priority order);

Fig. 8 is a flow chart of batch synchronization in the fourth embodiment of the present invention (using check code and synchronizing according to the order of priority of data blocks).

Detailed ways

Data synchronization refers to the process in which the backup system in the active/standby hot backup system maintains data consistency with the active system in real time. There are two situations, one is batch synchronization and the other is incremental synchronization.

The batch synchronization refers to the way that the backup system tries to synchronize all the data of the main system. It is used for abnormal situations such as communication interruption/system reset of the main and backup systems, and is also used for periodic synchronization to ensure that the backup system and Data synchronization between primary and active systems.

The incremental synchronization refers to the process of synchronizing the changed data block content to the standby system due to its own data changes (such as changes in the configuration data of the primary system caused by the configuration command sent by the user) when the primary system is working normally. .

The method discloses an optimized data synchronization method of the master and backup systems. The method adopts an N+1 (N>=1) master and backup hot backup protection mode.

The first embodiment of the present invention will be described in detail below with reference to FIG. 1 .

In the present invention, step S101 is first performed to divide the data to be synchronized in the active system into several data blocks. The specific operation is: Step 1. Determine the size of the data block according to comprehensive factors such as the data characteristics of the data to be synchronized, the data type, and the operating speed of the system. For example, when each configuration command in the configuration data to be synchronized in the system is a fixed-length M , when dividing the configuration data block, select the integer multiple N of M as the size of the data block (M*N); if the operation speed of the CPU is fast, the data block should be appropriately determined to be larger; the type of data is also called the inter-data Coupling, when dividing data blocks, divide the data with strong coupling into the same data block, and divide the data with weak coupling into different data blocks. For example, the service configuration data is divided into one data block, the protection type data is divided into one data block, and the maintenance type data is divided into one data block. Step 2, divide the data to be synchronized into several data blocks according to the size of the data block; for example, the total amount of configuration data to be synchronized is: T, then the configuration data can be divided into (T/(M*N)+1) data blocks.

After the synchronization data is divided into data blocks, the minimum unit of each backup operation is a data block. In a system, the data is divided into several data blocks, so that most modification operations only involve one data block at a time, so only the content of one data block needs to be synchronized, and if the capacity of the divided data block is small enough , which can speed up the synchronization process.

Then, step S102 is executed, and the main system judges whether the content of the data block changes (generally, the configuration data of the main system is changed due to the configuration command sent by the user), and if so, step S103 is executed for incremental synchronization; if the content of the data block has not If there is a change, step S104 is executed, and the main system initiates batch synchronization according to trigger conditions (such as communication interruption, system reset, or periodic synchronization, etc.).

The incremental synchronization process will be described in detail below in conjunction with FIG. 2 .

First, the main system executes step S201, detects that the content of some data blocks has changed, and incremental synchronization is required, then the main system executes step S202, obtains the number of the data block that needs to be incrementally synchronized, and then executes step S203, the The corresponding data blocks are synchronized to the standby system, and then the incremental synchronization process ends.

The batch synchronization process of the present invention will be described in detail below in conjunction with FIG. 3 .

When the primary system detects that the batch synchronization conditions are met, it first executes step S301, and initiates a batch synchronization command to the backup system; after receiving the command, the backup system executes step S302, and transmits the contents of each data block on the backup system to the primary system; then executes Step S303, after the primary system receives the data blocks sent by the standby system, it judges whether the content of these data blocks is consistent with the content of the corresponding data blocks on the primary system. That is, the inconsistent data block numbers are recorded; then step S305 is executed, that is, the data blocks corresponding to these data block numbers are synchronized to the backup system, and the batch synchronization process ends.

The following is the second embodiment of the present invention.

First, the main system divides the data to be synchronized into several data blocks. The specific operation is: Step 1. Determine the size of the data block according to comprehensive factors such as the data characteristics of the data to be synchronized, the data type, and the operating speed of the system; for example, when the system Each configuration command in the configuration data to be synchronized is a fixed length M. When dividing the configuration data block, select an integer multiple N of M as the size of the data block (M*N); if the operation speed of the CPU is faster, the The size of the data block is properly determined to be larger; the data category, also known as the coupling between data, when dividing the data block, the data with strong coupling is divided into the same data block, and the data with weak coupling is divided into different in the data block. For example, the service configuration data is divided into one data block, the protection type data is divided into one data block, and the maintenance type data is divided into one data block. Step 2, divide the data to be synchronized into several data blocks according to the size of the data block; for example, the total amount of configuration data to be synchronized is: T, then the configuration data can be divided into (T/(M*N)+1) data blocks. Step 3, according to the size of the data block and the operation speed of the system, select the corresponding type of check code for several data blocks; for example, when the size of the divided data block is 1K, if the speed of the CPU is faster, you can choose MD5, etc., which can be used for digital signatures Check code type; otherwise, you can choose the check code type that is often used to ensure data consistency between the receiving end and the sending end during transmission such as CRC-16.

Secondly, the active and standby systems calculate their corresponding check codes according to the data of each data block.

Then, if the main system determines that the content of the data block has changed (usually the configuration data of the main system changes due to the configuration command issued by the user), it will perform incremental synchronization; if the content of the data block has not changed, the main system will Initiate batch synchronization according to trigger conditions (such as communication interruption, system reset or periodic synchronization, etc.).

The process of incremental synchronization is shown in FIG. 2 , and the specific description is the same as that in the first embodiment.

The data batch synchronization process using check codes in the present invention will be described in detail below in conjunction with FIG. 4 .

When the primary system detects that the batch synchronization conditions are met, it first executes step S401, that is, initiates a batch synchronization command to the backup system; after receiving the command, the backup system executes step S402, calculates the check code of each data block in the backup system, and reports it to the master system; the primary system executes step S403 to calculate the check codes of each data block on it after receiving the check code reported by the standby system; Check whether the check code is consistent with the check code reported by the backup system; if they are consistent, the batch synchronization process will end, if they are not consistent, perform step S405 to record the number of data blocks whose check codes are inconsistent; then perform step S406 to convert these The data block corresponding to the number is synchronized to the standby system, and the batch synchronization process ends.

Using the check code method can further reduce the amount of data synchronization between the primary and backup systems, but it needs to increase the amount of calculation of the primary and backup systems, which is equivalent to the strategy of exchanging time for space commonly used in software design. In specific applications, it can be determined whether to use the check code method according to the amount of data to be synchronized. If the amount of data to be synchronized is not large, it is recommended not to use this method, but to use the method in the previous embodiment.

The following is the third embodiment of the present invention, which is a more optimized method for synchronizing data between the master and backup systems.

First, the main system divides the data to be synchronized into several data blocks; the specific operation process is the same as that of the first

Example.

When multiple data blocks need to be synchronized, after dividing the data blocks, it is necessary to determine the priority of each data block and formulate the synchronization order according to the importance of the data blocks to the system operation, for example: when there are configuration data blocks, the protection In the case of data blocks and maintenance data blocks, the priority of the configuration data can be positioned the highest, the priority of the protection data is second, and the priority of the maintenance data is the lowest; as shown in Figure 5, the main system will Example of data block partitioning, and priority assignment and synchronization to the standby system. The data of business configuration has the highest priority, which is marked as data block A110; the second is protection data, which is marked as data block B120; the second is maintenance data, which is marked as data block C130, and so on, according to the priority A>B> C>D>E>... to arrange.

Then execute the next step. When the configuration data of the main system changes due to the configuration command issued by the user, the main system judges that the content of the data block has changed, and then performs incremental synchronization; if the content of the data block does not change, Then the main system will initiate batch synchronization according to trigger conditions (such as communication interruption, system reset or periodic synchronization, etc.).

The process of incremental synchronization will be described in detail below in conjunction with FIG. 6 .

First, the main system executes step S601, detects that the content of some data blocks has changed, and incremental synchronization is required, then the main system executes step S602, obtains the number of data blocks that need to be incrementally synchronized, and then executes step S603, which will The data blocks corresponding to the number are synchronized to the standby system in descending order of the priority of the data blocks, and the ones with higher priorities are synchronized first, and then the incremental synchronization process ends.

The batch synchronization process of the present invention will be described in detail below in conjunction with FIG. 7 . When the primary system detects that the batch synchronization conditions are met, it first executes step S701 to initiate a batch synchronization command to the backup system; after receiving the command, the backup system executes step S702, and the backup system transmits the contents of each data block to the primary system; then executes the steps S703. After the primary system receives the data blocks sent by the standby system, it judges whether the content of these data blocks is consistent with the content of the corresponding data blocks on the primary system. Record the inconsistent data block numbers; then execute step S705, that is, the data blocks corresponding to these data block numbers are synchronized to the backup system in sequence from high to low according to the priority order of the data blocks, and the high priority is synchronized first, and then the batch synchronization process is ended .

The following is the fourth embodiment of the present invention, which is another more optimized method for synchronizing data between master and backup systems.

First, the main system divides the data to be synchronized into several data blocks; the specific operation process is the same as that of the second

Example.

Secondly, according to the importance of the data blocks to the system operation, determine the priority of multiple data blocks and formulate the synchronization order. For example: in the case of configuration data blocks, protection data blocks, and maintenance data blocks, three data blocks can be used. The priority of the configuration data is the highest, the priority of the protection data is second, and the priority of the maintenance data is the lowest; as shown in Figure 5, it is an example of dividing a data block in the active system, and assigning priority and synchronizing to the backup system . The data of business configuration has the highest priority, which is marked as data block A110; the second is protection data, which is marked as data block B 120; the second is maintenance data, which is marked as data block C130, and so on, according to the priority A>B >C>D>E>... to arrange.

Then, the active and standby systems calculate their corresponding check codes according to the data of each data block.

If the main system judges that the content of the data block has changed (usually the configuration data of the main system will change due to the configuration command issued by the user), it will perform incremental synchronization. The specific process is the same as that described in the third embodiment; If the content of the block does not change, the main system will regularly initiate batch synchronization according to trigger conditions (such as communication interruption, system reset or periodic synchronization, etc.).

The batch synchronization process in this embodiment will be described in detail below with reference to FIG. 8 .

When the primary system detects that the batch synchronization conditions are met, it first executes step S801, that is, initiates a batch synchronization command to the backup system; after receiving the command, the backup system executes step S802, calculates the check code of each data block in the backup system, and reports it to the master system; the main system executes step S803 to calculate the check code of each data block on it after receiving the check code reported by the standby system; check code is consistent with the check code reported by the standby system; if consistent, then end the batch synchronization process, if inconsistent, execute step S805, record the number of data blocks whose check codes are inconsistent; then execute step S806, and copy these The data blocks corresponding to the numbers are synchronized to the standby system in descending order of the priority of the data blocks, and the ones with higher priorities are synchronized first, and then the batch synchronization process ends.

In the above-mentioned third and fourth embodiments, since the main system performs data synchronization according to the priority level, that is, the data with high priority is synchronized first, and the data with lower priority is synchronized later, so even if there is a problem during the synchronization process In case of accidents, such as communication interruption, unexpected suspension of synchronization, etc., the backup system can use the data that has been synchronized to complete the basic functions, which greatly increases the reliability of the system.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto, any changes or variations that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (10)

1. A method for synchronizing data between active and standby systems, comprising: A. The main system divides the data to be synchronized into several data blocks; B. Determine whether the content of the data block has changed, and if so, initiate incremental synchronization; otherwise, initiate batch synchronization according to the trigger condition. 2. According to the data synchronization method of the master and backup systems according to claim 1, it is characterized in that the process of dividing the data to be synchronized into several data blocks specifically includes: 21), according to the data characteristics of the data to be synchronized in the system, the data category and the operating speed of the system to determine the size of the data block; 22). Divide the data to be synchronized into several data blocks according to the size of the data blocks. 3. According to the data synchronization method of the master and backup systems according to claim 2, it is characterized in that: The process of the incremental synchronization specifically includes: synchronizing the data block to the backup system; ending the incremental synchronization process; The batch synchronization process specifically includes: 31), the backup system uploads the data block to the main system; 32), the master system judges whether the content of the data block transmitted by the backup system is consistent with the content of the data block corresponding to the master system, if inconsistent, then synchronize the data block to the backup system, end the batch synchronization process, and continue to perform step B; if consistent, Then the batch synchronization process is ended, and step B is continued. 4. According to the data synchronization method of the master and backup systems according to claim 2, it is characterized in that the process of dividing the data to be synchronized into several data blocks also includes: According to the size of the data block and the operation speed of the system, select the corresponding type of check code for several data blocks. 5. The method for synchronizing data between the master and backup systems according to claim 4, wherein, between steps A and B, further comprising: The master and backup systems calculate the corresponding check code based on the data of each data block. 6. The method for synchronizing data in the master and backup systems according to claim 5, wherein the batch synchronization process specifically includes: 41), the standby system reports the calculated check code to the main system; 42), the main system judges whether the received check code is the same as the check code calculated by itself, if not, then synchronize the data block to the backup system, end the batch synchronization process, and continue to execute step B; if they are the same, then End the batch sync process and proceed to step B. 7. The method for synchronizing data between the master and backup systems according to any one of claims 1 to 6, characterized in that, between the steps A and B, it includes: Prioritize the data blocks to be synchronized according to the importance of the data. 8. According to the method for synchronizing the data of the primary and secondary systems according to claim 7, it is characterized in that, in the process of incremental synchronization and in step 32), the process of synchronizing the data block to the secondary system is specifically: The data blocks are synchronized to the standby system in descending order of the priority of the data blocks. 9. According to the data synchronization method of the master and backup systems according to claim 7, it is characterized in that, in step 42), the process of synchronizing the data block to the backup system is specifically: the data block is transferred from the data block according to the priority of the data block The sequence from high to low is synchronized to the standby system. 10. The method for synchronizing data between the master and backup systems according to claim 8 or 9, wherein the trigger conditions include: cycle synchronization, system reset, and communication interruption of the master and backup systems.

Images