JP2007179122A - Data delivery system and data delivery method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique that implements efficient database synchronization without adding new equipment to a data delivery system and requiring high load processing. <P>SOLUTION: A maintenance server 10 and one or more operating servers 20 all networked are provided with respective databases, and data is delivered from the database 11 of the maintenance server to the databases 24 of the operating servers. The maintenance server 10 comprises an output means 12 for outputting an export file 100 storing a data region of the database and a command file 101 for restoring it, and a transferring means 13 for transferring each file to the operating servers. The operating servers 20 each comprise a receiving means 21 for receiving each file, an analyzing means 23 for analyzing the command file 201 to acquire database restoration information, and an updating means 22 for updating the database 24 with the export file 200 according to the analysis result. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a data distribution system and a data distribution method for distributing data between servers provided with a database, and particularly to a technique for efficiently synchronizing the contents of a database.

  The conventional database synchronization adjustment between servers performs direct database connection from the maintenance server to the database of each of the plurality of operation servers from the maintenance server, and the database of the database is updated by the SQL of the maintenance server CPU. It is a mechanism to perform synchronous adjustment. For this reason, in an environment where there are a large number of operation servers, a corresponding processing time is required for the synchronization adjustment of the database, and since the database management becomes complicated, a technique for automating the management is known.

  For example, Patent Document 1 describes a database synchronization adjustment technique for a transmitting station host and a plurality of receiving station hosts connected to a network. This is a server configuration that has a hierarchical structure in which a transmitting station and a receiving station are interposed between a maintenance server that is a transmission server and an operation server, thereby distributing processing and reducing the processing load on each server. It is.

  Such a conventional method is a method in which the CPU of a maintenance server that is a transmission server is actively used to perform synchronization adjustment with databases in a plurality of operation servers that are reception servers by direct connection of the databases. In this method, most of the processing related to database synchronization adjustment is handled by the maintenance server, which is the transmission server. Therefore, the database synchronization adjustment processing is proportional to the performance of the transmission server and the number of operation servers. In order to reduce the processing time, the transmission server needs to have a high-performance server arrangement that takes into account the performance required for the original business maintenance and the performance required for database synchronization adjustment.

  Another method is to add a dedicated server for database synchronization adjustment, but this is intended for database synchronization adjustment in addition to the business server that should be prepared for the systemized business. The problem is that it leads to the arrangement of servers and requires extra running costs.

  Although the technique described in Patent Document 1 is a technique for improving the efficiency of synchronization adjustment of databases connected on a network, the method of arranging a relay server called a station increases the transmission side server or handles transmission. Since this is a technique for increasing the number of servers, it is not a technique for reducing running costs due to the increase in the number of servers installed.

JP 2000-105744 A

  The present invention was created in view of the above-described problems of the prior art, and efficiently adjusts the synchronization of a database without adding a new device to the data distribution system or requiring high-load processing. The purpose is to provide technology that realizes.

In order to achieve the above object, the present invention provides the following means.
That is, according to the first aspect of the present invention, a maintenance server and at least one operation server connected via a network each have a database, and data is transferred from the maintenance server database to the operation server database. A data distribution system capable of distributing

  And an output processing means for outputting an export file storing the data portion of the database and a command file storing at least restoration information necessary for restoring to the original database when the export file is read. Transfer processing means for transferring both the export file and the command file to the operation server.

On the other hand, the operation server includes an acceptance processing unit that receives the export file and the command file from the network, an analysis processing unit that analyzes the received command file and obtains restoration information of the database, and an analysis by the analysis processing unit Update processing means for updating the database of the operation server using the received export file according to the result.
By adopting this configuration, the data distribution system of the present invention performs database distribution update processing.

The invention according to claim 2 is characterized in that in the data distribution system, the operation server generates log data related to an update processing result by the update processing unit and transmits the log data to the maintenance server via a network. On the other hand, when the maintenance server receives the log data and at least the update processing result is a normal end, the maintenance server includes an end processing means for storing in the distribution server definition table that the update processing for the operation server has been completed.
Further, the transfer processing means refers to the distribution server definition table and performs the transfer processing on the operation server that has not been updated.

  The present invention can also be provided as a data distribution method. That is, according to the third aspect of the present invention, a maintenance server and at least one operation server connected via a network each have a database, and data is transferred from the maintenance server database to the operation server database. This is a data distribution method in a distributable data distribution system.

  In the method, the output processing means of the maintenance server includes an export file storing the data portion of the database, and a command file storing at least restoration information necessary for restoring the original database when the export file is read. An output processing step for outputting, a transfer processing unit for the maintenance server, a transfer processing step for transferring both the export file and the command file to the operation server, and a reception processing unit for the operation server, which transmits the export file and the command file from the network. A reception processing step of receiving a command file;

Further, the analysis processing means of the operation server analyzes the received command file and acquires the restoration information of the database, and the update processing means of the operation server performs the reception according to the analysis result of the analysis processing process. An update processing step of updating the database of the operation server using the exported file.
The database distribution update process is performed by the above steps.

  According to a fourth aspect of the present invention, there is provided a data distribution method comprising: a reply processing step in which the reply processing means of the operation server generates log data related to the update processing result after the update processing step and transmits the log data to the maintenance server via the network. And the maintenance server end processing means receives the log data and, at least if the update processing result is normal end, stores in the distribution server definition table that the update processing for the operation server has been completed. An end processing step, and the transfer processing step refers to the distribution server definition table and performs the transfer processing on an operation server that has not been updated.

  According to the first aspect of the present invention, it is possible to perform synchronization adjustment of a database by using a backup function provided in a normal database. This realizes data transfer without using a new device such as a server. This method reduces the load on each server and contributes to high-speed processing.

  According to the second aspect of the present invention, the completion result can be confirmed on the maintenance server side. In addition, database synchronization adjustment is performed when necessary, which contributes to traffic reduction and processing reduction.

As described above, according to the present invention, after the operator sets the command file and the delivery destination server, the delivery process is started, thereby performing the database synchronization adjustment as a series of automatic operations, and the completion result confirmation is also returned to the maintenance server. It is a mechanism that uses log files. Since the acquisition process that requires processing time is performed on each operation server side, it is possible to realize that each operation server that has completed transfer can execute the acquisition process in parallel while the maintenance server sequentially performs distribution processing. As a whole process, load distribution and processing time can be shortened.
Note that the present invention is a mechanism that implements a series of automatic operations for backup processing and restoration processing that are also required in the operation of business servers, so that it does not exceed the server performance required for business operations, and database synchronization This is a technique that achieves efficiency without requiring an additional server for coordination and management.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows a system configuration relating to the present invention.
This system is composed of a transmission server 10 that is a maintenance server, a reception server 20 that is at least one operation server, and a network connecting them, and the transmission server 10 includes a maintenance database 11. The transmission server 10 is formed by a known personal computer or the like, and includes a CPU that performs various processes as well known, a storage means such as a memory and a hard disk, and an input means such as a keyboard and a mouse.
The same applies to the receiving server.

The transmission server 10 includes a maintenance database 11 and an output processing unit 12 as an output processing unit, a transfer processing unit 13 as a transfer processing unit, and an end processing unit 14 as an end processing unit. This is realized by the cooperative action of the CPU and the like.
The maintenance database 11 may be configured to receive direct access from an external user, but is usually a master database serving as a basis for synchronously adjusting the database of the operation server database (operation database 24).

  The receiving server 20 includes a reception processing unit 21 that is a reception processing unit, an update processing unit 22 that is an update processing unit, an analysis processing unit 23 that is an analysis processing unit, and an answer processing unit 25 that is an answer processing unit. Department and an operation database 24. The operation database 24 has a function of receiving data from an external terminal device and providing data.

  The processing flow of each processing unit described above is as shown in FIG. 2. First, when the distribution processing start button is pressed (S1) in the transmission server, the processing of the distribution processing mechanism C1 starts. That is, the process of the transfer processing unit 13 is repeated as long as there is a receiving server 20 that is the next distribution destination in the branch process of the next distribution determination S2 through the process by the output processing unit 12 and the process by the transfer processing unit 13.

  Further, when there is a transmission from the transfer processing unit 13 of the distribution processing mechanism C1, the reception processing unit 21 of the receiving server 20 accepts this. Then, each process of the capture processing mechanism C2 is executed. That is, the processing by the analysis processing unit 23 and the processing by the update processing unit 22. Finally, the answer processing unit 25 transmits the result to the end processing unit (14) of the distribution processing mechanism C1. After the above processes are completed, the process is completed by pressing a distribution completion button.

  First, the transmission server 10 is provided with a start button for starting the distribution process and a completion button for completing the distribution process as described above, and these buttons are displayed on the monitor provided in the transmission server 10 and pressed with a mouse or the like. Configure.

When the start button is pressed, the output processing unit 12 generates the export file 100 with the contents of the database by DBMS (DataBase Management System). At this time, a command file 101 necessary for restoring from the export file 100 to the original database is also generated.
Then, the transfer processing unit 13 sends the export file 100 and the command file 101 to the receiving server 20 via the network.

A processing flowchart of the transfer processing unit 13 is shown in FIG. First, the transfer processing unit 13 reads the distribution server definition file 103 (S10), and when there is no data in the definition file (all the updated flags are 1), the process returns to idling (S14) for waiting for an answer, and there is data ( When there is an update flag of 0 (S11), the transfer target file is distributed (S12).
At that time, file transfer is performed toward the IP address of the receiving server 20. In the reception server 20, the reception processing unit 21 receives (S13).

  On the other hand, as will be described later, the update processing result is received as the log file 102 from the answer processing unit 25 of the receiving server 20, and the end processing unit 14 performs the processing in the lower part of FIG. That is, the distribution server definition file 103 is updated in the data update / read processing (S15) of the distribution server table, and when there are a plurality of receiving servers, idling (S17) is performed until there are no unanswered ones (S16). When there are no more items, an end message is displayed (S18) and the process ends.

Here, FIG. 6 shows the configuration of the distribution server definition file 103. The file 103 includes a transfer target file indicating the order of the receiving server of the receiving server (distribution order), the name of the distribution server (name), the IP address of the distribution server (distribution IP address), and the name of the file to be transferred (export file). A command file indicating the name of the file to be transferred (command file), and a delivery completion flag are stored.
The distribution server definition file 103 is taken into the distribution processing mechanism and used by the transfer processing unit 13.

The command file 201 received by the reception processing unit 21 of the reception server 20 is analyzed by the analysis processing unit 23. FIG. 4 shows a process flow diagram for this process.
That is, when the reception processing unit 21 receives, the analysis processing unit 23 is activated and the command file 201 is opened (S20).
As long as there is an unprocessed command (S21), the first processing section in the unprocessed processing order is registered in the memory (S22).

Further, the processing command execution path is registered in the memory (S23), and the log file name is registered in the memory (S24). When this is completed, the update processing unit (22) is activated. Thereafter, the presence / absence of an unprocessed command is confirmed again, and the processes of (S21) to (S25) are repeated as long as there are.
When there is no unprocessed command, the response processing unit 25 is activated (S26), and the log file 202 is transferred to the transmission server 10 (S27).

On the other hand, the update processing unit 22 activated by the activation of the update processing unit (S25) performs processing according to the processing flowchart shown in FIG.
First, the execution command is read (S30), and the processing divisions are confirmed in order. First, it is confirmed whether it is backup (S31), truncation (S32), import (S33), or return (S34).

Here, backup among the processing categories is processing in which DB information (object definition, table data, etc.) is extracted into a dump file by the DBMS standard tool “export” and used as a backup file.
The truncation is a high-speed deletion process for all data in the table, which is executed without acquiring the processing cancellation hold information (rollback) by the DBMS standard tool “Truncate”.
The import is a process of migrating information to the DB using the information of the dump file by the DBMS standard tool “import”.
Return is a process of responding that the above three processes have been completed.

When the processing category is backup, the operation DB 24 of the receiving server 20 is read and a predetermined application is executed (S35), and the backup file 203 and the log file 202 are generated on the hard disk.
When the processing category is truncation, a predetermined application is executed (S36), and the data in the operation DB 24 is deleted and the log file 202 is generated.
When the processing category is import, a predetermined application is executed (S37), the database export 200 is read, the data is saved (updated) in the operation DB 24, and the log file 202 is generated.

When the processing category is return, a predetermined application is executed (S38), an unprocessed no response (S39) is obtained, and a response is made to the analysis processing unit (S40).
In addition, after executing a predetermined application in all the processing sections, a response is similarly made to the analysis processing unit (S40).

The command file 201 has a structure as shown in FIG. That is, the processing classification is defined in accordance with the processing order, and processing commands (specifying a predetermined application) at that time, a parameter path at the time of execution, and a log file name to be generated are stored.
The command file 201 is used by the analysis processing unit 23 of the import processing mechanism, and the analysis result is processed by the update processing unit 22.

  After the log file transfer from all the distributed receiving servers is completed, the distribution processing completion is displayed as an end message to the operator. The processing is terminated when the operator presses the distribution processing completion button.

  The present invention is a technology for easily and quickly realizing data synchronization between servers connected via the Internet and the like, and as a synchronization method of various databases such as unit price / reference database synchronization and name list database synchronization that are referred to in an integration system. Can be used.

It is a system configuration diagram of the present invention. It is a flowchart of the process of each server in this invention. It is a flowchart which concerns on the delivery process by this invention. It is a flowchart which concerns on an answer process from the analysis process by this invention. It is a flowchart which concerns on the update process by this invention. It is explanatory drawing which shows the outline | summary of the delivery server definition table by this invention. It is explanatory drawing which shows the outline | summary of the command file by this invention.

Explanation of symbols

10: transmission server, 11: maintenance database, 12: output processing unit, 13: transfer processing unit, 14: end processing unit, 20: receiving server, 21: reception processing unit, 22: update processing unit, 23: analysis processing unit 24: Operation database 25: Answer processing unit 100: Export file 101: Command file 102: Log file 103: Distribution server definition file 200: Export file 201: Command file 202: Log file

Claims (4)

A maintenance server connected via a network and at least one operation server each have a database, and a data distribution system capable of distributing data from the maintenance server database to the operation server database,
The maintenance server
Output processing means for outputting an export file storing the data portion of the database, and a command file storing at least restoration information necessary for restoring the original database when the export file is read;
A transfer processing means for transferring both the export file and the command file to the operation server,
The operation server is
Reception processing means for receiving the export file and the command file from the network;
Analysis processing means for analyzing the received command file and obtaining restoration information of the database;
A data distribution system comprising: update processing means for updating the database of the operation server using the received export file according to the analysis result of the analysis processing means, and performing database distribution update processing. In the data distribution system, the operation server includes:
While providing log processing data related to the update processing result by the update processing means and sending to the maintenance server via a network,
The maintenance server is
The log data is received, and at least when the update processing result is a normal end, the log server includes an end processing means for storing in the distribution server definition table that the update processing for the operation server has been completed,
The data distribution system according to claim 1, wherein the transfer processing unit performs the transfer process with respect to an operation server that has not been updated by referring to the distribution server definition table. A data distribution method in a data distribution system in which a maintenance server and at least one operation server connected via a network each have a database, and data can be distributed from the database of the maintenance server to the database of the operation server Because
Output for outputting an export file storing the data portion of the database and a command file storing at least restoration information necessary for restoring to the original database when the export file is read by the output processing means of the maintenance server Processing steps,
A transfer processing step in which the transfer processing means of the maintenance server transfers both the export file and the command file to the operation server;
A reception processing step in which the reception processing means of the operation server receives the export file and the command file from the network;
An analysis processing step in which the analysis processing means of the operational server analyzes the received command file and acquires restoration information of the database;
In accordance with the analysis result of the analysis processing step, the update processing means of the operation server has an update processing step of updating the operation server database using the received export file, and performs database distribution update processing. A characteristic data distribution method. In the data distribution method,
The response processing means of the operation server generates the log data related to the update processing result after the update processing step, and transmits the log data to the maintenance server via the network,
An end processing step in which the end processing means of the maintenance server receives the log data and stores in the distribution server definition table that the update processing for the operation server has been completed at least when the update processing result is normal end Have
The data distribution method according to claim 3, wherein, in the transfer processing step, the transfer processing is performed for an operation server that has not been updated by referring to the distribution server definition table.

Images