JP2002288027A - Distributed operation supporting system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a distributed operation supporting system capable of being easily incorporated in a conventional database system. SOLUTION: This distributed operation supporting system is provided with a master distribution server 2 constructing the sub-database 2a as a subset replica of the operation database 1a constructed in the host computer 1, and connected to the host computer in a state that they can exchange the data with one another, and a slave distribution server 3 constructing the read-only- type sub-database 3a by the snapshot of the sub-database acquired from the master distribution server. A client terminal 4 acquires the data from the sub- database of the slave distribution server, creates the desired database, and transfers the latest data to the master distribution server.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention supports the daily operations of sales representatives by giving necessary information to terminals carried by sales representatives assigned to various locations through databases of distributed servers located in various locations. In addition, the present invention relates to a distributed business support system for performing appropriate business management based on information transferred from a sales representative.

[0002]

2. Description of the Related Art Conventionally, in a distributed system of a company or an institution that operates in a wide area, a comprehensive database in which company-wide business information and the like are comprehensively managed and stored in a host computer installed in a head office or a management center. Is built, and the distributed server installed at branch offices and service centers scattered in each region is ported from the comprehensive database to the database part of customers and products related only to the branch office or service center. A sub-database has been built. In such a distributed database, it is necessary to ensure that the contents of the database at the head office and the contents of the database at the branch office are the same. For example, if the database at the head office and the database at the branch office are updated independently, there is a disadvantage that separate databases exist at the head office and the branch office.

[0003]

In order to avoid such inconveniences, a system is required in which the updated contents of the sub-database of the branch office are immediately reflected in the integrated database of the head office. To maintain data consistency between multiple databases, only the source database updates or adds data, the destination only allows browsing,
A distributed database system has been proposed in which changes made to an original database are periodically reflected in a replication source.

[0004] However, in order to construct such a distributed database system, there is no problem if each computer constructing the system has easy conversion in terms of specifications, that is, OS and file structure. For example, if the host computer is a conventional large / medium-sized computer called a mainframe, and the computer that constructs a database distributed in each region is a general-purpose OS (such as WindowNT or OS2), the structure is completely different. It has been difficult to create a program for creating the above-mentioned consistency between two computers. In view of the above situation, an object of the present invention is to provide a distributed business support system that can be easily incorporated into a conventional database system and configured.

[0005]

In order to solve the above problems, a distributed business support system according to the present invention constructs a sub-database as a subset replica of a business database constructed in a host computer, and establishes a connection with the host computer. A parent distributed server connected so that data can be exchanged, and a child distributed server that constructs a read-only sub-database by a snapshot of the sub-database obtained from the parent distributed server, the child distributed server comprising: Data from the sub-database is sent to the client terminal so that a desired database is generated in the client terminal, and the parent distributed server is configured to receive update data from the client terminal.

In this configuration, a distributed server for constructing a distributed database is exchanged with a host computer by exchanging database files, that is, by exchanging data, a parent database for constructing a sub-database required for a predetermined business, for example, a customer database. The distribution server is divided into a distributed server and a child distributed server that constructs a local read-only sub-database by using a snapshot of the sub-database constructed on the parent distributed server.

An appropriate database is selected from a large number of sub-databases constructed by a host computer composed of a conventional office computer and the like, and the file is transferred (copied) to the parent distributed server to copy it. A desired sub-database is constructed on the type distribution server. This sub-database is accessible from the client terminal, and the client terminal can acquire and update necessary data. If this sub-database is updated, if it is left as it is, it will become inconsistent with the database on the host computer, so the file containing at least the updated data of the updated sub-database will be used for updating the comprehensive database. Is transferred from the parent distributed server to the host computer.

[0008] The relationship between the database of the parent distributed server and the database of the child distributed server is based on the so-called distributed database system technology. The database of the child distributed server is constructed by snapshot replication. The client terminal accessing the server cannot update or add data, and is only allowed to refer to the data.

With the above-described configuration, a distributed database system relating to a sub-database required by a specific department can be constructed while coexisting with a comprehensive database constructed in a host computer. By using the terminal, it is possible to support the work of members belonging to the relevant department anytime and anywhere.

In one preferred embodiment according to the present invention,
The parent distributed server transfers the replacement data (file) of the sub-database to the host computer in a batch process, temporarily prohibits the transfer of the update data from the client terminal during the batch process, and executes the subsequent batch process. In some cases, transfer of update data from the client terminal is started upon completion. Accordingly, for example, by transferring at least updated files to the host computer in a batch process at night, it is possible to prevent the client from accessing the server as much as possible. In some cases, when transfer processing of update data occurs from a client terminal during transfer processing between the parent distributed server / host computer,
Since the data transfer process from the client terminal can be started immediately after the end of the transfer process between the host computers, it is possible to prevent the transfer process of the update data from the client terminal from being unnecessarily delayed.

Further, in one preferred embodiment according to the present invention, before transferring the update data from the client terminal to the parent distributed server, an inquiry is made as to whether or not the key data of the sub-database has been changed. In this case, the update data is transferred after the key data change processing.
In this configuration, before the update data is transferred from the client terminal side, the presence or absence of the change is checked before the update data is transferred from the client terminal in case the key data such as the customer number is changed on the host computer or the parent distributed server side. If there is, the data before the change is replaced with the data after the change and then transferred. As a result, the data change is reflected in the client terminal in a timely manner, and the synchronization in the distributed database system becomes more reliable.

As one of suitable measures for timely reflection of such data change at the client terminal, in the present invention, the transfer of the key data change file from the parent distributed server to the child distributed server is performed in a batch mode. It has been proposed that the client terminal accesses the key data change file to make an inquiry about whether or not the key data has been changed. Other features and advantages according to the present invention will become apparent from the following description of embodiments with reference to the drawings.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of a distributed business support system according to the present invention is schematically shown in FIG.
In this system, an office computer installed at the head office builds a comprehensive database 1a as a host computer 1, and a server computer (hereinafter simply referred to as a server) installed at the head office and connected to the host computer 1 via an in-house LAN. Customer database 2 which is a sub-database of the comprehensive database 1a
a. The server 3 that builds the customer database 3a is also installed on the branch side. The branch-side customer database 3a is a copy of the head office-side customer database 2a, that is, the server 2 functions as a parent distributed server, and the server 3 functions as a child distributed server.
Customer databases 2a, 3a of these distributed servers 2, 3
A mobile terminal is assigned to each sales representative as a client terminal 4 that can access a. Generally, a mobile computer or a PDA is adopted as a portable terminal, but a general-purpose desktop personal computer may be used as the client terminal 4.

The parent distributed server 2 and the child distributed server 3 are connected through a communication line 6 via a router 5, and a snapshot of the customer database 2 a is copied from the parent distributed server 2 to the child distributed server every morning. Then, a read-only customer database 3a is also constructed on the branch side. Of course, the transfer of the snapshot may be performed a plurality of times at an appropriate time of day. Mobile terminal 4 as client terminal
Can be connected to a communication line 6 such as the Internet via a communication interface 7 such as a modem, and can access the parent distributed server 2 and the child distributed server 3 from anywhere at any time.
The customer information can be obtained from a. Needless to say, the desktop client terminal 4 installed in the head office or the branch is similarly connected to the parent distributed server 2 or the child distributed server 3 via an in-house LAN, thereby obtaining customer information from the respective customer databases 2a and 3a. Can be obtained.

The client terminal 4 is equipped with spreadsheet software or database software functioning as a front end of the customer database system, or a WWW browser, so that the latest customer information can be referenced and browsed immediately. Regarding addition / update information of data generated on the client terminal side, since the customer database 3a of the child distributed server 3 is of a read-only type,
The data is transferred to the parent distributed server 2, and the parent distributed server 2 performs processing on the customer database 2a based on the added / updated information.

As described above, based on the addition / update information transferred from the client terminal 4, the parent distributed server 2
The changed contents of the customer database 2a processed by the above must be reflected in the comprehensive database 1a of the host computer 1. The necessary data transfer between the host computer 1 and the parent distributed server 2 is performed as a batch process at night. Next, a flow in which a change in the contents of the customer database by the client terminal 4 is reflected in the comprehensive database 1a of the host computer 1 through the parent distributed server 2 will be described with reference to the flowchart in FIG.

First, addition / update of customer data generated through the sales work of the sales staff is generated as update / additional information in the mobile terminal 4 through input work to the mobile terminal (client terminal) 4 by the sales staff. (# 1
0).

Since the update / additional information generated in the portable terminal 4 must be reflected in the customer database 2a, preparation for transferring the update / additional information from the portable terminal 4 to the parent distributed server 2 is performed. (# 20). The preparation for transferring the update / additional information will be described later in detail.

When the transfer preparation is completed, the updated / added information is transmitted from the portable terminal 4 to the parent distributed server 2 via the portable telephone line, the gateway, the Internet, the router 5, and the communication line 6 composed of the company LAN. (# 30).

Based on the update / additional information received from the portable terminal 4, the parent distributed server 2 changes the contents of the customer database 3a (# 40), and changes the customer database 3a and the general database 1a of the host computer 1 side. Host computer 1 to maintain the integrity of
A file for update is created by converting the update information to be transferred to the file (# 50).

Parent distributed server 2 / host computer 1
Since the file transfer between them is batch-processed under negotiation at night, the file transfer waits until a predetermined time and the permission of the file transfer are obtained (# 60).
This file transfer negotiation will be described in detail later. When the batch processing time comes and the negotiation is established, the file transfer from the parent distributed server 2 to the host computer 1 is performed (# 70). The host computer 1 updates the customer database which is a part of the comprehensive database using the acquired update file (# 8)
0). The step (# 20 in FIG. 2) for preparing the transfer of the update / additional information from the portable terminal 4 to the parent distributed server 2 will be described in detail with reference to FIG. If the customer information loaded from the customer database 2a or 3a needs to be changed or newly added, the update / additional information to be transferred includes, for example, a physical table name to be updated, all field names to be updated, Update / additional data including operation mode designation such as change / deletion, record value, etc. is created (# 21). Before transferring the updated / added data to the parent distributed server 2, the key data change file created in the parent distributed server 2 or transferred from the parent distributed server 2 and stored in the child distributed server 3. The key data change file that is present is obtained from the server 2 or 3 (# 22),
It is checked whether or not the update / additional data to be transferred includes key data that has already been changed on the server side (# 23).

When the updated / added data already contains key data changed on the server side (Ye in # 23)
(s branch), and it is determined whether or not to transfer the update / additional information to the parent distributed server 2 in order to avoid unauthorized update processing such as double update (# 24). This determination is made based on the time stamp tag of the key data change process, for example, the date. When the time stamp tag is not included in both the server side and the terminal side, when the type stamp tag is included only in the server side, and when the same time stamp tag is included in both, transfer is possible. Otherwise, transfer is disabled, and transfer cancel processing is performed (# 2
8).

If it is determined that transfer is possible (# 24)
In the case of (Yes branch), the key data to be changed is rewritten in the database (list) created on the mobile terminal 4 side based on the key data change file (# 2).
4), update / additional data having correct key data is created again (# 25). When the update / additional data does not include the key data already changed on the server side (No branch in # 23) or when the update / additional data based on the key data change file is re-created (# 25) Then, a preparation is made to transfer the updated / added data to the parent distributed server 2 (# 26). When a data transfer session is established between the portable terminal 4 and the parent distributed server 2, data is transferred (# 30).

Next, negotiation of file transfer between the parent distributed server 2 and the host computer 1 as a batch process performed at night will be described with reference to FIG. First, it is checked whether it is a preset batch processing time (# 61), and when the time comes, negotiation is started. The negotiation starts when the host computer 1 issues a transfer request of the update file to the parent distributed server 2 (# 6).
2). Note that this file transfer may be performed at an arbitrary time, for example, by a manual command input, instead of the above-described regular processing performed at a predetermined time.

In response to the transfer request from the host computer 1, the parent distributed server 2 makes the following four responses: "BUSY", "OK", "ALREADY", and "NOTREADY". "BUSY" indicates that the parent distributed server 2 is currently acquiring data from the portable terminal 4 and does not accept the transfer request from the host computer 1, and the host computer issues a transfer request again after the waiting time. Must be done (# 63). "OK" indicates that the parent distributed server 2 has accepted the transfer request from the host computer 1, and
The update file is copied to the transfer folder. Further, during the copying, the session from the portable terminal 4 is blocked (# 64). “ALREADY” indicates that the update file has already been copied to the transfer folder, and prompts the host computer 1 to receive the transfer (# 6).
5). “NOTREADY” indicates that there is no update file this time, and waits until the next batch processing (# 6)
6). In either case of "OK" or "ALREADY", the file transfer process is subsequently executed (# 70).
This file transfer process is performed according to a well-known procedure of the FTP system.

In the above embodiment, the distributed servers 2 and 3
Although the sub-database constructed in the above was the customer database, of course, this may be any other database such as a database relating to schedule management, business instruction confirmation, delivery management, sales management, and the like, and a database combining these. Good. The child distributed server 3 is normally deployed in each area, and the portable terminal 4 can access any of the distributed servers 2 or 3 by roaming processing. Although the portable terminal is mainly described as the client terminal 4, a desktop personal computer, other personal computer, or a dedicated machine that is not carried around can also be used as the client terminal 4.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing one embodiment of a distributed business support system according to the present invention.

FIG. 2 is a flowchart showing that a change in the contents of a customer database by a client terminal is reflected in a comprehensive database of a host computer through a parent distributed server;

FIG. 3 is a flowchart showing preparation for transfer of update / additional information from a portable terminal to a parent distributed server;

FIG. 4 is a flowchart showing negotiation of file transfer between a parent distributed server / host computer.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Host computer 1a Comprehensive database 2 Parent distributed server 2a Customer database (sub-database) 3 Child distributed server 3a Read-only customer database (sub-database) 4 Client terminal (portable terminal) 5 Router 6 Communication line 7 Modem

Claims (4)

[Claims] A sub-database is constructed as a subset replica of a business database constructed in a host computer, and is acquired from the parent distributed server connected to the host computer in a data exchangeable manner. And a child distributed server that constructs a read-only sub-database by using a snapshot of the sub-database. The child distributed server is configured to store the sub-database in the client terminal so that a desired database is generated in the client terminal. 2. The distributed business support system according to claim 1, wherein the parent distributed server receives update data from the client terminal. 2. The parent distributed server transfers batch data of the sub-database to the host computer by batch processing, and temporarily prohibits transfer of update data from the client terminal during the batch processing. 2. The distributed business support system according to claim 1, wherein the transfer of the update data from the client terminal is started upon completion of the batch processing. 3. An inquiry as to whether or not key data of the sub-database has been changed before transfer of update data from the client terminal to the parent distributed server, and if key data has been changed, after key data change processing. 3. The distributed business support system according to claim 1, wherein update data is transferred. 4. The key data change file is transferred from the parent distributed server to the child distributed server in batches, and the client terminal accesses the key data change file to change the key data. 4. The distributed business support system according to claim 3, wherein an inquiry about presence or absence is made.