JP2710689B2 - Temporary center method in distributed database system - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a temporary center system in a distributed database system including a plurality of terminals connected to a network.

In recent years, a distributed processing system or a distributed database system (hereinafter, referred to as a distributed database system) has been constructed by a plurality of terminals connected to a network such as a public network or a private network. In such a system, the databases provided in each terminal are distributed in several places connected by a real-time network (communication network), and the processing capacity for handling the databases is also distributed.

Originally, a distributed database system stores most of the necessary data in its own database, but if the required data is on another terminal, it must receive the data via the network, which takes time to transfer. Improvement of such points is desired.

(Background technology)

In recent years, a database system has been constructed by a plurality of terminals having information processing functions connected to a network. Conventional database systems can be classified into a centralized type and a distributed type from the viewpoint of management.
The centralized database is a method in which all databases are centrally managed by one center host, and other terminals access the center host to search for necessary information.

This method has the disadvantage that the scale of the center host becomes large and accesses from other terminals are concentrated, so that the load on the center host becomes excessive. As a method for coping with this, a distributed database system in which a plurality of terminals connected to a network are provided with a database managed by each terminal is used.

A conventional distributed database system will be described with reference to FIG.

In the figure, T1 to T4 are terminals (T1 to T4) each having an information processing mechanism (CPU, memory, input / output device, file device, etc.).
To T4), F represents a file for storing a database provided in each terminal, and NW represents a network (public network, private network, etc.).

A in FIG. 16 shows the basic configuration of a distributed database system, and a plurality of terminals T1 to T4 have most of the data required for each in their own database. However, when it is located at another terminal (when it is known that there is related data elsewhere), the other terminals are sequentially inquired via the network to request data. For example, when the terminal T1 requests data from another terminal T2, the terminal T2 takes out the corresponding data from the file device, if any, transfers it via the network, and receives it at the terminal T1.

In the conventional distributed database system shown in A of Fig. 16 above, if each terminal does not hold all the data of the other terminals in its own database, it first searches the database of its own terminal and If there is no data, it is necessary to sequentially inquire other terminals via the network whether there is necessary data, and if so, it is necessary to receive the data from the terminal via the network. There is a problem that the load on the device becomes excessive.

Further, in the case of a system in which each terminal always retains all data, it is necessary to transmit the data to each terminal every time data is generated, and there is a problem that a load on the network is imposed in the same manner as described above.

Further, when the data transmitted to each terminal is revised, the revised data (revision data) must be transmitted to each terminal (this is called a missing operation).

As described above, in a distributed database system, each terminal operates a great amount of time to search and collect necessary data, and the use time of the network increases, and the load on the terminal and the load on the network are proportional to the number of terminals. There was a problem that it became huge.

On the other hand, instead of transferring data directly, there is a command transfer method in which data is expressed by a compressed command (code) to reduce the amount of information at the time of transfer. The structure is shown in Fig.
Shown in

Explaining the method, for example, at the terminal T2, "VOL1" is assigned as a compressed command for certain data. It is assumed that the terminal T1 requests the terminal T2 to transfer the data.

In this case, the terminal T2 transfers the assigned command “VOL1” and the number information “T2” indicating the terminal having this data as the additional information to the terminal T1, without transferring the data. When the terminal T1 receives the command, the terminal T1 searches its own database for data corresponding to the command.

At that time, if there is no data corresponding to the command in the own database, the corresponding terminal T2 is accessed by using the terminal number information "T2" added to the command, and the data corresponding to the command "VOL1" is obtained. Request. Thereby, target data can be obtained.

Similarly, when a command "VOL2" and "T3" indicating the location of data are sent from terminal T3 in response to a data request from terminal T1, the data can be obtained by the same procedure as described above.

In a database, a data group consisting of a plurality of related data or an integrated plurality of data may be processed in association with each other. In the database using the command shown in Fig. 16B above, is there any related data? There is a problem that it takes time and load on the network and the terminal to sequentially inquire each terminal as to whether or not to collect the data and perform the editing process.

In a database system, when there is a request to browse database data or a request to obtain data, which terminal (or individual)
Except for those that can be disclosed to the public, the eligibility of browsing is checked. In the conventional method, in each terminal, in response to a browsing request (data request) from the user (or the terminal), it is checked whether the user has a valid qualification. This check is usually a personal ID
(Identification number), a password, a keyboard, and the like. That is, a method is used in which a personal ID or the like registered in advance as a qualified person and a personal ID or the like input by the user are referred to, and if the two match, the browsing is permitted. However, according to this conventional method, as long as a valid personal ID or the like is input, the database can be browsed without requiring other qualifications.

For this reason, there is a problem that all information can be viewed from any personal ID regardless of the importance of the data. This problem is particularly acute in distributed database systems connected to public networks. That is, it is desirable that data other than the data to be kept secret (non-public data) can be viewed (opened) from any terminal. As a result, data can be viewed between terminals that are physically separated from each other, and processing using the data can be performed. Therefore, it is possible to fully utilize the features of being connected to the public network and being a distributed database system. . However, on the other hand, it is data that should be made available for browsing between different terminals (should be open data), but it cannot meet the requirement to restrict it to only those who have viewing qualifications such as personal IDs. That is, conventionally, there is a problem that the opening of the database is hindered because the viewing qualifications cannot be restricted according to the importance of the data.

The present invention eliminates the need to transmit all data via a base work to each terminal holding each database, and provides a temporary center system in a distributed database system that can efficiently obtain the latest data and reduce the network load. Its primary purpose is to provide.

Another object of the present invention is to provide a distributed database system in which a center having management information manages revised data and provides appropriate version number information to users.

It is another object of the present invention to enable a terminal connected to the center of a distributed database system to process data disclosed by its own terminal and data disclosed by another terminal separately.

The present invention further provides that, when a command is transferred in a distributed database system, the data corresponding to the required command can be obtained without asking each terminal which data is stored in the database of the terminal. An object of the present invention is to provide a temporary center system of a distributed database system that can efficiently obtain data having the same and related data with the same command.

A further object of the present invention is to provide a management system in a distributed database system which can treat related data dispersed and scattered as a single data when command transfer is used in the distributed database system.

In addition, the Center, which manages the security of data in a distributed database system, provides a provisional center system for a distributed database system that can restrict browsing according to the importance of published data and the qualifications of individuals. The purpose of. It is another object of the present invention to use a database by grouping terminals connected to a network of a distributed database system and to provide a temporary center system in a distributed database system for using data across groups. With the goal.

Still another object of the present invention is to provide a temporary center system in a distributed database system capable of distributing a load of a center function for holding and processing management information of the distributed database system and preventing danger at the time of failure. The purpose is to:

[Disclosure of the Invention]

The present invention provides a distributed database system comprising a plurality of terminals connected to a network, assigns one of the terminals as a table of contents management center for managing the entire table of contents of the published data, and provides a data providing terminal for publishing data. The main body of the data is stored in its own file and the table of contents and the terminal identification number are notified to the table of contents management center. The table of contents management center registers the table of contents notified from each data providing terminal, and sends the registered table of contents to the user terminal when there is a table of contents request from the user terminal. The user terminal can detect the required table of contents by looking at the table of contents and access the data providing terminal to obtain necessary data from the data providing terminal. Further, when a modified table of contents of the database published from the data providing terminal is registered, the table of contents management center manages the version number, and notifies the user terminal of an appropriate version number in response to the table of contents disclosure request. In addition, a terminal that requests a table of contents to the table of contents management center checks that the table of contents of its own published data is not entered from among all open tables of contents sent from the table of contents management center, and fetches only a table of contents published by other terminals. To do.

Further, according to the present invention, one of terminals of a distributed database system that performs command transfer is assigned as a command management center that manages commands, and terminal identification information that holds data corresponding to commands is registered in the command management center. The registration information is provided to the terminal upon request.
Further, the present invention allocates a common command to a related data group held in a database of one or more terminals, manages the command by a command management center, and allows a terminal to access a data group corresponding to the common command. make it easier.

The present invention further includes a command management center that matches and manages a plurality of different commands when performing command transfer in a distributed database system, and handles command matching at the same time in data processing corresponding to one command in a terminal. Data to be processed.

According to the present invention, the first rank is registered in an item such as a table of contents of the data according to the importance of the data opened in the center of the table of contents management and the command management, and identification information of a terminal (individual) which can be browsed and the corresponding information are stored. Information including the second rank is registered for the terminal (individual), and when there is a request from the terminal, the terminal (individual) permitted to browse is checked by checking the terminal and checking whether the rank satisfies the condition. Perform data security management for a limited amount.

Further, the present invention provides a distributed database system in which a group is composed of a plurality of terminals connected to a network, and a management center is provided within the group to perform data interaction, and data is exchanged between different groups. Use is performed by the inter-group management center.

According to the present invention, as the management center provided in the distributed database system, a center having various functions can be allocated according to each of the above-mentioned inventions. Is notified to other terminals, and necessary management information is collected from the other terminals so as to operate as a center, thereby distributing functions and coping with a center failure.

[Brief description of drawings]

1 is a basic configuration of an embodiment of the present invention, FIG. 2A is an embodiment configuration using a table of contents management center, FIG. 2B is an example of an operation sequence thereof, and FIGS. 3A to 3C are processing of version number management. Example of flow and file configuration, FIG. 4 is a diagram showing the interaction between a terminal for performing publisher identification and the center, FIG. 5 is an embodiment of a terminal-specific embodiment, and FIG. 6 is an implementation of a method using a command. FIG. 7 is a system configuration of an embodiment showing details of operation, FIG. 8 is a basic configuration of an embodiment of a real data management process using command matching, and FIG. 9A is a specific diagram of a real data management process. 9B to 9G are diagrams for explaining processing contents of the embodiment, FIG. 10 is a basic configuration of an embodiment for realizing security management, FIG. 11A is a table explanatory diagram, and FIG.
The figure shows the connection procedure. Fig. 11C shows the data management processing flow. Fig. 12 shows the system configuration in which terminals are grouped.
Figure A shows the configuration of an embodiment of a distributed database system with multiple groups, Figure 13B Database management processing flow between groups, Figure 14 shows the basic configuration of an embodiment that distributes the center load, and Figure 15A shows a specific example. FIG. 15B to FIG. 15I are diagrams for explaining each processing content of distributed management according to the embodiment configuration, and FIG. 16 is an explanatory diagram of a conventional example.

[Best mode for carrying out the invention]

Embodiments of the present invention for achieving the above objects will be described below in order.

FIG. 1 shows the basic configuration of Embodiment 1 of the present invention. In FIG. 1, reference numerals 1 denote all equivalent terminals each having an information processing function and a file 2 for storing a distributed database, and each terminal 1 is distinguished by reference numerals T1 to T4.
Here, T1 becomes the data providing terminal, T2,
T3 represents a user terminal. 2 is a file for storing the database of each terminal, 3 is a network such as a public network or a private network, and 4a is a table of contents (or title) of each data of a database of each terminal constituting the distributed database system.
A contents management center (temporary center) that manages the contents, and a contents file 5 that stores all contents (contents of published data) by pairing each contents with terminal identification information (terminal number) holding the data. This is a center file containing 50 files.

The table of contents management center 4a is one of the temporary centers according to the present invention. Here, the term “temporary center” is used, and its meaning will be described here.

According to the present invention, one or more terminals among a large number of terminals constituting the distributed database system are provided with management information (table of contents in the example of FIG. 1) on the distributed database system, and the information from other terminals is provided. The management information is checked in response to an inquiry (inquiry), and the center is assigned to perform a process of responding to the check result. A terminal having this center function is not fixed to a specific terminal, can be changed as needed, and a plurality of centers can be provided according to the type of management information. Center ”.

In the configuration of FIG. 1, a table of contents management center 4a is provided in the terminals constituting the distributed database system, and the terminal that provides data registers the table of contents (or "title") of data (material) in the table of contents management center. In this case, the actual contents of the data are stored at the data providing terminal, and the user terminal retrieves the table of contents from the table of contents management center, and retrieves the required data from the data providing terminal. It is. Further, the table of contents management center 4a includes a table of contents version management file 52 for managing the version of the table of contents in the table of contents management.

To explain the basic operation in the configuration of FIG. 1, when data is generated in the data providing terminal T1, the table of contents management center 4a is called via the network 3, and the table of contents of the data (information indicating the source of supply) Is transmitted to the table of contents management center 4a via the network 3 for registration. The data of the text of the data (material) is stored in the file 2 of the data providing terminal T1 and managed. Upon receiving the table of contents data, the table of contents management center 4a stores the data in the table of contents file 50 by the operation of the table of contents management means 121. The table of contents file 50 stores a table of contents transmitted from a plurality of data providing terminals.

When the user of the database accesses the table of contents management center 4a from the user terminal T3 via the network 3 and requests data from the table of contents requesting means 131, the table of contents management center 4
The table of contents management means a extracts all the tables of contents of the table of contents file 50 and transmits them to the user terminal T3. The user terminal T3 can know the list of the table of contents by a display device (not shown), and when one of the "table of contents" is selected, information indicating a source included in the "table of contents" is displayed. The data providing terminal T is transmitted from the text request means 132 via the network 10 using
Access 1 and request the text of the selected "Table of Contents".

The data providing terminal T1 retrieves the data of the text corresponding to the "table of contents" requested from the user terminal T3 by searching the file 2, and transmits the data to the user terminal T3.

FIG. 2A shows a specific embodiment of the distributed database system using the table of contents management center, and FIG. 2B shows an example of the operation sequence.

In the configuration of FIG. 2A, A and B are shown as the data providing terminals T1, but other configurations are the same as those of FIG.

The operation of the embodiment of FIG. 2A will be described below in the order of the operations shown in FIG. 2A with reference to the example of the operation sequence of FIG. 2B.

At the data providing terminal T1 (either A or B)
When data (material) is generated (or input), the data (table of contents and text) is stored in a file of the database of the terminal (1 in FIG. 2B).

At this time, if the user wants to use the data as a database on another terminal, the user accesses the table of contents management center 4a via the network 3 (2 in FIG. 2B). When the table of contents management center 4a is connected to the data providing terminal T1, the table of contents of the data to be registered as a database (or short letters or numbers such as a title) may be provided as provider information (terminal identification number) representing the terminal of the provider. As in the case of FIG. 2A, it is transmitted together with A or B) (3 in FIG. 2B). Upon receiving this, the table of contents management center 4a stores and manages the table of contents including the provider information in a file (4 in FIG. 2B).

When the user terminal T3 uses the database, it accesses the table of contents management center 4a and requests a table of contents when it is connected via the network (4 in FIG. 2B).

Upon receiving this request, the table of contents management center 4a extracts a table of contents (list) from the file to be managed and sends it to the user terminal T3 (5 in FIG. 2B).

When the user terminal T3 receives the list of contents, it displays it on the display screen of the display device as shown in FIG. 2A (6 in FIG. 2B).

The user of the user terminal T3 selects one table of contents from the list of tables of contents displayed on the display screen (performs the table of contents using input means such as a keyboard and a mouse). Then, the provider information (for example, A) of the data included in the table of contents is stored in the network 3
Is transmitted as a control signal for selecting the other party. As a result, the user terminal T3 accesses the data providing terminals T1, A in which the "table of contents" is registered by the operation of the network 3 (7 in FIG. 2B). After the connection, the data providing terminal requests the "text" of the selected "contents" information (2B
Then, the data of the "contents" of the "contents" is retrieved from the file managed by the data providing terminal (8 in FIG. 2) (9 in FIG. 2B).

When the data providing terminal T1 detects the corresponding "text", the data is transmitted to the requesting user terminal T3 (10 in FIG. 2B). In the user terminal 24, the "text"
The contents can be known by displaying or printing the contents.

In the configuration of Fig. 2A, if the data registered from the data providing terminal is to be modified (revised), the Table of Contents Management Center
For 4a, it is only necessary to send and register a table of contents including the number of revisions, and it is not necessary to transmit data every time revised data is generated unlike in the past.

According to this method, since it is not necessary to register the entire data as compared with the conventional method, it is possible to reduce the load on the network and the terminal that provides the data and the terminal that registers the data. In addition, since the user of the terminal makes a selection based on the table of contents, there is no need to receive unnecessary data (text), so that the access time can be reduced.

In addition, since the table of contents management center manages only the table of contents, the burden can be reduced.

Next, the operation principle of version number management in the table of contents will be described with reference to FIG.

In FIG. 1, when the user sends a modified table of contents of the database published by the user to the table of contents management center 4a from the data providing terminal T1, the table of contents management center 4a sequentially stores the modified table of contents in the temporary work file 51, Every time or every time the number of contents reaches the set number, the version number is added to the contents and transferred to the contents version number management file 52. When the table of contents version is requested from the user terminal T2 to the table of contents management center 4a together with the notification of the latest version of the terminal, the missing version after the latest version is retrieved from the table of contents management file 52 and queried. Transfer to the original terminal.

Based on the above principle, the processing flow of the
This will be described with reference to FIGS. 3A to 3B.

First, the processing flow of altering the table of contents (including registration, change, deletion, etc.) shown in FIG. 3A will be described. S1, S2 ... in the figure
Means Step 1, Step 2,...

Upon receiving a table of contents alteration request (S1 and S2 in FIG. 3A) from terminals A and B, the table of contents management center 4a opens the primary work file.
The modified contents of the terminal A and the modified contents of the terminal B are stored in 51 (S3). The state of storage in the primary work file 51 is as shown in the upper part of FIG. 3B. Here, S means a step.

Thereafter, when a certain time (for example, 24 hours) elapses, when the number of contents exceeds a predetermined number (for example, 100 items), or when a certain time (for example, 0:00) is reached, the update condition (S4 ), The entire table of contents stored in the primary work file 51 is transferred to the table of contents version number management file 52 (S5).

The storage state in the table of contents version number management file 52 is shown on the lower side of FIG. 3B. In this example, the version number update time is also registered along with the version number. Needless to say, the same version number is added to the table of contents transferred from the primary work file 51 to the table of contents version management file 52 together.

On the other hand, when the version number update condition is not satisfied in step S4, the table of contents management center 4a executes processing other than the table of contents version number management.

 Next, an example of a table of contents inquiry shown in FIG. 3C will be described.

For example, when a table of contents is queried from terminal A (S1
1) At that time, the latest version number 0003 of the terminal A is sent to the table of contents management center 4a. The table of contents management center 4a first determines whether or not the table of contents data exists in the primary work file 51 (S12). As a result, if it is determined that the table of contents is not stored, the version number 0004 to 0005 of the version number 0003 or later of the table of contents version management file 52 of the table of contents management center 4a (this version number 0005 is assumed to be the latest). The contents of the table of contents are transferred to the terminal A (S13).

When it is determined that the table of contents data exists in the primary work file 51, the version number 00 of the table
04 to 0005 to terminal A and the primary work file
The entire table of contents 51 is transferred to the terminal A as the version number 0006 (S1
Four). Here, for the version number 0003 inquired from the terminal, the latest version number 0005 stored in the table of contents version number management file 52
Not only is the previous version number 0004 transferred to the table of contents, because if the version number changes, it is not known whether the screen or the like desired on the terminal side is in the old version number or in the new version number. .

In this way, by managing the version numbers, the terminal-side user who requests the published table of contents can obtain only the table of contents other than the table of contents held by the user, thereby reducing the transfer time. In addition, the processing capacity of the table of contents management center is improved by the reduced time.

FIG. 4 shows the relationship between the terminal having the publisher identification function of the present invention and the center.

This terminal has a mechanism for distinguishing and processing data published by its own terminal and data published by other terminals when accessing the table of contents management center of the distributed database system.

In FIG. 4, 1 is a terminal, 21 and 22 are contents files provided in the terminal 1, 21 is a file for own station storing a table of contents of data held in the own station, and 22 is a file for data held in another terminal. A file for the other party in which the table of contents is stored, 4a indicates a file held by the table of contents management center, and 5 indicates a file held by the table of contents management center. The present invention is characterized in that the terminal 1 is provided with a publisher recognition unit 101.

First, the outline at the time of opening the database table of contents shown in FIG. 4A will be described. When the owner (user) of the terminal 1 wants to open his / her own database, the database is executed in the order of. That is, the table of contents of the database to be disclosed is extracted from the own station file 21 and transferred to the control unit 100.

In order to request the publication of the table of contents, the table of contents and the owner number (same as the terminal identification number) are transferred to the table of contents management center 4a.

The table of contents management center 4a stores the table of contents and the owner number sent from the control unit 400 in the file 5.

The control unit 400 of the table of contents management center 4a notifies the terminal 1 that the procedure for publishing the table of contents has been completed.

In this way, when the user of another terminal or the user of the same terminal wants to know the table of contents registered in the table of contents management center 4a, the processing shown in FIG. 4B is executed. That is, the terminal 1 requests the public table of contents to the table of contents management center 4a.

The control unit 400 of the table of contents management center 4a transfers the table of contents registered in the file 5 and its owner number to the terminal 1.

The control 100 of the terminal 1 passes the transferred table of contents and its owner number to the publisher recognizing unit 101. The publisher recognizing unit 101 compares the owner number of the received public table of contents with the user number of the terminal concerned, and if they match, discards the public table of contents (because it is the table of contents provided from the terminal itself), At the time, it is stored in the table of contents management file 22 for other persons and output to the input / output device as needed.

In this way, the table of contents of the database published by the user in the past is not duplicated.

Next, FIG. 5 shows a specific embodiment of the terminal 1 shown in FIG.

In FIG. 5, the control unit 100 is provided with a separation unit 100a, and the publisher recognition unit 101 is provided with a buffer 101a and a comparison unit 101b.
Is provided.

In operation, the public table of contents and its owner number sent from the table of contents management center are separated by the separating unit 100a and sent to the publisher recognizing unit 101. The publisher recognition unit 101
The table of contents is stored in the buffer 101a. Separation unit 10 in comparison unit 101b
The owner number (same as the terminal identification number) from 0a is compared with the user number (same as the terminal identification number) from the input / output device (not shown) to determine a match / mismatch.

As a result, if they match, the table of contents remaining in the buffer 101a is deleted without being output, while if they do not match, the table of contents in the buffer 101a is output and the table of contents for the other party is output under the control of the control unit 100. Store in file 22.

Such an operation is performed every time a public table of contents is sent from the table of contents management center 4a, and the file is stored in the file 22 for others.

In this way, it is possible to prevent the same user from receiving a table of contents previously published by a terminal user at the time of requesting the table of contents (when updating the table of contents). And the search time is reduced.

Next, in a distributed database system, the basic structure of an embodiment for solving the above-mentioned object of the present invention, which is to efficiently access data necessary for performing command transfer and efficiently obtain related data, will be described. It is shown in FIG.

FIG. 6A shows a configuration provided with a command management center, and FIG. 6B shows the principle of a common command system.

In FIG. 6, reference numeral 1 denotes terminals (T1 to T3) constituting a distributed database system, 2 denotes a file for storing a database of each terminal, 3 denotes a network, and 4b denotes a command management center (temporary center).

According to the present invention, in a distributed processing environment or a distributed database system, a center for managing data commands is provided to manage information on commands and terminals holding data corresponding to the commands, and to manage related data groups. A common command is set, and in response to an inquiry from a terminal, information on each terminal holding related data corresponding to the common command is obtained from the management center.

In FIG. 6A, each terminal T1 to T4 constitutes a distributed processing environment and a distributed database system, respectively. When a command is assigned to each of the data held by each terminal 1, it is registered in the command management center 4b. The command management center 4b includes a command management table 53 in which a terminal identification number is associated as a location of a command and data corresponding to the command. The command management center 4b searches the table 53 and responds with a terminal identification number in response to an inquiry by a command from each terminal. In the example of A., data corresponding to the command “COM” is held in the terminal T2, and data corresponding to the command “CON” is held in the terminal T3.

Next, in FIG. 6B, scattered mutually related data is expressed using a common command (compression code).
Unlike the conventional command transfer method, this command does not add a terminal identification number holding the data indicated by the command to the command, and the command is managed by the command management center 4b.

That is, the command management center 4b is provided with a common command management table 54 that stores terminal identification numbers of a plurality of data constituting a data group corresponding to the common command.

When receiving an inquiry from the terminal 1 by the common command, the command management center 4b refers to the table 54 and responds with the terminal identification number in which a plurality of related data are stored. Using this response information, the inquiring terminal 1 can sequentially collect the corresponding data from each terminal.

In this way, by providing a command management center for the system, it is possible to know the terminal holding the data corresponding to the command, and further, by using a common command, a group of related data can be stored at a plurality of locations. Even in the case of being scattered, all the locations can be known from the command management center.

FIG. 7 shows a system configuration of the embodiment showing details of the operation of the configuration shown in FIG.

7 correspond to the devices having the same reference numerals in FIG.

The operation of the embodiment will be described in the order of ~ shown in the figure.

When the terminal T3 wants to search and collect a related data group, it inquires of the command management center 4b.
Specifically, for example, in the case of a system in a company where the system shown in FIG. 7 is used, when data relating to a specific customer is scattered at each terminal, and the user wants to search, collect, and process the data. In that case, the command management center can use the "customer's name" without knowing the command.
You can contact 4b.

When the command management center 4b receives not a command but an attribute common to the data group (for example, the name of the customer) in the inquiry, the command management center 4b searches for a common command of the corresponding data group and detects that it is "VOL". You. At the time of inquiry, the common command is "VO
If the command management center 4b knows that the command is "L", the command management center 4b receives the common command.

When the common command "VOL" is obtained, the command management center 4b searches the common command management table 54 of the file device 5 and extracts the terminal identification information holding each data constituting the data group. The extracted terminal identification information (terminals T1 and T2 in the example in the figure) is transmitted to the inquiring terminal T3 as answer information.

The terminal T3 receiving the response information is the terminal T1, T2
Request data corresponding to the command “VOL”.

In response to this request, the terminals T1 and T2 extract the data corresponding to the command "VOL" from the respective files 21 and 22, and transfer the data to the terminal T3.

In this way, since the location of the data corresponding to the common command can be obtained collectively from the command management center, it is not necessary to inquire the terminal T4 in the above example.

Therefore, the conventional sequential inquiry method (Fig. 11
A.), it is possible to perform more efficient search and data collection. In addition, since the related data is expressed by a common command, compared with the conventional method of requesting each terminal for a command corresponding to each data (see B in FIG. 11).
Complex procedures can be avoided.

FIG. 8 shows a basic configuration of a real data management processing system of command matching for realizing the object of enabling the related data dispersed and scattered as single data according to the present invention.

In FIG. 8, 1 is a terminal (T1, T2), 2 is a file for storing actual data, 3 is a network, and 4c is a command link center (temporary center).

The terminal 1 shown in FIG. 8 is provided with means for assigning a command for expressing the actual data to each actual data and for matching different commands as necessary. The actual data stored in the file 2 is managed by the corresponding terminal 1 using a command expressing this. The command link center 4c collectively manages information (command matching information) on the matched commands in each terminal 1. For this reason,
Each terminal 1 notifies the command link center 4c of the contents each time a command is matched.

Command (Y) for terminal 1 (T2)
(For example, display processing) is input.
In response, the terminal T2 inquires of the command link center 4c whether there is a command matched with the command Y. Prior to this inquiry, a command (X) different from the command Y is matched with the command Y in another terminal 1 (T1). Then, the command link center 4c communicates the command X of the terminal T1 with the terminal
The fact that the command Y of T2 has been matched is created and held as command matching information.

The terminal T2 is informed by the command link center 4c that its command X has been matched to its own command Y, and is scattered between the different terminals T1 and T2 represented by the commands X and Y. Display processing is performed with the actual data X and Y to be used as single data.

In this manner, desired scattered and related actual data can be combined by matching commands, and related data can be processed for one of the matched commands without being aware of combining real data by processing input. Can be.

FIG. 8 shows a specific embodiment of the principle configuration shown in FIG.
It is shown in Figure 9A.

In FIG.9A, 1 is a terminal, 2 is a file provided in the terminal, 3
Represents a network.

The processing unit 5 of the terminal 1 performs various data processing (including command processing), requests the communication control unit 10 to communicate with other terminals, requests the learning control unit 11 to perform learning processing, Process the input code from 17. In addition, a command processing unit 13, a real data processing unit 14, and a display data processing unit 15 are provided for processing commands, real data, and display data.

The command processing unit 13 creates a command from the actual data stored in the actual data file 24 and assigns the command as a command expressing the actual data. This command is stored in the command file 23, and the file is managed by the command processing unit 13. The command processing unit 13 matches two different commands as necessary (declares that the different commands are the same), and
This is notified to the command link center 4c.

The actual data processing unit 14 inquires of the command link center 4c via the communication control unit 10 about the presence or absence of a command matched with the command to be processed in the command processing, and based on the response from the command link center 4c, To retrieve and retrieve real data. That is, based on the response, the actual data represented by the command to be processed is extracted from the actual data file 24 of the terminal using the command. Also,
The real data processing unit 14 registers (stores) the real data input from the own terminal 1 in the real data file 24 and manages it.

On the other hand, if the actual data represented so far does not exist in the actual data file 24 of the own terminal 1, a command is notified and the learning control unit 11 is requested to take out the actual data.

The display data processing unit 15 performs processing (realization processing) for display (realization) on the plurality of extracted real data, and sends it to the display unit 16. The display unit 16 converts the materialized data into displayable data and displays it on the display device 18.

The communication control unit 10 performs communication with another terminal or the command link center 4c via the network 3. This communication includes an inquiry to the command link center 4c at the request of the processing unit 12 and its response (reception of command matching information), a request of real data at the request of the learning control unit 11, and a response thereof (reception of real data). ).

Upon receiving the request from the processing unit 12, the learning control unit 11 requests the other terminal for the actual data represented by the matched command. In response to this request, the other terminal retrieves the corresponding actual data from the actual data file 24 and sends it to the requesting terminal 1. The plurality of extracted real data are written to the real data file 24 by the learning control unit 11.

The actual data written by the learning control unit 11 exists as the data of the terminal 1 in the actual data file 24 unless an erasing process is performed thereafter. Therefore, thereafter, the reference to the actual data can be performed without transferring the actual data itself. On the other hand, the command used for extracting the actual data is discarded without being stored in the command file 23 upon completion of the writing. That is, in the command file 23, only the commands allocated by the terminal 1 are stored.
The command obtained from 4c is not stored.

The actual data is substantially managed by the learning control unit 11, and the actual data processing unit 14, which manages the actual data file 24, cannot directly refer to the actual data. Also, the user and the processing unit 12 transmit the actual data to an actual data file.
I cannot (and cannot) be aware that it exists at 24.

In order for the actual data processing unit 14 to refer to the actual data, it is necessary to make an inquiry to the command link center 4c as described above, and access the actual data file 24 with the command obtained as a response. In this case, actual data is not transferred. Since this command is discarded after reference, the command link center 4c records the number of times the command has been referenced, and can use this as a charging means for referring to actual data.

The input control unit 17 receives an input from the input device 19 and sends an input code to the processing unit 12. The input device 19 is used for processing input of commands, input of matching of commands, input of actual data,
Used to input a data request to another terminal.

When a real data is requested by specifying a command from another terminal, the processing unit 12 causes the real data processing unit 14 to access the real data file 24 by the command, retrieve the real data, and transmit the data to the requesting terminal. .

Also, as mentioned above, Command Link Center 4c
Is actually one of the terminals 1, and any terminal connected to the network 3 may be the command link center 4c. That is, the command link center 4c only manages the command coincidence information. When the terminal 1 becomes the command link center 4c, the command matching information is stored in a dedicated file. When a command is specified from another terminal and the command matching information is inquired, the processing unit 12 or the command The processing unit 13 accesses the file of the command matching information by the arriving command, extracts the command matching information, and transmits the command matching information as a response to the inquiry source terminal.

For example, a display format screen (one “data”) of a certain related data is one terminal (T0).
Is created and released to each terminal, the command link center 4c for the relevant data is
It may be. It is not necessary to define a different command link center 4c for each data, and the terminal serving as the command link center 4c may be fixed to a host terminal (host computer).

In addition, each terminal inputs data (changes the data) to the format screen of the opened display, and attempts to match by a command. As a result, the actual data is dispersed and scattered,
In addition, it is possible to easily form a distributed database as single data by matching commands.

The operation in the configuration of FIG. 9A will be described with reference to FIGS. 9B to 9G. FIG. 9B is a flow chart of real data management processing, FIGS. 9C to 9E are diagrams for explaining command matching, FIG. 9F is a flow chart of management processing for a plurality of real data, and FIG. 9G is a description of extraction of real data. FIG.

The actual data management processing flow in FIG. 9B particularly shows a command matching process (integration of actual data, that is, formation of a distributed database). In this example, it is assumed that the terminal T0 operates as the command link center 4c and is connected to the network 3 together with the terminals T1 to T4.

Assuming that the terminal T1 creates and releases the actual data A, the terminals T2 to T4 are used for the actual data A without inquiring of the command link center 4c or the terminal T1.

9B will be described below with reference to FIGS. 9C to 9E.

First, a request for data A is input from the input device 16 to the terminal T2.

The processing unit 12 of the terminal T2 searches the real data file 24 using the command A, but since the real data A does not exist,
It requests the learning control unit 11 to extract the actual data A. The learning control unit 11 specifies the command A and
The real data A is requested to T1, and when the data is taken in from the terminal T1, it is written to the real data file 24.

The display data processing unit 15 performs a realization process of the real data A, converts the data into data that can be displayed by the display unit 16, and displays the data.
The data A is displayed at 18.

The data A is changed by adding new data to the data A from the input device 19 or the like.

An instruction to register “changed data A” as data C in the actual data file 24 is input from the input device 19. The command processing 13 assigns the command C to the data C and registers the command C in the command file 23.

A command to match command C with command A is input from input device 19.

The command processing unit 13 declares that the command C is to be matched with the command A, and notifies the command link center 4c of this. The command processing unit 13 creates command match information “C = A” or “C (T2) = A (T1)” and sends it to the command link center 4c.

The two real data A and C thus dispersed are in the 9D
As shown in the figure, they are matched and handled as one data. Note that the actual data A is actually processed by the terminal T2.
Also exists in the actual data file 24 of the.

The terminal T1 matches the actual data A and the actual data B existing in the actual data file 24 as shown in FIG. 9E. This can be realized, for example, by executing the processing.

The terminal T3 matches the command D with the command A in the same manner as in the processing and the like.

The terminal T4 matches the command E with the command A in the same manner as in the processing.

Each time the command link center 4c receives the command matching information, it stores it in a dedicated file.

The above is summarized as shown in FIG. 9C, and as a result, the code matching information is “A (T1) = B (T1) = C (T2)
= D (T3) = E (T4) ".

That is, data A to E scattered and dispersed to terminals T1 to T4
Are treated as one data.

FIG. 9F is a processing flow in the case of single processing of a plurality of actual data.

This processing flow shows the processing of the actual data matched by the processing shown in FIG. 9B, particularly the display processing.

A display request for data C is input from the input device 16 to the terminal T2.

The real data processing unit 14 of the terminal T2 inquires of the command link center 4c whether there is a command matched with the command C.

The command link center 4c searches for matching information on the command C, and obtains the command matching information “C (T2) =
A (T1) = B (T1) = D (T3) = E (T4) ", the commands A, B, D and E or A (T1), B (T1), D (T
3) and E (T4) are returned to the terminal T2.

When receiving the matched command, the processing unit of the terminal T2
12 requests the learning control unit 11 to retrieve the actual data corresponding to these. Using these, the learning control unit 11 requests the real data A and B to the terminal T1, the real data D to the terminal T3, and the real data E to the terminal T4. The real data processing unit 14 may extract the real data C from the real file 24.

The processing unit 12 or the real data processing unit 14 of each terminal T1, T3, T4 extracts the real data A and B, D, E corresponding to the requested command from each real data file 24, and returns them to the terminal T2.

Real data A, B, received by the learning control unit 11 of the terminal T2
D and E are registered in the actual data file 24.

The real data processing unit 14 extracts the real data A, B, D, and E from the real data file 24 by using the matched command, and sends it to the display data processing unit 15 together with the real data C. The display data processing unit 15 displays the actual data A to E on the display device 18.

The above is summarized as shown in FIG. 9G, and as a result, the actual data A to E having the mutually matched commands are obtained.
Are treated as one data (displayed). By specifying one of the matched commands, the same processing can be performed as one piece of data for all the actual data having the matched command without being conscious.

This is the same regardless of which of the terminals T1 to T4 specifies any of the commands A to E. However, each terminal T1 to
The commands that T4 can know are limited.

The display data processing unit 15 performs the realization processing in accordance with the display mode specified in the display request. To obtain the entire display of the actual data A to E (OR input),
To E are displayed.

At this time, if a plurality of data overlap at the same I position on the display screen, processing is performed to display the data in a window format. Further, the information may be displayed once and then sequentially displayed by a click operation using a pointing device or the like. On the other hand, when it is desired to display only matching parts of the actual data A to E (AND input), processing is performed so that only matching parts are displayed.

As described above, data can be written to the common (opened) data (format) A from any terminal, and can be displayed as one data from any terminal.

After the process of FIG. 9F, when the display request of the data C is input again from the terminal T2, the following process is performed.

First, the processing shown in FIG. 9F is performed. Thus, the matched commands A, B, D, and E are obtained. Next, the real data file 24 is searched by the commands A to E, and the real data A to E are extracted. Therefore, no data transfer is performed.

Next, data A to E are displayed in the same manner as the processing in FIG. 9F.

As described above, related data A to E can be obtained without data transfer. On the other hand, B to E which are not released cannot be referenced from terminals other than the terminal that created the data without passing through the command link center 4c.

Therefore, the security of the actual data is maintained, and the charging is not hindered.

According to the embodiment shown in FIGS. 9A to 9G,
In real data management processing in a distributed database,
By expressing data as commands and matching the commands, a plurality of real data scattered and distributed to a plurality of terminals can be handled as a single data, so that a distributed database system can be easily formed. . In addition, processing using a distributed database can be performed quickly and easily.

FIG. 10 shows the configuration of an embodiment for realizing the object of performing security management according to the degree of importance of published data in the distributed database system of the present invention.

In FIG. 10, 1 is a terminal (T1, T2), 3 is a network,
Reference numeral 4d denotes a management center (temporary center) for security management, and reference numeral 5 denotes a file provided in the management center and storing various management tables.

 The table provided in the file 5 will be described.

The table of contents list 57 is composed of information necessary for browsing data released to other terminals, for example, a list of items (item names) and storage positions of released data.

The browsing condition table 56 includes information for judging whether or not browsing is possible, that is, an item of open data and a rank of importance assigned to the item (this is referred to as a first rank (level)).

Management information table 55: composed of information for determining access qualification and browsability, that is, identification information such as a personal ID, a password, and a keyword, and the rank of the individual (this is referred to as a second rank (level)). I do.

The first and second ranks are set in a plurality of stages according to the importance of the release data and the identification information.

To explain the operation of FIG. 10, the terminal T1 inputs a browsing request designating the open data and inputs identification information in order to view the open data 2 held by another terminal. This input is sent to the management center 4d via the network 3. Based on this input, the management center 4d extracts the specified open data 2 and the first and second ranks corresponding to the input identification information from the browsing condition table 56 and the management information table 55, respectively. Next, the two ranks are collated, and whether the browsing is permitted or not is determined based on the result, and the browsing is permitted or rejected. For example, when the second rank is equal to or higher than the first rank, the terminal T1 is permitted to browse the table of contents list table 57. Specifically, the contents list table 57 is sent to the terminal T1 via the network 3. Thus, the terminal
T1 can obtain information for viewing the release data 2. Thereafter, the terminal T2 is accessed via the network 3 and the open data 2 is browsed.

On the other hand, if the second rank is smaller than the first rank,
Viewing of the table of contents list table 57 by the terminal T1 is not permitted.
Therefore, the terminal T1 cannot know the storage position of the release data 2 and cannot browse it.

As described above, the viewing qualification of the open data 2 can be restricted in accordance with the importance.

Explaining in more detail the contents of FIG. 10, the open data of the terminal T2 is composed of an item and actual data as shown in the figure, and even if the open data 2 is the actual data, the actual data is only transmitted to the terminal T2. Exists, the item exists in terminal T2 and management center 4d.

When the terminal 1 creates the data of the database, the terminal 1 retains the data in its own terminal, determines whether or not to release the data to another terminal, and performs necessary processing. That is, the unreleased data is not registered in the management center 4d. For the release data 2, the corresponding rank (first rank), own terminal number, storage location of the data, and the like are sent to the management center 4d via the network 3. Therefore, the first
The rank of indicates the degree of importance of the open data. For example, in an in-company database, the information may be "more than 3 years after joining the company (but browsing is permitted)", "more than section manager", and the like.

With respect to the terminal T1 and the open data 2 of the own terminal, the requested open data 2 is permitted to be browsed and transmitted to the other terminal T1 which has accessed the predetermined information such as the item and the storage position of the terminal T1.

The management center 4d is one of the terminals 1 connected to the network 3, and includes the following processing units.

The management information processing unit 40 performs management information table 55 based on the preset identification information of the terminal (personal user) and the preset rank (second rank) corresponding thereto.
, The qualification of the personal information is checked, and if the personal information is qualified (exists in the management information table 55), the corresponding second rank is extracted and sent to the browsing processing unit 41.

The browsing processing unit 41 creates and holds a browsing condition table 56 in accordance with the registration of the open data 2 from the terminal 1. In this case, the item transmitted as the release data and the corresponding rank (first rank) are registered in the viewing condition table 56.

Further, the browsing processing unit 41 responds to an input from the terminal 1
With reference to the viewing condition table 56, a rank (first rank) corresponding to the personal information is extracted.

The first rank is compared with the second rank from the management information processing 40, and when the result of the comparison is a predetermined value, for example, the second rank is “third year of joining a company” and the first rank is “ If you have been in the company for more than 3 years, you will be notified of your permission to view.

The table-of-contents management unit 42 transmits the contents of the table-of-contents list table 57 to the accessed terminal 1 in response to the notification of the browsing permission from the browsing processing unit 41. At this time, the contents transmitted are limited to those corresponding to (the items of) the open data 2 requested to be browsed by the terminal 1. In this way, browsing of other open data 2 can be prevented.

 FIG. 11A is an explanatory diagram of a table.

As shown in (A), the management information table 55 stores the name, personal ID (identification number), and rank for each user, and further stores a password, a keyword, and the like that can be used by this user.

(B) shows the browsing condition table 56, and the item and the rank are associated with each open table 2.

(C) shows a table of contents list 57 in which items, terminal numbers, storage positions, etc. are stored in association with each open table 2.

11B and 11C are a connection procedure explanatory diagram and a data management processing flow, respectively.

Hereinafter, the connection procedure and the data management processing will be described step by step.

The data created by the terminal T2 (see FIG. 10) is released as release data 2, and the necessary data described above is sent to the management center 4d.

The table of contents management unit 42 of the management center 4d registers the sent information in the table of contents position table 57, and the viewing processing unit 41 registers the sent items in the viewing condition table 56.

The terminal T2 checks the rank of the released data by the management center 4d.
Send to

The browsing processor 41 of the management center 4d registers the rank at a predetermined position in the browsing condition table 56.

The other terminal T1 issues a request to view the open data 2 to the management center 4d via the network 3.

The management center 4d requests the request source terminal T1 via the network 3 to input identification information.

The terminal T1 inputs identification information to the management center 4d via the network 3.

The management information processing unit 40 of the management center 4d searches the management information table 55 to determine the qualification of the identification information. If the user is qualified, the corresponding second rank is extracted from the table 55.

The browsing processing unit 41 searches the browsing condition table 56 and extracts a corresponding first rank.

The browsing processing unit 41 compares the first and second ranks and determines whether browsing is possible.

If browsing is permitted, the table of contents management unit 42 sets the table of contents list
Take out the contents of 57.

The management center 4d sends the contents to the request source terminal T1 via the network 3.

If the browsing is not permitted, the management center 4d does not transmit.

Thereafter, the terminal T1 requests the terminal T2 for release data using the received contents.

In this way, by managing the information for browsing the open data using the ranks in the management center, it is possible to limit the qualification for browsing according to the importance of the open data, so that the security of the open data is established. It can promote the release of data and mutually utilize the database.

Next, a technique for managing a database by grouping terminals that can use the database as a group with terminals connected to the network of the distributed database system and a technique for using data across groups will be described with reference to FIGS. This will be described with reference to FIGS. 13A to 13C.

 FIG. 12 shows an overall configuration diagram.

In FIG. 12, 1 is a terminal (T1, T2, T3), 2 is a database which is distributed and held in each terminal, 3 is a network, 4 is one of the terminals, and a management center (4) having a management function ( (Tentative center), 6 indicates the boundary of the group.

In this configuration, similarly to the above-mentioned distributed database system, a terminal 1 for distributing and holding a plurality of databases is connected to a network 3, and a predetermined number of terminals 1 form a group (group A in the example in the figure). I do. One terminal in the group A functions as the management center 4,
Have information necessary for management. Terminal T3 does not belong to group A. The information required for management includes which terminal holds which data and which data is open / non-open. The management information may include data such as a password for checking a terminal or a user.

The terminal 1 constituting the group A requests the management center 4 via the network 3 to search for data held by other terminals. When the network 3 is a public network, the management center 4 checks whether the requesting terminal 1 is a terminal in the A group (by receiving check data such as a password from the terminal 1) and rejects the check. If not, the data is searched based on the request, and the search result (for example, the terminal number holding the data) is notified to the request source. The requesting terminal 1 refers to the data from the database of the terminal 1 that holds the database based on the notification.

12, the terminal T1 is an independent computer. Therefore, the terminal T1 and the management center 4 are connected to each other via a network to form a computer network. The terminal T1 and the management center 4 are, for example, terminals that belong to one company and are dispersed in a region,
Therefore, one group is formed and the database is shared. On the other hand, the terminal T3 is connected to the network, but belongs to another company, for example.

The management center 4 has an information file for management. The information file stores management information for each data (information) of the database 2 or for each data held by one terminal. One piece of management information includes information name, management abbreviation, search keyword, comment storage list, terminal ID (terminal number), component ID (data storage position),
It consists of data items (items) of the database 2 such as charging data, erasing data, availability of release, and the like. Data of the database 2 is distributed and held in each terminal.

FIG. 13A shows an embodiment of a distributed database system having a plurality of groups.

In the figure, 1A, 1B, 4A, 4B, 2A, and 2B attached to each code represent terminals belonging to groups A and B, a management center, and a database, respectively, and 7 represents an inter-group management terminal.

In this example, the groups A and B respectively hold different databases 2A and 2B distributed to a plurality of terminals 1A and 1B, respectively. If the data in the database 2A of the group A is a predetermined one, it is also opened to terminals other than the group A. That is, the terminal 1B of the group B is also allowed to refer to. Conversely, the same applies to the database 2B of the group B.

Databases 2A and 2B between these different groups A and B
Is always performed via the inter-group management terminal 7. In that case, between terminals 1A and 1B, or management centers 4A and 4B
It is not allowed to send and receive data directly between

The inter-group management center 7 is composed of terminals connected to the network 3 and mutually has databases 2A and 2B.
Are provided in correspondence with a plurality of groups that allow reference to be performed. The inter-group terminal 7 is connected to the management center 4A, 4 in each group.
The same processing as B is performed between the groups A and B. That is, management for referencing the databases 2A and 2B between the groups A and B and management of communication between the groups A and B for the reference are performed.

Therefore, the information is collected from the inter-group management centers 4A and 4B and registered in an inter-group management information file (not shown).

The inter-group management terminal 7 is connected to the management centers 4A and 4B.
When there is a search request for referring to the databases 2B and 2A of the other groups B and A from the group, the inter-group management information file is searched and the result is notified to the request source management centers 4A and 4B.

Further, the inter-group management terminal 7 is connected to the management centers 4A and 4B.
Receives a data transfer request addressed to the management centers 4B, 4A of the other groups B, A, receives data from the request source and sends it to the transfer destination.

The above processing is performed between the inter-group management terminal 7 and each group A,
This is executed as a process between the management centers 4A and 4B of B.
Therefore, the terminals 1A and 1B and the inter-group management terminal 7 do not directly communicate with each other, but are always performed via the management centers 4A and 4B.

In the registration of management information, each management center 4A, 4
B notifies the inter-group management terminal 7 of only the management information on the released data among the management information owned by itself.

Next, a database management processing flow between the groups in FIG. 13A will be described with reference to FIG. 13B.

Search information for searching for data to be referred to, such as the name of information, is input from a user to one terminal 1A of the group A.

The terminal (request source terminal) 1A requests the management center 4A of the group A to which the terminal 1A belongs to perform a search for reference via the network 3.

Based on this request, the management center 4A searches for its own management information file.

As a result of the search, it is checked whether or not management information on the specified data exists (data managed by the management center 4A). That is, it is checked whether or not the data is data in the group A.

If the data is managed by the management center 4A, the requesting terminal 1A is notified via the network 3 of the terminal 1A (such as the terminal ID number) holding the data as a search result.

If the data is not included in the data managed by the management center 4A, the management terminal 4A requests the inter-group management terminal 7 via the network 3 for a search for referring to this data.

Based on this request, the inter-group terminal 7 searches for an inter-group management file. In this case, the terminal 7 between groups
Performs a search on the database 2B that is distributed and held by the group B other than the group A.

The inter-group management terminal 7 notifies the request source management center 4A of (the terminal ID number of the terminal 1B) holding the data as a search result.

The management center 4A notifies the request source terminal 1A of the search result via the network 3. If the data is non-open, the inter-group management terminal 7 notifies that no data exists (or that the data is not open) as a search result.

The request source terminal 1A accesses the terminal 1A or 1B holding the data via the network 3 based on the search result at the management center 4A or the inter-group management terminal 7,
Request the data.

At this time, in the case of terminal 1A, direct management center
The request is made without going through 4A.

On the other hand, for terminal 1B,
A, the request is made via the group management inter-management terminal 7 and the management terminal 4B.

The terminal 1A or 1B that has received the request sends the data to the requesting terminal 1A via the network.

Also in this case, the data is transmitted through the same processing path as the processing.

Next, a basic configuration of an embodiment of the present invention which enables distribution of a load for managing a distributed database system and distribution of danger at the time of failure will be described with reference to FIG.

In the figure, 1 is a terminal (T1 to T4), 2 is information unique to the terminal owned by the terminal, 3 is a network, and 5 is a management information file.

14, the management information processing unit 10 of the terminal T2 declares that the management of the database is performed for a predetermined item in the management information, And collects the information in the management information file 11. Thereafter, the management information processing unit 10 uses the management information file 11 storing the predetermined item,
The management of the database for the predetermined item in the distributed database system is performed. That is, when viewed from another terminal, this terminal T2 functions as a center for managing a predetermined item.

Declaring the above processing, that is, managing the database for one or more specific management information,
By collecting the management information, each terminal 1 can be a center of the management information of each item.

For example, terminal T1 manages the item of “personal ID” and terminal T3
Is "Table of Contents" and terminal T4 manages "Keywords".
The management information processing unit 10 of each terminal T1, T3, T4 has a personal ID
The management information or management information file 11 for the file, the table of contents, and the keyword is created.

As the management information center of the distributed database system, in addition to the above, a command management center (FIG. 6) for managing commands and common commands, and a command link center (FIG. 6) for managing matched data of commands. 8) etc. are naturally included.

An embodiment of FIG. 14 will be described with reference to FIGS. 15A to 15I.

In FIG. 15A, 1 is a terminal constituted by a processing device,
11 is a management information file, 2 is a database, and 3 is a network.

The processing unit constituting the terminal 1 is a central processing unit (CPU)
And an independent computer, and a plurality of terminals 1 constitute a computer network.

The terminal 1 creates and holds a database 2 as needed. The database 2 is provided with a table of contents and / or keywords. The storage location of the database 2 can be known from the table of contents and the keyword. The information about the table of contents includes, in addition to the table of contents, positional information of the database 2 and labels / levels (ranks) of the database 2. The information on the keyword includes the position information of the database and the label / level of the database 2 in addition to the keyword. Labels / levels are used to check the qualification of the accessed individual for data security.

In addition, the terminal 1 has personal ID information on the user who uses the terminal 1. This information includes personal ID and affiliation
Includes ID / Personal ID label / level.

The management information processing unit 10 includes a center declaration processing unit 10a, a search processing unit 10b, and a center change processing unit 10c.

The center declaration processing unit 10a performs a declaration for managing a database, and collects and stores data for certain items in the management information. That is, it declares (notifies) to each terminal 1 that the item is to be managed from that point in time together with its own terminal number, then collects the corresponding information from each terminal 1 notified, and stores the management information file 11 in the management information file 11. create. As a result, information on the item is concentrated on one terminal 1 from each terminal 1 in the group. Therefore, for this item, this one terminal 1 is the management center. As described above, the center declaration processing unit 10a establishes a management center for the item.

The search processing unit 10b searches its own management information file 11, checks based on the search result whether or not reference to the database 12 is permitted, and includes the database 2 to be referred to only when reference is permitted. The terminal 1 is requested to perform the search. When the management information file 11 is used for personal ID, table of contents, and keywords, when the management information file 11 is a command, a database name (table of contents), and a database keyword are input, respectively. Done. This input is sent to terminal 1 (first
The search can be performed from any of T1 to T4) in the figure, while the search is performed by the search processing unit 10b of the terminal 1 which is the management center of the item.
Only do.

Further, the search processing unit 10b checks whether or not the reference is permitted by comparing the label / level of the personal ID with the label / level of the database 2 designated to be referred to. The label / level of the personal ID is obtained from the management information file 11 for the personal ID.

For this purpose, the search processing unit 10b accesses the terminal 1 which is a personal ID management center. The label / level of the database 2 is obtained from the management information file 11 for its own table of contents or keywords. Therefore, the search processing unit 10b that performs this processing belongs to the terminal 1, which is a table of contents or a keyword management center.

From the above, the functions of each management center can be summarized as follows.

The personal ID management center collects information on personal IDs, such as personal IDs, affiliation IDs, labels / levels of the personal IDs, etc., from the terminals 1 of the group and manages them collectively.

Table of contents management center, each database in the group 2
It collects information on the table of contents and the table of contents such as labels / levels, and manages the table of contents for each database in the group. Further, the label / level of the database 2 is compared with the label / level of the accessed personal ID.

The keyword center performs the same processing as the index management center for the keywords of each database 2 in the group.

The center change processing unit 10c copies the function of the terminal 1 as a management center to another terminal 1, and changes the management center. For this reason, the center change processing unit 10c checks whether the function operation of the center managed by the change destination terminal 1 is possible, and transfers the own management information file 11 to the change destination terminal 1 if possible.

Next, the distributed management processing of the database will be specifically described with reference to FIGS. 15B to 15I.

In the following description, it is assumed that the terminals T1, T2, and T3 are a personal ID management center, a table of contents management center, and a keyword management center, respectively, and the database 2 to be referred to is in the terminal T2. It is assumed that the user refers to the database 2 from the terminal T4.

 First, processing for establishing each management center is performed.

As shown in FIG. 15B, the terminals T1, T2, and T3 declare that they are management centers for personal ID, table of contents, and keywords, respectively, and collect the corresponding information.

 This processing is performed as shown in FIG. 15E.

The system administrator or the like specifies an item of management information to the terminal 1 and inputs an instruction to establish a management center.

In response, the terminal 1 determines whether or not it has a function as a management center, that is, the management information processing unit 11.
Check to see if you have If not, the process ends.

If so, the center declaration processing unit 10a declares to each terminal 1 that it is a management center for the item, and collects the item via the network 3.

The center declaration processing unit 10a stores the collected information in the management information file 11, and establishes a management center.

As shown in FIG. 15B, when one terminal is a management center for only one item, concentration of load can be avoided.

 Next, the change of the management center will be described.

 The change of the management center shown as an example in Figure 15C.

 Table of contents management center is changed from T2 to T5.

This processing is performed by the center change processing unit 10c of the terminal T2 according to the processing flow of FIG. 15F.

The system administrator or the like specifies the change destination terminal T5 as the terminal T2 and inputs an instruction to change the management center.

The center change processing unit 10c checks via the network 3 whether the terminal T5 has a function as a management center.

If not, this is displayed to the system administrator or the like, and the processing is repeated.

If provided, the contents of the management information file 11 for the table of contents are transferred to the terminal T5.

The management information processing unit 10c of the terminal T5 provides the management information file 11 for the table of contents, stores the transferred contents, and establishes a new management center.

When the terminal 1 (for example, T1) already functioning as the management center is specified as the change destination, the terminal T
The first management information processing unit 10 can reject the transfer.

 Next, reference to the database 2 will be described.

As shown in FIG. 15D, the user of the terminal T4 inputs a personal ID, a command, and a keyword (or table of contents) of the database 2 to his / her terminal T4. Since the keyword was entered,
Terminal T4 is a keyword management command center, terminal T3 via the network, keyword, command, personal ID
At the same time, the own terminal number is notified (i in the figure).

The terminal T3 requests the terminal T1, which is a management center for the personal ID, to search for personal ID information via the network 3, together with the command, the personal ID and the own terminal number, and receives the result (ii in the figure). .

The search processing unit 10b of the terminal T1 performs the following processing shown in FIG. 15G.

The management information file 11 for the personal ID owned by itself is searched according to the command input.

As a search result, the terminal T3 is notified of the label / level and the like of the personal ID by the terminal number via the network 3.

The terminal T3 searches its own keyword management information file 11 by keyword, checks the qualification of the accessed personal ID, and finds out the database 2 found as a result of the search.
Is requested via the network 3 to the terminal T2 in which exists (iii in the figure).

The terminal T2 having received the request searches the database 11 and sends it to the access source terminal T4 instead of the terminal T3 via the network 3 (iv in the figure). Note that terminal T2
Does not perform processing as a table of contents management center at this time.

The operation as shown in FIG. 15D is executed by the processing flow shown in FIG. 15H and will be described below.

In response to the input of the keyword, the search processing unit 10b of the terminal T3 searches its own keyword management information file 11 to know the storage position, label / level, etc. of the database 2 corresponding to the keyword.

The search processing unit 10b checks the qualification by comparing the label / level of the database 2 with the label / level of the personal ID obtained from the terminal T1.

If you do not qualify (the label / level does not match)
This is notified via the network 3 to the search processing unit 10b.
Notifies the terminal T4 with the terminal number. in this case,
The process ends without the content of the database 2 being displayed on the terminal T4.

If the user is qualified, the search processing unit 10b notifies the storage position of the database 2 and the terminal number of the access source terminal T4 via the network 3.

The terminal T2 searches the database 2 according to the storage location.

Terminal T2 is database 2 retrieved by terminal number
To the terminal T4 via the network 3.

Thus, the contents of the database 2 to be referred to are displayed on the terminal T4 as the access source.

When the user specifies the database 2 by the table of contents instead of the keyword, the following processing is performed.

First, the terminal T4 sends a table of contents, a command, a personal ID,
Notify own terminal number.

The terminal T2 accesses the terminal T1, which is a personal ID management center, and obtains personal ID information.

Further, the terminal T2 searches its own table of contents management information file 11 and checks the qualification of the personal ID. Since the database 2 exists in its own terminal T2, the terminal T2 searches the database 2 and sends it to the terminal T4. send.

The processing flow when specified by the table of contents above is
Shown in the figure.

The search processing unit 10b of the terminal T2 according to the input of the table of contents (name)
Searches the table-of-contents management information file 11 to know the storage position, label / level, etc. of the database 2 corresponding to the table of contents.

The detection processing unit 10b checks the label / level of the database 2 against the label / level of the personal ID obtained from the terminal T1, and checks the qualification.

If there is no qualification, this is notified to the terminal T4 and the process ends.

If there is a qualification, the search processing unit 10b requests the terminal T2 where the database 2 exists to perform the search. Terminal
T2 searches the database 2 by the storage position.

Terminal T2 is database 2 retrieved by terminal number
Is sent to the terminal T4 via the network 3. As a result, the contents of the database 2 to be referred to are displayed on the terminal T4 as the access source.

As described above, in the distributed database system of the present invention, the database management function can be distributed to a plurality of terminals or can be changed to another terminal.
Concentration of management load on one center can be prevented, and the danger that the database cannot be used in the event of a failure can be prevented.

INDUSTRIAL APPLICABILITY The provisional center system of the distributed base data system according to the present invention can be used for various services in which a large amount of data is used as a common resource by a distributed database system connected to a network such as a public network or a private network. For example, in information processing, information provision services, and various businesses that handle data,
It is suitable for efficient operation with quick access to data while reducing the load on the network.

By grouping and building databases,
A database in a group can be effectively used by a distributed database system grouped by offices and related departments distributed over a wide area via a network.

In addition, since the security of the data is ensured by limiting the terminals (or individuals) that can access the data in the database to a certain range, the disclosure of the data can be promoted. Since the center having the management function of the distributed database system is shared by each terminal, the load on each terminal does not become excessive.

Continued on the front page (31) Priority claim number Japanese Patent Application No. 1-339369 (32) Priority date Hei 1 (1989) December 27 (33) Priority claim country Japan (JP) (31) Priority claim number Special Japanese Patent Application No. 2-4809 (32) Priority Date Heisei 2 (1990) January 12 (33) Priority claiming country Japan (JP) (31) Priority claim number Japanese Patent Application No. 2-56036 (32) Priority date Hei 2 (1990) March 7 (33) Priority claim country Japan (JP) (31) Priority claim number Japanese Patent Application No. 2-56037 (32) Priority Date Hei 2 (1990) March 7 (33) Priority Claiming country Japan (JP) (31) Priority claim number Japanese Patent Application No. 2-56038 (32) Priority date Hei 2 (1990) March 7 (33) Priority claiming country Japan (JP) (72) Inventor Kakehi Former 5-20-9-101, Eda Minami, Midori-ku, Yokohama-shi, Kanagawa-ken (72) Inventor Masahiko Murakawa 2-7-5, Chiyugaoka, Aso-ku, Kawasaki-shi, Kanagawa 5-21 Fujitsu Daini Yurigaoka Dormitory Room 421 (72) Invention Person: Mamoru Endo 2-1-1-1110 Minamidai, Sagamihara City, Kanagawa Prefecture Go 6-15 (72) Inventor Katsuo Sato 1-1-2 Minami Naruse, Machida City, Tokyo Plaza Naruse 406 (72) Inventor Akira Hashimoto 30-1-207 Matsufudai, Midori-ku, Yokohama, Kanagawa Prefecture (72) Invention Person Hiroshi Araki 3-9-1 Shimonoge, Takatsu-ku, Kawasaki City, Kanagawa Prefecture Room 458, Fujitsu Shimotomo-ryo (72) Inventor Yasumasa Takahashi 6-18-20 Minamikoiwa, Edogawa-ku, Tokyo (56) References JP-A-62-248048 (JP) JP-A-1-187632 (JP, A) JP-A-2-58165 (JP, A)

Claims (9)

(57) [Claims] In a distributed database system comprising a plurality of terminals connected to a network, an arbitrary terminal is managed from among the plurality of terminals, and a table of contents of the database is managed. The data providing terminal sets the temporary data center as a changeable temporary data center and publishes the data to other terminals. The data providing terminal stores the text of the generated data in its own file, notifies the table of contents and the terminal identification number to the temporary data center, and The center registers and manages the table of contents notified from each data providing terminal together with the terminal identification number in a table of contents file. The user terminal makes a table of contents request to the temporary center when data is needed, and obtains the entire table of contents. When one is selected from the list, the user accesses the data providing terminal that provided the table of contents, and the data providing terminal corresponds to the table of contents requested by the user terminal. Temporary center system in the distributed database system and transmitting the text data retrieved and the user terminal. 2. The temporary center for managing a table of contents according to claim 1, wherein the temporary center sequentially stores a modified table of contents of a database transmitted by a user sent from a data providing terminal in a temporary work file, and for each set period. When the version of the table of contents is added to the table of contents and transferred to the table of contents management file, and a version inquiry of the table of contents is received from the user terminal in response to a request to publish the table of contents, the shortage after the latest version of the terminal notified by the terminal is notified. A temporary center method in a distributed database system, wherein a version number is taken out from the contents version number management file and transferred to a user terminal. 3. The terminal according to claim 1, wherein the terminal, when requesting the public table of contents from the temporary center, receives the entire public table of contents and each terminal identification number sent from the temporary center, A temporary center system in a distributed database system, wherein a terminal identification number of the terminal is compared, and only when the two do not match, a corresponding public table of contents is stored in a third party table of contents management file provided in the terminal. 4. In a distributed database system comprising a plurality of terminals connected to a network, an arbitrary terminal among the plurality of terminals manages commands of the database, and other arbitrary ones as required. The data providing terminal, which is set as a temporary center that can be changed to the terminal and publishes data to other terminals, stores the text of the generated data in its own file and notifies the temporary center of its command and terminal identification number. The temporary center registers and manages the command and terminal identification number notified from each data providing terminal in a table, and the user terminal makes a command request to the temporary center when data is needed, and obtains the obtained terminal. A temporary center in a distributed database system characterized by accessing a data providing terminal using an identification number formula. 5. A command according to claim 4, wherein said temporary center for command management is provided with a common command assigned to a related data group held in a database of one or more terminals, and a command of said data group. A table is provided in which terminal identification numbers for retaining data are registered, and terminals using commands obtain terminal identification numbers for retaining data by inquiring of the temporary center.
A temporary center system in a distributed database system, wherein a corresponding data group is obtained using the command. 6. The terminal according to claim 4, further comprising: means for allocating a command expressing data held in the terminal, and matching different commands as necessary, wherein the information on the different matched commands is provided. The command link center that manages the command link is set as a temporary center, and in the processing of one command in one terminal, the terminal inquires of the temporary center of the command link whether or not there is a command matched to one command. A temporary center method in a distributed database system, wherein the terminal processes actual data represented by one command and a command matched with the command based on a response from the temporary center. 7. A temporary center for forming a group by a plurality of predetermined terminals among terminals connected to a network and managing a database distributed and held in said group. The temporary center searches for a database that is distributed and held based on a request from a terminal of the group, and the terminal that has made a request searches the database that is distributed and held based on the search result. A temporary center method in a distributed database system characterized by referring to. 8. The system according to claim 7, wherein a plurality of groups each including the predetermined terminal are provided, and a temporary center for managing a database distributed and held in each group is provided for each group. An inter-group management terminal that manages a database that is distributed and maintained, wherein each of the temporary centers searches for a database in the group based on a request from a terminal in the group to which the temporary center belongs; Based on a request from each temporary center, a search is made for a database following a group other than the group to which the temporary center belongs, and the requesting terminal is the group to which the terminal belongs based on the search result by the inter-group management terminal. Dispersion characterized by referring to databases of groups other than Provisional Center method in the database system. 9. The terminal according to claim 1, wherein the terminal includes a management information processing unit that processes a predetermined item of management information for managing a database, and a management information file that stores the predetermined item. After the terminal holds information about the item of the management information, and the management information processing unit issues a declaration to perform management of the database for the predetermined item to another terminal, the plurality of terminals transmit the predetermined item from the plurality of terminals. A temporary center system in a distributed database system, wherein the temporary center system collects and stores the collected information in the management information file and manages the database for the predetermined item using the management information file.