JPH03103942A - Distributed data base management system - Google Patents

Abstract

PURPOSE:To access a file at a very high speed by providing a local terminal data managing means and an other terminal data managing means and placing the other terminal file, which is frequently used by a local terminal, in the local terminal. CONSTITUTION:Each terminal 102 is provided with a local terminal data managing means 103 and an other terminal data managing means 104. A local terminal directory which controls the local terminal file belonging to the local terminal as the terminal itself and its access is stored in the local terminal data managing means 103, and an other terminal directory which controls the other terminal file belonging to another arbitrary terminal and the access of other terminal files related to all other terminals is stored in the other terminal data managing means 104. Thus, the access speed is very high because the other terminal file frequently used by the local terminal is placed in the local terminal, and contents of files and directories existing in respective terminals 102 are free from contradictions because they are always updated by the inter-terminal communication function.

Description

[Detailed Description of the Invention] [Summary] Regarding a distributed database management method for managing data on a plurality of terminals existing on a network such as a LAN, the present invention aims to shorten the access time of a directory or the access time of distributed data. In a distributed database management method in which a database is distributed to multiple terminals connected to a network and each terminal can mutually use the database, the purpose is to facilitate the search of distributed data. Each terminal stores this terminal file belonging to this terminal, which is its own terminal, and this terminal directory whose access is controlled, and controls the editing operation of this terminal file or this terminal directory by the user of this terminal. In response to a request for an unfinished file from any other terminal, the terminal file is sent to the other terminal, and in response to a change notification of this terminal file from any other terminal, the corresponding terminal file is sent to the other terminal. This terminal data management means changes this terminal file and sends each data to all other terminals when the change of this terminal file or this terminal directory on this terminal is confirmed, and another terminal file belonging to any other terminal and another terminal directory that controls access to other terminal files related to all other terminals, and when a user of this terminal requests editing of any other terminal file, accesses the other terminal directory and searches for another terminal file. , If the other terminal file is not in this terminal, a request for the other terminal file is sent to the corresponding another terminal to obtain the other terminal file, and if it is in this terminal, the other terminal file is sent by the user of this terminal. Controls the editing operation of the other terminal, changes the other terminal file or another terminal directory on this terminal in response to a change notification of another terminal file or another terminal directory from any other terminal, and changes the other terminal data on this terminal. and separate terminal data management means for transmitting the data to all other terminals when the change is confirmed.

[Industrial application field]

The present invention relates to a distributed data management system for managing data on a plurality of terminals existing on a network such as a LAN.

[Conventional technology]

In the recent highly information-oriented society, there is a large amount of data that is related in content but physically distributed, and there is a need for an environment in which this distributed data can be accessed in real time. . For this reason, fax and e-mail are used, but it is often difficult to know where the desired data is located. Furthermore, even if the location of the data is known, in order to obtain the data, humans must intervene and perform cumbersome tasks. Therefore, it has become necessary to clarify the location of dispersed data and to be able to acquire that data quickly and automatically.

In conventional distributed data storage, a system is adopted in which a single device for centrally managing directories is provided on the network, and actual data is distributed on each terminal. Furthermore, data can be searched using a directory search method or a keyword (sort key) search method.

[Problem to be solved by the invention]

However, in the case of the conventional example described above, when an attempt is made to search for necessary information from directories or keywords, the search requests are concentrated on the device that centrally manages the directories on the network, and the requests are sent to the network. This creates a problem in that the delay time issued and contention for access to the device frequently occur, and the user who made the request is forced to wait for a long time.

In addition, since data is searched from databases distributed on the network based on directories or keywords, it may take time to search for the data that the user truly requests, or it may be difficult to search for the data that the user truly requests. Information may not be found.

Furthermore, since all data is distributed to each terminal on the network, if you want to search for data that is not on your own terminal, you must access the database of the terminal that has that data through the network, which is actually difficult. The problem is that it takes time to access the data.

On the other hand, to avoid this, it is possible to copy the necessary data into your own database, but if the original data is changed, the original data and the copied data will not be consistent. There are problems.

An object of the present invention is to shorten the access time of a directory or the access time of distributed data, and to facilitate the search of distributed data.

[Means to solve the problem]

FIG. 1 is a block diagram of the present invention. The present invention
The present invention assumes a distributed database management method in which a data hash is distributed to a plurality of terminals 102 connected to a network 101 such as a local area network (local area network), and each terminal 102 can mutually use the respective data hase.

Each terminal 102 has a main terminal data management 103 and another terminal data management 104 stages.

The two-terminal data management stage 103 stores terminal files belonging to this terminal, which is its own terminal, and terminal directories that control access to the terminal files. First, the editing operation of the terminal file or the terminal directory by the user of the terminal is controlled. Further, in response to a request for a terminal file from any other terminal other than the own terminal, the terminal file is transmitted to the terminal. Furthermore, the corresponding terminal file on the terminal is changed in response to a change notification of the terminal file from any other terminal. Then, when the change of the terminal file or the terminal directory on the terminal is confirmed, each data is transmitted to all other terminals.

The separate terminal data management stage 104 stores a separate terminal directory that controls access to separate terminal files belonging to an arbitrary different terminal and other terminal files related to all different terminals. First, when a user of this terminal requests editing of any other terminal file, the other terminal directory is accessed to search for another terminal file. If the other terminal file does not exist in this terminal, a request for the other terminal file is sent to the corresponding other terminal to obtain the other terminal file.

Conversely, if the above-mentioned separate terminal file exists in this terminal, the editing operation of the separate terminal file by the user of this terminal is controlled.

Further, in response to a change notification of another terminal file or another terminal directory from any other terminal, the corresponding another terminal file or another terminal directory on this terminal is changed. Furthermore,
When changing the data of another terminal on this terminal is confirmed, the data is sent to all other terminals.

In addition to the above structure, in the present invention, the link management means 105
It is possible to create a structure with In this case, the Kibata unfinished file or another terminal file has logical link destination information consisting of terminal information, directory information, and file information indicating a link with other main terminal file or another terminal file related to each file. are doing.

When the user of this terminal selects arbitrary logical link destination information of this terminal file or another terminal file and makes a request to follow the link, the link management means 105 controls the link destination indicated by the logical link destination information. If this terminal is used, the selected logical link destination information will be stored in this terminal data management stage 103.
Output to.

On the other hand, if the link destination is another terminal, the selected logical link destination information is output to the other terminal data management means 104.

In this case, the terminal data management means 103 searches for the corresponding terminal file based on the logical link destination information from the link management means 105 and outputs it to the user of the terminal.

On the other hand, the separate terminal data management stage 104 searches for a corresponding separate terminal file based on the logical link destination information from the link management stage 105. If the other terminal file does not exist in this terminal, a request for the other terminal file is sent to the corresponding other terminal to obtain the other terminal file. Conversely, if it is in this terminal, it will be output to the user of this terminal.

Furthermore, in the present invention, the link management means 105, the main terminal data management stage 103, and the separate terminal data management stage 104 can be configured as shown below.

That is, when a user of this terminal inputs a request to edit arbitrary logical link destination information of this terminal file or another terminal file, the link management means edits the logical link according to the logical link destination information selected for editing. The destination information is output to this terminal data management means 103 or another terminal data management means 104.

On the other hand, this terminal data management means 103 or another terminal data management means 104 edits the logical link destination information of the corresponding this terminal file or another terminal file based on the logical link destination information from the link management means 105. do.

[For production]

When a user wants to access another terminal file belonging to another terminal other than the user's own terminal, the other terminal data management means 104 always manages directories of all other terminal files, and based on this, the other terminal can access the other terminal file belonging to another terminal other than the user's own terminal. Acquire files. In this case, if another terminal file or another terminal directory is changed on another terminal, compatriot communication is performed via the network 101, and the changed data in this terminal is managed as another terminal data. Sendan 1
Changed by 04. Conversely, if another terminal file is changed on this terminal, the other terminals will be notified through peer communication.

Furthermore, even if the user edits or otherwise changes the terminal data or terminal directory belonging to the terminal,
This is notified to other terminals through peer communication, and the terminal directories of all the other terminals or the terminal files being used in any of the terminals are changed.

With the above operation, other terminal files that are frequently used on this terminal can be kept on this terminal.
Allows very fast access.

Moreover, due to the communication function between the terminals, each file and directory 1.2 existing on each terminal 102 are always updated with the update 1.
Therefore, there will be no contradiction in content.

In addition to the above operation, if the link management 105 is provided, it is possible to easily search for other files related to this terminal file or another terminal file.
In this case as well, by passing control to the terminal data management means 103 or the separate terminal data management means 104, files and directories can be managed in the same manner as described above.

Furthermore, the logical link destination information of each file itself can be edited from the link management means 105 via the terminal data management means 103 or the separate terminal data management means 104.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings.

First, FIG. 2 is a diagram showing the overall structure of the network to which the present invention is applied. As shown in the figure, a plurality of terminals 202 are connected to a LAN (Local Area Network, hereinafter the same) 201, and a data base is distributed and arranged in each terminal 202. Further, the LANs 201 are connected to each other via an exchange 203, etc., and each terminal 202 is connected to the LAN to which it belongs.
Terminals 202 belonging to LANs other than AN 201 can also access.

FIG. 3 is a configuration diagram of the terminal 202 of FIG. 2 according to this embodiment.

The network server 301 is an interface that controls e-mail output to and from the LAN 201 .

The window surfer 302 is an interface that controls human output of various information to and from the user in a window-shaped display format.

15 16 The output management device 304 responds to requests from the file management device 308 on another machine and the data management device 306 on this machine, requests for files on other machines, or requests files on its own machine (hereinafter referred to as this machine). ) to distribute the latest version of the file on other machines.

The input management device 303 sends e-mails sent from other machines to a data management device 306 on this machine, a file management device 308 on another machine, or a directory management device 310 on another machine, depending on the type of e-mail. .

A data management device 306 on this machine edits data according to a user's request via the window server 302, and when data is changed, an output management device 304 is used to broadcast the change result to other machines. Send the information to. In addition, in accordance with the latest file request from the input management device 303, the corresponding file is extracted from the directory/file storage device 307 in this machine and sent to the output management device 304.

The file management device 30B on another machine stores only the information needed by the user (information accessed once by the user) among the latest file information sent from other machines every moment to its own separate machine file storage device 309. , and in response to a file display request sent from the directory management device 310 or link management device 305 on another machine, if that information exists in the other machine file storage device 309, the file is displayed on the window server. 30
If the file does not exist, the file is requested from the output management device 304. Also, window server 3
02, and when the file is changed, the information is sent to the output management device 304 in order to broadcast the change results to other machines.

The directory management device 310 on another machine searches for the latest directory sent from another machine, and in response to a request to access a file on another machine, the directory management device 310 manages the file on the other machine. to device 308;

When there is a request to search for related information by tracing logical links (described later) between files, the link management device 305 searches for related information when the related information is original data on this machine. Data management device 30 on this machine
If not, a request is sent to the file management device 308 on another machine.

FIG. 4 shows the storage contents of the directory/file storage device 307 in this machine. As shown in the figure, directory 1
, 2, 3, . . . , and the actual file bodies l, 2, . . . exist at the end of each tree. The contents of the file body will be described later.

FIG. 5 shows the storage contents of the separate machine file storage device 309. As shown in the figure, a tree of directories showing the machine name, file name, and storage location on the machine (in the case of the figure, in the directory ij in the directory 22 in the directory ll...in the directory nm) ), and the main body of the file. The contents of this file body will also be described later.

FIG. 6 shows the storage contents of the separate machine directory storage device 311. As shown in the figure, for each machine, machine name 1,
. . , 11 . . . A directory tree is stored in a frame format starting with n. Furthermore, at the end of the tree,
Unlike the case in FIG. 4, there is a file name instead of the file body.

FIG. 7 shows the contents of the file body in FIG. 4 or 5. As shown in the figure, information consisting of a machine name, directory name, and file name is added at the beginning as logical link destination information to link the file to other files related to it. Note that the linked file may be on this machine or on another machine. Further, one file can be linked to a plurality of files, and each link has logical link destination information as shown in the figure. The actual contents of the file are stored next to the logical link destination information.

On the other hand, FIG. 8 shows the format of e-mail that passes through the LAN 201 (network) 19 20 . First, the general format consists of a destination, a source, a type, an information source machine name, and an information body, as shown in FIG. 2(a). Under this format, as mail format 1, as shown in the same figure (b), all terminals (indicates broadcast) as the destination, the sender's machine name, the type as the directory change, and the sender's own machine as the information source machine name. name (same as the originating machine name) and has directory information as the body. In addition, as mail format 2, the same figure (
As shown in C), the destination is all terminals, the source machine name, the type is the file change, and the information source machine name is the name of the original copy source machine (yours or someone else) where the file was created. , has the directory information (sometimes absent) and the file body as the main body. Furthermore, as mail format 3, as shown in FIG. 3(d), the name of the machine requesting the latest file as the destination, the name of the sender machine, the name of the requesting machine as the information source machine name, and the directory information regarding the file requested as the main body. and file name.

The operation of the embodiment shown in FIGS. 2 to 8 having the above configuration will be explained according to the operation flowcharts shown in FIGS. 9 to 14. Here, FIG. 9 is an operation flowchart of the output management device 304, FIG. 10 is an operation flowchart of the input management device 303, and FIG. 11 is an operation flowchart of the data management device 306 on this machine.
12 is an operation flowchart of the file management device 30B on another machine, FIG. 13 is an operation flowchart of the directory management device 310 on another machine, and FIG. 14 is an operation flowchart of the link management device 305.

The terminal 202 in FIG. 3 performs processing corresponding to each of the 11 types of operations broadly classified as listed below.

(1) When a user requests to edit a file on this machine.

(2) When a user requests to save a file on this machine.

(2) 1■ In case of directory change.

21 22 (2) 1■ In case of file change.

(3) When information about files on this machine is input from another machine.

(3) In the case of a request for the latest file.

(3) 1■ In the case of file changes.

(4) When a user requests to edit a file on another machine on this machine.

(5) When a user requests to save a file on another machine.

(6) When change data for a file on a machine other than this machine is input from another machine.

(6) If there is an older version of the file.

(6) 1) If there is no previous version of the file and it is requested by this machine.

(6) 1) If there is no previous version of the file and it is not the one requested by this machine.

(7) When a user requests a search for a directory on another machine.

(8) When a user requests a file from another machine.

(8) 1) If the requested file exists on this machine.

(8) - ■When the requested file does not exist on this machine.

(9) When the directory change data of a machine other than this machine is manually changed from another machine.

00) When a user requests to follow a link.

(10) - ■ When the link destination is this machine.

00) 1 ■ When the link destination is another machine.

00) 1■1 (a) When the linked file exists on this machine.

00) 1■1 (b) When the linked file does not exist on this machine.

(11) When a user requests to edit a link.

(11) 1 ■ When the editing target is this machine.

(11) 1) When the editing target is on another machine.

The operations in each of the above cases (1) to (I1) will be sequentially explained using the operation flowchart 1 in FIGS. 9 to 14.

23 24 First, as the above-mentioned operation (1), a case where a user requests to edit a file on this machine will be described.

A user makes a request to edit a file on this machine via the window surfer 302 in FIG. 3, and this request is sent to the data management device 306 on this machine.

As a result, the determination of 3 1101 in FIG. 11 is NO, S
The determination in S1105 is also NO, the determination in S1107 is YES, and the process proceeds to 31108.

S]. 10B, in accordance with the user's request, the corresponding file existing in the internal directory/file storage device 307 of this machine is edited under the editing environment provided by the window server 302.

Next, as the above-mentioned operation (2), a case where a user requests to save a file on this machine will be described.

The user inputs a request to save the file being edited etc. via the window server 302, and this request is input to the data management device 306 on the machine shown in FIG.

As a result, the determination in S1101 becomes NO, the determination in S1105 becomes YES, and the process proceeds to S1106.

In S1106, the file requested to be saved is saved in the directory/file storage device 307 in this machine, and the contents of the file and directory information (see FIG. 4) are saved.
(see FIG. 7) is sent to the output management device 304.

In the output management device 304, in the operation flowchart of FIG. 9, the determination in S901 is YES, and the process proceeds to S902.

Here, if the directory has been changed in response to the operation in case (2)-1 above, the process proceeds to S903, and a directory change mail is sent according to mail format 1 in FIG. 8(b). The data is generated and sent to network server 301 and broadcast to all other terminals 202 (see FIG. 2) via LAN 201.

In addition, if the file has been changed in response to the operation in case (2)-1 above, the process proceeds to step S904, and a file change email is sent according to mail format 2 in Figure 8(C). data is created and sent to the network server 30.
1 and sent to all other terminals 20 via LAN 201.
2 (see Figure 2).

In reality, directory changes and file changes are often performed at the same time, so S901 to S903 in Figure 9
The loop of and the loop of S901 to S904 are executed continuously, and two mails for changing the directory and changing the file are successively output.

Next, as the operation (3) above, a case will be described in which information regarding a file on this machine is input from another machine.

When information regarding files on this machine is input from the terminal 202 of another machine via the LAN 201 and the network server 301, that information is input to the human resource management device 303.

As a result, the determination at 1001 in FIG. 10 becomes YES, and the process proceeds to S1002.

Now, corresponding to the operation in case (3) 1) described above, if a request for the latest file is input according to mail format 3 in FIG. 8(d), the determination in S1002 is NO, and
The determination in step 4 becomes YES, and the process advances to step S1005.

In step S1005, the directory information and file name included in the email are sent to the data management device 306 on this machine.
send to

As a result, in FIG. 11, the determination in S-1101 becomes YES, the determination in S1102 becomes YES, and S1
The process proceeds to step 103.

In step 51103, the corresponding file is extracted from the internal directory/file storage device 307 of this machine based on the sent directory information and file name, and the information is sent to the output management device 304.

As a result, in the output management device 304, in FIG.
The determination in S901 is YES, and the determination in S902 is NO.
Then, the process advances to step S904.

In S904, according to email format 2 in FIG. 8(C),
The latest file is prepared as file change mail data, sent to the network server 301 27 28 , and sent to the sender's terminal 202 via the LAN 201 . In this case, directory information is not added in mail format 2 shown in FIG. 8 (0).

On the other hand, in response to the operation in case (3)-1 above, the eighth
If a file change is input according to email format 2 in Figure (C), the determination in S1002 is No, S1004
The judgment of 31006 is No, the judgment of 31006 is YES, and S
The process advances to step 1007.

In S1007, the directory information and file contents (main body) included in the email are sent to the data management device 306 on this machine.

As a result, in FIG. 11, the determination of 31101 is Y.
The determination of ES, 31102 is No, and the process proceeds to 51104.

In S1104, the directory/file storage device 307 in this machine is stored based on the sent directory information.
and rewrites its memory contents with the contents of the sent file.

Next, as the above-mentioned operation (4), a case where a user requests editing of a file on another machine that is on this machine will be explained.

A user inputs a request to edit a file on another machine on this machine via the window server 302 in FIG. 3, and this request is manually input to the file management device 308 on the other machine.

As a result, the determination of 3 1201 in FIG. 12 is NO, S
The determination in 1206 is also No, the determination in 3120B is YES, and the process advances to S1209.

In S1209, the corresponding file existing in the separate machine file storage device 309 is edited under the editing environment provided by the window server 302 according to the user's request.

Next, as the above-mentioned operation (5), a case where a user requests to save a file on another machine on this machine will be explained.

The user inputs a request to save a file that was being edited on another machine via the window server 302, and this request is input to the file management device 308 on the other machine.

29 30 As a result, in FIG. 12, the determination in S1201 is No, the determination in S1206 is YES, and S120
Proceed to step 7.

In S1207, the file requested to be saved is saved in the separate machine file storage device 309, and the contents of the file and directory information (see FIG. 5) are sent to the output management device 304.

In the output management device 304, in the operation flowchart of FIG. 9, the determination in S901 is NO, and the determination in S905 is YES. ．． The determination in S906 is YES, and the process advances to step 3904.

In S904, as described above, file change mail data is created according to mail format 2 in FIG. 8(C), and sent to the network server 301.
to all other terminals 202 (see FIG. 2). In this embodiment, the directory of another machine cannot be changed, so no directory information is added in mail format 2 shown in FIG. 8(C).

Next, as the above-mentioned operation (6), a case will be described in which changed data of a file of a machine other than the present machine is manually input from another machine.

When change data for a file on a machine other than this machine is input from the terminal 202 of another machine via the LAN 201 and network server 301 in accordance with e-mail format 2 shown in FIG. input.

As a result, the determination of 1001 in FIG. 10 is YES, S
Each of the determinations in S1002, S1004, and S1006 becomes NO, and the process proceeds to S1008.

3 In step 1008, the directory information and file contents (main body) included in the email are sent to the file management device 308 on another machine.

As a result, in FIG. 12, the determination in S1201 becomes YES, and the process proceeds to S1202, where the separate machine file storage device 309 is accessed based on the sent directory information.

As a result, corresponding to (6)
The determination at step 1202 is YES, and the process advances to step S1203. In S1203, the old version of the file on the separate machine file storage device 309 is replaced with the new file sent.

On the other hand, if the old Harsion file corresponding to the changed file does not exist in the separate machine file storage device 309, the determination in S1202 is YES, and S12
Proceed to process 04.

Then, in response to the above-mentioned (6)-1, if the new file sent is the one requested by this machine, the file is stored in the separate machine file storage device 309 as a new file.

Conversely, in accordance with (6) 1) above, if the new file sent is not the one requested by this machine, no processing will be done.

Next, as the above-mentioned operation (7), a case where a user requests a search for a directory on another machine will be described.

A user manually issues a search request for a directory on another machine via the window server 302 in FIG. 3, and this request is input into the directory management device 310 on the other machine.

As a result, in FIG. 13, the determination in S1301 is N
O, the determination in S 1303 is YES, the determination in S 1304 is YES, and the process proceeds to step 3 1305.

In S1305, the directory name requested by the user is extracted from the separate machine directory storage device 311 (
(see FIG. 6) and is displayed via a window screen 302.

Next, as the above-mentioned operation (8), a case where a user requests a file from another machine will be described.

A user manually requests a file on another machine using directory information and a file name via the window server 302 in FIG. 3, and this request is sent to the directory management device 310 on the other machine.

As a result, in FIG. 13, the determination in S1301 is N
O, the determination in S 1303 is YES, the determination in S 1304 is NO, and the process proceeds to step 3 1306.

33 34 In S1306, the directory information and file name input by the user are sent to the file management device 308 on another machine.

As a result, in FIG. 12, S 1201, 312
06, S1208, and S L210 are NO, and S1217 is YES, and 3121
Proceeding to process B, another machine file storage device 309 is accessed based on the input directory information.

As a result, corresponding to (8) 1 above, if the requested file from another machine exists on this machine, 3.
The determination at step 121B is YES, and the process advances to step S1219. Here, the file is extracted from a database on another machine file storage device 309 and passed to the window server 302.

On the other hand, if the requested file from another machine does not exist on this machine, corresponding to (8) 1 above, 3 1
The determination in step 21B is No, and the process advances to S1220.

Here, the latest version of the file is requested from the output management device 304.

As a result, in FIG. 9, the determination in S901 is NO,
The determination in S905 is YES. ．． The determination in S906 is NO, and the process advances to S907.

In S907, according to email format 3 in FIG. 8(d),
An email requesting the latest file is created in the form of directory information and file name, and sent to the network server 301.
and is sent to the corresponding terminal 202 via the LAN 201.

As a result, the requested file can be imported into the separate machine file storage device 309 as a new file as a process corresponding to the above-mentioned (6)-1.
8) Window server 302 by processing corresponding to
passed to.

Next, as the operation (9) above, a case will be described in which data for changing the directory of a machine other than the present machine is input from another machine.

When a directory change is manually performed from the terminal 202 of another machine via the LAN 201 and the network server 301 according to the mail format 1 shown in FIG.

As a result, the determination of SIOOI in FIG. 10 becomes YES and 35 36 , and the determination of S 1002 becomes YES and S 10
Proceed to process 03. Here, the recipient's machine name and directory information included in the email are sent to the directory management device 310 on another machine.

As a result, in FIG. 13, the determination in S1301 becomes YES, and the process proceeds to S1302. Here, from the sent machine name of the other party, find the same machine name in another machine directory storage device 311 (see Figure 6) and replace that directory information with the sent directory information, or If it does not exist, add a new one.

Next, as the above-mentioned On) operation, a case where a user requests to follow a link will be described.

As mentioned above, the file body stored in the internal directory/file storage device 307 of this machine or the separate machine file storage device 309 in the format shown in FIG. 4 or FIG.
Logical link destination information is stored as shown in FIG. 7, and the user can search for a desired file by following this information. Specifically, a link destination is displayed for each file via the window server 302,
By selecting any link from these, the user can
You can follow the link. This request (selection) is then input to the link management device 305.

As a result, in FIG. 14, S1401 and S1
When both determinations in step 402 are YES, the process proceeds to step S1403, where logical link destination information (see FIG. 7) selected by the user is determined.

As a result, corresponding to the above-mentioned 00-■, if the link destination is this machine (which can be determined by the machine name in Figure 7), S
The determination in S1403 is YES, and the process advances to S1404.

In S1404, the directory information and file information (directory name and file name in FIG. 7) of the link destination selected by the user are sent to the data management device 306 on this machine, and a file search request is issued.

As a result, in FIG. 11, SIIOI, 3110
5 and S1107 are No, and S1109
37 38 The judgment is YES, and furthermore, the judgment of SIIIO is YES.
Then, the process proceeds to S1111.

In step S1111, the file determined by the sent directory name and file name is extracted from the directory/file storage device 307 in this machine and displayed via the window server 302.

On the other hand, corresponding to the above-mentioned 00)-1, if the link destination selected by the user is another machine, the judgment in the above-mentioned 31403 becomes NO, and the process proceeds to S1405.

In S1405, the machine name, directory name, and file name selected as logical link destination information are sent to the file management device 308 on another machine, and a file search request is issued.

As a result, in FIG. 12, S L201, S12
06 and S 1208 are NO, and S 12
The determination in S10 is YES, and the determination in S1211 is NO, and the process advances to S1213.

The S L213 uses the machine name, directory name, and file name sent to another machine file storage device 309.
Search for.

As a result, corresponding to the above aω1■1 (a), if the linked file exists on this machine, S1214
If the determination is YES, the process proceeds to S1215,
The file is displayed via window surfer 302.

Conversely, corresponding to 00)-■1(b) above, if the linked file does not exist on this machine, S1214
If the determination is NO, the process advances to step S1216, and a request to acquire the corresponding file is issued to the output management device 304.

As a result, in FIG. 9, S901 → S 905 → S 906 → S 9 as in the case of (8) 1) described above.
07 and follow email format 3 in Figure 8(d).
A file request email was created and Network Saha 3
The terminal 202 corresponding to the machine name sent to 01 and selected as logical link destination information via LAN201
sent to. As a result, the above (6)-1 or (8)
The linked file can be obtained in the same way as in case 1).

Finally, as the operation (11) mentioned above, the user
A case in which there is a request to edit 9 40 links will be explained.

Editing a link refers to a process of deleting, adding, or changing logical link destination information in the file body shown in FIG.

Edit requests manually entered by the user via the window surfer 302 are input to the link management device 305 .

As a result, in FIG. 14, the determination in S1401 is YES, the determination in S1402 is NO, and S140
The process proceeds to step 6, where the logical link destination information selected by the user is determined.

As a result, corresponding to item (11) above, if the link destination to be edited is this machine (can be determined by the machine name in Figure 7), the determination in S1406 is NO, and S
The process advances to step 1408.

In S1408, the machine name (main machine name), directory name, file name, and editing content, which are the logical link destination information selected to be edited, are sent to the data management device 306 on the main machine.

As a result, in FIG. 11, S 1101, SLl
05 and S1107 are No, and Sl109
(7) The determination is YES, and the determination of SIIIO is NO, and the process advances to S1112.

In S1112, the logical link destination information of the corresponding file is modified.

Conversely, corresponding to (11)-■ above, if the link destination to be edited is another machine, the determination in S1406 is YES and the process proceeds to S1407.

In S1407, the machine name, directory name, file name, and editing contents, which are the logical link destination information selected for editing, are transferred to the file management device 308 on another machine.
send to

As a result, in FIG. 12, S 1201, 312
06 and S 1208117) Each judgment is NO and S
1210 judgment is YES, further S1211 judgment is Y
ES, the process proceeds to S1212.

In S1212, the logical link destination information of the corresponding file is modified.

41 - 42 [Effects of the Invention] According to the present invention, other terminal files that are frequently used in the present terminal can be stored in the present terminal, and therefore can be accessed very quickly.

Moreover, each file and directory existing on each terminal is constantly updated by the communication function between the terminals, so there are no conflicts in content.

In addition, by controlling the logical link destination information added to each file with the link management means, it is easy to search for other related files from one terminal file or another terminal file. It is also possible to perform the same file control as above.

Furthermore, the link relationship between files can be easily edited by allowing the logical link destination information of each file to be edited by the terminal data management means or another terminal data management means via the link management means. becomes possible.

The combination of the above effects enables fast and easy access to distributed databases.

[Brief explanation of drawings]

Fig. 1 is a block diagram of the present invention, Fig. 2 is an overall configuration diagram of the network, Fig. 3 is a configuration diagram of each terminal according to this embodiment, and Fig. 4 is a directory/file storage device in this machine. 307, FIG. 5 is a diagram showing the contents of another machine file storage device 309, FIG. 6 is a diagram showing the contents of another machine directory storage device 311, and FIG. 7 is a diagram showing the contents of another machine directory storage device 311. A diagram showing the contents of the main body, Figure 8 (a
) to (d) are diagrams showing the format of mail passing through the network, FIG. 9 is an operation flowchart of the output management device 304, FIG. 10 is an operation flowchart of the human power management device 303, and FIG. 12(a) and 12(b) are operation flowcharts of the data management device 3043 44 6 on this machine, FIG. 12(a) and (b) are operation flowcharts of the file management device 308 on another machine, and FIG. 13 is a directory management device on another machine. FIG. 14 is an operation flowchart of the link management device 305. 101...Network, 102...Terminal, 103...This terminal data management means, 104...Another terminal data management means, 105...Link management means.

Claims (1)

[Claims] 1) A plurality of terminals (
In a distributed database management method in which a database is distributed to multiple terminals (102) and each of the terminals (102) can mutually use the database, each of the terminals (102) includes: This terminal file belonging to this terminal which is its own terminal. and the terminal directory whose access is controlled, control the editing operation of the terminal file or the terminal directory by the user of the terminal, and request the terminal file from any other terminal other than the own terminal. , the terminal file is sent to the other terminal, the corresponding terminal file on the terminal is changed in response to a change notification of the terminal file from any of the other terminals, and the terminal file on the terminal is changed. This terminal data management means (103) transmits the respective data to all of the other terminals when a change in the said terminal file or this terminal directory is confirmed; Stores another terminal directory that controls access to another terminal file related to another terminal, and when a user of the terminal requests editing of any of the other terminal files, accesses the other terminal directory and searches for the other terminal file. If the other terminal file is not in the terminal, a request for the other terminal file is sent to the corresponding other terminal to obtain the other terminal file, and if it is in the terminal, the user of the terminal Control the editing operation of the corresponding terminal file by
In response to a change notification of another terminal file or another terminal directory from any of the other terminals, change the corresponding another terminal file or another terminal directory on the said terminal, and confirm the change of said other terminal data on said terminal. Another terminal data management means (1) that transmits the data to all the other terminals at the same time.
04) A distributed database management method characterized by having the following. 2) The terminal file or another terminal file has logical link destination information consisting of terminal information, directory information, and file information indicating a link with the other terminal file or another terminal file related to each file. and when the user of the terminal selects any logical link destination information of the terminal file or another terminal file and requests to follow the link, the link destination indicated by the logical link destination information. If the terminal is this terminal, the selected logical link destination information is output to the terminal data management means, and if the link destination is another terminal, the selected logical link destination information is output to the other terminal data management means. It has a link management means (105) for outputting, and the terminal data management means (103) searches for a corresponding terminal file based on the logical link destination information from the link management means (105) and outputs the terminal file. The other terminal data management means (104) searches for a corresponding another terminal file based on the logical link destination information from the link management means (105), and the other terminal file is output to the user of the terminal. If the terminal file does not exist in the terminal, a request for the terminal file is sent to the corresponding other terminal to acquire the file, and if it exists in the terminal, the file is output to the user of the terminal. 2. The distributed database management method according to claim 1. 3) When the user of the terminal inputs a request to edit any logical link destination information of the terminal file or another terminal file, the link management means (105) selects the logical link selected for editing. According to the destination information, the logical link destination information is output to the main terminal data management means (103) or another terminal data management means (104), and the main terminal data management means (103) or another terminal data management means (104)
) edits the logical link destination information of the corresponding main terminal file or another terminal file based on the logical link destination information from the link management means (105). A distributed database management method.