JPH06266761A - Decentralized computer system - Google Patents

Abstract

PURPOSE:To provide a system which facilitates an access procedure and the exchange of data and is rich in extendability in a computer system composing a data base. CONSTITUTION:Data composing a data base as a whole is distributively stored in plural computers A, B an C connected via a network. Each of computers A, B and C stores at least a specific table and a distributed control table. In accordance with the designation of a prescribed key item related to the data of the specific table, stored corresponded data is retrieved in the computer corresponded to the designated key item, and the retrieval of corresponded data is requested to the computer corresponded to the designated key item in other computers.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distributed computer system in which a database is managed by a plurality of computers connected via a network, and in particular, a terminal device connected to one computer is stored in another computer. The present invention relates to a distributed computer system capable of retrieving stored data.

[0002]

2. Description of the Related Art As means for utilizing a database built in a computer, there is a system in which a plurality of computers are connected via a network and the databases are managed by the plurality of computers. In such a system, when using the database from a terminal device connected to a computer, it is possible to access not only the computer to which the terminal device is connected but also another computer connected via a network. it can.

Conventionally, in this type of system, the database of each computer is individually closed.
Each was independent. Therefore, each computer's database is managed by its own management system, and in order to access a specific computer from the terminal device of another computer, the search method (search formula, keyword) guaranteed by the management system of the computer. Etc.) was being searched.

[0004]

However, in the above conventional system, when accessing a computer other than the computer to which the terminal device is connected,
There is a drawback in that the access procedure becomes complicated because the search method by the database management system of the computer to be accessed must be followed.

Further, since the database is closed for each computer, it is difficult to exchange data between the computers that make up the system, and the system is expanded by adding a new computer to the network. There was a drawback that it was troublesome to do.

The present invention solves the above-mentioned conventional drawbacks, simplifies an access procedure to a computer other than a computer to which a terminal device is connected in managing a database, and saves data between computers constituting a system. An object of the present invention is to provide a computer system which is easy to replace and has high expandability.

[0007]

In order to achieve the above object, the present invention provides a plurality of computers connected via a network with data constituting one database as a whole, if necessary, for each computer. The above computers are at least,
A unique table made up of data unique to each computer that constitutes the database, and a distributed management table that manages the unique table of all computers connected to the network are stored, and the data of the unique table of the computer is: The entire table or each record in the table is associated with the key item associated with the computer that stores the data, and in the computer associated with the specified key item according to the designation of the key item, It is characterized in that the stored corresponding data is searched, and in the other computers, the computer associated with the specified key item is requested to search for the corresponding data.

[0008]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a distributed computer system according to an embodiment of the present invention.

In this embodiment, as shown in the figure, three computers (hereinafter referred to as "server machines" in the present embodiment) A, B, C connected by a wide area network (WAN).
And the local area network (L
An) connected terminal devices A1, A2, ... An
And a terminal device B connected to the server machine B by LAN.
, Bn and terminal devices C1, C2, ... Cn connected to the server machine C via a LAN.

Each of the server machines A, B and C has a configuration as shown in FIG. That is, a WAN communication device 1 for connecting to a WAN for exchanging commands and data, and a LAN communication device 2 for connecting to a LAN and receiving a search request from a terminal device or transmitting a search result.
And a storage device 3 that stores data that configures a database, and a CPU 4 that controls each of these units.

The storage device 3 of each of the server machines A, B and C stores data consisting of a plurality of tables. The tables stored in the server machines A, B, and C include a unique table storing data unique to each server machine and a distributed management table for managing the data of the unique tables of all the server machines.

The data stored in the unique tables of the server machines A, B, and C constitute one database as a whole, and are associated with a predetermined key item for each table or each record. Note that the server machines A, B, and C may be provided with a common table that stores data common to each server machine. In this case, the data stored in the common table also corresponds to the configuration data of the database. Become.

In the distributed management table of the server machines A, B and C, the key item of the unique table is associated with the server machine storing the table or record specified by the key item.

Examples of the unique tables stored in the server machines A, B and C are shown in FIGS. 5 is a unique table a stored in the server machine A, FIG. 6 is a unique table b stored in the server machine B, and FIG. 7 is a unique table c stored in the server machine C. In this example, the key item is provided for each record in the unique table.

Each of the peculiar tables a, b, and c is composed of a record consisting of predetermined fields, and at least a field indicating the type of the stored document is provided. That is, according to the field name “document type” in FIGS. 5 to 7, the document “product standard” and “application form” are stored in the specific table a, and the “material standard” and “production” are stored in the specific table b. The document about “method” and the document about “analysis method” are stored in the unique table c. In this example, such document type is used as a key item.

Next, FIG. 8 shows an example of the distributed management table commonly stored in the server machines A, B and C. The distributed management table x stores, for each record, the document type and the name of the server machine storing the document in association with each other. Therefore, if the document type is designated when searching the data from the terminal device, the server machine storing the document corresponding to the designated document type can be easily detected.

Next, the operation of this embodiment will be described.
The operation of the data retrieval from the terminal device A1 of the server machine A is shown in the flowchart of FIG. First, the terminal device A1 issues a search command requesting the server machine A to search for desired data (step S1).

The server machine A creates a search command according to the received search command (step S2) and refers to the document management table x in FIG. 8 to specify the corresponding server machine from the key item (step S3). If the server machine specified in step S3 is server machine A, CPU
4 retrieves the data requested for retrieval from the storage unit 3 of the server machine A and transmits it to the terminal device A1 (step S
8, S9).

On the other hand, when the server machine specified in step S3 is not the server machine A, the CPU 4 issues the search command created in step S2 to the specified server machine (step S5). The server machine receiving the search command searches for the data requested to be searched and returns the result to the server machine A (steps S6, S).
7). Then, the server machine A transmits the received search result to the terminal device A1 (step S9).

Here, as a specific search example, a case of searching for “analysis method of hydrogen ion concentration” from the terminal device A1 will be described with reference to the tables of FIGS. First, the terminal device A1 issues a search command {select, analysis method, KANRI, symbol = PH, document type = analysis method}. The tables of the unique tables a, b, and c stored in the server machines A, B, and C, respectively.
A common name “KANRI” is given as a bull name.

In response to this, the server machine A creates a search command: select from KANRI where symbol = PH and document type = analysis method. Therefore, it can be seen from the key item “Document type”, “Analysis method” that the data is in the server machine C. Then, since the requested data is in the server machine C, the server machine A sends the search command to the server machine C.

The server machine C retrieves the record 3, which is the corresponding data, from the unique table c and sends it to the server machine A. Then, the data received by the server machine A is transmitted to the terminal device A1. This is the end of the search for “analysis method of hydrogen ion concentration”.

Next, the operation for transferring data between server machines is shown in the flowchart of FIG.
First, the data to be transferred is extracted from the unique table of the server machine (step S1). Then, the extracted data is transmitted to the destination server machine and inserted into the unique table of the server machine (step S2).
Thereafter, the transferred data is deleted from the original unique table (step S3), and the server machine corresponding to the key item of the distributed management table is changed (step S4). It goes without saying that the change of the distributed management table is performed in all the server machines.

Here, as a specific modification, a case where the data of the "manufacturing method" is transferred from the server machine B to the server machine C will be described. Since the data of the “manufacturing method” exists in the unique table b of the server machine B, the data is extracted from this and inserted into the unique table c of the server machine C. Next, the data of “manufacturing method” is deleted from the unique table b. Finally, the server machine corresponding to the "manufacturing method" in the distributed management table of all the server machines A, B, and C is changed to C. This is the end of the movement of the "manufacturing method" data.

Next, an example in which a key item is provided for each unique table will be described. 9 to 13 show examples of the unique table, and FIGS. 14 and 15 show examples of the distribution management table. 9 and 10 are unique tables a1 and a2 stored in the server machine A, FIGS. 11 and 12 are unique tables b1 and b2 stored in the server machine B, and FIG. 13 is unique table stored in the server machine C. It is c1. The document type is used as the key item, as in the above example. However, in this example, since the key items are associated with each table, each table includes only data of the same document type. The distributed management table includes a table x in which the key items shown in FIG. 14 are associated with the server machine, and a table y in which the key items are associated with the unique table shown in FIG. In this example, the table name of the table a1 is "SEIHIN", the table name of the table a2 is "SINSEI", and the table b.
The table name of 1 is "GENRYOU", table b2
The table name of the table is "SEIZOU", and the table name of the table c1 is "BUNSEKI".

The processing procedure for data retrieval and data movement is no different from the above-mentioned example. However, in the data search, the unique tables corresponding to the key items are specified by referring to the distributed management tables x and y, respectively. In moving data, each unique table is moved from one server machine to another server machine. For example, when the data of “manufacturing method” is moved from the server machine B to the server machine C, the unique table b
1 is moved to the server machine C, and the unique table c
It will be 2. In this way, by creating a unique table for each document type and associating key items with each unique table, it is possible to set a field consisting of unique items for each unique table.

[0027]

As described above, in the distributed computer system of the present invention, data constituting one database as a whole is supplied to a plurality of computers connected via a network for each computer as necessary. Since it is stored in a distributed manner and managed by a single management system, the search method (search formula, keyword, etc.) can be changed even when a specific computer is accessed from the terminal device of another computer. There is an effect that you can easily search without.

Further, since one database is constructed by the entire computers connected through the network,
There is an effect that data can be easily exchanged between the computers constituting the system. Further, all the processing programs placed on each server machine and terminal device can be the same program, and it is not necessary to place a data-dependent program on each server machine or terminal device. .

In addition, the system is highly expandable, and new computers can be easily added to the network as the number of data and users increases. Can be expanded and the system can be easily introduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a distributed computer system according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a server machine shown in FIG.

FIG. 3 is a flowchart showing a data search processing procedure according to the present embodiment.

FIG. 4 is a flowchart showing a data transfer processing procedure according to the present embodiment.

FIG. 5 is a diagram showing a unique table.

FIG. 6 is a diagram showing a unique table.

FIG. 7 is a diagram showing a unique table.

FIG. 8 is a diagram showing a distributed management table.

FIG. 9 is a diagram showing a unique table.

FIG. 10 is a diagram showing a unique table.

FIG. 11 is a diagram showing a unique table.

FIG. 12 is a diagram showing a unique table.

FIG. 13 is a diagram showing a unique table.

FIG. 14 is a diagram showing a distributed management table.

FIG. 15 is a diagram showing a distributed management table.

[Explanation of sign]

 A server machine (computer) B server machine (computer) a server table (computer) a unique table a1 unique table a2 unique table b unique table b1 unique table b2 unique table c unique table c1 unique table x distributed management table y distributed management table

Front Page Continuation (72) Inventor Akito Kobayashi 1705-6 Miyagawa, Hikari-cho, Sago-gun, Chiba Prefecture

Claims (2)

[Claims] 1. A plurality of computers connected via a network stores data constituting one database as a whole by distributing the data for each computer as needed, and each computer is at least: A unique table made up of data unique to each computer that constitutes the database, and a distributed management table that manages the unique table of all computers connected to the network are stored, and the data of the unique table of the computer is The entire table or each record in the table is associated with the key item associated with the computer that stores the data. In the computer associated with the specified key item according to the designation of the key item, The stored corresponding data is searched and other computers are searched. In the computer, the distributed computer system is characterized in that the computer associated with the specified key item is requested to retrieve the corresponding data. 2. A plurality of computers connected via a network store data constituting a database as a whole by distributing the data for each computer as needed, and each computer stores the database. A common table composed of data common to all computers and a unique table composed of data unique to each computer,
A distributed management table that manages a unique table of all computers connected to the network is stored, and the data of the unique table of the computer is stored in the computer that stores the data for the entire table or for each record in the table. Corresponding to the corresponding key item, in response to the designation of the above key item, the computer corresponding to the designated key item retrieves the stored correspondence data, and the other computers specify the above designated item. A distributed computer system characterized in that a computer associated with a key item is requested to retrieve corresponding data.