JPH0444164A - Business processing system - Google Patents

Abstract

PURPOSE:To easily distribute a data processing by performing the data processing with a computer where a ledger registering the definition of each processing element is present thereon. CONSTITUTION:A defining means 100 which causes respective computers 150 for processing of processing elements to define the processing specifications of processing elements and storage means 110, 140, and 141 which analyze the defined processing specifications with respect to all computers 150 and generate the starting relation information indicating starting relations among processing elements to store them at every starting source computer 150 are provided. Consequently, the processing of transaction or the like accompanied with input data is automatically distributively executed by each computer 150 itself which defines the processing specifications. Thus, concentration of the load to a specific computer is prevented and the data transfer quantities between computers 150 are reduced.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an office processing system that performs office processing such as transaction processing using a plurality of computers forming a network, and particularly relates to an office processing system that performs office processing such as transaction processing using a plurality of computers forming a network. The present invention relates to an office processing system that minimizes data transfer between computers during cooperative execution.

[Prior Art] Conventionally, as an office processing system that performs transaction processing using a network configured by interconnecting a plurality of electronic computers (hereinafter sometimes simply referred to as "computers"), for example, Nikkei Computer, May 23, 1988
A processing system using a "distributed database" is known, which is described in "Distributed database whose commercialization begins for market formation", pages 101 to 116 of the Japanese issue.

FIG. 2 shows a processing system using a distributed database according to this prior art. As an example, Figure 2 shows two electronic computers (workstations, hereinafter referred to as II W S I
(sometimes abbreviated as +) is connected to the system.

A user causes the workstation “WS-1” 200 to execute a transaction processing program (hereinafter sometimes abbreviated as IfTPI+) 201 using the input/output device 204 . TP201 is "WS-1" 200 that functions as a distributed database management system.
database management system (hereinafter referred to as “DBMS”)
(abbreviated as “) 202 and workstation” W S
-2'' 210 DBMS 212 is used to retrieve and store data in databases 203 and 213 that function as distributed databases as described below, and proceed with office processing.

That is, in the TP201, when performing transaction processing, in order to access the target database,
Call BMS202. The DBMS 202 checks which workstation the target database is located on, and
When the target database is database 213,
Workstation where it exists “WS-2” 21
0 and requests data reading. 98MS212 received the request via communication,
After reading the required data from the database 213,
The data is sent to the DBMS 202 via communication. DBM
The DBMS 202 that received the data from S 212 sends the received data to the TP 20. The TP 201 processes the received data, returns the processed data to the database 213, and stores it in the DBMS 202.
Issue a storage request to. DBMS20 that received a storage request
2 communicates with the 98MS 212 to issue a request to store this data. 98MS received a storage request via communication.
212 stores data in a database 213;

The TP 201 repeats the above series of operations for the number of transaction data.

As described above, in the conventional technology, since data in a database distributed to each computer is handled, data is communicated between the computers in order to read and store data.

[Problems to be Solved by the Invention] In the distributed processing system using multiple computers according to the above-described conventional technology, file search processing and update processing (not including data processing such as calculation processing) must be performed in a distributed manner. is possible, but the operation.

Data processing such as processing is performed on the starting computer, that is, the computer at the hand of the user who inputs the processing request, so if the scale of one transaction process becomes large, it is necessary to There was a problem in that the load on the computer running the system (starting computer) became large.

In addition, in transaction processing that uses a large amount of transaction data, transaction data is used to retrieve the relevant data from a database distributed on each computer, collect it on the starting computer, process it, and then write it back to the local database. Since the process has to be repeated for the same number of times, there is a problem that the amount of data transferred increases and the processing time becomes longer.

The purpose of the present invention is to solve the above-mentioned problems of the prior art, and to enable distributed processing to be performed by multiple computers, including not only file search processing and update processing, but also data processing such as arithmetic processing and processing processing. The purpose of the present invention is to provide an office processing system that can perform transaction processing while reducing the load on each computer.

Another object of the present invention is to minimize the amount of data transferred between computers when updating data by calculating and processing data distributed among computers, thereby enabling efficient processing. The goal is to provide a business processing system that provides

[Means for Solving the Problems] In order to achieve the above object, the present invention has the following features.
In an office processing system that performs business by executing processing such as transactions consisting of a series of processing elements across multiple computers, each computer that has a ledger and defines the processing specifications of the processing element. The configuration is such that the processing of the processing elements is executed in a distributed manner.

Specifically, the present invention provides a definition means for defining processing specifications of each processing element in each computer that is to perform processing of the processing element in the office processing system that performs tasks such as transaction processing that spans a plurality of computers. Then, the defined processing specifications are analyzed for all computers (constituting the network) to generate startup relationship information that indicates the startup relationship between each processing element, and the generated startup relationship information is analyzed for each startup source computer. and analysis and storage means. In addition, when a process such as a transaction is started on a certain computer, the startup light processing element and the startup destination computer that are started by the processing element of the startup source are found by searching the startup related information, and the startup light that is found is Transfer data such as transactions (processing request data) for activating a processing element to the obtained activation destination computer, and execute the required processing (for the activation light processing element) on this activation destination computer. Have the means.

Note that the computer that defines the processing specifications is also a subtraction machine that has a ledger (database table) for the processing.

[Operation] According to the above configuration, in an office processing system that performs data processing in a distributed manner using a plurality of computers connected via a network, processing such as transactions associated with input data (also a computer with a ledger), Since the processing is automatically distributed and executed by each computer that has defined the processing specifications, the load is not concentrated on any particular computer. In addition, not only record data entry and exit, but also update processing, including data calculations and processing, can be executed on each computer that has a ledger and has been defined, reducing the amount of data transferred between each computer. can.

[Example code] Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a configuration diagram of an embodiment of an office processing system to which the present invention is applied and data processing is performed in a distributed manner. In Figure 1, WS-A and WS-B are computer (workstation) A and computer (workstation) connected to the network.
workstation) B. Each computer includes a specification definition section 100a.

100b, startup relationship analysis units 110a, 110b, execution control units 120a, 120b, execution processing units 130a, 130
b, specification definition information storage units 140a, 140b,
In addition, boot-related information storage units 141a and 141b are provided. Note that 150a and 150b are input/output units having a keyboard and a display (hereinafter referred to as computer A).
, B will be explained without subscripts a and b, such as 100 and 110].

First, the process specification definition operation performed in advance will be explained. The user uses the specification definition unit 1o○ to define the specifications for each workstation, which has its own ledger, and all processing elements executed by the workstation itself (the specifics of this specification definition). (See Figure 4 below for an example). This defined specification indicates in which computer the processing element is located and from which computer it is activated, and is stored as specification definition information in the specification definition information storage unit 140 of the workstation itself.

The defined specifications are also sent to the startup relationship analysis unit 110, and startup relationship information is generated based on the analysis results here. This activation related information includes the activation source specification string (processing specification necessary to activate the relevant processing element), the own specification name (the name of the relevant processing element, that is, the activation destination specification), of the defined specification described above, 1 specification is registered (starting source computer name), it is created for each computer (starting source computer) for which the starting source specification is registered, and is stored in the starting related information storage section 141 of each starting source computer. (Refer to FIG. 5, which will be described later, for a specific example of this activation-related information).

Next, an overview of the process execution will be explained.

At a certain workstation (e.g. WS-A),
When the defined specifications (for example, "A screen processing") are instructed to the execution control unit 120 (for example, 120a), the execution control unit 120 (120a) executes the execution control unit 130 (130
Instruct a) to execute this specification "A screen processing". The execution processing unit 130 (130a) starts executing the specified specification "A screen processing". At this time, the execution control unit 120
(120a) is the activation related information 141 (141a
), search for specifications (for example, 11B Ledger Processing) related to this Specification 11A Screen Processing'', and select the computer WS in which the searched Specification 11B Ledger Processing'' is registered.
-B's execution control unit 120 (120b) has this specification 1.
The execution control unit 120 (120b) instructs the execution of the 1B ledger process ″′.
30b) is instructed to execute the specification 11B ledger process ``'' search-selected in the above relationship.Execution processing unit 130 (1
30b) starts execution of ``Specification 11B Ledger Process''.

FIG. 3 is a flowchart showing details of the operation of the execution control unit 120. Execution control unit 12o (for example, computer W
In step 301, the execution control unit 120a) of S-A
When execution of one definition specification (for example, 11A screen processing'') is instructed, the execution processing unit 130 is instructed to execute the specified specification.

The execution control unit 120 (120a) also performs step 30.
2, the startup-related information (storage section 141a) is searched, and in step 303, one piece of startup-related information regarding an unsearched processing specification is retrieved using the name of the startup source specification as a key. In step 304, the one retrieved startup-related information The starting specification name of the related information is compared with the processing specification thread for which the above instruction was given, and in step 305, if the two specification names match, that is, the processing specification for which the above instruction was given (input surface processing) is executed. If the startup-related information to be the startup source is retrieved, in step 306, the specification recorded as the startup destination specification name in the retrieved startup-related information (for example, "B ledger processing") is used as the startup destination computer. An execution instruction is given to the execution processing unit 130 (130b) of the computer (for example, WS-B) registered as the name.

In this way, steps 302 to 306 are repeated until there is no unsearched information in the startup-related information (storage unit 141a), and execution instructions are issued to all startup destination specification names that must be sequentially processed. conduct.

4(a), (b), and (c) are definition examples (specific examples of specification definition information) stored in the specification definition information storage unit 140. FIG. “Enter surface processing” 500 of the processing element is WSA
Defined by The rrB ledger processing process 51° is as follows:
WS-B, and its definition is executed in conjunction with the execution of the A screen process 500 (i.e., the entry screen process 500
(is executed as the startup specification). The C ledger process 520 is defined in the WS-C, and is defined to be executed in conjunction with the execution of the A screen process 500. The contents of the activation related information (storage section 141a) of the WS-A at this time are as shown in FIG.

FIG. 5 shows an example of the table structure of the startup-related information file stored in the startup-related information (storage section 141a), and the table structure is based on the specification thread 401 that is the startup source,
A specification thread (starting destination specification name) 402 that must be started when the specification of the starting specification thread 401 is started,
This specification 402 is made up of three fields, including a registered computer thread (startup destination computer name) 403.

FIG. 6 is a system configuration diagram for explaining an example of execution when using the example definition and the example of startup related information shown in FIGS. 4 and 5 above. Transaction processing consisting of a series of processing elements is performed by computer WS-A (1600) (7
)1-the (input/output unit 1601). First, "A screen processing" 500 is started, and the W
S-A 1600 (7) Execution control unit 1620 is 11
Execute A screen processing "500" on WS-A 1600 (7)
The execution processing unit 1630 is instructed. Next, WS-A160
From the startup related information 1641 (see Figure 4 for details) of 0,
Search for specifications starting from "A screen processing" 500 (see Figure 5 for details), and select "IIB ledger processing" 510 from WS-B26.
00 also allows IIc ledger processing ``'520 to be performed in WS-
C3600 provides that each must be activated. Therefore, the execution control unit 26 of WS-82600
Start “B ledger processing” 510 for 20, and W
5-rpc to the execution control unit 3620 of the C3600
The activation of the ledger processing process 20 is respectively instructed. The execution control unit 2620 of the WS-82600 receives the startup instruction, and
“B ledger processing” 510 for the execution processing unit 2630
Instruct the startup and execute it. Similarly, the execution control unit 3620 of the WS-C 3600, which has received the activation instruction, instructs the execution processing unit 363o to activate the PC ledger process "520" and executes it.

In this way, "A screen processing" 5 of WS-A1600
Transaction data output by starting 00 (
Processing request data) is sent to the execution control unit 262 of the WS-82600 through the execution control unit 1620 of the WS-A1600.
0 and the execution control unit 3620 of the W5-C3600. In WS-82600, the execution control unit 2620 passes the data of the processing request to the execution processing unit 2630, and the execution processing unit 2630 executes the data in the database 2660 under its umbrella according to the definition of the specification definition information 2640 through the DBMS 2650 in the same WS. Performs necessary transaction data processing on the ledger (database table). Also,
At the same time, in the WS-C3600, the execution control unit 3620 passes the data of the processing request to the execution processing unit 3630, and the execution processing unit 3630 sends the affiliated database 3660 to the database 3660 according to the definition of the specification definition information 3640 through the DBMS 3650 in the same WS. Performs the necessary transaction data processing on the ledger (database table) inside. These transaction data processes include not only simple data input and output, but also data processing and arithmetic processing. Data obtained by executing transaction data processing (processing result data) is sent to the input/output unit 1601 of the WS-A by appropriately tracing the communication path in reverse, and is displayed or printed out.

As described above, according to this embodiment, the B ledger processing 510 and the C ledger processing 520 can be performed simultaneously, so that efficiency can be further improved. In addition, processing request data is transferred from WS-A1600 to WS-B2600 and WS-B2600.
Since it is only necessary to transfer transaction data to the C3600, the time required to complete transaction processing can be further shortened and efficiency can be increased.

['! [Effect of Akira] As explained in detail above, according to the present invention, in an office processing system that executes transaction processing consisting of a series of processing elements in a distributed manner using a plurality of computers forming a network, each processing element Since data processing can be performed by a computer that has a ledger, which is a computer that has registered the definition of It is possible to do this using the following methods.

Further, since the amount of data transferred between each computer is limited to the amount of transaction data generated, processing time can be shortened and processing efficiency can be improved.

[Brief Description of the Drawings] Fig. 1 is a configuration diagram of an embodiment of an office processing system that performs data processing in a distributed manner by applying the present invention, and Fig. 2 shows an office work processing system in which data processing is performed in a distributed manner according to the prior art. FIG. 3 is a flowchart illustrating the operation of processing startup to perform distributed data processing according to the present invention; FIG. 4 is a configuration diagram of a definition example for distributing data processing according to the present invention; FIG. 5
FIG. 6 is a diagram illustrating a management data configuration of activation-related information for distributing data processing according to the present invention, and FIG. 6 is an explanatory diagram showing an operation example for distributing data processing according to the present invention. WS-A, WS-B, WS-C...Workstation, 100a, 100b...Specification definition section, 110a, 110b...Startup relationship analysis section, 1
20a, 120b...execution control unit, 130
a, 130b...Execution processing unit, 140a, 14
0b...Specification definition information storage section, 141a, 14
1b...Start-up related information storage section, 1600, 26
00.3600...Workstation, 160
1...Input/output section, 1620.2620.362
0...Execution control unit, 1630.2630.363
0...Execution processing unit, 1640.2640.36
40... Specification definition information, 1641...
Start-up related information, 2650.3650...DBMS
, 2660.3660... Database.

Claims (1)

[Claims] 1. In an office processing system that performs business by processing transactions and the like consisting of a series of processing elements across multiple computers, the processing specification of each processing element is defined as the processing of the processing element. Analyze the definition means for each computer to define what should be done and the defined processing specifications for all computers, generate startup relationship information that shows the startup relationship between each processing element, and use the generated startup relationship information to identify the startup source. A business processing system characterized by comprising analysis and storage means for storing data for each computer. 2. When processing such as a transaction is activated on a certain computer, the activation destination processing element and activation destination computer that are activated by the activation source processing element are determined by searching the activation related information. Claim 1, further comprising execution means for transferring data such as a transaction for activating a destination processing element to a determined destination computer, and causing the required process to be executed on the destination computer. Administrative processing system described.