JPH01281557A - Load distributing system for computer processing - Google Patents

Abstract

PURPOSE:To distribute a load and to highly efficiently execute computer processing by providing the title system with a program group for mutually transmitting/receiving data and a processor for executing a program in parallel. CONSTITUTION:At the time of being started, an input processing program 5 in the processor 2 executes a program start instruction 7 for starting a slip printing processing program 6 to be immediately executed in parallel. After executing the input processing of items corresponding to a slip, the program 5 immediately writes data to be sent to an executing program 6 in a common memory area 8 by using a data transfer instruction 9 and continuously transfers its operation to the input processing of the succeeding slip. On the other hand, the program 6 always monitors whether data are inputted to the area 8 or not,and at the time of its input, receives the data by using the instruction 9 and starts the slip printing processing operation of the data. Consequently, the input processing of the program 5 and the slip printing processing of the program 6 can be executed in parallel.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a computer processing load distribution method, and particularly to a computer processing load distribution method for distributing the computer processing load among a plurality of processes.

[Conventional technology]

Conventional computer processing load balancing methods use instructions that can be executed in parallel within a single program to execute the processing of multiple instructions in parallel; There is a method in which the processes are executed in parallel and the data processed by each process is transferred manually.

[Problem to be solved by the invention]

However, in the former of the conventional computer processing load distribution methods described above, the instructions that can be executed in parallel are limited to only some instructions (for example, in the Kobol language, the WRITE instruction and 5END instruction), and the number of instructions that can be executed in parallel is limited. This means that a large number of instructions cannot be processed in parallel.

Therefore, the former method has the disadvantage that the effect of distributing the computer processing load is limited to the processing of some instructions that can be executed in parallel.

In addition, in the latter of the conventional computer processing load balancing methods mentioned above, multiple programs to be processed in parallel are manually input, and the data processed by each program is handed over again manually, which is time-consuming and may cause problems. An object of the present invention is to provide a computer processing load distribution method that can efficiently execute target computer processing by distributing the load to the processing of multiple programs that are executed in parallel. It's about doing.

[Means to solve the problem]

The computer processing load distribution method of the present invention includes (A) a processing device that can execute a plurality of programs in parallel; (C) a common memory area that can transfer data to and from a plurality of programs running on the processing device; (D> (E) a data transfer instruction capable of transferring data between the executed program and the common memory area by executing each of the plurality of programs being executed on the processing device; (E) the program startup instruction; By executing as necessary, several programs are activated, and the programs are transferred to each other while passing necessary data between them using the common memory area and the data transfer instruction. D'ifre is a group of programs that can perform the target computer processing by sharing the load with the processing of each program by executing them in parallel.

[Example]) Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of a computer processing load distribution method according to the present invention.

The system shown in FIG. 1 is a system that performs computer processing published by From the file stored in 3, using the product code as a key, the product name, unit price, etc. are successively retrieved, and the processing device 2 collects the items necessary for the slip, such as the product code, product name, unit price, and total amount per quantity!! By doing so, it is possible to print out the information on the slip using the slip printing/output device 4.

In the above operation, the processing device 2 has a function that can execute multiple programs in parallel, so it performs input processing that controls input from the keyboard 1 and reading from the magnetic disk device 3. The input processing program 5, which is a program that performs the processing, and the slip printing processing program 6, which is a program that performs slip printing processing that controls slip editing and printing output by the slip printing output device 4, can be executed in parallel.

At this time, the operations of the input processing program 5 and the (invoice printing processing program 6) are performed as follows.

First, when the input processing program 5 is started on the processing device 2, it immediately executes a program starting instruction 7 that starts a slip printing processing program 6, which is another program that is executed in parallel on the processing device 2.

After the input processing program 5 performs the input processing of the items for each slip, the input processing program 5 immediately sends the data for each slip to the common memory area 8 to the slip printing processing program 6 running in the processing device 2. The data transfer command 9 is used to write data to the data transfer command 9, and the process continues to input the next slip.

On the other hand, the slip printing processing program 6 is stored in the common memory area 8.
It always monitors whether or not data has been input, and when the data is input to the slip printing processing program 6, it receives it using a data transfer command 9 and starts the operation of slip printing processing of the received data.

As a result, the input processing of the input processing program 5 and the slip printing processing of the slip printing processing program 6 are executed in parallel. This is done by receiving completion data indicating that the slip has been completed.

In this way, the input processing program 5 and the slip printing processing program 6 that constitute the program group are activated, and these programs are executed in parallel while passing necessary data between them. By sharing the load of processing each program,
8. As described above, the computer processing load distribution method of this embodiment distributes the load to the processing of multiple programs that are executed in parallel. This allows computer processing to be carried out efficiently.

[Effects of the Invention] As explained above, the computer processing load distribution method of the present invention is capable of efficiently executing target computer processing by distributing the load among the processing of multiple programs executed in parallel. It has the effect of being able to.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of a computer processing load distribution method according to the present invention. ■...Keyboard, 2...Processing device,
3...Magnetic disk device, 4...Voucher printing output device, 5...Manual processing program, 6
・・・・・・Voucher printing processing program, 7・・・・・・
Program startup instruction, 8... common memory area,
9...Data transfer command. Represented Patent Attorney Atsushi Uchi

Claims (1)

[Scope of Claims] (A) A processing device that can execute multiple programs in parallel; (B) A program that is executed on the processing device to another program that is executed in parallel on the processing device. a program startup instruction that can start a program; (C) a common memory area that can transfer data between multiple programs running on the processing device; (D) a program startup instruction that can start a program; a data transfer instruction capable of transferring data between the executed program and the common memory area by executing it in each of the plurality of programs being executed; (E) executing the program startup instruction as necessary; By starting several programs, the programs are executed in parallel while passing necessary data between them using the common memory area and the data transfer instruction. A computer processing load distribution method comprising: a group of programs capable of performing targeted computer processing by sharing the load among the processing of each program.