JPS62286137A - Decentralized file processing system - Google Patents

Abstract

PURPOSE:To improve the processing performance of a decentralized processing system by copying previously a file to receive an access and storing it in a computer system set at the access side when the read-only accesses last at a rate more than a fixed level. CONSTITUTION:When a computer 2b gives an access request to a computer 2a, the output device 5a of a computer system 1A outputs the data to the computer 2b of a computer system 1B set at the access side when the read-only accesses last at a fixed rate. The storing device 6b of the system 1B receives the data from the device 5a and produces the copy of a file 4a to receive an access within the external memory device 3b of the system 1B as a copy file 41a. Hereafter the access is given to the file 41a in place of the access given to the file 4a of the system 1A via a communication circuit 1.

Description

Detailed Description of the Invention 3. Detailed Description of the Invention [Industrial Application Field] This invention provides computer systems interconnected via a network,
The present invention relates to a distributed file processing method that allows files stored in external storage devices to be freely accessed via a network.

[Conventional technology]

Figure 2 shows a configuration diagram of a conventional distributed file processing method.
In the figure, 1 is the communication line that constitutes the network, 2
a to 2C are the computers connected to each other via the communication line 1; 3a to 3C are external storage devices such as disk devices provided in each of the calculations i2a to 2c; and 4a to 4c are the respective computers. The files IA to IC stored in the external storage devices 3a to 3C are computer systems.

Next, the operation will be explained.

For example, when calculation m2a attempts to access file 4C stored in external storage device 3C provided in computer system 1C, computer 2a outputs a processing request to computer 2C, and sends the file via communication line 1. 4C is taken into the computer 2a.

This file can be imported via the moderate communication line 1 that outputs processing requests.

Similarly, even when the computers 2b and 2c access a file stored in the external storage device 3C provided in another computer system IA, the above procedure is followed.

[Problem that the invention seeks to solve]

In this way, conventional distributed file processing methods output moderate processing requests that require files owned by other computer systems, and import files via communication lines, so the communication lines are used frequently and the performance of the communication lines is affected. There was a problem that the data processing speed was determined by

This invention was made to solve the above-mentioned problems, and aims to reduce the frequency of line use and improve processing performance.

[Means for solving problems]

For this reason, the distributed file processing method according to the present invention copies and stores the file to be accessed in the computer system on the accessing side in advance when read-only access continues for a certain percentage or more, and then transfers the file to the other computer via a communication line. This system is characterized in that the copied file is accessed instead of accessing a file in an external storage device provided in the system.

[Effect]

In this invention, when read-only access continues for a certain percentage or more, a copy of the file to be accessed is generated and stored in advance in the computer system on the accessing side.

Then, instead of accessing a file in an external storage device provided in another computer system via a communication line, the copied file is accessed.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. 1st
The figure is a block diagram of a distributed file processing system showing an embodiment of the present invention. In the figure, when read-only access continues for a certain percentage or more, data to that effect is output to the computer on the accessing side. The output devices 6a to 6c are storage devices that copy files to be accessed based on data from the output devices in advance and store them in a disk device provided in the computer system to be accessed. The same components are given the same numbers and their explanations are omitted.

Next, the operation will be explained.

The premise is that the computer system 1B is the computer system to be accessed, the computer system IA is the computer system to be accessed, and the computer system IB is accessing the file 4a stored in the external storage device 3a provided in the computer system IA. Consider a case where

First, when the computer 2b outputs a file access request to the calculation m2a, the output device 5a in the computer system IA detects whether read-only access continues at a certain rate, and performs read-only access at a certain rate or higher. If the dedicated access continues, data to that effect is output to the computer 2b of the accessing computer system IB.

When the storage device 6b in the computer system IB receives the data output from the output device 5a, it stores a copy of the file 4a to be accessed in the external storage i! of the computer system IB. It is generated as a coca aisle 4La in the device 3b.

From now on, instead of accessing the file 4a of the calculation stage system IA via the communication line 1, the copy file 4a is accessed via the communication line 1.
Access 1a.

Here, the computer 2b is the file 4 of the computer system IA.
When accessing a for writing, all storage devices 6a to 6c are notified, and the copied copy file 41a is
Erase.

In the above embodiment, the storage device 6b generates the copy file ILa in its own computer system IB, but the copy file 41a may be generated by a storage device in another computer system.

〔Effect of the invention〕

As explained above, the distributed file processing method according to the present invention copies and stores the file to be accessed in the computer system on the accessing side in advance when read-only access continues for a certain percentage or more, and then transfers the file to the other computer via a communication line. Instead of accessing the files in the external storage device provided in the computer system, the copied files are accessed, so there is no need to access files in other computer systems via communication lines, and file processing is simplified. This has the effect of improving processing performance.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG. 2 is a block diagram of a conventional file distribution processing system. 1... Communication line, IA-IC... Computer system,
2a-2C...Computer, 3a-3C...External storage device, 4a-4C...File, 5a-5C...Output device, 6a-6b...Storage device, 41a...Copy file . Agent Masuo Oiwa (2 idiots) 11 Figure 2

Claims (1)

[Claims] Multiple computer systems equipped with disk devices storing files are connected to each other via a network, and one computer system can access files in an external storage device provided in another computer system via the network. In a distributed file processing system configured as above, when read-only access continues for a certain percentage or more, the accessing computer system copies and stores the file to be accessed in advance, and then transfers the file to other computer systems via a communication line. A distributed file processing method characterized in that the copied file is accessed instead of accessing a file in an external storage device provided in the computer.