JPH0573393A - Decentralized file control system - Google Patents

Abstract

PURPOSE:To maintain the same contents for subordinate source files of plural slave work stations at all times by gathering the contents of files updated on plural slave work station and generating a total difference, and putting it in the source files of the respective slave work stations. CONSTITUTION:A master work station(MWS) process gathers difference information of the respective slave work stations(SWS) and generates the total difference, which is transferred from the MWS 1 to the respective SWSs 2. An SWS process is started from the MWS 1 to compare the subordinate source file 4 with a work station update file 5 and generate a difference, whose difference information is transferred to the MWS 1. The SWSs 2 once receiving the total difference information put it in the subordinate source files 4 and copy them to the work station update files 5. Consequently, the respective slave work stations have the same contents of the subordinate source files.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a file management system for files distributed in a plurality of workstations.

[0002]

2. Description of the Related Art Conventionally, as a file update management technique, a plurality of file updaters update the same file from a terminal connected to one workstation.

In order to refer to and update a file from a plurality of distributed workstations,
SUN Microsystems NFS (Network)
There is a File System). In this NFS, a file of a workstation positioned as a server can be referred to as a file of its own workstation from a plurality of distributed workstations through a network.

[0004]

However, the technique for updating a file in a single workstation inevitably increases the response time because the files are simultaneously accessed from a plurality of terminals. In addition, when accessing a file from a location where the network capacity is low, the transfer of file information is slow.

Further, in the NFS which refers to and updates a file from a plurality of distributed workstations,
Since it is always visible as a file on my workstation, updates are immediately reflected in the original file, which is a security issue. In addition, since information regarding updates is transferred in the network, it is inevitable that sufficient network capacity is required and the load on the network becomes high.

[0006]

According to the distributed file management method of the present invention, files are distributed and arranged in each of a master workstation and a plurality of slave workstations connected via a network, and the master workstation is an original document. Managing a file, each of the slave workstations managing a copy of the original file and a sub-original file for each of the slave workstations after updating based on the copy of the original file; First means for instructing each of the slave workstations to periodically collect the difference between the copy of the original file and the sub-original file of each of the slave workstations; and each of the slave workstations. The difference collected from each And a third means for transferring the total difference to the target slave workstation; each of the slave workstations is transferred from the master workstation. Fourth means for receiving the entire difference and updating a copy of the original file of its own; fifth means for copying the updated copy of the original file to the sub-original file; and A sixth means for updating and a seventh means for comparing the copy of the original file and the updated sub-original file by the instruction of the master workstation to generate the difference and transferring it to the master workstation. Prepare

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described with reference to the drawings.

Referring to FIGS. 1 and 2 showing an embodiment of the present invention, a master workstation 1 and a plurality of slave workstations 2 are connected to a network 6, respectively. The master workstation 1 manages the original file 3 and the slave workstation 2
Each manages a sub-original file 4 and a workstation update file 5.

The master workstation process is activated periodically and registered slave workstations 2
To each slave workstation 2 in turn
The difference information 7 is collected (S10 in FIG. 2). When collecting the difference information 7 of all target slave workstations 2,
An overall difference is created (S11), and the overall difference information 8 is transferred from the master workstation 1 to each slave workstation 2 (S12). On the other hand, the slave workstation process is started from the master workstation 1 and compares the sub-original file 4 with the workstation update file (after workstation update) 52 to generate a difference (S20). After the generation, the difference information 7 is transferred to the master workstation 1. Overall difference information 8 transferred after processing by the master workstation 1
Is received by the slave workstation 2, the entire difference is included in the sub-original file 4 (S21), and after inclusion, is copied to the workstation update file (after copying the sub-original file) 51 (S22). Slave workstation 2 has workstation update file 5
Is updated to create a workstation update file (after workstation update) 52 (S23). The processes of the master workstation 1 and the slave workstation 2 are repeated, and the original file 3 of the master workstation 1 and the sub-original file 4 of the slave workstation 2 are always kept to have the same contents.

[0010]

As described above, according to the present invention, the contents of files updated by a plurality of slave workstations are collected and the entire difference is stored in the sub-original file of each slave workstation, so that Sub-original files on multiple slave workstations can be kept the same. Further, by transferring the difference, the load on the network can be reduced as compared with transferring the entire file.

The distributed file management system of the present invention is a LAN (Local Area Network) housed in one site.
It is more effective not only for file management of a plurality of workstations connected to rk), but also for file management of workstations at a plurality of sites using a wider network.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing an embodiment of the present invention.

FIG. 2 is a diagram for explaining the operation in the embodiment.

[Explanation of symbols]

 1 Master workstation 2 Slave workstation 3 Original file 4 Sub-original file 5,51,52 Workstation update file 6 Network

Claims (1)

[Claims] 1. A file is distributed in each of a master workstation and a plurality of slave workstations connected via a network, the master workstation manages an original file, and each of the slave workstations Managing a copy of the original file and a sub-original file for each slave workstation after updating based on the copy of the original file; and further, the master workstation issues an instruction to each of the slave workstations and the slave First means for periodically collecting a difference between the copy of the original file of each of the workstations and the sub-original file and the difference collected from each of the slave workstations are integrated to generate an overall difference. Second means to do, before Third means for transferring the entire difference to the target slave workstation; each of the slave workstations receives the entire difference transferred from the master workstation and copies its own original file. And a fifth means for copying the updated copy of the original file to the secondary original file, a sixth means for updating the secondary original file, and an instruction of the master workstation. And a seventh means for comparing the copy of the original file with the updated secondary original file to generate the difference and transfer it to the master workstation.