JP2003099297A - System and method for file distribution, program, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To lighten the load of distribution of a file to a client through a network on the network. SOLUTION: When the file is distributed from a distribution source server 10 to a plurality of clients 30 through a plurality of relay servers 20, the distribution source server 10 transfers a difference file to the respective relay servers 20, which transfer the difference files to one another, and not the difference file, but a file restored on each relay server 20 is transferred from the relay server 20 to each client 30 connected to the relay server 20. Specially, when a plurality of files (file group) are distributed, an extracting process for the difference file based upon a comparison between files of the file group 13 which are distributed this time is performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a file distribution technique using a network in a computer, and more particularly to a file distribution technique suitable for reducing the load on the network due to file distribution.

[0002]

2. Description of the Related Art For example, there are file groups that are frequently updated, such as virus definition files of antivirus software. When distributing such a file group to another computer via a network, the file group is compressed so that it can be distributed in a shorter time, and, for example, JP-A-10-124419 and JP-A-10-28917.
It is common practice to reduce the data size by transferring only the difference, as in the technique disclosed in Japanese Patent Laid-Open No. 2 or Japanese Patent Laid-Open No. 11-228818.

However, these techniques distribute files one by one from a distribution server to a plurality of clients, and when there are a large number of clients or the distance between the server and the client is long. In the case where the line is thin, the server or the network cannot process the data and the data transfer time becomes long.

As described above, instead of directly distributing a file from one distribution source server to all clients, a plurality of servers are provided and files to be distributed are transferred to each server, There is also a technology that shortens the data transfer time by distributing the file from the server to the nearby client.

However, even in the technique for distributing files using a plurality of servers in this way, the environment of a large number of clients is different, for example, because the OS (operating system) is different, the version of the current virus definition file is different, etc. Not necessarily the same, so a large number of distribution files to apply to all clients,
Must be prepared for each server. As a result, the amount of data transferred when transferring files to each server increases,
Distribution will take time.

As described above, in both the conventional file distribution technology by difference and the technology for distributed distribution of files using a plurality of servers, the load on the network is large, and as a result, the execution of other tasks and applications is performed. There is a danger of hindering

[0007]

The problem to be solved is that the conventional technique cannot reduce the network load associated with file distribution.

An object of the present invention is to solve these problems of the prior art and to enable the distribution of a large number of files via a network without hindering the execution of other tasks and applications.

[0009]

In order to achieve the above object, according to the present invention, (1) a plurality of first server computers (distributor servers) are connected via a second server computer (relay server). When distributing a file to a client computer, a difference file is transferred from the distribution source server to each relay server, and a difference file is also transferred between each relay server so that each relay server can transfer to each client connected to that relay server. Is not a diff file,
Transfer the restored files on each relay server. (2) In addition, when distributing a plurality of files (file group), the distribution source server distributes this time together with extraction of a difference file based on a comparison process between the file group distributed this time and the file group previously distributed. Extraction of the difference file based on the comparison processing of files in the file group, the difference files extracted by the respective difference extraction processing, and information for restoring the file group to be distributed this time from this difference file (difference information) Is transferred to the relay server, and each relay server restores a file group based on the respective difference file and the difference information and transfers the file group to the client.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram showing a configuration example of a file distribution system according to the present invention, and FIG. 2 is a block diagram showing a network configuration example suitable for applying the file distribution system in FIG. 3 is a flowchart showing an example of processing operation of the file distribution system in FIG. 1, FIG. 4 is an explanatory diagram showing a specific example of file creation by the processing in FIG. 3, and FIG. 5 is created by the file distribution system in FIG. It is explanatory drawing which shows the specific example of a difference information file.

FIG. 2 shows a file distribution form when a file is distributed using a relay server. First, the conventional problem will be described with reference to FIG.
For example, the clients 1c to 7 in each of the sites A1 to G7
Consider a case where the files 1b to 7b are distributed from the distribution source server 1a or the relay servers 2a to 7a to c and 1d to 7d.

In this case, each of the clients 1c to 7c and 1d to the distribution source server 1a or the relay servers 2a to 7a.
Since only the files necessary for the clients 1c to 7c and 1d to 7d are distributed to 7d, the transfer data amount is not so large. Further, the network in each of the sites A1 to G7 is a relatively high-speed network such as a LAN, and is unlikely to cause a problem in terms of load on the network.

On the other hand, the data transfer amount between the distribution source server 1a and the relay servers 2a to 7a among the sites A1 to G7 is at least the distribution source server 1a and the relay server 2a to.
All clients 1c to 7c and 1d to 7 under 7a
Very many because it needs to send a distribution file for d.

Further, the network between the sites is
Since it is slower than the internal network such as a private line and the Internet, the larger the transfer amount, the more serious the problem.

Now, let us consider a case where the relay servers 2a to 7a are not used. In this case, the file is directly distributed from the distribution source server 1a to all the clients 1c to 7c and 1d to 7d, which is certainly a problem when the relay servers 2a to 7a are used. At first glance, the large amount of data transferred during the period is eliminated.

However, another site B2 other than the clients 1c and 1d in the site A1 of the distribution source server 1a.
Since a low-speed network line has to be used for file distribution to each of the clients 2c to 7c and 2d to 7d in G7, the transfer efficiency is very poor as compared with the form using the relay servers 2a to 7a, and the transfer time is longer. Worsen. In particular, when the number of clients 1c to 7c and 1d to 7d is large, the problem becomes more prominent, which is not a good idea.

When the relay servers 2a to 7a are used, it is necessary to prepare different distribution files for a large number of clients having different environments such as OS, so that the substantial size of the distribution files increases. The point is the neck.

Therefore, in this example, in the distribution source server 1a, the difference files between all the previously distributed files and the files to be distributed this time, and further the difference files between the files in the files to be distributed this time. Is created and transferred to each of the relay servers 2a to 7a.

Then, in each of the relay servers 2a to 7a, a file group to be distributed this time is restored from the difference file of this time and the distribution file of the previous time (restored), and transferred to each of the clients 2c to 7c, 2d to 7d, Moreover, the restored file group is retained to restore the next difference file.

By scheduling the transfer processing of the difference file from the distribution source server 1a to each of the relay servers 2a to 7a as described above, the distribution contents are made uniform among the relay servers 2a to 7a.

The configuration of the file distribution system that performs such processing operations will be described below with reference to FIG.

In FIG. 1, reference numeral 10 is a distribution source server, and 20 is a distribution source server.
Is a relay server and 30 is a client.
A display device (not shown), an input device, an external storage device, and an optical disk drive device for reading out programs and data recorded on an optical disk such as a CD-ROM, and L
A communication device such as an AN card is provided, and a computer processing function for performing processing according to the present invention is mounted.

That is, each of the distribution source server 10, the relay server 20, and the client 30 installs the program and data stored in the optical disc or the like into the external storage device via the optical disc drive device or the like, and then, the external storage device. Read from main memory to CPU
By processing in (Central Processing Unit),
The distribution source server 10 is equipped with a difference file creation unit 11 and a difference file transfer unit 12, the relay server 20 is equipped with a difference file reception unit 21, a difference file restoration unit 22, and a restoration file transfer unit 23, and A restoration file receiving unit 31 is mounted on the client machine 30 and executes a file distribution process according to the present invention.

As described above, the distribution source server 10 has the difference file creation unit 11 and the difference file transfer unit 12, and the difference file creation unit 11 distributes the file group 13 distributed this time and the file group 14 distributed last time. And take the difference,
The difference file group 15 is created. At the same time, the difference information file 16 is created.

In the difference information file 16, information indicating how a difference is created for each difference file group 15 is written, and in the relay server 20, the distribution file (this time restored file) from the difference file group 24 is written. Group 2
It is necessary when restoring 7).

In the difference file transfer unit 12 of the distribution source server 10, the relay server 2 of the difference file group 15 and the difference information file 16 created by the difference file creation unit 11
Transfer to 0. At this time, the difference file transfer unit 1
2 transfers the difference file to each of the relay servers 2a to 7a in accordance with the schedule so that the distribution contents among the relay servers 2a to 7a shown in FIG.

The relay server 20 includes the difference file receiving unit 2
1, difference file restoration unit 22, restoration file transfer unit 23
First, the difference file receiving unit 21 receives the difference file group and the difference information file transferred from the distribution source server 10, and stores them as the difference file group 24 and the difference information file 25.

Next, in the difference file restoring section 22,
Difference file group 2 received by the difference file receiving unit 21
4, the difference information file 25, and the previously restored file group 26 restored and saved last time are used to create the currently restored file group 27. This restored file group 27
Is substantially the same as the current distribution file group 13 to be distributed this time in the distribution source server 10.

Then, the restored file transfer unit 23 transfers the currently restored file group 27 restored by the difference file restoration unit 22 to the client 30 accordingly. still,
If it is necessary to transfer the data from this relay server 20 to another relay server, the difference file group 24 and the difference information file group 25 instead of the currently restored file group 27.
To transfer.

In the client 30, the restore file receiving unit 31 receives the present restore file transferred from the relay server 20, and replaces the old file as the restore file 32.

Next, referring to FIG. 3, the difference file group 1 by the difference file creating unit 11 in the distribution source server 10
5 and the creation processing operation example of the difference information file group 16 will be described.

First, regarding a certain distribution file in the file group to be distributed this time (step 101),
All the files distributed last time are compared to see if they have the same contents (step 102). This looks at the entire contents of the file, not the file name.

If there is the same content, the file name of the previous distribution file having the same content is entered as the difference information file 16 (step 103). If there is no file with the same contents in the files distributed the previous time, then the same comparison is performed for all the other files to be distributed this time (step 104). That is, a comparison process is performed between the files to be distributed this time.

Here again, if there is the same content,
As the difference information file 16, the file name of the current distribution file having the same content is entered (step 105).

If there is no file having the same contents in any of the files, the file having the most matching portion is selected from both files (the previously distributed file group and the current distribution target file group) (step 106). And
As the difference information file 16, the file name of the partial matching file and the matching portion are entered (step 107), and the difference file is created by removing the partial matching portion from the distribution target file (step 108).

The above process is performed for all distribution files this time (step 109) to obtain the difference file group 15 and the difference information file 16.

Next, such a difference file creating unit 11
A specific example of the difference file creation processing by
It will be explained based on.

In the example shown in FIG. 4, a difference file creation process is performed on the distribution target file group (files 20A to 20D) this time to obtain a reduced difference file group (files 20c and 20d) by removing the overlapping portion. There is.

That is, the file 20A in the file group to be distributed this time has the same contents as the file 10A in the previously distributed file group, and the file 20A
B has the same contents as this file 20A, and there is no difference file for each.

The file 20C partially overlaps with the above-mentioned file 20B, and also the file 20D.
Regarding, regarding file 10B of the previously distributed files
And a part of the file 20c and the file 20d, which are reduced by removing the respective overlapping parts, are created as a difference file group.

FIG. 5 shows an example of the difference information file 16 created by the process of creating such a difference file group.
In FIG. 5, for each distribution target file on the left side,
On the right side, information indicating whether the same file or partial matching file exists, information indicating whether the same / partial matching file is in the previous distribution file or this distribution file, the file name of the file , And in the case of a partial match, information indicating where the partial match files match, where the partial file matches, the difference file name, and the part where the matching part of the partial match file is omitted. Information is described.

For example, in the case of the distribution file 20C, there is a partial match between the 100th byte and the 200th byte of the file 20B, which is the distribution file of this time, and the partial matching part is 10th of the file 20c, which is the difference matching file.
It is indicated that it is omitted from the 0th byte.

Regarding the difference file restoration process in the difference file restoration unit 22 in the relay server 20, the difference file group created in the difference file creation process in the difference file creation unit 11 in the distribution source server 10 described above Based on the difference information file and the previous restoration file group (which is substantially the same as the previous distribution file group), if you follow the procedure completely opposite to the above difference file creation process, it will be the same as this distribution file It is possible to obtain a set of restored files.

As described above with reference to FIGS. 1 to 5, in this example, when a file is distributed from the distribution source server 10 to a plurality of clients 30 via a plurality of relay servers 20, the distribution source The difference file is transferred from the server 10 to each relay server 20, the difference file is also transferred between the relay servers 20, and the difference file is transferred from each relay server 20 to each client 30 connected to each relay server 20. Instead, transfer the restored files on each relay server.

In particular, when distributing a plurality of files (file groups) in the distribution source server 10, a difference file is extracted based on a comparison process between the file group 13 distributed this time and the file group 14 previously distributed. ,
To extract a difference file based on a comparison process between files of the file group 13 to be distributed this time, and to restore the difference file group 15 extracted by each difference extraction process and the file group to be distributed this time from this difference file And the information (difference information file 16) of (1) is transferred to each relay server 20, and each relay server 20 distributes this time based on the difference file group 24 and the difference information file 25 received and stored respectively. The group is restored (currently restored file group 27) and transferred to the client 30.

As a result, it becomes possible to efficiently distribute a specific file to all computers on the network.

The present invention is not limited to the examples described with reference to FIGS. 1 to 5, and various modifications can be made without departing from the scope of the invention. For example, in this example,
Although the procedure for transferring the appropriate restoration file from the relay server 20 to the client 30 is used, the client 3
The procedure may be such that the relay server 20 is accessed from the 0 side to obtain the desired restored file this time.

In this example, the transfer of a plurality of files (group of files) has been described, but the present invention can also be applied to the transfer of a single file.

Further, the computer configuration of the relay server 20 in this example may be a computer configuration without a keyboard or an optical disk drive. Further, as the recording medium of the program, an FD (Flexible Disk) or the like may be used instead of the optical disc.
Further, regarding the installation of the program, the program may be downloaded and installed via the network via the communication device.

[0051]

According to the present invention, when a file is distributed from a file distribution source server to a client via a relay server via a network, the load on the network can be reduced, and other jobs and applications can be reduced. It is possible to avoid the risk of hindering the execution of the above, and it is possible to distribute a large number of files via the network without hindering the execution of such other business and applications.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration example of a file distribution system according to the present invention.

FIG. 2 is a block diagram showing an example of a network configuration suitable for applying the file distribution system in FIG.

3 is a flowchart showing an example of processing operation of the file distribution system in FIG.

FIG. 4 is an explanatory diagram showing a specific example of file creation by the processing in FIG.

5 is an explanatory diagram showing a specific example of a difference information file created by the file distribution system in FIG.

[Explanation of symbols]

1 to 7: sites A to G, 1a: distribution source server, 2a to 7
a: relay server, 1b to 7b: files, 1c to 7c,
1d to 7d: client, 10: distribution source server, 1
1: difference file creation unit, 12: difference file transfer unit,
13: current distribution file group, 14: previous distribution file group, 15: difference file group, 16: difference information file,
20: Relay server, 21: Difference file receiving unit, 22:
Difference file restoration unit, 23: restoration file transfer unit, 2
4: difference file group, 25: difference information file, 26:
Previously restored files, 27: Currently restored files, 3
0: client, 31: restoration file receiving unit, 32:
Restore file.

Claims (7)

[Claims] 1. A system for distributing a file from a distribution source server to a plurality of clients via one or more relay servers via a network, wherein the distribution source server stores a file to be distributed this time. A relay server is provided with a difference file from a previously distributed file, and means for generating difference information used to restore the file distributed this time from the previously distributed file using the difference file and transferring the difference information to the relay server. Includes a means for receiving the transferred difference file and difference information, a means for transferring the received difference file and difference information to another relay server, a received difference file and difference information and the previously distributed file. Based on the above, the means for restoring the file distributed this time and the restored file as the previously distributed file A file distribution system characterized in that means for storing and means for transferring the restored file to the client are provided, the difference file and difference information are transferred between the servers, and the restored file is transferred from the server to the client. . 2. A system for distributing a plurality of files (a group of files) from a distribution source server to a plurality of clients via one or more relay servers via a network, the distribution source server comprising: , Comparing the file group distributed this time with the file group distributed last time, and restoring the file group distributed this time from the first differential file group and the file group previously distributed using the first difference file group. The means for generating the first difference information used for this purpose and the file groups distributed this time are compared with each other, and the second difference file group and the file group distributed this time using the second difference file group Means for generating the second difference information used to restore the file group to be distributed, and the generated first and second difference file groups and the first and second
And means for transferring the difference information of the above to the relay server,
The relay server receives the transferred first and second difference file groups and the first and second difference information, and receives the first and second difference file groups and the first and second difference files. Means for transferring the difference information of the other relay server, and the received first,
Means for restoring the file group to be distributed this time based on the second difference file group, the first and second difference information, and the previously distributed file, and saving the restored file group as the previously distributed file group Means for transferring the restored file group to the client, transferring the first and second differential file groups and the differential information between the servers, and transferring the restored file group from the server to the client. A file distribution system characterized in that 3. A file distribution method of a system for distributing a plurality of files (a group of files) from a distribution source server to a plurality of clients via one or more relay servers via a network, the distribution comprising: In the original server, a difference file between the file to be distributed this time and the file distributed last time, and a procedure of generating difference information used to restore the file distributed this time from the file previously distributed using the difference file, and The difference file and the difference information are transferred to the relay server, and the relay server receives the transferred difference file and the difference information, and the received difference file and the difference information are transferred to another relay server. Based on the procedure to transfer to the server, the received difference file and difference information and the previously distributed file The steps of restoring the file to be distributed this time, saving the restored file as the previously distributed file, and transferring the restored file to the client are executed, and the difference file and difference information between the servers are executed. File distribution method, and the restored file is transferred from the server to the client. 4. A file distribution method of a system for distributing a file group from a distribution source server to a plurality of clients via one or more relay servers via a network, wherein the distribution source server is In order to compare the file group to be distributed with the previously distributed file group and restore the first distributed file group and the previously distributed file group from the previously distributed file group using the first differential file group The procedure for generating the first difference information to be used is compared with the file groups to be distributed this time, and the second difference file group and the second difference file group are used to distribute from the file group distributed this time. A procedure for generating the second difference information used to restore the file group, and the generated first and second difference file groups and the first and second
And the procedure of transferring the difference information of the above to the relay server, and the relay server transfers the first and second transferred information.
And a procedure for receiving the first and second difference information, and a procedure for transferring the received first and second difference file groups and the first and second difference information to another relay server, A procedure for restoring the file group to be distributed this time based on the received first and second difference file groups, the first and second difference information and the previously distributed file, and distributing the restored file group to the previous time. Stored as a file group and a procedure for transferring the restored file group to the client, transfer the first and second difference file groups and difference information between the servers, and restore from the server to the client. A file distribution method characterized by transferring a group of files that have been created. 5. The file distribution method according to claim 3 or 4, wherein in order to make the distribution contents identical among a plurality of relay servers, from the distribution source server to the relay server. Distributing a file to a file, the method comprising the step of scheduling a transfer process. 6. The computer according to any one of claims 3 to 5.
A program for executing each step in the file distribution method described in any of 1. 7. The computer according to any one of claims 3 to 5.
A computer-readable recording medium recording a program for executing each step in the file distribution method described in any one of 1.