JP2000222268A - Method for synchronizing file for plural computers - Google Patents

Abstract

PROBLEM TO BE SOLVED: To synchronize the contents of file groups on computers in a relatively short network connection time by disconnecting the computers from a network and updating files, and connecting the computers to the network as required. SOLUTION: For example, a slave computer A triggers the synchronization of files with a master computer. After a line is connected, the master computer side obtains identification information and final synchronization date and time information of the slave computer A and extracts difference contents (difference file group) of files updated on the master computer side after the final synchronization date and time. The slave computer A, on the other hand, sends a difference file group updated by itself after the date and time of the final synchronization with the master computer. After the contents of the difference file group are all sent to the master computer side, the difference file group extracted on the master computer side is sent.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of exchanging files on two or more computers independently of each other via a network, and exchanging information on the updated contents between the computers. Regarding file synchronization methods that synchronize the contents of files held on each other's computers, especially Sales For
In a sales support system such as ce Automation, each computer is normally disconnected from the network to perform file updates, and if necessary, connected to each other via the network, the computers are connected to each other with a relatively short network connection time. Even if the form of the data management unit used by the users is different, users can exchange information about file updates between two parties without being aware of the difference, and synchronize the contents of the file group on each other's computer The preferred file synchronization method.

[0002]

2. Description of the Related Art A method of matching data between two or more computers is disclosed, for example, in Japanese Patent Application Laid-Open No. 9-29.
As described in the “file update processing method” of Japanese Patent No. 4180, by exchanging logging data of a file group owned by a plurality of servers with each other, the configuration of the file group owned by each server is matched. Methods are known.

[0003]

The above-mentioned conventional method is suitable for matching the file group configuration in a one-to-one manner with the file group owned by each server in a short network connection time. However, if the relationship between the contents of the files in the database on one computer and the contents of the files in the database on the other computer is a superset and subset relationship to each other, There is a problem that a server that manages files of a superset cannot simply extract from the contents of logging data as update logging information for a group of files of a subset managed by a specific computer.

[0004] In addition, this method is to be aware of the contents of logging data if the database management units of a plurality of computers are unified with the same widely available DBMS (database management system) application product. However, it is possible to match the file configuration with a standard function such as replication at the level of the DBMS applications installed on each other, but when the database management unit has a DBMS application with different specifications on each computer. It is necessary to create a syntax analysis process, a common linguistic process, and a distribution process of logging information at a higher application level in accordance with the type of the DBMS application.
As the types of applications increased, there was a problem that software development for realizing the system became more difficult.

[0005] An object of the present invention is not to simply exchange logging data between respective computers in synchronizing a file group of two or more computers, but to compare differences between respective computers in their respective databases. By exchanging actual data and transmitting update summary information created based on the difference actual data from the master computer to the other computer,
The independence from the database management unit is improved, and the master computer starts another process during the communication time for exchanging the actual difference data information, and transmits the update summary information to be provided to the other computer. The purpose is to reduce the communication time between computers by creating in parallel based on data information, predicting which information transmission time is shorter, and switching to the shorter data transmission time. I do.

[0006]

In order to achieve the above object, according to the present invention, first, a master computer for managing the entire file group and a file group having information which is a subset of the file group held by the master database are managed. One or more slave computers to be connected and a network typified by a WAN connecting them to each other are prepared. The network between the master computer and the slave computers is usually disconnected, and while the network is disconnected, each computer independently performs a file group update process.

When synchronizing the contents of a group of files commonly managed by the computers, a slave computer (slave computer A) first triggers a file synchronization on the master computer. Specifically, a line is connected from the slave computer A to the master computer. After the line is connected, the master computer acquires the identification information of the slave computer A and the last synchronization date and time information.

[0008] After that, based on the information, from among the file group held by the master computer side among the file group held by the slave computer A and the master computer, the master computer A sets the master computer A after the last synchronization date and time. The difference contents of the file updated on the computer side (referred to as a difference file group) are extracted.

On the other hand, the slave computer A transmits the last synchronization date and time information and the identification information, and then successively transmits a group of difference files updated by the slave computer A itself after the last synchronization date and time with the master computer. When all the contents of the difference file group are transmitted to the master computer, the master computer starts transmitting the difference file group extracted by the master computer to the slave computer A.

During the transmission, on the master computer side,
Based on the information of the difference file group of the slave computer A received earlier, the difference information with the computer A is reflected in the master database managed by the own computer (referred to as synchronization processing). After the completion of the synchronization processing, the master computer creates synchronization summary information for enabling the slave computer A to execute the synchronization processing in a short time.

The synchronization summary information is information that summarizes information related to updating for matching the contents of a file group commonly managed by the master computer and the slave computer A. Since this information is summarized information, it is more compact than the real file information of the difference (difference file group), and when transmitted to the slave computer, can be shorter than the time required to transmit the difference file.

For this reason, if the synchronization summary information creation is completed while the master computer transmits the difference file group to the slave computer A, the line usage time will be equal to the remaining information of the difference file group. Either the time to continue transmission or the time to transmit synchronization summary information from the present time is predicted and selected.

If the transmission of the synchronization summary information is shorter, the transmission of the difference file group is immediately stopped, and the transmission is switched to the transmission of the synchronization summary information. As soon as the transmission of information from the master computer to the slave computer A is completed, the line is disconnected, and the slave computer A starts the synchronization process.

As described above, the file group is exchanged between the master computer and the slave computer A while minimizing the line connection time between the computers, and furthermore, the files are created based on the actual data and the actual data. By exchanging information, which does not depend on the specifications of the database management unit, called update summary information between computers, the independence from the database management unit is increased.
The application can easily be adapted to an information processing system using the method.

[0015]

Next, an embodiment of the present invention will be described in detail with reference to a system configuration diagram and a workflow drawing.

FIG. 1 is a block diagram of a system in which the present invention is realized. FIGS. 2 and 3 are flowcharts showing the file group synchronization processing according to the present invention.

In FIG. 1, 1 is a master computer, 1
0 is the database management unit M of the master computer, 11
Denotes a data processing unit M, 12 denotes a communication unit, and 13 denotes a master database for storing a group of files of the master computer. 2 is a slave computer A and 3 is a slave computer B. Although two slave computers are illustrated, in this figure, the slave computer A is the first slave computer as viewed from the master server, and the slave computer B is the nth slave computer as viewed from the master server. There may actually be three or more computers.

20 is a communication unit on the slave computer A side, 21 is a data processing unit S of the slave computer A,
Reference numeral 22 denotes a database management unit A of the slave computer A, and reference numeral 23 denotes a database A for storing files of the slave computer A. 24 is a communication unit on the slave computer B side, 25 is a slave computer B
The data processing units S and 26 on the side are database management units B of the slave computer B, and 27 are slave computers B
Is a database B for storing a group of files.
The master computer, the slave computer A, and the slave computer B normally update data independently. However, the master computer and the slave computer A are connected to each other via the line 4 in order to occasionally synchronize files with the master computer.

The contents of processing performed by the data processing unit S existing in the slave computer A and the slave computer B are the same. The database management unit A of the slave computer A and the database management unit B of the slave computer B depend on the specification of the database to be managed, and may differ between the slave computers. The contents of the file group managed by the slave computer are usually a subset of the contents of the file group managed by the master computer.

Therefore, each slave computer manages only a group of files required by each slave computer, and the required database size can be made smaller than the database size managed by the master computer. Further, each slave computer does not always match the contents of the file group managed by another slave computer.

That is, if a file common to slave computers exists in a file group, a file unique to each slave computer also exists. On the other hand, the file group managed by the master computer has a file group managed by all slave computers that the master computer follows.

For this reason, when the update processing is performed independently on each computer without connecting the line, the line is sometimes connected to maintain the consistency of the file group between the master computer and the slave computer. It is necessary to exchange information about the contents of the file groups and synchronize the file groups with each other.

Next, a specific flow of the synchronization process of the present invention will be described with reference to the flowcharts of FIGS.

FIGS. 2 and 3 show the flow of the procedure of the file synchronization process in FIG. When performing file synchronization processing, the slave computer triggers the master computer.

Referring to FIG. 2, when a slave computer A under the master computer starts synchronization processing upon an instruction from a user operating the own computer or an event generated by a timer or the like (step 30), first, the slave computer A starts. A performs a line connection request process to the master computer (step 31), and connects the network between the slave computer A and the master computer.
(Steps 32 and 33).

In this figure, the request source is the slave computer A, but it goes without saying that any slave computer may be the request source of the synchronous processing.

Network connection processing (steps 32 and 3)
According to 3), the master computer identifies that the synchronous processing request source is the slave computer A (step 34). A method in which the master computer provides information for identifying the other party from the slave computer A to the master computer side, a method in which the master computer identifies from the transmission source number of the line, and the like are conceivable, but any method may be used. When the master computer identifies that the requesting partner is the slave computer A, the master computer requests another process to extract a difference file group with the slave computer A from the file group on the master computer side. (Step 35).

In the master computer, process 2 is generated (step 36). From the master database, a file group used by the slave computer A and updated after the last synchronization date and time with the slave computer A A group is extracted (step 37). By the way, in step 37, "Slave computer A
32, Step 3
It may be obtained from the slave computer A in the process of 3, or may be held by the master computer itself.

On the other hand, the slave computer A side performs network connection processing with the master computer (step 3).
When 2) is completed, a file group updated on the slave computer A side after the last synchronization date and time with the master computer, that is, a difference file group is transmitted to the master computer side (step 38). The master computer starts receiving the difference file group of the slave computer A (step 39). At this time, in the process 2 on the master computer, if the extraction of the difference file group is completed in the above-described step 37, the difference file group on the master computer side is temporarily stored until the transmission of the difference file group from the slave computer A is completed. (Step 40).

When the transmission of the differential file group from the slave computer A is completed (step 41), the slave computer A notifies the master computer that the transmission has been completed (step 42). When the master computer receives the transmission completion notification from the slave computer A (step 43), the process 1 first starts a synchronization process for the database in the master computer based on the difference file group of the slave computer A (step 43). Step 46).

Next, the process 2 starts transmitting the held difference file group of the master computer to the slave computer A (step 44). The slave computer A receives the information of the difference file group of the master computer (step 45). When the process 2 completes the transmission of all the sub file groups to the slave computer (steps 49 and 50), the process 2 notifies the slave computer A that the transmission of the difference file group is completed (step 51). The slave computer A receives this notification (step 52), and then performs network disconnection processing between the computers (steps 53 and 54).

After performing the cutting process, the process 2 ends.
(Step 62), the slave computer A performs a synchronization process while matching the information of the difference file group received from the master computer with the information of the file group on its own database (Step 61). When the synchronization processing is completed, the date and time of the last synchronization with the master computer is recorded, and a series of synchronization processing on the slave computer side ends (step 89).

Focusing again on the process 1 of the master computer, the process returns to the step 46. When the database synchronization process is completed (step 55), the process 1 of the master computer completes the synchronization of its own database. Synchronization summary information for the database of the slave computer A is created based on the information (step 56). Since the synchronous summary information is the summary information obtained by matching the databases based on the information of both the file groups respectively managed by the slave computer A and the master computer, the synchronization summary information is smaller than the total size of the information of the difference file groups of both computers. Is also small. The synchronization summary information is created in the data processing unit on the master computer. When the slave computer receives the synchronization summary information, the information is analyzed by the data processing unit, and an update execution instruction is sent to the database management unit based on the analysis result to execute the data update process.

When the creation of the synchronization summary information is completed, process 1 of the master computer checks whether communication with the slave computer A is currently ongoing. That is, it is confirmed whether the difference file group is still being transmitted from the master computer to the slave computer and the line is in use (step 57). If all the difference files on the master computer have already been transmitted to the slave computer and the line has been disconnected, the slave computer A can perform synchronization processing. A
There is no need to newly transmit synchronization summary information to the side. Therefore, the synchronization summary information in the database of the slave computer A is discarded (step 59), and the synchronization processing ends.
(Step 60).

If the difference file group on the master computer side is still being transmitted to the slave computer A, the estimated time (t1) required until the transmission of the difference file group from the current time is completed and the created synchronization Time required to transfer the encrypted summary information to slave computer A
(t2) is calculated (step 58), and which information transfer is completed earlier is compared (step 70 in FIG. 3). If the transfer of the synchronization summary information is expected to be completed earlier, a request is made to the process 2 to interrupt the transfer of the differential file group currently being executed (step 7).
1).

In step 70, the estimated time required for completing the transmission of the group of differential files (t1)
In the case where is shorter, there is no time merit due to newly switching the information to be transmitted from the group of difference files to the synchronization summary information. Therefore, the synchronization summary information in the database of the slave computer A is discarded (step 59 in FIG. 2), and the synchronization process ends (step 60).

In step 70, the estimated time (t) required to complete the transmission of the differential file group from the present time is shown.
When 1) is equal to the expected time (t2) required to transfer the generated synchronization summary information to the slave computer A, the process branches to step 59 in the flowchart of FIG. Needless to say, it is also possible to branch to.

Returning to the aforementioned branch step 70, expecting that the transfer of the synchronization summary information will be completed earlier, the process 1
Requests the process 2 during the transfer of the difference file group information to interrupt the transfer (step 71), the process 2 receives the transfer interruption request from the process 1 (step 72), and transfers the data of the difference file group to the slave computer A. Is interrupted (step 73), and the process is terminated (step 74).
On the other hand, the slave computer A receiving the group of difference files detects interruption of information transmission from the transmission source master computer (step 47 in FIG. 2), and suspends reception in response to the interruption (step 48). After the reception is interrupted, the difference file group information that has been received up to now is discarded. (Step 83 in FIG. 3).

On the other hand, the process 1 of the master computer requests the process 2 to suspend the transfer of the difference file group (step 71). The process 2 transmitting the difference file group to the slave computer A (step 49 in FIG. 2)
When the transfer interruption request from the process 1 is received (step 72 in FIG. 3), the data transfer to the slave computer A is interrupted (step 73). Transfer interruption completed in process 2
After (Step 74), the process 2 notifies the process 1 that the transfer interruption process has been completed (Step 75).
Process 2 ends (step 76).

In the process 1 receiving the transfer interruption processing completion notification from the process 2 (step 77), transmission of the synchronization summary information to the slave computer A is started (step 78). (Step 84). When the transmission of the synchronization summary information to the slave computer A is completed (step 79), the master computer notifies the slave computer A of the transmission completion.
(Step 80), the slave computer A receives the transmission completion notification (Step 85). Thereafter, a network disconnection process is performed between the master computer and the slave computer A (steps 81 and 86). The master computer having completed the network disconnection process ends a series of synchronization processes (step 82).

On the other hand, the slave computer A disconnects the network from the master computer (step 86).
After completing the above, based on the synchronization summary information from the master computer, the synchronization processing of the database of the own computer is performed (step 87), the last update date and time with the master computer are recorded, and a series of synchronization processing ends. (Step 89).

[0042]

As described above, according to the present invention,
Since the information exchanged between the master computer and the slave computer is independent of the specifications of the database management unit, for example, a large-scale and expensive database management application is adapted to the master computer, and the client PC is a notebook PC. It is possible to easily adapt an inexpensive simple database application by using a computer with low specifications such as.

When an application product is implemented by the method of the present invention, many commercial database management systems can be easily supported. Also, in the process of synchronizing files between the two computers, the master computer performs parallel processing as much as possible in addition to merely exchanging the difference files, and creates relatively compact synchronous summary information with a small capacity. By comparing the transmission time of the synchronization summary information with the transmission time of the difference file and selecting the shorter one, the file can be synchronized with the shortest possible line connection time.

[Brief description of the drawings]

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

FIG. 2 is an exemplary flowchart showing the procedure of a file synchronization process in FIG. 1;

FIG. 3 is an exemplary flowchart showing the procedure of a file synchronization process in FIG. 1;

[Brief description of reference numerals]

1. Master computer, 2. First slave computer A, 3. nth slave computer B, 4,
... Line, 10 ... Database management unit M on master computer side, 11 ... Data processing unit M on master computer side
12, 20, 24: communication unit, 13: master database, 21, 25: data processing unit S, 22: database management unit A, 23 on the first slave computer A side
Database A on the first slave computer A side,
26: database management unit B on the n-th slave computer B side, 27: n-th slave computer B
Side database B.

Continued on the front page (72) Inventor Kazunari Morita 890 Kashimada, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture, Ltd.Information Systems Division, Hitachi, Ltd. Within the System Division (72) Kazuo Kage, Inventor 890 Kashimada, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture F-Term (Information System Division, Information Systems Division, Hitachi, Ltd.) 5B045 BB19 BB28 BB47 DD17 5B082 AA01 GA04 GA14 GB02 HA03 HA05 5B089 GA21 JA32 JB23 KA05 KA08 KA12 KC23 KC60 KG10

Claims (1)

[Claims] 1. A file group having the same configuration is owned by each of two or more computers (or information processing apparatuses) linked by a line, and another computer holds update processing of a file group held by one computer. File update processing method also applied to a group of files to be
At least one computer performs an update process on the file, and selects either the difference information of the update performed on the file or the data summarizing the update process by selecting the shorter communication time required for transmission and reception. Performing, immediately after the transmission and reception is completed, disconnect the line, after that, the computer receiving the difference information of the update process or the file summarizing the content of the update process, the file group held on its own computer, After receiving the difference information of the received update process or the information summarizing the content of the update process, the information with the shorter reception time is received, the line is disconnected, and the file group is updated based on the information, so that the communication time is increased. And that each computer holds difference information of the update process or information summarizing the content of the update process. By exchanging the data processing unit, a method of synchronizing files between a plurality of computers, characterized in that to absorb the difference in the specifications of the database management unit.