JP2001236256A - Distributed arrangement system for electronic information and distributed arrangement system for data base and remote management system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide one or plural systems arranged at a remote place on which a part or whole part of the original text of electronic information is placed as a manuscript for establishing both high speed information synchronization processing between the original text and the manuscript using file transfer and successive change reflection processing using a change notifying message, and for realizing an operation including the change of the original text by the other system even during the information synchronization processing by the system having the manuscript. SOLUTION: In information synchronization processing, a synchronization information file 115 prepared based on an original text 108 is file transferred 116. Also, a system on which a manuscript 103 is placed is provided with an access exclusion control mechanism 104 for waiting for access to the manuscript 103 by the other processing during the information synchronization processing. Thus, it is possible to prevent the competition of the information synchronization processing with change reflection processing 114 to receive a change notification message 113 issued according to the change of the original text 108 by the other system, and to reflect it on the manuscript 103 while normally operating the system on which the original text 108 is placed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a method for remotely arranging part or all of the original electronic information via an information transfer network such as a LAN (Local Area Network) or a WAN (Wide Area Network). The present invention relates to a distributed information system in which copies are copied as a manuscript to one or a plurality of systems and used for an application arranged in or near the system in which the manuscript is placed, thereby preventing a reduction in response performance due to a concentration of accesses to the original. .

[0002]

2. Description of the Related Art Conventionally, information synchronization between a database original and a database manuscript is performed between a system in which a database as electronic information is set as an original and a system in which a part or all of the database original is set as a manuscript. In general, file transfer is used to achieve high-speed operation. FIG. 2 shows an example.

[0005] In FIG. 2, a system A includes an application 201 and a database management function unit 202, and stores a database copy 203.

[0004] On the other hand, the system B is composed of an original database 2
04, and a database management function unit 205 for managing the original database 204.

[0005] Application 201 in system A
Requests the database management function unit 202 in the system A when referring to the database. The database management function unit 202 in the system A reads information from the database copy 203 in the system A and returns a response to the application 201. Original database 20
Since no communication is performed with the system B storing No. 4, a high-speed response can be realized in the case of referring only to the database.

The application 2 in the system A
01, when the database is changed, first, the system A
Request to that effect from the database management function unit 202 in the server. The database management function unit 202 in the system A
A database change process is executed by issuing a transaction request 206 carrying a change request from the application 201 to the system B and receiving the response 207.

The database management function unit 2 in the system B
05, based on a write-type transaction received by a transaction request 206 issued from the system A or another remote system,
04, a database update history called a journal 208 is created and stored. When a change is made directly to the database original 204 in the system B, the journal 208 is created and stored.

Therefore, the database management function unit 202 in the system A reflects such changes in the database copy 203 in the own system in case that the original database 204 is changed by a system other than the own system. Then, the journal 208 in the system B is periodically transferred to the own system using the file transfer 209, and the database copy 203 in the own system is updated based on the transferred journal 210.

As described above, when updating the database, even if the database copy 204 in the system A in which the change request has been made is first changed after the database original 203 in the system B is changed,
Using 210, the changes are reflected in the database copy 204.

When the database manuscript 203 is destroyed due to a serious failure of the system A, or when the database manuscript 203 is newly created, the database management function unit 202 in the system A performs data extraction. By issuing a transaction request carrying the request to the system B, the database management function unit 205 in the system B is requested to extract data.

The database management function unit 2 in the system B
05 extracts the synchronization information 211 based on the transaction received by the transaction request issued from the system A or another remote system, so the database management function unit 202 in the system A
The extracted synchronization information 211 is transferred to the own system using the file transfer 212, and the transferred synchronization information 2 is transferred.
13 based on the database manuscript 20 in the own system.
By synthesizing / updating the information 3, the information is synchronized. Note that the synchronization information 211 and 213 here are actually
This is the contents of the original database 204 at the time when the data extraction request is received.

In the example of FIG. 2, after the synchronization information 211 is extracted, even if the contents of the database original 204 are changed by a request from a system other than the system A, the journal 208 is regularly transferred and Since the contents (the journal 210) are reflected in the database manuscript 203, the database manuscript 203 can be updated without missing any changes.

[0013] Such a change of the database original 204 that occurred during the information synchronizing process is performed using the database copy 20.
FIG. 3 shows an example of a sequence in which the sequence is reflected in No. 3.

The database management function unit 2 in the system A
In 02, the data reflection function unit first issues a file transfer request 301 to transfer the synchronization information 211 as shown in FIG. This file transfer request 30
1, the synchronization information 211 is transferred from the system B to the system A. That is, in the database management function unit 205 in the system B, the data extraction unit extracts and transfers the synchronization information 211, and when the transfer is completed successfully, transmits a completion response 302.

Therefore, in the database management function unit 202 in the system A, when the data reflection function unit receives the completion response 302 to the file transfer request 301,
Since the synchronization information 211 has been successfully transferred as the synchronization information 213, the database manuscript 203 is updated based on the synchronization information 312 (S303). The above is the information synchronization processing.

In the database management function section 202 of the system A, when the information synchronization processing by the data reflection function section is completed, the data link function section is periodically driven. As shown in FIG. 3, the data link function unit transfers the journal 208 in the system B to the system A by the file transfer 304 and 305, and updates the database manuscript 204 based on the transferred journal 210 ( S306). The above is the database copy updating process.

Here, as shown in FIG. 3, while the information synchronization process is being performed by the data reflection function unit of the database management function unit 202 in the system A, the original database 204 is stored in the system B. Is updated (S307). Although the update content of the database original 204 in S307 is not reflected in the synchronization information 211, when the data linkage function unit of the database management function unit 202 is driven in the system A after the information synchronization process ends. Then, the database copy 203 is reflected in the database copy 203 by the database copy update processing.

Therefore, in the example of FIG. 3, the period of the periodic drive of the data interlocking function unit is from the end of the information synchronization process to the time when the changed contents of the database original 204 are reflected on the database copy 203. The delay is caused by the minute 308.

By the way, the present inventors have devised a system in which the update is performed sequentially by a change notification, instead of the periodic update by reading the journal. FIG. 4 shows an example of the application.

In the example of FIG. 4, the application 201 in the system A accesses only the database copy 203 in the system A when referring to the database, as in the example of FIG.

In the example of FIG. 4, when the application 201 in the system A changes the database, the application 201 requests the database management function unit 202 in the system A to that effect. The change to the database original 204 via the database management function unit 205 in the system B in response to the response 207 is the same as in the example of FIG. 1, but in the example of FIG. Immediately after safely changing the database original 204, the database manuscript 2 in the own system is immediately changed.
03 will also be changed.

In the example of FIG. 4, the database management function unit 205 in the system B
, A database change notification 401 is immediately issued to notify the system A (and other remote systems) of the change. The database management function unit 202 in the system A updates the database copy 203 in its own system based on the received database change notification 401.

It is conceivable that the database management function unit 202 in the system A is executing information synchronization processing when the database change notification 401 is received. In such a case, the database management function unit 202 Is
The received change notification message 401 is stored in the message queue 402, and after the information synchronization process ends,
Change notification message 401 from message queue 402
And the extracted database change notification 401
, The database manuscript 203 in the own system is updated.

In the example of FIG. 4, the database copy 203 in the system A and the original database 2 in the system B
FIG. 5 shows an example of an information synchronization processing sequence for synchronizing information with the information 04.

Here, the information synchronization process is realized by activating a transaction process for each section (record) of information recorded in the database.

That is, as shown in FIG. 5, first, the information synchronization process starts a transaction process (S5).
01). When the transaction process is started, information of one record is acquired from the original database 203 in the system B using a transaction (read-out transaction) (S502, S503), and based on the acquired contents, Database manuscript 20 in system A
4 is updated for one record (S504).

When one record is updated and the transaction processing is completed, the execution right returns to the information synchronization processing (S505).

In the information synchronization process, since the process 506 for one record has been completed, the transaction process is subsequently started, and the process 507 for the next record is performed. As described above, the information synchronization process synchronizes information between the database copy 203 and the database original 204 by repeating the process for all records.

In the example shown in FIG. 5, during the process 507 for the second record, a change is made to the database original 204 in the system B (S508).
At this time, the contents of the change made from the system B to the system A are transmitted by the change notification message 509.

As is clear from the definition, the transaction processing uses a mechanism such as occupation of the execution right to prevent contention for data access due to interruption of other processing. After the processing 507 for the record of the
The database copy 203 is updated based on (S510).

In the example shown in FIG. 5, the database original 2
Even if the change that occurred in 04 is related to the third record, the latest information is read from the original database 203 in the system B for each record. The process 511 does not overwrite the data before the change.

As described above, if the change of the database original 204 is successively reflected in the database copy 203 by using the change notification message 401, the data linkage function unit as described in the example of FIG. There is no delay caused by the drive interval 308.

However, the transaction processing is
In general, since a function such as conflict exclusion during execution or rollback for returning data to the original value at the time of abnormal termination is provided, reliability is high, but processing overhead is large, and therefore, as shown in the example of FIG. When information synchronization processing is performed using transaction processing, the execution time becomes longer as the number of records in the database is larger.

Therefore, the method described with reference to FIGS. 2 and 3 (the method of performing high-speed information synchronization processing using file transfer) and the method described with reference to FIGS. 4 and 5 (the sequential method) In order to make use of the advantages of the method of updating the copy of the digitized information in the above-mentioned method, it is conceivable to combine both methods as shown in FIG. However, a simple combination of both methods results in the following problems.

That is, as shown in FIG. 7, when a change is made to the original database 204 in the system B during the information synchronization processing using file transfer (S701) (S702), the change is made. The contents of the change are transmitted by the change notification message 703 from the system B to the system A, and the database copy 20 in the system A is transmitted.
3 (S704), in the information synchronization processing (S705) to be executed subsequently, the information transferred from the system B to the system A before the change occurs changes the content reflected in S704. However, the information before the change may be overwritten.

FIG. 8 shows a measure for avoiding such a situation, which can be easily devised from the current technology.

Conventionally, when there is a concern that a plurality of systems may access the shared information at the same time and cause some inconsistency, the writing of the shared information is prohibited while a certain system is accessing the shared information. The idea was general.

FIG. 8 is a diagram showing an example of the information synchronization processing sequence in the example of FIG.

In FIG. 8, a shaded portion 812 in the system B indicates that the database original 204 is set to the change prohibition state during the information synchronization processing.

As shown in FIG. 8, in the information synchronization processing, first, the transaction processing is started, and the state attribute of the database original 204 in the system B is changed to the change prohibited state (S801 to S804). After that, in the information synchronization process, the information is synchronized using the file transfer (S805 to S807), and finally, the transaction process is started again, and the database original 2 in the system B is started.
04 is changed to a changeable state (S80).
8-S811).

In this way, the original database 204 in the system B is prohibited from being changed by the system A.
During the transition to 12, even if a system other than the system A attempts to change the database original 204, the process ends in an error (S813).

[0042]

As described above, FIG.
In the example described with reference to FIG. 3 and FIG. 3, the process of reflecting the change of the database original 204 into the database copy 203 is performed by periodically reading the journal. Accordingly, there is a problem that the time required for the change of the database original 204 to be reflected in the database 203 becomes longer.

This is, for example, when applied to a large-scale network management system, the time from the occurrence of a failure in the managed system until the application detects a change in the failure state of the managed system held in the original database. Causes the problem that the length becomes longer.
Therefore, if the period for executing the process of reading the journal is shortened, the process of reading the journal is performed frequently, and the communication that accompanies the information transfer network resources when there is no failure in the managed system. Problem arises.

Also, as described above, in the example described with reference to FIGS. 4 and 5, since the change of the database original 204 is sequentially reflected on the database copy 203 by the change notification message 401, for example, Even when the present invention is applied to a network management system, a change in the failure state of the managed system can be quickly detected in the application. However, since the information synchronization process is executed for each record using a transaction, the communication time is longer than when the information synchronization process is performed by file transfer, and as a result, the time required for the information synchronization process is increased. There is a problem that becomes longer.

In the example of FIG. 8, the problems seen in the above two examples are solved. However, as shown in FIG. 9, when the system C equivalent to the system A shares the database original 204 of the system B, the database original 204 is in a change-prohibited state during the synchronization process of the system A. Transaction 2 from system C
06 has the problem of being rejected.

Therefore, an object of the present invention is to combine a high-speed information synchronization process using a file transfer, a sequential change reflection process using a change notification message, and a copy of electronic information. An object of the present invention is to enable an operation including a change of the original of electronic information by another system even when one or a plurality of systems are in the process of synchronizing information.

[0047]

In order to achieve the above object, according to the present invention, a system for placing a copy of digitized information temporarily stores a change notification message received during execution of information synchronization processing. After the information synchronization process is performed, the contents of the change by the change notification message temporarily stored are reflected in the manuscript. In other words, until the information synchronization processing is completed, the change reflection processing using the change notification message is waited in the copy placing system.

Therefore, the high-speed information synchronization processing using file transfer and the sequential change reflection processing using the change notification message can be performed without any restriction on the function of the system for placing the original electronic information. It is possible to avoid writing conflicts with the manuscript and prevent inconsistency in the contents of the manuscript.

[0049]

Embodiments of the present invention will be described below with reference to the drawings. (First Embodiment) First, a first embodiment of the present invention will be described.

The first embodiment shows an example in which the present invention is applied to a network management system, and FIG.
It is a lineblock diagram of a network management system concerning an embodiment.

In FIG. 1, E where a database copy is placed
M (Element Manager) is a GUI (Graphical User In
terface) process 101, Manager process 102, and MIB (Management Informa) which is a database manuscript under the control of Manager process 102.
tion Base) 103, between EM and other systems,
CMIP (Common Manager) that performs processing related to inter-system communication based on ISO / IEC International Standard 9596-1
t Information Protocol) communication process 105 and FTP (File) for performing a file transfer process related to the information synchronization process.
Transfer Protocol) is a workstation on which a client process 106 is mounted.

The Manager process 105 is based on the ISO
/ POSIX of IEC International Standard 9945-1 (Portable
Pthread Mute, which is a multi-thread mechanism conforming to the Operating System Interface) standard, and is an inter-thread exclusive control process conforming to the POSIX standard
Using x104, access conflict avoidance for the MIB 103 is performed.

In FIG. 1, an ADM (Add Drop Multiplexer) in which an original database is placed is composed of tens to hundreds of "paths" which are time-division multiplexed and pass through a backbone digital transmission line for performing high-speed digital data communication. This is a transmission device having a function (cross-connect function) of branching communication data called to a lower-speed branch digital transmission path or changing the transmission path by exchanging each other. AD
M has a remote monitoring and control interface built on a real-time OS (Operating System) that enables monitoring and control from a remotely located EM.

The remote monitoring control interface includes a CMIP communication task 107 for performing processing related to inter-system communication with the EM, an XC (Cross Connection) map 108 which is an original database and stores setting information of a cross-connect function. That manages the XC map 108
It comprises a task 109 and an FTP server task 110 for performing a file transfer process related to the information synchronization process.

As shown in FIG. 10, two EMs are installed for backup in the event of a system failure, and both are always operating, so that the user can use any EM.

In the example of FIG. 1, the user operates the EM GUI.
When the user attempts to change the setting of the cross-connect function by operating the screen, the Manager process 102 that has received the change request from the GUI process 101 performs the cross-connect function setting change transaction request 111 by inter-system communication via the CMIP communication process 105. Issue

In the ADM, the CMIP communication task 10
A that has received transaction request 111 via
The gent task 102 includes a transaction request 111
After updating the XC map 108 based on the content of
Based on the contents of the C map 108, the setting is changed for the hardware related to the cross connect function. When the setting change is completed, the Agent task 109 sends the transaction response 112 via the CMIP communication task 107.
return it.

Upon receiving the transaction response 112, the Manager process 102 determines that the transaction processing has been completed normally, updates the MIB 103, and responds to the GUI process 101 to that effect. After that, the GUI process 101 displays on the GUI screen to notify the user that the setting change has been normally completed.

As described above, when the user changes the setting of the cross-connect function using one of the two EMs, the Agent task 109 issues a change notification message 113 to both EMs. I do.

The EM Man of the EM not used by the user
In the ager process 102, the change notification message 11
As soon as 3 is received, the notification receiving thread 114 for executing the change notification receiving process is started. Based on the change notification message 113, the notification receiving thread 114
Update the IB103.

Even if a failure occurs in one of the two EMs, for example, the EM (system 0) in FIG. 10, the user can use the other EM (system 1) to perform cross-connect. Function settings can be changed. Then, after the repair of the EM (system 0) is completed, information mismatch occurs between the MIB 103 in the EM (system 0) and the XC map 108 in the ADM. Therefore, EM (0 series) EM
The Manager process 102 executes an information synchronization process.

In the first embodiment, as shown in FIG.
The information transfer relating to the information synchronization process is performed by transferring the path information file 115 created in the ADM to the FTP file process 116 by the FTP client process 106 in the EM and the FTP server task 110 in the ADM. This is performed by transferring the information as a route information file 117 into the server.

FIG. 11 is a sequence diagram when the ADM issues a change notification message 113 while the EM is executing the information synchronization process.

In FIG. 11, a detailed description of the inter-task communication inside the ADM is omitted in order to more clearly show the features of the present invention.

In the EM, first, as shown in FIG. 11, the synchronization thread in the Manager process 102 for executing the information synchronization processing immediately starts the processing, and immediately after the processing starts, the Pthread Mut generated in advance when the process is started.
ex104 is locked by the pthread_mutex_lock function (S1101). PthreadM
If the tex 104 is not locked by another thread, the execution right of the process immediately returns to the synchronization thread. In FIG. 11, PthreadMutex
A hatched portion 1102 applied to the line representing 104 indicates a period during which the synchronization thread locks the Pthread Mutex 104.

The synchronization thread is a Pthread Mu
After locking tex104, by inter-process communication,
It requests the FTP client process 106 to transfer the path information file (S1103). FTP client process 10 that has received a transfer request for the path information file
6 issues an FTP file transfer request message 1104 to the FTP server task 110 in the ADM, and sends the route information file 115 (1
105) is stored in the EM as a path information file 117. When the transfer of the route information file is completed, FTP
The client process 106 notifies the synchronization thread of the completion of the file transfer by the inter-process communication (S
1106) After completing the file transfer, the synchronization thread reads the transferred path information file 117 and updates the MIB 103 (S1107).

When the updating process of the MIB 103 is completed, the synchronization thread locks the lock on the PthreadMutex 104 with the pthread_mutex_unloc.
Release by the k function (S1108).

Here, the synchronization thread is Pthread
When the XC map 108 is changed while the mutex 104 is locked, the ADM issues a change notification message 113 (1109) as shown in FIG. In the EM, the notification receiving thread 114 is activated by the inter-process communication from the CMIP communication process 105 that has received the change notification message 1109 (S1).
110), the activated notification receiving thread 114 starts processing in parallel with the synchronization thread.

That is, the notification receiving thread 114 sets p
Pth by the thread_mutex_lock function
An attempt is made to lock the read mutexes 104 (S1111), but since the Pthread mutexes 104 are locked by the synchronization thread, the pth
The processing is blocked when the read_mutex_lock function is called. Pth by synchronization thread
After unlocking the read mutex 104 (S11
08), the execution right is pthread_mutex_loc
The k function returns to the notification receiving thread 114 (S111).
2).

The notification receiving thread 114 executes the Pthrea
After locking the d Mutex 104, the MIB 103 is updated based on the change notification message 1109 (S1).
113) When the update is completed, Pthread Mute
Lock on x104, pthread_mute
Release by the x_unlock function (S1114),
The change notification receiving process ends. In FIG. 11,
A white rectangle 1115 drawn on the line representing the Pthread Mutex 104 indicates that the notification receiving thread 114
This indicates a period during which the thread Mutex 104 is locked.

FIG. 12 is a sequence diagram when the GUI process 101 issues a route information change request during the execution of the information synchronization process in the EM.

In FIG. 12, as in FIG. 11, a detailed description of communication between tasks in the ADM is omitted.

In FIG. 12, the sequence relating to the synchronization thread in the Manager process 102 for executing the information synchronization processing is the same as that in FIG. 11 (S110).
1,1102, S1103,1104,1105, S1
106 to S1108).

Here, the synchronization thread is Pthread
When the GUI process 101 issues a request for changing the path information by inter-process communication while the Mutex 104 is locked (S1201), a transaction thread is generated and activated in the Manager process 102.

As shown in FIG. 12, as soon as the transaction thread is started, the pthread_mu
By the tex_lock function, Pthread Mute
Attempts to lock x104, but Pthread M
Since the tex 104 is locked by the synchronization thread, the process is blocked when the pthread_mutex_lock function is called, and the Pthread
After waiting for the release of the lock of the d Mutex 104, and after the synchronization thread releases the lock of the Pthread Mutex 104 (S1108), the execution right becomes pthrea.
The process returns to the transaction thread from the d_mutex_lock function (S1202).

The transaction thread that has resumed the processing performs the cross-connect function setting change transaction processing for the ADM (S1203).
After updating the MIB 103 (S1204), the Pthr
lock on the mutex 104
Release by the function "ad_mutex_unlock" (s
1205) The completion of the path information change processing is notified to the GUI process 101 by inter-process communication (S12).
06). (Embodiment 2) Next, a second embodiment of the present invention will be described.

The second embodiment shows an example in which the present invention is applied to a network management system. FIG. 13 is a configuration diagram of a network management system according to the second embodiment.

The difference between the second embodiment and the first embodiment is that the EM in the first embodiment is constituted by two workstations connected via an information transfer network. Is a point. That is, in the second embodiment, as shown in FIG. 13, a GUI application 1306 that performs the function of the GUI process 101
Are implemented in an HMI (Human Machine Interface) workstation, and other functions are implemented in a Server workstation connected to the ADM via a management information transfer network. Therefore, in the second embodiment, the functions realized by the single Manager process 102 having the multi-thread mechanism in the first embodiment are replaced by the transaction processing process 1301, the MIB control process 1302, and the synchronization process 1303. , A notification receiving process 1304.

In the first embodiment, Pthread
In the second embodiment, the semaphore (Semaphore) 130, which is an inter-process exclusion control mechanism provided as standard by the Unix-based OS, is used to avoid access conflicts performed using the Mutex 104 in the second embodiment, as shown in FIG.
5 can be realized.

FIG. 14 is a sequence diagram when the ADM issues a change notification message 113 while the synchronization process 1303 mounted on the EM (Server) is executing information synchronization processing.

The synchronization process 1303 is first executed in FIG.
As shown in FIG.
The IB control process 1302 performs a semaphore value subtraction process on the semaphore 1305 generated by giving the semaphore value = 1 at the time of activation by using a semop function specifying -1 as an argument (S1401). A period 1402 in which the semaphore 1305 has the semaphore value = 0 due to the subtraction processing.
When another process tries to perform the subtraction process of the semaphore value of the semaphore 1305 in the same manner, the process is blocked when the semop function is called.

The synchronization process 1303 continues with the first
Similarly to the case of the embodiment, the transfer of the path information file is performed (S1403), the MIB 103 is updated (S1404), and the semaphore value addition processing of the semaphore 1305 is performed by the semop function specifying +1 as an argument. Performed (S1405).

Here, while the semaphore value = 0 by the synchronization process 1303, the XC map 108
Is changed, the ADM, as shown in FIG.
A change notification message 113 (1406) is issued. E
In M (Server), a change notification message 1
The contents of the change notification message 1406 are transmitted from the CMIP communication process 105 that has received 406 to the notification reception process 1304 by inter-process communication (S14).
07).

Therefore, the notification receiving process 1304 is
The semaphore 1
An attempt is made to subtract the semaphore value of 305, but since the semaphore value of the semaphore 1305 has already been set to 0 by the synchronization process 1303, the process does not return from the semop function, and the semaphore value waits for addition (S1).
408).

After the semaphore value adding process of the semaphore 1305 by the synchronization process 1303 (S1405), the notification receiving process 1304 succeeds in the semaphore value subtracting process (S1409), and updates the MIB 103 (S1409).
410), after adding the semaphore value of the semaphore 1305 using the semop function designating +1 as an argument (S1411), the change notification receiving process ends.

FIG. 15 shows a case where the GUI application 1306 mounted on the EM (HMI) issues a route information change request while the synchronization process 1303 mounted on the EM (Server) is executing information synchronization processing. FIG.

In FIG. 15, the synchronization process 1303
Is the same as that of FIG. 14 (S1401, 1402, S1403 to S140).
5).

Here, the synchronization thread is Pthread
While locking the Mutex 104, the transaction processing process 1301 that has received the route information change request message 1501 from the GUI application 1306 uses the semaphore 1305 with the semop function specifying -1 as an argument, as shown in FIG. Attempt to subtract the semaphore value of
Since the semaphore value of the semaphore 1305 is 0 by 303, the process does not return from the semop function, and the semaphore waits for the semaphore value addition (S1502).

Transaction processing process 1301
After the semaphore value addition processing of the semaphore 1305 by the synchronization process 1303 (S1405), the semaphore value subtraction processing succeeds (S1503). S15
05), and after adding the semaphore value of the semaphore 1305 by the semop function specifying +1 as an argument (S1506), the GUI application 130
6 returns a route information change response message 1507.

By the way, the present inventors have stated that the information synchronization process according to the present invention is completed in a shorter time as compared with the case where a read-out transaction is used. Was measured and confirmed.

FIG. 16 is a configuration diagram of the pseudo environment at that time.

Referring to FIG. 16, a WS (Work) provided with a pseudo-Manager 1601 for managing a copy of digitized information.
station) (A) and WS (B) provided with a pseudo Agent 1602 for managing the original electronic information,
Ethernet LA with basic bandwidth of 0 megabit
N.

In the pseudo environment shown in FIG.
63 held by the pseudo Agent 1602 using a CMIP transaction of a read type from the
When the original of the digitized information having zero data was read, the required time was about 118 seconds.

On the other hand, in the simulated environment shown in FIG. 16, the time required to transfer an electronic file having exactly the same contents as the original to the FTP file is about 2 seconds, and the contents of the electronic file are all converted by the pseudo manager 1601. The time required for reading was about 2 seconds, and the total time required was about 4 seconds.

Thus, in the simulated environment, the time required to transfer the information of the original using the file transfer is about 30 times less than in the case of using the read-type transaction. all right.

As described above, according to the present embodiment, every time a change occurs in the contents of the original of the digitized information, a change notification message is sent from the system in which the original is placed to the system in which the copy of the digitized information is placed. By issuing, the changes can be reflected in the manuscript one by one, and the information synchronization process between the original and the manuscript using file transfer and the processes other than the information synchronization process during the information synchronization process can be performed. By combining with the lock function that prohibits access to the manuscript, even if the system where the manuscript is placed is in the process of synchronizing information, other systems can access the original. . Since the information transfer at the time of performing the information synchronization process is not file-type transaction but file transfer, the information synchronization process can be completed in a short time.

The lock function is a lock function corresponding to the system to be applied (the Pthread Mute described above).
x104, semaphore 1305, etc.).

[0098]

As described above, according to the present invention,
In addition to a high-speed information synchronization process using file transfer and a sequential change reflection process using a change notification message, one or more systems having database manuscripts are performing the information synchronization process. Also, an operation including a change of the original database by another system can be performed.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a network management system according to a first embodiment.

FIG. 2 is an explanatory view showing an example of a conventional regular update distributed arrangement type database system.

FIG. 3 is an information synchronization sequence diagram of a DB copy and a DB original in the example of FIG. 2;

FIG. 4 is an explanatory diagram showing an example of a conventional sequential update type distributed arrangement type database system.

FIG. 5 is an information synchronization sequence diagram between the DB copy and the DB original in the example of FIG. 4;

FIG. 6 is an explanatory diagram showing an example of a sequentially updated distributed arrangement type database system that can be easily devised from the prior art.

FIG. 7 is a sequence diagram showing a problem that occurs when the example in FIG. 4 is combined with information synchronization processing using file transfer.

FIG. 8 is a sequence diagram showing a method for solving the problem shown in FIG. 7;

FIG. 9 is an explanatory diagram of a problem in the example of FIG. 8;

FIG. 10 is a configuration diagram of a network management system according to the first embodiment.

FIG. 11 is a sequence diagram illustrating waiting for a change notification receiving process in the first embodiment.

FIG. 12 is a sequence diagram illustrating waiting for a change transaction process according to the first embodiment;

FIG. 13 is a configuration diagram of a network management system according to a second embodiment.

FIG. 14 is a sequence diagram illustrating waiting for a change notification receiving process according to the second embodiment;

FIG. 15 is a sequence diagram illustrating waiting for a change transaction process according to the second embodiment;

FIG. 16 is a schematic diagram of a pseudo test system used for measurement for confirming the effect of the present embodiment.

[Explanation of symbols]

101: GUI (Graphical User Interface) process, 102: Manager process, 103: MIB
(Management Information Base), 104 ... Pthr
ead Mutex, 105 ... CMIP (Common Manage
ment InformationProtocol) communication process, 106 ... F
TP (File Transfer Protocol) client process, 107 CMIP communication task, 108 XC (Cross
s Connection) map, 109 ... Agent task, 1
10: FTP server task, 111: transaction request, 112: transaction response, 113: change notification message, 114: notification receiving thread, 115,
117: path information file, 116: FTP file transfer, 1301: transaction processing process, 130
2. MIB control process, 1303 synchronization process,
1304: Notification receiving process, 1305: Semaphore (S
emaphore), 1306 GUI application, 1601 pseudo-Manager, 1602 pseudo-A
gent, 201: application, 202: database management function unit, 203: database manuscript, 204
... Database original, 205 ... Database management function part, 206 ... Transaction request, 207 ... Response, 2
08, 210: journal, 209, 212: file transfer, 211, 213: synchronization information, 401: database change notification, 402: message queue.

 ──────────────────────────────────────────────────の Continuing on the front page (72) Inventor Yuichi Takahashi 216 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside the Hitachi, Ltd.Communications Division (72) Masayuki Iijima 180 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Sun (72) Inventor Yoshiyuki Yanaka 180 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture F-term (reference) 5B045 BB47 DD17 5B082 AA01 FA17 GB02 HA03 HA05 5B089 GA21 GB01 HA01 HA06 HB04 JA12 KB04 LB01 ME08

Claims (4)

[Claims] (1) A part or all of the original electronic information is
If the original copy is changed by the system where the original is changed, the contents of the change are sent to the system where the copy is copied by a change notification message. Sequential transmission, the manuscript placing system reflects the change notified by the change notification message in own manuscript, and the manuscript placing system synchronizes information between own manuscript and the original manuscript When performing the information synchronization process, transfer the information necessary for information synchronization using the file transfer from the system where the original is placed to the system where the copy is placed, and the system where the copy is placed However, if the change notification message is received during the information synchronization process, the received change notification message is temporarily stored, and A method of distributing and distributing digitized information, characterized in that, after completion of the information synchronization processing, the contents of the change by the change notification message temporarily stored are reflected in the own copy. 2. A part or all of an original of a database is
Necessary for information synchronization in an information synchronization process for synchronizing information between the database manuscript and the database original by copying it as a database manuscript on one or a plurality of systems and using the application. Is a distributed system in which the transfer of important information is performed by using a file transfer from the system in which the database original is placed to the system in which the database copy is located, wherein the system in which the database original is placed is a change request for the database original. Change means for changing the database original based on the received change request; and, when the database original is changed, change notification means for sequentially transmitting the contents of the change to a system in which the database manuscript is placed by a change notification message. And having The system that places the database manuscript receives the change notification message while the information synchronization process is being performed, and a change reflecting unit that reflects the change content transmitted by the change notification message in its own manuscript. In this case, the access exclusive control means relating to the manuscript, the change request receiving means for receiving a change request for the original database, and the change received by the change request receiving means for awaiting the change reflecting process by the change content reflecting means. A request for notifying the system that stores the original database of the database, and a change request unit for changing its own database copy based on the change request received by the change request receiving unit, the access exclusive control unit includes: While the information synchronization process is being performed, the change request A database distributed arrangement system, wherein when a change request is received, the change request processing by the change request means is made to wait. 3. A remote management system in which part or all of management information in a managed system is copied as management information manuscripts on one or a plurality of management systems and used for an application. A change means for receiving a change request for the management information and changing the management information based on the received change request; and when the management information is changed, the contents of the change are sequentially sent to the management system by a change notification message. A change notifying means for transmitting, and when the management system synchronizes information between the management information and a management information manuscript in the management system, the management information includes synchronization information necessary for information synchronization. Synchronization information extracting means for extracting the management information and the management information of the management system based on the synchronization information. Information synchronization means for synchronizing information with the report manuscript; change reflection means for reflecting the change content transmitted by the change notification message in its own management information manuscript; information synchronization by the information synchronization means Access exclusion control means for the management information manuscript for waiting for the change reflection processing by the change content reflection means when the change notification message is received while the conversion processing is being performed; and Change request receiving means for receiving a change request; and notifying the change request received by the change request receiving means to the managed system, and copying its own management information copy based on the change request received by the change request receiving means. Change request means for changing, wherein the information synchronization means transfers the synchronization information from the managed system. The above-mentioned access exclusion control means performs the change when the change request receiving means receives a change request while the information synchronization processing is being performed by the information synchronization means. A remote management system for waiting for a change request process by a request unit. 4. A management system in which part or all of management information in a managed system is copied as a management information copy and used for application, wherein the management information transmitted from the managed system is provided. Receiving a change notification message indicating the contents of the change, the change reflecting means for reflecting the contents of the change in its own management information manuscript, and synchronizing information between the management information and its own management information manuscript When the synchronization information required for information synchronization transmitted from the management system is received, information between the management information and its own management information manuscript is obtained based on the synchronization information. The information synchronization means for synchronizing the information, and the change content reflection means when the change notification message is received while the information synchronization processing is being performed by the information synchronization means. An access exclusion controller for the management information manuscript for waiting for a change reflection process, a change request receiver for receiving a change request for the management information, and a change request received by the change request receiver for the managed system. And a change request unit that changes its own management information copy based on the change request received by the change request receiving unit. The information synchronization unit performs the synchronization from the managed system. The information transfer is performed by using a file transfer, and the access exclusion control unit is configured to perform the information synchronization process when the change request receiving unit receives the change request while the information synchronization process is performed. A change request process by the change request means.