JP2001159985A - Duplex device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a duplex device capable of relieving processing burden of a processor. SOLUTION: When a processing to change an internal state by communicating with terminals 21, 22 is executed, processors 11, 12 store information regarding the internal state after the change as an update log including processor information (machine IDs 'A', 'B') and update information (sequence number). The update log is simultaneously transmitted to other processors. In addition, when an unstored update log is received from other processors, the processors 11, 12 update the internal state according to the update log and store the update log. The terminals 21, 22 can also be constituted so as to select the processors 11, 12 on the basis of a priority level.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a duplexer.

[0002]

2. Description of the Related Art Conventionally, as a duplication device for duplicating a processing device, for example, a duplication device in which a common storage means is shared by a plurality of processing devices is known. However, this method is expensive because the shared storage means is expensive and the storage means which is relatively easy to fail must be duplicated. Further, there is known a duplex apparatus in which, for example, two processing apparatuses are provided, one of which is set as a main processing apparatus, and the other is set as a standby processing apparatus. In this duplex device, the main processing device normally executes processing by communicating with the terminal device, and stores the processing result in its own storage means. In addition, the processing result stored in the storage unit of the main processing device is transmitted to the preprocessing device, and the processing result stored in the storage device of the preprocessing device is updated. However, the duplexer compares the processing result stored in the storage unit of the main processing unit with the processing result stored in the storage unit of the preliminary processing unit, and compares the processing result stored in the storage unit of the preliminary processing unit. Since the result is updated, the processing load on the update processing of the processing device is large.

[0003]

SUMMARY OF THE INVENTION The present invention has been made in order to solve such a problem, and an object of the present invention is to provide a duplexing apparatus capable of reducing the processing load on each processing unit constituting the duplexing apparatus. The purpose is to provide.

[0004]

According to a first aspect of the present invention, there is provided a duplex apparatus as set forth in claim 1. In the duplex device according to claim 1,
When each processing device communicates with the terminal device and executes a process of changing the internal state, the information on the internal state after the change is stored in the storage unit as an update log including the processing device information and the update information, The update log is transmitted to another processing device. When each processing device receives the update log, the processing device determines whether the received update log is an unstored update log based on the processing device information and the update information included in the update log. When it is determined that the update log has not been stored, the internal state is updated based on the update log, and the update log is stored in the storage unit. As described above, information about the internal state after the change is transmitted in units of update logs, and it is determined whether or not the update log has not been stored based on the processing device information and the update information included in the update log. The processing load related to the update processing of the device can be reduced. Further, each processing device can execute processing in parallel, so that processing performance can be improved. According to a second aspect of the present invention, there is provided a duplex apparatus as set forth in claim 2. In the duplex device according to the second aspect, since the update information is the sequence number, each processing device can easily determine whether the received update log is an unstored update log. This allows
The processing load on each processing device can be reduced. Also,
According to a third aspect of the present invention, there is provided a duplex apparatus as described in claim 3. In the duplex device according to the third aspect, the terminal device selects a processing device based on the determined priority. As a result, the load on each processing device can be dispersed, and the processing load on each processing device can be reduced. A fourth aspect of the present invention is a duplexer as described in claim 4. In the duplex device according to the fourth aspect, when each processing device communicates with the terminal device and executes the process of changing the internal state, the information on the internal state after the change includes the processing device information and the update information. The information is stored as an update log, and the update log is transmitted to another processing device. Further, when each processing device communicates with the terminal device, if the processing device is not the highest priority processing device of the terminal device (if it is a non-highest priority processing device), it executes the processing before executing the process. Synchronization processing is performed with the highest-priority processing device of the terminal device to make the internal state the same as the internal state of the highest-priority processing device. This allows each processing device to execute the synchronization processing with the highest priority processing device only when the processing device itself is not the highest priority processing device of the terminal device with which it communicates. The communication load can be reduced, and the communication load at the normal time can be reduced. Further, a fifth invention of the present invention relates to claim 5
Is a duplexing device as described in (1). In the duplex device according to the fifth aspect, the terminal device determines a priority order based on its own ID. Thus, the priority of each terminal device can be easily determined. A sixth aspect of the present invention is a duplexer as described in claim 6. In the duplex device according to claim 6, each processing device includes:
When the process of changing the internal state is executed, the information about the changed internal state is stored as an update log including the processing device information and the update information, and the update log is transmitted to the management device. When receiving the update log, the management device determines whether the received update log is an unstored update log based on the processing device information and the update information included in the update log. When it is determined that the update log has not been stored, the internal state is updated based on the update log, and the update log is stored in the storage unit. This allows each processing device to transmit the minimum necessary information, and the management device determines whether or not the update log has not been stored based on the processing device information and the update information included in the update log. Therefore, a duplexer can be configured without increasing the processing load related to the update processing of each processing device and the management device. Further, detailed data of each processing device can be collected in the management device. A seventh aspect of the present invention is a duplexer as described in claim 7. If the duplexing device according to claim 7 is used,
Since the update log is stored in the management device for each processing device, it is easy to perform a data totaling process for each processing device.
Thereby, the processing load on the management device can be reduced.

[0005]

Embodiments of the present invention will be described below with reference to the drawings. In this specification, “duplication”
Is used to mean that two or more devices that store the same update log are provided. FIG. 1 is a schematic diagram of a first embodiment of a duplexer according to the present invention. In the embodiment shown in FIG. 1, the processing devices 11 and 12 and the terminal device 2
1, 22 are provided. Although FIG. 1 shows two processing devices and two terminal devices, the number of processing devices and terminal devices can be variously changed. The processing devices 11 and 12 execute processing by communicating with the terminal devices 21 and 22, respectively.
Then, when the process of changing the internal state is performed, for example, when the process of changing the database or information stored in the external storage unit or the internal storage unit is performed, the information on the internal state after the change is The information is stored in the storage unit as an update log including processing device information indicating the processing device that has performed the processing and update information indicating the order of processing in the processing device. Further, when each of the processing devices 11 and 12 receives the update log, it determines whether the received update log is an unstored update log based on the processing device information and the update information included in the update log. When it is determined that the update log has not been stored, the internal state is updated based on the update log, and the update log is stored in the storage unit. As a result, the internal state of each processing device is kept the same, and the processing devices are duplicated. By duplicating the processing devices, even if an abnormality occurs in any one of the processing devices, the processing can be continuously performed by another processing device.

For example, when the processing device 11 communicates with the terminal device 21, performs a predetermined calculation process on the data A, obtains the data B, and executes a process of overwriting and storing the data B on the data A. The processing device 11 stores information indicating that data B is to be overwritten on data A and stores an update log including processing device information indicating the processing device 11 and update information indicating the order of processing in the processing device 11. at the same time,
It transmits to the update log processing device 12. In this case, the information indicating that the data B is overwritten on the data A and stored corresponds to the information indicating the changed internal state. Processing unit 12
Determines whether or not the received update log is not stored based on the processing device information and the update information included in the received update log when the update log is received. When it is determined that the update log has not been stored, the internal state is updated based on the update log. In this case, data B is overwritten on data A and stored. At the same time, the update log is stored. As the information on the internal state after the change, a processing result and the number of a subroutine for processing the processing result may be used. In this case, for example, a subroutine of overwriting and storing is used.

Normally, when a process for changing an internal state is executed by a processing device, a process for searching and referring to a current internal state and storing a result of performing an appropriate calculation process or the like as a next internal state is executed. There is a need to. In the duplexer of the present embodiment, each time the processing device changes the internal state, the processing device stores information on the changed internal state as an update log and transmits the update log to another processing device. . If the received update log is an unstored update log, the other processing device updates the status of the part that is not based on the update log and stores the update log. For this reason, it is not necessary for the other processing devices to perform processing for searching and referring to the internal state, calculation processing, and the like. In the case of the above example, the other processing devices need only perform the process of overwriting data B with data A and storing the data. Therefore, the processing load on the processing device can be reduced. Further, the processing device 11 and the processing device 12 can each execute processing in parallel. Thereby, the processing performance is improved.

An example of the update log will be described below. First, a unique device ID is assigned to devices whose internal states are to be kept the same (hereinafter, referred to as “member devices”). In the present embodiment, the member devices are the processing device 11 and the processing device 12, and the device ID “A” is assigned to the processing device 11 and the device ID “B” is assigned to the processing device 12. The device IDs “A” and “B” correspond to processing device information indicating the processing device that performed the processing. Further, when each of the processing devices 11 and 12 executes the process of changing the internal state, it generates update information indicating the order of the processes. The update information only needs to be able to determine the order of processing for changing the internal state. In the present embodiment, sequence numbers 1, 2, 3,... Are generated as update information. Each of the processing devices 11 and 12 includes a terminal device 21,
In the case where a process of changing the internal state is executed by communicating with the communication device 22, the information on the internal state after the change is processed by the processing device information (the device ID in the present embodiment) and the update information (the sequence No.) is stored in its own storage means as an update log containing the number. At the same time, the update log is transmitted to another processing device. When the other processing device receives the update log, the other processing device determines whether the received update log based on the processing device information and the update information included in the received update log is an update log not stored in its own storage unit. Judge. When it is determined that the received update log is an unstored update log, the internal state is updated based on the update log, and the update log is stored in the storage unit. In FIG. 1, update logs A1, A2, and B are stored in storage units of the processing devices 11 and 12.
1 and B2 are stored. That is, this indicates that the processing device 11 has executed the processing related to the update logs A1 and A2, and the processing device 12 has executed the processing related to the update logs B1 and B2.

Further, at an appropriate timing, the processing device 1
A backup of an internal state called a snapshot is created for 1 and 12, that is, for the member devices.
A snapshot number is assigned to this snapshot. The snapshot number only needs to be able to determine the order of the snapshots. In the present embodiment, sequence numbers such as 1, 2, 3,... Are used as snapshot numbers. In FIG. 1, the update log A
1 and A2 and a snapshot including the update logs B1 and B2 as a snapshot number 1 for each of the processing units 11 and 12
Is created in the storage means. The update information can be generated for each snapshot. For example, even in the snapshot of snapshot number 2,
Update logs A1, A2, B1, B2, etc. can be used.

Each of the processing units 11 and 12 must be set to the same internal state as that of the other processing units at the time of startup, such as when power is turned on. Therefore, each processing device 1
By performing "synchronous processing" at the time of startup,
Before executing the normal processing, the snapshot and the update log stored in another processing apparatus are copied to make the internal states match. In this specification, “synchronous processing” is used to mean processing for making the internal state of a processing device the same as the internal state of another processing device.

The synchronization process is performed, for example, in the following procedure. The activated processing device transmits a synchronization processing signal to another processing device. The processing device receiving the synchronization processing signal,
A synchronous processing signal is transmitted with a response signal including the snapshot number (latest snapshot number) of the latest snapshot among the snapshots stored in its own storage means and the latest update log in the latest snapshot. Is sent to the processing device. Note that instead of transmitting the latest update log, update information of the latest update log may be transmitted. The processing device that has received the response signal (the activated processing device) first determines whether there is a processing device that stores a snapshot newer than the latest snapshot stored in its own storage means. . For example, if the snapshot number is larger as the snapshot is newer, it is determined whether or not any response signal includes the latest snapshot number larger than its own latest snapshot number.

The latest snapshot number larger than the latest snapshot number is included in any one of the response signals (a newer snapshot than the latest snapshot is stored in another processing device).
If it is determined that the latest snapshot is stored, the snapshot and the update log of the processing device storing the newest snapshot are copied. For example, for a processing device that stores a snapshot with the highest snapshot number,
It transmits a transmission request signal for a snapshot and an update log (unstored snapshot and unstored update log) that are not stored in the storage unit of its own processing device. The processing device that has received the transmission request signal transmits the unstored snapshot and the unstored update log to the processing device that has transmitted the transmission request signal. In this case, the unstored snapshot can be determined by comparing the latest snapshot numbers of both processing devices. The unstored update log can be determined by comparing the update information (latest update information) included in the latest update log in the snapshot with the same snapshot number of both processing devices. Therefore, as the transmission request signal, the snapshot number of the unstored snapshot and the update information of the unstored update log can be transmitted, and the transmission request signal including its own latest snapshot number and the update information of the latest update log can be transmitted. Can also be sent.

When the latest snapshot number of the other processing device is the same as its own latest snapshot number, the latest update log in the latest snapshot of the other processing device is more than the latest update log in its own latest snapshot. Is also new or not. For example, it is determined whether or not the update information of the latest update log in the latest snapshot of another processing device is larger than the update information of the latest update log in its latest snapshot. If the update information of the latest update log in the latest snapshot of another processing device is larger than the update information of the latest update log in the latest snapshot of the own device, the update log of the processing device storing the latest update log Copy and store. For example, a transmission request signal for an update log not stored in its own storage unit (unstored update log) is transmitted to the processing device storing the update log of the largest update information. The processing device that has received the transmission request signal transmits the unstored update log to the processing device that has transmitted the transmission request signal.

When a plurality of duplicated processing devices are operating normally, the internal state of each processing device should be kept the same at the next startup. That is, the latest snapshot number is the same, and the update information of the latest update log in the snapshot with the latest snapshot number is also the same. In such a case, when it is confirmed that the latest snapshot number and the latest update information match, the synchronization processing ends, and normal processing can be started immediately. By executing such a synchronization process, the internal state of all member devices becomes the same.

After performing the synchronization process to make the internal states of all the processing units the same, each of the processing units 11 and 12 creates an update log and stores its own storage every time the processing for changing the internal state is executed. And the other processing devices 12, 11
Send to Thus, when the internal state is changed in one of the processing devices, the internal state of another processing device is similarly changed. For example, when the storage state of the storage units of the processing devices 11 and 12 is in the state shown in FIG. 1 and the processing unit 11 communicates with the terminal device 21 to execute the process of changing the internal state, the storage unit of the processing device 11 Is changed from the state shown in FIG. 2A to the state shown in FIG. 2B. That is, when the processing device 11 executes the process of changing the internal process, the processing device 11 stores the update log A3 regarding the changed internal state in the storage unit. Then, the processing device 11 transmits the update log A3 to the processing device 12. Thereby, the processing device 12 updates the internal state based on the update log A3, and stores the update log A3 in the storage unit. That is, the storage state of the storage means of the processing device 12 is also the same as that shown in FIG.
The state shown in FIG. As a result, the internal state of all the processing devices 11 and 12 is kept the same.

When a plurality of processing devices can simultaneously change the same internal state, the synchronization process should be performed not only when each processing device is started but also when each processing device is operating. preferable. For example, immediately before each processing device changes the internal state, synchronization processing is performed at predetermined time intervals or the like. In this case, first, confirmation processing is performed to determine whether or not there is a snapshot (unstored snapshot) or update log (unstored update log) not stored in its own storage means in another processing device. As the confirmation processing, for example, the latest snapshot number and the latest update log are transmitted to another processing device. Note that instead of transmitting the latest update log, update information of the latest update log may be transmitted. The other processing unit compares the latest snapshot number of its own with the latest snapshot number of the other party, or compares the latest update log of its own with the latest update log of the other party, to store the snapshots not stored in the other party. It is determined whether there is an unstored update log. Then, if there is no unstored snapshot or unstored update log in another processing device, the synchronization process ends and the process is immediately executed. On the other hand, if there is an unstored snapshot or unstored update log in another processing device, the unstored snapshot or unstored update log is copied from the other processing device.

In the above description, each of the processing units 11 and 12 is configured to create and store a snapshot. However, it is also possible to omit the creation of a snapshot and store only an update log. . In this case,
Each of the processing devices 11 and 12 performs only a synchronization process on the update log.

In the first embodiment, the processing units 11, 1
2 can communicate with either of the terminal devices 21 and 22, so that it is possible to achieve redundancy while enabling parallel processing. In this case, if the priority order in which the terminal devices 21 and 22 select the processing devices 11 and 12 is not determined, communication may be concentrated on one of the processing devices 11 and 12 and the processing load may be concentrated. Therefore, the terminal device 2
If the priority order for selecting the processing devices 11 and 12 is determined for each of the processing devices 1 and 22, the processing load on the processing devices 11 and 12 can be distributed. A second embodiment capable of dispersing the processing load of the processing devices 11 and 12 will be described below. In the present embodiment, the terminal devices 21 and 22
When communicating with the processing devices 11 and 12, first, a connection request signal is transmitted to the processing device with the highest priority (the processing device with the highest priority). If the connection request signal is not received, the connection request signal is transmitted to the processing device having the next highest priority. In the same manner, communication is performed with the processing devices 11 and 12 that have received the connection request signal.

The processing devices 11 and 12 are connected to the terminal devices 21 and 22, respectively.
Various methods are conceivable as a method of determining the priority for selecting. For example, a method is possible in which priorities are set in advance in the terminal devices 21 and 22 so that the processing loads of the processing devices 11 and 12 are substantially equal. This method
This is effective when the number of processing devices 11 and 12 and the processing load of each processing device 11 and 12 are known in advance. Further, a method is also possible in which the terminal devices 21 and 22 automatically determine the priority according to the number of the processing devices 11 and 12 and the processing load of the processing devices 11 and 12. This method uses the processing apparatus 1
When the number of 1, 12 is not known, or when each processing device 11, 12
This is effective in cases where the processing load of the system tends to fluctuate.

An example of a method for automatically determining the priority in each terminal device will be described with reference to FIG. The duplexer shown in FIG. 3 includes three duplicated processors 31 to 33 for processing bank deposits and terminal devices 41 and 42 for performing bank deposit operations (bank deposit instructions). I have.
The processing devices 31 to 33 have “bank deposit” as the service name.
Is set. Service numbers from "1" to
“3” is set. The terminal devices 41 and 42 search for a processing device that can execute the bank deposit processing at the time of activation or when performing a bank deposit operation such as a deposit operation or a withdrawal operation. For example, the service name “bank deposit” is transmitted from the terminal devices 41 and 42 to the communication line. When the processing devices 31 to 33 receive the service name “bank deposit”,
It determines whether the received service name matches its own service name, and if they match, sends a response signal. The response signal includes, for example, address signals of the processing devices 31 to 33, service numbers “1” to “3” set in the processing devices 31 to 33, and the like.

Upon receiving the response signal, the terminal devices 41 and 42 determine the priority of selecting the processing devices 31 to 33 based on the response signal. As a method of determining the priority based on the response signal, for example, there is a method of prioritizing the order of the service numbers. Alternatively, each terminal device 41, 42
There is a method in which a remainder obtained by dividing the ID by the number of processing devices 31 to 33 is set as a priority. For example, when the remainder is “0”, the processing devices 31, 32, and 33 are selected in this order, and when the remainder is “1”, the processing devices 32, 33, and 31 are selected in this order, and the remainder is “2”. In the case, the processing devices 33, 31, 32
Select in order. When this method is used, each of the processing devices 31 to 31
33 can be evenly distributed. Also, the processing devices 31 to 33 periodically send the terminal devices 41 and 4.
2 may report the processing load status of each of the processing devices 31 to 33, and this processing load may be given a priority. For example, when the terminal devices 41 and 42 communicate with the processing devices 31 to 33, the processing devices 31 are arranged in ascending order of the processing load at that time.
Select ~ 33. When this method is used, each processing device 3
The processing loads of 1 to 33 can be dynamically distributed.

When the method of determining the priority order is used for the terminal device, it is possible to reduce the communication load at the normal time. That is, since the terminal device connects in order from the processing device with the highest priority, as long as the terminal device communicates with the processing device with the highest priority, an update log related to the terminal device may not be generated in another processing device. Absent. Therefore, the processing device performs the synchronization process during operation (for example, before executing the process) only when the processing device is not the highest priority processing device of the communication terminal device, and performs communication with the terminal device. If the processing device has the highest priority, the synchronization process during operation can be omitted. In this way, the processing device reduces the communication load during operation by configuring the processing device to perform the synchronization process during operation only when the processing device is not the highest priority processing device of the terminal device that is performing communication. can do.

When the synchronization process is performed, first, a confirmation process is performed to determine whether or not there is an unstored update log in its own storage means in another processing device. If you configure it to copy the stored update log,
The communication load during operation can be further reduced.
Further, the information about the internal state after the change is stored as an update log including the processing device information and the update information. However, the information about the internal state after the change may be stored as an update log including the update information. In this case, each processing device
It is necessary to transmit the update log and the processing device information to another device and store the update log for each processing device information.
As described above, the process of “storing the information on the internal state after the change as an update log including the update information and transmitting the update log and the processing device information to another device” is also described in the claims. The information about the subsequent internal state is stored as an update log including the processing device information and the update information, and the update log is transmitted to another device. "

In the above description, the duplication device is configured by storing the update log in each processing device. However, the duplication device may be configured by storing the update log in the processing device and the management device. 2. Description of the Related Art Conventionally, for example, a management system is known in which data such as sales trends of stores provided in each district (site) is managed by a management device provided in a headquarters. In the conventional management system, each store and the management device are connected via a low-speed and expensive communication line such as a telephone line.
For this reason, the amount of communication is reduced by using a method of compressing and transmitting communication data, or a method of transmitting the sum of sales results on the day, for example, instead of detailed data. In this case, it is necessary to create a transmission / reception program for each data, and the development cost is high. Further, detailed data cannot be aggregated. If the duplexer of the present invention is applied to such a management system, there is no need to create a transmission / reception program for each data, and the development cost is reduced. In addition, the processing device may transmit the update log to the management device in units of update logs, and the management device may determine whether the update log is not stored based on the processing device information and the update information included in the update log. Therefore, a duplexer can be configured while reducing the processing load on the processing device and the management device. Further, the communication amount can be suppressed to a necessary minimum, and detailed data can be totaled.

FIG. 4 shows a schematic configuration diagram of a third embodiment in which a duplexer is constituted by a processing device and a management device. The duplexer of the present embodiment includes processing devices 51 to 53 such as store servers provided for each region (site), and a management device 61 provided in the headquarters or the like. The processing devices 51 to 53 and the management device 61 include a telephone line,
They are connected by communication lines such as the Internet and wireless lines. In some cases, one of the processing devices 51 to 53 is also used as the management device 61. The processing devices 51 to 53 are given unique device IDs (processing device information). In the present embodiment, the processing device 51 has “Site 1 / Server 1”, the processing device 52 has “Site 2 / Server 1”, and the processing device 53 has
"Site 3 / Server 1" is assigned to the server. The management device 61 has a management file for storing an update log for each of the processing devices 51 to 53.

When each of the processing devices 51 to 53 communicates with a terminal device such as a clearing terminal device (for example, a register in each store) and executes processing for changing internal processing, the processing device 51 to 53 relates to the internal state after the change. The information is stored as an update log including processing device information indicating the processing device that has performed the processing and update information indicating the order of processing in the processing device. Synchronous processing with the management device 61 is performed at the same time as the storage or at an appropriate interval or before the management device 61 refers to the data of the corresponding processing device. That is, each of the processing devices 51 to 53 detects update logs generated by each of the processing devices 51 to 53 that the management device 61 does not have, and collectively sends the update logs.
The management device 61 updates the internal state in association with the processing device information based on the received update log, and stores the received update log in the management file in association with the processing device information. For example, the storage state of the storage unit of the processing device 51 is as shown in FIG.
In the state shown in FIG.
When the book is sold, the storage state of the storage unit of the processing device 51 is changed to the state shown in FIG. That is,
In the storage unit of the processing device 51, information indicating that sales information indicating that “three bottles of drinking water were sold at 9:35” is additionally stored is stored in the processing device information “ It is stored as an update log “NA4” including “site 1 / server 1” and update information “4”. And the processing device 5
1 transmits the update log “NA4” to the management device 61. When receiving the update log “NA4”, the management device 61 updates the internal state based on the update log “NA4” and stores the update log “NA4”. That is,
Sales information indicating that "three bottles of drinking water were sold at 9:35" is additionally stored in the management file corresponding to the processing device information "site 1 / server 1". This allows
The storage state of the management file corresponding to “Site 1 / Server 1” of the management device 61 is also the state shown in FIG. Thus, the storage state of the storage unit of the management device 61 is
The storage state of each of the processing units 51 to 53 is the same as the storage state.

Update logs stored in the processing devices 51 to 53 and the management device 61 when the processing devices 51 to 53 and the management device 61 are activated or when the processing devices 51 to 53 are operating (for example, during regular communication). Are synchronized. In this case, the management device 61 includes the processing devices 51 to 53
And the update log stored in the management file corresponding to each of the processing devices 51 to 53.

As described above, the same update log as the update log stored in the storage unit of each of the processing units 51 to 53 is stored in the storage unit of the management unit 61 for each of the processing units 51 to 53. You. Therefore, the management device 61 can easily total the data for each processing device. In this case, since each of the processing devices 51 to 53 transmits the update log to the management device, the communication amount is small. Therefore, detailed data can be transmitted from each of the processing devices 51 to 53 to the management device 61. In the third embodiment, the management device 61 stores the update log of each of the processing devices 51 to 53 for each of the processing devices. However, it is not always necessary to store the update log for each of the processing devices. Even in this case, since the update log includes the processing device information, data for each processing device can be totaled. Further, the data regarding the internal state after the change is stored as an update log including the processing device information and the update information. However, the data regarding the internal state after the change is stored as an update log including the update information, and the update log and the processing device information are stored. Can be transmitted to the management device. As described above, the process of “storing the information on the internal state after the change as the update log including the processing update information and transmitting the update log and the processing device information” is also described in the claim. Is stored as an update log including processing device information and update information, and the update log is transmitted. "

In the above embodiment, the case where the processing device that communicates with the terminal device and executes the predetermined processing is duplicated has been described. However, the duplexing device of the present invention is applicable to the case where various devices are duplicated. Can be. In this case, a device that executes the process of changing the internal state corresponds to the processing device of the present invention, and a device that performs a processing operation or a processing instruction corresponds to the terminal device of the present invention. Further, each processing device transmits the update log every time the process for changing the internal state is executed, but the processing device may be configured to transmit the update log at a predetermined time or at an arbitrary time. Various methods can be used to copy the unstored snapshot or the unstored update log.

[0030]

As described above, the use of the duplexer according to the first aspect can reduce the processing load on each processor and improve the processing performance. Further, the use of the duplexer according to the second aspect makes it possible to reduce the processing load on each processor. Further, the load of each processing device can be distributed by using the duplexing device according to the third aspect. In addition, by using the duplexing device according to the fourth aspect, the processing load on each processing device can be reduced, and the communication load during normal operation can be reduced. Further, the use of the duplexer according to the fifth aspect makes it possible to easily determine the priority of each processing device. Further, by using the duplication device according to claim 6, the duplication device can be configured while reducing the processing load on each processing device, and detailed data can be collected in the management device. Further, the use of the duplexing device according to claim 7 facilitates data totaling processing for each processing device.

[Brief description of the drawings]

FIG. 1 is a schematic view of a first embodiment of a duplexer according to the present invention.

FIG. 2 is a diagram illustrating a state where a storage state of a storage unit of the processing device according to the first embodiment is changed.

FIG. 3 is a schematic view of a second embodiment of a duplexer according to the present invention.

FIG. 4 is a schematic view of a third embodiment of a duplexer according to the present invention.

FIG. 5 is a diagram illustrating a state in which a storage state of a storage unit of a processing device according to a third embodiment is changed.

[Explanation of symbols]

 11, 12, 31 to 33, 51 to 53 Processing device 21, 22, 41, 42 Terminal device 61 Management device

Claims (7)

[Claims] An information processing apparatus comprising: at least two processing devices; and a terminal device capable of communicating with the processing devices.
When a process of changing the internal state is performed by communication with the terminal device, information on the changed internal state is stored as an update log including processing device information and update information, and the update log is stored in another process. A duplex device that transmits an update log that has not been stored from another processing device to an internal state based on the update log and stores the update log when the update log is transmitted to another device. 2. The duplexer according to claim 1, wherein:
The duplication device, wherein the update information is a sequence number. 3. The duplexing device according to claim 1, wherein the terminal device selects a processing device based on a priority when communicating with the processing device. 4. At least two processing devices, and a terminal device capable of communicating with the processing devices, wherein the processing devices are:
When a process of changing the internal state is performed by communication with the terminal device, information on the changed internal state is stored as an update log including processing device information and update information, and the update log is stored in another process. When transmitting to the device and communicating with the terminal device, if the terminal device is not the highest priority processing device of the terminal device, a synchronization process is performed with the highest priority processing device before executing the process. And when receiving an unstored update log from another processing device, updates the internal state based on the update log and stores the update log. 5. The duplexing device according to claim 3, wherein the priority is determined based on an ID of each terminal device. 6. A processing device comprising: a plurality of processing devices; and a management device capable of communicating with the processing device, wherein when the processing device executes a process of changing an internal state, information on the internal state after the change is provided. Is stored as an update log including update information, and the update log and the processing device information are transmitted to the management device. When the management device receives an unstored update log from each processing device, the management device A duplexer that updates the internal state based on the information and stores the update log. 7. The duplex device according to claim 6, wherein:
The management device is a duplex device that stores an update log for each processing device.