JP7164175B2 - DISTRIBUTED FILE DEVICE, FAILOVER METHOD, PROGRAM AND RECORDING MEDIUM - Google Patents

Description

The present invention relates to a distributed file device, failover method, program and recording medium.

There are two methods of data management in a distributed file device between an active device and a standby device (backup device): synchronous copy and asynchronous copy. Of the two methods, synchronous copy is preferable as a backup. However, a synchronous copy takes time to process if both are remote, for example in different geographically separated cities, so that the same disaster does not affect both the working and standby devices. requires. Therefore, asynchronous copying is preferable for shortening the processing time (see, for example, Patent Document 1).

JP 2018-136596 A

However, the asynchronous copy has a problem that when the execution of the application is handed over from the active device to the standby device due to the occurrence of a failure, it is not clear at what point in time the data copied from the active device to the standby device.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide an asynchronous copy distributed file device in which the point in time of data copied from an active device to a standby device is clarified, and a failover method using the distributed file device.

In order to achieve the above object, the distributed file device of the present invention
a first device executing an application;
a second device that takes over execution of the application when the first device fails;
including
The first device includes a first control unit, a first storage unit, and a reflection execution unit,
The second device includes a second control unit, an unreflected data storage unit, and a second storage unit,
The first control unit and the second control unit are connectable via a communication network,
The first control unit registers data related to execution of the application in the first storage unit with a flag indicating that the data has not been reflected in the second device, and adds a transmission time to the data. Send to the second control unit,
the second control unit registers the data to which the transmission time is attached in the unreflected data storage unit;
The reflection execution unit, when a preset trigger condition is satisfied, instructs the second control unit to reflect data that has not been reflected in the second device to the second device. to
When receiving the instruction, the first control unit instructs the second control unit to reflect the unreflected data;
Upon receiving the instruction, the second control unit changes the transmission time attached to the unreflected data registered in the unreflected data storage unit to the checkpoint time, and then transfers the data to the registering in the second storage unit, and transmitting a completion signal for reflecting the data to the first control unit;
The first control unit sets a flag indicating that the data has not been reflected in the second device and clears the flag of the data registered in the first storage unit when the completion signal is received.

The failover method of the present invention is
a first device executing an application;
a second device that takes over execution of the application when the first device fails;
using a distributed file device containing
The first device includes a first control unit, a first storage unit, and a reflection execution unit,
The second device includes a second control unit, an unreflected data storage unit, and a second storage unit,
The first control unit and the second control unit are connectable via a communication network,
The first control unit registers data related to execution of the application in the first storage unit with a flag indicating that the data has not been reflected in the second device, and adds a transmission time to the data. a non-reflected data processing step for transmitting to the second control unit;
an unreflected data registration step in which the second control unit registers the data to which the transmission time is attached in the unreflected data storage unit;
The first control unit, when the reflection execution unit satisfies a preset trigger condition, instructs the second control unit to reflect data that has not been reflected in the second device to the second device. a first reflection instruction step of instructing to
a second reflection instruction step in which, when receiving the instruction, the first control unit instructs the second control unit to reflect the unreflected data;
When receiving the instruction, the second control unit changes the transmission time attached to the unreflected data registered in the unreflected data storage unit to the checkpoint time, and then transfers the data to the a reflection step of registering the data in a second storage unit and transmitting a reflection completion signal of the data to the first control unit;
a registration step in which, when the completion signal is received, the first control unit raises a flag indicating that the data has not been reflected in the second device and clears the flag of the data registered in the first storage unit;
A method comprising:

According to the present invention, it is possible to provide an asynchronous copy distributed file device in which the point in time of data copied from an active device to a standby device is clarified, and a failover method using the distributed file device.

FIG. 1 is a block diagram showing an example configuration of a distributed file device according to the first embodiment. FIG. 2 is a block diagram showing an example of data flow in a steady state in which the first device does not issue a data reflection instruction to the second device in the distributed file device of the first embodiment. FIG. 3 is a block diagram showing an example of data flow in the distributed file system according to the first embodiment in a state in which a data reflection instruction has been issued from the first device to the second device. FIG. 4 is a block diagram showing an example configuration of a distributed file device according to the second embodiment. FIG. 5 is a block diagram showing an example configuration of a distributed file device according to the third embodiment. FIG. 6 is a block diagram showing an example configuration of a distributed file device according to the fourth embodiment. FIG. 7 is a block diagram showing an example configuration of a distributed file device according to the fifth embodiment. 8 is a block diagram showing an example of the hardware configuration of the first device in the distributed file system of the first embodiment; FIG. 9 is a block diagram showing an example of a hardware configuration of a second device in the distributed file system of the first embodiment; FIG. FIG. 10 is a flow chart showing an example of processing in the distributed file device according to the first to third embodiments. FIG. 11 is a flow chart showing an example of processing in the distributed file system according to the fourth embodiment. FIG. 12 is a flow chart showing an example of processing in the distributed file system according to the fifth embodiment. FIG. 13 is a flow chart showing another example of processing in the distributed file system of the fifth embodiment.

In the present invention, for example, the first device may be the active device and the second device may be the standby device (backup device).

In the present invention, "failover" means, for example, that the second device (standby device) takes over the execution of an application on the failed first device (active device).

In the distributed file system and failover method of the present invention, the preset trigger condition may be elapse of a predetermined time or arrival of a predetermined time.

In the distributed file system and the failover method of the present invention, the unreflected data storage unit has a data amount threshold, and the preset trigger condition is such that the amount of data registered in the unreflected data storage unit is It may be that the threshold value is exceeded.

In the distributed file system and failover method of the present invention, the preset trigger condition may be that the application issues a checkpoint acquisition request to the first control unit.

In the distributed file system of the present invention, when part or all of the unreflected data to which the transmission time is attached and which is registered in the unreflected data storage unit is lost, the first controller controls the second device and sending a difference between the data registered in the first storage unit and the data registered in the unreflected data storage unit to the second control unit. may be According to this aspect, it is possible to restore the state before the unreflected data is lost from the unreflected data storage unit in a short period of time.

In the failover method of the present invention, when part or all of the unreflected data with the transmission time registered in the unreflected data storage unit is lost, the first control unit instructs the second device to a difference transmission step of transmitting to the second control unit the difference between the data registered in the first storage unit and the data registered in the unreflected data storage unit with a flag indicating that it is not reflected; may contain.

In the distributed file system of the present invention, when a failure occurs in the first device and the second device takes over the execution of the application, the second control unit satisfies either condition 1 or condition 2 below. It may be a mode of selecting.
(Condition 1)
Only the reflected data with the checkpoint time registered in the second storage unit is used for execution of the application.
(Condition 2)
The application stores the reflected data with the checkpoint time registered in the second storage unit and the unreflected data with the transmission time registered in the unreflected data storage unit. used for execution.

In the failover method of the present invention, when a failure occurs in the first device and the second device takes over execution of the application, the second control unit selects either condition 1 or condition 2 below. A selection step may be included.
(Condition 1)
Only the reflected data with the checkpoint time registered in the second storage unit is used for execution of the application.
(Condition 2)
The application stores the reflected data with the checkpoint time registered in the second storage unit and the unreflected data with the transmission time registered in the unreflected data storage unit. used for execution.

In the distributed file system of the present invention, when the second control unit selects the condition 1, the time required for the second device to start the application, and when the condition 2 is selected, the second control unit It is also possible to predict the time required for the device to start the application.

In the failover method of the present invention, the second control unit controls the time required for the second device to start the application when the condition 1 is selected, and the second device when the condition 2 is selected. to launch the application.

A program of the present invention is a program capable of executing the failover method of the present invention on a computer.

A recording medium of the present invention is a computer-readable recording medium recording the program of the present invention.

Next, an embodiment of the present invention will be described with reference to FIGS. 1 to 13. FIG. The present invention is not limited or restricted by the following embodiments. In addition, in FIGS. 1 to 13, the same parts are denoted by the same reference numerals. The descriptions in each embodiment can be used with each other.

[Embodiment 1]
FIG. 1 is a block diagram showing the configuration of an example of a distributed file device according to this embodiment. As shown in FIG. 1, the distributed file system 10 includes a first device 11 that executes an application 20 and a second device 12 that takes over execution of the application 20 when the first device 11 fails. In the distributed file system 10, for example, the first device 11 may be the active device and the second device 12 may be the standby device (backup device). The first device 11 includes a first control unit 11a, a first storage unit 11b, and a reflection execution unit 11c. The second device 12 includes a second control section 12a, an unreflected data storage section 12b, and a second storage section 12c. The first control unit 11 a and the second control unit 12 a can be connected via a communication network 30 . The communication network 30 is not particularly limited, and a known network can be used. For example, it may be wired or wireless. The communication line network 30 includes, for example, the Internet line, WWW (World Wide Web), telephone line, LAN (Local Area Network), WiFi (Wireless Fidelity), and the like. The first device 11 and the second device 12 may be, for example, personal computers (PCs).

FIG. 8 illustrates a block diagram of the hardware configuration of the first device 11. As shown in FIG. The first device 11 has, for example, a CPU (Central Processing Unit) 111, a memory 112, a storage device 114, an input device 115, a display 116, a communication device 117, and the like. Each unit of the first device 11 is interconnected via a bus 113 .

FIG. 9 illustrates a block diagram of the hardware configuration of the second device 12. As shown in FIG. The second device 12 has, for example, a CPU (Central Processing Unit) 121, a memory 122, a storage device 124, an input device 125, a display 126, a communication device 127, and the like. Each part of the second device 12 is interconnected via a bus 123 .

The CPU 111 is responsible for overall control of the first device 11 . In the first device 11, the CPU 111 executes, for example, the program of the present invention and other programs, and reads and writes various information. Specifically, for example, the CPU 111 functions as the first control unit 11a and the reflection execution unit 11c.

The CPU 121 is responsible for overall control of the second device 12 . In the second device 12, the CPU 121 executes, for example, the program of the present invention and other programs, and reads and writes various information. Specifically, for example, the CPU 121 functions as the second control unit 12a.

The bus 113 can also be connected to external equipment, for example. Examples of the external device include an external storage device (external database, etc.), a printer, and the like. The first device 11 can be connected to a communication network (not shown), for example, by a communication device 117 connected to the bus 113, and can also be connected to the external device via the communication network. Similarly, the bus 123 can also be connected to external equipment, for example. Examples of the external device include an external storage device (external database, etc.), a printer, and the like. The second device 12 can be connected to a communication network (not shown), for example, by a communication device 127 connected to the bus 123, and can also be connected to the external device via the communication network.

The memory 112 includes, for example, main memory, which is also referred to as main memory. When the CPU 111 performs processing, for example, the memory 112 reads various operating programs such as the program of the present invention stored in a storage device 114 to be described later, and the CPU 111 receives data from the memory 112 and processes the programs. to run. The main memory is, for example, RAM (random access memory). Memory 112, for example, further includes ROM (read only memory). Similarly, memory 122 includes, for example, main memory, which is also referred to as main memory. When the CPU 121 performs processing, for example, the memory 122 reads various operation programs such as the program of the present invention stored in a storage device 124 described later, and the CPU 121 receives data from the memory 122 and processes the programs. to run. The main memory is, for example, RAM (random access memory). Memory 122, for example, further includes ROM (read only memory).

The storage device 114 is also called a so-called auxiliary storage device, for example, in contrast to the main memory (main storage device). As described above, the storage device 114 stores operating programs including the program of the present invention. The storage device 114 includes a first storage section 11b. Storage device 114 includes, for example, a storage medium and a drive that reads from and writes to the storage medium. The storage medium is not particularly limited, and may be, for example, a built-in type or an external type. Examples include DVD, flash memory, memory card, etc., and the drive is not particularly limited. The storage device 114 may be, for example, a hard disk drive (HDD) in which a storage medium and drive are integrated.

The storage device 124 is also called a so-called auxiliary storage device, for example, in contrast to the main memory (main storage device). As described above, the storage device 124 stores operating programs including the program of the present invention. The storage device 124 includes an unreflected data storage section 12b and a second storage section 12c. Storage device 124 includes, for example, a storage medium and a drive that reads from and writes to the storage medium. The storage medium and the drive are the same as in the storage device 114 described above. The storage device 124 may be, for example, a hard disk drive (HDD) in which a storage medium and drive are integrated.

The first device 11 further has an input device 115 and a display 116, for example. The input device 115 is, for example, a touch panel, keyboard, mouse, or the like. Examples of the display 116 include an LED display and a liquid crystal display. Similarly, the second device 12 also has an input device 125 and a display 126, for example. The input device 125 is, for example, a touch panel, keyboard, mouse, or the like. The display 126 may be, for example, an LED display, a liquid crystal display, or the like.

In the first device 11, the memory 112 and the storage device 114 can also store access information and log information from users, and information obtained from an external database (not shown). Similarly, in the second device 12, the memory 122 and storage device 124 can store access information and log information from users, as well as information obtained from an external database (not shown).

Next, an example of processing in the distributed file device 10 will be described.

First, in the distributed file system 10, an example of processing in a steady state in which the first device 11 does not issue a data reflection instruction to the second device 12 will be described with reference to the block diagram of FIG. 2 and the flowchart of FIG. .

First, the first control unit 11a registers data related to the execution of the application 20 in the first storage unit 11b with a flag indicating that it has not been reflected in the second device 12 (as unreflected data). The transmission time is attached to the data and transmitted to the second control unit 12a (S1).

Next, the second control unit 12a registers the data with the transmission time in the unreflected data storage unit 12b (S2).

Next, in the distributed file system 10, an example of processing in a state in which the first device 11 issues a data reflection instruction to the second device 12 will be described with reference to the block diagram of FIG. 3 and the flowchart of FIG.

First, when a preset trigger condition is satisfied, the reflection execution unit 11c performs first control to instruct the second control unit 12a to reflect data that has not been reflected in the second device 12 to the second device 12. The part 11a is instructed (S3). According to the present embodiment, since asynchronous copying is employed in which an instruction to reflect data to the second device 12 is given when a preset trigger condition is satisfied, for example, the first device 11 and the second device 12 Even if the copy is remote, the processing time can be shortened compared to synchronous copy. In the present embodiment, the preset trigger condition is the elapse of a predetermined period of time (for example, 12 hours, 24 hours, etc.), and the instruction is given periodically, for example, every 12 hours, every 24 hours, etc. be. Further, in the present embodiment, the preset trigger condition is a predetermined time (for example, 1:00, 20:00), etc., for example, periodically at 1:00, 20:00, etc. Instructions may be given.

Next, when receiving the instruction, the first control unit 11a instructs the second control unit 12a to reflect the unreflected data (S4).

Next, when receiving the instruction, the second control unit 12a changes the transmission time attached to the unreflected data registered in the unreflected data storage unit 12b to the checkpoint time, and then is registered in the second storage unit 12c, and a signal indicating completion of reflection of the data is transmitted to the first control unit 11a (S5).

Next, when the completion signal is received, the first control unit 11a raises a flag indicating that the data has not been reflected in the second device 12, and clears the flag of the data registered in the first storage unit 11b (S6). ).

According to this embodiment, the transmission time is added to the data that has not yet been reflected in the second device 12, and after the transmission time is converted to the checkpoint time when the data is reflected in the second device 12, the data is stored in the second storage unit 12c. By registering, it is clarified at what point in time the data copied from the first device 11 to the second device 12 is.

[Embodiment 2]
FIG. 4 is a block diagram showing the configuration of an example of the distributed file device of this embodiment. As shown in FIG. 4, in the distributed file device 10 of the present embodiment, the unreflected data storage unit 12b has a data amount threshold, and the preset trigger condition is registered in the unreflected data storage unit 12b. This is the same as the distributed file device 10 of the first embodiment except that the amount of data processed exceeds the threshold.

Next, an example of processing in the distributed file system 10 of this embodiment will be described with reference to the block diagrams of FIGS. 2 to 4 and the flowchart of FIG.

First, as in the first embodiment, processing and registration of unreflected data are performed (S1 and S2).

Next, when a preset trigger condition is satisfied, the reflection execution unit 11c instructs the second control unit 12a to reflect the data that has not been reflected in the second device 12 to the second device 12. The controller 11a is instructed (S3). In this embodiment, when the amount of data with the transmission time registered in the unreflected data storage unit 12b exceeds the threshold value, the reflection executing unit 11c is instructed to the first control unit 11a to instruct the second control unit 12a to reflect to the second device 12.

Next, as in the first embodiment, instructions (S4), reflection (S5), and registration (S6) are performed from the first control unit 11a to the second control unit 12a.

According to the present embodiment, as in the first embodiment, the processing time can be shortened compared to the synchronous copy, and the point in time of the data copied from the first device 11 to the second device 12 can be clarified. .

[Embodiment 3]
FIG. 5 is a block diagram showing the configuration of an example of the distributed file device of this embodiment. As shown in FIG. 5, the distributed file device 10 of the present embodiment is implemented except that the preset trigger condition is that the application 20 issues a checkpoint acquisition request to the first control unit 11a. It is the same as the distributed file device 10 of form 1.

Next, an example of processing in the distributed file system 10 of this embodiment will be described with reference to the block diagrams of FIGS. 2, 3 and 5 and the flowchart of FIG.

First, as in the first embodiment, processing and registration of unreflected data are performed (S1 and S2).

Next, when a preset trigger condition is satisfied, the reflection execution unit 11c instructs the second control unit 12a to reflect the data that has not been reflected in the second device 12 to the second device 12. The controller 11a is instructed (S3). In this embodiment, when the application 20 issues a checkpoint acquisition request to the first control unit 11a, the reflection execution unit 11c causes the second device 12 to reflect the data that has not been reflected in the second device 12 to the second device 12. The first controller 11a is instructed to instruct the controller 12a.

Next, as in the first embodiment, instructions (S4), reflection (S5), and registration (S6) are performed from the first control unit 11a to the second control unit 12a.

According to the present embodiment, as in the first embodiment, the processing time can be shortened compared to the synchronous copy, and the point in time of the data copied from the first device 11 to the second device 12 can be clarified. .

[Embodiment 4]
FIG. 6 is a block diagram showing the configuration of an example of the distributed file device of this embodiment. As shown in FIG. 6, the distributed file device 10 of the present embodiment loses part or all of the unreflected data to which the transmission time is attached, which is registered in the unreflected data storage unit 12b. The control unit 11a calculates the difference between the data registered in the first storage unit 11b with a flag indicating that it has not been reflected in the second device 12 and the data registered in the unreflected data storage unit 12b. It is the same as the distributed file device 10 of the first embodiment except that it is transmitted to the control unit 12a.

Next, an example of processing in the distributed file system 10 of this embodiment will be described with reference to the block diagrams of FIGS. 2, 3 and 6 and the flowchart of FIG.

First, as in the first embodiment, processing and registration of unreflected data are performed (S1 and S2).

Next, when part or all of the unreflected data with the transmission time registered in the unreflected data storage unit 12b is lost, the first control unit 11a detects that the data has not been reflected in the second device 12. is set and the difference between the data registered in the first storage unit 11b and the data registered in the unreflected data storage unit 12b is transmitted to the second control unit 12a (S7). FIG. 11 shows an example in which this difference transmission step (S7) is performed after registration of unreflected data (S2). This may be performed at any timing after the loss of part or all of the unreflected data marked with .

Next, in the same manner as in the first embodiment, an instruction from the reflection execution unit 11c to the first control unit 11a (S3), an instruction from the first control unit 11a to the second control unit 12a (S4), and a reflection (S5) , and registration (S6). In this embodiment, the instruction (S3) from the reflection execution unit 11c to the first control unit 11a may be performed in the same manner as in the second or third embodiment.

According to this embodiment, it is possible to recover the state before the unreflected data is lost from the unreflected data storage unit 12b in a short time.

[Embodiment 5]
FIG. 7 is a block diagram showing the configuration of an example of the distributed file device of this embodiment. As shown in FIG. 7, in the distributed file device 10 of this embodiment, when a failure occurs in the first device 11 and the second device 12 takes over the execution of the application 20, the second control unit 12a performs the following conditions: This is the same as the distributed file device 10 of the first embodiment except that either one of condition 1 and condition 2 below is selected.
(Condition 1)
Only the reflected data with the checkpoint time registered in the second storage unit 12c is used for the execution of the application 20. FIG.
(Condition 2)
The application 20 transfers the reflected data with the checkpoint time registered in the second storage unit 12c and the unreflected data with the transmission time registered in the unreflected data storage unit 12b. used for execution.

Next, an example of processing in the distributed file system 10 of this embodiment will be described with reference to the block diagrams of FIGS. 2, 3 and 7 and the flowchart of FIG.

First, as in the first embodiment, processing and registration of unreflected data (S1 and S2), instructions from the reflection execution unit 11c to the first control unit 11a (S3), 12a (S4), reflected (S5), and registered (S6). In this embodiment, the instruction (S3) from the reflection execution unit 11c to the first control unit 11a may be performed in the same manner as in the second or third embodiment. Further, in this embodiment, when part or all of the unreflected data with the transmission time registered in the reflected data storage unit 12b is lost, at an arbitrary timing after that, in the fourth embodiment, , the difference transmission step (S7) may be carried out.

Next, when a failure occurs in the first device 11 and the second device 12 takes over the execution of the application 20, the second control unit 12a selects either condition 1 or condition 2 (S8). .

In the present embodiment, as exemplified in FIG. 13, prior to the selection, the second control unit 12a controls the time required for the second device 12 to start the application 20 when the condition 1 is selected, The time required for the second device 12 to activate the application 20 when the condition 2 is selected may be predicted (S9). Thus, the selection can be made after considering the processing times for both the condition 1 and the condition 2.

Although the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes can be made to the configuration and details of the present invention within the scope of the present invention that can be understood by those skilled in the art.

According to the present invention, for example, it is possible to provide an asynchronous copy distributed file device in which the point in time of data copied from the active device to the standby device is clarified, and a failover method using the distributed file device.

10 distributed file device 11 first device 11a first control unit 11b first storage unit 11c reflection execution unit 12 second device 12a second control unit 12b unreflected data storage unit 12c second storage unit 20 application 30 communication network 111, 121 CPUs
112, 122 memory 113, 123 bus 114, 124 storage device 115, 125 input device 116, 126 display 117, 127 communication device

Claims (16)

a first device executing an application;
a second device that takes over execution of the application when the first device fails;
including
The first device includes a first control unit, a first storage unit, and a reflection execution unit,
The second device includes a second control unit, an unreflected data storage unit, and a second storage unit,
The first control unit and the second control unit are connectable via a communication network,
The first control unit registers data related to execution of the application in the first storage unit with a flag indicating that the data has not been reflected in the second device, and adds a transmission time to the data. Send to the second control unit,
the second control unit registers the data to which the transmission time is attached in the unreflected data storage unit;
The reflection execution unit, when a preset trigger condition is satisfied, instructs the second control unit to reflect data that has not been reflected in the second device to the second device. to
When receiving the instruction, the first control unit instructs the second control unit to reflect the unreflected data;
Upon receiving the instruction, the second control unit changes the transmission time attached to the unreflected data registered in the unreflected data storage unit to the checkpoint time, and then transfers the data to the registering in the second storage unit, and transmitting a completion signal for reflecting the data to the first control unit;
When the completion signal is received, the first control unit raises a flag indicating that the data has not been reflected in the second device, and clears the flag of the data registered in the first storage unit.
Distributed File Device. wherein the preset trigger condition is elapse of a predetermined time or arrival of a predetermined time;
2. The distributed file system according to claim 1. the unreflected data storage unit has a data amount threshold,
wherein the preset trigger condition is that the amount of data registered in the unreflected data storage unit exceeds the threshold;
2. The distributed file system according to claim 1. The preset trigger condition is that the application requests the first control unit to acquire a checkpoint,
2. The distributed file system according to claim 1. When part or all of the unreflected data with the transmission time registered in the unreflected data storage unit is lost, the first control unit provides a flag indicating that the unreflected data has not been reflected in the second device. and transmitting the difference between the data registered in the first storage unit and the data registered in the unreflected data storage unit to the second control unit;
5. A distributed file device according to any one of claims 1 to 4. When a failure occurs in the first device and the second device takes over execution of the application, the second control unit selects either condition 1 or condition 2 below.
A distributed file device according to any one of claims 1 to 5.
(Condition 1)
Only the reflected data with the checkpoint time registered in the second storage unit is used for execution of the application.
(Condition 2)
The application stores the reflected data with the checkpoint time registered in the second storage unit and the unreflected data with the transmission time registered in the unreflected data storage unit. used for execution. A time required for the second device to activate the application when the second control unit selects the condition 1, and a time required for the second device to activate the application when the condition 2 is selected. predict the time and
7. The distributed file system according to claim 6. a first device executing an application;
a second device that takes over execution of the application when the first device fails;
using a distributed file device containing
The first device includes a first control unit, a first storage unit, and a reflection execution unit,
The second device includes a second control unit, an unreflected data storage unit, and a second storage unit,
The first control unit and the second control unit are connectable via a communication network,
The first control unit registers data related to execution of the application in the first storage unit with a flag indicating that the data has not been reflected in the second device, and adds a transmission time to the data. a non-reflected data processing step for transmitting to the second control unit;
an unreflected data registration step in which the second control unit registers the data to which the transmission time is attached in the unreflected data storage unit;
The first control unit, when the reflection execution unit satisfies a preset trigger condition, instructs the second control unit to reflect data that has not been reflected in the second device to the second device. a first reflection instruction step of instructing to
a second reflection instruction step in which, when receiving the instruction, the first control unit instructs the second control unit to reflect the unreflected data;
When receiving the instruction, the second control unit changes the transmission time attached to the unreflected data registered in the unreflected data storage unit to the checkpoint time, and then transfers the data to the a reflection step of registering the data in a second storage unit and transmitting a reflection completion signal of the data to the first control unit;
a registration step in which, when the completion signal is received, the first control unit raises a flag indicating that the data has not been reflected in the second device and clears the flag of the data registered in the first storage unit;
failover methods, including wherein the preset trigger condition is elapse of a predetermined time or arrival of a predetermined time;
9. The failover method according to claim 8. the unreflected data storage unit has a data amount threshold,
wherein the preset trigger condition is that the amount of data registered in the unreflected data storage unit exceeds the threshold;
9. The failover method according to claim 8. The preset trigger condition is that the application requests the first control unit to acquire a checkpoint,
9. The failover method according to claim 8. When part or all of the unreflected data with the transmission time registered in the unreflected data storage unit is lost, the first control unit provides a flag indicating that the unreflected data has not been reflected in the second device. and transmitting the difference between the data registered in the first storage unit and the data registered in the unreflected data storage unit to the second control unit,
A failover method according to any one of claims 8 to 11. When a failure occurs in the first device and the second device takes over execution of the application, the second control unit selects either condition 1 or condition 2 below.
A failover method according to any one of claims 8 to 12.
(Condition 1)
Only the reflected data with the checkpoint time registered in the second storage unit is used for execution of the application.
(Condition 2)
The application stores the reflected data with the checkpoint time registered in the second storage unit and the unreflected data with the transmission time registered in the unreflected data storage unit. used for execution. A time required for the second device to activate the application when the second control unit selects the condition 1, and a time required for the second device to activate the application when the condition 2 is selected. including a prediction step of predicting the time required for
14. The failover method according to claim 13. A program executable on a computer for the method according to any one of claims 8 to 14. A computer-readable recording medium recording the program according to claim 15.

Images