JP2019168807A - Control device, control method, control program and control system - Google Patents

Abstract

To provide a control device, a control method, a control program and a control system which allow for reducing load on an entire system by preventing unnecessary retries when receiving an access request to a database during the execution of switching processing.SOLUTION: The control device comprises: a monitoring unit which monitors states of a plurality of instances which an in-memory database has; and a control unit which, in response to accepting a request for transaction processing with a specific instance out of the plurality of instances as an object, suspends the accepted request for transaction processing if a monitoring result of the monitoring unit indicates that the specific instance is in the middle of switching processing.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a control device, a control method, a control program, and a control system that control transaction processing.

  In recent systems, etc., “high speed” that processes a large amount of transactions in a short time and “high reliability” that supports 24-hour operation are required. As a technique for realizing the above-mentioned “high speed” and “high reliability”, there is an in-memory database that performs data storage mainly on a main memory.

  Generally, in a system using an in-memory database, a master instance (master) and a slave instance (slave) are configured for the purpose of redundancy. An instance is a state in which a predefined computer program, data structure, and the like are expanded on a main memory and can be processed and executed.

  When the business application (business AP) updates the master data, the update is transferred to the slave as an update log. By reflecting the update log on the slave data synchronously or asynchronously, the slave is made a replica of the master. If a failure occurs on the master side, the master and slave can be switched, and what was operating as a slave can be operated as the master, so that processing can continue even if a failure occurs on the master side. It becomes possible. Hereinafter, the process of switching between the master and the slave is referred to as a switching process.

  Switching processing may also occur during transaction execution. The transaction involved in the switching process ends abnormally or repeats unnecessary retries until the switching process is completed.

  Patent Document 1 is an example of a technique related to switching processing in a cluster configuration database having a plurality of instances. The database access control unit according to Patent Document 1 issues SQL to the database instance based on the database access request. The retry possibility determination unit executes a rollback process when the execution result of the SQL transmitted from the database system ends abnormally. When the rollback process fails, the database access control unit switches the database instance (switching process).

JP 2011-086067 A

  The technique disclosed in Patent Document 1 issues an SQL upon accepting an access request, performs a rollback process when the execution result of the SQL is abnormally terminated, and performs a switching process when the rollback process fails. carry out. However, the technique disclosed in Patent Document 1 does not consider a case where an access request is received while the switching process is being performed. Therefore, the technique disclosed in Patent Document 1 may transmit an access request received while the switching process is being performed to an instance that is a target of the access request. Since the instance during the switching process cannot accept the access request, the access request transmitted to the instance during the switching process fails. Furthermore, when the database access control unit repeatedly retries the access request, the load on the entire system increases.

  The invention of the present application solves the above-described problems, and can prevent unnecessary retries when a request for access to a database is received while switching processing is being performed, and can reduce the load on the entire system and control method A control program and a control system are proposed.

  The control device according to the present invention has a monitoring unit that monitors the states of a plurality of instances included in the in-memory database, and a monitoring unit that receives a transaction processing request for a specific instance among the plurality of instances. A control unit that suspends the received transaction processing request when the monitoring result indicates that the specific instance is undergoing switching processing.

  The control method of the present invention monitors the state of a plurality of instances in the in-memory database, and when a transaction processing request for a specific instance among the plurality of instances is received, the monitoring result is: If switching processing is being performed in the specific instance, the accepted transaction processing request is suspended.

  The computer program of the present invention monitors the state of a plurality of instances in the in-memory database, and when a transaction processing request for a specific instance among the plurality of instances is received, the monitoring result is: When switching processing is being performed in the specific instance, the computer is caused to execute processing for holding the received transaction processing request.

  According to the present invention, it is possible to prevent unnecessary retries when a database access request is received during the switching process, and to reduce the load on the entire system.

2 is a diagram illustrating a configuration of a control system 500. FIG. 5 is a flowchart showing an operation of a data access control unit 330. 5 is a flowchart showing the operation of a monitoring unit 320. 5 is a flowchart showing the operation of a retry control unit 340. It is a figure which shows the specific example of the element mentioned as a confirmation result of a management process. It is a figure which shows the specific example of the check table. It is a figure which shows the specific example of the operating state table. 2 is a diagram illustrating a configuration of a control device 1. FIG. 3 is a flowchart showing the operation of the control device 1. It is a figure which shows the computer which implement | achieves this invention.

<First Embodiment>
A first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating a configuration of a control system 500 according to the first embodiment.

  The control system 500 includes a business application 100 (hereinafter referred to as business AP 100), an in-memory database system 200 (hereinafter referred to as in-memory DB 200), and a control device 300. The business AP 100, the in-memory DB 200, and the control device 300 are connected via a network or the like and can communicate with each other.

  The business AP 100 issues an access request in response to a request from a user or the like. Specifically, the business AP 100 issues an access request including a transaction for updating or referring to data in the in-memory DB 200.

  The in-memory DB 200 includes a processing unit 210, an instance 220, an instance 230, and an instance 240.

  When receiving the access request, the processing unit 210 performs, for the access request, exclusive control processing between records, commit processing, rollback processing, execution processing of the request, and the like. For example, when the received access request is a data update for a specific instance, the processing unit 210 performs an update process on the target instance. In addition, when a failure occurs in the master, the processing unit 210 performs a switching process for switching between the master and the slave.

  The instance 220 includes a user table 221, a management process 222, and a check table 223.

  The user table 221 is a table that is updated or referred to by performing data access. The information shown in the user table 221 may be any information as long as it is updated or referred to by the user or the like.

  The management process 222 manages the state of the instance 220. The monitoring unit 320 to be described later confirms the management process 222, so that the state of the instance 220 managed by the management process 222 can be determined. As the state of the instance 220 that can be determined by the monitoring unit 320 confirming the management process 222, for example, there are three types of states: “active”, “stopped”, and “failed”. Although the monitoring unit 320 has been described as confirming the management process, the present application is not limited to this, and the processing unit 210 or the like may confirm the state of the instance 220 as long as the state can be determined.

  The check table 223 shows a result of periodic access by the monitoring unit 320 described later to confirm the state of the instance 220. Specifically, the check table 223 indicates the date and time when the monitoring unit 320 accessed and the access result in a corresponding log format. For example, there are three types of access results: “normal end”, “abnormal end of timeout”, and “abnormal end other than timeout”. Although the monitoring unit 320 has been described as accessing periodically, the present application is not limited to this, and the processing unit 210 may access the instance 220 as long as the instance 220 can be accessed. Although the instance 220 has been described as having a check table, the present application is not limited to this, and the instance other than the instance 220 may have the check table 223 as long as the monitoring unit can confirm.

  “Normally completed” indicates that the monitoring unit 320 has accessed the instance 220 and the access has been completed normally, for example, when a response is returned within a certain time for the access.

  “Abnormal termination of timeout” indicates that the access is terminated abnormally, for example, when the monitoring unit 320 accesses the instance 220 and no response is returned within a predetermined time.

  “Abnormal termination other than timeout” indicates that the monitoring unit 320 accessed the instance 220 and terminated abnormally due to a reason other than timeout. Examples of causes other than timeout include cases where the monitoring unit 320 cannot access the instance 220, but are not limited thereto.

  Like the instance 220, the instance 230 and the instance 240 also have a user table, a management process, and a check table. The configurations of the user table, the management process, and the check table included in the instance 230 and the instance 240 are the same as the configurations of the user table 221, the management process 222, and the check table 223, and thus description thereof is omitted.

  The control device 300 includes an operating state table 310, a monitoring unit 320, a data access control unit 330, a retry control unit 340, and a registration unit 350.

  The operation state table 310 records the state of each instance. As the state of each instance, for example, there are three types of “in operation”, “stopped”, and “switching process”.

  The monitoring unit 320 confirms the management process of each instance and the access result of the latest date and time in the check table at regular intervals. When the monitoring unit 320 confirms the management process and the check table for all the instances, the monitoring unit 320 determines the state of each instance according to the operation condition, and updates the operation state table 310 using the determined result. The operating conditions will be described later.

  In addition, the monitoring unit 320 uses the operating state table 310 to detect the completion of the switching process. Specifically, when updating the operating state table 310, the monitoring unit 320 confirms whether there is an instance whose state has changed from “switching process” to “running”. When there is an instance whose state has changed from “switching process” to “running”, the monitoring unit 320 recognizes that the switching process has been completed for the instance, and indicates that the switching process has been completed for the instance. A processing completion notification is transmitted to the retry control unit 340 described later.

  The data access control unit 330 receives an access request from the business AP 100 or the retry control unit 340, and performs a confirmation process on the received access request using the operation state table 310. The confirmation process is a process in which the data access control unit 330 confirms the state of the instance that is the target of the access request.

  As a result of the confirmation processing, the data access control unit 330 transmits the access request to the processing unit 210 when the instance that is the target of the access request is “in operation”. The processing unit 210 performs data access on the instance that is the target of the received access request.

  As a result of the confirmation processing, the data access control unit 330 does not transmit the access request to the processing unit 210 when the instance targeted for the access request is “stopped”. An access request that is not transmitted to the processing unit 210 is not implemented in the instance that is the target of the access request.

  As a result of the confirmation processing, when the instance that is the target of the access request is “switching processing”, the data access control unit 330 transmits the access request to the retry control unit 340 without transmitting it to the processing unit 210.

  The retry control unit 340 holds the access request received from the data access control unit 330. Specifically, the retry control unit 340 holds the access request by registering the access request received from the data access control unit 330 in the registration unit 350. The pending access request means that the access request is suppressed by registering the access request in the registration unit 350.

  When the retry control unit 340 receives the switching process completion notification from the monitoring unit 320, the retry control unit 340 acquires an access request for accessing the instance for which the switching process has been completed from the registration unit 350, and transmits the acquired access request to the business AP 100. To do.

  When receiving the access request from the retry control unit 340, the business AP 100 issues the received access request again. The data access control unit 330 may perform confirmation processing on the access request issued again by the business AP 100, or may directly transmit the request to the processing unit 210 without performing confirmation processing. The business AP 100 may issue the received access request again after performing the rollback process on the received access request, or may issue the access request again without performing the rollback process.

  Note that there may be a case where the instance that is the target of the access request is different from the instance that is the target of the previous access request due to the switching process, but the business AP 100 issues the access request to the appropriate target instance again. Is possible.

  The registration unit 350 registers the access request received from the data access control unit 330 by the retry control unit 340. In other words, as a result of the confirmation processing performed by the data access control unit 330, if the instance that is the target of the access request is “switching processing”, the access request is registered in the registration unit 350. The registration unit 350 may be configured in any manner that can register an access request, such as a buffer or a queue.

  In addition, although FIG. 1 demonstrated the case where the control apparatus 300 had the registration part 350, this application is not restricted to this. For example, the registration unit 350 may be included in the retry control unit 340, or may be configured outside the control device 300 as long as it can communicate with the retry control unit 340.

  When the retry control unit 340 receives the switching process completion notification from the monitoring unit 320, the retry control unit 340 may directly transmit the access request registered in the registration unit 350 to the processing unit 210 without sending it to the business AP 100. In other words, when the retry control unit 340 receives the notification of completion of the switching process, the retry control unit 340 may release the suspension of the access request registered in the registration unit 350 and transmit it to the processing unit 210. When the retry control unit 340 transmits an access request to the processing unit 210 instead of the business AP 100, the data access control unit 330 may or may not perform confirmation processing on the access request. good.

  The retry means that when the access request fails, the access request similar to the failed access request is made again. The failure of the access request means that the access request reaches the request destination, for example, when the access request is not transmitted to the processing unit 210 or when the access request is not accepted due to the switching process or the like. This is the case when there was not.

[Operation]
Next, main operations in the first embodiment will be described.

  FIG. 2 is a flowchart showing the operation of the data access control unit 330.

  The data access control unit 330 receives the access request issued by the business AP 100 (S11).

  The data access control unit 330 uses the operation state table 310 to confirm the state of the instance that is the target of the access request (S12). Note that the processing in S12 is referred to as confirmation processing.

  When the state of the access request destination instance is “in operation” (“in operation” in S13), the data access control unit 330 transmits the access request to the processing unit 210 (S14).

  If the state of the access request destination instance is “stopped” (“stopped” in S13), the data access control unit 330 does not transmit the access request to the processing unit 210 (S15).

  If the state of the access request destination instance is “switching process” (“switching process” in S13), the data access control unit 330 transmits the access request to the retry control unit 340 (S16).

  Next, the main operation of the monitoring unit 320 will be described with reference to FIG.

  The monitoring unit 320 confirms a management process for a specific instance among a plurality of instances (S31).

  The monitoring unit 320 confirms the check table for the specific instance (S32).

  If the monitoring unit 320 has not confirmed the management processes and check tables of all instances among the plurality of instances, the monitoring unit 320 repeats the processes of S31 and S32 until the management processes and check tables of all instances are confirmed (S33). No).

  If the monitoring unit 320 confirms the management process and check table of all instances (Yes in S33), the monitoring unit 320 determines the state of each instance according to the operating conditions described later (S34).

  The monitoring unit 320 updates the operating state table 310 using the result determined in S34 (S35).

  When the operating unit table 310 is updated, the monitoring unit 320 checks whether there is an instance whose state has changed from “switching process” to “running”. When there is an instance whose state has changed from “switching process in progress” to “being activated” (Yes in S36), the monitoring unit 320 transmits a switching process completion notification to the retry control unit 340 (S37).

  Next, the main operation of the retry control unit 340 will be described with reference to FIG.

  The retry control unit 340 registers the access request received from the data access control unit 330 in the registration unit 350 (S51).

  When the retry control unit 340 receives the switching process completion notification from the monitoring unit 320 (S52), the retry control unit 340 obtains an access request that targets an instance for which the switching process has been completed, among access requests registered in the registration unit 350. (S53).

  The retry control unit 340 transmits the acquired access request to the business AP 100 (S54).

[Concrete example]
Next, main specific examples of the first embodiment will be described.

  In this specific example, the instance 220 is operated as a master instance, and the instance 230 and the instance 240 are operated as slave instances of the instance 220.

  Next, specific examples of operating conditions are shown. The operating condition is a condition for the monitoring unit 320 to periodically check the management process and check table of each instance and determine the state of each instance. The state of each instance includes, for example, “in operation”, “stopped”, and “switching process”. The method of determining the state of each instance based on the management process confirmation result, the check table confirmation result, and the operating condition is particularly effective in determining “switching process” as an example of the instance state.

  Examples of elements that can be cited as the result of management process confirmation by the monitoring unit 320 include “starting”, “stopping”, and “failing”.

  FIG. 5 shows a specific example 250 of elements cited as a result of management process confirmation by the monitoring unit 320.

  As a result of checking the check table by the monitoring unit 320, for example, there are “normal end”, “abnormal end of timeout”, and “abnormal end other than timeout”.

  FIG. 6 shows a specific example of the check table 223 that the instance 220 has. The date and time 224 indicates the date and time when the monitoring unit 320 accessed the instance 220. The access result 225 indicates an access result obtained when the monitoring unit 320 accesses the instance 220.

  The operating condition for determining “in operation” in the state of each instance is that the confirmation result of the management process is “active” and the confirmation result of the check table is “normal end”.

  The operating condition for determining “stopped” in the state of each instance is that the check result of the management process is “stopped” and the check result of the check table is “abnormal termination other than timeout”. When the management process confirmation result is “stopped”, the target instance may be set to “stopped” regardless of the check table confirmation result.

  The operation condition for determining “switching process” in the state of each instance is that the master instance satisfies the condition A or B, and the specific slave instance satisfies the condition C.

  Condition A is that the management process confirmation result is “failing” and the check table is “abnormal termination other than timeout”.

  Condition B is that the confirmation result of the management process is “Active” and the check table is “Abnormal termination due to timeout”.

  Condition C is that the confirmation result of the management process is “being activated” and the check table is “abnormal termination due to timeout”.

  FIG. 7 shows a specific example of the operating state table 310. The operating state table 310 has an instance name 311 and a state 312.

  The column of the instance name 311 shows the name of the instance to be monitored by the monitoring unit 320.

  The state 312 column shows the state of the target instance.

  In the specific example illustrated in FIG. 7, the state of the instance 220 and the instance 230 indicates that switching processing is being performed, and the state of the instance 240 indicates that it is operating.

  Next, as a specific example in the second embodiment, a process when the user makes an access request for updating the data of the instance 220 while the switching process is executed in the instance 220 and the instance 230. explain. As a specific example of the instance state, an operation state table 310 shown in FIG. 7 is used.

  The user transmits an access request to the data access control unit 330 using the business AP 100.

  The data access control unit 330 receives the access request transmitted by the business AP 100.

  The data access control unit 330 checks the state of the instance 220 using the operation state table 310 shown in FIG. Upon confirming the operation state table 310 shown in FIG. 7, the data access control unit 330 transmits the access request to the retry control unit 340 because the instance 220 is “switching process”.

  When the retry control unit 340 receives a data access request from the data access control unit 330, the retry control unit 340 registers the access request in the registration unit 350.

  The monitoring unit 320 confirms the management process of each instance and the access result of the latest date and time in the check table at regular intervals. Specifically, when the monitoring unit 320 confirms the instance 220, the instance 230, and the instance 240, the monitoring unit 320 determines the state of each instance according to the operation condition, and updates the operation state table 310 using the determined result.

  In this specific example, it is assumed that the monitoring unit 320 updates the state 312 of the instance 220 and the instance 230 in the operation state table 310 from “switching process” to “active”. The monitoring unit 320 recognizes that the switching processing of the instance 220 and the instance 230 has been completed as a result of updating the operating state table 310 as described above, and transmits a switching processing completion notification to the retry control unit 340.

  When the retry control unit 340 receives the switching process completion notification from the monitoring unit 320, the retry control unit 340 acquires an access request for the instance 220 from the registration unit 350 and transmits the access request to the business AP 100.

  When receiving the access request from the retry control unit 340, the business AP 100 issues the received access request again.

  Thus, the present invention can prevent unnecessary retries when a request for access to the database is received during the switching process, and can reduce the load on the entire system.

  In addition, when an access request is received during the switching process, the present invention can immediately recognize when the instance is inaccessible by performing a confirmation process before transmitting directly to the target instance of the access request. is there.

  Furthermore, by introducing the present invention, it becomes possible to control transactions flexibly according to the operating state of the database.

  If the processing unit 210 performs access to the target instance and the access fails, the processing unit 210 sends the access request again to the processing unit 210 in S14 illustrated in FIG. You may transmit to the data access control part 330. FIG. Further, the data access control unit 330 may perform confirmation processing again on the access request received from the processing unit 210. As a result of the confirmation processing, when the access target instance is being switched, the data access control unit 330 transmits the access to the retry control unit 340.

  As a result, even when the switching process occurs in the implementation target instance immediately before the processing unit 210 performs the access and the execution of the access fails, the registration unit does not immediately retry the access request. 350 can be registered. Furthermore, since the access request is registered in the registration unit 350 until the switching process is completed, unnecessary retry can be prevented and transaction control can be performed while reducing the load on the entire system.

  In the present application, the data access control unit 330 and the retry control unit 340 have been described separately, but the present application is not limited to this. In the present application, for example, the data access control unit 330 and the retry control unit 340 can be combined into one and configured as a control unit. In this case, the control unit can perform operations of the data access control unit 330 and the retry control unit 340.

  When a plurality of access requests are registered in the registration unit 350, the retry control unit 340 may transmit the plurality of access requests to the business AP 100 one by one when receiving the switching completion notification. These access requests may be collectively sent to the business AP 100.

  In the present application, the number of instances is three, but the number of instances may be any number. Further, although the present application has been described with the configuration of one master and two slaves, the present application is not limited to this, and the number of masters and slaves can be arbitrarily changed.

  Note that the user table and the operating state check table may be newly generated when the in-memory DB 200 is started, or each instance may have before starting.

<Second Embodiment>
A second embodiment of the present invention will be described with reference to the drawings. The configuration of the control device 1 in the second embodiment is shown in FIG.

[Constitution]
First, the configuration of the control device 1 will be described. As illustrated in FIG. 8, the control device 1 according to the present embodiment includes a monitoring unit 2 and a control unit 3.

  The monitoring unit 2 monitors the states of a plurality of instances included in the in-memory database.

  When the control unit 3 receives a transaction processing request for a specific instance among a plurality of instances, and the monitoring result of the monitoring unit 2 is being switched at the specific instance, Hold the accepted transaction processing request.

  The monitoring unit 2 in the second embodiment corresponds to the monitoring unit 320 in the first embodiment, and the control unit 3 in the second embodiment includes the data access control unit 330 and the retry in the first embodiment. This corresponds to the control unit 340. That is, the control unit 3 in the second embodiment can realize the configurations and operations of the data access control unit 330 and the retry control unit 340 in the first embodiment within a consistent range.

[Operation]
Next, the operation of the second embodiment will be described with reference to FIG.

  FIG. 9 is a flowchart showing the operation of the control device 1 in the second embodiment.

  The monitoring unit 2 monitors the states of a plurality of instances included in the in-memory database (S1).

  When the control unit 3 receives a transaction processing request for a specific instance among a plurality of instances, and the monitoring result of the monitoring unit 2 is being switched at the specific instance, The accepted transaction processing request is suspended (S2).

  As a result, the present invention according to the second embodiment can prevent unnecessary retries when the access request to the database is received during the switching process, and can reduce the load on the entire system. .

[program]
The program in the present invention is a program for causing a computer to execute the processing of the present invention. By installing and executing this program in a computer, the control device, control system, and control method of the present invention can be realized.

  Here, the computer which implement | achieves the control apparatus 1, the control system 500, and the control method by running the program in this invention is demonstrated using FIG.

  FIG. 10 is a block diagram illustrating an example of a computer that implements the control device 1, the control system 500, and the control method according to the present invention.

  As shown in FIG. 10, the computer 110 includes a CPU 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. These units are connected to each other via a bus 121 so that data communication is possible.

  The CPU 111 performs various operations by expanding the program (code) of the present invention stored in the storage device 113 in the main memory 112 and executing them in a predetermined order. The main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory). The program according to the present invention is provided in a state of being stored in a computer-readable recording medium 120. The program in the present invention may be distributed on the Internet connected via the communication interface 117.

  Examples of the storage device 113 include a hard disk drive and a semiconductor storage device such as a flash memory. The input interface 114 mediates data transmission between the CPU 111 and the input device 118. Examples of the input device 118 include a keyboard, a mouse, and a touch panel. The display controller 115 is connected to the display device 119 and controls display on the display device 119.

  The data reader / writer 116 mediates data transmission between the CPU 111 and the recording medium 120, and reads a program from the recording medium 120 and writes a processing result in the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and another computer.

  Specific examples of the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), or magnetic storage media such as a flexible disk (Flexible Disk). . Furthermore, a specific example of the recording medium 120 includes an optical storage medium such as a CD-ROM (Compact Disk Read Only Memory).

  So far, the embodiment for carrying out the present invention has been described, but the present invention is not limited to the above-described embodiment. For example, the order of processing in the flowcharts described in this specification is merely an example, and can be changed within a consistent range when carrying out the present invention. In addition, the configuration described in this specification is merely an example, and the configuration can be changed within a consistent range when the present invention is implemented. That is, the present invention can apply various modes that can be understood by those skilled in the art within the scope of the present invention when carrying out the present invention. Modifications of the above embodiments, combinations thereof, and the like are also included in the present invention. Is included in the technical scope.

DESCRIPTION OF SYMBOLS 1 Control apparatus 2 Monitoring part 3 Control part 100 Business application 111 CPU
112 Main memory 113 Storage device 114 Input interface 115 Display controller 116 Data reader / writer 117 Communication interface 118 Input device 119 Display device 120 Recording medium 200 In-memory database system 210 Processing unit 220 Instance 221 User table 222 Management process 223 Check table 230 Instance 231 User table 232 Management process 233 Check table 240 Instance 241 User table 242 Management process 243 Check table 250 Specific examples of elements listed as management process confirmation results 300 Controller 310 Operating state table 320 Monitoring unit 330 Data access control unit 340 Retry Control unit 350 Registration unit 500 Control system

Claims (10)

A monitoring unit that monitors the status of multiple instances of the in-memory database;
When a request for transaction processing for a specific instance among the plurality of instances is received and the monitoring result of the monitoring unit is being switched at the specific instance, A control unit that holds the request;
A control device comprising: The monitoring unit detects completion of the switching process,
The control unit transmits the suspended transaction process to the application that issued the transaction process when the monitoring unit detects the completion of the switching process.
The control device according to claim 1. A registration unit for registering the transaction processing request;
When the control unit receives the transaction processing request and the monitoring result of the monitoring unit is being switched in the specific instance, the control unit registers the received transaction processing request in the registration unit. When the monitoring unit detects the completion of the switching process, the registered transaction processing request is acquired from the registration unit.
The control device according to claim 1 or 2. The state of the plurality of instances includes running, stopping, or switching processing,
Each of the plurality of instances includes a confirmation table indicating a result of access by the monitoring unit, and a management process indicating that each instance itself is being started, stopped, or failed.
The monitoring unit determines a state during the switching process using the confirmation table and the management process.
The control device according to any one of claims 1 to 3. An operating state table showing the states of the plurality of instances;
The monitoring unit determines the state of the plurality of instances using the confirmation table and the management process, and updates the operation state table using the determined result.
The control device according to claim 4. The monitoring unit detects completion of the switching process based on the operating state table;
The control device according to claim 5. The monitoring unit detects completion of the switching process,
The control unit transmits the suspended transaction process to the specific instance when the monitoring unit detects the completion of the switching process.
The control device according to any one of claims 1 to 6. An in-memory database with multiple instances;
A control device according to any one of claims 1 to 7,
An application for issuing a transaction processing request for a specific instance of the plurality of instances and transmitting the request to the control device;
A control system comprising: Monitor the status of multiple instances of the in-memory database,
When a request for transaction processing targeting a specific instance among the plurality of instances is received, if the result of the monitoring is switching processing at the specific instance, the received request for transaction processing is Hold,
Control method. Monitor the status of multiple instances of the in-memory database,
When a request for transaction processing targeting a specific instance among the plurality of instances is received, if the result of the monitoring is switching processing at the specific instance, the received request for transaction processing is Hold,
A computer program that causes a computer to execute processing.

Images