JP2014052687A - Virtual machine management system, management server, and virtual machine management method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically and safely stop a virtual machine even when there is a need to immediately stop the virtual machine.SOLUTION: A virtual machine management system 1 includes: a plurality of uninterruptible power supplies 30 for supplying electric power to connected devices when a power failure occurs; a plurality of virtual servers 20 connected to each of the uninterruptible power supplies 30 and having one or a plurality of virtual machines 21; and a management server 10 for managing the order of stopping the virtual servers 20 and the virtual machines 21. When a power failure occurs, the management server 10 executes live migration which migrates a virtual machine 21 operating on a predetermined virtual server 20 to the other virtual server 20 on the basis of a preset stop order of the virtual machines 21 and a stop order of the uninterruptible power supplies 30, to sequentially stop the virtual machines 21 from a machine operating on the virtual server 20 connected to an uninterruptible power supply 30 which is high in the stop order of the uninterruptible power supplies 30.

Description

  The present invention relates to a virtual machine management system, a management server, and a virtual machine management method, and in particular, management capable of automatically controlling the stop order of virtual machines operating on a virtual server when a power failure occurs. The present invention relates to a system, a management server, and a management method.

  Conventionally, in a virtual machine computer system that simultaneously operates a plurality of virtual machines on a virtual server, various methods can be used to continue the operation of the virtual machine even when power supply to the virtual server is interrupted due to a power failure or the like. Is used.

  For example, in Patent Document 1, in a virtual machine computer system, when a power failure occurs due to a power failure or the like for a plurality of virtual machines activated on a virtual server, the virtual machine computer system is activated on a predetermined virtual server. A technique is described in which execution of a virtual machine is continued by failing over the virtual machine to another virtual server.

  In this virtual machine computer system, when a fail-safe device detects a virtual server in which a power failure has occurred, the virtual machine on the virtual server is failed over to another virtual server, and after the movement of the virtual machine is completed, Since the virtual server in which the failure has occurred is shut down, the virtual server in which the power failure has occurred can be safely shut down.

  Patent Document 2 describes a technique for indicating a stop order when an administrator manually stops a plurality of virtual machines in a virtual machine computer system.

JP 2009-282714 A JP 2009-245316 A

  By the way, when stopping a running virtual machine, if there are linkages or dependencies between multiple virtual machines running on the virtual server or between programs running on each virtual machine, Considering this relationship, stop control for virtual machines must be performed in order.

  If the virtual machine continues to run due to failover, the virtual machine moves on the virtual server, so stop control for the virtual machine is performed in consideration of which virtual server the virtual machine is running on. Must be done in order.

  Such ordered virtual machine stop control is not limited to when a virtual machine is systematically stopped, for example, in the event of a sudden power failure. Even in a situation where it is not possible to perform the process, it is necessary to automatically control the stop order of virtual machines in consideration of the dependency between virtual machines and programs.

  However, in the system described in Patent Document 1, when a power failure occurs, a virtual machine running on a predetermined virtual server is failed over to another virtual server, thereby enabling continuous operation by the virtual machine. However, when the power failure state continues and the virtual machine needs to be stopped, the virtual machine stop process is performed without considering the priority order. For this reason, there is a problem in that it is impossible to control the stop order of virtual machines in consideration of dependency between virtual machines or programs.

  On the other hand, when the virtual machines are manually ordered and stopped using the technique described in Patent Document 2, if a planned power failure occurs, the virtual machine can be stopped while confirming the stop order. However, when a power outage occurs, it is almost impossible for the administrator to always wait and monitor the management server. Therefore, when the administrator is absent, stop processing that ignores the stop order is automatic. Done. Therefore, in the event of a sudden power failure, it is not possible to control the stop order of virtual machines taking into account the dependency between virtual machines or programs.

  Therefore, the present invention has been made in view of the problems in the above-described conventional technology, and even when the virtual machine needs to be stopped suddenly and stopped, the virtual machine can be automatically and safely stopped. An object is to provide a virtual machine management system, a management server, and a virtual machine management method.

  In order to achieve the above object, the present invention provides a virtual machine management system comprising: a plurality of uninterruptible power supply devices that supply power to a connected device when a power failure occurs; A plurality of virtual servers that are connected to each of the power failure power supply devices and on which one or a plurality of virtual machines operate, and a management server that manages the stop order of the virtual servers and the virtual machines, When a power failure occurs, another virtual server for a virtual machine that operates on a predetermined virtual server based on the preset stop order for each virtual machine and the stop order for each uninterruptible power supply The virtual migration is executed on the virtual server connected to the upper uninterruptible power supply based on the stop order of the uninterruptible power supply. Sequentially from the thin, and executes the stopping process.

  According to the present invention, the live migration processing is executed based on the stop order for each virtual machine and the stop order of each uninterruptible power supply, and the stop order is higher based on the stop order of the uninterruptible power supply. In order to execute sequential stop processing from virtual machines running on virtual servers connected to the uninterruptible power supply, even if it is necessary to stop and stop the virtual machines immediately, stop them from the virtual machine with the highest stop order. It is possible to stop the virtual machine automatically and safely.

  In the virtual machine management system, the management server includes a stop order registration management table in which stop orders of virtual machines operating on the plurality of virtual servers, which are preset by a user, and the plurality of uninterruptible power supplies are registered. Usable time management table in which device usable times are registered, the plurality of uninterruptible power supply units, the plurality of virtual servers connected to each of the plurality of uninterruptible power supply units, and the plurality of virtual servers And the connection management table registered in association with each other. Thereby, the management server can manage the stop order of the virtual server and the virtual machine according to the stop order based on the usable time of the uninterruptible power supply.

  In the virtual machine management system, the management server includes an uninterruptible power supply connection management unit that manages a connection state of the uninterruptible power supply and the virtual server based on the connection management table, and a stop order registration management table. A virtual machine management unit that manages the stop order of the virtual machines, a power state management unit that monitors a power state of the plurality of uninterruptible power supply units and determines whether or not the power failure has occurred, and An operable time measuring unit that calculates usable times of a plurality of uninterruptible power supply devices, a live migration processing unit that executes the live migration process on a virtual machine that operates on the virtual server, the virtual server, and the virtual server A stop control unit that executes the stop process for the virtual machine, and the result of the monitoring by the power state management unit When it is determined that the power failure has occurred, the usable time of the plurality of uninterruptible power supply devices is calculated, and based on the calculated usable time of the plurality of uninterruptible power supply devices, the usable Updating the information registered in the time management table, determining the stop order of the plurality of uninterruptible power supply units based on the information registered in the available time management table, and determining the plurality of uninterruptible power supplies The information registered in the connection management table is rearranged in order from the top of the stop order according to the stop order of the power supply device, the live migration process is executed based on the rearranged connection management table, and the live migration process On the basis of the execution result of the virtual machine, the virtual machine information operating on each virtual server in the connection management table is updated, and the connection is updated. Based on the management table, according stop order of the plurality of uninterruptible power supply, the stop process with respect to the virtual machines running on the plurality of virtual servers and plurality of virtual servers can be performed sequentially.

  In the virtual machine management system, the live migration processing unit compares the stop order of each virtual machine with the stop order of each uninterruptible power supply based on the rearranged connection management table, and the stop order is The upper virtual machine is moved onto the virtual server connected to the uninterruptible power supply having the higher stop order, and the lower-order virtual machine is moved to the uninterruptible power supply having the lower stop order. It can be moved onto the connected virtual server. Accordingly, the virtual machine can be appropriately moved to the virtual server according to the stop order set for the virtual machine.

  In the virtual machine management system, the stop control unit is an uninterruptible power supply having a higher stop order among the one or more virtual machines operating on the plurality of virtual servers based on the connection management table. The stop process is executed by sequentially stopping virtual machines operating on virtual servers connected to the server, and sequentially stopping all virtual machines operating among the virtual servers that have stopped operating from the virtual servers that have stopped. be able to. Thereby, according to the stop order of a virtual machine and a virtual server, a virtual machine and a virtual server can be stopped appropriately.

  In the virtual machine management system, the operable time measuring unit can calculate the usable time based on power consumption amounts of the plurality of virtual servers connected to the plurality of uninterruptible power supply apparatuses.

  Further, the present invention provides a plurality of virtual servers connected to each of a plurality of uninterruptible power supply devices that supply power to a connected device when a power failure occurs, and each of the plurality of virtual servers A management server that manages the stop order of one or a plurality of virtual machines operating on the server, and when the power failure occurs, the stop order for each virtual machine set in advance and each uninterruptible power Based on the stop order of the power supply unit, a live migration process for moving a virtual machine operating on a predetermined virtual server to another virtual server is executed, and based on the stop order of the uninterruptible power supply unit, the stop order Is executed sequentially from the virtual machine operating on the virtual server connected to the host uninterruptible power supply. According to the present invention, as in the case of the above-described invention, even when the virtual machine needs to be stopped suddenly and stopped, it is possible to stop the virtual machine from a higher rank virtual machine and to stop the virtual machine automatically and safely. Is possible.

  Further, the present invention provides a plurality of uninterruptible power supply devices that supply power to a connected device when a power failure occurs, and is connected to each of the plurality of uninterruptible power supply devices. A virtual machine management method in a virtual machine management system comprising a plurality of virtual servers on which the virtual machine operates, and a management server that manages the virtual server and the stop order of the virtual machines, when the power failure occurs In addition, live migration processing for moving a virtual machine operating on a predetermined virtual server to another virtual server based on a preset stop order for each virtual machine and stop order of each uninterruptible power supply And based on the stop order of the uninterruptible power supply, the stop order is the order from the virtual machine operating on the virtual server connected to the upper uninterruptible power supply. , And executes the stopping process. According to the present invention, as in the case of the above-described invention, even when the virtual machine needs to be stopped suddenly and stopped, it is possible to stop the virtual machine from a higher rank virtual machine and to stop the virtual machine automatically and safely. Is possible.

  As described above, according to the present invention, it is possible to automatically and safely stop a virtual machine even when it is necessary to urgently and stop the virtual machine.

It is a block diagram which shows one Embodiment of the virtual machine management system which concerns on this invention. It is a block diagram which shows an example of a structure of a management server. It is a basic diagram for demonstrating a connection management table. It is a basic diagram for demonstrating a stop rank registration management table. It is an approximate line figure for explaining a battery usable time management table. It is a flowchart for demonstrating the flow of an environment setting process. It is a flowchart for demonstrating the flow of a power failure monitoring process. It is a flowchart for demonstrating the flow of a live migration process. It is a flowchart for demonstrating the flow of a stop process. It is an approximate line figure for explaining a connection management table at the time of performing live migration processing.

  Next, an embodiment for carrying out the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows an embodiment of a virtual machine management system according to the present invention. This virtual machine management system 1 is roughly divided into a management server 10, a plurality of virtual servers 20, 20,. , And an uninterruptible power supply 40. In the example illustrated in FIG. 1, a case where three virtual servers 20A, 20B, and 20C and three uninterruptible power supply devices 30A, 30B, and 30C are used is illustrated.

  In the following description, when there is no need to particularly distinguish each of the virtual servers 20A, 20B, and 20C, it is simply referred to as “virtual server 20”, and each of the uninterruptible power supply devices 30A, 30B, and 30C is particularly designated. When it is not necessary to distinguish, it is simply referred to as “uninterruptible power supply 30”.

  When the virtual servers 20A, 20B, and 20C are connected to the uninterruptible power supply devices 30A, 30B, and 30C, and a power failure due to a power failure or the like occurs in a power source to which power is normally supplied, the power supply is cut off In addition, power is supplied from the uninterruptible power supply devices 30A, 30B, and 30C.

  On the virtual server 20A, one or a plurality of virtual machines 21A, 21A,. Similarly, one or more virtual machines 21B, 21B,... Operate on the virtual server 20B, and one or more virtual machines 21C, 21C,. Works. In this example, three virtual machines 21A1, 21A2, and 21A3 operate on the virtual server 20A, three virtual machines 21B1, 21B2, and 21B3 operate on the virtual server 20B, and three on the virtual server 20C. A case where the virtual machines 21C1, 21C2, and 21C3 operate is shown.

  In the following description, the virtual machines 21A1, 21A2, and 21A3, the virtual machines 21B1, 21B2, and 21B3, and the virtual machines 21C1, 21C2, and 21C3 are simply referred to as “virtual machines 21”.

  As shown in FIG. 2, the management server 10 includes an uninterruptible power supply connection management unit 11, a virtual machine management unit 12, a power state management unit 13, a battery operable time measurement unit 14, a live migration processing unit 15, and a stop control unit. 16 and manages and controls the virtual server 20 described above, the virtual machine 21 operating on each virtual server 20, and the uninterruptible power supply 30.

  The uninterruptible power supply connection management unit 11 manages the connection state between each virtual server 20 and the uninterruptible power supply 30 based on a connection management table (see FIG. 3) to be described later, and on each virtual server 20. The virtual machine 21 that operates is managed. The uninterruptible power supply connection management unit 11 periodically accesses each virtual server 20 at a predetermined timing, and acquires information on the virtual machines 21 operating on each virtual server 20.

  The virtual machine management unit 12 manages the virtual machines 21 that are running on each virtual server 20 and the stop order of each virtual machine 21 based on a stop order registration management table (see FIG. 4) described later.

  The power supply state management unit 13 periodically accesses the uninterruptible power supply 30 connected to each virtual server 20 at a predetermined timing, monitors the power supply state of each uninterruptible power supply 30, and supplies power such as a power failure Determine whether a failure has occurred.

  The battery operable time measuring unit 14 measures the battery usable time of the uninterruptible power supply 30 connected to each virtual server 20, and uses the battery usable time management table (to be described later) for the battery usable time obtained by the measurement. (See FIG. 5). The battery operable time measuring unit 14, based on the power consumption of the virtual server 20 connected to each uninterruptible power supply 30 when the power supply state management unit 13 determines that a power outage is occurring, The battery usable time of 30 is calculated.

  The live migration processing unit 15 determines a predetermined virtual server 20 based on information on the virtual machine 21 running on each virtual server 20 and the battery usable time of the uninterruptible power supply 30 connected to each virtual server 20. A live migration process is performed in which the virtual machine 21 activated above is moved to another virtual server 20 without stopping the operation.

  The stop control unit 16 performs a stop process on the virtual server 20 and the virtual machine 21 running on the virtual server 20 based on the stop order.

  Returning to FIG. 1, when the management server 10 is connected to the uninterruptible power supply 40 and a power failure occurs due to a power failure or the like, and the normal power supply is cut off, Supplied.

  Next, various tables used in the management server 10 will be described.

  The connection management table is a table for managing the connection state of each uninterruptible power supply 30 and the virtual server 20, and the virtual machine 21 operating on each virtual server 20, and as shown in FIG. Information relating to the power supply device 30, the virtual server 20, and the virtual machine 21 operating on the virtual server 20 is stored in association with each other. Information associated with each other is arranged from the top according to the stop order of the uninterruptible power supply 30 based on the battery usable time of each uninterruptible power supply 30.

  In the example of FIG. 3, the uninterruptible power supply 30A, the virtual server 20A, and the virtual machines 21A1, 21A2, and 21A3 that operate on the virtual server 20A are associated with each other, and the uninterruptible power supply 30B, the virtual server 20B, and the virtual server Virtual machines 21B1, 21B2, and 21B3 that operate on 20B are associated with each other, and virtual machines 21C1, 21C2, and 21C3 that operate on uninterruptible power supply 30C, virtual server 20C, and virtual server 20C are associated with each other. In addition, information associated with the virtual server 20A to be stopped first is arranged at the top, and information associated with the virtual servers 20B and 20C is sequentially arranged at the lower order according to the stop order.

  The information stored in the connection management table is obtained by the battery operable time measuring unit 14 of the management server 10 accessing the virtual servers 20 periodically, and the virtual machines 21 operating on the virtual servers 20. It is updated based on the information.

  The stop order registration management table is a table for managing the stop order of the virtual machines 21, and stores the stop order and each virtual machine 21 corresponding to the stop order in association with each other as shown in FIG. . In the example of FIG. 4, virtual machines 21A1, 21B1, and 21C1 are associated with “stop order 1”, and virtual machines 21A2, 21B2, and 21C2 are associated with “stop order 2”, and “stop order 3”. Are associated with virtual machines 21A3, 21B3 and 21C3.

  The stop order of each virtual machine 21 is set by the user when each virtual machine 21 is started, and is registered in the stop order registration management table by the virtual machine management unit 12 of the management server 10 based on this setting information. The stop order of each virtual machine 21 may be set for each virtual machine 21 to be started, for example, or a plurality of virtual machines 21 may be set for the same stop order as shown in FIG. Also good. When a plurality of virtual machines 21 are set for the same stop order, the plurality of virtual machines 21 set to the same stop order are stopped simultaneously.

  The battery usable time management table is a table for managing the battery usable time of each uninterruptible power supply 30, and stores the battery usable time of each uninterruptible power supply 30, as shown in FIG. . FIG. 5 shows a case where the battery usable time of the uninterruptible power supply 30A is T1, the battery usable time of the uninterruptible power supply 30B is T2, and the battery usable time of the uninterruptible power supply 30C is T3. An example is shown.

  The battery usable time of each uninterruptible power supply 30 stored in the battery usable time management table is updated at the timing when the battery usable time is calculated by the battery operable time measuring unit 14 of the management server 10.

  Next, the operation of the virtual machine management system 1 having the above configuration will be described with reference to the flowcharts shown in FIGS. In the virtual machine management system 1 according to the present embodiment, the environment setting process for registering the relationship between the uninterruptible power supply 30, the virtual server 20, and the virtual machine 21 is performed, and then based on the state of each uninterruptible power supply 30. Perform power outage monitoring processing. When a power failure such as a power failure occurs, a live migration process is performed to move the activated virtual machines 21 on each virtual server 20, and the activated virtual machines 21 are sequentially stopped. Perform stop processing.

  First, the flow of the environment setting process will be described with reference to FIG. In step S1, the uninterruptible power supply connection management unit 11 of the management server 10 displays connection information between each virtual server 20 preset by the user and the uninterruptible power supply 30 connected to the virtual server 20 in the connection management table. Register with.

  Next, in step S2, when the virtual machine 21 is started on each virtual server 20, the stop order is set by the user for each started virtual machine 21, and the virtual machine management unit 12 sets the set stop Based on the information indicating the rank, the stop rank of each virtual machine 21 is registered in the stop rank registration management table.

  Next, in step S <b> 3, the uninterruptible power supply connection management unit 11 accesses each virtual server 20 and acquires information on the virtual machine 21 that is running on each virtual server 20. Then, the uninterruptible power supply connection management unit 11 registers the acquired information of the virtual machine 21 in the connection management table (step S4). Through the processing in steps S1 and S4, each uninterruptible power supply 30, the virtual server 20 connected to the uninterruptible power supply 30, and the virtual machine 21 running on the virtual server 20 are associated with each other.

  Thereafter, the processes shown in steps S3 and S4 are periodically executed at a predetermined timing. Thereby, the contents of the connection management table are updated when the virtual machine 21 moves to a different virtual server 20 or when the virtual machine 21 is newly activated.

  Next, the flow of the power failure monitoring process will be described with reference to FIG. First, the power supply state management unit 13 of the management server 10 periodically accesses each uninterruptible power supply 30 at a predetermined timing, and determines whether or not a power failure has occurred based on the power supply state of the uninterruptible power supply 30. Judgment is made (step S11). When the power supply state management unit 13 determines that no power failure has occurred (step S11; No), the process returns to step S11.

  On the other hand, when the power state management unit 13 confirms the uninterruptible power supply 30 in the power failure state and determines that a power failure has occurred (step S11; Yes), the battery operable time measuring unit 14 determines that the uninterruptible power supply 30 The battery usable time is calculated based on the power consumption of the virtual server 20 connected to, and the battery usable time registered in the battery usable time management table is updated (step S12).

  Next, in step S13, the uninterruptible power supply connection management unit 11 displays the information registered in the connection management table based on the battery usable time of each uninterruptible power supply 30 registered in the battery usable time management table. Rearranges.

  Here, the stop of each uninterruptible power supply 30 is performed in the order of the short battery usable time. Therefore, the uninterruptible power supply connection management unit 11 rearranges the information in the connection management table from the top in the order of short battery usable time based on the battery usable time updated in step S12.

  Next, the live migration processing unit 15 executes live migration processing for the active virtual machine 21 based on the rearranged connection management table (step S14). The details of the live migration process will be described later.

  After the live migration process is executed, the management server 10 checks the usable time of each uninterruptible power supply 30 based on the battery usable time management table and determines whether or not the usable time has passed ( Step S15).

  When it is determined that the usable time has elapsed (step S15; Yes), the stop control unit 16 stops the virtual server 20 connected to the uninterruptible power supply 30 and the virtual machine 21 running on the virtual server 20. Processing is performed (step S16), and the processing returns to step S11. Details of the stop process will be described later.

  On the other hand, if it is determined that the usable time has not elapsed (step S15; No), the process returns to step S11.

  Next, the flow of live migration processing will be described with reference to FIG. First, the live migration processing unit 15 of the management server 10 sets each virtual server 20 based on the connection management table rearranged based on the battery usable time of each uninterruptible power supply 30 registered in the battery usable time management table. The virtual machine 21 being started up and the stop order of the virtual machines 21 registered in the stop order registration management table are confirmed (steps S21 and S22).

  Next, the live migration processing unit 15 compares the stop order set for the virtual machine 21 with the stop order of the uninterruptible power supply 30 connected to the virtual server 20 on which the virtual machine 21 operates. Live migration for moving the machine 21 to another virtual server 20 is executed (step S23). Specifically, for example, the live migration processing unit 15 moves the virtual machine 21 with the earlier stop order to the virtual server 20 connected to the uninterruptible power supply 30 with a short battery usable time, The virtual machine 21 with the slow stop order is moved to the virtual server 20 connected to the uninterruptible power supply 30 with a long usable time.

  Then, after the live migration is executed, the uninterruptible power supply connection management unit 11 updates the connection management table based on the information of the live migration that has been executed (step S24).

  Next, the flow of the stop process will be described with reference to FIG. First, the stop control unit 16 of the management server 10 is based on the connection management table, and information on the virtual server 20 connected to the uninterruptible power supply 30 with an earlier stop order and the virtual machine 21 running on the virtual server 20 is displayed. Is acquired (step S31).

  Next, the stop control unit 16 performs a stop process on the virtual machine 21 based on the acquired information (step S32). Then, the stop control unit 16 determines whether there is another virtual machine 21 running on the virtual server 20 connected to the same uninterruptible power supply 30 (step S33), and the other virtual machine If it is determined that 21 exists (step S33; Yes), the process returns to step S32, and a stop process is executed for the virtual machine 21.

  On the other hand, when it is determined that there is no other virtual machine 21 on the virtual server 20 connected to the same uninterruptible power supply 30 and the stop processing for all the virtual machines 21 is completed (step S33; No). The process proceeds to step S <b> 34, and the stop control unit 16 executes a stop process for the virtual server 20.

  Hereinafter, according to the stop order of the uninterruptible power supply 30, the stop control unit 16 sequentially performs the virtual server 20 connected to the uninterruptible power supply 30 in the power failure state and the virtual machine 21 running on the virtual server 20, The stop process shown in FIG. 9 is executed.

  Here, the operation of the virtual machine management system 1 will be described more specifically. Here, a case where a power failure occurs in the virtual machine management system 1 having the configuration illustrated in FIG. 1 and the uninterruptible power supply devices 30A, 30B, and 30C are in a power failure state will be described as an example. In this example, it is assumed that the connection management table is registered in advance as shown in FIG. 10A, and the stop order registration management table is registered in advance as shown in FIG.

  In this case, when the power supply state management unit 13 detects the power failure state of the uninterruptible power supply devices 30A, 30B, and 30C, the management server 10 uses the battery operable time measuring unit 14 based on the power consumption of each uninterruptible power supply device 30. The battery usable time is calculated, and the information in the battery usable time management table is updated for each uninterruptible power supply 30 based on the calculation result.

  Here, when it is assumed that the battery usable time of each uninterruptible power supply 30 is “T1”, “T2”, and “T3” in ascending order, the management server 10 is connected by the uninterruptible power supply connection management unit 11. Rearrange the information registered in the management table.

  Next, the management server 10 determines the uninterruptible power supply unit based on the connection management table and the stop order registration management table that are rearranged by the live migration processing unit 15 based on the battery usable time of each uninterruptible power supply unit 30. The stop order of 30 and the stop order of the virtual machines 21 running on each virtual server 20 are compared.

  At this time, of the virtual machines 21A1 to 21A3 that are running on the virtual server 20A connected to the uninterruptible power supply 30A with the stop order “1”, the virtual machine 21A1 has no stop order as shown in FIG. Since it is the same as the power failure power supply 30A, live migration is not performed. Since the stop order is “2” as shown in FIG. 4, the virtual machine 21A2 moves to the virtual server 20B connected to the uninterruptible power supply 30B with the stop order “2” by live migration. Since the stop order is “3” as shown in FIG. 4, the virtual machine 21A3 moves to the virtual server 20C connected to the uninterruptible power supply 30C with the stop order “3” by live migration.

  Similarly, among the virtual machines 21B1 to 21B3 that are running on the virtual server 20B connected to the uninterruptible power supply 30B with the stop order “2”, the virtual machine 21B1 has the stop order “ Since it is “1”, the virtual machine 20A moves to the virtual server 20A connected to the uninterruptible power supply 30A having the stop order “1” by live migration. The virtual machine 21B2 does not perform live migration because the stop order is the same as that of the uninterruptible power supply 30B as shown in FIG. Since the stop order is “3” as shown in FIG. 4, the virtual machine 21B3 moves to the virtual server 20C connected to the uninterruptible power supply 30C with the stop order “3” by live migration.

  Furthermore, among the virtual machines 21C1 to 21C3 that are running on the virtual server 20C connected to the uninterruptible power supply 30C with the stop order “3”, the virtual machine 21C1 has the stop order “1” as illustrated in FIG. Therefore, the virtual machine 20A moves to the virtual server 20A connected to the uninterruptible power supply 30A whose stop order is “1” by live migration. Since the stop order is “2” as shown in FIG. 4, the virtual machine 21C2 moves to the virtual server 20B connected to the uninterruptible power supply 30B with the stop order “2” by live migration. The virtual machine 21C3 does not perform live migration because the stop order is the same as that of the uninterruptible power supply 30C as shown in FIG.

  And the management server 10 performs live migration with respect to all the virtual servers 20 currently started on the virtual server 20 connected to the uninterruptible power supply 30 in a power failure state, and then the uninterruptible power supply connection management unit 11 Thus, the information in the connection management table is updated as shown in FIG.

  As described above, according to the present embodiment, by performing live migration processing based on the stop order of the uninterruptible power supply 30 and the stop order of each virtual machine 21, the uninterruptible battery life is short. On the virtual server 20 connected to the power supply device 30, the virtual machine 21 with the earlier stop order is started, and on the virtual server 20 connected to the uninterruptible power supply 30 with a long battery usable time, The slow virtual machine 21 is activated.

  Therefore, by stopping the corresponding virtual machines 21 in order from the shortest battery usable time of the uninterruptible power supply 30, the virtual machines 21 can be stopped in a preset stop order, and the virtual machines are suddenly stopped and stopped. This makes it possible to automatically and safely stop the virtual machine even when it is necessary to do so.

DESCRIPTION OF SYMBOLS 1 Virtual machine management system 10 Management server 11 Uninterruptible power supply connection management part 12 Virtual machine management part 13 Power supply state management part 14 Battery operation possible time measurement part 15 Live migration processing part 16 Stop control part 20 (20A, 20B, 20C) Virtual server 21 (21A1, 21A2, 21A3, 21B1, 21B2, 21B3, 21C1, 21C2, 21C3) Virtual machine 30 (30A, 30B, 30C) Uninterruptible power supply 40 Uninterruptible power supply

Claims (8)

A plurality of uninterruptible power supplies that supply power to connected devices in the event of a power failure; and
A plurality of virtual servers connected to each of the plurality of uninterruptible power supply units, on which one or a plurality of virtual machines operate;
The virtual server and the management server that manages the stop order of the virtual machine,
The management server
When the power failure occurs, another virtual machine operating on a predetermined virtual server is selected based on the preset stop order for each virtual machine and the stop order of each uninterruptible power supply. Execute the live migration process to move to the server,
Based on the stop order of the uninterruptible power supply, the virtual machine management is characterized in that stop processing is executed sequentially from the virtual machine operating on a virtual server connected to the higher order uninterruptible power supply. system. The management server
A stop order registration management table in which stop orders of virtual machines operating on the plurality of virtual servers preset by the user are registered;
An available time management table in which available times of the plurality of uninterruptible power supply units are registered;
The plurality of uninterruptible power supply devices, the plurality of virtual servers connected to each of the plurality of uninterruptible power supply devices, and the one or more virtual machines operating on each of the plurality of virtual servers are mutually connected. The virtual machine management system according to claim 1, further comprising a connection management table registered in association with the virtual machine management system. The management server
Based on the connection management table, the uninterruptible power supply connection management unit for managing the connection state of the uninterruptible power supply and the virtual server;
Based on the stop order registration management table, a virtual machine management unit that manages the stop order of the virtual machines;
A power state management unit that monitors a power state of the plurality of uninterruptible power supply units and determines whether or not the power failure has occurred;
An operable time measuring unit that calculates the usable time of the plurality of uninterruptible power supply devices,
A live migration processing unit that executes the live migration process on a virtual machine operating on the virtual server;
A stop control unit that executes the stop process for the virtual server and the virtual machine,
As a result of the monitoring by the power state management unit, when it is determined that the power failure has occurred,
The usable time of the plurality of uninterruptible power supply devices is calculated by the operable time measuring unit, and the available time management table is based on the calculated usable time of the plurality of uninterruptible power supply devices. The registered information is updated,
The uninterruptible power supply connection management unit determines the stop order of the plurality of uninterruptible power supplies based on the information registered in the usable time management table, and the determined uninterruptible power supplies. According to the stop order of devices, the information registered in the connection management table is sorted in order from the top of the stop order,
The live migration processing unit executes the live migration process based on the rearranged connection management table,
The uninterruptible power supply connection management unit updates the information of the virtual machine operating on each virtual server in the connection management table based on the execution result of the live migration process,
The stop control unit sequentially executes the stop processing for the plurality of virtual servers and the virtual machines operating on the plurality of virtual servers according to the stop order of the plurality of uninterruptible power supply units based on the connection management table. The virtual machine management system according to claim 2, wherein:   The live migration processing unit compares the stop order of each virtual machine with the stop order of each uninterruptible power supply based on the rearranged connection management table, and the virtual machine with the highest stop order is Move the virtual machine connected to the uninterruptible power supply having the higher stop order and move the virtual machine having the lower stop order to the virtual server connected to the uninterruptible power supply having the lower stop order The virtual machine management system according to claim 3, wherein the virtual machine management system is moved to a virtual machine management system.   The stop control unit is based on the connection management table, on the virtual server connected to the uninterruptible power supply having the highest stop order among the one or more virtual machines operating on the plurality of virtual servers. The stop process is executed by sequentially stopping virtual machines that operate in a plurality of virtual servers and sequentially stopping all virtual machines that operate among the virtual servers that have stopped. 5. The virtual machine management system according to 3 or 4.   6. The operable time measuring unit calculates the available time based on power consumption of the plurality of virtual servers connected to the plurality of uninterruptible power supply devices. The virtual machine management system described in 1. When a power failure occurs, a plurality of virtual servers connected to each of a plurality of uninterruptible power supply devices that supply power to the connected devices, and 1 operating on each of the plurality of virtual servers A management server that manages the stop order of one or more virtual machines,
When the power failure occurs, another virtual machine operating on a predetermined virtual server is selected based on the preset stop order for each virtual machine and the stop order of each uninterruptible power supply. Execute the live migration process to move to the server,
A management server that sequentially executes stop processing from the virtual machines that operate on a virtual server connected to an uninterruptible power supply having a higher stop order based on the stop order of the uninterruptible power supply. When a power failure occurs, a plurality of uninterruptible power supply devices that supply power to the connected device, and one or a plurality of virtual machines are connected to each of the plurality of uninterruptible power supply devices A virtual machine management method in a virtual machine management system comprising a plurality of virtual servers and a management server that manages the stop order of the virtual servers and the virtual machines,
When the power failure occurs, another virtual machine operating on a predetermined virtual server is selected based on the preset stop order for each virtual machine and the stop order of each uninterruptible power supply. Execute the live migration process to move to the server,
Based on the stop order of the uninterruptible power supply, the virtual machine management is characterized in that stop processing is executed sequentially from the virtual machine operating on a virtual server connected to the higher order uninterruptible power supply. Method.

Images